International Journal of Diabetes Mellitus
Volume 2, Issue 2 , Pages 122-124, August 2010

Medication adherence in low income elderly type-2 diabetes patients: A retrospective cohort study

  • I. Patel

      Affiliations

    • Clinical, Social and Administrative Sciences, College of Pharmacy, University of Michigan at Ann Arbor, 428 Church Street, Ann Arbor, MI 48109-1065, USA
    • Corresponding Author InformationCorresponding author. Tel.: +1 734 936 1505; fax: +1 734 615 8171.
    • Both the authors have contributed equally to the project.
    • ,
  • J. Chang

      Affiliations

    • Clinical, Social and Administrative Sciences, College of Pharmacy, University of Michigan at Ann Arbor, 428 Church Street, Ann Arbor, MI 48109-1065, USA
    • Both the authors have contributed equally to the project.
    • Tel.: +1 734 936 1505; fax: +1 734 615 8171.
    • ,
  • R.A. Shenolikar

      Affiliations

    • Health Management Innovations, GlaxoSmithKline, 5 Moore Drive, Durham, NC 27709, USA
    • Tel.: +1 919 483 8980; fax: +1 989 483 0611.
    • ,
  • R. Balkrishnan

      Affiliations

    • Center for Medication Use, Policy, and Economics, The University of Michigan, 428 Church Street, Ann Arbor, MI 48109-1065, USA
    • Department of Clinical, Social and Administrative Sciences, The University of Michigan, 428 Church Street, Ann Arbor, MI 48109-1065, USA
    • Department of Health Management and Policy, The University of Michigan, 428 Church Street, Ann Arbor, MI 48109-1065, USA
    • Tel.: +1 734 764 7203; fax: +1 734 615 8171.

Received 8 April 2010; accepted 8 May 2010. published online 21 June 2010.

Article Outline

Abstract 

The study objective was to determine the age associated medication adherence among low income type-2 diabetes patients enrolled in Medicaid. This was a retrospective cohort study consisting of patients aged 18–44years (n=681), 45–64years (n=2327) and 65+ years (n=161), respectively newly starting antidiabetic medication between July 2001 and June 2002. Medication adherence was measured as medication possession ratio using prescription refill patterns. Multiple regression analyzes showed that compared to age group 18–44years, age groups 65+ and 44–64years had significantly higher adherence rate by 13.4% and 12.5%, respectively. Better oral antidiabetic medication adherence was associated with increased age.

Keywords: Type-2 diabetes, Medicaid, Metformin, Sulphonylurea, Tzd, Drug compliance

 

Back to Article Outline

1. Introduction 

Diabetes is one of the major reasons for mortality and morbidity in the U.S. In 2007, about 23.6 million people were suffering from diabetes, making it the seventh cause of death nationally. Direct and indirect medical costs estimated with diabetes were about $116 and $58 billion, respectively [1], [2], [3]. It is a metabolic disorder, and the root cause of many micro vascular and macro vascular complications [4]. Studies have shown that an increase in the number of medications might present hurdles in adhering to OHDs in case of patients who are elderly, socially deprived and have many co-morbidities [4], [5]. This paper examines the effect of starting a new medication therapy on medication adherence of patients suffering from type-2 diabetes.

Back to Article Outline

2. Materials and methods 

2.1. Study subjects 

This was a retrospective cohort study, which used data of continuously eligible type-2 diabetes enrollees of the North Carolina Medicaid program from July 1, 2000 to June 30, 2000 across age groups 18–44years, 45–64years and 65+ years. Patients were identified using at least 1 or more ICD-9 code (International Classification of Diseases, Ninth Revision (ICD-9), Clinical Modification) for type-2 diabetes (250.xx) and one NDC code (National Drug Code) for antidiabetic medication. Patients on insulin therapy alone and <18years were excluded. Patients using monotherapy and not having used any of the oral anti diabetic drugs like metformin or sulphonylurea or TZD in the previous year were included [3], [6].

2.2. Adherence 

The medication possession ratio (MPR) based on the Med-Total approach by Steiner et al. was calculated as the days of antidiabetic prescription supply dispensed divided by the number of days between these prescription refills [7]. The observation period began with the first date of dispensing within each year and ended as the dispensing date of the last prescription [3], [6].

2.3. Sensitivity analyses 

The natural logarithm of Medication possession ratio (MPR) was the outcome variable, as the distribution of MPR was found to be skewed (as determined by the Shapiro–Wilk test) [3], [6].

∗Outcome [ln(MPR)]=f (age indicator, treatment group indicator, demographic and clinical confounders).

All the data analysis was conducted using the STATA software (StataCorp LP, Texas).

Back to Article Outline

3. Results 

Table 1 gives the descriptive characteristics of the study population across different age groups. In the case of medication therapy, metformin and sulphonylurea were used more by the age group 18–44years, whereas TZD was used more by the age group 65+ years. Medication adherence was highest for the age group 65+ years. People in the age group of 45–64years had higher healthcare costs in year 1, and more co-morbidities.

Table 1. Descriptive characteristics of the study population (comparison of age groups 18–44years versus 45–64years versus 65+ years).
VariableAge group: 18–44years (n=681) mean (standard deviation)Age group: 45–64years (n=2327) mean (standard deviation)Age group: 65+ years (n=161) mean (standard deviation)
Baseline characteristics male gender (%)23.4926.6424.22
Whites (%)35.0935.8833.54
African Americans(%)55.8046.0646.58
Hispanics (%)0.730.510.62
Others (%)8.3717.5319.25
Metformin (%)7.636.665.59
Sulphonylurea (%)42.8736.2227.32
TZD (%)49.4857.1167.08
Total healthcare costs in year 1 ($)8588.54 (15015.29)9480.18 (14174.72)7891.39 (11176.03)
No of co morbidities1.35 (1.49)2.48 (1.91)2.35 (1.93)
No of physician visits26.51 (21.15)28.03 (21.80)29.39 (24.75)
Study outcome adherence rate to new medication (year 2)0.46 (0.3)0.58 (0.32)0.59 (0.33)

Adherence rate of 0.58 for age group 45–64years is same as adherence rate of 58%.

Table 2 gives an OLS multiple log-linear regression was conducted to measure the association between medication adherence and patients’ information. Compared with age group 18–44, the age group 45–64 and over 65 had higher adherence by 12.5% and 13.4%, respectively. Metformin and Tzd users had almost 34.5% lower and 2.4% higher medication adherence, as compared to sulfonylurea users (p<0.05), respectively. Each additional co-morbidity and physician visit was associated with about 1.3% and 0.1% decrease in adherence (p<0.05), respectively.

Table 2. Comparison of adherence rate between different age groups using multiple regression analysis.
Predictor variableMedication possession ratio (natural log) estimated coefficient (standard error)95% Confidence interval of estimated coefficient
Age group: 45–64years0.125 (0.009)a[0.106–0.143]
Age group: 65+ years0.134 (0.018)a[0.096–0.171]
Males0.02 (0.008)a[0.003 to −0.038]
African Americans−0.051 (0.008)a[−0.067 to −0.034]
Hispanics0.111 (0.041)a[−0.192 to −0.029]
Others−0.046 (0.011)a[−0.069–0.023]
High healthcare costs in year 1 (>$16,423)0.042 (0.009)a[0.023–0.061]
TZD0.024 (0.008)a[0.008–0.04]
Metformin−0.345 (0.01)a[−0.366 to −0.324]
Number of co morbidities−0.013 (0.002)a[−0.018 to −0.009]
Number of physician visits−0.001 (0.0002)a[−0.001 to −0.0007]
Constant0.552 (0.012)[0.527–0.576]

Reference group: female gender, white race, sulfonylureas, <$16,423 costs in year 1.

aIndicates significance at the 0.05 level for a t test examining whether beta=0.

Back to Article Outline

4. Discussion 

The study assessed the medication adherence in the North Carolina Medicaid program participants aged 18–65+ years suffering from type-2 diabetes. When drug usage was examined across age groups, 65+ patients showed maximum usage of TZD since these people might have more severe form of the disease compared to the other age groups. TZDs are used only when the disease progresses to a more severe stage and sulphonylurea alone cannot manage the disease. Patients aged 18–44years showed a maximum use of sulphonylurea and metformin. It may be that most of them could no longer control their diabetes with exercise and diet alone, and had recently switched to medications, thereby using sulphonylurea for treatment in the initial stages. Also, if patients were obese then metformin might be favorable since research has shown that it works best in obese patients aged below 60years. Also, metformin is a comparatively cheaper drug, and might lead to weight loss [8], [9]. The majority of patients in all age groups were African Americans. Previous studies have shown that African Americans have a lower SES compared to whites, hence it is justified that they are mostly enrolled in Medicaid [3], [6].

In multiple regression analysis, a significant association was found between age, medication therapy and medication adherence. Comparatively, superior adherence to TZDs was seen in patients aged 18 and above due to a high risk for complications arising out of their poverty status [10].

Due to the retrospective nature of the study, actual medication intake was not observed. Medication adherence was used as a proxy to measure medication taking behavior. Prescription refilled was assumed as prescription taken. Medication switching and combination were not captured by the medication possession ratio. Procured medication records at the time of discharge from hospitals were not covered in the study. The observational nature of the study does not permit causal inferences. There was limited generalizability due to the selection of a pre defined group. Information on Hba1c level, Body Mass index, duration of diabetes and therapy related factors like dosing, complexity of regimen and desire for quicker results could not be obtained in the study [9], [12]. Total healthcare costs were used as a proxy to assess the severity of the disease [3], [6].

Back to Article Outline

5. Conclusion 

The side effects associated with the medication, age progression, knowledge and beliefs and understanding of the complex drug regimens by the elderly might affect medication adherence in ways which are unclear and need to be researched further [10], [8]. Understanding the predictors of adherence will help in the development of medication management therapies which might help reduce healthcare costs and the disease burden associated with elderly patients [10], [11].

Back to Article Outline

Conflict of interest statement 

The authors have no conflict of interest to declare.

Back to Article Outline

Acknowledgement 

This study is not supported by any funding.

Back to Article Outline

References 

  1. National Diabetes Information Clearing House. National diabetes statistics; 2007. Available from: http://www.diabetes.niddk.nih.gov/DM/PUBS/statistics/#costs [cited 2009 April 13].
  2. Cowie CC, Rust KF, Bryd-Holt D, Eberhardt MS, Saydah S, Geiss LS et al. Prevalence of diabetes and impaired fasting glucose in adults – United States, 1999–2000. MMWR [serial online] 2003 September 5 [cited 2009 April 13]; 52(35): 833–8337. Available from: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5235a1.htm.
  3. Shenolikar RA, Balkrishnan R, Camacho FT, Whitmore T, Anderson R. Race and medication adherence in Medicaid enrollees with type-2 diabetes. J Natl Med Assoc. 2006;98(7):1071–1077
  4. Donnan PT, MacDonald TM, Morris AD. Adherence to prescribed oral hypoglycemic medication in a population of patients with type-2 diabetes: a retrospective cohort study. Diabetes Med. 2002;19:279–284
  5. Balkrishnan R. Predictors of medication adherence in the elderly. Clin Ther. 1998;20(4):764–771
  6. Shenolikar RA. Racial differences in oral antidiabetic medication adherence, healthcare utilization and costs associated with type-2 diabetes in Medicaid enrollees [dissertation] Columbus (OH): Ohio State University. Diabetes Care. 2008;31:e5;(2)
  7. Steiner JF, Koepsell TD, Fihn SD, Inui TS. A general method of compliance assessment using centralized pharmacy records. Med Care. 1998;26:814–823
  8. Robert J, Glynn JR, Monane M, Gurwitz JH, Choodnovskiy I, Avorn J. Aging comorbidity and reduced rates of drug treatment for Diabetes Mellitus. J Clin Epidemiol. 1999;52(8):781–790
  9. Dailey G, Kim MS, Lian JF. Patient compliance and persistence with antihyperglycemic drug regimens: evaluation of a Medicaid patient population with type-2 diabetes mellitus. Clin Ther. 2001;23:1311–1320
  10. Balkrishnan R, Rajagopalan R, Shenolikar RA, Camacho FT, Whitmire JT, Anderson RT. Healthcare costs and prescription adherence with introduction of thiazolidinedione therapy in Medicaid type-2 diabetic patients: a retrospective data analysis. Curr Med Res Opin. 2004;20:1633–1640
  11. Balkrishnan R, Rajagopalan R, Camacho FT, Huston SA, Murray FT, Anderson RT. Predictors of medication adherence and associated health care costs in an older population with type-2 diabetes mellitus: a longitudinal cohort study. Clin Ther. 2003;25:2958–2971
  12. Paes AH, Bakker A, Soe-Agnie CJ. Impact of dosage frequency on patient compliance. Diabetes Care. 1997;20:1512–1517

PII: S1877-5934(10)00025-1

doi:10.1016/j.ijdm.2010.05.003

International Journal of Diabetes Mellitus
Volume 2, Issue 2 , Pages 122-124, August 2010

Article Tools