BACKGROUND: Statistical models for analysis of correlated count data are important for answering
epidemiological issues that involve taking individual count measurements repeatedly over time
through the use of longitudinal studies. Conventional regression models for this type of data are
inadequate and can lead to inappropriate conclusions and inference. Longitudinal studies in Public
Health involve evaluation and monitoring of patients with infectious diseases, such as HIV/AIDS, to
assess their health status, to check the effectiveness of the treatment, and to make prognosis about
the evolution of the disease, including interdependencies of clinical manifestations. The purpose
of this article is to describe various statistical strategies for the analysis of longitudinal count data
with emphasis on how to choose the most suitable model for the data and in the interpretation of
the results.
METHODS: We illustrate the applicability of various statistical strategies by evaluating the effect of
associated factors on lymphocyte CD4+T cell count in HIV seropositive patients in Salvador, Bahia,
Brazil. We describe the Poisson and Negative Binomial models using the multilevel (ML) approach and
the generalized estimations equations (GEE) for the analysis of longitudinal count data.
RESULTS: The interpretations of the results derived from ML and GEE differ and thus their direct
comparison should be avoided.
CONCLUSION: We believe that the statistical methods for the analysis of longitudinal studies with
correlated count data can be useful to address several important issues in public health, especially in
helping to monitor patients and checking the effectiveness of treatments.