Date of Completion


Embargo Period



Tania B. Huedo-Medina, Michael M. Copenhaver, Amy Gorin

Field of Study

Health Promotion


Master of Science

Open Access

Open Access


Importance: Different ART treatments to HIV patients has been shown to be effective in improving a variety of disease outcomes, including CD-4 cell count improvements and viral suppressions. In addition to these improvements, investigators are also interested to evaluate the association between ART treatments and the body habitus changes of the HIV patients. It is unclear that which body habitus change outcomes are more greatly associated with what type of ART treatments and how those effects could vary depending on context and sample characteristics.

Objective: To obtain overall effect sizes for the body habitus change outcome variables (weight, body mass index, trunk fat, limb fat and lipid parameters) and explain the variability among different ART treatments/interventions (NRTI, NNRTI, PI and combination HAART medicines) considering all treatments clustered within the current literatures using multilevel approach for meta-analysis.

Data Sources: Five electronic databases (PubMed, Scopus, CINAHL, PYCHINFO and Cochrane Controlled Trials register) were searched starting on July 6, 2015 using a comprehensive Boolean Search Strategy.

Study Selection: Studies were included if pre-and post- intervention measurements of body habitus changes were reported and NRTI, NNRTI, PI, and combination HAART medicines LPV/r vs ABC/3TC + LPV/r; LPV/r +ZDV/3TC vs LPV/r NVP; PI compared alone or, in combination with other ARTs and NNRTI compared alone or, in combination with other ARTs were used as treatments to HIV patients. Data from 20 studies (N = 376) were included.

Data Extraction and Synthesis: Two independent researchers identified studies that met the inclusion criteria and coded methodological, participant, and treatment characteristics using a pilot tested structured coding form. Inter-rater agreement was evaluated using Cohen’s kappa coefficient statistic and effect size was calculated as standardized mean difference as all outcome variables were continuous.

Main Outcomes and Measures: Weighted mean effect size under random-effects assumption were obtained and modeled after pooling the individual standardized mean differences for each study on the five-body habitus change outcomes.

Results:There were significant beneficial effects of the LPV/r combination category on weight and BMI [d+ = 0.44, 95% CI (0.18 to 0.71)]; [d+ = 0.94, 95% CI (0.53 to 1.3)] respectively. PI combination category has beneficial effect [d+ = 0.44, 95% CI (0.2601 to 0.6311)] for limb fat. Level of HDL significantly increased in all three combination categories but more in favor of NNRTI combination category with large impact [d+ = 0.85, 95% CI (0.54 to 1.15)], p-value of

Conclusions and Relevance: LPV/r combined HAART is more effective with moderate increase of weight and BMI. PI combined HAART has significant impact on limb fat improvement which could be used to maintain body proportion in HIV patients as limb fat loss is a major concern. NNRTI combined treatment category showed large impact on HDL which might be prescribed as HDL reduces cardiovascular risk.

Major Advisor

Dr. Tania B. Huedo-Medina