Main Findings It seems there are demographics disproportionately affected by clinical progression, most prevalently by age, but these are very slight effects. The demographic groups look a little different on the survival plots, but the differences are not strong enough to confidently say one group had a higher risk once we account for treatment and starting health. Baseline immune health: Almost everyone starts with a CD4:CD8 ratio below 1, which shows immune system imbalance. People who start with lower CD4 and lower ratio tend to reach the study endpoint sooner. Treatment results: The ZDV + ddI combination shows the largest improvement in CD4 by week 20. The CD4:CD8 ratio improves slightly, while CD8 changes are inconsistent and harder to interpret. Model takeaway: Predictions are much stronger when we model week-20 values using baseline + treatment, compared to modeling change alone. In short, where you start matters a lot. More days of anti-retroviral therapy increases Karnofsky score
Log in or sign up for Devpost to join the conversation.