No 'magic bullet,' but compliance increases with multiple interventions

Need for positive doctor-patient communication

There is no "magic bullet" to get patients to comply with doctors' orders, but researchers have discovered that a shotgun approach yields results.

A study from the Johns Hopkins School of Public Health in Baltimore shows that a combination of compliance approaches including direct education from doctors and other health care providers, group processes, family support, behavioral modalities, and provider-targeted interventions can have a significant effect on outcomes.

The results were particularly dramatic for patients with chronic diseases like asthma, diabetes, and hypertension. Those least affected were patients with acute infections needing antibiotic treatment.

"The more comprehensive the program, the more effective the outcome," says Debra Roter, DrPH, a Johns Hopkins professor of health policy and management and lead author of the article on the study published in the August issue of Medical Care. "The most powerful combinations included all three elements: educational, behavioral, and affective."

In a study which speaks as much to changing behavior among physicians and other health care providers as to patient behavior, Roter and her team conclude, "More information giving and more positive talk, but less negative talk and fewer questions overall (but more questions about patient compliance) predicted higher patient compliance. The effort it takes to build rapport with a patient can lead to a significant improvement in the level of compliance."

Marianna Sockrider, MD, DrPH, assistant professor of pediatrics and pediatric asthma specialist at Baylor College of Medicine in Houston, says she was not surprised by the need for more positive doctor-patient communication.

"If you go back to your doctor and he or she never comments, 'Things have been going so well since you've been on this medicine,' [it's because] doctors tend to focus more on the problem."

Medical recommendations to patients can range across a spectrum from consuming a single pill to following a complex drug regimen, returning for follow-up appointments to major lifestyle modifications.

Need to follow instructions

Past studies have shown that 30% to 60% of all patients for whom drugs are prescribed fail to follow instructions. Most researchers agree than half of all patients for whom drugs are prescribed do not get full benefits of the medications because they do not take them according to orders. In about a third of all cases, researchers say, lack of compliance with instructions can seriously jeopardize patients' health.

And while doctors frequently suspect patients are not taking their medications as prescribed, patients rarely volunteer the information and doctors usually don't ask.

One monitoring mechanism involved pharmacy reports on pill counts and prescription refills, although Roter points out that a prescription refill is no guarantee the patient has actually consumed the medication.

A wide variety of compliance incentives was offered, including:

r One-on-one and group teaching.

r The use of written and audio-visual materials.

r Mailed materials.

r Telephone instruction.

r Behavioral modeling and contracting.

r Packaging and dosage modifications or tailoring.

r Rewards.

r Mail and telephone reminders.

r Family support.

r Counseling.

r Supportive home visits.

It is of little importance which of the above methods is employed to persuade patients to higher compliance levels, the Roter study shows.

Rather, multiple approaches are clearly the key. Educational interventions that used two or more different ways of learning helped the most patients, perhaps reflecting the fact that people learn in different ways, Roter says. She says it is important for health care providers not to think of a particular patient as "compliant" or "noncompliant."

Each patient is a composite of "complex behaviors," Roter says. "A patient who is consistent in taking medications may not be as good at keeping appointments. The more explicit the discussion around compliance, the better the opportunity for patient and physician to come to an understanding."

She also discovered that mailing a form letter to remind patients about appointments or tests increased patient compliance as much as making time-consuming phone calls. Conversely, offers of rewards or incentives for keeping appointments was shown to be largely ineffective.

Sockrider supports the multiple-incentive concept but advocates a more individualized approach. She says a number of variables govern compliance, including belief systems about a disease pattern, financial considerations, side effects, or simple lack of education about the use of a drug.

"For more serious and long-standing ideas such as 'I'm not sure I have really bought into the idea of needing this medicine every day' or 'How will it work for me?' - those two things really require very different strategies," Sockrider says. "Depending on the individual within the population, you need to use different resources or different strategies to try and help the patient become compliant or adopt the kind of program he or she needs to keep the current disease well." she concludes. "There's not just one approach that works."

Roter says compliance improvement is "very impressive" in many categories. For example, 86% of diabetics given indirect compliance measures "derived substantial benefits."

"This is the biggest effect size I have ever seen," she says.

Roter's study is based on information pulled from a database including 153 published studies included in 162 published articles in English- language journals between 1979 and 1994. Of those, 55 studies involved adult subjects (ages 18 to 65), seven looked at elderly populations, and 59 included both adults and the elderly. Children, adolescents, and their caretakers were the focus of 22 studies and 10 targeted providers.

Among the criteria for a study's inclusion in the Roter project included:

r At least one systematic effort was used to influence or improve compliance.

r Compliance was quantitatively measured.

r Sample size was 10 or more (in fact, sample sizes ranged from 20 to more than 75,000 subjects).

r The intervention and its focus were described in detail.

r Results were weighed against a control group.

While most of the studies (58%) looked at only one compliance measure, included studies looked at as many as five.

Roter notes than even in areas where only small improvements in compliance were noticed, those results are important. "This modest advantage can carry appreciable clinical and practical implications when the outcome is as important as survival," she says. "When these results are interpreted in terms of an important and valued outcome such as lives saved or hospital visits averted, however, even very small effect sizes were clinically impressive."