Incidence of Severe Acute Malnutrition After Treatment: A Prospective Matched Cohort Study in Sokoto, Nigeria 2018-2019

This dataset corresponds to the baseline questionnaire, which was the long questionnaire administered at the first home visit for both OTP-cured and community control children. The dataset contains data at the level of the household, as well as data at the level of the recruited child and their mother (which is also at the level of the household given that only one child per household was selected). The dataset includes the data from all modules of the baseline questionnaire except the household roster which is Section B of the household questionnaire.

This dataset has been anonymised for confidentiality purposes and therefore all identifying information have been removed from the data.

The unique identifier of this dataset is 'child_id'. This dataset can be linked to all other datasets using the unique child ID (child_id). It can also be linked to the health facility dataset using the facility ID (hf_id).

There are some non-response observations in the data. A response of 'don't know' is coded as '999' in the dataset; a missing response (i.e. a response should have been provided but was not) is denoted as '.a' in the dataset; and a skipped response (i.e. a valid non-response due to a skip in the questionnaire) is denoted as '.' in the dataset.

Note that in addition to the questions in the baseline questionnaire, the 'baseline_main' dataset also includes selected constructed indicators prefixed by n_ (these were constructed by data analysts using the survey data). These constructed indicators can be used to set up a survival analysis model and are included in the dataset to save data users time as they might require complex reshaping of data (but they could be generated by data users if preferred). The constructed variables include the following:
- The date of the OTP-cured child's recruitment into the study (on the day of their discharge from the health facility): 'n_date_recruit'
- The date of the baseline questionnaire for the OTP-cured child and the community control child (for the community control child this would represent their date of recruitment into the study): 'n_date_baseline'
- A dummy variable to indicate if the child developed SAM at any point during the 6 months follow-up, or if the child did not develop SAM (either lasted until the end of the study SAM-free, or dropped out of the study earlier): 'n_event_outcome'
- A dummy variable to indicate if the child died during the 6 months follow-up: 'n_died'
- A dummy variable to indicate if the child dropped out during the 6 months follow-up (household no longer consented to continue the study, or household moved away, etc.): 'n_dropout'
- A dummy variable to indicate if the child lasted until the end of the study period SAM-free: 'n_end_study'
- The number of the round (or visit) when the child exited the study (this would be the visit when the child developed SAM, or the final 12th visit if the child remained SAM-free, or the round before the child dropped out of the study and could no longer be traced): 'n_round_exit'
- The date of the interview at the round (or visit) when the child exited the study: 'n_date_followupexit'
- A variable indicating the number of days that elapsed from the date of joining the study until the outcome event (i.e. until either developing SAM, or dropping out, or lasting until the end of the study SAM-free): 'n_days'

Note that the datasets and questionnaires for the recruitment of the OTP-cured children and community control children have not been included (these questionnaires solely assessed the eligibility of the children to be recruited into the study). However, relevant indicators from these questionnaires have been included in the 'baseline_main' dataset. These are the sex of the child (ch_gender), the age of the child in months (ch_age), the level of education of the mother (ma_education), and the date of the recruitment of the OTP-cured child which is their date of discharge from the health facility (n_date_recruit).

Similarly, note that the data from the scanned OTP and Ration cards for the OTP-cured children has not been uploaded for public use due to data quality concerns. The data suffers from many missing observations, given that this data was not directly collected by the enumerators but relied on health facility staff filling in the OTP and Ration cards for the treated children. Only one variable from these cards is included in the uploaded data and this is the MUAC of the child at the time of their admission into OTP. This is represented by variable 'muac_entry' which is included in the 'baseline_main' dataset.

This dataset corresponds to the baseline questionnaire, which was the long questionnaire administered at the first home visit for both OTP-cured and community control children. The dataset contains data at the level of the household member and includes the data from the household roster which is Section B of the household questionnaire.

This dataset has been anonymised for confidentiality purposes and therefore all identifying information have been removed from the data.

The unique identifier of this dataset is 'child_id' + 'serial_number'. This dataset can be linked to all other datasets using the unique child ID (child_id).

There are some non-response observations in the data. A response of 'don't know' is coded as '999' in the dataset; a missing response (i.e. a response should have been provided but was not) is denoted as '.a' in the dataset; and a skipped response (i.e. a valid non-response due to a skip in the questionnaire) is denoted as '.' in the dataset.

This dataset corresponds to the baseline questionnaire, which was the long questionnaire administered at the first home visit for both OTP-cured and community control children. The dataset contains data at the level of the mother's births and includes the data from the reproductive history roster which is Section B of the mother questionnaire.

This dataset has been anonymised for confidentiality purposes and therefore all identifying information have been removed from the data.

The unique identifier of this dataset is 'child_id' + 'mb_10'. This dataset can be linked to all other datasets using the unique child ID (child_id).

There are some non-response observations in the data. A response of 'don't know' is coded as '999' in the dataset; a missing response (i.e. a response should have been provided but was not) is denoted as '.a' in the dataset; and a skipped response (i.e. a valid non-response due to a skip in the questionnaire) is denoted as '.' in the dataset.

This dataset corresponds to the follow-up questionnaire, which was administered at each follow-up home visit after the first (baseline) home visit for both OTP-cured and community control children. All follow-up visits for each recruited child are included in this dataset. The dataset contains data at the level of the household, as well as data at the level of the recruited child and their mother (which is also at the level of the household given that only child per household was selected). The dataset includes the data from all modules of the follow-up questionnaire.

This dataset has been anonymised for confidentiality purposes and therefore all identifying information have been removed from the data.

The unique identifier of this dataset is 'child_id' + 'round_fu'. This dataset can be linked to all other datasets using the unique child ID (child_id). It can also be linked to the health facility dataset using the facility ID (hf_id).

There are some non-response observations in the data. A response of 'don't know' is coded as '999' in the dataset; a missing response (i.e. a response should have been provided but was not) is denoted as '.a' in the dataset; and a skipped response (i.e. a valid non-response due to a skip in the questionnaire) is denoted as '.' in the dataset.

Note that not all modules from this questionnaire were administered at every follow-up. As mentioned in the protocol above, at every follow-up the child's MUAC was measured and data was collected on illnesses they may have experienced in the preceding two weeks and their breastfeeding status. However, at the exit round/visit (which is either the visit when the child has developed SAM or the final 12th visit if the child had remained SAM-free), additional modules were administered to the mother of the child to collect data on mother's employment, deaths in the household, household food security, mother's knowledge on feeding, mother's networks, mother's reproductive history, and child and household dietary diversity. The follow-up questionnaire indicates which modules and questions were asked at every follow-up and which were only asked at the exit round. Note that for the children that dropped out of the study either because household withdrew consent for participation or because household moved away, the exit round modules were not administered (that is because the enumerator did not know that the last round they visited the household was going to be the final one).

This dataset corresponds to the health facility questionnaire, which was administered once in each health facility on the first day the interview team visited the health facility to recruit OTP-cured children. The dataset contains data at the level of the health facility and includes data from all modules of the health facility questionnaire.

This dataset has been anonymised for confidentiality purposes and therefore all identifying information have been removed from the data.

The unique identifier of this dataset is 'hf_id'. This dataset can be linked to all other datasets using the health facility ID (hf_id).

There are some non-response observations in the data. A response of 'don't know' is coded as '999' in the dataset; a missing response (i.e. a response should have been provided but was not) is denoted as '.a' in the dataset; and a skipped response (i.e. a valid non-response due to a skip in the questionnaire) is denoted as '.' in the dataset.