Correspondence (letter to the editor): methodological problems.

W. Kirschner,K. Friese,A. Scheffler

DOI: https://doi.org/10.3238/arztebl.2011.0460b

2011-07-01

Deutsches Ärzteblatt international

Abstract:When considering the common problem of publication bias it is to be welcomed that the evaluation results from this study were published. In view of the fact that 1887 of 7469 pregnant women (25%) mentioned at least one abnormal test value, but only 803 (10.8%) sought a medical examination and only 642 (9%) were treated with whatever medications, it becomes obvious that a reduction in the preterm birth rate cannot always be achieved. One of the study’s problems is that it is impossible to figure out, even as a tendency, why vaginal pH self-testing is not efficacious. This is due to an accumulation of methodological problems in the intervention and evaluation. Interventions have to be well planned and controlled with regard to their structures, processes, and outcomes. The aim of planning and conducting interventions is to generate, in a case-control approach, intervention groups that are not selected by health or social status. In the reported study, the case group included a sample that was without exception self-selected. In the context of the study, it was not possible to control the structures and processes; the intervention was conducted under the pregnant women’s own direction. Whether and how the pH test glove was used was elicited only by means of a questionnaire. The control group was not questioned. Only by doing so it would have been possible, however, to recognize the selection in the groups and possibly to standardize these. The present study does not raise any doubt in the individual medical benefits of the glove. However, any further administration of the glove to study populations should always be accompanied by a robust evaluation.