Analysis of Repeated Measurements of HIV Viral Load as a Continuous Variable while Accounting for Missing Values

Recent HIV research predominantly uses Single Measure Frameworks (SMF), focusing solely on the latest viral load data and overlooking missing values. This study explored repeated measures frameworks to assess factors affecting viral load copies while accounting for missing data. The analysis involved 1670 records of HIV patients,using the generalized linear mixed models (GLMM). All variables, except for treatment regimen changes and adherence rating, were recorded at patients treatment enrollment. A GLMM was applied to data before and after imputation accounting for the repeated nature of the HIV viral load copies over time. The best-fitting model, selected for discussion, was the GLMM fitted to multiply imputed data. Gender and adherence rating did not significantly affect viral load copies. The analysis included other variables such as patient age, marital status, treatment duration, WHO clinical stages, and facility ownership. Results show that viral load copies were higher among currently or formerly married individuals ({beta} = 0.49, 0.30; SE = 0.042, 0.052; p = 0.0000). In contrast, viral load copies were lower for patients with longer treatment durations ({beta} = -0.01; SE = 0.001; p = 0.0000) and those receiving treatment at a private facility ({beta} = -0.196; SE = 0.077; p = 0.0000). The study highlights the significance of recognizing repeated data patterns in longitudinal settings and addressing missing values in health research. It proposes a similar investigation in controlled environments to evaluate SMF and RMF in presence of missing values. Author summaryThis work was conceptualized by, Susan Habert Sendege. The author was responsible for the initial drafting of the paper, including the analysis, discussion, conclusions, and recommendations. Saint Kizito and Symon Peter played a pivotal role in shaping the study and all provided numerous hours of time for proofreading and editing.