Medication non-adherence and poor glycaemic control in patients with type 2 diabetes mellitus

Medication non-adherence and poor glycaemic control in patients with type 2 diabetes mellitus

diabetes research and clinical practice 97 (2012) 377–384 Contents available at Sciverse ScienceDirect Diabetes Research and Clinical Practice journ...

286KB Sizes 0 Downloads 38 Views

diabetes research and clinical practice 97 (2012) 377–384

Contents available at Sciverse ScienceDirect

Diabetes Research and Clinical Practice journ al h ome pa ge : www .elsevier.co m/lo cate/diabres

Medication non-adherence and poor glycaemic control in patients with type 2 diabetes mellitus Elke Raum a,1,*, Heike U. Kra¨mer a,1, Gernot Ru¨ter b, Dietrich Rothenbacher c, Thomas Rosemann d, Joachim Szecsenyi e, Hermann Brenner a a

Division of Clinical Epidemiology and Aging Research, German Cancer Research Center, Im Neuenheimer Feld 581, D-69120 Heidelberg, Germany Practice of General Medicine, Blumenstrasse 11, D-71726 Benningen/Neckar, Germany c Institute of Epidemiology and Medical Biometry, Ulm University, Helmholtzstrasse 22, D-89081 Ulm, Germany d Department of General Practice and Health Services Research, University of Zu¨rich, University Hospital of Zu¨rich, Pestalozzistrasse 24, CH-8091 Zu¨rich, Switzerland e Department of General Practice and Health Services Research, University Hospital Heidelberg, Vossstrasse 2, Geb. 37, D-69115 Heidelberg, Germany b

article info

abstract

Article history:

Aims: Our main aim was to analyse gender differences in the association of adherence and

Received 25 January 2012

poor glycaemic control (PGC) in a cohort of patients with type 2 diabetes mellitus in Germany.

Received in revised form

Methods: Baseline data of the DIANA-study, a prospective cohort study of type 2 diabetes

22 May 2012

mellitus patients in South-West Germany, were analysed. Information on medication

Accepted 28 May 2012

adherence and factors related to PGC was obtained by self-administered questionnaire.

Published on line 2 July 2012

PGC was defined as HbA1c  7.5%. Bivariate and multivariate analyses using log-binomial

Keywords:

ence and PGC.

Primary care

Results: 624 men and 518 women were included in the analyses. In total, 147 men (24%) and

Type 2 diabetes mellitus

114 women (23%) reported non-adherence to medication. In men, PGC was found in 37% of

regression were employed to assess overall and gender-specific associations of non-adher-

Glycaemic control

the participants reporting non-adherence and in 19% reporting adherence (adjusted preva-

Adherence

lence ratio (PR) = 1.90, 95%-CI: 1.46–2.49). In women, PGC was found in 19% of the partici-

Compliance

pants reporting non-adherence and in 18% reporting adherence (adjusted PR = 0.97, 95%-CI: 0.65–1.46). Conclusions: Our results show gender-specific differences in the association of adherence and PGC. This underlines the need for efforts to improve glycaemic control in patients with type 2 diabetes mellitus with a particular focus on men. # 2012 Elsevier Ireland Ltd. All rights reserved.

1.

Introduction

In Germany, approximately 8% of the total population has type 2 diabetes mellitus (type 2 DM) [1]. Since type 2 DM is a chronic disease requiring long-term treatment and a high quality care in the ambulatory setting, efficient patient self-management is crucial [2]. An important aspect of patients’ self-management is

medication adherence, i.e. ‘the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen’ [3]. It was previously reported that medication adherence in diabetes patients ranges widely [4] depending on age [5], lower educational level and less affluent economic status [6], depression [7,8] and other cognitive/physical impairments as well as on regimen complexity [9], dosing frequency [10],

* Corresponding author. Tel.: +49 622142 1304; fax: +49 622142 1302. E-mail address: [email protected] (E. Raum). 1 Equally contributing first authors. 0168-8227/$ – see front matter # 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.diabres.2012.05.026

378

diabetes research and clinical practice 97 (2012) 377–384

fear of hypoglycaemia or of bothersome side effects [11]. Illness and medication beliefs are also of relevance [12]. Discordant results have been reported for gender as a risk factor for medication adherence [8,13,14]. A retrospective cohort study showed that adherence with medication for diabetes treatment decreases over time [15]. Overall, adherence has a strong association with an increased risk of medical complications, e.g. risk of severe cardiovascular diseases, and mortality as well as poor quality of life leading to major health care and economic implications [6,16]. Studies on non-adherence to prescribed medication in patients with type 2 DM in primary care are limited [6]. Only few studies have investigated the association between adherence and glycaemic control in general practice and studies focusing on gender-related differences in this context are still lacking [17]. We aimed to investigate the total and gender-specific prevalences of self-reported medication adherence and its associations with poor glycaemic control (PGC) (HbA1c  7.5%) in a large cohort of type 2 diabetes outpatients in Germany, taking into account a wide range of socio-demographic characteristics and lifestyle factors as well as diabetes- and health services-related factors.

2.

Materials and methods

2.1.

Study design and study population

This analysis is based on data from the baseline examination of the DIANA study (Type 2 Diabetes Mellitus: New Approaches to Optimize Medical Care in General Practice), an epidemiological prospective cohort study with type 2 DM patients conducted in the Ludwigsburg-Heilbronn area located in South-West Germany. The study was initiated in 2008 to address (short- and long-term) diabetes-related outcomes and to evaluate potentials for health care improvements in

patients with type 2 DM. Participants with physician diagnosed type 2 DM aged 18 and older were recruited according to a standardized protocol by 38 general practitioners (GPs) during regular practice visits between October 2008 and March 2010. Inclusion criteria beside type 2 DM were having sufficient knowledge of the German language and given written informed consent. We excluded nursing home residents as well as patients seen by the GPs for palliative or emergency care only. The study protocol was approved by the Ethics Committees of the medical faculty of the University of Heidelberg (reference S186/2008) and of the Chamber of Physicians of Baden-Wu¨rttemberg (reference B-2008-168). A total of 1146 eligible patients with physician diagnosed type 2 DM participating in the DIANA study completed a selfadministered standardized questionnaire at baseline. A blood sample was collected by the recruiting physicians and HbA1c was assessed by a central laboratory, using ion exchange high pressure liquid chromatography (G8, Tosoh Biosciences). Of the 1146 participants at baseline, 4 were excluded since there was no information on glycaemic control (HbA1c level). In total, data of 1142 participants were analysed.

2.2.

Definition of key variables

Medication non-adherence to all medications was assessed using the 4-item self-report medication adherence questionnaire developed by Morisky et al. [18], which is a wellestablished and frequently used instrument in large-scale studies in primary care using close-ended questions with yes or no responses (questions are displayed in Table 1). The sum score was calculated ranging from 0 (full adherence) to 4 (poor adherence). Patients were grouped either being adherent (zero points) or non-adherent (1–4 points). Classification of glycaemic control was based on baseline HbA1c levels defining 7.5% as poor glycaemic control (PGC) according to the International Diabetes Federation guideline [19].

Table 1 – Description of the study population.

Age Mean (standard deviation) Age in years 59 60–69 70–79 80 Number of years of school education 9 10–12 13 Marital status Single/widowed/divorced Married Occupational status Employed Retired Housewife Other

Total

Men (n = 624, 54.6%)

n (%)

n (%)

n (%)

68.3 (10.3)

67.2 (10.1)

69.7 (10.4)

220 321 469 132

(19.3) (28.1) (41.1) (11.6)

137 193 239 55

(22.0) (30.9) (38.3) (8.8)

Women (n = 518, 45.4%)

83 128 230 77

(16.0) (24.7) (44.4) (14.9)

825 (73.6) 197 (17.6) 99 (8.8)

433 (70.9) 104 (17.0) 74 (12.1)

392 (76.9) 93 (18.2) 25 (4.9)

320 (28.2) 816 (71.8)

112 (18.1) 508 (81.9)

208 (40.3) 308 (59.7)

235 720 79 58

158 403 0 36

77 317 79 22

(21.5) (65.9) (7.2) (5.3)

(26.5) (67.5) (0) (6.0)

(15.6) (64.0) (16.0) (4.4)

379

diabetes research and clinical practice 97 (2012) 377–384

Table 1 (Continued ) Total

Men (n = 624, 54.6%)

n (%)

n (%)

Smoking 546 (48.0) Never 454 (40.0) Ex-smoker Current smoker 138 (12.1) Alcohol consumption 421 (37.9) Abstainer Body mass index (kg/m2) <25 154 (13.5) 440 (38.6) 25–<30 354 (31.0) 30–<35 35 193 (16.9) Hours of total physical activity per week 232 (20.3) <5 326 (28.6) 5–<10 10–<23 274 (24.0) 288 (25.2) 23 Time since diagnosis of diabetes (self-reported) 334 (29.3) 5 years 6–10 years 330 (28.9) 226 (19.8) 11–15 years 252 (22.1) 16 years Disease management program (DMP) participation 865 (80.4) Yes Diabetes education training (1) 659 (57.7) Yes Appointments with the general practitioner (GP) (last 3 months) 454 (38.8) 1 460 (40.4) 2–3 226 (19.8) 4 Number of current medications 300 (26.3) 3 267 (23.4) 4–5 6–8 326 (28.6) 249 (21.8) 9 History of depression (self-reported) 146 (12.8) Yes Health status 102 (8.9) Excellent or very good 655 (57.4) Good 344 (30.2) Less good Poor 40 (3.5) Medication non-adherencea Yes 261 (23.8) Poor glycaemic control (HbA1c  7.5%) Yes 237 (20.8)

Women (n = 518, 45.4%) n (%)

174 (28.0) 355 (57.2) 92 (14.8)

372 (72.0) 99 (19.2) 46 (8.9)

139 (22.3)

282 (54.4)

76 255 194 98

(12.2) (40.9) (31.1) (15.7)

78 185 160 95

(15.1) (35.7) (30.9) (18.3)

121 164 164 160

(19.4) (26.3) (26.3) (25.2)

111 162 110 128

(21.4) (31.1) (21.2) (24.7)

174 196 121 133

(27.9) (31.4) (19.4) (21.3)

160 134 105 119

(30.9) (25.9) (20.3) (23.0)

471 (79.4)

394 (81.6)

360 (57.7)

299 (57.7)

260 (41.7) 245 (39.3) 119 (19.1)

194 (37.6) 215 (41.7) 107 (20.7)

182 143 177 122

118 124 149 127

(29.2) (22.9) (28.4) (19.6)

65 (10.4) 62 366 174 21

(10.0) (58.6) (27.9) (3.4)

(22.8) (23.9) (28.8) (24.5)

81 (15.7) 40 289 170 19

(7.7) (55.8) (32.8) (3.7)

147 (24.4)

114 (23.0)

144 (23.1)

93 (18.0)

Number may not always add up to total because of missing values for some items. a Morisky score questionnaire: 1. Do you ever forget to take your medicine? 2. Are you careless at times about taking your medicine? 3. When you feel better do you sometimes stop taking your medicine? 4. Sometimes if you feel worse when you take the medicine, do you stop taking it?

Participants reported socio-demographic characteristics and lifestyle factors including education level (9 years, 10–12 years, 13 years of school education), marital status (single, married), occupational status (employed, retired, housewife, other), smoking history (never, ex-smoker, current smoker), alcohol consumption (abstainer, other), height and weight. Latter were used to calculate body mass index (BMI) in kg/m2 (<25 kg/m2, 25 to <30 kg/m2, 30 to <35 kg/m2, 35 kg/m2). Physical activity was explored by the International Physical Activity Questionnaire (IPAQ) for hours of walking, moderateintensity and vigorous-intensity physical activity and hours of sitting per day [20]. For this analysis, patients were classified

according to hours of total physical activity per week (<5, 5 to <10, 10 to <23, 23). Due to high proportions of ‘‘don’t know’’ answers on the hours of the different types of physical activity (ranging from 5.3% to 16.8%), the median level for the respective category was imputed in those cases. Time since type 2 DM was categorized into the following categories: 5 years, 6–10 years, 11–15 years and 16 years. Participation in a disease management program for type 2 DM was reported by the recruiting GPs. Participation in type 2 DM education training was asked for in the participants’ questionnaire by the question: ‘Have you ever taken part in diabetes education training?’ (yes/no). Other type 2 DM-related aspects, such as

380

diabetes research and clinical practice 97 (2012) 377–384

number of visits to the GP during the last 3 months, time since diagnosis, total number of medications were documented by the participants. Medications (prescribed or over-the-counter medication) were asked for with the question ‘Which medications do you take on a daily basis?’ (name and/or pharmaceutical identification number) ‘How frequently do you take them?’ History of physician diagnosed depression (yes/no) was self reported by the participants. For the assessment of the general health status the first question of the short-form-12 (SF-12) questionnaire (‘How would you describe your general health status?’) with a scale ranging from ‘poor’, ‘less good’ and ‘good’ to ‘very good’ and ‘excellent’ was used [21].

2.3.

Statistical analysis

We first described the study population according to sociodemographic factors and other covariates considered in this analysis. Next, prevalence of medication non-adherence and PGC were determined for the total study population and separately for gender. Third, prevalences of medication nonadherence and PGC were analysed by levels of covariates, and bivariate associations were tested for statistical significance by x2-tests. Finally, overall and gender-specific associations of medication non-adherence with PGC were assessed by using multiple log-binomial regression, with PGC as the dependent variable and medication non-adherence and the aforementioned covariates as independent variables. All patients were included in this analysis independent of their medication. Covariates were included if they were related to PGC in bivariate analyses ( p < 0.1). Furthermore, effect modification by gender was assessed using Breslow-Day test. For statistical testing, a two-sided alpha level of 5% was applied, and the SAS software (version 9.2, SAS Institute, Cary, N.C., USA) was used.

3.

Results

Overall, 624 men (54.6%) and 518 women (45.4%) participated in this study (Table 1). Mean age (SD) was 68.3 (10.3 years) (median age: 70 years). Men were on average younger than women (men: 67.2  10.1 years; women: 69.7  10.4 years). The vast majority of patients had completed up to 9 years of school education (73.6%) and were married (71.8%). Generally, more men than women were still married (men: 81.9%; women: 59.7%). Due to the high mean age only about one fourth of the participants were still employed. Current smoking was reported by 14.8% and 8.9% of male and female patients, respectively. Almost half of the participants were obese (47.9%) with a BMI  30 kg/m2; 15.7% of the men and 18.3% of the women had a BMI  35 kg/m2. Physical activity <5 h and 23 h per week, respectively, was reported by 20.3% and 25.2% of the participants. Overall, 29.3% of the participants were diagnosed with diabetes within 5 years before recruitment and 22.1% more than 15 years before recruitment. Four out of five patients (80.4%) were enrolled in disease management program, and the majority of both men and women (57.7%) reported that they had diabetes education training. Appointments with the GP during the last 3 months were generally common: 40.4% of the participants reported 2–3 appointments and 19.8%

reported even more than 4 appointments. Nearly all participants (97.1%) were treated with anti-diabetic medications (not shown in the table) and about half of the participants reported taking 6 or more prescribed or over-the-counter medications per day. Almost one of four women reported taking 9 or more medications compared to one of five men. A history of physician diagnosed depression was reported by 12.8% of the participants and was more common among women (15.7%). Two thirds of the participants reported either an excellent or very good to good general health status, with small disparities between genders. Medication non-adherence was found in 23.8% of all participants, with only marginal differences between men and women (24.4% and 23.3%, respectively). The HbA1c values ranged from 4.7% to 12.2% with an arithmetic mean of 6.9% (SD  1.1%). In total, 20.8% showed PGC (HbA1c  7.5%). Significantly more men than women had HbA1c levels 7.5% (men: 23.1%; women: 18.0%) ( p = 0.03). Table 2 shows the results of bivariate analyses of the covariates with medication non-adherence and with PGC, respectively: Medication non-adherence was more common among younger patients with a lower school education and those still employed, and was also associated with being unmarried and less physically active, having had diabetes education training, taking less medication and reporting a history of depression. Except for physical activity, numbers of medications and depression, the same factors were also associated with PGC. PGC was also more common among smokers than among nonsmokers, among patients with a BMI  25 kg/m2 and those with a long duration of diabetes. As shown in Table 3, 29.5% of non-adherent patients also had poor glycaemic control compared to only 18.4% of adherent patients (prevalence ratio (PR) = 1.60; 95%-CI: 1.27– 2.03). Non-adherent men had a twofold higher risk to have PGC (37.4%) compared to adherent men (18.7%, crude PR = 2.00; 95%-CI: 1.51–2.66), whereas adherent and non adherent women (18.1% and 19.3%, respectively) did not differ in regard to PGC prevalence (crude PR = 1.07; 95%-CI: 0.69–1.65). After adjustment for the covariates the strength of the association between PGC and patient self-reported medication nonadherence was slightly attenuated, but it remained statistically significant for the entire study population (fully adjusted PR = 1.54; 95%-CI: 1.24–1.93) as well as for men (PR = 1.90; 95%CI: 1.46–2.49), while in women there was no significant association (PR = 0.97; 95%-CI: 0.55–1.46). The interaction of the association between medication non-adherence and PGC by gender was highly statistically significant ( p = 0.01).

4.

Discussion

In this cohort of diabetes patients recruited in the primary care setting in Germany, self-reported medication non-adherence was present in 23.8% of patients with no gender-specific differences, which is in accordance with prevalences reported by two reviews [22,23]. Almost 21% of the patients had PGC defined as HbA1c  7.5%, with a higher prevalence among men than among women. Both medication non-adherence and PGC were more common among younger patients and those still

381

diabetes research and clinical practice 97 (2012) 377–384

Table 2 – Self-reported medication non-adherence and poor glycaemic control according to covariates: results of bivariate analysis. Variables of interest

n

Non-adherencea n (%)

Age in years 220 77 (37.4) 59 60–69 321 73 (23.9) 469 87 (18.9) 70–79 132 24 (19.2) 80 Number of years of school education 825 175 (22.1) 9 197 50 (25.9) 10–12 99 29 (31.2) 13 Marital status 320 85 (27.6) Single/widowed/divorced 816 173 (22.0) Married Occupational status 235 71 (32.6) Employed 720 134 (19.1) Retired 79 18 (23.4) Housewife 58 22 (40.0) Other Smoking Never 546 115 (22.0) 454 104 (23.6) Ex-smoker 138 41 (31.5) Current smoker Alcohol consumption 421 93 (23.0) Abstainer 721 168 (24.2) Other Body mass index (kg/m2) <25 154 32 (21.9) 440 94 (22.3) 25–<30 354 76 (22.4) 30–<35 193 59 (30.9) 35 Hours of total physical activity per week 232 61 (27.1) <5 326 90 (28.5) 5–<10 274 53 (19.9) 10–<23 288 52 (19.2) 23 Time since diagnosis of diabetes (self-reported) 334 80 (25.8) 5 years 6–10 years 330 79 (25.1) 226 47 (21.2) 11–15 years 252 55 (21.9) 16 years Disease management program (DMP) participation 865 192 (23.0) Yes 211 51 (25.9) No Diabetes education training (1) Yes 659 172 (26.6) 483 89 (19.7) No Appointments with the general practitioner (GP) (last 3 months) 454 104 (24.1) 1 2–3 460 105 (23.8) 226 50 (22.4) 4 Number of current medications 300 81 (30.8) 0–3 4–5 267 56 (21.1) 326 65 (20.1) 6–8 249 59 (23.9) 9 Depression (self-reported) 146 51 (35.7) Yes 995 210 (22.0) No Health status Excellent or very good 102 17 (17.5) 655 140 (22.4) Good 344 94 (28.0) Less good 40 10 (25.0) Poor a

2

Poor glycaemic controla

P-value

<0.0001

n (%) 68 68 75 26

(30.9) (21.2) (16.0) (19.7)

p-value

0.0001

0.10

158 (19.1) 53 (26.9) 25 (25.2)

0.03

0.05

86 (26.9) 151 (18.5)

0.001

<0.0001

72 128 12 18

(30.6) (17.8) (15.2) (31.0)

<0.0001

0.07

96 (17.6) 103 (22.7) 37 (26.8)

0.02

0.63

97 (23.0) 140 (19.4)

0.14

0.09

19 71 81 66

(12.3) (16.1) (22.9) (34.2)

<0.0001

0.03

43 74 54 60

(18.5) (22.7) (19.7) (20.8)

0.70

0.51

44 67 52 74

(13.2) (20.3) (23.0) (29.4)

0.0001

0.39

184 (21.3) 43 (20.4)

0.78

0.009

155 (23.5) 82 (17.0)

0.01

0.89

96 (21.2) 88 (19.1) 53 (23.5)

0.41

0.01

55 60 66 56

(18.3) (22.6) (20.3) (22.5)

0.54

0.0003

36 (24.7) 200 (20.1)

0.20

0.11

19 129 77 11

0.50

X -test. Number may not always add up to total because of missing values for some items.

(18.6) (19.7) (22.4) (27.5)

382

diabetes research and clinical practice 97 (2012) 377–384

Table 3 – Association between self-reported medication non-adherence and poor glycaemic control. PGCb

All Non-adherence

Men Non-adherence

Women Non-adherence a b c

PR (95%-CI)c

n (%)

Crude

Adjusted for age and sex

Adjusted for multiple covariatesa

No Yes

154 (18.4) 77 (29.5)

Ref. 1.60 (1.27–2.03)

1.50 (1.18–1.90)

1.54 (1.24–1.93)

No Yes

85 (18.7) 55 (37.4)

Ref. 2.00 (1.51–2.66)

1.77 (1.33–2.37)

1.90 (1.46–2.49)

No Yes

69 (18.1) 22 (19.3)

Ref. 1.07 (0.69–1.65)

1.04 (0.67–1.61)

0.97 (0.65–1.46)

Adjusted for sex, marital status, body mass index, time since diabetes diagnosis and diabetes education training. PGC = poor glycaemic control. PR = prevalence ratio; CI = confidence interval.

employed or not being married. Medication non-adherence was also strongly related to a history of depression, whereas PGC seemed to be more influenced by physiologic and diabetes-related factors, such as BMI or diabetes duration. Long term control of blood glucose levels which is routinely measured by HbA1c is one of the most important targets of management of patients with diabetes mellitus. Good blood glucose control can be achieved when patients follow their diet, take their medication properly and exercise. Consequently, non-adherence to treatment regimens likely results in suboptimal blood glucose control. We focused on self-reported adherence because patients are normally able to report, if they take their medication or not and therefore self-report is potentially the most accurate record of what a given patient has done, especially when inquired not in the doctors office or by the GPs but in an neutral setting and with the ascertainment of anonymity given with a self-administered questionnaire. To our knowledge, this is the first study in a general practice setting on the association of self-reported medication non-adherence with PGC that adjusted for a variety of patientrelated confounders and stratified for gender. We found an overall 1.55 higher risk for PGC in non-adherent compared to adherent patients and a significant effect-modification by gender: There was no association between non-adherence and PGC for female patients, whereas non-adherent male patients had a twofold higher risk for PGC compared to adherent male patients. Although the number of studies on the association of medication adherence with glycaemic control in type 2 DM patients is still very limited, reported proportions of adherence varied widely [4,9]. Much of this variation can be explained either by differences in the measurement of medication adherence or by study design. However, prevalence of poor adherence to medication in patients with type 2 DM in our study is similar to the prevalence found in previous studies conducted in managed care settings [24,25]. We focused on the overall medication adherence because patients with type 2 DM frequently have additional significant comorbidities, such as hypertension, coronary heart disease or chronic heart failure, which are also generally treated by oral medications, and adherence to other medications may often interfere with adherence to diabetes medication. In

addition, beside the good glycaemic control the optimal management of diabetes-related comorbidity is of importance as large proportion of patients with diabetes suffers a fatal cardiovascular event [26]. Unexpectedly and in contrast to a recent analysis of a large retrospective prescription database of type 2 DM patients in the USA [27], we found a slightly u-shaped relationship between total number of medications and non-adherence. This might be explained by the cross-sectional design of our study and reflecting the fact that GPs might have reduced the number of medications (using rather fixed combinations than two or more mono-therapies) due to known or suspected nonadherent behaviour of their patients. Our finding of higher PGC rates among younger than older patients confirm observations of a previous study conducted in the same area approximately 9 years earlier [28]. The observations of increase in adherence with age and a positive association between non-adherence and age are furthermore consistent with results of a retrospective study in Oregon, USA, analysing medical and pharmacy claims from a managed care plan [29]. We found married patients to be less often non-adherent and to have less often PGC than single or widowed patients. This finding is in line with observations of lower mortality of married compared to unmarried older adults [30]. Our finding of a higher prevalence of PGC among current or former smoking patients with type 2 DM are in line with results of the Western New York Health study showing that current smoking was associated with a negative influence of smoking on fasting glucose level [31]. Similar to the DETECT (Diabetes Cardiovascular Risk Evaluation: Targets and Essential Data for Commitment of Treatment) study from Germany [32] we found an association between history of depression and non-adherence of type 2 diabetes-patients, although adherence was – in contrast to our study – only assessed with one question. The DETECT study using a HbA1c-level of 7% as definition for poor glycaemic control also found a twofold higher risk for poor glycaemic control for patients with depression. In our study more patients (24.7%) with depression had poor glycaemic control (defined as a HbA1c-level of 7.5%) than patients without depression (20.1%), but this difference was not statistically significant.

diabetes research and clinical practice 97 (2012) 377–384

Long-term improvements in glycaemic control by adequate use of diabetes medication as well as participation in a longterm disease management program have been inconsistent regarding the effect of structured patient education program [33]. In our study participation in a German disease management program for diabetes was likewise not related to medication adherence or glycaemic control. Our bivariate results on diabetes education training showing that participants who had already attended at least one diabetes education training reported medication nonadherence and had PGC more often than those who had never attended such training seem surprising and counterintuitive on first view. A possible explanation for these cross-sectional, crude associations might be reverse causality: Most likely, diabetes education training was more often recommended, prescribed and used by patients having difficulties with medication adherence and glycaemic control. In addition, duration of diabetes and its character of progression (fast, medium, slowly) might also be potential confounders in this scenario. Specific strength and potential limitations of this study should be recognized. The major strength of this study is that the study population constituted of a relatively large sample of elderly type 2 DM patients in primary health care of rural and urban areas in Germany. The HbA1c level of each participant was evaluated by a single central laboratory. Poor glycaemic control was defined by HbA1c level of equal or larger than 7.5% according to the International Diabetes Federation guideline. We assessed adherence to all prescribed medications, since type 2 DM is a complex disease that often goes along with major comorbidity. Regarding potential limitations, information on socio-demographic, partly medical history, lifestyle and diabetes- and health services-related factors relied on patient self-reports only. Also the information on medication based on patient self-report may be less precise than, for example, medication prescription or dispensing data or physician information The Morisky questionnaire is often used because it is brief, inexpensive and applicable in various settings (as in ours), but also certain weaknesses are attributed to the questionnaire, such as inadequate reliability, poor distributional properties, proneness for recall and response bias due social desirability or inaccurate memory, leading to an overestimation of adherence. Due to financial and organisational limitations we were not able to use other adherence measurements like ascertainment of the medication or its metabolites in the blood samples or electronic drug monitoring devices, pill counts or pharmacy refills. Two reviews [34,35] reported good concordance between selfreported adherence questionnaires and objective measurements of adherence, so the Morisky questionnaire seemed to be an appropriate and feasible way to measure adherence in our study population, although some information may have been lost when categorizing the patients into only two groups (adherent and non-adherent). Non-adherence may have been underreported to some extent by the patients due to social desirability. Finally, the results presented here were based on a crosssectional analysis, which does not allow firm quantitative conclusions to be drawn with respect to the temporal and causal sequence of the associated factors. Generalisability

383

may also be limited by the fact that only patients of a defined area of South-West Germany were included. In conclusion, PGC and medication non-adherence require more attention in the management of type 2 DM. Our results show important differences in PGC of both men and women. Men appear to be particularly prone to PGC due to medication non-adherence, and efforts to improve glycaemic control are needed in men in particular. Good self-management including tight metabolic control – measured by glycosylated haemoglobin (HbA1c) during regular physician visits – is of particular relevance for both, optimal treatment and (long-term) prognosis, and increased medication adherence is associated with a reduction of HbA1c levels and a decrease of all-cause hospitalization and all-cause mortality [16]. Further research should aim to investigate how medication adherence and glycaemic control can be improved under conditions of routine clinical care, paying particular attention to potential gender-specific differences.

Conflict of interest The authors declare that they have no conflict of interest.

Grant support This study was supported by a grant of the Federal Ministry of Education and Research (study identification number: 01GX0746). Results were presented as poster ‘‘Medication non-adherence and poor glycaemic control in patients with Type 2 diabetes mellitus: First results of the DIANA-Study’’ (in German) during the annual meeting of the German College of General Practitioners and Family Physicians from 22 to 24 September 2011 in Salzburg, Austria.

Acknowledgements We wish to thank all the participants of the DIANA study and our cooperating study partners. The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

references

[1] Hauner H, Hanisch J, Bramlage P, Steinhagen-Thiessen E, Schunkert H, Jo¨ckel KH, et al. Prevalence of undiagnosed Type-2-diabetes mellitus and impaired fasting glucose in German primary care: data from the German Metabolic and Cardiovascular Risk Project (GEMCAS). Exp Clin Endocrinol Diabetes 2008;116:18–25. [2] King DK, Glasgow RE, Toobert DJ, Strycker LA, Estabrooks PA, Osuna D, et al. Self-efficacy, problem solving, and social-environmental support are associated with diabetes self-management behaviors. Diabetes Care 2010;33:751–3.

384

diabetes research and clinical practice 97 (2012) 377–384

[3] Cramer JA, Roy A, Burrell A, Fairchild CJ, Fuldeore MJ, Ollendorf DA, et al. Medication compliance and persistence: terminology and definitions. Value Health 2008;11:44–7. [4] Cohen HW, Shmukler C, Ullman R, Rivera CM, Walker EA. Measurements of medication adherence in diabetic patients with poorly controlled HbA(1c). Diabet Med 2010;27:210–6. [5] Bezie Y, Molina M, Hernandez N, Batista R, Niang S, Huet D. Therapeutic compliance: a prospective analysis of various factors involved in the adherence rate in type 2 diabetes. Diabetes Metab 2006;32:611–6. [6] Odegard PS, Capoccia K. Medication taking and diabetes: a systematic review of the literature. Diabetes Educ 2007;33:1014–29. [7] Lin EH, Katon W, Von Korff M, Rutter C, Simon GE, Oliver M, et al. Relationship of depression and diabetes self-care, medication adherence, and preventive care. Diabetes Care 2004;27:2154–60. [8] Nau DP, Aikens JE, Pacholski AM. Effects of gender and depression on oral medication adherence in persons with type 2 diabetes mellitus. Gend Med 2007;4:205–13. [9] Corsonello A, Pedone C, Lattanzio F, Lucchetti M, Garasto S, Carbone C, et al. Regimen complexity and medication nonadherence in elderly patients. Ther Clin Risk Manag 2009;5:209–16. [10] Odegard PS, Gray SL. Barriers to medication adherence in poorly controlled diabetes mellitus. Diabetes Educ 2008;34:692–7. [11] Grant RW, Devita NG, Singer DE, Meigs JB. Polypharmacy and medication adherence in patients with type 2 diabetes. Diabetes Care 2003;26:1408–12. [12] Mahler C, Hermann K, Horne R, Jank S, Haefeli WE, Szecsenyi J. Patients’ beliefs about medicines in a primary care setting in Germany. J Eval Clin Pract 2012;18:409–13. [13] Geisel-Marbaise S, Stummer H. Diabetes adherence – Does gender matter? J Public Health 2010;18:219–26. [14] Rivellese AA, Riccardi G, Vaccaro O. Cardiovascular risk in women with diabetes. Nutr Metab Cardiovasc Dis 2010;20:474–80. [15] Hertz RP, Unger AN, Lustik MB. Adherence with pharmacotherapy for type 2 diabetes: a retrospective cohort study of adults with employer-sponsored health insurance. Clin Ther 2005;27:1064–73. [16] Ho PM, Rumsfeld JS, Masoudi FA, McClure DL, Plomondon ME, Steiner JF, et al. Effect of medication nonadherence on hospitalization and mortality among patients with diabetes mellitus. Arch Intern Med 2006;166:1836–41. [17] Alvarez Guisasola F, Tofe Povedano S, Krishnarajah G, Lyu R, Mavros P, Yin D. Hypoglycaemic symptoms, treatment satisfaction, adherence and their associations with glycaemic goal in patients with type 2 diabetes mellitus: findings from the Real-Life Effectiveness and Care Patterns of Diabetes Management (RECAP-DM) Study. Diabetes Obes Metab 2008;10(Suppl. 1):25–32. [18] Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care 1986;24:67–74. [19] International Diabetes Federation. Clinical Guidelines Task Force Global guideline for Type-2 Diabetes. Brussels: International Diabetes Federation; 2005. [20] Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity

[21]

[22] [23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35:1381–95. Ware Jr J, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:220–33. Rubin RR. Adherence to pharmacologic therapy in patients with type 2 diabeters mellitus. Am J Med 2005;118:27S–34S. Lerman I. Adherence to treatment: the key for avoiding long-term complications of diabetes. Arch Med Res 2005;36:300–6. Grant R, Adams AS, Trinacty CM, Zhang F, Kleinman K, Soumerai SB, et al. Relationship between patient medication adherence and subsequent clinical inertia in type 2 diabetes glycemic management. Diabetes Care 2007;30:807–12. Schmittdiel JA, Uratsu CS, Karter AJ, Heisler M, Subramanian U, Mangione CM, et al. Why don’t diabetes patients achieve recommended risk factor targets? Poor adherence versus lack of treatment intensification. J Gen Intern Med 2008;23:588–94. Sarwar N, Gao P, Seshasai SR, Gobin R, Kaptoge S, Di Angelantonio E. Emerging Risk Factors Collaboration Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet 2010;375:2215–22. Thayer S, Arondekar B, Harley C, Darkow TE. Adherence to a fixed-dose combination of rosiglitazone/glimepiride in subjects switching from monotherapy or dual therapy with a thiazolidinedione and/or a sulfonylurea. Ann Pharmacother 2010;44:791–9. Rothenbacher D, Ru¨ter G, Saam S, Brenner H. Younger patients with type 2 diabetes need better glycaemic control: results of a community-based study describing factors associated with a high HbA1c value. Br J Gen Pract 2003;53:389–91. Rozenfeld Y, Hunt JS, Plauschinat C, Wong KS. Oral antidiabetic medication adherence and glycemic control in managed care. Am J Manag Care 2008;14:71–5. Johnson NJ, Backlund E, Sorlie PD, Loveless CA. Marital status and mortality: the national longitudinal mortality study. Ann Epidemiol 2000;10:224–38. Rafalson L, Donahue RP, Dmochowski J, Rejman K, Dorn J, Trevisan M. Cigarette smoking is associated with conversion from normoglycemia to impaired fasting glucose: the Western New York Health Study. Ann Epidemiol 2009;19:365–71. Dirmaier J, Watzke B, Koch U, Schulz H, Lehnert H, Pieper L, et al. Diabetes in primary care: prospective associations between depression, nonadherence and glycemic control. Psychother Psychosom 2010;79:172–8. Deakin T, McShane CE, Cade JE, Williams RD. Group based training for self-management strategies in people with type 2 diabetes mellitus. Cochrane Database Syst Rev 2005;2:CD003417. Garber MC, Nau DP, Erickson SR, Aikens JE, Lawrence JB. The concordance of self-report with other measures of medication adherence. Med Care 2004;42:649–52. Shi L, Liu J, Koleva Y, Fonseca V, Kalsekar A, Pawaskar M. Concordance of adherence measurement using selfreported adherence questionnaires and medication monitoring devices. Pharmacoeconomics 2010;28:1097–107.