Article Text

Download PDFPDF

Short report
Do young women engage in greater sexual risk behaviour with biological fathers of their children?
  1. Michele R Decker1,
  2. Shang-En Chung2,3,
  3. Jonathan M Ellen4,5,
  4. Susan G Sherman4
  1. 1Department of Population, Family & Reproductive Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
  2. 2Division of General Pediatrics and Adolescent Medicine, Department of Pediatrics, Johns Hopkins School of Medicine, Baltimore, Maryland, USA
  3. 3Center for Child and Community Health Research, Johns Hopkins School of Medicine, Baltimore, Maryland, USA
  4. 4Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
  5. 5All Childrens Hospital, Johns Hopkins Medicine, St Petersburg, Florida, USA
  1. Correspondence to Dr Michele R Decker, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe St., E4142, Baltimore, MD 21212, USA; mdecker{at}


Objectives Qualitative research suggests that young women's relationships with the biological fathers of their children (BFRs), known colloquially as ‘baby daddy’ relationships, enable risk for pregnancy and STI/HIV. Our study compared partner characteristics and sexual risk within dyads based on BFR, among a sample of young women in Baltimore, Maryland, USA.

Methods We conducted secondary analysis of survey data collected in 2011–2013 from heterosexually experienced youth ages 15–24 in Baltimore, Maryland, USA. Analyses are limited to women with at least one recent (past 6 months) sex partner (n=171 participants, reporting on 271 relationships). Using generalised estimating equations with logit function for correlated binary responses, we evaluate associations of BFR with partner characteristics, sexual risk behaviour and contraceptive non-use.

Results At least one BFR partner was reported by 25.2%. Male partners in BFRs were more likely to have been incarcerated or arrested. BFRs were more often characterised by women as ‘main’ versus ‘casual’ partners (adjusted OR (AOR) 3.92, 95% CI 1.19 to 12.9). In adjusted analyses, BFR was associated with condom non-use for vaginal (AOR 12.3, 95% CI 3.92 to 38.7) and anal (AOR 3.32, 95% CI 1.34 to 8.22) intercourse. While BFR was associated with contraceptive non-use (AOR 2.21, 95% CI 1.01 to 4.84), this association attenuated to non-significance after adjusting for partnership type (AOR 2.06, 95% CI 0.91 to 4.67).

Conclusions While few differences in BFR partner characteristics emerged, significantly greater risk for unprotected intercourse was identified within BFR relationships. Findings suggest that the relationship context of a shared child heightens sexual risk for the young women most affected by STI.


Statistics from


In the USA, young women have the highest rates of chlamydia and gonorrhoea,1 and adolescent childbearing persists at 31.3 births per 1000.2 Adolescent sexual risk occurs within the context of broader environments; influences include partnership factors, gendered power dynamics, and family and community norms. Within these broader contexts, the partner relationship is highly relevant. Unique dynamics exist between mothers and the fathers of their children, giving rise to the term ‘baby daddy’.3 Definitions are inconsistent and varied; while popular media often depicts individuals who are no longer involved, the limited research available suggests that these partners often maintain some level of relationship, ranging from occasional sexual activity to cohabitating partnerships. Women's relationships with the biological fathers of their children (BFRs) are often characterised by sexual access and tolerance of sexual risk such as multiple partnering to maintain good terms for the child's sake.4 ,5 For example, quantitative evidence from urban adolescent mothers indicates they are significantly less likely to use condoms consistently if they are sexually active with the baby's father, and rate their motivation to use condoms as significantly higher with new boyfriends relative to their child(ren)'s father.6 Qualitative research with adolescent, African-American mothers indicates BFRs as virtually the only relationship for which condoms are not used.7 These dynamics hold distinct implications for STI, HIV and subsequent pregnancies. Yet surprisingly little research has quantified the extent to which BFRs confer unique risk for sexual risk. One study with STI-infected adolescent mothers found that BFR increased risk for condom and contraceptive non-use.8 Less is known about how BFRs compare with non-BFRs in broader populations of adolescents. No research to date has sought to disentangle potential differences in the partners within BFR from the relationship context of BFR. BFR partners themselves may differ from partners in non-BFRs, that is, BFR may be a marker for higher-risk individuals. Thus, analysis that compares relationships, and adjusts for both individual-level and partner-level confounders, is essential to understanding the extent to which BFR confers unique sexual risk for young women. Our study sought to compare partner characteristics and sexual risk across relationships based on BFR among a sample of heterosexually experienced young women.



This cross-sectional study uses survey data collected in 2011–2013 from heterosexually experienced male and female youth ages 15–24 in Baltimore, Maryland, USA (N=352) to understand gender norms, partner selection and STI/HIV risk among urban youth (SG Sherman, under review. The role of ethnicity, class, and gender norms in STI risks among adolescents in Baltimore, Maryland). Details are available elsewhere (SG Sherman, under review). A representative sample of low-class and middle-class African-American and white youth was recruited from census block groups, with some groups oversampled. Following informed consent and parental consent for minors, data were collected via audio computer-assisted self-interview. Eligibility criterion for the current analysis was reporting at least one sexual partnership in the BFR past 6 months. Given our aim, we restricted the sample to women. Of the entire baseline sample of women (n=220), a total of 171 reported a recent sex partner and are thus included in our analysis (final n=171 participants, who reported on a total of n=271 relationships).


All measures were self-reported. Respondent demographics assessed included age, race, highest level of parental education and age at first sex. Participants completed a partnership index for each sexual partner in the past 6 months. For each of the partnership in turn, participants reported on the following: having a shared baby (to determine BFR), casual or main partner, sexual behaviours, condom non-use for vaginal and anal sex, index concurrency (ie, participant reporting anal/vaginal sex with another while with a partner) and contraceptive use (ie, pregnancy prevention methods inclusive of hormonal (ie, oral, patch, injections and vaginal ring) and barrier methods (ie, condoms)). For each partner enumerated, participants self-reported that partner's age, enabling calculation of participant–partner age differences, perceived partner concurrency and that partner's risk profile (eg, HIV/STI history, sold drugs, gang involvement, incarceration).


BFR prevalence was calculated for the sample and by demographics using the participant as the unit of analysis; characteristics were compared between participants with at least one BFR partner and participants without BFR partner using χ2 and regression models. Subsequently, we contrasted partner characteristics and relationship type based on BFR (vs non-BFR) to explore potential differences in partners themselves based on BFR. Because the partner relationships are nested within individuals, differences were assessed via generalised estimating equations with exchangeable correlation matrix option and logit function for correlated binary responses. Separate models were constructed for each respective partner characteristic and adjusted for participant age and race as potential demographic confounders. Next we compared sexual risk within relationship based on BFR to explore potential differences in behaviour within the BFR context. Separate models were constructed for each respective sexual risk behaviour, adjusting for participant age and race. Finally, sexual risk models were then adjusted for partner characteristics and relationship-type factors significantly associated with BFR at p<0.05. The sample size varied with small amounts of missing data (<2%). All analyses were performed using SAS V.9.3 (SAS Institute, Cary, North Carolina, USA).


At least one BFR partner was reported by 25.2% (43/171 participants). BFR was slightly more common among older adolescents (28.0% among ages 19–24 vs 10.7% among ages 15–18; p=0.05). No other differences were detected based on demographic characteristics (data not shown). At the relationship level (n=271 relationships; table 1), BFR was associated with male partner having ever been arrested and incarcerated (adjusted OR (AOR) 2.01, 95% CI 1.04 to 3.88; AOR 2.45, 95% CI 1.22 to 4.94, respectively), adjusting for respondent characteristics. BFRs were more likely classified as ‘main’ versus ‘casual’ partnerships (AOR 3.92, 95% CI 1.19 to 12.9). No additional partner risk characteristics were associated with BFR. We compared sexual risk behaviour within the relationship and found BFR was associated with contraceptive non-use in analyses adjusted for demographics (AOR 2.21, 95% CI 1.01 to 4.84); this association attenuated to non-significance when additionally adjusting for partnership type (AOR 2.06, 95% CI 0.91 to 4.67). Adjusting for individual and partner characteristics, and relationship type, BFR was also associated with unprotected vaginal (AOR 12.3, 95% CI 3.92 to 38.7) and anal intercourse (AOR 3.32, 95% CI 1.34 to 8.22).

Table 1

Partner characteristics and sexual risk behaviour by relationship type (BFR vs non-BFR; n=271 relationships)


In this first study to explore BFR prevalence, one in four young, heterosexually experienced women reported a recent BFR partner. Findings from this analysis of relationships nested within individuals demonstrate significantly greater sexual risk behaviour in the forms of unprotected vaginal and anal intercourse based on BFR among the young women at greatest risk for STI in the USA. Though past qualitative reports characterise non-marital BFRs as tolerant of sexual risk,4 current results found comparable levels of partner concurrency in BFRs and non-BFRs. Together, the prevalence of partner concurrency (28.6% within BFRs), coupled with the elevated risk of unprotected sex, suggests significant and sustained STI/HIV risk to women in BFRs. While contraceptive non-use was also more common in BFRs, this association attenuated into non-significance after adjusting for main partnership status. These data underscore the relevance of relationship context, specifically the context of a shared child, in young women's sexual risk.

We found minimal evidence of elevated risk among male BFR partners themselves. Partners themselves in BFRs were comparable to non-BFR partners in most domains including partner age difference and partner STI/HIV infection, though male BFR partners were more likely to have been arrested and incarcerated. BFRs were more likely classified by young female participants as main partnerships; however, this characterisation may not be reciprocated by the male partner and does not imply monogamy.

Together, findings suggest that the BFR relationship context confers significant risk for subsequent STI/HIV and pregnancy, independent of individual and partner-level confounders. Qualitative evidence from adolescent mothers in non-marital BFRs speaks to a sense of inevitability of a continued sexual relationship without condom use,4 providing direction for understanding the current findings.

This exploratory study is characterised by several limitations. We lack a nuanced understanding of the type and nature of BFR, including marital and co-habitation status, and partner characterisation of the relationship. Sexual risk may vary across these factors. Statistical power was limited to detect differences for rare experiences such as partner injection drug use and gang involvement. Some estimates were unstable, for example, that for unprotected vaginal sex. Several additional risk determinants were not available, including partner condom refusal and intimate partner violence, both of which are common among adolescent mothers and associated with sexual risk.9 Lastly, we relied on participants to report partners’ risk behaviours.

Among the young women at greatest risk for STI/HIV, BFRs are common and may impart STI/HIV risk through unprotected intercourse. Findings confirm and extend qualitative reports of unique sexual risk dynamics where a shared child exists.4–7 By contrasting risk behaviour across relationships nested within individuals, findings extend beyond past analyses focused on individuals alone and suggest that the relationship context of BFR gives rise to risk for unprotected sex and shapes decision-making. With >270 000 births to adolescent women annually,2 findings hold significant implications. Sexual health interventions and family planning and STI/HIV providers must address BFRs as a context for unprotected sex. Policy efforts targeting young mothers for STI/HIV risk reduction and pregnancy prevention should address BFR dynamics to maximise impact.



  • Handling editor Jackie A Cassell

  • Contributors MRD conceptualised the analyses and led the writing. SGS and JME designed and implemented the parent study, provided input on analysis and contributed to writing. SC conducted analyses and contributed to writing. All authors reviewed and approved the manuscript as submitted.

  • Funding National Institute of Allergy and Infectious Diseases (P30AI094189), National Institute of Child Health and Human Development (R01HD057789).

  • Competing interests None declared.

  • Ethics approval Johns Hopkins Bloomberg School of Public Health Institutional Review Board (FWA # 0000287).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.