This study was conducted between July 2012 and October 2013 at six sentinel, public health facilities in Lusaka Province, which were included in the pilot vaccine roll-out. Three facilities were in Lusaka District, the most densely populated, urban district in Zambia, and one additional facility was in each of the three remaining districts, Kafue, Chongwe and Luangwa, which contain a lower density, more rural population. Facilities were purposively selected. We selected facilities with in-patient departments, sufficient under-5 patient volume, space to support study activities, and geographical consistency with the pilot vaccine roll-out.

The study team conducted active surveillance for children presenting with diarrhea at each of the study sites. A child was considered eligible if they met the following criteria: age 0 – 59 months; admitted to the in-patient department (IPD) or under care in the outpatient department (OPD) at the time of screening; and caregiver verbal confirmation of child presenting with diarrhea (defined as the passing of three or more abnormally loose stools in the past 24 hours). Additionally, children were determined to be eligible if there was an indication of potentially severe diarrhea by confirmation of at least one of the following symptoms assessed through physical examination by the study nurse or verbal confirmation by the caregiver: sunken eyes, loss of normal skin turgor, intravenous rehydration prescribed/administered, blood in stool, and hospitalization for diarrhea or dysentery. Study nurses targeted enrolment of younger children, attempting to enroll approximately two-thirds of the sample from children under the age of 24 months.

As part of the larger evaluation of diarrheal disease, caregivers who voluntarily consented to participate completed a 3-part assessment: Part 1 immediately upon enrolment, targeted to occur as soon as possible after the child entered care, Part 2 at the child’s discharge from health services (same-day if an OPD patient), and Part 3 thirty days after enrolment. Part 1 data utilized for the vaccine effectiveness evaluation included individual and household demographics (e.g. age, sex, caregiver education level, family size), maternal health status, the child’s symptoms, history and treatment of the current illness and child health background. The child’s vaccination history and growth trajectory were recorded from the under-5 card (child health record). If the under-5 card was unavailable, verbal confirmation of vaccination status was requested from the care giver. The child’s health status during the study assessment was documented based on a combination of health facility record review for the current visit and a clinical examination of the child by the study nurse. A stool sample was collected for laboratory assessment. Part 2 included a review of the child’s health status and treatment received while in care. Part 3 determined child vital status at 30 days post-enrolment.

A combination of Parts 1 and 2 provided the 7 components required for assessment of the child’s disease severity using a Vesikari score (diarrheal stool and vomiting frequency and duration based on caregiver report, treatment based on admission status, temperature and dehydration based on clinical assessment).[13]

Stool specimens were refrigerated and transported from the study sites to the CIDRZ laboratory. They were tested for rotavirus antigen by enzyme immunoassay (EIA) using a monoclonal antibody solid phase sandwich. The Meridian Rotaclone EIA kit for the detection of rotavirus antigen in fecal samples (CAT No. 696004) was used for the analysis.

The study was approved by research ethics authorities at the Zambian Ministry of Health, the University of Zambia, the University of Alabama at Birmingham, and the University of North Carolina at Chapel Hill.

Children who presented with moderate to severe diarrhea who were age-eligible to have received rotavirus vaccine and had under-5 card confirmed vaccination records were included in the vaccine effectiveness analysis. Children whose stool specimen tested positive for rotavirus by EIA were classified as cases while those with EIA negative results were controls. A child was considered age-eligible for the vaccine if the rotavirus vaccine was available in the child’s health facility when the child was 6 weeks of age. A child was considered vaccinated if s/he received a minimum of one dose of vaccine at least 14 days before presentation to the study site.

Sociodemographic and clinical characteristics of cases and controls were compared using χ² analyses for categorical variables. For continuous variables, medians were compared using the Wilcoxon rank sum test. We reviewed age at vaccination, rotavirus vaccine coverage by rotavirus result, and rotavirus vaccine uptake compared to pentavalent vaccine (DTP, Hib and HeP B) uptake.

We used unconditional logistic regression, controlling for month and year of birth, month and year of admission, and clinical site to calculate the adjusted odds ratio (aOR) for rotavirus vaccination compared to no vaccination. To account for residual confounding related to age given the small sample size, we restricted our results to children aged 6 months or older. Vaccine effectiveness was expressed as (1-aOR) x 100.

Among all 1,506 children under the age of 5 years enrolled in the study in Lusaka, rotavirus detection between July 2012 and October 2013 showed two peaks, one around October (a hot, dry month before the rainy season) and one in June (a cool, dry month after the rainy season). (Figure 1.)

We excluded 904 children enrolled in diarrhea surveillance from the vaccine effectiveness analysis, of whom 885 (98%) were excluded because they were age-ineligible to receive vaccine either because they were enrolled in the study prior to vaccine rollout at their district or were too old to receive the vaccine. A total of 125 rotavirus positive cases and 404 rotavirus negative controls that had under-5 card-confirmed vaccination status were included in the case control analysis. (Figure 2)

Cases and controls were similar in age, gender, chronic disease prevalence, and household demographics. The rotavirus cases were more often brought in by a caregiver other than their mother and reported higher median expenditure on food and non-food items in the child’s household 30 days prior to the child’s presentation at the health facility. (Table 1)

Compared to antigen-negative children, those who were rotavirus positive had more severe diarrhea (including a greater duration and maximum number of diarrhea episodes), were more likely to have vomiting and for a longer duration, were in worse clinical condition on arrival at the facility, and had a higher Vesikari score compared to controls. (Table 2) Only 20% of cases and controls were hospitalized.

Vaccine coverage among cases and controls was high at 70% for at least 1 dose and 58% for 2 doses of rotavirus vaccine. While cases had lower pentavalent vaccine coverage, there were no significant differences between cases and controls for rotavirus vaccine coverage. (Table 3)

Rotavirus vaccination timing was consistent with recommended schedules, compared to pentavalent vaccine administration. (Figure 3) The median age at rotavirus and pentavalent vaccine administration was 6 weeks for dose 1 and 11 weeks for dose 2. Among children vaccinated against rotavirus, 79% received their first dose within 2 weeks of the recommended age for vaccination and 64% received their second dose within 2 weeks of the recommended age. Among children receiving pentavalent, 84% and 71% received their first and second doses, respectively, within 2 weeks of the recommended age.

Against rotavirus diarrhea of all severity, the adjusted two-dose vaccine effectiveness among children ≥6 months of age was 26% (95% CI, −30%, 58%). Against rotavirus diarrhea that required hospitalization, it was 56% (95% CI, −34%, 86%). Vaccine effectiveness against very severe rotavirus (Vesikari ≥15) for children with two doses was 48% (95% CI, −163%, 90%). (Table 4)