gms | German Medical Science

Background/research question: Informed consent is an important need to conduct medical research ethically correct. In Switzerland, since 2014 the human research act (HRA) [1] regulates patient information and consent in research. Unique for a law on human research in Europe and within good clinical practice (GCP), the HRA of Switzerland requests informed consent for non-interventional further use of health-related personal data to conduct retrospective observational studies. An exemption is only possible on the base of Article 34, if it is impossible or disproportionately difficult to obtain consent. Several hospitals introduced a broad consent (general consent) allowing researchers to use health-related personal data generated through the regular treatment at the hospital for arising research questions. The aim of this study was to gain information on the process of getting informed consent for an observational retrospective cohort study compared to data available through the general consent as well as to discuss the meaning of disproportionate difficulties.

Methods: We analyzed results of consenting process within the Bern IVF cohort study. During set up of a control group we contacted eligible women for consent. We identified possible controls through the ultrasound department and extracted address data from medical records. We sent patient information, informed consent and an answer envelop by postal mail and asked them to send back the singed consent. For group A we made an effort to find out valid addresses in case of return and to remind participants (first postal reminder; second phone reminder). For group B we did nothing of that. We checked for conformity between general consent and study consent.

Results: In group A (n=389), 255 (65.5%) received a postal reminder and 227 (58.4%) a phonecall. 160 (41.1%) gave written consent and 14 per phone. 32 (8.2%) refused to participate, 30 (7.7%) were lost to FU and 153 did not replay. Group B (n=135), 17 (12.4%) gave written consent.

Conclusion: Response rates are low in group A and extremely low, if no effort is made. This reduces data available for research and raises questions about potential selection bias. We need to discuss what we consider as a disproportionate effort and how we can facilitate these processes by providing actual address data or a wider interpretation of the general consent.

Competing interests: The authors do not have any conflict of interests