(30) COVID-19 pandemic-induced changes in baseline characteristics of newly recruited intrauterine device users in an ongoing large multinational cohort study (EURAS-LCS12)

Background: During the COVID-19 pandemic, in some countries women were advised to use self-administered shorter-acting contraceptives rather than long-acting methods (e.g., intrauterine devices [IUD], implants) until usual healthcare access resumed.

Objective: To investigate whether government restrictions due to the pandemic had an influence on the recruitment intensity or characteristics of new study participants in an ongoing cohort study of IUD users.

Method: The EURAS-LCS12 study is an ongoing prospective multinational cohort study to assess the risk of unintended pregnancy with IUDs. The study population comprises women from ten European countries with a newly inserted IUD recruited via healthcare professionals during routine clinical practice. Potential selection bias was investigated by comparing baseline characteristics of women recruited before (i.e., 2019) and during (i.e., 2020) the pandemic. Standardized differences (SD) were calculated to measure the effect size between the two cohorts. The impact of government strictness due to the pandemic on recruitment intensity was assessed via the Oxford Stringency Index (SI).

Results: Recruitment numbers were lower and in high concordance with the SI within the first months of the pandemic. In total, 9,788 women were recruited in 2019 and 8,949 women in 2020. Overall, baseline characteristics were comparable between the two cohorts (SD < 0.2). Yet, the proportion of women aged < 20 years was slightly higher in women recruited during the pandemic (10.5% vs. 8.8%). Furthermore, women recruited in 2020 were more likely first-time IUD users (69.8% vs. 67.4%) and had a lower household income more frequently (57.8% vs. 51.7%) compared to 2019. However, the proportion of women living single was higher in 2019 than 2020 (27.2% vs. 23.8%).

Conclusions: Our analyses show no indication that the overall study conduct was affected by COVID-19 pandemic related government restrictions. Recruitment intensity dropped and was highly associated with the SI in the beginning of the pandemic however, there was no relevant evidence of selection bias in the study population.