*2.2. Analytical Framework Development*

To collect relevant and useful information from the mapping, we developed an analytical framework based on the Gavi programmatic guidance [19] and the UNICEF mapping of JA reports [15]. Table 1 provides a list of the variables used to summarise the pro-equity strategies planned by each country to reach ZD/under-immunised children. Adapting the definition used by Dadari and colleagues [15], we defined a pro-equity intervention as any tailored or targeted approach designed to reach underserved/vulnerable populations or communities with immunisation. All other interventions planned to be implemented throughout the country or not targeted at the priority groups or areas identified as being most vulnerable were not included. We created categories of interventions by thematic areas (grouping interventions that were similar and had the same purpose) to analyse which types of pro-equity interventions countries planned to implement. We validated the categories against those used by UNICEF and the Gavi programmatic guidance to ensure they were comprehensive and aligned. When a thematic area was identified that did not fit in any of the existing categories, it was brought to an internal working group to determine whether to create new categories. A complete list of intervention categories and their definitions can be found in Table A1.


**Table 1.** Variables included in the analytical framework and their corresponding response type.

**Table 1.** *Cont.*


### *2.3. Searching and Data Extraction*

The search strategy consisted of a manual screening of specific sections of the HSS proposals and keyword searches. Data on interventions planned to reach the target populations were usually found in the "Objectives of the proposal" and/or "Description of Activities" sections. Different keywords were used to answer specific questions, such as "sustainability" and "gender" to identify whether those topics were addressed, for example. Relevant information was extracted into an Excel database, in which each row represented a different proposal and each column a variable of the analytical framework (see Table 1 for the variables included). One researcher extracted these data from all HSS proposals during February and March 2022. To run the correlation analysis in R, we also created a separate database listing all the pro-equity intervention categories included in the proposals.

#### *2.4. Data Analysis*

The quantitative analysis consisted of descriptive statistics (counts, proportions, and frequencies) and was conducted on Microsoft Excel PivotTables, version 2206. Furthermore, we performed a correlation matrix using the R expand function to find the correlations of interventions in each country. This was conducted to find which interventions are often planned to be implemented together in the countries. Lastly, we performed inductive and deductive thematic coding based on information and observations noted throughout data extraction for open-ended variables. The coding was conducted by constructing a matrix in Excel.
