BACKGROUND: Fathers are intricately bound to the experience of adolescent mothers and their children. Yet, fathers of children born to adolescent mothers, particularly within the context of HIV, remain neglected in the literature. These exploratory analyses provide insight into the characteristics of fathers of children born to adolescent mothers affected by HIV in South Africa. SETTING: Eastern Cape Province, South Africa. DESIGN: Cross-sectional data from a prospective cohort study. PARTICIPANTS: Young mothers (10-24 years of age) and their children (0-68 months). All mothers completed detailed study questionnaires, including standardised and study-specific measures, relating to their self, their children and the fathers of their children. Summary statistics are presented based on maternal self-report of father characteristics. χ tests and t-tests (Fisher's exact/Kruskal-Wallis tests, where appropriate) were additionally used to explore sample characteristics (including father characteristics, maternal experience and child characteristics) according to paternal age and father involvement in childcare (defined by responses to four maternal self-report questions). Father characteristics were also explored according to maternal HIV status and maternal mental health status. RESULTS: 40% of fathers were adolescents (10-19 years) at the birth of their children. Overall, father involvement was low (19.5%). Compared with noninvolved fathers, involved fathers were more likely to be older when their child was born (21 years vs 20 years, t=4.30, p=0.04), to be in a relationship with the mothers of their children (74.8% vs 47.2%, χ=40.8, p≤0.0001), to reside with their children and their mothers (14.7% vs 3.7%, χ=49.3, p≤0.0001) and to attend the first antenatal appointment (4.3% vs 1.5%, χ=5.21, p=0.02). A quarter (25.4%; 227/894) of the adolescent mothers in the sample were living with HIV. The prevalence of maternal HIV was found to be higher among adolescent mothers of children born to older fathers compared with adolescent fathers (31.7% vs 15.9%, χ=28.3, p≤0.001). Likewise, depressive symptoms were more prevalent among adolescent mothers of children born to older fathers compared with adolescent fathers (9.9% vs 5.3%, χ=6.08, p=0.01). Adolescent mothers reporting poor mental health were less likely to be in a relationship with the fathers of their children (41.8% vs 54.1%, χ=7.32, p=0.03) and more likely to experience domestic violence perpetrated by the fathers of their children (8.2% vs 3.3%, χ=6.07, p=0.01) and to engage in arguments about finances with the fathers of their children (30.0% vs 17.0%, χ=10.8, p=0.001). While some differences in individual subscales were identified, overall composite scores of child cognitive development did not differ according to father age or father involvement. CONCLUSIONS: Analyses provide the first preliminary description of the fathers of children born to adolescent mothers affected by HIV in South Africa. Fathers are inherently tied to the experiences of adolescent mothers and their children. Father involvement with their children was low. Further research is required to explore the potential barriers to father involvement and pathways to overcome these. Efforts to bolster father engagement, such as the inclusion of fathers within maternal and child service provision, may have benefits for fathers, adolescent mothers and their children. There was a high prevalence of adolescent fatherhood in the study. Adolescent fathers may have specific needs requiring tailored intervention for adolescent parent families. The need for the inclusion of fathers within policy, programming and research remains.