Relationship of Polypharmacy to HIV RNA Suppression in People Aged ≥ 50 Years Living With HIV

MM Murray; J Lin; A Buros Stein; ML Wilcox; J Cottreau; M Postelnick; FJ Palella


HIV Medicine. 2021;22(8):742-749. 

In This Article


This was a cross-sectional analysis of PLWH aged ≥ 50 years as of 1 June 2013, prescribed ART, and seen at least once in the Northwestern Infectious Disease Center between 1 June 2013 and 31 May 2015. Pertinent variables included: age, gender, race, number of clinic visits, plasma HIV RNA, CD4 T-lymphocyte count (cells/μL), complete medication list, patient self-reported adherence to ART, presence of patient self-reported AEs, length of time on ART, history of opportunistic infections, comorbidities, low-density lipoprotein, blood pressure and fasting blood glucose. Self-reported ART adherence was collected if there was any mention of self-reported adherence in the medical record. This study was reviewed and approved by the Northwestern University and Midwestern University Institutional Review Boards with a waiver of informed consent.

Medications included in this study were any oral medications, including ART (e.g. tablets, capsules or liquids), injectable medications, patches, eye drops, scheduled inhalers and nasal sprays. Fixed dose combination medication preparations including more than one active ingredient (e.g. abacavir/lamivudine or hydrochlorothiazide/lisinopril) were each counted as one medication. If a patient was taking abacavir and lamivudine as separate tablets, they were counted as two medications. Any over-the-counter medications, topical formulations (e.g. creams or ointments), "as-needed" inhalers, "as-needed" nasal sprays and any prescriptions with the indication of "as-needed" were excluded. Type of ART regimen was categorized as based upon protease inhibitors (PIs), integrase strand transfer inhibitors (INSTIs), nonnucleoside reverse transcriptase inhibitors (NNRTIs), nucleoside reverse transcriptase inhibitors (NRTIs) and other type (patient was taking more than one of the following classes: PI, INSTI and NNRTI). An undetectable plasma HIV RNA (viral load) was defined as < 20 HIV-1 RNA copies/mL. The length of time on ART was documented in years, as noted in the medical record. The presence of DDIs was found utilizing the University of Liverpool HIV Drug Interactions Checker ( and DDIs were indicated as pharmacokinetic or pharmacodynamic. Cardiovascular disease (CVD) was defined as one or more of the following: hypertension, atrial fibrillation, ischaemic heart disease and congestive heart failure. Pulmonary disease was defined as asthma or chronic obstructive pulmonary disease. "Lipids at goal" was defined as a low-density lipoprotein (LDL) concentration < 100 mg/dL within the study period. "Blood pressure at goal" was defined as a blood pressure reading ≤ 140/80 mmHg within the study period. "Blood glucose at goal" was defined as a fasting blood glucose concentration < 100 mg/dL within the study period. All data were taken from the most recent clinic appointment and laboratory results.

Statistical Analysis

The primary outcome measure was the presence of an undetectable plasma HIV RNA. Univariate and multivariate logistic regressions were used to analyse the effect of variables of interest on the presence of an undetectable plasma HIV RNA. Logistic regression with the minimum P-value approach was used to identify a threshold number of medications (ART plus non-ART medications) above which there was a statistical difference in plasma HIV RNA detectability. This cut-off point was then used to group patients for further analyses. Demographic variables were summarized descriptively. Differences between patients were analysed for variables of interest with Mann–Whitney U tests, χ 2 tests and Fisher's exact tests, as appropriate. Logistic regression estimates were presented as odds ratio (OR) and corresponding 95% confidence interval (CI). The significance level was set at the 0.05 level. All analyses were completed with R version 3.6.0.