Clinical opinionEducationThe use of a comparability scoring system in reporting observational studies
Section snippets
Comparability scoring criteria
Ideally, comparative analyses should take into consideration the following criteria regarding circumstances of health care. The definition of the criteria is shown in Table 1.
Checklist-based scoring system
We propose a scoring checklist system (Table 1) to evaluate the similarity or comparability of baseline characteristics between the comparison groups, thus allowing for more valid conclusions.
The first 4 criteria may be used for all comparative analyses (maximum score 8), whereas all 6 criteria may be used for historical control studies (maximum score 12). If both the historical control and study groups have the same geographic setting (score 2), the same health care setting (score 2), the same
Examples and statistical analysis
We propose that comparative analyses should properly account for differences in the comparability score (Table 1). In a regression framework to assess the efficacy of the treatment on the outcome, several options are available and one of them is a simple weighting of the analysis by the comparability score, derived from the criteria shown in Table 1. The degree of correction should be inversely related to the total score; weighting the analysis based on the comparability score may have no
Conclusions
We have described a comparability-based scoring system for circumstances of health care to determine the degree of comparability between comparison groups in observational studies. Using examples from the published literature, we showed that this scoring system is useful in assessing baseline similarities or dissimilarities between comparison groups in observational studies or studies using historical controls that then can be used to correct the statistical analysis and produce valid
References (21)
- et al.
Equity in maternal, newborn, and child health interventions in countdown to 2015: a retrospective review of survey data from 54 countries
Lancet
(2012) - et al.
Pain relief for childbirth: the preferences of pregnant women, midwives and obstetricians
Women Birth
(2013) - et al.
The attitudes of Canadian maternity care practitioners towards labour and birth: many differences but important similarities
Obstet Gynaecol Can
(2009) - et al.
The fetal biophysical profile in patients with premature rupture of the membranes: an early predictor of fetal infection
Am J Obstet Gynecol
(1985) - et al.
The use of fetal biophysical profile improves pregnancy outcome in premature rupture of the membranes
Am J Obstet Gynecol
(1987) - et al.
Maternal seizure disorder and risk of adverse pregnancy outcomes
Am J Obstet Gynecol
(2013) - et al.
Case-control studies: research in reverse
Lancet
(2002) Studying a study and testing a test. How to read the medical evidence
(2005)- US Department of Health and Human Services. Draft guidance for industry, clinical investigators, and Food and Drug...
- et al.
Regional differences in prescribing quality among elder veterans and the impact of rural residence
J Rural Health
(2013)
Cited by (0)
The authors report no conflict of interest.
Reprints not available from the authors.