ORI GIN AL PA PER
Attitudes of Muslim Physicians and Nurses TowardReligious Issues
Sina Hafizi • Harold G. Koenig • Mohammad Arbabi •
Mohammad Pakrah • Amene Saghazadeh
� Springer Science+Business Media New York 2013
Abstract There is a growing body of evidence that suggests a positive role for religious
involvement in physical and mental health. Studies have shown that attitudes of physicians
toward religion affect their relationship with patients and their medical decisions, and in
this way may ultimately affect treatment outcomes. Attitudes of nurses toward religion
could also influence whether or not they address patients’ unmet spiritual needs. To assess
attitudes of physicians and nurses toward religion and how these attitudes vary by edu-
cation level and demographic characteristics, a total of 800 physicians, medical students,
and nurses from some of the largest hospitals in Tehran, Iran, were approached, of whom
720 completed questionnaires (148 nurses, 572 medical students and physicians). The
survey questionnaire included the Duke University Religion Index (DUREL), Hoge
Intrinsic Religiosity Scale, a brief measure of Negative Religious Coping (NRCOPE), and
the brief Trust/Mistrust in God Scale. Religious attitudes and practices were compared
between physicians (medical students and physicians) and nurses. Regression analysis
revealed that except for intrinsic religiosity, physicians were not less religious than nurses
on any other dimension of religiosity. Training level (year of training) was a predictor of
religiosity, with those having less training being the most religious. The findings suggest
S. Hafizi (&) � M. PakrahSchool of Medicine, Tehran University of Medical Sciences, No. 7, Al-e-Ahmad Highway,P. O. Box 14395-578, Tehran, Irane-mail: [email protected]
H. G. KoenigCenter for Spirituality, Theology and Health, Duke University Medical Center, Durham, NC, USA
H. G. KoenigKing Abdulaziz University (KAU), Jidda, Saudi Arabia
M. ArbabiDepartment of Psychiatry, Tehran University of Medical Sciences, Tehran, Iran
A. SaghazadehSports Medicine Research Center, Neuroscience Institute, Tehran University of Medical Sciences,Tehran, Iran
123
J Relig HealthDOI 10.1007/s10943-013-9730-1
that there are few religious differences between nurses and physicians in Iran. However,
religiosity may become less as the training level increases. Lack of emphasis in training on
the important role that religion plays in health care may result in a decrease in religious
involvement and the development of negative attitudes toward religion over time (dis-
placed by a focus on the technological aspects of health care).
Keywords Spirituality � Religious involvement � Medical staff � Muslim
Introduction
During the last decade, more attention has been placed on the link between religion/
spiritual (R/S) issues and health in the medical literature (Koenig 2012). Considerable
research has sought to determine whether R/S affect patients’ health, what factors moderate
this relationship, how to go about assessing R/S in medical and health settings, and in
particular, how the R/S commitments of the clinicians influence their practices (Curlin
2008).
With regard to the high prevalence of religiosity and religious attendance of people in
eastern and especially Middle East countries and the major impact of R/S issues in their
everyday life, there is a limited number of published papers available in medical literature
about the attitude of medical staffs and professionals toward R/S issues in these countries
(Koenig et al. 2012). This is in contrast to European and North American countries where
research on R/S and health is a well-known topic of research in their medical health
settings. There are studies that suggest positive attitude of physicians toward patients’
religious beliefs could positively influence doctor–patient relationship (MacLean et al.
2003). Religion and religious beliefs are proposed to be associated with better mental and
physical health (Koenig 2009), and in large sample studies which have been conducted
mostly in the USA, it has been shown that physicians and nurses who are more religious
and spiritual are much more likely to address R/S issues in their practice (Curlin et al.
2006; Stern et al. 2011). Therefore, due to lack of studies looking into the status of
religious beliefs of medical health-care providers, in this study, we asked a sample of
physicians, medical students, and nurses practicing in Iranian hospitals to fill out ques-
tionnaire on religion and religious attendance. We aimed to find answer to questions
regarding the attitude of physicians and nurses toward R/S issues and factors that affect/
mediate this attitude. We hypothesized that as the more experienced physicians and nurses
have had more contacts with patients, they may have more positive attitudes toward R/S.
So we expected a positive correlation between training level and number of years in current
status and religiosity. We also hypothesized that as nurses usually have a closer rela-
tionship with the patients, they should be more religious in comparison with physicians.
Methods
Participants and Design
This study surveyed 720 health professionals: 148 nurses and 572 physicians. The mean
age was 24.61 (SD = 6.61). The general characteristics of the sample are shown in
J Relig Health
123
Table 1. The participants were informed that participation was voluntary and their
responses would not be individually identified. They were from some of the largest hos-
pitals in Tehran, Iran. This study was approved by ethics in research committee of Tehran
University of Medical Sciences.
From 800 persons that were initially approached, 720 persons accepted to participate in
this study, given a response rate of 90.7 %. Six potential subjects were excluded due to the
missing data. The participants were asked for age, experience years, major (physician or
nurse), marital status, number of years in current status, and training level and were
administered the following measures:
Duke University Religion Index (DUREL) is a five-item questionnaire measuring three
core dimensions of religious attendance including organizational religious activity (ORA/one
item), non-organizational religious activity (NORA/one item), and intrinsic religiosity (IR/
three items) (Koenig et al. 1997). The inclusiveness of DUREL, along with its brevity, makes
it a valuable measure to be used in large-scale surveys (Koenig and Bussing 2010). Cron-
bach’s alpha coefficient, an indicator of reliability of the questionnaire, was measured to be
0.85. In this study, we used a validated Farsi version of the DUREL (Hafizi et al. 2013).
Hoge Intrinsic Religiosity Index is a well-known measure of intrinsic religiosity. The
measure asks respondents to rate their intrinsic religiosity on a 1–5 Likert-like scale, giving
a total score of 10–50. Reliability analyses of the questionnaire showed Cronbach’s alpha
coefficient to be 0.88.
Brief Trust/Mistrust in God Scale is a six-item self-report measure of positive and
negative beliefs about God. It consists of three items for positive and three items for
negative beliefs. It has satisfactory internal consistency and test–retest reliability (Rosm-
arin et al. 2011). Internal consistency of the measure was assessed using Cronbach’s alpha
coefficient for Trust (TIG)/Mistrust (MIG) subscales. The Cronbach’s alphas were 0.90 and
0.92 for Trust/Mistrust in God subscales, respectively.
Table 1 Demographic anddescriptive characteristics of theparticipants
Hoge IR Hoge IntrinsicReligiosity Scale, TIG Trust inGod Scale, MIG Mistrust in GodScale, NRCOPE NegativeReligious Coping Scale, DURELDuke University Religion Index,ORA organizational religiousactivities, NORA non-organizational religiousactivities, IR intrinsic religiosity
Characteristics Mean (SD)
Total number 720
Female (%) 61.7
Age (year) 24.61 (6.61)
Major (%)
Nurse 20.6
Physician 79.4
Married (%) 23.2
Training level 3.68 (2.37)
Experience years 7.92 (6.58)
Hoge IR 18.60 (5.96)
God trust 12.31 (3.05)
God mistrust 4.60 (2.77)
NRCOPE 4.19 (3.21)
DUREL
ORA 2.95 (1.60)
NORA 4.36 (2.08)
IR 11.28 (3.14)
Total 18.60 (5.96)
J Relig Health
123
Brief Negative RCOPE (NRCOPE). The NRCOPE (seven-item measure) measures the
negative reappraisal of God’s power, demonic reappraisal, spiritual discontent, punishing
God reappraisal, and interpersonal religious discontent (Pargament et al. 2011). Previously
studies have shown the association of NRCOPE and poor mental health such as depression
and anxiety (Pargament et al. 2002). Cronbach’s alpha coefficient of NRCOPE was 0.73.
Analyses
Descriptive statistics described the general characteristics of the sample. The independent
sample t test and Spearman’s correlation were used to compare nurses and physicians on
different measures. Two linear regression models were used to examine predictor of
outcomes. In the first step, major (i.e., physician and nurse) was examined alone as a
predictor. In the second step, demographic variables (i.e., age, gender, marital status, and
training level) were included in the model. Significance level was set at 0.05.
Results
The general characteristics of the participants are summarized in Table 1. Of the 720
subjects, 61.7 % were female. The females in our sample were significantly more religious
than men on the DUREL (DUREL total score: p = 0.033, Cohen’s d = 0.17/ORA:
p = 0.039, Cohen’s d = 0.16/NORA: p \ 0.001, Cohen’s d = 0.27/IR: p = 0.005,
Cohen’s d = 0.21), Hoge IR (p = 0.035, Cohen’s d = 0.16), and God trust scale
(p = 0.024, Cohen’s d = 0.20). They also had significantly higher (p = 0.005, Cohen’s
d = 0.22) scores on NRCOPE in comparison with the men. Women were significantly
more likely to be married than men (Pearson’s v2= 22.32, p \ 0.01). Married subjects were
more religious than non-married participants by all religious measures (DUREL total
score: p = 0.018, Cohen’s d = 0.18/NORA: p = 0.051, Cohen’s d = 0.15/IR: p = 0.002,
Cohen’s d = 0.24) and Hoge IR (p = 0.052, Cohen’s d = 0.14) scores.
There was an inverse (r = -0.169, p \ 0.001) correlation between age and NRCOPE
scores. Training level was correlated with God mistrust (r = 0.160, p \ 0.001) and
inversely correlated with DUREL total (r = -0.142, p \ 0.001), Hoge IR (r = -0.080,
p = 0.05), God trust scale (r = -0.184, p \ 0.001), and NRCOPE (r = -0.201,
p \ 0.001). There was a significant positive correlation between DUREL total scores and
Hoge IR (r = 0.727, p \ 0.001), and God trust scale (r = 0.576, p \ 0.001). The
NRCOPE was positively correlated with God mistrust (r = 0.159, p \ 0.001) and inver-
sely correlated with God trust (r = 0.137, p \ 0.001) scale scores (Table 2).
As shown in Table 3, the Hoge IR (p = 0.003, Cohen’s d = 0.26), DUREL (total score:
p = 0.018, Cohen’s d = 0.53/NORA: p \ 0.001, Cohen’s d = 0.56/IR: p \ 0.001,
Cohen’s d = 0.55), and God trust (p = 0.003, Cohen’s d = 0.29) were significantly higher
for nurses compared to physicians. The percentage of married nurses was significantly
(p \ 0.01) higher than married physicians.
Table 4 shows the results of multivariable analysis. Nurses were significantly more
religious compared to physicians according to the DUREL total score (b = 2.80,
p \ 0.0001), Hoge IR score (b = 2.36, p = 0.004), and God trust scale score (b = 0.81,
p = 0.005). After including other demographic variables into the model (model 2), the
coefficient for major decreased and became statistically non-significant for all religious
measures except the Hoge IR score (b = 2.91, p = 0.035).
J Relig Health
123
Ta
ble
2S
pea
rman
’sco
rrel
atio
nco
effi
cien
tb
etw
een
age,
trai
nin
gle
vel
,y
ears
incu
rren
tst
atu
s,an
dm
easu
res’
sco
res
23
45
67
89
10
MIG
1-A
ge
0.8
16
**
*0
.35
7*
*-
0.0
58
0.0
73
0.0
50
0.0
17
0.0
10
-0
.169
**
*-
0.0
71
0.0
95
*
2-T
rain
ing
lev
el–
0.5
17
**
-0
.169
**
*-
0.0
79
*-
0.1
04
**
-0
.142
**
*-
0.0
80
*-
0.2
01
**
*-
0.1
84
**
*-
0.1
60
**
*
3-Y
ears
incu
rren
tst
atu
s–
–0
.009
0.1
30
0.0
67
0.0
75
-0
.05
4-
0.0
08
0.0
92
0.1
02
4-O
RA
––
–0
.582
**
*0
.62
0*
**
0.8
30
**
*0
.54
6*
**
0.0
33
0.4
19
**
*-
0.3
20
**
*
5-N
OR
A–
––
–0
.64
9*
**
0.8
12
**
*0
.57
5*
**
0.0
49
0.4
82
**
*-
0.3
53
**
*
6-I
R–
––
––
0.9
04
**
*0
.74
8*
**
0.0
48
0.5
75
**
*-
0.4
16
**
*
7-D
UR
EL
(To
tal
sco
re)
––
––
––
0.7
27
**
*0
.046
0.5
76
**
*-
0.4
16
**
*
8-H
og
eIR
––
––
––
–0
.056
0.4
87
**
*-
0.3
78
**
*
9-N
RC
OP
E–
––
––
––
–-
0.1
37
**
*0
.15
9*
**
10
-TIG
––
––
––
––
–-
0.6
41
**
*
Ho
ge
IRH
og
eIn
trin
sic
Rel
igio
sity
Sca
le,T
IGT
rust
inG
od
Sca
le,M
IGM
istr
ust
inG
od
Sca
le,N
RC
OP
EN
egat
ive
Rel
igio
us
Cop
ing
Sca
le,D
UR
EL
Du
ke
Un
iver
sity
Rel
igio
nIn
dex
,O
RA
org
aniz
atio
nal
reli
gio
us
acti
vit
ies,
NO
RA
non-o
rgan
izat
ional
reli
gio
us
acti
vit
ies,
IRin
trin
sic
reli
gio
sity
*p\
0.0
5;
**
p\
0.0
1;
**
*p\
0.0
01
J Relig Health
123
Except for Hoge IR score, training level was a significant predictor for all other mea-
sures (DUREL: b = -0.574, p \ 0.0001/God trust: b = -0.318, p \ 0.0001/God mis-
trust: b = 0.221, p = \ 0.0001/NRCOPE: b = -0.268, p \ 0.0001). Marital status was
also a significant predictor for God trust (b = 0.849, p = 0.018), Hoge IR (b = 2.193,
p = 0.036), and God mistrust (b = -0.911, p = 0.006).
Discussion
To our knowledge, this is the first study to examine the attitudes of Muslim physicians and
nurses toward religion using established measures of religious involvement. Female or
married nurses and physicians were more religious than male or single ones. Nurses were
not significantly more religious than physicians by most measures used in this study.
However, training level was an independent predictor of religiosity and was inversely
related to it.
Table 3 Difference of measures’ scores between physicians and nurses
Measures Physician (SD) Nurse (SD) P value Cohen’s d t Eta-square (g2)
Hoge IR 33.85 (8.65) 36.25 (9.32) 0.003 0.26 2.96 0.012
TIG 12.14 (3.17) 12.98 (2.47) 0.003 0.29 2.95 0.012
MIG 4.62 (2.81) 4.50 (2.64) 0.651 0.04 -0.45 0.000
NRCOPE 4.29 (3.24) 3.88 (3.12) 0.171 0.12 -1.37 0.002
DUREL
ORA 2.91 (1.64) 3.13 (1.40) 0.158 0.14 1.41 0.002
NORA 4.17 (2.12) 5.22 (1.58) \0.001 0.56 5.55 0.041
IR 10.98 (3.19) 12.58 (2.52) \0.001 0.55 5.57 0.041
Total 18.06 (6.10) 20.92 (4.53) \0.001 0.53 5.24 0.036
Hoge IR Hoge Intrinsic Religiosity Scale, TIG Trust in God Scale, MIG Mistrust in God Scale, NRCOPENegative Religious Coping Scale, DUREL Duke University Religion Index, ORA organizational religiousactivities, NORA non-organizational religious activities, IR intrinsic religiosity
Table 4 Regression analysis regarding the score of the physicians and nurses in different scales
Measures Model 1a unstandardized coefficient Model 2b unstandardized coefficient
DUREL (Total score) 2.80 (p \ 0.0001) R2 = 0.034 1.70 (p = 0.064) R2 = 0.061
Hoge IR 2.36 (p = 0.004) R2 = 0.010 2.91 (p = 0.035) R2 = 0.023
NRCOPE -0.475 (p = 0.12) R2 = 0.002 -0.832 (p = 0.10) R2 = 0.047
TIG 0.81 (p = 0.005) R2 = 0.010 0.53 (p = 0.25) R2 = 0.059
MIG -0.90 (p = 0.73) R2 = - 0.01 0.25 (p = 0.55) R2 = 0.029
Hoge IR Hoge Intrinsic Religiosity Scale, TIG Trust in God Scale, MIG Mistrust in God Scale, NRCOPENegative Religious Coping Scale, DUREL Duke University Religion Index, ORA organizational religiousactivities, NORA non-organizational religious activities, IR intrinsic religiosity
** Presented unstandardized coefficient and P values are for majora In model 1, major (i.e., physician and nurse) was included as the predictor variableb In model 2, major (i.e., physician and nurse) plus demographic variables (i.e., age, gender, marital status,and training level) were included as independent variables
J Relig Health
123
The finding that females or married nurses and physicians are generally more religious
than male or single ones is consistent with the results of other studies (Stark 2002; Watson
et al. 1988).
There was no relationship between number of years as a nurse or physician and reli-
giosity. This finding, along with the inverse relationship between religiosity and training
level, argues against the primary hypothesis that nurses and physicians with more expe-
rience would be more religious.
Furthermore, nurses may be not, in general, more religious than physicians. This also
argues against our initial hypothesis that as nurses have closer relationship with patients,
they should be more religious (Christopher 2010). One possible explanation is that nurses
do not receive the proper training on how to address spiritual issues in their relationship
with patients. Although no previous studies have compared religiosity in nurses and
physicians, those on separate samples of physicians and nurses have revealed the impor-
tance of addressing spiritual needs of the patients. These studies found that religiosity of
physicians and nurses increases the likelihood of addressing religious issues in the doctor–
patient and nurse–patient relationships (Christopher 2010; Curlin et al. 2006). Studies have
also shown that the religious beliefs of physicians influence their decisions on morally
controversial issues such as end of life care (Curlin 2008). Since most studies have shown
that patient’s religious and spiritual needs are often ignored, addressing these needs could
have positive impact on treatment outcomes (Pearce et al. 2012).
Also in contrast to our primary hypotheses, we did not find a positive relationship
between training level and religiosity scores, and if anything, there was a significant
independent inverse relationship. This finding could be interpreted in a number of ways.
The negative psychological effects of confronting death and human suffering in clinical
settings, particularly in religious patients, may give the impression that their beliefs have
nothing to do with their ability to cope. This could force nurses and physicians to shift their
attention to the technical aspects of care rather than to the spiritual. Another explanation
may be the dehumanizing aspects of training that nurses and physicians are exposed to,
such that their attention becomes focused on the more biological and technical aspects of
care (which may also ultimately influence their own personal religious beliefs and prac-
tices). Beside this lack of proper training, a lack of guidance from mentors and supervisors
may also exacerbate the problem. Another reason is that the high workload may force the
physicians and nurses to suppress their religious feelings in clinical settings that negatively
affect their personal attitudes toward religion. Nurses and physicians may also be con-
cerned that religious conflicts between patients and themselves that could adversely affect
the treatment process and patients’ adherence. Negative feedback that they may receive
from the patients while addressing spiritual issues could also affect their personal attitude
toward these issues.
The findings of this study are important for health-care providers and health policy
makers. Studies have shown that a large number of patients have religion and spiritual
needs which are not being met mostly (Koenig 2012). On the other hand, religious phy-
sicians and nurses are more likely to address religious issues in their practice (Curlin et al.
2006; Kevern 2012). Physicians and nurses should be trained on how to provide spiritual
care or at least how to assess for spiritual needs that may be present. They should also be
encouraged to acknowledge the patients’ beliefs that support them in coping with pain and
suffering (Koenig et al. 2000).
A future direction for this study might be to assess how and to what extent the religi-
osity and religious affiliation of physicians and nurses affect their decision making, their
relationship with patients, and their treatment outcomes. Another area of future research
J Relig Health
123
would be to assess the influence that a training program on spirituality and medicine has on
attitudes of nurses and physicians toward addressing religious issues with patients.
The results of this study should be interpreted with caution due to following limitations.
First, the cross-sectional nature of this study does not allow any interpretation about the
causality of the relationship between different variables. Second, the self-report nature of
the assessments utilized in this study is another limitation. Third, all of our participants
were Shia Muslims, which limits generalizability of our findings across other religions.
References
Christopher, S. A. (2010). The relationship between nurses’ religiosity and willingness to let patients controlthe conversation about end-of-life care. Patient Education and Counseling, 78(2), 250–255.
Curlin, F. A. (2008). Commentary: A case for studying the relationship between religion and the practice ofmedicine. Academic Medicine, 83(12), 1118–1120.
Curlin, F. A., Chin, M. H., Sellergren, S. A., Roach, C. J., & Lantos, J. D. (2006). The association ofphysicians’ religious characteristics with their attitudes and self-reported behaviors regarding religionand spirituality in the clinical encounter. Medical care, 44(5), 446.
Hafizi, S., Memari, A., Pakrah, M., Mohebi, F., Saghazadeh, A., & Koenig, H. G. (2013). The DukeUniversity Religion Index (DUREL): Validation and reliability of the Farsi version. PsychologicalReports, 112(1), 151–159.
Kevern, P. (2012). Who can give ‘spiritual care’? The management of spiritually sensitive interactionsbetween nurses and patients. Journal of Nursing Management.
Koenig, H. G. (2009). Research on religion, spirituality, and mental health: A review. Canadian Journal ofPsychiatry, 54(5), 283–291.
Koenig, H. G. (2012). Commentary: Why do research on spirituality and health, and what do the resultsmean? Journal of religion and health, 1–8.
Koenig, H. G., Al Zaben, F., & Khalifa, D. (2012). Religion, spirituality and mental health in the West andthe Middle East. Asian Journal of Psychiatry, 5, 180–182.
Koenig, H. G., & Bussing, A. (2010). The duke university religion index (DUREL): A five-item measure foruse in epidemological studies. Religions, 1(1), 78–85.
Koenig, H. G., Nicklin, D. E., Kaufman, A. S., Hite, R. W., Cameron, J. R., Hall, D. E., et al. (2000).Medicine and religion. New England Journal of Medicine, 343(18), 1339–1342.
Koenig, H., Parkerson Jr, G. R., & Meador, K. G. (1997). Religion index for psychiatric research. TheAmerican Journal of Psychiatry, 154(6), 885–886.
MacLean, C. D., Susi, B., Phifer, N., Schultz, L., Bynum, D., Franco, M., et al. (2003). Patient preference forphysician discussion and practice of spirituality. Journal of General Internal Medicine, 18(1), 38–43.
Pargament, K., Feuille, M., & Burdzy, D. (2011). The brief RCOPE: Current psychometric status of a shortmeasure of religious coping. Religions, 2(1), 51–76.
Pargament, K. I., Tarakeshwar, N., Ellison, C. G., & Wulff, K. M. (2002). Religious coping among thereligious: The relationships between religious coping and well-being in a national sample of Presby-terian clergy, elders, and members. Journal for the Scientific Study of Religion, 40(3), 497–513.
Pearce, M. J., Coan, A. D., Herndon, J. E., Koenig, H. G., & Abernethy, A. P. (2012). Unmet spiritual careneeds impact emotional and spiritual well-being in advanced cancer patients. Supportive Care inCancer, 1–8.
Rosmarin, D. H., Pirutinsky, S., & Pargament, K. I. (2011). A brief measure of core religious beliefs for usein psychiatric settings. The International Journal of Psychiatry in Medicine, 41(3), 253–261.
Stark, R. (2002). Physiology and faith: Addressing the ‘‘universal’’ gender difference in religious com-mitment. Journal for the Scientific Study of Religion, 41(3), 495–507.
Stern, R. M., Rasinski, K. A., & Curlin, F. A. (2011). Jewish physicians’ beliefs and practices regardingreligion/spirituality in the clinical encounter. Journal of Religion and Health, 50(4), 806–817.
Watson, P., Howard, R., Hood Jr, R. W., & Morris, R. J. (1988). Age and religious orientation. Review ofreligious research, 271–280.
J Relig Health
123