Prospective comparison of two models of integrating early infant male circumcision with maternal child health services in Kenya: The Mtoto Msafi Mbili Study

As countries scale up adult voluntary medical male circumcision (VMMC) for HIV prevention, they are looking ahead to long term sustainable strategies, including introduction of early infant male circumcision (EIMC). To address the lack of evidence regarding introduction of EIMC services in sub-Saharan African settings, we conducted a simultaneous, prospective comparison of two models of EIMC service delivery in Homa Bay County, Kenya. In one division a standard delivery package (SDP) was introduced and included health facility-based provision of EIMC services with community engagement for client referral versus in a different division a standard package plus (SDPplus) that included community-delivered EIMC services. Babies 1–60 days old were eligible for EIMC. A representative sample of mothers and fathers of baby boys at 16 health facilities was surveyed. We examined differences between mothers and fathers in the SDP and SDPplus divisions and identified factors associated with EIMC uptake. We report adjusted prevalence ratios (aPR). Of 1660 mothers interviewed, 1501 (89%) gave approval to contact the father, and 1259 fathers (84%) were interviewed. The proportion of babies circumcised was slightly greater in the SDPplus division than the SDP division (27.3% vs 23.7%), but the difference was not significant (p = 0.08). In adjusted analyses, however, the prevalence of babies being circumcised was greater in the SDPplus division (aPR = 1.23, 95% CI:1.04–1.45) and the factors associated with a baby being circumcised were the mother having received information about EIMC (during pregnancy, aPR = 4.81, 95% CI: 2.21–3.42), having discussed circumcision with the father if married or cohabiting (aPR = 5.39, 95% CI: 3.31–8.80) or being single (aPR = 5.67, 95% CI: 3.31–9.69), perceiving herself to be living with HIV (aPR = 1.39, 95% CI: 1.15–1.67), or having a post-secondary education (aPR = 1.33, 95% CI: 1.04–1.69), and the father being Muslim (aPR = 1.85, 95% CI: 1.29–2.65) or circumcised (aPR = 1.34, 95% CI: 1.13–1.59). The median age of 2117 babies circumcised was 8 days (IQR: 1–36), and the median weight was 3.6 kg (IQR: 3.2–4.4). There were 6 moderate adverse events (AEs) (0.28%); 5 severe AEs (0.24%), all involving an injury to the glans penis, requiring hospitalization and corrective surgery; and one death probably related to the procedure. There were no AEs among the 365 procedures performed outside health facilities. Information and education campaigns must reach members of the general population, especially men and fathers, who are influential to the EIMC decision. Serious AEs using the Mogen clamp are rare, but do occur and require efficient, reliable emergency back-up. Our results can assist countries considering scale-up of EIMC services for HIV prevention as their adult VMMC programs mature.