PROSPECTIVE EVALUATION OF A FLOW-CHART USING A RISK ASSESSMENT FOR THE DIAGNOSIS OF STDS IN PRIMARY HEALTH-CARE CENTERS IN LIBREVILLE, GABON

Background: The serious impact of STDs on women and children in partic ular, and the linkage between STDs and HIV infection are a profound co ncern to public health worldwide. One of the main strategies against S TDs is based on early diagnosis and treatment. However, this approach is limited by the lack of appropriate laboratory facilities. A syndrom ic approach has been recommended by the WHO but needs to be evaluated under field conditions. A preliminary cross sectional study on STD pre valence and risk factors in Libreville showed that 13.5% of pregnant w omen had gonococcal and/or chlamydial infection which justifies system atic screening of STDs. Based on the results of this study, different flow charts with or without a risk factor assessment (score) were desi gned. The flow chart with the best performances for diagnosing chlamyd ial or gonococcal cervical infection and routinely acceptable, was a s core algorithm, based on two risk factors (age and marital status) and four simple clinical signs (pelvic or lumbar pain, vaginal discharge and its characteristics). Sensitivity and specificity were 76.9% and 4 0.4% respectively. Thus, the objective of this study was to evaluate t his strategy under field conditions. Methods: A prospective study amon g pregnant women attending antenatal clinics was done. The score was a pplied to each woman by a midwife and a physician, and specimens were collected for the reference laboratory tests. Validation of the algori thm was done by comparing the performances with the gold standard labo ratory diagnosis. Results: 646 pregnant women were enrolled. The preva lence of cervical infection was 11.3%. The sensitivity and specificity of this algorithm recorded by the midwives were 73.3% and 54.8%, resp ectively and by the physician 76.7% and 50.6%. The proportion of women correctly classified by the midwives and by the physician was not sig nificantly different. Conclusion: The score applied was well accepted by healthcare workers and patients, and was routinely practised. Resul ts obtained by the midwives and by the physician were similar. Thus, t he use of flow charts which adds a risk assessment to the syndromic ap proach for diagnosing cervical infections is feasible. However, the pe rformances of such flow charts need to be improved before being used r outinely.