Characteristics of Study Participants
Between January 2010 and January 2013, there were 1624 infants registered in the EID clinic at Mbarara Regional Referral hospital. One third of these infants (33%) were excluded from analysis because the mother’s ISS clinic identification number to link the infant and mother’s data was missing in the EID record. Also, 1.6% of the remaining 1091 infants, was left out because they were dead or moved to other clinics i.e., they did not contribute towards obtaining the desired outcome. A cohort of 1073 mother-infant pairs enrolled into PMTCT and followed up for 18 months was therefore analysed (Table 1).
Proportion of HIV-exposed infants lost to follow-up
Out of the total cohort of 1073 infants analysed, 515 (48%; 95% CI: 45% to 51%) were LTFU by age 18 months; they did not complete follow-up to the point of being discharged and were not declared deceased. The findings revealed that lost to follow-up was 261 (24%) at 3 months, 308 (29%) at 6 months while 436 (40.6%) and 515 (48%) were lost to follow up at 12 and 18 months into care respectively (Table 2).
Proportion of mother to child transmission (sero-conversions)
Out of the 515 infants who were LTFU, 7 infants (1.4%) had tested positive for HIV and 508 had an unknown HIV status (had negative 1st or 2nd PCR results but didn't return for final testing). Of the 558 who were discharged, 20 infants (3.6%) tested positive for HIV and were referred for care into the ART clinic.
Factors associated with loss to follow up of HIV-exposed infants
Mother’s age, distance from home to hospital and family planning use were found to be associated with loss to follow up following the bivariate analysis conducted. Compared to mothers within the ages 18-23 years, p=0.002 (OR=0.6; 95% CI 0.44-0.82), mothers with age >30 years were least likely to have their infants lost to follow up. Again, those with 25-29 years were less likely to have their infants lost to follow up compared to mothers with 18-23 years of age, p=0.05 (OR=0.57; CI 0.42-0.76). Long distances from home that took mothers more than 1 hour to get to the hospital were associated with infant LTFU, p=0.05 (OR=1.31, 95% CI 0.99-1.73). Infants whose mothers were not using any family planning method were more likely to be lost to follow up, p=0.01 (OR=1.48, 95% CI 1.09-2.02).
Multivariable analysis indicated that young age of mother; long distance from home to hospital; and failure to use family planning methods were independently associated with loss to follow-up (Table 3). Compared to those infants whose mothers were in the 18-23 years age category, the odds of LTFU were lower among those whose mothers fell in the 24-29 age category. The odds of loss to follow-up were higher among babies whose mothers were reported to come from faraway places to hospital at enrolment in ISS clinic compared to those whose mothers were coming from near.
The odds of loss to follow-up were higher among babies whose mothers reported they did not use family planning compared to those whose mothers were using family planning at enrolment in ISS clinic (Table 3)
Findings from in-depth interviews
Factors identified from qualitative interviews were grouped facility-level and client barriers.
Facility level factors
Mothers pointed out waiting time as one of the reasons why they did not return their babies to the facility for follow-up. One respondent had this to say:
“The other thing is that we spend a lot of time here. You try to come early so that by midday you are going back home but you find yourself returning home in the evening. When it comes to lunch time, the health workers close and go for lunch, by the time they come back, time has already elapsed and yet you wanted to go look for some money to feed the children,” The young mother explained.
Waiting time was also mentioned by one health worker as a hindrance to child follow-up. A nurse who works with the HIV-exposed infants’ clinic when quizzed on why mothers were not returning their children for follow-up, submitted that: “It might be the long queue at the clinic especially for working class mothers.”
Client based factors:
Distance to health facility
Travelling long distances to the health facility was reported as a barrier to adhering to EID clinic schedules. A mother from Bushenyi had this to say:
“I come all the way from Bushenyi with a child. We spend a whole day here with nothing for my child to eat. Sometimes, the health workers are slow and by the time we are out of the queue, the child is already very hungry. So, when I thought about the long journey to and from the health facility and how very challenging it is, I decided to leave my child at home since I was not breastfeeding and the child’s first results were negative,” she justified.
Short childbirth spacing and multiple visits from single household:
When children are born too close to each other, this can affect their HIV care because the burden of carrying two children to care is enormous for the mother. This is worsened if mother and children have different follow-up dates.
“Like I said, my second child came a bit too early after the first. So, the first didn’t come to the clinic the number of times she was supposed to have come. She came in once. But the second one because she was sick, I had to bring her, and then bring myself, and then bring her… you couldn’t ask for the same day, you would get disturbed…being called here and there. So that was challenging. When my daughter was still sick, I had to bring her every month, then bring myself as well. But when she got better, that kind of changed. That’s why I didn’t even bother bringing her for discharge,” the young mother explained.