HIV JournalView

HIV JournalView: August 2012

September 6, 2012

 < Prev  |  1  |  2  |  3  |  4  |  5  |  Next > 

Comparing Adherence to Two Different HIV Antiretroviral Regimens: An Instrumental Variable Analysis

Richard E. Nelson et al. AIDS Behav 2012; epublished ahead of print. Read the abstract.

There is a cottage industry of research that aims to show how bad doctors are, especially compared to nurses, when it comes to a host of skills, including gaining patient trust, hand washing, and paying for coffee with exact change when the line at the counter is really long. (I made the last one up, but I suspect it is true). "Survey says!" that we M.D. types also flub it when predicting who will be adherent to HIV meds. Yet, we regularly make treatment decisions based upon our ability to size up a patient's adherence potential.


For most of us, when confronted with a patient whose ability to adhere to an antiretroviral regimen we judge to be dodgy, it is common to prescribe a protease inhibitor (PI)-based regimen. Ritonavir (Norvir)-boosted PIs are potent and also provide a good measure of indemnification against drug resistance, as the development of resistant mutants on such regimens is biologically difficult and therefore uncommon. However, with a pill- and bottle-count that, unlike single-tablet formulations, are greater than one, PI-based regimens are more challenging to take.

This preference for PI-based regimens in patients perceived to be less likely to adhere introduces channeling bias that complicates any study that compares adherence between different antiretroviral classes. In this analysis of adherence to PI-based versus non-nucleoside reverse transcriptase inhibitor (NNRTI)-based regimens in 1,676 HIV-infected patients (929 on PI, 747 on NNRTI) cared for by the U.S. Veteran's Administration (VA), an attempt was made to minimize the effects of channeling bias by adjusting for the clinician's default or "go to" regimen type. Thus, they examined what regimen each clinician tended to use in patients in whom there was no clear indication for one regimen or another and accounted for this preference in the final analysis.

Without the adjustment and without looking at those who started a PI- or an NNRTI-based regimen, adherence, as measured by pharmacy refill, was lower among those on PIs; 20% of patients on a PI had an estimated adherence of >80%, while 35% of the patients on an NNRTI had adherence above this level. Adjusting for the clinician preference, as well as patient factors, led to no major differences, and in fact the association between PI therapy and non-adherence became stronger.

This is an interesting and novel study, but it is not without limitations: It was done in the VA (and therefore is not representative of people living with HIV who are outside the health care system), there is no sure-fire way to eliminate channeling bias, and there was no discussion of regimen efficacy or the development of drug resistance, and pharmacy refills are useful but crude measures of adherence. But, it does suggest that, all things being equal, adhering to PI-based regimens is more difficult than adhering to NNRTI-based regimens. Something to think about as you are considering what to prescribe and your Spidey Sense starts tingling that there is the threat of non-adherence.

 < Prev  |  1  |  2  |  3  |  4  |  5  |  Next > 

This article was provided by TheBodyPRO. It is a part of the publication HIV JournalView.

No comments have been made.

Add Your Comment:
(Please note: Your name and comment will be public, and may even show up in
Internet search results. Be careful when providing personal information! Before
adding your comment, please read's Comment Policy.)

Your Name:

Your Location:

(ex: San Francisco, CA)

Your Comment:

Characters remaining:


The content on this page is free of advertiser influence and was produced by our editorial team. See our content and advertising policies.