Evaluation of stool frequency and stool form as measures of HIV-related diarrhea

Purpose: In the highly active antiretroviral therapy (HAART) era, HIV-related diarrhea remains common. Our aim was to evaluate stool frequency and form as measures of HIV-related diarrhea. Method: Forty-eight HIV-infected persons with self-reported diarrhea were studied. In Analysis 1, self-reported retrospective and 7-day prospective measurement of stool frequency and form were compared using Spearman's correlation coefficient. In Analysis 2, diarrhea was measured during two 8-hour study periods in a subgroup (n = 20) using stool weight (Wt), diarrhea symptom score (Sx Score), stool frequency (SP-freq), and stool form using the Bristol Stool Form Scale (SP-BSFS). SP-freq and SP-BSFS were modeled alone and in combination to predict Wt and Sx Score. Results: In Analysis 1, correlation between measures of stool frequency was r(s) = 0.62 (p < .0001) but was r(s)= 0.16 (p =.26) between measures of stool form. In Analysis 2, the two-predictor model best predicted Wt, whereas the model using SP-freq only performed as well as the two-predictor model to predict Sx Score. Conclusion: Prospective measurement of stool frequency performed well; in some situations, it may be used alone to measure severity of HIV-related diarrhea. Our findings may be used to design more rigorous clinical trials in HIV.