Compliance Corner

Research helps identify those at risk for non-adherence

Identify those at risk using 3 elements

Behavioral compliance interventions typically are time-consuming and costly, but it might work more efficiently if clinicians quickly could identify the patients most in need of such assistance.

New research suggests a 3-part strategy for identifying HIV patients who are at risk for poor adherence and who might benefit from a behavioral intervention.

"For those of us working with HIV patients, we’ve seen tremendous gains in the use of highly active antiretroviral treatment [HAART]," says David Ramstad, PsyD, MBA, director of training at the Carl T. Hayden VAMC in Phoenix, AZ.

"But a subset of people are failing, and most of the research seems to be focused on people who are compliant, and those are not the people we’re having problems with," Ramstad says. "The biggest number of those who are not doing well are those who have poor adherence."

HIV patients most at risk for poor medication adherence are those who are abusing substances and those who are depressed, Ramstad notes.

"One of the things we also know is when people come into the clinic, they don’t say, I’m not taking my medicine,’" he says.

So Ramstad and co-investigator Lagen Biles, a former intern at the VAMC, investigated the use of a quick procedure for identifying patients who need adherence assistance.

"We reviewed the guidelines that are available, and they didn’t seem realistic in terms of the amount of time a physician has when seeing patients in a clinic," Ramstad says. "So we came up with quick measures that could be implemented in almost any setting and implemented by anyone—the receptionist could do one part, and the nurse could do a part."

The 3 elements used to identify those at risk for poor adherence and the study’s recommended assessment method are as follows:

  • Substance use: collect urine drug screens;
  • Depression: administer the Beck Depression Inventory (BDI);
  • Continued transmission risk: provider inquiry of risk exposures and adherence to treatment recommendations.1

The last category was selected because investigators were interested in identifying people who might be re-exposed through additional exposures, whether through sex or needles, Ramstad says.

The BDI is an 18-item measure that gives a score for a person’s level of depression, and it could be administered while a patient is waiting for the doctor’s visit to begin.1

The urine screening, which could be administered along with other laboratory draws, will easily tell if someone is using substances, he explains.

Investigators created a 3-question survey for assessing transmission risk, and the questions can be asked by the nurse or physician when the patient is brought back into the office, Ramstad says. The questions are as follows:

  • Have you shared any needles with someone else?
  • Have you engaged in any sexual activity that might have re-exposed you to HIV?
  • Are you taking an active stance to try to prevent transmission to HIV?

"What we’re finding in our clinic, which is probably true in most clinics with established patients, is our patients are very straightforward and honest with us," Ramstad says. "This may not be true when someone is going to a clinic where they don’t have an established relationship with providers."

Patients may minimize their risk factors if they are afraid of a lecture or of being treated badly, Ramstad says.

Investigation into the use of the screening methods continues, Ramstad says.

"We’re trying to establish in a more psychometric fashion the degrees to which these approaches will help us predict which people have better adherence for HAART," Ramstad says.

"We’re still in the process of data collection, but essentially, we’re finding the data does support our hypothesis that depression is an indicator for poor compliance," Ramstad says. "We’re finding that at mild-to-moderate levels of depression it does not impact adherence, so it’s only for more severe cases of depression."

This finding confirms previous research showing that when people have more severe depression the illness itself impacts a people’s functional level and their ability to manage their lives, Ramstad notes.

"They’re overwhelmed and don’t have the energy and resources to manage it," he says. "That was consistent with what anecdotally we expected to find."

With substance use, even mild abuse of alcohol and drugs significantly impacts adherence, Ramstad says.

"Anecdotally, it makes sense," he says. "If someone was out partying all night and sleeps in, then he misses his morning dose."

The data reveal mixed results on the transmission risk factors, Ramstad says.

"But we do know, not from our data, but from other’s, that people who are using substances are more likely to engage in risk behaviors because their judgment is impaired," he says. "They’re out and having a good time and meet someone, and maybe if they were sober they wouldn’t engage in risky behavior, but that’s from other people’s data."

A chief advantage to using this type of screening for poor adherence is it will give clinicians a faster indication of problems than if they were relying on viral load and t-cell data, Ramstad says.

The study indicates that a proven adherence strategy is motivational interviewing in which clinicians use the stages of change model.1

Broken into its 5 essential elements, the provider using motivational interviewing can do the following, according to the study:

  • Express empathy: show understanding of ambivalence toward change;1
  • Develop a discrepancy: point out discrepancy between behavior and desire to change;1
  • Avoid arguing: avoid confrontation—it strengthens maladaptive behavior;1
  • Roll with resistance: revisit the discrepancy of goals and behavior;1
  • Self-efficacy: support personal responsibility and support their successes.1

The next phase of research is to conduct an outcomes study with motivational interviewing to see if the behavioral intervention works with the patients who are found at risk through the 3-part screening process, Ramstad says.

"We hypothesize this is the approach that will work with them," Ramstad says. "It’s consistent because it comes out of the old, client-centered approach, which is nonjudgmental and allows people to gain on inner strengths."


  1. Ramstad D, Biles L. Strategies to Increase Predictive Validity of Behavioral Compliance with HAART. Presented at the Treatment & Management of HIV Infection in the United States Conference, held Sept. 15-18, 2005, in Atlanta, GA.