In this study, 34.5% study participants had sexual debut before ages of 18 years. This finding was much higher than the report of study conducted in China (17%) (17). However, it was lower (49.8%) than that of the study from Nigeria (27). This difference might be explained by the educational, economic, and healthcare disparities between countries. For instance, Chinese inhabitants might enjoy better socioeconomic prosperity and better healthcare facilities that provide SRH services than Ethiopia. In addition, the study from Nigeria was conducted among reproductive age group women while the present study only included young female. So, due to the age difference, the respondents from the study conducted in Nigeria might be more susceptible to recall bias than the present study participants. Thus, the study in Nigeria might overestimate the magnitude of RSB and could be a reason for the difference.
Nearly one-fourth (23.97%) of the study participants had multiple sexual partners during the last six months. Similar finding have been reported among female internal migrant workers in China (28).
This study found that 64.4% of the respondents had sex without condom or inconsistently used condom in their most recent sexual intercourse during the last six months. Likewise, the study conducted among unmarried female migrants from Shanghai reported nearly similar finding, (66.8%) (29). On the other hand, another study from China reported higher (84%) prevalence of sex without condom or inconsistent use of condom among female internal migrants workers (28). The discrepancy might be due to the fact that the study in China measured condom use over the last three most recent sexual intercourses while this study measured only the most recent sexual intercourse. This might underestimate the prevalence of sex without condom or inconsistent use of condom in the present study.
The magnitude of risky sexual behavior among study participants in this study was 71.9 %. This finding was lower than the results reported from Tiss Abay town of Amhara regional state of Ethiopia (80.4%) (21). The disparity could be due to the fact that the present study was conducted among all types of internal migrants while the study from Tiss Abay town was done mainly among rural to urban migrants. In addition, Tiss Abay town is semi-urban and the study showed that the migrants in small urban areas acquire reproductive health knowledge sooner than migrants in the metropolitan area (30).
More than one-third (34.5%) of the study participants watched pornography at least once over the last six months. This finding was higher than that of the study conducted in Jimma town of Ethiopia among preparatory students (25). This study was done among internal migrants while the study from Jimma town was among preparatory students who were combination of both migrants and non-migrants. This could be the reason for the difference because of the fact that an internal migrant has less control from family so that they could engage themselves in activities that deviate from their culture (31).
Nearly three-fourth (73.0%) of the study participants used social media and the majority (54.9%) of these used it many times daily. In addition, about (48.4%) of the respondents who were active on the social media indulged in sexting with men over the last six months. This finding was higher than the result of the studies conducted in the United States (32,33).
This study found that, one-fourth (25.8%) of the study participants were raped by their sexual partners or strangers over the last six months preceding the data collections period. The studies conducted in Northwest Ethiopia among street females (34) and that conducted in Southern Ethiopia among university students (35) found similar results even though they reported life time and one-year prevalence of rape, respectively.
Regarding the discussion and communication with family about SRH, only one-fifth (19.5%) of the study participants ever discussed about SRH with at least one of their family members. Even though the study population was different, an earlier study conducted among internal migrant children in Ethiopia reported that 21% of the migrant parents talked to them about HIV and only 11% reported discussions on sex (8).
The majority (156{58.4%}) of the respondents expressed the perceived risks of contracting HIV. However, this finding was lower than the result reported from China, (71.0%)(36). The difference might be that the present study used only young age groups while the study from China was conducted among all age groups. Older people might have higher risk perception than younger age group.
Even though it was only in bivariate analysis, age of the study participants has statistically significant association with sexual behavior, [COR 1.97(95%; CI; 1.15-3.37)]. After adjustment for potential covariates, age of the study participants has no statistically significant association with sexual behavior of the respondents’ [AOR 1.47(95%; CI; 0.41-5.37)]. Consistent with this study finding, younger age was reported as predictors of internal migrants’ risky sexual behavior on previous studies (36,37).
The unadjusted analysis showed that alcohol use was associated with risk sexual behavior of young female internal migrants, [COR 2.98(95%; CI; 1.69-5.23)]. The reason might be due to the nature of alcohol in decreasing attention to safe sex practices, altering rational decision-making and increasing risk-taking behaviors (38,39).
Recent technology is facilitating the tendency to have multiple sexual partners and practical sexual behavior as reported in previous studies (32,33). In this study, the respondents who shared sexually explicit texts, images or/and videos with their sexual partners over social media were about 3.5 times more likely to experience risky sexual behavior than those never sexted, [AOR 3.47(95%; CI;1.10-11.94)]. Consistent with this study finding, sexting of any kind was associated with higher rates of engaging in a variety of sexual behaviors but the impact of image-sexting are more prominent than pure texting (40). In contrary, study from United States reported that sexting was not associated with risky sexual behaviors (32). Further studies are recommended for a better understanding of the association between sexting and risky sexual behavior.
In bivariate analysis, the respondents who were forced to have sex over the last six months preceding the field data collection were 4.65 times more likely to have risky sexual behavior than their counterparts, [COR 4.63(95%; CI; 2.01-10.68)]. It is not surprising that rape victims were vulnerable to risky sexual behavior in previous studies (34,35).
Young women who had received sexuality education through schools, parents or other family members were less likely to engage in risky sexual behaviors in previous studies (41,42). Consistent with the former studies, unadjusted analysis showed that school-based sexuality and reproductive health information predict risky sexual behavior of the study participants in this study, [COR 0.52(95%; CI; 0.28-0.97)]; However, this association was not more established on multivariate analysis, [AOR 0.34(95%; CI; 0.07-1.79)].
Statistically significant association was established between respondents who exchanged sex for money despite the fact that there was no significant association on the adjusted analysis. The unadjusted analysis implies that the study participants who gave sex in exchange for money were nearly four times more likely to have risky sexual behavior than their counterparts, [COR 3.92(95%; CI; 1.49-10.34)]. Similar to this study, transactional sex was associated with risky sexual behavior according to the study conducted in Northwest Ethiopia (21).
Previous studies finding reported that the women who reported hunger were more likely to engage in transactional sex which was significantly exposed them to unprotected sexual intercourse(18,43). The unadjusted finding of this study also found statistically significant association, [COR 3.36(95%; CI; 1.57-7.19)]. A possible explanation for this result is that migrants are economically disadvantaged, so that they might have been participated in risky behavior to support their daily life.
The study participants who reported feeling of embarrassment to buy condom were about eight times more likely to have risky sexual behavior than their counterparts on adjusted analysis, [AOR 8.28(95%; CI; 2.10-32.62)]. In agreement with this finding, previous studies identified embarrassment to buy condom as a key risk factor in young people’s sexual behaviour (41,44). Moreover, the respondents who had negative attitude about the use of condom for steady and loving relationships were about six times more likely to have risky sexual behavior than those who had positive attitude about the importance of condom for the same purpose on multivariate analysis, [AOR 5.72(95%; CI; 1.47-22.24)]. Young people believe that condoms do not play a role in a relationship based on love and they are less likely to perceive themselves at risk of contracting HIV/AIDS or unwanted pregnancy according to the report from a previous study (45).
The women who reported ability to refuse sex without condoms were 85% less likely to have risky sexual behavior than their counterparts in this study, [AOR 0.15(95% CI; 0.04-0.57)].In agreement with this study, finding from Cameroon showed that self-efficacy was protective factor of sexual behavior (46).
The perceived risks of getting pregnant has statistically significant association with risky sexual behavior both on bivariate and multivariate analysis, [COR 0.21(95%; CI; 0.12-0.37)] and [AOR 0.05(95%; CI; 0.01-0.23)], respectively. This means that the women who reported perceived risks of getting pregnant were 95% less likely to have risky sexual behavior than their counterparts after adjustment for potential confounders.
Limitation of the study
This study results should be interpreted while considering several limitations. First, we cannot draw causal conclusions owing to the cross-sectional design of the study. Secondly, given that the sample population was limited to one city, the investigator could not assert that it is representative of all sexually-active unmarried female migrant workers in Ethiopia. Thirdly, this study was limited in its reliance on self-reported data because it is a sensitive topic; however, we tried our best to obtain unbiased responses. Despite these limitations, this study identified a range of major reproductive health issues related to young female internal migrants that affect their sexual behavior which was probably underexplored in Ethiopia by previous studies.