Counseling Helps HIV Patients Stick to Treatment

HIV patients receiving a year-long counseling program on problem solving showed better adherence to drug therapy in a randomized study, researchers said.

In the 180-patient trial, those assigned to the program, called Managed Problem Solving, were significantly more likely (odds ratio 1.78, 95% CI 1.07 to 2.96) to be classified in a higher adherence category compared with participants who received usual care, according to Robert Gross, MD, of the University of Pennsylvania in Philadelphia, and colleagues.

There was also a strong trend toward a greater likelihood of having undetectable viral loads in patients receiving the program (OR 1.48, 95% CI 0.94 to 2.31), the researchers reported online in JAMA Internal Medicine.

Gross and colleagues noted that the program was resource-intensive — involving 16 one-on-one counseling sessions followed by monthly telephone calls — but because the estimated cost per participant was “substantially less than $1,000 per year,” the increase in treatment adherence made it almost cost-effective, they argued.

As in other chronic illnesses, taking the prescribed medications regularly on schedule proves difficult for many HIV patients, Gross’ group wrote. Among the common barriers in this population are poor health literacy and social support, substance abuse, depression, and the complexity of their daily medication regimen.

Because these barriers vary from patient to patient, “personalized strategies … hold the greatest promise for success,” the researchers asserted.

The Managed Problem Solving program was designed to provide intensive assistance to patients in identifying and overcoming barriers to treatment adherence. It was delivered by staff members with no special training in psychology or counseling; they were required only to have a college degree and some experience with healthcare patients.

Patients in the study met with the counselor initially for 60 to 90 minutes, with three more in-person sessions held monthly and lasting up to 45 minutes. In between these sessions, the counselor sought to talk with the patient weekly by telephone, and then monthly for another 9 months.

During the early sessions, the counselor and patient “brainstormed” methods for overcoming barriers and discussed ways to implement them. Later on, the discussions turned to reviewing whether the chosen methods had helped as well as new problems that may have come up.

Counselors also reminded patients during the follow-up phone calls about refilling prescriptions.

The control group received usual care, which consisted of a single meeting with a pharmacist about their drug regimen and, if patients wanted, a pill organizer.

Patients had a mean age of 42, with about 60% male and 85% black. One-third had incomes below $5,000. Drug and “hazardous” alcohol use were present in 26% and 18%, respectively. Some 40% were treatment-naive when entering the study.

Adherence was measured with electronic monitors attached to the medication bottles, which could detect when the containers were opened. If a patient opened a bottle, it was assumed that he or she took the medication.

The authors set categories of adherence, ranging from greater than 95% to 70% or less.

The OR of 1.78 for patients receiving the intervention being in a higher adherence category applied at every follow-up point during the year-long study, they indicated. That figure was calculated on an intent-to-treat basis.

Dropouts were common, however. A total of 56 of the 180 enrollees left the study prematurely, including 33 assigned to the problem-solving program. Nearly half the dropouts disappeared without explanation.

When the authors examined outcomes among participants who completed the study, the OR for being in a higher adherence category at any follow-up point was 2.33 (95% CI 1.35 to 4.05) for the program. The researchers found no evidence that the effect changed over time, and it did not appear to depend on factors such as age, sex, race, or baseline viral load or CD4-positive cell count.

The per-protocol analysis also showed that the odds of having undetectable virus at any follow-up point was significant for patients receiving the intervention (OR 1.98, 95% CI 1.15 to 3.41).

There was a close relationship between adherence and virologic response: for each 25% increment in the proportion of prescribed doses actually taken, the OR for undetectable viral load was 1.99 (95% CI 1.64 to 2.41) in a conservative analysis (imputing a value of “detectable virus” for patients with missing data on viral load).

Gross’ group did not perform a detailed cost-benefit analysis. But they noted that previous studies had indicated that, in the HIV population, an intervention leading to a 10% improvement in adherence is cost-effective if it costs less than $1,000 per year per patient.

With each counselor earning $50,000 per year and following 20 study participants at 15% of total effort — and including the $150 annual cost for the pill-bottle monitors — the cost easily met that standard, the researchers indicated.

The study had some limitations: Participants could not be blinded to the study arm and were recruited from academic specialty HIV clinics where services for adherence may be greater than those in general medical clinics.

In an accompanying commentary, physicians at Massachusetts General Hospital in Boston said the study represented “significant progress toward a model of effective long-term adherence support.”

But drug therapy for HIV is lifelong, and “calling every HIV-infected person on treatment every month of their life may not be feasible or necessary,” wrote David Bangsberg, MD, and Jessica Haberer, MD.

Instead, they suggested that a more practical approach may be lifelong adherence monitoring — using pill-bottle monitors or other technologies — with interventions such as the current one reserved for patients with lapses in adherence.

“Advances in just-in-time adherence monitoring and support delivered when and where it is needed may transform treatment from success measured over years to success measured over a lifetime,” Bangsberg and Haberer wrote.