Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
For peer review only
A Prospective Multicenter Study of the Incidence of Adverse Drug Events in Saudi Arabia Hospitals: The (ADESA) Study
Journal: BMJ Open
Manuscript ID bmjopen-2015-010831
Article Type: Research
Date Submitted by the Author: 15-Dec-2015
Complete List of Authors: Aljadhey, Hisham; College of Pharmacy, King Saud University, Medication Safety Research Chair Mahmoud, Mansour Ahmed, Yusuf; King Saud University, Medication Safety Research Chair Sultana, Razia; Specialized Medical Center Hospital Zouein, Salah; Specialized Medical Center Hospital Alshanawani, Sulafa ; King Saud Medical City Mayet, Ahmed; King Khaled University Hospital
Alshaikh, Mashael; King Khaled University Hospital Kalagi, Nora; King Saud University, Medication Safety Research Chair altawil, esra; King Khaled University Hospital El Kinge, Abdul Rahman ; Specialized Medical Center Hospital Arwadi, Abdulmajid; Specialized Medical Center Hospital Alyahya, Maha; King Salman Hospital Murray, Michael; Purdue University and Regenstrief Institute Bates, David; Division of General Medicine
<b>Primary Subject Heading</b>:
Epidemiology
Secondary Subject Heading: Pharmacology and therapeutics, General practice / Family practice,
Intensive care, Research methods, Surgery
Keywords: Adverse Drug Events (ADEs), prospective cohort study, Saudi Arabia
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on D
ecember 29, 2020 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. D
ownloaded from
For peer review only
1
A Prospective Multicenter Study of the Incidence of Adverse Drug Events in Saudi Arabia Hospitals: The (ADESA) Study
Hisham Aljadhey1, Mansour A Mahmoud1, Yusuf Ahmed1, Razia Sultana2, Salah Zouein2, Sulafah Alshanawani3, Ahmed Mayet4, Mashael Alshaikh4, Nora Kalagi1, Esraa Al Tawil4, Abdul Rahman Agha El Kinge2, Abdulmajid Arwadi2, Maha Alyahya5, Michael
D Murray6, David Bates7
1 Medication Safety Research Chair, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia
2Specialized Medical Center, Riyadh, Saudi Arabia
3King Saud Medical City, Riyadh, Saudi Arabia
4King Khaled University Hospital, Riyadh, Saudi Arabia
5King Salman Hospital, Riyadh, Saudi Arabia
6Purdue University and Regenstrief Institute, Indianapolis, United States
7Harvard Medical School and Brigham and Women's Hospital, Boston, United States.
Number of words: 2413
Corresponding author: Hisham Aljadhey, Pharm D, PhD Director of Medication Safety Research Chair Dean, College of Pharmacy, King Saud University P.O.Box 2475, Riyadh 11451, Saudi Arabia Email: [email protected] Phone: +966 530039008
Page 1 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
2
ABSRACT
Background: Few studies have investigated the epidemiology of adverse drug events
(ADEs) in developing countries.
Objective: To determine the incidence of ADEs and assess their severity and
preventability in four Saudi hospitals which represent public, private, teaching, small,
and large hospitals.
Method: We performed a prospective cohort study of patients admitted to medical,
surgical and intensive care units of four hospitals. These hospitals are a 900 beds tertiary
teaching hospital, a 400 beds private hospital, a 1400 beds large government hospital and a
350 beds small government hospital. Incidents were collected by pharmacists and
reviewed by independent clinicians. Reviewers classified the identified incidents into
ADEs, potential ADEs and medication errors and determined their severity and
preventability.
Result: We followed 4041 patients from admission to discharge. Of those 3985
patients had complete data and were analyzed. The mean age of patients in the
analyzed cohort was 43 (±19.5) years. A total of 1676 incidents were identified by
pharmacists during medical chart review. Clinicians reviewers accepted 1531(91.4%) of
the incidents found by pharmacists (245 ADEs, 677 PADEs and 609 medication errors
with low risk to cause harm). The incidence of ADEs was 6.1 (95% CI, 5.4-6.9) per 100
admissions and 7.9 (95% CI, 6.9 – 8.9) per 1000 patient days. The occurrence of ADEs
was most common in the intensive care units 149 (60.8%) followed by medical
67(27.3%) and surgical units 29(11.8%). In terms of severity, 129 (52.7%) of the ADEs
Page 2 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
3
were significance, 91 (37.1%) were serious, 22 (9%) were life-threatening and 3 (1.2%)
were fatal.
Conclusion: We found that ADEs were common in Saudi hospitals, especially in the
ICUs, and they caused significant morbidity and mortality. Future studies should focus
on investigating the root causes of ADEs at the prescribing stage and development and
testing of interventions to minimize harm from medications.
Key words: Adverse Drug Events (ADEs), prospective cohort study, Saudi Arabia
Page 3 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
4
INTRODUCTION
Adverse drug events (ADEs) are major cause of morbidity, mortality, extra
healthcare costs and hospitalization (1-3). An ADE is defined as an injury occurring from
the use of drugs (4). They are largely preventable and occur mostly at the prescribing
stage of medication use process (2, 4, 5). The incidence of ADEs reported in the
literature vary significantly between countries. This is largely because of the differences
in practices, training, study methodology and patient safety initiatives among the
particular countries settings. Early in 1995 Bates et al from the USA reported an
incidence of 6.5 per 100 admissions (2) while using the same methods a study in
Japan reported an incidence of 17 per 100 admissions (4). In Saudi Arabia one single
hospital study reported an incidence of 8.5 per 100 admissions (5) and a cross-sectional
study in morocco reported an incidence of 4.2 per 100 admissions (6). A recent
international study using hospital datasets estimated the prevalence of ADEs to be
3.2% in England, 4.8% in Germany and 5.6% in the USA (7). It is important to mention
that the incidence of preventable ADEs was estimated by a population based study to
be 13.8 per 1000 person-years(8).
Despite the evidence that ADEs are common and threatening patient life, little
attempts had been made in Saudi Arabia to detect and estimate the incidence of ADEs
in hospitalized patients. Such lack of studies will hinder the designing of prevention
strategies and improve patient safety. To date, one prospective chart review study had
been conducted in a single teaching hospital in the Saudi settings (5). Therefore we
thought to study this issue with a larger sample from different hospitals which might
have different practices or strategies of managing patients to come up with a better
Page 4 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
5
estimate of ADEs. The objective of our study was to estimate the incidence of ADEs in
Saudi hospitals and determine their severity and preventability.
Page 5 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
6
METHODS
Study design and patient population
The Adverse Drug Events in Saudi Arabia (ADESA) study was a prospective cohort
study involving four hospitals from Riyadh, Saudi Arabia. These hospitals are a 900 beds
tertiary teaching hospital, a 400 beds private hospital, a 1400 beds large government hospital
and a 350 beds small government hospital. We randomly selected medical, surgical and
intensive care units (ICUs) from these hospitals and excluded obstetrics and pediatric units
because of the minimum use of medications in these units. Further inclusion was patients older
than 12 years of age admitted to these units during the four month study period and patients
admitted for more than 24 hours. None of these hospitals had electronic medical records or
decision support system. All the hospitals utilized paper-based system where physician notes
including prescribed medications and nursing notes including daily administered medications
are handwritten and kept in patient files. Whereas, medication orders are sent to inpatient
pharmacy and dispensed through the unit dose systems.
Definitions
We defined ADE as an injury caused by a medication and may be preventable or not
preventable (2, 9). The non-preventable ADEs are also known as adverse drug reactions
defined by the World Health Organization as “a response to a drug which is noxious and
unintended, and which occurs at doses used in man for prophylaxis, diagnosis, or therapy of
disease, or for the modifications of physiological function”(10). Preventable ADEs are those who
result from medication errors at any stage of medication use process. A potential ADEs is an
error that carries the risk of causing injury related to the use of a medication but harm did not
occur(9).
Page 6 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
7
Data collection and classification of incidents
The method used to collect data was described in details elsewhere (5) In summary, data
was collected daily by trained clinical pharmacists. In addition all nurses working in the particular
units were invited to attend a monthly in-service to increase their awareness about ADEs
reporting. The pharmacists reviewed patients’ medical charts of all admitted patients in each of
the participating units to report demographic characteristics of patients including comorbidity
and number of medications. When incidents found the pharmacists wrote a detailed description
of the incidents and patient history related to the incidents.
Two independent clinicians, who were not involved in the data collection process, were
provided with a study manual that contains study terminology and a guide on the assessment of
the severity and preventability of an incident with example of incidents and their severity
classifications(2). The severity of the incidents was categorized as significant, serious, life
threatening or fatal. The study manual served as a guide for the reviewers to independently
review the incidents and decide on inclusion of incidents and further classify them as ADEs,
potential ADEs or ME and assess their severity and preventability. In case of disagreement
between physicians the final decision was confirmed through discussion between both
reviewers. The primary outcomes of this study were incidence of ADEs, potential ADEs and
medication errors as defined in the earlier section of methodology. The secondary outcomes
were the severity of these events, their preventability, and the associated risk factors.
Data Analysis:
We calculated the overall incidence per 100 admissions and crude rate per 1000 patient
days with 95% Confidence Interval (CI). In addition the incidence was calculated by hospital and
by units. Continuous variables are presented as mean ± standard deviation (SD) and
categorical variables as number and percentages. The Inter-rater reliability was assessed using
Page 7 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
8
kappa statistic for assessment of the presence of an ADE and its preventability and severity. To
evaluate the univariate association of potential risk factors with ADEs, we used univariate
logistic regression. Variables which were found to be significance in the univariate analysis were
included in the multivariate logistic regression final model. Statistical analyses were conducted
using the Statistical Package for Social Science (SPSS) software (SPSS Inc., Chicago, IL),
version 22.0.
Page 8 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
9
RESULTS
Demographic Characteristics of the patients
Medical charts of 4,041 patients were reviewed by clinical pharmacists. Complete
data of 3,985 patients were analyzed (Table 1). The length of hospital stay for these
patients was 30,996 days. The study was conducted in four hospitals in Riyadh, Saudi
Arabia (977 patients from a teaching Hospital, 2033 patients from a private Hospital;
683 patients from large government Hospital; and 292 patients from a small government
hospital). Majority of the patients were male 2102 (52.7%) (Table1). Patients were
admitted to three different services (Medicine; 1352 patients, surgery; 1771 patients and
Intensive Care Units; 862 patients). The mean length of hospital stay was 8.1 ± 10.2
days and the mean age of patient was 43.4 ± 19.0 years (Table 2).
Incidents review and classification
Chart review by pharmacist in the four hospitals identified 1676 cases. The
physician reviewed these cases and finally accepted 1531 (91.3%) of the cases. These
cases were classified as 609 (39.8%) medication errors with low risk of harm, 677
(44.2%) potential adverse drug events (PADEs) and 245 (16%) cases were judged to
be ADEs. Among ADEs 85 (34.7%) were preventable and 160 (65.3%) were judged to
be non-preventable (Table 2). Majority of the preventable ADEs occurred in the
prescribing stage 75 (88.2%) followed by administering 7(8.2%), dispensing 2(2.4% and
monitoring 1(1.2%). One hundred twenty nine (52.7%) of all ADEs were significant, 91
(37.1%) were serious, 22 (9%) were life-threatening and 3 (1.2%) were fatal. Of the 85
preventable ADEs 36 (42.4%) were significant, 38 (44.7%) were serious, 10 (11.9%)
were life-threatening and 1 (1.2%) was fatal. Among potential ADEs 213 (31.9%) were
Page 9 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
10
intercepted by the medical staff. Regarding severity of potential ADEs 383 (56.6%) were
significant, 271 (40%) were serious and 23 (3.4%) were life-threatening.
Overall incidence of ADEs and Medication Errors
The incidence of ADEs per 100 admissions was 6.1(95% CI 5.4 – 6.9), and the
incidence per 1000 patient days was 7.9(95% 6.9 – 8.9) (Table 2). The incidence of
potential ADEs was 16.9 (95% CI 15.7 – 18.3) per 100 admissions and 21.8 (95% CI
20.2 – 23.5) per 1000 patient days. The incidence of medication Errors with low risk to
cause harm was 15.3 (95% CI 14.1 – 16.5) per 100 admissions and 19.6 (95% CI 18.7
– 21.2) per 1000 patient days. The incidence of preventable ADEs was 2.1(95% CI 1.7
– 2.6) per 100 admissions and 2.7 (95% CI 2.2 – 3.3) per 1000 patient days. The
incidence of not preventable ADEs was 4.0 (95% CI 3.4 – 4.6) per 100 admissions and
5.1 (95% CI 4.4 – 6.0) per 1000 patient days. The incidence of intercepted potential
ADEs was 5.3 (95% CI 4.6 – 6.1) per 100 admissions and 6.8 (95% CI 5.9 – 7.8) per
1000 patient days (Table 2). Incidents most commonly occurred in the prescribing stage
1288 (84.1%) followed by dispensing stage 69 (4.5%) and administering 43 (2.8%).
Table 3 shows the distribution of incidents among the four hospitals. The incidence of
ADEs was most common in the small government hospital 96 (6.3%) while the
incidence of PADEs was predominantly higher in private hospital 367 (23.9%).
Medication errors were mostly seen in private hospital 367 (23.9%).
Classification of the 1531 Incidents by service
The incidence of ADEs was higher in the ICUs 13.7(95% CI 11.6-16.1) per 1000
patients days and 17.4(95% CI 14.7-20.3) per 100 admissions followed by the medical
Page 10 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
11
units 6.1(95% CI 4.7-7.7) per 1000 patients days and 4.8(95% CI 3.8-6.1) per 100
admissions (Table 4).
Agreement of physician’s reviewers on the classification of the incidents
The kappa value for the presence of ADEs was 0.71, for the presence of
medication errors it was 0.67 and for the presence of potential ADEs was 0.60. The
kappa value for preventability of ADEs was 0.68 (definitely or probably preventable vs.
definitely or probably not preventable). For the severity of ADEs the kappa score was
0.74 (fatal vs. significant, serious or life-threatening), 0.63 (Life-threatening vs.
significant, serious or fatal), 0.53 (Significant vs. serious, life-threatening or fatal), 0.48
(Serious vs. life-threatening or fatal).
Factors Associated with ADEs
Factors significantly associated with ADEs included older age; (OR, 1.013;
95%CI, 1.004 – 1.021) more number of medications (OR, 1.070; 95%CI, 1.018 – 1.125),
greater length of hospital stay (OR, 1.026; 95%CI, 1.016 – 1.036) and admission to the
ICUs and medicine units respectively (OR, 3.131; 95%CI, 1.937 – 5.063) and (OR,
1.729; 95%CI, 1.086 – 2.755) (Table 6).
Page 11 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
12
DISCUSSION
In this study we evaluated the incidence of ADEs and found that ADEs were
common with one third caused by medication errors and therefore were judged to be
preventable. Errors resulting from preventable ADEs were most common in the
prescribing stage followed by dispensing and administration. Majority of the preventable
ADEs were judged to be serious. The incidence of ADEs reported in our study was
similar to previous studies (2, 5). However, a higher incidence was reported in Japan
(4). The differences between our study result and this study could be the longer length
of hospital stay in Japan and differences in healthcare systems between both countries.
The similarity between our findings and the US study could be because of the similarity
in the healthcare systems between both countries.
We included 3985 patients and found 245 ADEs of which 35% were judged to be
preventable. Gurwitz and colleagues (8) identified 546 ADEs during 2403 nursing home
residence admissions and reported that 51% of the observed ADEs were preventable.
Bates et al (2) identified incidence of ADEs in 4031patients and found 247 ADEs of
which 28% were judged preventable with more serious ADEs to be preventable. In 2009
hug et al(11) assessed the occurrence of ADEs in 1200 patient from six community
hospitals and identified 180 ADEs of which 75% were preventable. Recently a
multicenter cohort study of 3459 patients identified 1010 ADEs and found that 14% of
the identified ADEs were preventable (4).
Page 12 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
13
Regarding potential ADEs we noticed only one third of the events were
intercepted. It is noteworthy to highlight that three of the four hospitals had clinical
pharmacists monitoring patient treatments and most of the intercepted potential ADEs
were in those hospitals.
Our study revealed that ADEs were associated with admission to ICUs and older
age. Consistent with our results other studies also reported that admission to ICUs (4,
12) and older age (4, 8) as major factors associated with ADEs. In support to this
finding perhaps especial care should be given to elderly who are admitted to ICUs
because of the added risk of combining two risk factors.
Several important basic medication safety practices are not adopted in most
Saudi hospitals (13, 14) . Therefore, there are opportunities for improving the safe use
of medications and preventing ADEs in hospitals in Saudi Arabia. On a national level,
the Saudi Food Drug Authority may lead efforts to prevent adverse drug reactions and
the Saudi Medication Safety Center may lead initiatives to prevent medication errors.
For example, the use of pharmacist to ascertain complete medication histories at
admission and provide discharge counseling reduced the incidence of ADEs (15, 16).
Although reporting is a good tool to identify and prevent ADEs, underreporting is a
common challenge in Saudi hospitals(17)
Future research could focus on investigating the causes of ADEs that occur
during the medication use process especially prescribing. Causes of ADEs need to be
identified through both qualitative and quantitative approaches. A systematic approach
need to be used to classify these causes. Using methods similar to the ones used in this
Page 13 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
14
study, the benefits of Interventions can be estimated and compared to baseline. This
study is limited in not including hospitals in small towns and rural areas. However, it is
expected to find higher incidence in these areas and the objective from this study is not
only to get the incidence but to understand the severity, preventability, and seriousness
of these ADEs.
In conclusion, ADEs are common in Saudi hospitals, especially in the ICUs, and
they caused significant morbidity and mortality. Although there are variations in the
incidence of ADEs between countries the possibilities of preventing it are higher;
therefore interventions that were proved to be effective in other countries should be
tested in Saudi Arabia. Such interventions may include but not limited to,
implementations of computerized physician entry (CPOE) with clinical decision support
system(11) involvement of clinical pharmacists as part of the medical team during
physicians rounds(18-20) and changing the currently available paper-based system to
electronic medical records(21).
Contribution: HA, DW and MM designed the study. HA wrote the manuscript, MA and
YA contributed in the data analysis and management. All authors contributed to the data
collection process, the study idea and design and approved the final manuscript.
Conflict of Interest: None declared
Data sharing statement: No additional data are available.
Funding: This study was funded by the National Plan for Science and Technology (09-
BIO708-02)
Page 14 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
15
Acknowledgment: We would like to thank research assistants, Emad Zalloum, Trig
Allam, Maram Abuzaid, Umm Hani Sayeda, Shamailah Osmani, Aishah Nor, Nesreen
al-shabr, Sultan Al-Harbi, for their help during the data collection process.
Page 15 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
16
References
1. Bates DW, Spell N, Cullen DJ, Burdick E, Laird N, Petersen LA, et al. The costs of
adverse drug events in hospitalized patients. Adverse Drug Events Prevention Study Group. JAMA : the journal of the American Medical Association. 1997;277(4):307-11. Epub 1997/01/22.
2. Bates DW, Cullen DJ, Laird N, Petersen LA, Small SD, Servi D, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. JAMA : the journal of the American Medical Association. 1995;274(1):29-34. Epub 1995/07/05.
3. Al Hamid A, Ghaleb M, Aljadhey H, Aslanpour Z. A systematic review of hospitalisation resulting from medicine related problems in adult patients. British journal of clinical pharmacology. 2013. Epub 2013/11/29.
4. Morimoto T, Sakuma M, Matsui K, Kuramoto N, Toshiro J, Murakami J, et al. Incidence of Adverse Drug Events and Medication Errors in Japan: the JADE Study. J Gen Intern Med. 2011;26(2):148-53.
5. Aljadhey H, Mahmoud MA, Mayet A, Alshaikh M, Ahmed Y, Murray MD, et al. Incidence of adverse drug events in an academic hospital: a prospective cohort study. Int J Qual Health C. 2013;25(6):648-55.
6. Benkirane R, Pariente A, Achour S, Ouammi L, Azzouzi A, Soulaymani R. Prevalence and preventability of adverse drug events in a teaching hospital: a cross-sectional study. Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit. 2009;15(5):1145-55. Epub 2010/03/11.
7. Stausberg J. International prevalence of adverse drug events in hospitals: an analysis of routine data from England, Germany, and the USA. BMC health services research. 2014;14:125. Epub 2014/03/14.
8. Gurwitz JH, Field TS, Harrold LR, Rothschild J, Debellis K, Seger AC, et al. Incidence and preventability of adverse drug events among older persons in the ambulatory setting. Jama-J Am Med Assoc. 2003;289(9):1107-16.
9. Morimoto T, Gandhi TK, Seger AC, Hsieh TC, Bates DW. Adverse drug events and medication errors: detection and classification methods. Qual Saf Health Care. 2004;13(4):306-14.
10. WHO WHO. Definitions. [cited 2014]; Available from: http://www.who.int/medicines/areas/quality_safety/safety_efficacy/trainingcourses/definitions.pdf
11. Hug BL, Witkowski DJ, Sox CM, Keohane CA, Seger DL, Yoon C, et al. Adverse Drug Event Rates in Six Community Hospitals and the Potential Impact of Computerized Physician Order Entry for Prevention. J Gen Intern Med. 2010;25(1):31-8.
12. Benkirane RR, Abouqal R, Haimeur CC, SS SECEK, Azzouzi AA, Mdaghri Alaoui AA, et al. Incidence of adverse drug events and medication errors in intensive care units: a prospective multicenter study. Journal of patient safety. 2009;5(1):16-22. Epub 2009/11/19.
13. Aljadhey H, Alhusan A, Alburikan K, Adam M, Murray MD, Bates DW. Medication safety practices in hospitals: A national survey in Saudi Arabia. Saudi Pharm J. 2013;21(2):159-64. Epub 2013/08/21.
14. Alkhani S, Ahmed Y, Bin-Sabbar N, Almogirah H, Alturki A, Albanyan H, et al. Current practices for labeling medications in hospitals in Riyadh, Saudi Arabia. Saudi Pharm J. 2013;21(4):345-9. Epub 2013/11/15.
Page 16 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
17
15. AbuYassin BH, Aljadhey H, Al-Sultan M, Al-Rashed S, Adam M, Bates DW. Accuracy of the medication history at admission to hospital in Saudi Arabia. Saudi Pharm J. 2011;19(4):263-7.
16. Al-Ghamdi SA, Mahmoud MA, Alammari MA, Al Bekairy AM, Alwhaibi M, Mayet AY, et al. The outcome of pharmacist counseling at the time of hospital discharge: an observational nonrandomized study. Ann Saudi Med. 2012;32(5):492-7.
17. Alshaikh M, Mayet A, Aljadhey H. Medication error reporting in a university teaching hospital in Saudi Arabia. Journal of patient safety. 2013;9(3):145-9. Epub 2013/02/02.
18. Kucukarslan SN, Peters M, Mlynarek M, Nafziger DA. Pharmacists on rounding teams reduce preventable adverse drug events in hospital general medicine units. Arch Intern Med. 2003;163(17):2014-8.
19. Schnipper JL, Kirwin JL, Cotugno MC, Wahlstrom SA, Brown BA, Tarvin E, et al. Role of pharmacist counseling in preventing adverse drug events after hospitalization. Arch Intern Med. 2006;166(5):565-71. Epub 2006/03/15.
20. Leape LL, Cullen DJ, Clapp MD, Burdick E, Demonaco HJ, Erickson JI, et al. Pharmacist participation on physician rounds and adverse drug events in the intensive care unit. Jama-J Am Med Assoc. 1999;282(3):267-70.
21. Liu M, Hinz ERM, Matheny ME, Denny JC, Schildcrout JS, Miller RA, et al. Comparative analysis of pharmacovigilance methods in the detection of adverse drug reactions using electronic medical records. J Am Med Inform Assn. 2013;20(3):420-6.
Page 17 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
18
Tables
Table1. Demographic Characteristics of 3985 patients admitted to four Hospitals in Riyadh Frequency (%) Mean (±SD)
Gender
Male 2102 (52.7) -
Female 1883(47.3) -
Hospital type
Hospital 1(Teaching Hospital) 977(24.5) -
Hospital 2 (Private Hospital) 2033(51.1) -
Hospital 3(Large government Hospital) 683(17.1) -
Hospital 4 (Small government Hospital) 292(7.3) -
Service
Medicine 1352(33.9) -
Surgery 1771(44.5) -
Intensive Care Unit (ICU) 862(21.6) -
Age - 43.4 (±19.0)
Length of Hospital Stay - 8.1 (±10.2)
Charlson’s Comorbidity Score - 1.1 (±1.4)
Number of Medications - 2.5 (±2.9)
Table 2. Overall Incidence
Total number
of incidents
=1531
% Incidence per 100 admissions
(95% CI)
Crude rate per 1000 patient days (95%
CI)
Medication Errors with low risk to cause harm 609 39.8 15.3(14.1 – 16.5) 19.6 (18.7 – 21.2)
Potential ADEs 677 44.2 16.9 (15.7 – 18.3) 21.8 (20.2 – 23.5)
Intercepted Potential ADEs (N=213) 5.3 (4.6 – 6.1) 6.8(5.9 – 7.8)
Not intercepted Potential ADEs (N=464) 11.6 (10.6 – 12.7) 14.9 (13.6 – 16.3)
ADEs (Harm) 245 16 6.1 (5.4 – 6.9) 7.9 (6.9 – 8.9)
Preventable ADEs (N=85) 2.1(1.7 – 2.6) 2.7 (2.2 – 3.3)
Non-preventable ADEs (N=160) 4.0 (3.4 – 4.6) 5.1 (4.4 – 6.0)
ADEs, adverse drug events
Table 3. Classification of ADEs by Hospitals Type
Units ADEs Length of hospital
stay
ADEs crude rate per 1000 patient days (95% CI)
Number of admissions
ADEs Incidence per 100 admissions
(95% CI)
Hospital 1
83 9585 8.7 (6.9-10.6) 977 8.5(6.8-10.4)
Hospital 2
53 9032 5.9 (4.4-7.6) 2033 2.6 (2.1-3.3)
Hospital 3
13 6613 2.1(1.1-3.2) 683 2.1(1.1-3.2)
Hospital4
96 5766 16.6(13.5-20.2) 292 32.9(27.7-38.4)
Hospital 1= Teaching hospital, Hospital 2=Private hospital, Hospital 3= small government hospital, Hospital 4=Large government hospital
Table4. Classification of ADEs incidence by Type of Services
Units ADEs Patient days, No
ADEs Crude rate per 1000 patient days (95% CI)
Number of admissions
ADEs Incidence per 100 admissions
(95% CI)
Medical 66 10767 6.1(4.7-7.7) 1352 4.8(3.8-6.1)
Surgical 29 9310 3.1(2.1-4.4) 1771 1.6(1.1-2.3)
ICU 150 10919 13.7(11.6-16.1) 862 17.4(14.7-20.3)
ADEs, adverse drug events
Page 18 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
19
Table 5. Factors Associated with ADEs
Factor Unadjusted Odds Ratio
95% CI Adjusted Odds Ratio
95% CI
Lower Upper Lower Upper
Age 1.024 1.017 1.032 1.012 1.003 1.021
Number of medications 1.193 1.148 1.240 1.062 1.008 1.119
Charlson’s comorbidity index 1.251 1.158 1.352 1.041 0.937 1.157
Length of hospital stay 1.042 1.034 1.051 1.025 1.015 1.035
Gender (Male)a 0.848 0.642 1.121 - - -
ICUb 7.786 5.259 11.528 3.276 2.005 5.354
Medicineb 2.539 1.664 3.872 1.736 1.078 2.796
CI, Confidence Interval, Reference categories: Femalea, Reference categories: Surgery
b
Page 19 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
STROBE 2007 (v4) Statement—Checklist of items that should be included in reports of cohort studies
Section/Topic Item
# Recommendation Reported on page #
Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract Reported on page 1
(b) Provide in the abstract an informative and balanced summary of what was done and what was found Reported on page 1
Introduction
Background/rationale 2 Explain the scientific background and rationale for the investigation being reported Reported on page 4
Objectives 3 State specific objectives, including any prespecified hypotheses Reported on page 5
Methods
Study design 4 Present key elements of study design early in the paper Reported on page 6
Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data
collection
Reported on page 1
Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up Reported on page 7
(b) For matched studies, give matching criteria and number of exposed and unexposed
Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if
applicable
Data sources/
measurement
8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe
comparability of assessment methods if there is more than one group
Bias 9 Describe any efforts to address potential sources of bias
Study size 10 Explain how the study size was arrived at
Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and
why
Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding Reported on page 7
(b) Describe any methods used to examine subgroups and interactions
(c) Explain how missing data were addressed
(d) If applicable, explain how loss to follow-up was addressed
(e) Describe any sensitivity analyses
Results
Page 20 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on December 29, 2020 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. Downloaded from
For peer review only
Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, confirmed
eligible, included in the study, completing follow-up, and analysed
Reported on page 9
(b) Give reasons for non-participation at each stage
(c) Consider use of a flow diagram
Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential
confounders
Reported on page 9
(b) Indicate number of participants with missing data for each variable of interest
(c) Summarise follow-up time (eg, average and total amount) Reported on page 9
Outcome data 15* Report numbers of outcome events or summary measures over time
Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (eg, 95% confidence
interval). Make clear which confounders were adjusted for and why they were included
Reported on page 11
(b) Report category boundaries when continuous variables were categorized
(c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period
Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses
Discussion
Key results 18 Summarise key results with reference to study objectives Reported on page 12
Limitations
Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from
similar studies, and other relevant evidence
Reported on page 14
Generalisability 21 Discuss the generalisability (external validity) of the study results Reported on page 14
Other information
Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on
which the present article is based
Reported on page 14
*Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies.
Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE
checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at
http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org.
Page 21 of 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on December 29, 2020 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. Downloaded from
For peer review only
A Prospective Multicenter Study of the Incidence of Adverse Drug Events in Saudi Arabia Hospitals: The (ADESA) Study
Journal: BMJ Open
Manuscript ID bmjopen-2015-010831.R1
Article Type: Research
Date Submitted by the Author: 15-Apr-2016
Complete List of Authors: Aljadhey, Hisham; College of Pharmacy, King Saud University, Medication Safety Research Chair Mahmoud, Mansour Ahmed, Yusuf; King Saud University, Medication Safety Research Chair Sultana, Razia; Specialized Medical Center Hospital Zouein, Salah; Specialized Medical Center Hospital Alshanawani, Sulafa ; King Saud Medical City Mayet, Ahmed; King Khaled University Hospital
Alshaikh, Mashael; King Khaled University Hospital Kalagi, Nora; King Saud University, Medication Safety Research Chair altawil, esra; King Khaled University Hospital El Kinge, Abdul Rahman ; Specialized Medical Center Hospital Arwadi, Abdulmajid; Specialized Medical Center Hospital Alyahya, Maha; King Salman Hospital Murray, Michael; Purdue University and Regenstrief Institute Bates, David; Division of General Medicine
<b>Primary Subject Heading</b>:
Epidemiology
Secondary Subject Heading: Pharmacology and therapeutics, General practice / Family practice,
Intensive care, Research methods, Surgery
Keywords: Adverse Drug Events (ADEs), prospective cohort study, Saudi Arabia
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on D
ecember 29, 2020 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. D
ownloaded from
For peer review only
1
Incidence of Adverse Drug Events in Public and Private Hospitals in Riyadh, Saudi Arabia: The (ADESA) Prospective Cohort Study
Hisham Aljadhey1, Mansour A Mahmoud1, Yusuf Ahmed1, Razia Sultana2, Salah Zouein2, Sulafah Alshanawani3, Ahmed Mayet4, Mashael K Alshaikh4, Nora Kalagi1,
Esraa Al Tawil4, Abdul Rahman El Kinge5, Abdulmajid Arwadi5, Maha Alyahya6, Michael D Murray7, David Bates8
1 Medication Safety Research Chair, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia
2Specialized Medical Center, Department of Pharmacy, Riyadh, Saudi Arabia
3King Saud Medical City, Department of Pharmacy, Riyadh, Saudi Arabia
4King Khaled University Hospital, Department of Pharmacy, Riyadh, Saudi Arabia
5Specialized Medical Center, Department of Internal Medicine, Riyadh, Saudi Arabia
6King Salman Hospital, Department of Pharmacy, Riyadh, Saudi Arabia
7Purdue University and Regenstrief Institute, College of Pharmacy, Indianapolis, United States
8Harvard Medical School and Brigham and Women's Hospital, Department of Medicine, Boston, United States.
Number of words
Text: 2987
Abstract: 299
Corresponding author: Hisham Aljadhey, Pharm D, PhD Director of Medication Safety Research Chair Dean, College of Pharmacy, King Saud University
Page 1 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
2
Supervisor of Pharmacy Services at King Saud University Medical CityP.O.Box 2475, Riyadh 11451, Saudi Arabia Email: [email protected] Phone: +966 530039008
Page 2 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
3
ABSRACT
Objectives: To determine the incidence of ADEs and assess their severity and
preventability in four Saudi hospitals.
Design: Prospective cohort study
Setting: The study included patients admitted to medical, surgical and intensive care
units of four hospitals in Saudi Arabia. These hospitals include a 900 bed tertiary
teaching hospital, a 400 bed private hospital, a 1400 bed large government hospital and
a 350 bed small government hospital.
Participants: All patients (≥12 years) admitted to the study units during four month.
Primary and secondary outcome measures: Incidents were collected by pharmacists
and reviewed by independent clinicians. Reviewers classified the identified incidents as
ADEs, potential ADEs (PADEs) and medication errors and then determined their
severity and preventability.
Results: We followed 4041 patients from admission to discharge. Of those, 3985
patients had complete data for analysis. The mean age of patients in the analysed
cohort was 43 (±19.5) years. A total of 1676 incidents of ADEs were identified by
pharmacists during the medical chart review. Clinician reviewers accepted 1531(91.4%)
of the incidents identified by the pharmacists (245 ADEs, 677 PADEs and 609
medication errors with low risk to cause harm). The incidence of ADEs was 6.1 (95% CI,
5.4-6.9) per 100 admissions and 7.9 (95% CI, 6.9 – 8.9) per 1000 patient-days. The
occurrence of ADEs was most common in the intensive care units 149 (60.8%) followed
Page 3 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
4
by medical 67(27.3%) and surgical units 29(11.8%). In terms of severity, 129 (52.7%) of
the ADEs were significant, 91 (37.1%) were serious, 22 (9%) were life-threatening and
three (1.2%) were fatal.
Conclusions: We found that ADEs were common in Saudi hospitals, especially in the
ICUs, causing significant morbidity and mortality. Future studies should focus on
investigating the root causes of ADEs at the prescribing stage and development and
testing of interventions to minimise harm from medications.
Key words: adverse drug events (ADEs), prospective cohort study, hospitals, Saudi
Arabia
Strength and limitations
• This study is one of the largest studies investigating the incidence of ADEs in the
Middle East.
• This study is limited by a lack of hospitals from small towns and rural areas and
that these settings have an even higher incidence of ADEs.
• Our study findings are not generalisable to overall Saudi Arabia because the
study was conducted in Riyadh only.
Page 4 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
5
INTRODUCTION
Adverse drug events (ADEs) are major cause of morbidity, mortality, and
increased healthcare costs and hospitalization (1-3). An ADE is defined as an injury
caused by a medication (4). They are largely preventable and occur mostly at the
prescribing stage of the medication use process (2, 4, 5). The incidence of ADEs
reported in the literature varies significantly between countries largely because of the
differences in available drug products, practices, training, study methodology and
patient safety initiatives among countries. Early in 1995 Bates et al from the United
States (U.S) reported an incidence of 6.5 per 100 admissions (2); however, while using
the same methods a study in Japan reported an incidence of 17 per 100 admissions (4)
suggesting real differences between these two countries. In Saudi Arabia, a single
hospital study reported an incidence of 8.5 per 100 admissions (5) and a cross-sectional
study in Morocco reported an incidence of 4.2 per 100 admissions (6). A recent
international study using hospital datasets estimated the prevalence of ADEs to be
3.2% in England, 4.8% in Germany and 5.6% in the U.S (7). It is important to mention
that the incidence of preventable ADEs was estimated by a population-based study to
be 13.8 per 1000 person-years (8).
Despite the evidence that ADEs are common and could be life threatening, little
attempt has been made in Saudi Arabia to detect and estimate the incidence of ADEs in
hospitalised patients. Such a paucity of research hinders the development of prevention
strategies to improve patient safety. To date, one prospective chart review study has
been conducted in a single teaching hospital in the Saudi setting (5). Therefore we
sought to estimate incident ADEs with a larger patient sample from different hospitals
Page 5 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
6
with diverse settings with varying practices and strategies for managing patients.
Therefore, the objective of our study was to estimate the incidence and risk factors
associated with ADEs in Saudi hospitals and determine their severity and preventability.
Page 6 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
7
METHODS
Study design and setting
The Adverse Drug Events in Saudi Arabia (ADESA) project was a four month
prospective cohort study involving four hospitals with diverse settings. These hospitals
included a 900 bed tertiary teaching hospital, a 400 bed private hospital, a 1400 bed
large government hospital and a 350-bed small government hospital. We randomly
selected medical, surgical and intensive care units (ICUs) from these hospitals and
excluded obstetrics and pediatric units because of the lower frequency of use of
medications within these units. We included patients older than 12 years of age
admitted for more than 24 hours during the four-month study period. None of the
hospitals had electronic medical records or decision support systems. Instead, hospitals
utilized paper-based systems where physician notes (including prescribed medications)
and nursing notes (including daily administered medications) were handwritten and kept
in patient charts. Medication orders were sent to the inpatient pharmacies and
dispensed using unit dose systems.
Definitions
Each incident was defined as an ADE (preventable and non-preventable), PADE that
were classified as either intercepted or non-intercepted or a medication errors with low
risk to cause harm. We defined ADE as a preventable injury that was caused by a
medication (2, 9). Non-preventable ADEs, also known as adverse drug reactions
(ADRs), are defined by the World Health Organization as “a response to a drug which is
noxious and unintended, and which occurs at doses used in man for prophylaxis,
Page 7 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
8
diagnosis, or therapy of disease, or for the modification of physiological function”(10).
Preventable ADEs were those that result from medication errors at any stage of the
medication use process (2). A PADE was an error that carries a risk of causing injury
related to the use of a medication but harm did not occur, either because of specific
circumstances or because the error was intercepted (9). Intercepted PADEs were those
that had the potential to cause injury but did not reach the patient because they were
intercepted by someone during the medication use process; Non-intercepted potential
ADEs were those with the potential to cause harm but failed to do so after the
medication reached the patient (9). Medication errors with a low risk to cause harm
included those with minimal risk to cause ADEs or PADEs. Comorbidities were
determined using Charlson's Comorbidity Index, which is a method of classifying
comorbidities of patients according to the International Classification of Disease (ICD).
Each comorbidity class has an associated weight of 1, 2, 3 or 6. The sum of all weights
results in a single comorbidity score for each patient with higher scores predictive of
adverse outcomes such as mortality or high resource use.
Data collection and classification of incidents
Data were collected as described in details elsewhere (5). Briefly, trained clinical
pharmacists collected data each day during the study period. In addition, all nurses
working in the particular units were invited to attend monthly in-service presentations
about the study to increase their awareness about ADEs reporting. The pharmacists
reviewed patients’ medical charts of all admitted patients in each of the participating
units to report demographic characteristics of patients, comorbidity and the number of
Page 8 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
9
medications. When incidents were noted the pharmacists wrote a detailed description of
each incident and captured the relevant patient characteristics and event history.
Two independent clinicians who were not involved in the data collection process
were provided with a study manual that contained study terminology and a guide on the
assessment of the severity and preventability of an incident. The manual included
examples of incidents and their severity classifications. The severity of the incidents
was categorized as significant, serious, life threatening or fatal using a methodology
developed by the Brigham and Women’s Hospital’s Center for Patient Safety Research
and Practice (2). The study manual served as a guide for the reviewers to
independently review the incidents and decide on inclusion of incidents and further
classify them as ADEs, PADEs or medication errors with low risk to cause harm. They
were then able to assess severity and preventability. Preventability categories were
defined as follows: definitely preventable, probably preventable, definitely not
preventable or probably not preventable (2). In the event that there was disagreement
on the classification of the incidents, the clinicians called for a meeting to decide
whether to include or exclude the incidents. The primary outcomes of this study were
incidence of ADEs, PADEs and medication errors with low risk to cause harm, as
defined previously. The secondary outcomes were the severity of events, their
preventability, and associated risk factors. The research and ethics committees of the
four hospitals approved this study.
Data Analysis:
Page 9 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
10
We calculated the overall incidence per 100 admissions and crude rate per 1000
patient- days with 95% confidence intervals (CI). In addition the incidence was
calculated by hospital and by unit type. Continuous variables are presented as mean ±
standard deviation (SD) and categorical variables as number and percentage. Inter-
rater reliability was assessed using the kappa statistic for assessment of the presence
of an ADE and its preventability and severity. To evaluate the univariate association of
potential risk factors with ADEs, we used univariate logistic regression. The variables
included in the univariate analysis were age, gender, Charlson’s comorbidity index
weight, length of hospital stay, number of medications, and service type. Variables
found to be statistically significant (P < 0.05) in the univariate analysis were included in
the multivariate logistic regression final model. Statistical analyses were conducted
using the Statistical Package for Social Science (SPSS) software IBM SPSS Statistics
for Windows, Version 22.0. Armonk, NY: IBM Corp
Page 10 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
11
RESULTS
Demographic Characteristics of the patients
Clinical pharmacists reviewed the medical charts of 4,041 patients. Complete
data of 3,985 patients were analysed (Table 1). The total length of hospital stay for
patients was 30,996 days. The study was conducted in four hospitals in Riyadh, Saudi
Arabia (977 patients from a teaching hospital; 2033 patients from a private hospital, 683
patients from large government hospital, and 292 patients from a small government
hospital). Male patients were a slight majority (52.7%). The patients were admitted to
one of the three services (Medicine, 1352 patients; surgery, 1771 patients; and
Intensive Care Units, 862 patients). The mean length of the hospital stay was 8.1 ± 10.2
days and the mean age of the patients was 43.4 ± 19.0 years (Table 2).
Incidents review and classification
The pharmacists’ chart review s in the four hospitals identified 1676 cases of
ADEs, PADEs and medication errors. Physician reviewed and accepted 1531 (91.3%)
of the cases, which were classified as 609 (39.8%) medication errors with low risk of
harm, 677 (44.2%) PADEs, and 245 (16%) cases of ADEs. Among the ADEs 85
(34.7%) were deemed preventable and 160 (65.3%) were judged to be non-preventable
(Table 2). The majority of the preventable ADEs occurred in the prescribing stage (75 ;
88.2%) followed by administering stage (7 ;8.2%), dispensing stage (2 ;2.4% and
monitoring stage (1; ;1.2%). One hundred twenty nine (52.7%) of the ADEs were
significant, 91 (37.1%) were serious, 22 (9%) were life-threatening, and three (1.2%)
were fatal. Of the 85 preventable ADEs, 36 (42.4%) were significant, 38 (44.7%) were
serious, 10 (11.9%) were life-threatening and one (1.2%) was fatal. Among PADEs, 213
Page 11 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
12
(31.9%) were intercepted by the medical staff. Regarding severity of PADEs, 383
(56.6%) were significant, 271 (40%) were serious and 23 (3.4%) were life-threatening.
Overall incidence of ADEs and medication errors with low risk to cause harm
The incidence of ADEs per 100 admissions was 6.1(95% CI 5.4 – 6.9) and the
incidence of potential ADEs was 16.9 (95% CI 15.7 – 18.3) per 100 admissions (Table
2). The incidence of medication errors with low risk to cause harm was 15.3 (95% CI
14.1 – 16.5) per 100 admissions and the incidence of preventable ADEs was 2.1(95%
CI 1.7 – 2.6) per 100 admissions. The incidence of non-preventable ADEs was 4.0
(95% CI 3.4 – 4.6) per 100 admissions and the incidence of intercepted potential ADEs
was 5.3 (95% CI 4.6 – 6.1) per 100 admissions (Table 2). Incidents of preventable
ADEs, PADEs and medication errors with low risk to cause harm most commonly
occurred in the prescribing stage 1288 (84.1%) followed by dispensing stage 69 (4.5%)
and the administering stage (43; 2.8%). Table 3 shows the distribution of incidents
among the four hospitals. Examples of PADEs at different stages of the medication use
process are listed in Appendix A. The incidence of ADEs was higher in the large
government hospital 16.6 (95% CI 13.5-20.2) per 1000 patient-days and 32.9(95% CI
27.7-38.4) per 100 admissions followed by the teaching hospital 8.7 (95% CI 6.9-10.6)
per 1000 patient-days and 8.5(95% CI 6.8-10.4) per 100 admissions (Table 3).
The incidence of PADEs was predominantly higher in the private hospital (367;
23.9%). Medication errors were mostly seen in the private hospital (367; 23.9%).
Classification of ADEs Incidents by service type
Page 12 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
13
The incidence of ADEs was higher in the ICUs at 13.7 (95% CI 11.6-16.1) per
1000 patients-days and 17.4 (95% CI 14.7-20.3) per 100 admissions, followed by the
medical units 6.1 (95% CI 4.7-7.7) per 1000 patients-days and 4.8, (95% CI 3.8-6.1) per
100 admissions (Table 4).
Medication classes involved in ADEs, PADEs and medication errors with low risk to cause harm
Anticoagulants (21.6%) and antibiotics (20.8%) were the most common medication
classes associated with ADEs. Medication classes most commonly associated with
PADEs were antibiotics (31.3%) followed by anticoagulants (17.3%) and
antihypertensives (9.3%) (Table 5).
Agreement of physician’s reviewers on the classification of the incidents
The kappa value for the presence of ADEs was 0.71; for the presence of
medication errors it was 0.67, and for the presence of potential ADEs it was 0.60. The
kappa value for preventability of ADEs was 0.68 (definitely or probably preventable vs.
definitely or probably not preventable). For the severity of ADEs, the kappa value was
0.74 (fatal vs. significant, serious or life-threatening), 0.63 (life-threatening vs.
significant, serious or fatal), 0.53 (significant vs. serious, life-threatening or fatal), 0.48
(serious vs. life-threatening or fatal) (Table 6).
Factors associated with ADEs
Factors significantly associated with ADEs included age; (OR, 1.012; 95%CI,
1.003 – 1.021) number of medications (OR, 1.062; 95%CI, 1.008 – 1.119), length of
hospital stay (OR, 1.025; 95%CI, 1.015 – 1.035) and admission to the ICUs and
medicine units (OR, 3.276; 95%CI, 2.005 – 5.354) and (OR, 1.736; 95%CI, 1.078 –
Page 13 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
14
2.796), respectively (Table 7). Gender was not significantly associated with ADEs (p =
0.248); therefore, it was not included in the multivariate analysis.
Page 14 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
15
DISCUSSION
In this study we evaluated the incidence of ADEs and found that ADEs were
common with one third caused by medication errors judged to be preventable. Errors
resulting from preventable ADEs were most common in the prescribing stage followed
by the dispensing and administration stages. The majority of the preventable ADEs
were judged to be serious. The incidence of ADEs reported in our study was similar to
that found in previous studies (2, 5). However, a higher incidence was reported in Japan
(4). The differences between our study results and the results from the Japanese study
could be the longer length of hospital stay in Japan and differences in healthcare
systems between countries. The similarity between our findings and the US study could
be because of our use of the same methods for data collection, ADE detection, and
event classification and the similarities in the healthcare systems. The incidence in the
prescribing stage in the current study was higher than those reported in Malaysia
(25.15%) (11), Indonesia (20.4%) (12) and Thailand (1%)(13).
The kappa values reported in our study range from substantial to moderate
according to the measure of the strength of agreement suggested by Landis & Koch
(1977)(14). The lowest level of agreement in the current study was reported for
judgment regarding the severity of the incidents (0.48 and 0.52). However a similar
study reported kappa values lower than those found in our study (0.32 and 0.37) (2).
We included 3985 patients and found 245 ADEs, of which 35% were judged to
be preventable. Gurwitz and colleagues (8) identified 546 ADEs during 2403 nursing
home residence admissions and reported that 51% of the observed ADEs were
Page 15 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
16
preventable. Bates et al (2) determined the incidence of ADEs in 4,031 patients and
found 247 ADEs, of which 28% were deemed preventable. In 2009, Hug et al. (15)
assessed the occurrence of ADEs in 1200 patients from six community hospitals and
identified 180 ADEs, of which 75% were preventable. Recently a multicentre cohort
study of 3,459 patients identified 1010 ADEs and found that 14% of the identified ADEs
were preventable (4).
Regarding PADEs we noticed that only one third of the events were intercepted.
It is noteworthy to highlight that three of the four hospitals had clinical pharmacists
monitoring patient treatments and most of the intercepted PADEs were in those
hospitals.
Our study revealed that ADEs were associated with admission to ICUs and older
age. Consistent with our results, other studies also reported that admission to ICUs (4,
16) and older age (4, 8) were major factors associated with ADEs. In support of this
finding, perhaps especial care should be given to elderly who are admitted to ICUs
because of the added risk of combining two risk factors.
Several important basic medication safety practices are not widely adopted in
most Saudi hospitals (17, 18) . Therefore, there are opportunities for improving the safe
use of medications and preventing ADEs in hospitals in Saudi Arabia. On a national
level, the Saudi Food Drug Authority may lead efforts to prevent ADRs and the Saudi
Medication Safety Centre may lead initiatives to prevent medication errors. For
example, the use of pharmacists to ascertain complete medication histories at
admission and provide discharge counseling reduced the incidence of ADEs (19, 20).
Page 16 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
17
Although reporting is a good tool to identify and prevent ADEs, underreporting is a
common challenge in Saudi hospitals (21).
There is lack of literature about incident medication errors in Southeast Asian
(22) and the Middle Eastern countries (23). Future research could focus on investigating
the causes of ADEs that occur during the medication use process, especially at the
prescribing stage. More research is needed on the causes of ADEs using both
qualitative and quantitative methodologies using standard definitions of events and
severity classification. Using methods similar to the ones used in this study, the benefits
of interventions to prevent ADEs can be estimated and compared to a rigorously
determined baseline.
This study is one of the largest studies investigating the incidence of ADEs in the
Middle East. This study is limited by a lack of hospitals from small towns and rural areas
and that these settings have an even higher incidence of ADEs. Finally, our study
findings are not generalisable to overall Saudi Arabia because the study was conducted
in Riyadh only.
In conclusion, ADEs are common in Saudi hospitals, especially in the ICUs,
causing significant morbidity and mortality. While there are variations in the incidence of
ADEs among countries, there are prospects for preventing them. Interventions that are
effective in other countries should be tested in Saudi Arabia. Such interventions may
include but are not limited to implementations of computerised physician entry (CPOE)
with a clinical decision support system (15), involvement of clinical pharmacists as part
of the medical team during physicians rounds (24-26), medication reconciliation to
Page 17 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
18
obtain accurate medication histories at hospital admission, unit transfers during
hospitalization, and discharge from the hospital (27) and changing the currently
available paper-based system to electronic medical records system (28).
Contribution: HA, DB and MM designed the study. HA wrote the manuscript, MA and
YA contributed in the data analysis and management. All authors contributed to the data
collection process, the study idea and design and approved the final manuscript.
Conflict of Interest: None declared
Data sharing statement: No additional data are available.
Funding: This study was funded by the National Plan for Science and Technology (09-
BIO708-02).
Acknowledgment: We would like to thank research assistants, Emad Zalloum, Trig
Allam, Maram Abuzaid, Umm Hani Sayeda, Shamailah Osmani, Aishah Nor, Nesreen
al-shabr, Sultan Al-Harbi, for their help during the data collection process.
Page 18 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
19
References
1. Bates DW, Spell N, Cullen DJ, et al. The costs of adverse drug events in hospitalized patients.
Adverse Drug Events Prevention Study Group. JAMA. 1997;277(4):307-11.
2. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug
events. Implications for prevention. ADE Prevention Study Group. JAMA. 1995;274(1):29-34.
3. Al Hamid A, Ghaleb M, Aljadhey H, et al. A systematic review of hospitalisation resulting from
medicine related problems in adult patients. Br J Clin Pharmacol . 2013.
4. Morimoto T, Sakuma M, Matsui K, et al. Incidence of Adverse Drug Events and Medication Errors
in Japan: the JADE Study.J Gen Intern Med. 2011;26(2):148-53.
5. Aljadhey H, Mahmoud MA, Mayet A, et al. Incidence of adverse drug events in an academic
hospital: a prospective cohort study. Int J Qual Health Care. 2013;25(6):648-55.
6. Benkirane R, Pariente A, Achour S, et al. Prevalence and preventability of adverse drug events in
a teaching hospital: a cross-sectional study. East Mediterr Health J.. 2009;15(5):1145-55.
7. Stausberg J. International prevalence of adverse drug events in hospitals: an analysis of routine
data from England, Germany, and the USA. BMC health serv res. 2014;14:125.
8. Gurwitz JH, Field TS, Harrold LR, et al. Incidence and preventability of adverse drug events
among older persons in the ambulatory setting. JAMA. 2003;289(9):1107-16.
9. Morimoto T, Gandhi TK, Seger AC,et al. Adverse drug events and medication errors: detection
and classification methods. BMJ Qual Saf . 2004;13(4):306-14.
10. WHO WHO. Definitions [cited 2014]. Available from:
http://www.who.int/medicines/areas/quality_safety/safety_efficacy/trainingcourses/definitions.pdf
11. Abdullah DC, Ibrahim NS, Ibrahim MI. Medication errors among geriatrics at the outpatient
pharmacy in a teaching hospital in Kelantan. Malays J Med Sci. 2004;11:52-8.
12. Ernawati DK, Lee YP, Hughes JD. Nature and frequency of medication errors in a geriatric ward:
an Indonesian experience. The Clin Risk Manag. 2014;10:413-21.
13. Sangtawesin V, Kanjanapattanakul W, Srisan P, et al. Ingchareonsunthorn P. Medication errors
at Queen Sirikit National Institute of Child Health. J Med Assoc Thai. 2003;3:S570–5.
14. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics.
1977;33(1):159-74.
15. Hug BL, Witkowski DJ, Sox CM, et al. Adverse Drug Event Rates in Six Community Hospitals and
the Potential Impact of Computerized Physician Order Entry for Prevention. J Gen Intern Med.
2010;25(1):31-8.
16. Benkirane RR, Abouqal R, Haimeur CC, et al. Incidence of adverse drug events and medication
errors in intensive care units: a prospective multicenter study. J Patient Saf. 2009;5(1):16-22.
17. Aljadhey H, Alhusan A, Alburikan K, et al. Medication safety practices in hospitals: A national
survey in Saudi Arabia. Saudi Pharmaceutical Journal. 2013;21(2):159-64.
18. Alkhani S, Ahmed Y, Bin-Sabbar N, et al. Current practices for labeling medications in hospitals in
Riyadh, Saudi Arabia. Saudi Pharm J . 2013;21(4):345-9.
19. AbuYassin BH, Aljadhey H, Al-Sultan M, et al. Accuracy of the medication history at admission to
hospital in Saudi Arabia. Saudi Pharm J . 2011;19(4):263-7.
20. Al-Ghamdi SA, Mahmoud MA, Alammari MA, et al. The outcome of pharmacist counseling at the
time of hospital discharge: an observational nonrandomized study. Ann Saudi Med. 2012;32(5):492-7.
21. Alshaikh M, Mayet A, Aljadhey H. Medication error reporting in a university teaching hospital in
Saudi Arabia. J Patient Saf. 2013;9(3):145-9.
22. Salmasi S, Khan TM, Hong YH, et al. Medication Errors in the Southeast Asian Countries: A
Systematic Review. PloS one. 2015;10(9):e0136545.
Page 19 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
20
23. Alsulami Z, Conroy S, Choonara I. Medication errors in the Middle East countries: a systematic
review of the literature. Eur JClin Pharmacol. 2013;69(4):995-1008.
24. Kucukarslan SN, Peters M, Mlynarek M, et al. Pharmacists on rounding teams reduce
preventable adverse drug events in hospital general medicine units. Arch Intern Med.
2003;163(17):2014-8.
25. Schnipper JL, Kirwin JL, Cotugno MC, Wahlstrom SA, Brown BA, Tarvin E, et al. Role of
pharmacist counseling in preventing adverse drug events after hospitalization. Arch Intern Med.
2006;166(5):565-71.
26. Leape LL, Cullen DJ, Clapp MD, et al. Pharmacist participation on physician rounds and adverse
drug events in the intensive care unit. JAMA. 1999;282(3):267-70.
27. Tam VC, Knowles SR, Cornish PL, Fine N, Marchesano R, Etchells EE. Frequency, type and clinical
importance of medication history errors at admission to hospital: a systematic review. Can Med Assoc J.
2005;173(5):510515.
28. Liu M, Hinz ERM, Matheny ME, Denny JC, Schildcrout JS, Miller RA, et al. Comparative analysis of
pharmacovigilance methods in the detection of adverse drug reactions using electronic medical records.
J Am Med Inform Assoc. 2013;20(3):420-6.
Page 20 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
21
Tables
Table1. Demographic Characteristics of 3985 patients admitted to four Hospitals in Riyadh Frequency (%) Mean (±SD)
Gender
Male 2102 (52.7) -
Female 1883(47.3) -
Hospital type
Hospital 1(Teaching Hospital) 977(24.5) -
Hospital 2 (Private Hospital) 2033(51.1) -
Hospital 3(Large government Hospital) 683(17.1) -
Hospital 4 (Small government Hospital) 292(7.3) -
Service
Medicine 1352(33.9) -
Surgery 1771(44.5) -
Intensive Care Unit (ICU) 862(21.6) -
Age, years - 43.4 (±19.0)
Length of Hospital Stay, days - 8.1 (±10.2)
Comorbidities (Charlson’s index weight) - 1.1 (±1.4)
Number of Medications - 2.5 (±2.9)
Table 2. Overall Incidence of ADEs, potential ADEs and medication errors with low risk t cause harm
Total number
of incidents
=1531
% Incidence per 100 admissions
(95% CI)
Crude rate per 1000 patient-days (95%
CI)
Medication Errors with low risk to cause harm 609 39.8 15.3(14.1 – 16.5) 19.6 (18.7 – 21.2)
Potential ADEs (PADEs) 677 44.2 16.9 (15.7 – 18.3) 21.8 (20.2 – 23.5)
Intercepted Potential ADEs (N=213) 5.3 (4.6 – 6.1) 6.8(5.9 – 7.8)
Not intercepted Potential ADEs (N=464) 11.6 (10.6 – 12.7) 14.9 (13.6 – 16.3)
ADEs (Harm) 245 16 6.1 (5.4 – 6.9) 7.9 (6.9 – 8.9)
Preventable ADEs (N=85) 2.1(1.7 – 2.6) 2.7 (2.2 – 3.3)
Non-preventable ADEs (N=160) 4.0 (3.4 – 4.6) 5.1 (4.4 – 6.0)
Medication errors with low risk to cause harm include those medication errors with low risk to cause ADEs or PADEs. ADEs, adverse drug events
Table 3. Classification of ADEs by Hospital Type
Units ADEs Length of hospital
stay
ADEs crude rate per 1000 patient- days (95% CI)
Number of admissions
ADEs Incidence per 100 admissions
(95% CI)
Hospital 1
83 9585 8.7 (6.9-10.6) 977 8.5(6.8-10.4)
Hospital 2
53 9032 5.9 (4.4-7.6) 2033 2.6 (2.1-3.3)
Hospital 3
13 6613 2.1(1.1-3.2) 683 2.1(1.1-3.2)
Hospital4
96 5766 16.6(13.5-20.2) 292 32.9(27.7-38.4)
Hospital 1= Teaching hospital, Hospital 2=Private hospital, Hospital 3= small government hospital, Hospital 4=Large government
hospital
Table4. Classification of ADEs incidence by Type of Services
Units ADEs Patient-days, No
ADEs Crude rate per 1000 patient-days (95% CI)
Number of admissions
ADEs Incidence per 100 admissions
(95% CI)
Medical 66 10767 6.1(4.7-7.7) 1352 4.8(3.8-6.1)
Surgical 29 9310 3.1(2.1-4.4) 1771 1.6(1.1-2.3)
ICU 150 10919 13.7(11.6-16.1) 862 17.4(14.7-20.3)
ADEs, adverse drug events
Page 21 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
22
Table 5. Medication classes involved in ADEs, PADEs and medication errors with low risk to cause harm
Medication classes ADEs, n (%)
(n= 245)
PADEs n (%)
(n= 677)
Medication errors with low risk to cause harm n (%)
(n= 609)
Antibiotics 51(20.8) 212(31.3) 193(31.7)
Anticoagulants 53(21.6) 117(17.3) 92(15.1)
Antihypertensives 49(20) 63(9.3) 61(10)
NSAIDs 11(4.5) 41(6.1) 55(9)
GI-medicines 4(1.6) 43(6.4) 39(6.4)
Antidiabetics 5(2) 32(4.7) 26(4.3)
Steroids 14(5.7) 12(1.8) 17(2.8)
Electrolytes 7(2.9) 19(2.8) 6(1)
Cardiovasculars 4(1.6) 15(2.2) 11(1.8)
Dyslipidemic agents 7(2.9) 16(2.4) 6(1)
Analgesics 4(1.6) 9(1.3) 15(2.5)
Antiasthmatics 5(2) 14(2.1) 6(1)
Antituberculosis 3(1.2) 11(1.6) 7(1.1)
Vitamins 0 8(1.2) 5(0.8)
Antifungals 1(0.5) 4(0.6) 6(1)
Antiseizures 5(2) 3(0.4) 2(0.3)
Antipsychotics 1(0.5) 3(0.4) 6(1)
Thyroid agents 0 5(0.7) 3(0.5)
Antivirals 4(1.6) 2(0.3) 1(0.2)
Antihistamines 1(0.5) 3(0.4) 1(0.2)
Sedatives 4(1.6) 1(0.1) 0
Anticancers 0 3(0.4) 2(0.3)
Others 12(4.9) 41(6.1) 49(8)
Page 22 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
23
Table 6. Interrator reliability of the incident type and their severity and preventability
Severity Agreement % Kappa value
Exclude vs ADEs, PADE or medication error 56.4 0.63
ADEs vs PADE, medication error or exclude 93.6 0.71
PADEs vs ADE, medication error or exclude 70.9 0.62
Medication error vs ADEs, PADE or exclude 93.7 0.67
Preventable vs non-preventable ADEs 92.2 0.68
Fatal vs Life-threatening, serious or significant 100 0.74
Life-threatening vs fatal serious or significant 100 0.63
Serious vs fatal, life-threatening or significant 60.5 0.48
Significant vs fatal, life-threatening or serious 84.2 0.52
Table 7. Factors Associated with ADEs
Factor Unadjusted Odds Ratio
95% CI P value Adjusted Odds Ratio
95% CI P value
Lower Upper Lower Upper
Age 1.024 1.017 1.032 <0.001 1.012 1.003 1.021 0.009
Number of medications 1.193 1.148 1.240 <0.001 1.062 1.008 1.119 0.023
Charlson’s comorbidity index weight
1.251 1.158 1.352 <0.001 1.041 0.937 1.157 0.452
Length of hospital stay 1.042 1.034 1.051 <0.001 1.025 1.015 1.035 <0.001
Gender (Male)a 0.848 0.642 1.121 0.248 - - - -
ICUb 7.786 5.259 11.528 <0.001 3.276 2.005 5.354 <0.001
Medicineb 2.539 1.664 3.872 <0.001 1.736 1.078 2.796 0.023
CI, Confidence Interval, Reference categories: Femalea, Reference categories: Surgery
b
Page 23 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
24
Appendix A.
Examples of PADEs
Prescribing
The physician ordered lisinopril, 5 mg tablet, once daily even though the patient was not
hypertensive and had no indication for the drug. The order was intended for another patient and
the nurse intercepted the error.
Transcribing
A 74-year old man with diabetes mellitus, hypertension and recurrent urinary tract infections
was admitted to the medical ward. Meropenem, 500 mg intravenously, every eight hours was
ordered. In the morning round, the infectious disease consultant verbally asked the intern to
change meropenem to imipenem 500 mg intravenously every six hours. However, the intern
mistakenly transcribed it as meropenem. This error was caught, corrected, and noted as an
error in the patient’s medical record.
Dispensing
An order of metoclopramide 10 mg was sent to the pharmacy. The nurse obtained the drug from
the pharmacy, but from the appearance of the solution, she suspected that the preparation was
not metoclopramide. The nurse contacted the pharmacy and the pharmacist found that it was
the wrong medication, although the label was stated that it was metoclopramide.
Administering
A nurse handled two capsules for two different patients in Room #8 and Room #9. She almost
accidentally gave the wrong medication (switched) to each patient. However, the patient in
Page 24 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
25
Room # 9 knew her medication and she said, “This is not my medicine,” and the error was
intercepted by the patient.
Page 25 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
26
Figure 1. Study flow chart
Study included 3,985 patients admitted to four
hospitals over four months
Incidents identified by pharmacists
1676
Incidents rejected by reviewers
145
Incidents accepted by reviewers
1531
Medication errors 1286
Adverse drug events (ADEs)
245
Non-preventable ADEs 160
Preventable ADEs
85
Risk to cause harm (Potential ADEs)
677
Low risk to cause harm
609
Non-intercepted 464
Intercepted 213
Page 26 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
STROBE 2007 (v4) Statement—Checklist of items that should be included in reports of cohort studies
Section/Topic Item
# Recommendation Reported on page #
Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 6
(b) Provide in the abstract an informative and balanced summary of what was done and what was found 2
Introduction
Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 5
Objectives 3 State specific objectives, including any prespecified hypotheses 5
Methods
Study design 4 Present key elements of study design early in the paper 6
Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data
collection
6
Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up 6
(b) For matched studies, give matching criteria and number of exposed and unexposed
Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if
applicable
6
Data sources/
measurement
8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe
comparability of assessment methods if there is more than one group
7
Bias 9 Describe any efforts to address potential sources of bias
Study size 10 Explain how the study size was arrived at 6
Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and
why
8
Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 8
(b) Describe any methods used to examine subgroups and interactions 8
(c) Explain how missing data were addressed
(d) If applicable, explain how loss to follow-up was addressed
(e) Describe any sensitivity analyses
Results
Page 27 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on December 29, 2020 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. Downloaded from
For peer review only
Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, confirmed
eligible, included in the study, completing follow-up, and analysed
10
(b) Give reasons for non-participation at each stage 10
(c) Consider use of a flow diagram 24
Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential
confounders
10
(b) Indicate number of participants with missing data for each variable of interest
(c) Summarise follow-up time (eg, average and total amount) 6
Outcome data 15* Report numbers of outcome events or summary measures over time 10
Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (eg, 95% confidence
interval). Make clear which confounders were adjusted for and why they were included
12
(b) Report category boundaries when continuous variables were categorized
(c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period
Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses
Discussion
Key results 18 Summarise key results with reference to study objectives 14
Limitations
Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from
similar studies, and other relevant evidence
15 -16
Generalisability 21 Discuss the generalisability (external validity) of the study results 16
Other information
Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on
which the present article is based
17
*Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies.
Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE
checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at
http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org.
Page 28 of 28
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on December 29, 2020 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. Downloaded from
For peer review only
Incidence of Adverse Drug Events in Public and Private Hospitals in Riyadh, Saudi Arabia: The (ADESA) Prospective
Cohort Study
Journal: BMJ Open
Manuscript ID bmjopen-2015-010831.R2
Article Type: Research
Date Submitted by the Author: 01-Jun-2016
Complete List of Authors: Aljadhey, Hisham; College of Pharmacy, King Saud University, Medication Safety Research Chair Mahmoud, Mansour
Ahmed, Yusuf; King Saud University, Medication Safety Research Chair Sultana, Razia; Specialized Medical Center Hospital Zouein, Salah; Specialized Medical Center Hospital Alshanawani, Sulafa ; King Saud Medical City Mayet, Ahmed; King Khaled University Hospital Alshaikh, Mashael; King Khaled University Hospital Kalagi, Nora; King Saud University, Medication Safety Research Chair altawil, esra; King Khaled University Hospital El Kinge, Abdul Rahman ; Specialized Medical Center Hospital Arwadi, Abdulmajid; Specialized Medical Center Hospital Alyahya, Maha; King Salman Hospital Murray, Michael; Purdue University and Regenstrief Institute
Bates, David; Division of General Medicine
<b>Primary Subject Heading</b>:
Epidemiology
Secondary Subject Heading: Pharmacology and therapeutics, General practice / Family practice, Intensive care, Research methods, Surgery
Keywords: Adverse Drug Events (ADEs), prospective cohort study, Saudi Arabia
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open on D
ecember 29, 2020 by guest. P
rotected by copyright.http://bm
jopen.bmj.com
/B
MJ O
pen: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. D
ownloaded from
For peer review only
1
Incidence of Adverse Drug Events in Public and Private Hospitals in Riyadh, Saudi Arabia: The (ADESA) Prospective Cohort Study
Hisham Aljadhey1, Mansour A Mahmoud1, Yusuf Ahmed1, Razia Sultana2, Salah Zouein2, Sulafah Alshanawani3, Ahmed Mayet4, Mashael K Alshaikh4, Nora Kalagi1,
Esraa Al Tawil4, Abdul Rahman El Kinge2, Abdulmajid Arwadi2, Maha Alyahya5, Michael D Murray6, David Bates7
1 Medication Safety Research Chair, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia
2Specialized Medical Center, Riyadh, Saudi Arabia
3King Saud Medical City, Riyadh, Saudi Arabia
4King Khaled University Hospital, Riyadh, Saudi Arabia
5King Salman Hospital, Riyadh, Saudi Arabia
6Purdue University and Regenstrief Institute, Indianapolis, United States
7 Harvard Medical School and Brigham and Women's Hospital, Boston, United States.
Number of words
Text: 3059
Abstract: 299
Corresponding author: Hisham Aljadhey, Pharm D, PhD Director of Medication Safety Research Chair Dean, College of Pharmacy, King Saud University Supervisor of Pharmacy Services at King Saud University Medical CityP.O.Box 2475, Riyadh 11451, Saudi Arabia Email: [email protected] Phone: +966 530039008
Page 1 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
2
ABSRACT
Objectives: To determine the incidence of ADEs and assess their severity and
preventability in four Saudi hospitals.
Design: Prospective cohort study
Setting: The study included patients admitted to medical, surgical and intensive care
units of four hospitals in Saudi Arabia. These hospitals include a 900 bed tertiary
teaching hospital, a 400 bed private hospital, a 1400 bed large government hospital and
a 350 bed small government hospital.
Participants: All patients (≥12 years) admitted to the study units during four month.
Primary and secondary outcome measures: Incidents were collected by pharmacists
and reviewed by independent clinicians. Reviewers classified the identified incidents as
ADEs, potential ADEs (PADEs) and medication errors and then determined their
severity and preventability.
Results: We followed 4041 patients from admission to discharge. Of those, 3985
patients had complete data for analysis. The mean age of patients in the analysed
cohort was 43 (±19.5) years. A total of 1676 incidents of ADEs were identified by
pharmacists during the medical chart review. Clinician reviewers accepted 1531(91.4%)
of the incidents identified by the pharmacists (245 ADEs, 677 PADEs and 609
medication errors with low risk to cause harm). The incidence of ADEs was 6.1 (95% CI,
5.4-6.9) per 100 admissions and 7.9 (95% CI, 6.9 – 8.9) per 1000 patient-days. The
occurrence of ADEs was most common in the intensive care units 149 (60.8%) followed
Page 2 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
3
by medical 67(27.3%) and surgical units 29(11.8%). In terms of severity, 129 (52.7%) of
the ADEs were significant, 91 (37.1%) were serious, 22 (9%) were life-threatening and
three (1.2%) were fatal.
Conclusions: We found that ADEs were common in Saudi hospitals, especially in the
ICUs, causing significant morbidity and mortality. Future studies should focus on
investigating the root causes of ADEs at the prescribing stage and development and
testing of interventions to minimise harm from medications.
Key words: adverse drug events (ADEs), prospective cohort study, hospitals, Saudi
Arabia
Page 3 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
4
Strength and limitations
• This study is one of the largest studies investigating the incidence of ADEs in the
Middle East.
• This study is limited by a lack of hospitals from small towns and rural areas and
that these settings have an even higher incidence of ADEs.
• Our study findings are not generalisable to overall Saudi Arabia because the
study was conducted in Riyadh only.
Page 4 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
5
INTRODUCTION
Adverse drug events (ADEs) are major cause of morbidity, mortality, and
increased healthcare costs and hospitalization (1-3). An ADE is defined as an injury
caused by a medication (4). They are largely preventable and occur mostly at the
prescribing stage of the medication use process (2, 4, 5). The incidence of ADEs
reported in the literature varies significantly between countries largely because of the
differences in available drug products, practices, training, study methodology and
patient safety initiatives among countries. Early in 1995 Bates et al from the United
States (U.S) reported an incidence of 6.5 per 100 admissions (2); however, while using
the same methods a study in Japan reported an incidence of 17 per 100 admissions (4)
suggesting real differences between these two countries. In Saudi Arabia, a single
hospital study reported an incidence of 8.5 per 100 admissions (5) and a cross-sectional
study in Morocco reported an incidence of 4.2 per 100 admissions (6). A recent
international study using hospital datasets estimated the prevalence of ADEs to be
3.2% in England, 4.8% in Germany and 5.6% in the U.S (7). It is important to mention
that the incidence of preventable ADEs was estimated by a population-based study to
be 13.8 per 1000 person-years (8).
Despite the evidence that ADEs are common and could be life threatening, little
attempt has been made in Saudi Arabia to detect and estimate the incidence of ADEs in
hospitalised patients. Such a paucity of research hinders the development of prevention
strategies to improve patient safety. To date, one prospective chart review study has
been conducted in a single teaching hospital in the Saudi setting (5). Therefore we
sought to estimate incident ADEs with a larger patient sample from different hospitals
Page 5 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
6
with diverse settings with varying practices and strategies for managing patients.
Therefore, the objective of our study was to estimate the incidence and risk factors
associated with ADEs in Saudi hospitals and determine their severity and preventability.
Page 6 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
7
METHODS
Study design and setting
The Adverse Drug Events in Saudi Arabia (ADESA) project was a four month
prospective cohort study involving four hospitals with diverse settings. These hospitals
included a 900 bed tertiary teaching hospital, a 400 bed private hospital, a 1400 bed
large government hospital and a 350-bed small government hospital. We randomly
selected medical, surgical and intensive care units (ICUs) from these hospitals and
excluded obstetrics and pediatric units because of the lower frequency of use of
medications within these units. We included patients older than 12 years of age
admitted for more than 24 hours during the four-month study period. None of the
hospitals had electronic medical records or decision support systems. Instead, hospitals
utilized paper-based systems where physician notes (including prescribed medications)
and nursing notes (including daily administered medications) were handwritten and kept
in patient charts. Medication orders were sent to the inpatient pharmacies and
dispensed using unit dose systems.
Definitions
Each incident was defined as an ADE (preventable and non-preventable), PADE that
were classified as either intercepted or non-intercepted or a medication errors with low
risk to cause harm. We defined ADE as any injury resulting from medical interventions
related to a drug and includes both ADR in which no error occurred and complications
resulting from medication errors (2, 9, 10). Non-preventable ADEs, also known as
adverse drug reactions (ADRs), are defined by the World Health Organization as “a
Page 7 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
8
response to a drug which is noxious and unintended, and which occurs at doses used in
man for prophylaxis, diagnosis, or therapy of disease, or for the modification of
physiological function”(11). A non-preventable ADE is an injury with no error in the
medication process. An example of this would be an allergic reaction in a patient not
previously known to be allergic to that particular medication. Preventable ADEs were
those that result from medication errors at any stage of the medication use process (2).
An example of this would be an anaphylactic reaction to an antibiotic that the patient is
known to be allergic to. Preventability was further classified into definitely
preventable/non-preventable and probably preventable/non-preventable. A PADE was
an error that carries a risk of causing injury related to the use of a medication but harm
did not occur, either because of specific circumstances or because the error was
intercepted (9). Intercepted PADEs were those that had the potential to cause injury but
did not reach the patient because they were intercepted by someone during the
medication use process; Non-intercepted potential ADEs were those with the potential
to cause harm but failed to do so after the medication reached the patient (9).
Medication errors with a low risk to cause harm included those with minimal risk to
cause ADEs or PADEs. Comorbidities were determined using Charlson's Comorbidity
Index, which is a method of classifying comorbidities of patients according to the
International Classification of Disease (ICD). Each comorbidity class has an associated
weight of 1, 2, 3 or 6. The sum of all weights results in a single comorbidity score for
each patient with higher scores predictive of adverse outcomes such as mortality or
high resource use.
Data collection and classification of incidents
Page 8 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
9
Data were collected as described in details elsewhere (5). Briefly, trained clinical
pharmacists collected data each day during the study period. In addition, all nurses working in
the particular units were invited to attend monthly in-service presentations about the study to
increase their awareness about ADEs reporting. The pharmacists reviewed patients’ medical
charts of all admitted patients in each of the participating units to report demographic
characteristics of patients, comorbidity and the number of medications. When incidents were
noted the pharmacists wrote a detailed description of each incident and captured the relevant
patient characteristics and event history.
Two independent clinicians who were not involved in the data collection process, were
provided with a study manual that contained study terminology and a guide on the assessment
of the severity and preventability of an incident. The manual included examples of incidents and
their severity classifications. The severity of the incidents was categorized as significant,
serious, life threatening or fatal using a methodology developed by the Brigham and Women’s
Hospital’s Center for Patient Safety Research and Practice (2). The study manual served as a
guide for the reviewers to independently review the incidents and decide on inclusion of
incidents and further classify them as ADEs, PADEs or medication errors with low risk to cause
harm. They were then able to assess severity and preventability. Preventability categories were
defined as follows: definitely preventable, probably preventable, definitely not preventable or
probably not preventable (2). In the event that there was disagreement on the classification of
the incidents, the clinicians called for a meeting to decide whether to include or exclude the
incidents. The primary outcomes of this study were incidence of ADEs, PADEs and medication
errors with low risk to cause harm, as defined previously. The secondary outcomes were the
severity of events, their preventability, and associated risk factors. The research and ethics
committees of the four hospitals approved this study.
Data Analysis:
Page 9 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
10
We calculated the overall incidence per 100 admissions and crude rate per 1000 patient-
days with 95% confidence intervals (CI). In addition the incidence was calculated by hospital
and by unit type. Continuous variables are presented as mean ± standard deviation (SD) and
categorical variables as number and percentage. Inter-rater reliability was assessed using the
kappa statistic for assessment of the presence of an ADE and its preventability and severity. To
evaluate the univariate association of potential risk factors with ADEs, we used univariate
logistic regression. The variables included in the univariate analysis were age, gender,
Charlson’s comorbidity index weight, length of hospital stay, number of medications, and service
type. Variables found to be statistically significant (P < 0.05) in the univariate analysis were
included in the multivariate logistic regression final model. Statistical analyses were conducted
using the Statistical Package for Social Science (SPSS) software IBM SPSS Statistics for
Windows, Version 22.0. Armonk, NY: IBM Corp.
Page 10 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
11
RESULTS
Demographic Characteristics of the patients
Clinical pharmacists reviewed the medical charts of 4,041 patients. Complete
data of 3,985 patients were analysed (Table 1). The total length of hospital stay for
patients was 30,996 days. The study was conducted in four hospitals in Riyadh, Saudi
Arabia (977 patients from a teaching hospital; 2033 patients from a private hospital, 683
patients from large government hospital, and 292 patients from a small government
hospital). Male patients were a slight majority (52.7%) (Table1). The patients were
admitted to one of the three services (Medicine, 1352 patients; surgery, 1771 patients;
and Intensive Care Units, 862 patients). The mean length of the hospital stay was 8.1 ±
10.2 days and the mean age of the patients was 43.4 ± 19.0 years (Table 1).
Incidents review and classification
The pharmacists’ chart review s in the four hospitals identified 1676 cases of
ADEs, PADEs and medication errors. Physician reviewed and accepted 1531 (91.3%)
of the cases, which were classified as 609 (39.8%) medication errors with low risk of
harm, 677 (44.2%) PADEs, and 245 (16%) cases of ADEs (Figure 1). Among the ADEs
85 (34.7%) were deemed preventable and 160 (65.3%) were judged to be non-
preventable (Table 2). The majority of the preventable ADEs occurred in the prescribing
stage (75 ; 88.2%) followed by administering stage (7 ;8.2%), dispensing stage (2 ;2.4%
and monitoring stage (1; ;1.2%). One hundred twenty nine (52.7%) of the ADEs were
significant, 91 (37.1%) were serious, 22 (9%) were life-threatening, and three (1.2%)
were fatal. Of the 85 preventable ADEs, 36 (42.4%) were significant, 38 (44.7%) were
serious, 10 (11.9%) were life-threatening and one (1.2%) was fatal. Among PADEs, 213
Page 11 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
12
(31.9%) were intercepted by the medical staff. Regarding severity of PADEs, 383
(56.6%) were significant, 271 (40%) were serious and 23 (3.4%) were life-threatening.
Overall incidence of ADEs and medication errors with low risk to cause harm
The incidence of ADEs per 100 admissions was 6.1(95% CI 5.4 – 6.9) and the
incidence of potential ADEs was 16.9 (95% CI 15.7 – 18.3) per 100 admissions (Table
2). The incidence of medication errors with low risk to cause harm was 15.3 (95% CI
14.1 – 16.5) per 100 admissions and the incidence of preventable ADEs was 2.1(95%
CI 1.7 – 2.6) per 100 admissions. The incidence of non-preventable ADEs was 4.0
(95% CI 3.4 – 4.6) per 100 admissions and the incidence of intercepted potential ADEs
was 5.3 (95% CI 4.6 – 6.1) per 100 admissions (Table 2). Incidents of preventable
ADEs, PADEs and medication errors with low risk to cause harm most commonly
occurred in the prescribing stage 1288 (84.1%) followed by dispensing stage 69 (4.5%)
and the administering stage (43; 2.8%). Table 3 shows the distribution of incidents
among the four hospitals. Examples of PADEs at different stages of the medication use
process are listed in Appendix A. The incidence of ADEs was higher in the large
government hospital 16.6 (95% CI 13.5-20.2) per 1000 patient-days and 32.9(95% CI
27.7-38.4) per 100 admissions followed by the teaching hospital 8.7 (95% CI 6.9-10.6)
per 1000 patient-days and 8.5(95% CI 6.8-10.4) per 100 admissions (Table 3).
The incidence of PADEs was predominantly higher in the private hospital (367;
23.9%). Medication errors were mostly seen in the private hospital (367; 23.9%).
Classification of ADEs Incidents by service type
Page 12 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
13
The incidence of ADEs was higher in the ICUs at 13.7 (95% CI 11.6-16.1) per
1000 patients-days and 17.4 (95% CI 14.7-20.3) per 100 admissions, followed by the
medical units 6.1 (95% CI 4.7-7.7) per 1000 patients-days and 4.8,(95% CI 3.8-6.1) per
100 admissions (Table 4).
Medication classes involved in ADEs, PADEs and medication errors with low risk to cause harm
Anticoagulants (21.6%) and antibiotics (20.8%) were the most common medication
classes associated with ADEs. Medication classes most commonly associated with
PADEs were antibiotics (31.3%) followed by anticoagulants (17.3%) and
antihypertensives (9.3%) (Table 5).
Agreement of physician’s reviewers on the classification of the incidents
The kappa value for the presence of ADEs was 0.71; for the presence of
medication errors it was 0.67, and for the presence of potential ADEs it was 0.60. The
kappa value for preventability of ADEs was 0.68 (definitely or probably preventable vs.
definitely or probably not preventable). For the severity of ADEs, the kappa value was
0.74 (fatal vs. significant, serious or life-threatening), 0.63 (life-threatening vs.
significant, serious or fatal), 0.53 (significant vs. serious, life-threatening or fatal), 0.48
(serious vs. life-threatening or fatal) (Table 6).
Factors associated with ADEs
Factors significantly associated with ADEs included age; (OR, 1.012; 95%CI,
1.003 – 1.021) number of medications (OR, 1.062; 95%CI, 1.008 – 1.119), length of
hospital stay (OR, 1.025; 95%CI, 1.015 – 1.035) and admission to the ICUs and
medicine units (OR, 3.276; 95%CI, 2.005 – 5.354) and (OR, 1.736; 95%CI, 1.078 –
Page 13 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
14
2.796), respectively (Table 7). Gender was not significantly associated with ADEs (p =
0.248); therefore, it was not included in the multivariate analysis.
Page 14 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
15
DISCUSSION
In this study we evaluated the incidence of ADEs and found that ADEs were
common with one third caused by medication errors judged to be preventable. Errors
resulting from preventable ADEs were most common in the prescribing stage followed
by the dispensing and administration stages. The majority of the preventable ADEs
were judged to be serious. The incidence of ADEs reported in our study was similar to
that found in previous studies (2, 5). However, a higher incidence was reported in Japan
(4). The differences between our study results and the results from the Japanese study
could be the longer length of hospital stay in Japan and differences in healthcare
systems between countries. The similarity between our findings and the US study could
be because of our use of the same methods for data collection, ADE detection, and
event classification and the similarities in the healthcare systems. The incidence in the
prescribing stage in the current study was higher than those reported in Malaysia
(25.15%) (12), Indonesia (20.4%) (13) and Thailand (1%)(14).
The kappa values reported in our study range from substantial to moderate
according to the measure of the strength of agreement suggested by Landis & Koch
(1977)(15). The lowest level of agreement in the current study was reported for
judgment regarding the severity of the incidents (0.48 and 0.52). However a similar
study reported kappa values lower than those found in our study (0.32 and 0.37) (2).
We included 3985 patients and found 245 ADEs, of which 35% were judged to
be preventable. Gurwitz and colleagues (8) identified 546 ADEs during 2403 nursing
Page 15 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
16
home residence admissions and reported that 51% of the observed ADEs were
preventable. Bates et al (2) determined the incidence of ADEs in 4,031 patients and
found 247 ADEs, of which 28% were deemed preventable. In 2009, Hug et al. (16)
assessed the occurrence of ADEs in 1200 patients from six community hospitals and
identified 180 ADEs, of which 75% were preventable. Recently a multicentre cohort
study of 3,459 patients identified 1010 ADEs and found that 14% of the identified ADEs
were preventable (4).
Regarding PADEs we noticed that only one third of the events were intercepted.
It is noteworthy to highlight that three of the four hospitals had clinical pharmacists
monitoring patient treatments and most of the intercepted PADEs were in those
hospitals.
Our study revealed that ADEs were associated with admission to ICUs and older
age. Consistent with our results, other studies also reported that admission to ICUs (4,
17) and older age (4, 8) were major factors associated with ADEs. In support of this
finding, perhaps especial care should be given to elderly who are admitted to ICUs
because of the added risk of combining two risk factors.
Several important basic medication safety practices are not widely adopted in
most Saudi hospitals (18, 19) . Therefore, there are opportunities for improving the safe
use of medications and preventing ADEs in hospitals in Saudi Arabia. On a national
level, the Saudi Food Drug Authority may lead efforts to prevent ADRs and the Saudi
Medication Safety Centre may lead initiatives to prevent medication errors. For
example, the use of pharmacists to ascertain complete medication histories at
Page 16 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
17
admission and provide discharge counseling reduced the incidence of ADEs (20, 21).
Although reporting is a good tool to identify and prevent ADEs, underreporting is a
common challenge in Saudi hospitals (22)
There is lack of literature about incident medication errors in Southeast Asian
(23) and the Middle Eastern countries (24). Future research could focus on investigating
the causes of ADEs that occur during the medication use process, especially at the
prescribing stage. More research is needed on the causes of ADEs using both
qualitative and quantitative methodologies using standard definitions of events and
severity classification. Using methods similar to the ones used in this study, the benefits
of interventions to prevent ADEs can be estimated and compared to a rigorously
determined baseline.
This study is limited by a lack of hospitals from small towns and rural areas and it
that these settings have an even higher incidence of ADEs. Finally, our study findings
are not generalisable to overall Saudi Arabia because the study was conducted in
Riyadh only.
In conclusion, ADEs are common in Saudi hospitals, especially in the ICUs,
causing significant morbidity and mortality. While there are variations in the incidence of
ADEs among countries, there are prospects for preventing them. Interventions that are
effective in other countries should be tested in Saudi Arabia. Such interventions may
include but are not limited to implementations of computerised physician entry (CPOE)
with a clinical decision support system (16), involvement of clinical pharmacists as part
of the medical team during physicians rounds (25-27), medication reconciliation to
Page 17 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
18
obtain accurate medication histories at hospital admission, unit transfers during
hospitalization, and discharge from the hospital (28) and changing the currently
available paper-based system to electronic medical records system (29).
Contribution: HA, DB and MDM designed the study. HA and MAM wrote the
manuscript, MAM and YA contributed in the data analysis and management. All authors
contributed to the data collection process, the study idea and design and approved the final
manuscript.
Conflict of Interest: None declared
Data sharing statement: No additional data are available.
Funding: This work was supported by the National Plan for Science and Technology
(09-BIO708-02).
Acknowledgment: We would like to thank research assistants, Emad Zalloum, Trig
Allam, Maram Abuzaid, Umm Hani Sayeda, Shamailah Osmani, Aishah Nor, Nesreen
al-shabr, Sultan Al-Harbi, for their help during the data collection process.
Page 18 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
19
References
1. Bates DW, Spell N, Cullen DJ, et al. The costs of adverse drug events in hospitalized patients.
Adverse Drug Events Prevention Study Group. JAMA 1997;277:307-11.
2. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug
events. Implications for prevention. ADE Prevention Study Group. JAMA 1995;274:29-34.
3. Al Hamid A, Ghaleb M, Aljadhey H, et al. A systematic review of hospitalisation resulting from
medicine related problems in adult patients. Br J Clin Pharmacol 2013;78:202-17.
4. Morimoto T, Sakuma M, Matsui K, et al. Incidence of adverse drug events and medication errors
in Japan: the JADE study. J Gen Intern Med 2011;26:148-53.
5. Aljadhey H, Mahmoud MA, Mayet A, et al. Incidence of adverse drug events in an academic
hospital: a prospective cohort study. Int J Qual Health C 2013;25:648-55.
6. Benkirane R, Pariente A, Achour S, et al. Prevalence and preventability of adverse drug events in
a teaching hospital: a cross-sectional study. East Mediterr Health J 2009;15:1145-55.
7. Stausberg J. International prevalence of adverse drug events in hospitals: an analysis of routine
data from England, Germany, and the USA. BMC Health Serv Res 2014;14:125.
8. Gurwitz JH, Field TS, Harrold LR, et al. Incidence and preventability of adverse drug events
among older persons in the ambulatory setting. JAMA 2003;289:1107-16.
9. Morimoto T, Gandhi TK, Seger AC, et al. Adverse drug events and medication errors: detection
and classification methods. BMJ Qual Saf 2004;13:306-14.
10. Bates DW, Boyle DL, Vliet MVV, et al. Relationship between Medication Errors and Adverse Drug
Events. J Gen Intern Med 1995;10:199-205.
11. WHO. WHO Definitions [cited 2014]. Available from:
http://www.who.int/medicines/areas/quality_safety/safety_efficacy/trainingcourses/definitions.pdf
12. Abdullah DC, Ibrahim NS, Ibrahim MI. Medication errors among geriatrics at the outpatient
pharmacy in a teaching hospital in Kelantan. Malays J Med Sci 2004;11:52-8.
13. Ernawati DK, Lee YP, Hughes JD. Nature and frequency of medication errors in a geriatric ward:
an Indonesian experience. Therapeutics and clinical risk management 2014;10:413-21.
14. Sangtawesin V, Kanjanapattanakul W, Srisan P, et al. Medication errors at Queen Sirikit National
Institute of Child Health. J Med Assoc Thai 2003;3:S570–5.
15. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics.
1977;33:159-74.
16. Hug BL, Witkowski DJ, Sox CM, et al. Adverse Drug Event Rates in Six Community Hospitals and
the Potential Impact of Computerized Physician Order Entry for Prevention. J Gen Intern Med
2010;25:31-8.
17. Benkirane RR, Abouqal R, Haimeur CC, et al. Incidence of adverse drug events and medication
errors in intensive care units: a prospective multicenter study. J Patient Saf 2009;5:16-22.
18. Aljadhey H, Alhusan A, Alburikan K, et al. Medication safety practices in hospitals: A national
survey in Saudi Arabia. Saudi Pharm J 2013;21:159-64.
19. Alkhani S, Ahmed Y, Bin-Sabbar N, et al. Current practices for labeling medications in hospitals in
Riyadh, Saudi Arabia. Saudi Pharm J 2013;21:345-9.
20. Abuyassin BH, Aljadhey H, Al-Sultan M, et al. Accuracy of the medication history at admission to
hospital in Saudi Arabia. Saudi Pharm J 2011;19:263-7.
21. Al-Ghamdi SA, Mahmoud MA, Alammari MA, et al. The outcome of pharmacist counseling at the
time of hospital discharge: an observational nonrandomized study. Ann Saudi Med 2012;32:492-7.
22. Alshaikh M, Mayet A, Aljadhey H. Medication error reporting in a university teaching hospital in
Saudi Arabia. J Patient Saf 2013;9:145-9.
Page 19 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
20
23. Salmasi S, Khan TM, Hong YH, et al. Medication Errors in the Southeast Asian Countries: A
Systematic Review. PloS one 2015;10:e0136545.
24. Alsulami Z, Conroy S, Choonara I. Medication errors in the Middle East countries: a systematic
review of the literature. Eur J Clin Pharmacol 2013;69:995-1008.
25. Kucukarslan SN, Peters M, Mlynarek M, et al. Pharmacists on rounding teams reduce
preventable adverse drug events in hospital general medicine units. Arch Intern Med 2003;163:2014-8.
26. Schnipper JL, Kirwin JL, Cotugno MC, et al. Role of pharmacist counseling in preventing adverse
drug events after hospitalization. Arch Intern Med 2006;166:565-71.
27. Leape LL, Cullen DJ, Clapp MD, et al. Pharmacist participation on physician rounds and adverse
drug events in the intensive care unit. JAMA 1999;282:267-70.
28. Tam VC, Knowles SR, Cornish PL, Fine N, Marchesano R, Etchells EE. Frequency, type and clinical
importance of medication history errors at admission to hospital: a systematic review. CMAJ
2005;173:510515.
29. Liu M, Hinz ERM, Matheny ME, Denny JC, Schildcrout JS, Miller RA, et al. Comparative analysis of
pharmacovigilance methods in the detection of adverse drug reactions using electronic medical records.
JAMIA 2013;20:420-6.
Page 20 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
21
Tables
Table1. Demographic Characteristics of 3985 patients admitted to four Hospitals in Riyadh Frequency (%) Mean (±SD)
Gender
Male 2102 (52.7) -
Female 1883(47.3) -
Hospital type
Hospital 1(Teaching Hospital) 977(24.5) -
Hospital 2 (Private Hospital) 2033(51.1) -
Hospital 3(Large government Hospital) 683(17.1) -
Hospital 4 (Small government Hospital) 292(7.3) -
Service
Medicine 1352(33.9) -
Surgery 1771(44.5) -
Intensive Care Unit (ICU) 862(21.6) -
Age, years - 43.4 (±19.0)
Length of Hospital Stay, days - 8.1 (±10.2)
Comorbidities (Charlson’s index weight) - 1.1 (±1.4)
Number of Medications - 2.5 (±2.9)
Table 2. Overall Incidence of ADEs, potential ADEs and medication errors with low risk t cause harm
Total number
of incidents
=1531
% Incidence per 100 admissions
(95% CI)
Crude rate per 1000 patient-days (95%
CI)
Medication Errors with low risk to cause harm 609 39.8 15.3(14.1 – 16.5) 19.6 (18.7 – 21.2)
Potential ADEs (PADEs) 677 44.2 16.9 (15.7 – 18.3) 21.8 (20.2 – 23.5)
Intercepted Potential ADEs (N=213) 5.3 (4.6 – 6.1) 6.8(5.9 – 7.8)
Not intercepted Potential ADEs (N=464) 11.6 (10.6 – 12.7) 14.9 (13.6 – 16.3)
ADEs (Harm) 245 16 6.1 (5.4 – 6.9) 7.9 (6.9 – 8.9)
Preventable ADEs (N=85) 2.1(1.7 – 2.6) 2.7 (2.2 – 3.3)
Non-preventable ADEs (N=160) 4.0 (3.4 – 4.6) 5.1 (4.4 – 6.0)
Medication errors with low risk to cause harm include those medication errors with low risk to cause ADEs or PADEs. ADEs, adverse drug events
Table 3. Classification of ADEs by Hospital Type
Units ADEs Length of hospital
stay
ADEs crude rate per 1000 patient- days (95% CI)
Number of admissions
ADEs Incidence per 100 admissions
(95% CI)
Hospital 1
83 9585 8.7 (6.9-10.6) 977 8.5(6.8-10.4)
Hospital 2
53 9032 5.9 (4.4-7.6) 2033 2.6 (2.1-3.3)
Hospital 3
13 6613 2.1(1.1-3.2) 683 2.1(1.1-3.2)
Hospital4
96 5766 16.6(13.5-20.2) 292 32.9(27.7-38.4)
Hospital 1= Teaching hospital, Hospital 2=Private hospital, Hospital 3= small government hospital, Hospital 4=Large government
hospital
Table4. Classification of ADEs incidence by Type of Services
Units ADEs Patient-days, No
ADEs Crude rate per 1000 patient-days (95% CI)
Number of admissions
ADEs Incidence per 100 admissions
(95% CI)
Medical 66 10767 6.1(4.7-7.7) 1352 4.8(3.8-6.1)
Surgical 29 9310 3.1(2.1-4.4) 1771 1.6(1.1-2.3)
ICU 150 10919 13.7(11.6-16.1) 862 17.4(14.7-20.3)
ADEs, adverse drug events
Page 21 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
22
Table 5. Medication classes involved in ADEs, PADEs and medication errors with low risk to cause harm
Medication classes ADEs, n (%)
(n= 245)
PADEs n (%)
(n= 677)
Medication errors with low risk to cause harm n (%)
(n= 609)
Antibiotics 51(20.8) 212(31.3) 193(31.7)
Anticoagulants 53(21.6) 117(17.3) 92(15.1)
Antihypertensives 49(20) 63(9.3) 61(10)
NSAIDs 11(4.5) 41(6.1) 55(9)
GI-medicines 4(1.6) 43(6.4) 39(6.4)
Antidiabetics 5(2) 32(4.7) 26(4.3)
Steroids 14(5.7) 12(1.8) 17(2.8)
Electrolytes 7(2.9) 19(2.8) 6(1)
Cardiovasculars 4(1.6) 15(2.2) 11(1.8)
Dyslipidemic agents 7(2.9) 16(2.4) 6(1)
Analgesics 4(1.6) 9(1.3) 15(2.5)
Antiasthmatics 5(2) 14(2.1) 6(1)
Antituberculosis 3(1.2) 11(1.6) 7(1.1)
Vitamins 0 8(1.2) 5(0.8)
Antifungals 1(0.5) 4(0.6) 6(1)
Antiseizures 5(2) 3(0.4) 2(0.3)
Antipsychotics 1(0.5) 3(0.4) 6(1)
Thyroid agents 0 5(0.7) 3(0.5)
Antivirals 4(1.6) 2(0.3) 1(0.2)
Antihistamines 1(0.5) 3(0.4) 1(0.2)
Sedatives 4(1.6) 1(0.1) 0
Anticancers 0 3(0.4) 2(0.3)
Others 12(4.9) 41(6.1) 49(8)
Page 22 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
23
Table 6. Interrator reliability of the incident type and their severity and preventability
Severity Agreement % Kappa value
Exclude vs ADEs, PADE or medication error 56.4 0.63
ADEs vs PADE, medication error or exclude 93.6 0.71
PADEs vs ADE, medication error or exclude 70.9 0.62
Medication error vs ADEs, PADE or exclude 93.7 0.67
Preventable vs non-preventable ADEs 92.2 0.68
Fatal vs Life-threatening, serious or significant 100 0.74
Life-threatening vs fatal serious or significant 100 0.63
Serious vs fatal, life-threatening or significant 60.5 0.48
Significant vs fatal, life-threatening or serious 84.2 0.52
Table 7. Factors Associated with ADEs
Factor Unadjusted Odds Ratio
95% CI P value Adjusted Odds Ratio
95% CI P value
Lower Upper Lower Upper
Age 1.024 1.017 1.032 <0.001 1.012 1.003 1.021 0.009
Number of medications 1.193 1.148 1.240 <0.001 1.062 1.008 1.119 0.023
Charlson’s comorbidity index weight
1.251 1.158 1.352 <0.001 1.041 0.937 1.157 0.452
Length of hospital stay 1.042 1.034 1.051 <0.001 1.025 1.015 1.035 <0.001
Gender (Male)a 0.848 0.642 1.121 0.248 - - - -
ICUb 7.786 5.259 11.528 <0.001 3.276 2.005 5.354 <0.001
Medicineb 2.539 1.664 3.872 <0.001 1.736 1.078 2.796 0.023
CI, Confidence Interval, Reference categories: Femalea, Reference categories: Surgery
b
Page 23 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
215x279mm (300 x 300 DPI)
Page 24 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
Appendix A: Examples of PADEs
Prescribing
The physician ordered lisinopril, 5 mg tablet, once daily even though the patient was not hypertensive and had no indication for the
drug. The order was intended for another patient and the nurse intercepted the error.
Transcribing
A 74-year old man with diabetes mellitus, hypertension and recurrent urinary tract infections was admitted to the medical ward.
Meropenem, 500 mg intravenously, every eight hours was ordered. In the morning round, the infectious disease consultant verbally
asked the intern to change meropenem to imipenem 500 mg intravenously every six hours. However, the intern mistakenly
transcribed it as meropenem. This error was caught, corrected, and noted as an error in the patient’s medical record.
Dispensing
An order of metoclopramide 10 mg was sent to the pharmacy. The nurse obtained the drug from the pharmacy, but from the
appearance of the solution, she suspected that the preparation was not metoclopramide. The nurse contacted the pharmacy and the
pharmacist found that it was the wrong medication, although the label was stated that it was metoclopramide.
Administering
A nurse handled two capsules for two different patients in Room #8 and Room #9. She almost accidentally gave the wrong
medication (switched) to each patient. However, the patient in Room # 9 knew her medication and she said, “This is not my
medicine,” and the error was intercepted by the patient.
Page 25 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on Decem
ber 29, 2020 by guest. Protected by copyright.
http://bmjopen.bm
j.com/
BM
J Open: first published as 10.1136/bm
jopen-2015-010831 on 12 July 2016. Dow
nloaded from
For peer review only
STROBE 2007 (v4) Statement—Checklist of items that should be included in reports of cohort studies
Section/Topic Item
# Recommendation Reported on page #
Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 6
(b) Provide in the abstract an informative and balanced summary of what was done and what was found 2
Introduction
Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 5
Objectives 3 State specific objectives, including any prespecified hypotheses 5
Methods
Study design 4 Present key elements of study design early in the paper 6
Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data
collection
6
Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up 6
(b) For matched studies, give matching criteria and number of exposed and unexposed
Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if
applicable
6
Data sources/
measurement
8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe
comparability of assessment methods if there is more than one group
7
Bias 9 Describe any efforts to address potential sources of bias
Study size 10 Explain how the study size was arrived at 6
Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and
why
8
Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 8
(b) Describe any methods used to examine subgroups and interactions 8
(c) Explain how missing data were addressed
(d) If applicable, explain how loss to follow-up was addressed
(e) Describe any sensitivity analyses
Results
Page 26 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on December 29, 2020 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. Downloaded from
For peer review only
Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, confirmed
eligible, included in the study, completing follow-up, and analysed
10
(b) Give reasons for non-participation at each stage 10
(c) Consider use of a flow diagram 24
Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential
confounders
10
(b) Indicate number of participants with missing data for each variable of interest
(c) Summarise follow-up time (eg, average and total amount) 6
Outcome data 15* Report numbers of outcome events or summary measures over time 10
Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (eg, 95% confidence
interval). Make clear which confounders were adjusted for and why they were included
12
(b) Report category boundaries when continuous variables were categorized
(c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period
Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses
Discussion
Key results 18 Summarise key results with reference to study objectives 14
Limitations
Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from
similar studies, and other relevant evidence
15 -16
Generalisability 21 Discuss the generalisability (external validity) of the study results 16
Other information
Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on
which the present article is based
17
*Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies.
Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE
checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at
http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org.
Page 27 of 27
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
on December 29, 2020 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2015-010831 on 12 July 2016. Downloaded from