Although none of the participants were selected based on their STIs seeking behavior, 70% and 50% of women presented with at least one active or treatable STI. It is generally assumed that STI prevalence is higher in urban residents [24
]. However, these findings show that the burden of STIs in rural areas may be underestimated. The STI prevalences found were higher than those reported from other rural areas in sub-Saharan Africa [25
]. The high migration rate might partially explain the high frequency of STIs in these women [4
Vaginal discharge showed no correlation with the presence of infection, which suggests that syndromic management alone is unlikely to have a major public health impact in controlling STIs and HIV transmission in women [27
On the contrary, the presence of anogenital warts was significally correlated with the presence of any treatable STI. Most pregnant women had either an active (75.6%) or treatable (55%) STI. These frequencies were higher than what has been reported in the region [7
] and in a urban area of northern Mozambique (51%) [31
]. The most frequent STI was Trichomonas vaginalis which was found in 31% of women, and taking into account that the direct examination only detects approximately a 60%–70% of infections it is reasonable to assume that we are underestimating its prevalence. The high prevalence of gonorrhea, CT, and syphilis are particularly alarming due to their potential impact on the newborn. At the time of the study, the only control program in place was that of syphilis prevention. These results call for different approaches to STIs prevention of the main STIs in both pregnant women and their partners. In Mozambique, current guidelines recommend ciprofloxacin in cases of vaginal discharge [32
]. However, this drug is rarely available and kanamycin plus erythromycin is the combination still most frequently used. Sensitivity for kanamycin was not tested but it can be assumed to be similar to that of gentamicin (9%), which may partly explain the high frequency of gonoccocal infection observed. These are still the only available results on NG sensitivity in the country, and reflect the difficulties that countries with limited resources face in reconciling their health policies with some recommendations.
CT infection prevalence was 8% overall and the 14–20 years old group had a higher prevalence of CT infection (10%) than that reported from rural Tanzania (2.4% of female adolescents) [33
]. This discrepancy may be explained by the recruitment sources of our study being more likely for our participants to have had more sexual partners. These data are consistent with the evidence that girls and young women are more susceptible to CT infection than older women, with the consequent risk, among others, of infertility [34
Syphilis prevalence between 1.6% and 9.8% in rural areas [16
] to 18.3% in Maputo has been reported among women in Mozambique [11
]. In this study, syphilis was detected in 12% of all women and in 10% in the youngest age group. These figures are also higher than those from rural areas of South Africa (8%) [8
] and Tanzania (9.1% of all women, 6.6% of adolescents) [35
]. Again, the different selection of individuals may explain the different rates found in the current study. However, syphilis prevalence was also high in the oldest age groups, where most women were recruited from the community and therefore not exposed to the potential selection bias.
The overall prevalence of HSV-2 antibodies was high (83%). Among 14–20 year old women the prevalence was 56%, higher than that reported in another study from rural Africa (27%) [36
]. HSV-2 is of much interest in Africa because it increases the risk of HIV transmission [37
]. Although the information on HSV-2 is based on seroprevalence of antibodies, HSV-2 seropositivity has been proposed as a marker of sexual risk behavior among adolescents [36
]. These observations support the latter, and suggest that there is high-risk sexual behavior at very young ages.
The overall prevalence of HPV infection was high, especially in the youngest age group where over 50% were infected. HIV and co infection with other STIs are highly prevalent in this population, and this is known to increase the risk of progression from cervical HPV infection to cervical neoplasia and invasive cancer [38
]. The HIV seroprevalence (12%) found was still lower than that reported from other areas of Southern Africa [39
]. However, it could be foreseen that with this high STIs burden, the epidemic could rise dramatically if no effective STIs prevention is implemented. HIV prevalence estimates among ANC attendees in this area have reached 24% confirming this suspicion [40
]. The risk factors for STI detection identified are consistent with those observed in other studies [41
]. However, the indirect information obtained through questionnaires makes the data on behavioral risk factors subject to bias and difficult to reproduce [42
]. The main limitation of this study is that many women were not selected from the community and that the size and age distribution of the three recruitment sources are not comparable. These may limit the extrapolation of these findings to the general population. However, the similar rate of syphilis found among the community-based recruited women suggests that the potential selection bias may not be relevant.
Our data is still to date the only available information on the prevalence of STIs and cervical cancer among women in Mozambique. A study performed in 2004 in an urban area of Mozambique reported a remarkable decline in STIs, which was attributed to successful implementation of STIs prevention strategies in that area [31
We have to mention that one of the limitations of the study is that it may be underpowered for some of the tested associations.
In conclusion, the burden of disease associated with STIs borne by African women in rural areas, and the implications for the acquisition and transmission of HIV and cervical neoplasia, are enormous. Reproductive health programs do not easily reach adolescents and older women, so specific approaches to target them should be envisioned and implemented. In pregnant women, there is a need for a more aggressive approach to avoid the harmful effects associated with STIs on mothers and children. For HPV infections and cervical cancer, recent development of vaccines [44
] offers great hope, but the challenge is that they become affordable for developing countries. There is an urgent need to develop interventions in order to avoid the burden of morbidity due to STIs and the further spread of HIV in rural African populations.