*2.3. Study Design and Procedure*

We analyzed the child-level longitudinal immunization records in the SEIR from 1 January 2019 to 31 December 2022, for all 30 districts of Sindh. Data from District Khairpur and District Dadu were not shown for children who had received their vaccinations in 2019, as the SEIR was launched in these districts in 2020. We extracted data related to the demographic profile (gender, age, and maternal literacy level), immunization history (vaccines, date of administration, and geo-coordinates of vaccine administration site); modality of immunization service delivery (fixed, routine outreach, or enhanced outreach), and geographical location of household (district, UC, urban vs. rural area, rural vs. remote-rural area and slums vs. non-slums) of children from the 2019–2022 birth cohorts enrolled in the SEIR. Out of 1130 UCs in the province, 464 were classified as urban, and 666 as rural. Within the rural UCs, 88 were classified as remote-rural UCs, and within the urban UCs, 89 were classified as slum areas. An slum area was defined as a contiguous settlement where the inhabitants are characterized as having inadequate housing and basic services. Slum UCs were defined as having >75% population living in poverty. The slum area analysis was limited to EPI-identified slums in the eight districts of Karachi and Hyderabad [18]. All slum UCs in Karachi and Hyderabad were in urban areas. Remote-rural UCs were classified according to the Government of Sindh's School Education and Literacy Department classification of hard-area UCs that were located in remote coastal, desert, or mountainous areas [19]. Remote-rural UCs were mostly concentrated in the eastern and western peripheries of the province; urban UCs were found within the cities of Karachi and Hyderabad; and the rest of the remaining UCs in the province were predominantly rural (Supplementary, Figure S1). In addition to the geo-location data, we also extracted the gender profile of vaccinators who used the SEIR across the province.
