B

Kp

UBa

b

Ic

d

e

a

A

R

R

2

A

P

K

D

B

U

V

K

T

S

E

tV

f

1d

p r i m a r y c a r e d i a b e t e s 1 ( 2 0 0 7 ) 195–198

avai lab le at www.sc iencedi rec t .com

journa l homepage: ht tp : / /www. int l .e lsev ierhea l th .com/ journa ls /pcd /

rief Report

nowledge of blood pressure targets amongatients with diabetes�

sha Subramaniana,b,c,∗, T.P. Hoferd,e, M.L. Klamerusd,.J. Zikmund-Fisherd,e, M. Heislerd,e, E.A. Kerrd,e

Roudebush VA Medical Center, Indianapolis, IN, United StatesDivision of General Medicine & Geriatrics, Department of Medicine, Indiana University School of Medicine,

ndianapolis, IN, United StatesIU Center for Health Services & Outcomes Research, Regenstrief Institute, Inc., Indianapolis, IN, United StatesVeterans Affairs Health Services Research & Development Center of Excellence, Ann Arbor, MI, United StatesDepartment of Internal Medicine, University of Michigan, Ann Arbor, MI, United States

r t i c l e i n f o

rticle history:

eceived 20 February 2007

eceived in revised form

4 September 2007

ccepted 26 September 2007

ublished on line 31 October 2007

eywords:

iabetes

lood pressure

nited States Department of

eterans Affairs

nowledge

argets

a b s t r a c t

Introduction: Tight blood pressure (BP) control is the single most important intervention to

prevent cardiovascular mortality among patients with diabetes mellitus (DM). However, little

is known about how many patients have specific target BP levels or the factors associated

with patients’ knowledge of these targets.

Objectives: (1) To determine what proportion of patients with diabetes have BP targets; (2) To

determine patient characteristics associated with having a BP target.

Methods: Cross-sectional, anonymous survey of 500 randomly selected outpatients with

hypertension and DM receiving care in any Veterans Health Administration outpatient clinic

in 2003. We examined multivariate associations between patient characteristics and having

targets for BP. Covariates included age, race, gender, and education level; and factors specific

to diabetes and BP treatment, including medication use, diabetes duration, and number of

visits to diabetes healthcare providers in the previous year.

Results: Three hundred and seventy-eight (80%) patients responded. Although most (91%)

had blood glucose targets, fewer than 60% reported having a BP target. In multivariate anal-

elf-monitoring

ducation

yses, college education was associated with having a BP target (AOR 1.97 [95% CI: 1.16–3.34]).

Conclusions: Less than two-thirds of diabetic, hypertensive patients had BP targets. Encourag-

ing patients to set target BPs may promote hypertension self-management in this high-risk

Less educated patients may especially benefit from interventions to

patient population.

increase awareness of BP targets.

Published by Elsevier Ltd.

� Support: Department of Veterans Affairs, the Veterans Health Administration, Health Services Research and Development IIR 02-225-1,he QUERI DIB-98-001 and the Michigan Diabetes Research and Training Center Grant P60DK-20572 from the NIDDK. Michele Heisler is aA HSR&D Career Development awardee.∗ Corresponding author at: Roudebush VAMC, IF 122, 250 University Boulevard, Indianapolis, IN 46202, United States. Tel.: +1 317 278 1602;

ax: +1 317 278 0911.E-mail address: usubrama@iupui.edu (U. Subramanian).

751-9918/$ – see front matter Published by Elsevier Ltd.oi:10.1016/j.pcd.2007.09.001

e t e s

they did not know what to do to control their BP.

196 p r i m a r y c a r e d i a b

1. Introduction

Meticulous blood pressure (BP) control is the most importantway to reduce cardiovascular complications and overall mor-tality in patients with type 2 diabetes mellitus (DM) [1–3].Yet, one in three DM patients has uncontrolled BP [4]. Arecent study in the US Veterans Health Administration (VHA)found only 50% of DM patients had BP levels <140/90 mmHg[5]. Accordingly, the VHA National Clinical Practice Guide-lines Council made improved BP control a priority for qualityimprovement [6].

Achieving BP control requires patient self-managementand effective clinical management. Strategies such as goalsetting and self-monitoring facilitate acquisition of desiredbehaviors [7]. Patients’ lack of knowledge of their blood glucosetarget has been associated with worse control [8,9]. Similarly,having a BP target is the first step to involve patients in BPcontrol. While previous studies have emphasized patients’self-management with respect to glucose control, less workhas considered self-management to improve BP control. Arecent meta-analysis has shown that self-monitoring alsoimproves control of blood pressure by improving medicationadherence, as patients become more involved in their care[10]. The study by Cheng et al. assessing BP awareness amonghospitalized patients with coronary artery disease found thathaving DM was a risk factor for poor knowledge of BP tar-gets [11]. However, we know little about whether patients withDM are as, more, or less likely to have BP targets than bloodglucose targets, nor which patient factors are associated withpatients’ having BP targets. Understanding the characteristicsof patients who currently have BP targets could identify factorsfor interventions to improve understanding of the importanceof BP control. Our objectives, therefore, were to determine: (1)the proportions of DM patients with BP and glucose targets;and (2) factors associated with having BP targets.

2. Methods

2.1. Sample and recruitment

We conducted a cross-sectional anonymous survey; the sur-vey sample, obtained using national VHA data, consisted of500 randomly selected outpatients with both DM and hyper-tension who received outpatient primary care at any VHAoutpatient clinic in fiscal year 2003. These outpatient clin-ics provide the majority of primary care to patients with DM.Hypertension was established by ICD-9 codes 401.xx [12], inFY02 or FY03; DM was established by ICD-9 codes 250.xx, 357.2,362.0, 366.41 [13–15]; both associated with two outpatient vis-its (or 1 inpatient visit) and at least 1 non-ancillary outpatientvisit in FY03 [16]. Study procedures were approved by the Insti-tutional Review Board of the Ann Arbor VA Medical Center.

2.2. Procedures

In 2003, each patient was mailed a cover letter and sur-vey. As per the modified Dillman technique [17], all patients

1 ( 2 0 0 7 ) 195–198

received a reminder letter approximately 1 week later, andnon-respondents were sent a second survey 2 weeks after thereminder. Upon return, surveys were immediately separatedfrom envelopes, anonymizing all responses.

2.3. Survey questions

Patients were asked whether they had ‘a target number fortheir BP to be at or below’ (1 = yes, 0 = no). Patients respond-ing ‘yes’ were asked to list their target. Patients were alsoasked if they had a target number for their blood glucoselevel to be at or below. Patients responding ‘yes’ were askedto list their target morning, pre-prandial blood glucose level.We also assessed socio-demographic characteristics includ-ing age, race, gender, and education level; and factors specificto diabetes and BP treatment, including medication use, dia-betes duration, and number of visits to diabetes healthcareproviders in the previous year. We asked patients to list theirtop three health concerns and to assess their agreement withthe following question: “I do not understand what to do tocontrol my BP” (rated from strongly disagree to strongly agree).

2.4. Statistical analysis

We constructed univariate, bivariate and multivariate modelsto examine our study findings. Our multivariate models werebased on our a priori hypotheses and (did not use stepwiseregression to determine what variables to include in the mul-tivariate models) examined the association between patientcharacteristics and having a BP target, adjusting for age, race,education level, diabetes duration, insulin use, BP medicationuse, and number of provider visits. All analyses were con-ducted using Stata 9.0 (College Station, Texas).

3. Results

Overall response rate was 80% (378 returned; 94 not returned;16 returned due to address problems; 10 denied having dia-betes; 2 deceased). Rates of missing values were 16% for havinga BP target and <8% for all other variables. Demographic char-acteristics are shown in Table 1. Less than half of the subjectshad any college education. Eighty percent were or have beenon, medications to treat both BP and diabetes, yet only 59% hada target level for BP. There were no differences in the demo-graphic characteristics of patients who reported having a BPtarget versus those who did not [data not shown]. Of those whoknew what blood glucose was, 91% had a glucose target level.Among those reporting a BP target, 84% had a self-reportedVHA guideline-congruent systolic goal (≤135 mmHg) and 94%had a VHA guideline-congruent diastolic goal (≤80 mmHg) [7].While 86% listed ‘controlling blood sugar’ among their topthree health concerns, only 55% listed ‘controlling BP’ in theirtop three concerns. While just over half of the patients mon-itored their BP at home, 25% agreed or strongly agreed that

In multivariate analyses, having a college education wasassociated with having a BP target (AOR = 1.97; 95% CI,1.16–3.34) (Table 2).

p r i m a r y c a r e d i a b e t e s 1 ( 2 0 0 7 ) 195–198 197

Table 1 – Characteristics of respondents (N = 378)

Mean ± S.D.

Age (years) 68.9 ± 9.8Duration of diabetes (years) 12.6 ± 10.1Visits to diabetes providers in the past year 5.3 ± 6.0

% Number

Male 98 366Race

White 85 310Black 11 39Other 4 15

EducationSome college or more 45 162High school or less 55 199

Diabetes treatmentDiet and exercise only 10 39Oral medication only 57 214Insulin and oral medication 14 54Insulin only 18 67

Take medications to treat blood pressure 90 322/359Have a target blood pressure level 59 188/317Have a guideline-congruent systolic goal (≤135 mmHg) 84 157/188Have a guideline-congruent diastolic goal (≤80 mmHg) 94 176/188Have a target blood glucose level 91 276/305Controlling BP is among top three health concerns 55 151/275

ontro

4

Obtfpgrthscah

h

Controlling blood glucose is among top three health concernsMonitor blood pressure at homeAgree with the statement “I do not understand what I need to do to c

. Discussion

ur study describes patient awareness of targets for BP andlood glucose and identifies patient education as the only fac-or among those we studied, associated with reporting targetsor BP control among this high-risk population. Almost allatients on medications to treat both BP and diabetes had alucose target, yet nearly half did not have a BP target. Thisesult is not surprising: physicians, like patients, have tendedo prioritize glucose rather than BP control [18]. However, sinceaving BP targets is essential for self-monitoring, and effectiveelf-monitoring has been clearly associated with improved BPontrol [19], our study suggests that clinicians should place

dditional emphasis on BP goals to help patients self-manageypertension.

We found that lower education was associated with notaving a BP target, consistent with recent studies assessing

Table 2 – Multivariate logistic regression model examining facto

Age (years)Race (white)College educationDuration of diabetes (years)Diabetes (insulin) treatmentBP treatment (on medication)Number of visits to diabetes providers in the past year

86 237/27554 192/356

l my BP” 25 74/299

knowledge of BP targets among patients with coronary arterydisease and diabetes [11] and assessing glycemic control in DM[20]. Thus, interventions to increase awareness of the impor-tance of a BP target may be especially important for this group.

Our findings are limited by our study design. First, we canonly comment on the associations between key variables, nottheir directionality or causality. Second, we were unable tolink to medical record data to examine associations betweentargets, self-monitoring, and actual BP control. However, allrespondents had a diagnosis of hypertension, 90% were onanti-hypertensive medications, and 50% reported that eithertheir systolic BP was >135 mm or diastolic BP was >80 mmHgat last measurement. Further, those reporting that their lastBP measurement was >135/80 were no more likely to list BP

among their top three health concerns, nor were they morelikely to have a BP target. Finally, our study was conductedamong predominantly male veterans and may not be gener-alizable to women or other settings. Indeed, because quality

rs associated with patients reporting a BP target (N = 257)

Adjusted odds ratios and confidence intervals

0.98 (0.96–1.01)1.21 (0.56–2.58)1.97 (1.16–3.34)0.98 (0.95–1.01)1.46 (0.78–2.72)1.51 (0.52–4.32)0.99 (0.95–1.04)

e t e s

r

198 p r i m a r y c a r e d i a b

of VHA care for diabetes has been shown to be better than inother settings [14,21], the magnitude of the problem may belarger in non-VHA settings.

Less than two-thirds of diabetic, hypertensive patients hadBP targets. Encouraging patients to set target BPs may pro-mote hypertension self-management in this high-risk patientpopulation, and this type of goal setting may be an importantpart of patient educational programs. Less educated patientsmay benefit from interventions to increase awareness of BPtargets.

Conflict of interest

The authors have no financial conflicts of interest to disclose.

Acknowledgement

The views expressed in this article are those of the authorsand do not necessarily represent the views of the Departmentof Veteran Affairs.

e f e r e n c e s

[1] L. Hansson, A. Zanchetti, S.G. Carruthers, B. Dahlof, D.Elmfeldt, S. Julius, et al., Effects of intensive blood-pressurelowering and low-dose aspirin in patients withhypertension: principal results of the Hypertension OptimalTreatment (HOT) randomised trial. HOT Study Group, Lancet351 (1998) 1755–1762.

[2] V. Snow, K.B. Weiss, C. Mottur-Pilson, The evidence base fortight blood pressure control in the management of type 2diabetes mellitus, Ann. Intern. Med. 138 (2003) 587–592.

[3] D.R. Berlowitz, A.S. Ash, E.C. Hickey, M. Glickman, R.Friedman, B. Kader, Hypertension management in patientswith diabetes: the need for more aggressive therapy,Diabetes Care 26 (2003) 355–359.

[4] J.B. Saaddine, M.M. Engelgau, G.L. Beckles, E.W. Gregg, T.J.Thompson, K.M. Narayan, A diabetes report card for theUnited States: quality of care in the 1990s, Ann. Intern. Med.136 (2002) 565–574.

[5] E.A. Kerr, D.M. Smith, M.H. Hogan, et al., Comparing clinicalautomated, medical record and hybrid data sources for

diabetes quality measures, Jt. Comm. J. Qual. Improv. 28(2002) 555–565.

[6] VA Office of Quality and Performance, July 22, 2005, DiabetesMellitus Clinical Practice Guidelines, http://www.oqp.med.va.gov/cpg/DM/DM base.htm (accessed July 21, 2006).

1 ( 2 0 0 7 ) 195–198

[7] R. Cameron, J.A. Best, Promoting adherence to healthbehavior change interventions: recent findings frombehavioral research, Patient Educ. Couns. 10 (1987)139–154.

[8] S. Panja, B. Starr, K.M. Colleran, Patient knowledge improvesglycemic control: is it time to go back to the classroom? J.Investig. Med. 53 (5) (2005) 264–266.

[9] C.S. Blaum, L. Velez, R.G. Hiss, J.B. Halter, Characteristicsrelated to poor glycemic control in NIDDM patients incommunity practice, Diabetes Care 20 (1997) 7–11.

[10] F.P. Cappuccio, S.M. Kerry, L. Forbes, A. Donald, Bloodpressure control by home monitoring: meta-analysis ofrandomized trials, BMJ 329 (2004) 145–151.

[11] S. Cheng, J.H. Lichtman, J.M. Amatruda, G.L. Smith, J.A.Mattera, S.A. Roumanis, H.M. Krumholz, Knowledge of bloodpressure levels and targets in patients with coronary arterydisease in the USA, J. Hum. Hypertens. 19 (10) (2005)769–774.

[12] M.K. Goldstein, R.W. Coleman, S.W. Tu, R.D. Shankar, M.J.O’Connor, M.A. Musen, S.B. Martins, P.W. Lavori, M.G.Shlipak, E.Z. Oddone, A.A. Advani, P. Gholami, B.B. Hoffman,Translating research into practice: organizational issues inimplementing automated decision support for hypertensionin three medical centers, J. Am. Med. Inform. Assoc. JAMIA11 (2004) 368–376.

[13] D.R. Miller, M.M. Safford, L.M. Pogach, Who has diabetes?Best estimates of diabetes prevalence in the Department ofVeterans Affairs based on computerized patient data,Diabetes Care 27 (Suppl. 2) (2004) B10–B21.

[14] E.A. Kerr, R.B. Gerzoff, S.L. Krein, et al., A comparison ofdiabetes care quality in the Veterans Affairs HealthcareSystem and commercial managed care: results from theTRIAD Study, Ann. Intern. Med. 141 (2004) 272–281.

[15] P.L. Hebert, L.S. Geiss, E.F. Tierney, M.M. Engelgau, B.P. Yawn,A.M. McBean, Identifying persons with diabetes usingMedicare claims data, Am. J. Med. Qual. 14 (6) (1999)270–277.

[16] S.L. Krein, M.L. Klamerus, S. Vijan, J.L. Lee, J.T. Fitzgerald, A.Pawlow, et al., Case management for patients with poorlycontrolled diabetes: a randomized trial, Am. J. Med. 116 (11)(2004 Jun) 732–739.

[17] D.A. Dillman, Mail and Internet Surveys: The TailoredDesign Method, John Wiley and Sons, Inc., New York, NY,2000.

[19] R.J. McManus, M.A. Roalfe, R.A. Oakes, S. Bryan, H.M.Pattison, F.D. Hobbs, Targets and self monitoring inhypertension: randomized controlled trial and costeffectiveness analysis, BMJ 331 (2005) 466–467.

[21] S.M. Asch, E.A. McGlynn, M.M. Hogan, R.A. Hayward, P.Shekelle, L. Rubenstein, et al., Comparison of quality of carefor patients in the Veterans Health Administration andpatients in a national sample, Ann. Intern. Med. 141 (12)(2004) 938–945.