Skip to main content
Download PDF

Prevalence and associated factors of condomless sex among adolescents and young adults in Liberia: a multilevel analysis using data from the 2019-2020 Demographic and Health Survey

BMC Public Health volume 25, Article number: 1511 (2025) Cite this article

Abstract

Background

Condom use is an essential component of strategies to improve the sexual and reproductive health of adolescents and young adults (AYA). However, it remains a challenge for many Sub-Saharan African countries, including Liberia. This study aimed to examine the effects of individual and contextual factors on condomless sex within the past 12 months among AYA in Liberia.

Methods

A secondary analysis was conducted using data from the 2019-2020 Liberia Demographic and Health Survey (2019-20 LDHS). Sexually active AYA were included in the study. A simultaneous assessment of the effects of individual and community characteristics on unprotected sex was conducted using a multilevel mixed-effects logistic regression model. The adjusted odds ratios (aORs) and their 95% confidence intervals (CIs) for condomless sex were estimated.

Results

Of the 2,260 AYA included in the analysis, 68.3% were female, and 40.0% were living in poor households. Their mean age (± SD) was 19.3 (±2.6) years. Only 31.6% reported a history of HIV testing. The prevalence of condomless sex was 83.1%. Individual and contextual factors explained 71.4% of the variation in condomless sex among AYA. In the multivariable analysis, condomless sex was less likely among males (aOR = 0.36 [0.27–0.47]), those with moderate (aOR = 0.67 [0.48–0.94]) or high (aOR = 0.46 [0.31–0.67]) media exposure, and those with occasional partners (aOR = 0.61 [0.39–0.96]). Having a professional activity was associated with higher odds (aOR = 1.52 [1.17–1.97]). Contextual factors associated with lower odds included high community-level education (aOR = 0.65 [0.43–0.98]), urban residence (aOR = 0.54 [0.37–0.78]), and living in South-Eastern B region (aOR = 0.52 [0.3–0.93]; reference = North-Western).

Conclusion

The study shows a high prevalence of condomless sex in Liberia. Condom promotion strategies must take into account individual and contextual factors such as gender and regional inequalities.

Peer Review reports

Background

Sexually transmitted infections (STIs), including HIV infection, remain a major reproductive health problem in sub-Saharan Africa (SSA). The World Health Organization (WHO) estimates that in 2017, Africa had the highest regional prevalence of each of the major curable STIs, including syphilis, gonorrhea, chlamydia and trichomoniasis [1]. The human papilloma virus (HPV) and the herpes simplex virus (HSV-2), two incurable STI, are also widespread in that region [2]. Adolescents and young adults (AYA), particularly those living in SSA, have a higher risk of STIs [3]. This is due to their risky sexual behaviors, including early sexual activity, multiple sexual partnerships and condomless sex, as shown in studies conducted in several countries in this region of Africa [4,5,6]. In 2021, 1.7 million adolescents (10–19 years) were living with HIV/AIDS worldwide; among them, 90% were from the WHO African Region [7]. A study of demographic data from 27 countries in this part of Africa between 2010 and 2018 found that the prevalence of SR-STI among adolescent girls and young women was 6.62% [8]. In West Africa, several studies have reported the prevalence of STIs in AYA [8,9,10], including 14,1% for SR-STI among the adolescent girls and young women in Mali, 14,1% for SR-STIs among the adolescent girls and young women in Mali [9], 3.6% for human papillomavirus (HPV) in Côte d’Ivoire [11] and 21.96% for herpes simplex virus (HSV) in Nigeria [12].

But beyond the numbers, it is important to understand the factors that prevent young people from using condoms. These factors include the stigma associated with condom use, myths and misconceptions (such as the idea that condoms reduce pleasure or are only for casual sex), lack of knowledge, cultural norms, cost or limited access, and power dynamics within relationship [13].

The prevalence of risky sexual behaviors among AYA reported in the scientific literature in sub-Saharan Africa highlights the magnitude of this problem [14,15,16]. Indeed, a study conducted in Ghana in 2022 found that 79% of young women and 68% of young men did not use a condom during last sexual intercourse [14]. Several other studies on risky sexual behavior among young people have examined how factors at the individual level (such as gender, age, educational level, etc.) [17,18,19], as well as those at the contextual level (such as region, place of residence, etc.) [20,21,22], affect young people’s sexual behavior.

In 2022, 31,000 adults over the age of 15 years were living with HIV in Liberia, with a prevalence of 1.1% [23]. For the 15–24 age group, data on HIV and STI prevalence are limited in Liberia. In addition, this age group has increased biological vulnerability, making the study of these sexual behaviors even more important. Therefore, there is a need for the development of tailored strategies to change risky sexual behavior among AYA to curb the spread of HIV and other STIs. However, studies on sexual risk behavior, including condom use during sex in this group are rare..

This study aimed to assess the prevalence of condomless sex among AYA living in Liberia and to identify the individual and contextual factors associated with high-risk sex.

Methods

Data source and study design

This was a secondary analysis of data from the 2019-2020 Liberia Demographic and Health Survey (2019-20 LDHS). The 2019-20 LDHS is the fifth DHS conducted in Liberia and provides up-to-date estimates of key demographic and health indicators needed by program managers, policymakers and implementers to monitor and evaluate the impact of existing policies and programs and to design new health policy initiatives in Liberia. Data collection took place from 16 October 2019 to 12 February 2020. The sampling for the 2019-20 LDHS is based on a stratified, two-stage area survey. The first step was to select clusters constituting the enumeration areas (EAs), with a probability proportional to their size in each EA. A total of 325 clusters were selected. The second step consisted of systematic sampling of households. A household census was undertaken in all selected clusters. In each cluster, 30 households were selected by systematic equal probability sampling. The total sample size was 9745 households. The results from this sample are representative at the national, urban (Greater Monrovia and all other urban areas) and rural levels, including each of the five regions. In each household, all females aged 15–49 and males aged 15–59 who were either permanent residents of the selected households or visitors who stayed in the households the night before the survey were interviewed. Structured questionnaires were used to collect data from respondents through face-to-face interviews. A total of 29,014 individuals aged 15–59 years, of whom 24,765 were women, participated in the survey. Details on the methodology have been described in the final report of the LDHS 2019-2020 [24].

The entire survey dataset was downloaded from https://dhsprogram.com/data/dataset/. A reconstruction process was carried out to obtain the final database used for this study. This was the standard individual dataset containing sociodemographic and behavioral characteristics of household members, as well as characteristics of the households that participated in the survey. In total, two datasets, including Individual Recode (IR) and Men’s Recode (MR), were used to construct the database for our analysis. For this analysis, males and females respondents aged 15–24 years and sexually active at the time of the LDHS survey, with valid data on the variable of interest, were included. Sexual behavior was assessed based on reports of male or female condom use at last sexual intercourse within the past 12 months. The total sample size for the age group was 3,949 (Fig. 1).

Fig. 1
figure 1

Flow chart of the study population

Full size image

Definition of variables

Dependent variable

The outcome of interest was condomless sex, a dichotomous variable defined as non-systematic use of a condom during the last sexual intercourse of the past twelve (12) months preceding the survey [18].

Explanatory variables

The following explanatory variables were selected from the DHS data:

Individual variables

  • Sociodemographic characteristics of participants: Age (15–19 years and 20–24 years), sex (female and male), marital status (in union and not union), education level (no formal education, primary, secondary and above), read perfectly (No, Yes), employment status (does not work, works), religion (Christian, Muslim, other), wealth level (poor, medium, rich), level of exposure to mass media (low, medium, high), gender of head of household (female and male).

  • Sexual behaviors: age of first sexual intercourse (before 16 years and 16 to 24 years), multiple sexual partners (defined as at least 2 sexual partners), last sexual partner (occasional sexual partner, boyfriend/girlfriend, spouse).

  • HIV knowledge (low, medium, high): This is a composite variable constructed from six questions on STI/HIV/AIDS. The HIV knowledge score (minimum = 0; maximum = 6) was constructed by assigning a value of 1 to true answers and 0 for false answers. Three questions addressed false modes of transmission (each coded 1 for No and 0 for Yes): 1) “Can you get AIDS through witchcraft or supernatural means?” 2) “AIDS transmission can occur by being bitten by a mosquito”, and 3) “AIDS transmission can occur by sharing food with a person who has AIDS”. The other three questions asked about knowledge to reduce the risk of contracting HIV (each coded 1 for Yes and 0 for No): 1) “using condoms would reduce the risk of contracting AIDS”, 2) “having only one sexual partner would reduce their risk of contracting AIDS”, and 3) “it is possible for a person who appears healthy to have the AIDS virus”. Thus, the level of knowledge was established as follows: low (score =< 2), medium (score between 3 and 4) and high ((score >= 5).

  • Sexually transmitted infection history variables: STIs in the last 12 months (No, Yes), ever tested for AIDS (No, Yes)

Contextual variables

Contextual variables are community-based variables. The LDHS-2019-2020 data include identifiers for the primary survey units (PSUs) where each respondent resided. Each PSU (cluster) had between 20 and 30 households. For the purposes of this study, clusters were considered representative of the community from which the individuals came. Thus, place of residence (rural, urban), region, and aggregate variables (education, poverty, and community-level media exposure) were included in this analysis:

  • Community-level education: Clusters whose median number of years of education is less than the median number of years of education of the study population are classified as having a low level of education. Otherwise, they are classified as having a high level of education.

  • Poverty at the community level: Clusters whose proportion of the poor is lower than the proportion of the poor in the study population are classified as having a low level of poverty. Otherwise, they are classified as having a high level of poverty.

  • Media exposure at the community level: Clusters whose proportion of mass media exposure is less than the proportion of mass media exposure of the study population are classified as having a low level of mass media exposure. Otherwise, they are classified as having a high level of mass media exposure.

Statistical analyses

A descriptive analysis was performed. The frequencies and weighted percentages were generated. The comparison according to the variable of interest was performed, and the association was tested using Chi-square or Fischer’s exact tests, if appropriate. The explanatory variables associated with the variable of interest with a p value of less than 0.25 were introduced into the regression model, thus constituting the main multivariable model. Additionally, variables such as age, which have a high risk of confounding, were kept in the model regardless of the degree of statistical significance.

At the multivariate analysis level, a two-level multivariable logistic regression was used. The first level examined the relationship between individual variables and condomless sex. The second level examined the effects of aggregate community-level factors on condomless sex. To assess the effects of individual and community characteristics on condomless sex, a two-stage mixed-effects logistic regression model is appropriate and was therefore fitted.

Multivariate analysis strategy

Four models were fitted as follows: Model 0 or the empty model. In this model, no effect of the explanatory variable on the dependent variable was applied, condomless sex. It is used to determine the initial within-cluster and between-cluster variance of condomless sex or the unconditional variance. If the variance is statistically non-zero, then a multilevel model can be conducted. Model I allows the block addition of individual variables to determine the effect of individual characteristics on condomless sex. Model II allows the block addition of contextual variables only. It assesses the effect of contextual characteristics on condomless sex. In Model III, individual and contextual variables were introduced simultaneously to determine their combined fixed and random effects on condomless sex. Variance inflation factors were estimated to assess the risk of multicollinearity between the variables [25].

The final model results were presented as adjusted odds ratios (aORs) with their corresponding 95% confidence intervals (CIs). Parameters such as the intraclass correlation (ICC) [25, 26], the median odds ratio (MOR) [26] and the proportional variation of variance (PCV) [26, 27] were estimated to assess the contribution of the random part of the model.

Results

Characteristics of the participants

Out of a total of 3,949 AYA, 2,260 individuals residing in 324 clusters were included in the analysis (Fig. 1). Table 1 presents the characteristics of the participants. The mean age (± SD) of the participants was 19.3 (± 2.6) years, and 68.3% were female. More than half of the participants (57.2%) had a secondary education or higher, 51.8% had a low literacy level and 49.8% were living in rural areas. In addition, 49.1% of these AYA were employed, and 40.0% were living in poor households. Most of these households were headed by men (56.7%). More than a fifth.of the AYA (20.7%) had a low level of exposure to the mass media.

Table 1 Characteristics of adolescents and young adults (N= 3274)
Full size table

Regarding sexual behavior, 83.1% of AYA reported condomless sex during their last sexual intercourse, and 18.2% reported multiple sexual partners (2 sexual partners or more). The median age at first sexual intercourse was 16 years. Approximately 56.2% of the respondents had good knowledge about HIV. However, only 31.6% had been tested for HIV, and approximately 26.7% of the participants reported a history of STIs.

In the bivariate analysis, almost all the sociodemographic characteristics were associated with condomless sex, with the exception of sex of the head of household, SR-STIs and history of HIV testing (Table 1). Being female and having a professional activity were associated with a high prevalence of condomless sex (p<0.001). The prevalence of condomless sex was lower among those with secondary education or higher (p<0.001) or those with high literacy (p<0.001). Low levels of mass media exposure were associated with a high prevalence of condomless sex (p<0.001). Regarding sexual behavior, AYA who had sex before the age of 15 and those with only one sexual partner were more likely to report condomless sex (p<0.001). Those for whom the last sexual intercourse was with a boyfriend or girlfriend and those who had been tested for HIV were more likely to report condomless sex (p<0.001). Good knowledge of HIV was significantly associated with a low prevalence of condomless sex (p<0.001).

At the contextual level, all characteristics tested were associated with condomless sex. High levels of poverty, high levels of exposure to mass media, and low levels of education were associated with a high prevalence of condomless sex (p<0.001). With regard to residence, those living in rural areas were more likely to report condomless sex at last intercourse (p<0.001).

Factors associated with condomless sex among AYA in Liberia (Table 2)

Table 2 Individual and contextual factors associated with unprotected sex among adolescents and young adults in Liberia (multilevel logistic regression)
Full size table

Model III was considered for determining factors associated with condomless sex among AYA, as it had the lowest AIC.

For individual factors, being male (aOR = 0.36 [0.27–0.47]), having moderate (aOR = 0.67 [0.48–0.94]) or high (aOR = 0.46 [0.31–0.67]) exposure to mass media (reference = low), and having an occasional partner (aOR = 0.61 [0.39–0.96]) were associated with lower odds of condomless. In contrast, having a professional activity was associated with higher odds of condomless sex (aOR = 1.52 [1.17–1.97]).

Among contextual factors, living in a community with a high level of education (aOR = 0.65 [0.43–0.98]), living in urban area (aOR = 0.54 [0.37–0.78]), and living in South-Eastern B region (aOR = 0.52 [0.3–0.93]; reference = North-Western) were associated with lower odds of condomless sex.

Random effect analysis and model fitness comparison

The random effect model’s assessment was conducted using ICC, PCV, and MOR. The prevalence of condomless sex varied significantly across clusters. In the null model, there was statistically significant variability in the odds of reporting condomless sex among AYA. In that, 13% of the variance in reporting condomless sex was explained by the variation in characteristics between clusters (ICC = 0.130). The within-cluster variation was reduced to 4.09% in model III, which included both individual and community factors. The variance in reporting condomless sex could therefore be explained by cluster differences.

Furthermore, the proportional change in variance (PCV) was found to be highest in the final model, indicating that both individual- and community-level variables accounted for 71.4% of the variation in condomless sex. The median odds ratio (MOR) in model III showed that if an AYA moved from a cluster with a low risk of condomless sex to a cluster with a high risk of condomless sex, the median increase in the odds of condomless sex would increase by 42% (MOR = 1.42) (Table 3).

Table 3 Assessment of multilevel models
Full size table

Discussion

This study, based on a nationwide survey in Liberia conducted in 2019-2020, reported condomless sex and identified its associated factors (individual and contextual) among AYA.

Indeed, the prevalence of condomless sex among AYA in this study was found to be higher (86.8%) than those reported in several studies in sub-Saharan Africa [4,5,6], where the prevalence ranged between 30% and 77%. This high prevalence could be explained by the fact that in African societies, sexuality remains taboo at this age and is generally not discussed much within the family [28]. Additionally, in some relationships, people may begin to neglect condom use after a few dates [29]. Factors contributing to this laxity include high mutual trust, lack of awareness of risks, increased comfort, and insufficient communication regarding condom use. Therefore, it is important to promote communication for change in young people, including with peers or parents, to improve awareness of safe or protective behaviors regarding sexual reproductive health.

The prevalence of condomless sex in this study varied between regions, with the exception of South East Region. AYA in the North Central South and South East B regions were less likely to report unprotected sex than those in the North West region. The North-West region is the region that houses the capital of Liberia. This seems paradoxical since in Africa, capital cities and their surroundings areas often have the greatest number of socio-educational and health infrastructures and better access to preventive sexual health care [30].

Other contextual factors, including place of residence, administrative region and community education level, were associated with condomless sex.

Rural residents were more likely to report condomless sex. This appears to reflect the disparity in condom availability between rural and urban areas [30, 31]. This is partly due to the low coverage of rural areas with sexual health prevention services. To reach the rural community, it is necessary to involve leaders in behavior change actions. In addition, there is a need to improve, including reproductive health services in primary health care facilities.

The level of community education influences condom nonuse. Indeed, the high level of education in the community is a factor in the low prevalence of condomless sex. Previous studies have shown that residential area is another key factor in condom use [30]. This fact seems obvious, as a community with a high level of education certainly has a sufficient level of knowledge to take control of its sexual health.

At the level of individual factors, the results showed that male gender, having a professional activity, wealth level of household, media exposure and type of sexual partner were associated with a decreased likelihood of condomless sex.

Male participants were less likely to have condomless sex compared to female participants. This result is in line with those reported in the studies conducted by Sunday et al. in Nigeria and Evans et al in South Africa [5, 6]. This gender difference in the prevalence of condomless sex could be explained by the existence of negative stereotypical representations of condom use that are still widespread in Africa, particularly among girls, who were considered promiscuous when they used condoms [32]. These representations may constitute real obstacles to condom use in this region. In addition, most of the time, girls are financially dependent on their male partner and therefore have little decision-making power, including whether to use condoms [30]. This finding could also support the fact that in the present study, respondents from wealthy households were less likely to have condomless sex than their peers from poor households.

The impact of professional situation on condom use is confirmed by this study. Indeed, AYA with professional activity were less likely to report condomless sex. This result is similar to the result of the Rwenge JM study, in which occupation in general did not play a negative role in condomless sex [4].

The influence of exposure of AYA to mass media in the household varies according to the level of exposure. Indeed, respondents from households with medium exposure were less likely to report condomless sex. This fact is certainly due to the information on sexual practices and sexual health disseminated by the mass media [30].

The main limitations of this study are the existence of social desirability bias and recall bias resulting from self-reporting of risk behavior. This inevitably leads to an under- or overestimation of behaviors. Furthermore, the study used cross-sectional surveys, so it is not possible to infer causality of the effect of individual and contextual factors on condomless sex [33]. Finally, with regard to contextual factors, as the duration of exposure to the community was not taken into account, it was impossible to assess a possible cumulative effect [25].

Nevertheless, the abovementioned limitations do not call into question the results of this work, as the data are from a nationally representative survey with high response rates. In addition, it has the merit of having highlighted individual and contextual factors to be considered in the design and implementation of interventions tailored to AYA to improve their condom use.

Conclusion

The prevalence of condomless sex in Liberia was high in this study. The inclusion of contextual factors in the analysis helped to highlight factors associated with condomless sex and to reduce the disproportionate importance of individual factors. These results provide a compass for developing strategies to reduce condomless sex by considering individual factors such as gender, occupation and household wealth. However, contextual factors such as place of residence, region and community education level should be considered. Measures such as sex education programs and behavior change communication are needed to reduce the prevalence of condomless sex. Further analysis of these data could provide insight into condomless sex among men compared to women. In addition, to better understand condomless sex, qualitative studies would be useful.

Data availability

Datasets are available from https://dhsprogram.com/data/available-datasets.cfm Liberia 2020 Standard DHS.

References

  1. Regional Committee for Africa, 67. Global health sector strategy on sexually transmitted infections 2016–2021: implementation framework for the African Region. World Health Organization. Regional Office for Africa; 2017. https://apps.who.int/iris/handle/10665/260232. Accessed 7 Sept 2023.

  2. World Health Organisation. Sexually transmitted infections (STIs). World health organization. 2023. https://www.who.int/news-room/fact-sheets/detail/sexually-transmitted-infections-(stis). Accessed 7 Sept 2023.

  3. Laurenzi CA, Du Toit S, Ameyan W, et al. Psychosocial interventions for improving engagement in care and health and behavioural outcomes for adolescents and young people living with HIV: a systematic review and meta-analysis. J Int AIDS Soc. 2021;24:e25741.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Rwenge JM. Comportements sexuels parmi les adolescents et jeunes en Afrique Subsaharienne Francophone et facteurs associés. Afr J Reprod Health. 2013;17:49–66.

    PubMed  Google Scholar 

  5. Adedini SA, Mobolaji JW, Alabi M, et al. Changes in contraceptive and sexual behaviours among unmarried young people in Nigeria: evidence from nationally representative surveys. PLoS One. 2021;16:e0246309.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Muchiri E, Odimegwu C, De Wet N. HIV risk perception and consistency in condom use among adolescents and young adults in urban Cape Town, South Africa: a cumulative risk analysis. S Afr J Infect Dis. 2017;32:105–10.

    Google Scholar 

  7. World Health Organisation. Adolescent and young adult health, 28 April 2023. 2023. https://www.who.int/news-room/fact-sheets/detail/adolescents-health-risks-and-solutions.. Accessed 7 Sept 2023.

  8. Dadzie LK, Agbaglo E, Okyere J, et al. Self-reported sexually transmitted infections among adolescent girls and young women in sub-Saharan Africa. Int Health. 2022;14:545–53.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Appiah CK, Dowou RK, Balame SK, et al. Self-reported sexually transmitted infections among adolescent girls and young women in Mali: analysis of prevalence and predictors. BMJ Open. 2023;13:e069226.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Seidu A-A, Agbaglo E, Dadzie LK, et al. Self-reported sexually transmitted infections among sexually active men in Ghana. BMC Public Health. 2021;21:993.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Tchounga B, Horo A, Boni S, et al. Human papilloma viruses infection among adolescent females perinatally infected with HIV in Côte d’Ivoire. Sex Transm Infect. 2021;97:238–43.

    Article  PubMed  Google Scholar 

  12. Abimbola O, Adedokun A. Sexually transmitted infections among young patients at two general outpatient clinics in Southwest, Nigeria: assessment of knowledge and risk factors. Res J Health Sci. 2021;9:369–77.

    Google Scholar 

  13. Engelbert Bain L, Amu H, Enowbeyang Tarkang E. Barriers and motivators of contraceptive use among young people in Sub-Saharan Africa: a systematic review of qualitative studies. PLoS One. 2021;16:e0252745.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Manu A, Ogum-Alangea D, Azilaku JC, et al. Risky sexual behaviours and HIV testing among young people in Ghana: evidence from the 2017/2018 Multiple Indicator Cluster Survey. Reprod Health. 2022;19:125.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Wado YD, Bangha M, Kabiru CW, et al. Nature of, and responses to key sexual and reproductive health challenges for adolescents in urban slums in sub-Saharan Africa: a scoping review. Reprod Health. 2020;17:149.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Seidu AA, Ahinkorah BO, Dadzie LK, et al. A multi-country cross-sectional study of self-reported sexually transmitted infections among sexually active men in sub-Saharan Africa. BMC Public Health. 2020;20:1884.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Challa S, Manu A, Morhe E, et al. Multiple levels of social influence on adolescent sexual and reproductive health decision-making and behaviors in Ghana. Women Health. 2018;58:434–50.

    Article  PubMed  Google Scholar 

  18. Odimegwu CO, Ugwu NH. A multilevel mixed effect analysis of neighbourhood and individual level determinants of risky sexual behaviour among young people in South Africa. Reprod Health. 2022;19:119.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Asrese K, Mekonnen A. Social network correlates of risky sexual behavior among adolescents in Bahir Dar and Mecha Districts, North West Ethiopia: an institution-based study. Reprod Health. 2018;15:61.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Abdul R, Gerritsen AAM, Mwangome M, et al. Prevalence of self-reported symptoms of sexually transmitted infections, knowledge and sexual behaviour among youth in semi-rural Tanzania in the period of adolescent friendly health services strategy implementation. BMC Infect Dis. 2018;18:229.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Somefun OD, Odimegwu C. The protective role of family structure for adolescent development in sub-Saharan Africa. PLoS One. 2018;13:e0206197.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Odimegwu C, Somefun OD, Chisumpa VH. Regional differences in positive sexual behaviour among youth in sub-Saharan Africa. J Biosoc Sci. 2019;51:254–72.

    Article  PubMed  Google Scholar 

  23. UNAIDS Liberia. Country factsheets Liberia 2022. UNAIDS. 2022. https://www.unaids.org/en/regionscountries/countries/liberia. Accessed 25 May 2022.

  24. Liberia Institute of Statistics and Geo-Information Services - LISGIS, Minsitry of Health - MOH, and ICF. 2021. Liberia Demographic and Health Survey 2019-20. Monrovia: LISGIS/MOH/ICF. https://dhsprogram.com/methodology/survey/survey-display-537.cfm. Accessed 25 May 2022.

  25. Abba MS, Nduka CU, Anjorin S, et al. Influence of contextual socioeconomic position on hypertension risk in low-and middle-income countries: disentangling context from composition. BMC Public Health. 2021;21:1–13.

    Article  Google Scholar 

  26. Merlo J, Chaix B, Ohlsson H, et al. A brief conceptual tutorial of multilevel analysis in social epidemiology: using measures of clustering in multilevel logistic regression to investigate contextual phenomena. J Epidemiol Community Health. 2006;60:290–7.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Walubita T, Beccia AL, Boama-Nyarko E, et al. Complicating narratives of sexual minority mental health: an intersectional analysis of frequent mental distress at the intersection of sexual orientation, gender identity, and race/ethnicity. LGBT Health. 2022;9:161–8.

    Article  PubMed  Google Scholar 

  28. Hien H, Somé DA, Méda N, et al. Caractéristiques de la communication parents-adolescentes sur la sexualité et le VIH à Bobo-Dioulasso, Burkina Faso. Santé Publique. 2012;24:343–51.

    Article  PubMed  Google Scholar 

  29. Ouédraogo S, Diallo I, Sarigda M, et al. Connaissances des infections sexuellement transmissibles et pratiques sexuelles des scolaires de la ville de ouagadougou au Burkina Faso. Revue Marocaine de Santé Publique. 2022;9. https://revues.imist.ma/index.php/RMSP/article/view/28258. Accessed 14 Oct 2023.

  30. Ali MM, Merdad L, Bellizzi S. Socioeconomic variations in risky sexual behavior among adolescents in 14 sub-Saharan Africa countries who report ever having had sex. Int J Equity Health. 2021;20:11.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Borges ALV, Duarte LS, Lay AAR, et al. Individual and context correlates of the oral pill and condom use among Brazilian female adolescents. BMC Womens Health. 21. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12905-021-01447-6. Epub ahead of print 2021.

  32. Aventin Á, Gordon S, Laurenzi C, et al. Adolescent condom use in Southern Africa: narrative systematic review and conceptual model of multilevel barriers and facilitators. BMC Public Health. 2021;21:1228.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Amegbor PM, Borges SS, Pysklywec A, et al. Effect of individual, household and regional socioeconomic factors and PM2. 5 on anaemia: a cross-sectional study of sub-Saharan African countries. Spat Spatiotemporal Epidemiol. 2022;40:100472.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

The authors of this paper acknowledge DHS for freely providing data to conduct this research.

Funding

The authors did not receive any funding for this study.

Author information

Authors and Affiliations

  1. Département d’Epidémiologie et Recherche Clinique, Institut Pasteur de Côte d’Ivoire, Abidjan, Côte d’Ivoire

    Kouamé Mathias N’Dri

  2. Université Bernard Lédéa Ouedraogo de Ouahigouya, Ouahigouya, Burkina Faso

    Ter Tiero Elias Dah

  3. Association African Solidarité, Ouagadougou, Burkina Faso

    Ter Tiero Elias Dah

  4. Université de Ouagadougou, UFR Sciences de la Santé, Laboratoire de Santé Publique, Ouagadougou, Burkina Faso

    Ter Tiero Elias Dah

  5. CHU SO Lomé, Unité de Recherche en Santé des Populations (URESAP), Lomé, Togo

    Aboubakari Nambiema, Bayaki Saka & Issifou Yaya

  6. Association de la Diaspora pour le Recherche et la Promotion de la Santé en Afrique (ADREPSA), Angers, France

    Aboubakari Nambiema

  7. Departement de Biophysique, Biostatistiques, Informatique, Unité de Formation et de Recherches en Sciences Pharmaceutiques et Biologiques, Université Félix Houphouët Boigny de Cocody, Abidjan, Côte d’Ivoire

    Djedjro Franck Renaud Meless

  8. Service de Dermato-vénéréologie, CHU Sylvanus Olympio, Université de Lomé, Lomé, Togo

    Bayaki Saka

  9. Société Togolaise de Dermatologie (SOTODERM), Lomé, Togo

    Bayaki Saka

  10. Patient-Reported Outcomes Research (PROQOL), Health Economics Clinical Trial Unit (URC-ECO), Hotel-Dieu Hospital, AP-HP, Paris, France

    Issifou Yaya

  11. ECEVE, UMR-S 1123, Paris Cité University, Inserm, Paris, France

    Issifou Yaya

Authors
  1. Kouamé Mathias N’Dri
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Ter Tiero Elias Dah
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Aboubakari Nambiema
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Djedjro Franck Renaud Meless
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Bayaki Saka
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Issifou Yaya
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

KMN: Conceptualization, data curation, formal analysis, writing of original draft, review and editing, ETTD: reviewing and editing, AN: reviewing and editing, DFRM: reviewing and editing, BS: reviewing and editing, IY: Conceptualization, supervision, reviewing and editing. All the authors have read and approved the final manuscript.

Corresponding author

Correspondence to Bayaki Saka.

Ethics declarations

Ethics approval and consent to participate

Non applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

N’Dri, K.M., Dah, T.T.E., Nambiema, A. et al. Prevalence and associated factors of condomless sex among adolescents and young adults in Liberia: a multilevel analysis using data from the 2019-2020 Demographic and Health Survey. BMC Public Health 25, 1511 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-025-22730-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-025-22730-3

Keywords

BMC Public Health

ISSN: 1471-2458

Contact us