Table 1.

Risk of bias assessment criteria for individual studies

No.	Criteria	Attributes	Risk of bias
1.	Definition of maternal mortality	❖ ICD-10 maternal death definition [85], or similar	Low
		❖ No or unclear definition provided	High
2.	Definition of neonatal mortality	❖ Death of live-born within the first 28 days of life, or similar	Low
		❖ No or unclear definition provided	High
3.	Completeness of ascertainment of maternal and neonatal deaths	❖ Prospective recording of mortality data ❖ Mixed methods cross-referencing facility records ❖ Demographic surveillance system with frequent rounds ❖ Survey based on recall of maternal or neonatal deaths ≤ 6 months previously	Low
		❖ Survey using direct or indirect sisterhood estimation methods ❖ Demographic surveillance system with infrequent rounds.	High
4.	Completeness of ascertainment of live births	❖ Prospective recording of births data ❖ Use of census < 5 years old for live births	Low
		❖ Use of census ≥ 5 years old for live births ❖ Live births data source not stated or unclear	High
5.	Sampling technique/design	❖ Census ❖ Vital registration ❖ Survey using nationally representative sample ❖ Systematic analysis involving the use of data collected from the above method(s)	Low
		❖ Design or sampling techniques not stated or unclear ❖ Provincial or sub-national sample used	High
6.	Data quality	❖ Data provide enough information for the study	Low
		❖ Insufficient data provided or unclear	High