Which method of adherence measurement is most suitable for daily use to predict virological failure among immigrant and non-immigrant HIV-1 infected patients?

In industrialized countries, virological failure occurs more often among HIV-infected immigrant patients. Non-adherence is the most credible explanation. We compared adherence of immigrant patients with that of non-immigrant patients in the Netherlands, and investigated which method of adherence measurement is most suitable for daily use to predict virological treatment failure: testing knowledge of the current regimen, a quantitative adherence interview, pharmacy prescription refill ratio (dispensed medication divided by prescribed medication, DM/PM), and plasma drug levels. Included were 61 immigrants and 81 non-immigrants. Virological failure did occur more often in immigrants than in non-immigrants (19.7% (12/61) versus 8.6% (7/81), p=0.056), especially among previously naive patients (19.6% (11/56) versus 0% (0/54), p<0.01). There were no differences between both groups on any of the four adherence measures. Virological failure was associated with reporting stopping medication when not feeling well (OR=12, 95%CI=1.9-77.7, p=0.02), and, among naive patients, also with a DM/PM < 0.85 (Odds Ratio=5.1, 95%Confidence Interval=1.2-22.3, p=0.03). Although our study confirmed a much higher virological failure rate among immigrants, we were unable to identify clear differences in adherence between immigrants and non-immigrant patient, although virological failure was associated with stopping medication when not feeling well and a low DM/PM. Unstructured treatment interruptions are a likely explanation of the findings. Interventions should be aimed at preventing patients to stop medication. A DM/PM below 0.85 can be indicative for patients who did stop medication and are at risk for virological failure.