Data sources

The main sources of data used for capturing Maternal and Newborn Health (MNH) quality indicators include:

  1. Vital registration
  2. Special surveys and studies
  3. Routine health information systems

These provide a range of quantitative and qualitative data and information which can be used for analysis. The following pages provide a brief overview of each, and of common merits and weaknesses.

Vital registration systems

A vital registration system is one that collects, collates, tabulates and processes data on vital events such as births and deaths. The data is usually captured by a civil registration system and co-ordinated by the National Statistics Office in a country.

Occasionally, special registries may exist of reportable diseases. For example, some countries may have a national obstetric registry system where all maternal deaths are reported. Perinatal registry systems also exist.

Unfortunately, the proportion of the world’s population covered by high quality vital registration systems remains low and in many low and middle income countries they do not exist (AbouZahr et al. 2007). Some countries may have a complete lack of a vital registration system (or a breakdown of one which was previously in place). In others, a lack of resources (or political will) means that the system is insufficient to properly register and collate the vital events which are registered. In addition to structural problems, where there is a weak vital registration system, people may be less inclined to register vital events at all.

Check your knowledge*: What reasons can you think of that may prevent people from registering vital events?

Take a closer look – Vital Registration Systems

Merits: Ongoing data collection and availability, relatively inexpensive. Data collection often standardised.

Weaknesses: Can be incomplete or inaccurate. Quality is dependent on the degree of missing, incomplete or inaccurate data.

Special surveys

Special surveys are those that are commissioned to collect specific information. There are a range of methods to collect data, including:

  • Census
  • Direct observation
  • Facility assessments
  • Population based studies using questionnaires, interviews or focus group discussions

The surveys can also use routine health information systems data and health service records, including administrative records such as out of stock cards, facility registers and clinical case notes.

Some surveys are designed to compare data across countries, for example the demographic and health surveys done by the DHS Progam. Some allow for in-country trend analysis through comparison with previous (same methodology) surveys.

Take a closer look – Special Surveys

Merits: Depends on the type of survey, but can collect rich data for specific purposes.

Weaknesses: Can be costly to undertake (in comparison with routine systems) or have to be separately financed.

Routine health information systems

Routine health information data are collected through the health system. Generally this refers to the health management information system (HMIS) although sometimes the term is used to include vital registration systems and maternity registries.

In the health system, data are usually obtained from health service records (including clinical case records, registers, handheld cards, administrative and service records). Data collation, cleaning and validation processes vary with country context, as do processes for analysis and dissemination.

Take a closer look – Routine health information systems:

Merits: Relatively inexpensive and efficient use of data compared to surveys. Provides countrywide data at all levels of the health system. Ownership by government and a wide group of stakeholders.

Weaknesses: Quality of data may not be assured. Data on coverage may be more available than indepth data on quality

The range of data may not be comprehensive, for example disaggregation by sex or age may not be possible when using routine health information data.

Data sources was last modified: June 15th, 2015 by Adrian Bannister

References - View all