Understanding the difference between data, benchmarking, and research

The members of the DOR work together to gather, verify and make available reliable datasets and provide benchmarking data. Members of the DOR may apply as individual investigators for access to the data for research purposes through the usual process, but with respect to MANA Stats, the Division of Research is primarily concerned with supporting the infrastructure necessary for reliable data collection, contributor support, creation and oversight of data review protocols, and facilitation of access to the datasets access.

Data (or dataset): In lay terms, data means the information gathered. It may be verified or reviewed by a team of trained data reviewers to ensure accuracy, but raw data alone does not contain any analysis of the information. For example, the 2.0 dataset contains all of the pregnancy, labor, birth, and newborn information recorded in the 24,000+ charts that midwives then entered into the MANA Stats database from 2004 to 2009.

Benchmarking: Benchmarking means the reporting of basic statistics for key items in a dataset. For example, one benchmark is the overall cesarean rate for all of the clients of midwives contributing the MANA Stats dataset, pooled together.  Midwives might use this information to compare their own homebirth or birth center practice data to the average rate reported by all midwives. Other benchmarking data include indicators of safety and rates of obstetric procedures and interventions. The DOR has been reporting benchmarks to Midwives Alliance members at annual conferences for several years.

Research: Research studies pose a question and systematically and investigate the question using data. For example, a research study might use data to answer a question like: “Is laboring at home in water associated with a lower risk of cesarean section, compared to laboring at home but not in water?” In this study example, the researcher would analyze all the birth records of clients who started labor intending to have a homebirth. Dividing them into the group who used water immersion vs. the group who did not, the researcher would then statistically compare the cesarean rates of the two groups to look for a pattern. The researcher might also compare the water-immersion group’s c-section rate to the c-section rate for similar women giving birth in a hospital without the opportunity for water immersion.