Summary
Introduction
No one (or almost no one) enjoys taking medications, yet most of us have to, at one time or another. Failure to take prescribed drugs can have serious consequences, as a series of reports in the Archive of Internal Medicine point out. We summarize here the findings in two of the most important ones - non-adherence to medication in patients after hospitalization for a heart attack, and in patients with diabetes.
Discontinuing medications after a heart attack
Denver researchers used data from patients enrolled at 19 hospitals in the Prospective Registry Evaluating Myocardial Infarction: Event and Recovery (PREMIER) study. All patients with a verified myocardial infarction (heart attack) between January 2003 and June 2004 were registered. Data collected at baseline came from the medical records, an interview with the patient, and hospital discharge records.
The focus of this study was on three medications - aspirin, beta-blockers, and statins - that are well-recognized therapies to prevent a recurrence of heart attack. Medication use was obtained from the hospital records and telephone interviews done at 1, 6, and 12 months after discharge. Survival status at one year was ascertained using the Social Security Death Master file.
Out of 2500 patients enrolled, 70% were discharged from hospital with all three medications (aspirin, a beta-blocker, and a statin). Their average age was 60 years, 70% of them were male, and 80% were white. Over 1500 completed the one-month interview.
In the one-month reports it emerged that 66% continued taking all three medications, 12% discontinued all three, 4% discontinued two, and 18% discontinued taking one medication. Those who stopped all meds were more likely to be older, less well educated, and have more co-existing disorders, such as chronic lung disease.
The 1 in 8 patients who discontinued taking all 3 drugs at one month had a lower one-year survival rate compared with those who continued to take at least one of the 3 medications - 88.5% vs. 97.7%.
The analyses show that stopping medication after a heart attack is common (12% of patients after one month) and more than triples the risk of dying within one year. The findings are similar to those reported in another large published study. No solid information is offered regarding the causes or reasons for stopping medication, or the mechanism(s) for increased mortality.
Nonadherence to medication in diabetics
The same group of researchers used the Kaiser Permanente of Colorado diabetes registry to provide information on prescribed oral antidiabetic, high blood pressure, and statin medications, using their automated pharmacy records. Adherence to prescribed medication was obtained from filled prescription records; this allowed calculation of the actual medication purchases against the prescribed medication amounts. Adherence was expressed as the proportion of days covered (PDC), with a PDC calculated for each of the three categories of medications and for all three counted together.
As in the first study, baseline data was obtained from medical records during 2003. Hospitalizations and deaths were collected from January 2004 to April 2005. Data from 11,500 patients were analyzed. Their baseline age averaged 64, half were male, and most of them had co-existing conditions such as high blood pressure, high cholesterol, and coronary heart disease.
For the analysis, all patients were classified as "medication adherent" or "medication non-adherent", the latter depending on whether the PDC was below 80%. i.e. they were not taking their prescribed medications for at least 20% of the time.
It was found that 21% of the patients were non-adherent. In general, they were younger and had fewer co-existent conditions compared to adherent patients. During the follow-up period, non-adherent patients had higher HbA1c levels (i.e. their diabetes was less-well controlled), higher blood pressure, and higher LDL-cholesterol; these changes reflect omission of the three specific drug types studied.
Non-adherent patients were found to have a higher rate of hospitalization and all-cause mortality than adherent patients (23% vs. 19%, and 6% vs. 4%, respectively). When allowances were made for possible interfering factors, such as age, presence of high blood pressure, coronary heart disease, and so on, the rates of hospitalization and mortality were still significantly greater in the non-adherent patients. The results were similar when using the PDCs for the individual types of medications.
This study shows that non-adherence to three of their medications caused diabetics to have increased hospitalizations and mortality, compared with those who took their drugs as prescribed. The non-adherence rate was 21%, or one in five patients.
How to improve medication adherence
Another article in the Archives of Internal Medicine reports on the role of cost in non-adherence to prescribed medications in elderly and disabled Medicare recipients1. This is obviously a large factor, but not the only one, in non-compliance with the doctor's recommended therapy. An editorial in the same issue2 tries to suggest the likely causes and remedies for the problem that today costs the country an estimated $100 billion a year.
The patient is not the whole problem. The physician and economic factors each play considerable roles. Physicians often omit critical information when prescribing a new drug. They should improve their educational responsibilities to the patient, informing her/him of the purpose of the drug, the dosing schedule, possible adverse effects, and the risks associated with not taking the drug. The pharmacist can reinforce this information when providing the drug.
Having to take several drugs, or dosing several times daily, is another factor in non-compliance. Drug combinations are a step towards combating this. Going further, a polypill approach has been mooted, but has not been taken up yet. It would involve combining, say, generic forms of aspirin, an ACE inhibitor, and a statin in a single once-a-day pill. (See the first link below).
It may be possible to persuade patients to be non-adherent in a rational way. Thus physicians can concentrate on using especially beneficial drugs and cut back on the less beneficial ones. Or explain to the patient the relative importance of the drugs being taken, so that, if forced by economic circumstances to cut back, the patient will only eliminate the less relevant ones.
Particularly distressing is the switch some patients make from prescribed drugs to alternative therapies recommended by friends or unreliable information sources (e.g. the Internet). The proof of effectiveness demanded for prescription drugs should convince all but the most illogical of the benefits they offer over herbs and supplements. It all comes back to patient education - which is something we try to help provide in these pages.
