96 4 CHAPTER 4 2022, 19, 1046 4 of 15 rics department to fill in the questionnaire once on behalf of their practice or department. Via the link to the questionnaire, professionals could accept or decline participation. Those who accepted participation were forwarded to the questionnaire. The primary midwifery care practices and obstetrics departments that did not respond to the invitation received a reminder e-mail in February 2020. In the first section of the questionnaire, we asked what practices and departments of which MCC they were a member and to list the other primary midwifery care practices that were a member of their MCC. In the case of membership to multiple MCCs, respondents were asked to answer the questionnaire with regards to the MCC with which they collaborated most. We used these answers to group the primary midwifery care practices and obstetrics departments into MCCs. The data was anonymized and stored in a secured, password-protected digital storage system at the Department of Midwifery Science at the Vrije Universiteit Amsterdam. 2.5. Analyses First, we grouped the respondents at the level of MCCs using Excel software [34]. We combined the answers of the primary midwifery care practices and obstetrics departments that were members of the same MCC. The answers that were inconsistent or unclear were categorized in three categories: ‘contradictory’, ‘variation within the MCC’, and ‘unclear’. The category ‘contradictory’ was used if the answers among the respondents within a MCC contradicted. The category ‘variation within the MCC’ was used if members of the same MCC indicated that a variation in the division of tasks and responsibilities within the MCC existed. We used the category ‘unclear’ if the combined answers of the respondents were unclear. Then, we identified different MRS by grouping the different MCCs with the same division of tasks and responsibilities between primary care midwives and obstetricians. The answers about the level of satisfaction were combined in three ways: by calculating the average score of the primary midwifery care practices belonging to the same MCC, the average score of the obstetrics departments belonging to the same MCC, and the average score of both. We grouped the data at the level of MCCs in a two-stage process to minimize the bias. First, the researchers BG and CV independently categorized the data of the first five MCCs. This resulted in some disagreements about how to categorize contradictory and unclear answers and answers indicating different policies within an MCC. After BG and CV reached consensus on the classification, they independently combined the data of another five MCCs, resulting in no differences. The remaining data were categorized by BG. The data were analyzed using STATA software [35]. We used descriptive statistics (n,%) to report the characteristics of the MCCs, the organization of tasks and responsibilities of primary care midwives and obstetricians within MCCs, the distribution of MRS across regions, and the level of satisfaction of MCCs. To examine the distribution of different MRS across regions, the MCCs were divided into five regions: ‘north’, ‘east’, ‘south’, ‘southwest’ and ‘northwest’ [36,37]. The level of satisfaction was categorized in binary categories based on the mean score for level of satisfaction for primary care midwives within a MCC, obstetricians within an MCC, and the overall score of an MCC. We combined the categories ‘very satisfied’ and ‘satisfied’ and the categories ‘a little satisfied’, ‘neutral’, ‘a little unsatisfied’, ‘unsatisfied’, and ‘very unsatisfied’ into the categories ‘very satisfied’ and ‘not very satisfied’ [38]. The chi-square test of independence and the Fisher’s exact test were calculated to examine the relation between the MRS and level of satisfaction. p-values < 0.05 were considered significant. 3. Results 3.1. Characteristics of the Included Primary Midwifery Care Practices and Obstetrics Departments The questionnaire was completed by 312 (55%) primary midwifery care practices and 53 (72%) obstetrics departments. Participation was declined by 68 (12%) primary midwifery care practices and seven (9%) obstetrics departments (Figure 1). To check the national representation in our study sample, in June 2020, we geographically mapped the primary