OBJECTIVE--To measure the prevalence of chlamydial genital infection in Ethiopian women attending gynaecological, obstetric and family planning clinics; to identify the epidemiological, social and economic factors affecting the prevalence of infection in a country where routine laboratory culture and serological tests for chlamydial species are unavailable; to determine the risk factors for genital chlamydial infection in those with serological evidence of other sexually transmitted diseases. SUBJECTS--1846 Ethiopian women, outpatient attenders at two teaching hospitals and a mother and child health centre in Addis Ababa, Ethiopia. SETTING--Gynaecological outpatient department, antenatal, postnatal and family planning clinics. METHODS--Sera were tested for type-specific anti-chlamydial antibodies using purified chlamydial antigens (C. trachomatis A-C (CTA-C), C. trachomatis D-K (CTD-K), Lymphogranuloma venereum (LGV1-3), and C. pneumoniae (CPn)), in a micro-immunofluorescence test. The genital chlamydia seropositivity was analysed against patient's age, clinic attended, ethnic group, religion, origin of residence, age at first marriage and first coitus, income, number of sexual partners, duration of sexual activity, marital status/profession, obstetric and contraceptive history, and seropositivity for other sexually transmitted diseases. RESULTS--Overall exposure to chlamydia species was found in 84%, genital chlamydial infection in 62%, and titres suggestive of recent or present genital infection in 42% of those studied. Genital chlamydial infection was highest (64%) in family planning and lowest (54%) in antenatal clinic attenders. Exposure to genital chlamydia species was influenced by ethnic group and religion. Those married and sexually active under 13 years of age had greater exposure (69%) to genital chlamydial infection than those first sexually active aged over 18 (46%). Prevalence of infection was highest in those with more than five sexual partners (78%) and in bargirls (84%). The lowest income groups had a higher prevalence (65%) of genital chlamydial infection than the wealthiest (48%). Multivariate analysis showed the most important factors to be age at first coitus, religion, prostitution and present age of the woman in that order. Risk for genital chlamydial infection was increased in those with seropositivity for syphilis, gonorrhoea, HSV-2 but not HBV infection. CONCLUSION/APPLICATION--Chlamydial genital infections are highly prevalent in both symptomatic and asymptomatic Ethiopian women. The high prevalence of infection reported reflects a complexity of socioeconomic factors: very early age at first marriage and first coitus, instability of first marriage, subsequent divorce and remarriage or drift into prostitution, all of which are influenced by ethnic group, religion and poverty--together with transmission from an infected group of prostitutes by promiscuous males to their wives, lack of diagnostic facilities and inadequate treatment of both symptomatic and asymptomatic men and women. The problem of chlamydial disease in Ethiopia needs to be addressed urgently in the context of control of STD.