Some Diabetics Are Not Taking Their Medication

Summarized by Robert W. Griffith, MD
January 3, 2007

Summary

African American type-2 diabetics, who have a higher burden of illness and mortality than White Americans, are less likely to take their oral medications than other races.

Introduction

We recently wrote about the ill-effects of not taking prescribed medications in heart attack and diabetic patients (see the first link below). Economic factors are often responsible for this non-compliance, or non-adherence, as it's being called. But race is another possible factor. In the USA, African Americans, Latinos, and Native Americans diabetics have a 50% to 100% higher mortality than white Americans, and this is associated with lower adherence to medication.

A study reported in the Journal of the National Medical Association has tried to uncover the reasons for this discrepancy. It comes from the Ohio State University College of Pharmacy; here's a summary of their findings.

What was done

Data were obtained from the North Carolina Medicaid program database between 2000 and 2003. Type-2 diabetic patients over 18 who maintained Medicaid eligibility for at least 12 months within the three study years were identified. Patients who were over 65 were excluded, as they generally had incomplete data, as were those exclusively treated with insulin, as they were the most serious cases and could not be compared with those on oral antidiabetic meds. This resulted in a list of 17,600 patients.

"New starts" among these patients were identified by examining claims for 3 types of oral drugs (thiazolidinediones, sulfonylureas, and metformin) where there were no claims for any of these meds in the previous 12 months. Patients taking more than one antidiabetic medication were excluded, leaving 3,169 in the final analysis.

Prescription refill patterns were used to identify gaps in medication supply, and hence non-adherence to therapy. A medication possession ratio (MPR) was calculated as the days of prescription dispensed divided by the number of days between each refill. This ratio represents the adherence rate for a particular treatment.

The analyzed subjects were classified according to race into three groups: African American (1,527 patients), Whites (1,128), and Others Races (514).

What the analysis showed

The average age of the African Americans was significantly lower than those of the other groups (47.5 years, vs. 49 and 51 years, respectively). There were more men in the Whites group than in the African Americans and the Other Races (32% vs. 23.5% and 18%, respectively).

Adherence rates for new diabetic medications were as follows:

African Americans 0.54
Whites 0.59
Other races 0.56

The difference between Whites and African Americans - about 8.5% - was statistically significant. When adjustments were made for factors such as age, gender, comorbidities (coexistent health disorders), and type of therapy, it was found that the adherence rate of African Americans was 12% lower than that for Whites. Metformin users had a 62% decrease in adherence rate compared with those starting on sulphonylureas or thiazolidinediones.

What these findings suggest

There are obviously several factors for discrepancies in the frequency, severity, and outcome of type-2 diabetes in different races. The risk factors for diabetes include obesity, sedentary lifestyle, family history, high cholesterol levels, and high blood pressure. And all of these can vary in frequency from one race to another. But diagnosis and prognosis without treatment are of little value. So possible factors affecting the success of treatment are clearly very important in helping to explain racial differences in outcome. Adherence to prescribed therapy is a good starting point.

This study has revealed quite considerable adherence differences between African Americans and Whites in respect to oral antidiabetic drugs. The authors of the study suggest that further work should be done to determine the reasons for this race-related difference in adherence. Two obvious starting points are treatment affordability and insufficient patient education. There is little doubt that African Americans, in general, have less access to health insurance funding, or that their health education probably lags behind that of Whites. But there may be other factors responsible for lessened adherence to medication. Beliefs and attitudes about disease may be more extreme than those of other races; it's possible, too, that attitudes to drug-related adverse effects may have a greater influence on discontinuation of a specific medication.

The report summarized here represents a further example of the problems encountered in prescribing. It's not enough for the physician to write a prescription, hand it to the patient, and assume the drug will be taken as long as necessary. Adherence to medication may be inadequate for a variety of reasons, and the disease outcome will be jeopardized. In this case, it's type-2 diabetes; in others, it's cardiovascular problems, or depression, or epilepsy, or osteoporosis, or cancer. Compliance with medications is clearly an important step in good health. As Dr C Everett Koop has said: "Drugs don't work in patients who don't take them."

Source

• Race and medication adherence in Medicaid enrollees with type-2 diabetes. RA. Shenalikar, R. Balkrishnan, FT. Camacho,  et al., J Natl Med Assoc, 2006, vol. 98, pp. 1071--1077

Related Links
Not taking Your Meds Can Be Dangerous to Your Health
US Diabetics Are Not Always Well Treated
Non-Compliance of Diabetics in Canada