Objective: To determine where and with what symptoms women seek care for reproductive tract infections (RTI) in Morocco and to guide allocation of resources for training and treatment for RTIs.
Methods: A primary healthcare centre (PHC), a family planning centre (FPC), and a specialty dermatovenereology clinic (SC) were selected in each of three urban areas. Women with symptoms of vaginal discharge, lower abdominal or pelvic pain, or genital lesions (genital ulcer or warts) underwent interviews, physical examinations, serological testing for human immunodeficiency virus (HIV) and syphilis, and collection of vaginal fluid for microscopic examination, and urine for detection of Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (GC) by ligase chain reaction (LCR).
Results: Over 8 months, 1238 women enrolled, including 61.8% at PHCs, 34.8% at FPCs, and 3.4% at SCs. Overall, 54% complained of vaginal discharge, of whom 8.8% had GC or CT infection and 30.1% had trichomoniasis (TV) or bacterial vaginosis (BV); 24.9% complained of lower abdominal pain with or without vaginal discharge, of whom 7.3% had GC or CT and 22.6% had TV or BV. GC or CT infections were found in 10.1% of PHC and 5.4% of FPC patients; while TV and/or BV infections were found in 28.7% and 22.8%, respectively. GC or CT infection was associated with perceived risk behaviours of the male partner (for example, belief partner is unfaithful) more often than with reported risk behaviours of the women themselves. For vaginal infections, a modified World Health Organisation (WHO) test algorithm for vaginal discharge involving risk assessment plus speculum and bimanual examination was 98.0% sensitive at PHCs and 90.8% at FPCs, with positive predictive value (PPV) of 33.4% at PHCs and 26.8% at FPCs. For GC or CT infections this algorithm was 60.6% sensitive at PHCs and 85.7% sensitive at FPCs; but PPV was only 9.9% and 9.0% respectively, little higher than the background prevalence of these infections. An RTI algorithm (Morocco specific) had comparable sensitivity and PPV for vaginal infection, and for cervical infection was less sensitive but had much higher PPV (26.9% for PHCs and 26.7% for FPCs).
Conclusion: Women with complaints of vaginal discharge and/or lower abdominal pain presented to PHC and FP clinics, not to SCs. PHCs and FPCs should therefore receive resources for management of vaginal discharge. Both the test algorithm and the new RTI algorithm were useful in allocating treatment for vaginal infection, but only the RTI algorithm discriminated in selecting women with cervical infection. Even with the RTI algorithm, which limited treatment for cervical infection to risk assessment positive patients with signs of cervical infection or PID, the PPV for cervical infection was low, potentially resulting in frequent overtreatment and problems of partner notification.