Long tradition of midwifery education in Sweden

Swedish midwifery education has a long tradition with more than 300 years of educated midwives. Historically, the Swedish midwife worked in women’s homes by assisting during labor and birth. Antenatal care was introduced during the 1930s and was fully developed at the same time that the majority of births were moved to hospitals. From having a good amount of continuity from a midwife caregiver, the care became fragmented. Swedish midwives usually work either at antenatal outpatient clinics or at hospital-based labor and postnatal wards. The field of midwifery currently covers all aspects of women’s sexual and reproductive health⁶.

Swedish midwifery education is a post-nursing education. After completing three years of nursing education at the university level with a Bachelor’s degree, an additional 18 months of midwifery education is required. Therefore, Swedish midwives have double authorizations in nursing and in midwifery. The midwifery education also includes a one-year Master’s degree in reproductive health. All Swedish midwifery students are required to assist at least 50 births before graduation⁶.

Swedish Association of Midwives, member of ICM

The Swedish Association of Midwives, founded in 1886, is a member of the International Confederation of Midwives (ICM), a non-governmental organization that has existed for over 100 years and currently includes 136 midwifery organizations from 117 countries. ICM defines a midwife as follows: ‘A midwife is a person who has successfully completed a midwifery education program based on the ICM Essential Competencies for Midwifery Practice and the framework of the ICM Global Standards for Midwifery Education, recognized in the country where it is located; who has acquired the requisite qualifications to be registered and/or legally licensed to practice midwifery and use the title “midwife”, and who demonstrates competency in the scope of practice of the midwife’. The basic competencies, first established in 2002 and thereafter updated several times, suggest that antenatal care includes assessing the health status of women and fetal wellbeing; monitoring pregnancy; and promoting healthy behavior and guidance. For intrapartum care, the competencies cover skills to promote normal, physiological and safe labor and birth and prevent complications; provide care for the newborn; and initiate breastfeeding. Postnatal and newborn care also require skills in detecting and treating complications in women and newborns^7-9.

COMICE reflects competencies of midwifery skills

Based on the basic competencies of midwifery skills developed by ICM, an international project was initiated in 2015. The project was named COMICE (Confidence Of Midwifery students on selected midwifery Competencies at completion of Education: a multi-centric study); a questionnaire was developed in an Indian province^10-12 and thereafter tested in seven African countries (Tanzania, Kenya, Uganda, Zimbabwe, Zambia, Malawi, and Somaliland)^13-15. The questionnaire comprises the areas of antenatal, intrapartum and postpartum/newborn care.

Rational

In most countries, the extensive questionnaire underwent a principal component analysis (PCA) to investigate underlying factors or components and find relevant dimensions. However, this analysis was not previously performed in Sweden. It is highly recommended that instruments are tested and validated when used in a new population¹⁶. The aim was to psychometrically test and adapt the COMICE questionnaire in a Swedish context and identify the underlying components of the instrument.

Study design and setting

A cross-sectional national study with an adaptation and psychometric evaluation component where students enrolled in all Swedish midwifery programs (n=13) were invited to participate. All 13 midwifery programs in Sweden participated in the study. The HEIs providing midwifery education were situated in universities belonging to a medical faculty (n=7) or at universities without a medical faculty (n=6). Data collection took place from January 2016 to January 2017 to coincide with the semester endings (January and June).

Participants

A convenience sample of all midwifery students in Sweden, in their final semester of their education in 2016–2017 who were present during the time of data collection were eligible and invited to participate.

Data sources and measurement

After receiving information about the study through the Swedish Midwives Association’s Educational Board, a formal request was sent out to the program directors at the universities, who accepted to participate in the recruitment and negotiated about the best period for data collection. One of the authors (LB) travelled to all the universities and was available in the classrooms for clarification and questions about the content of the questionnaire when the students were filling it in. It took approximately one hour to complete the questionnaire. The COMICE questionnaire was used to investigate the Swedish midwifery students’ confidence prior to their final exam. The questionnaire measured midwifery students’ self-assessment of confidence against four selected areas of ICM competencies: antenatal (22 skills), intrapartum (48 skills), postpartum (16 skills) and newborn (19 skills) care.

The origin of the questionnaire comes from an Indian study by Sharma et al.¹⁰. It was initially translated from the ICM skills statements in English to Gujarati, then reviewed by a group of midwifery teachers who rated the skill statements. The questionnaire was subsequently pilot tested with a group of students¹⁰.

Adaptation of the COMICE questionnaire

To translate and transform the COMICE from English to the Swedish language and cultural context, we were inspired by the back-translation methodology of Brislin¹⁷. Accordingly, several principal steps were taken to secure a proper, idiomatically correct and functionally equivalent Swedish version of the COMICE (Table 1). After obtaining permission to use and translate the questionnaire into Swedish, the transformation was agreed upon after discussing wording and concepts in relation to education and language understanding. Thereafter, a blinded back-translation from Swedish to English was performed. Finally, comparisons were made between the two versions and thereafter pilot tested.

In addition to the ICM competencies displayed in the COMICE in relation to confidence, some background information was collected. These questions covered the students’ age, sex, whether they had children, years of experience as a nurse and whether their midwifery school was connected to a medical university.

Ethics

The study was approved by the university ethics committee (Ref. 2015/1850). The respondents were informed that only the researchers had access to the data, and that the findings would be published without identifiable information on the participants. The students were free to refuse participation and could leave at any time without further explanation.

Analysis

Descriptive statistics were used to present the students’ background information (frequencies and percentages). The large data set was subjected to several principal component analyses (PCAs) for each of the areas of antenatal, intrapartum, and postpartum and newborn care, to reduce the number of items and gather information about the interrelationships among the variables¹⁸.

The Kaiser-Meyer-Olkin (KMO) value¹⁹ was first checked to assess the factorability of the data sets. The KMO value should be >0.6 for sufficient factorability, and Bartlett’s test of sphericity should be statistically significant (p<0.05)²⁰.

The number of components to retain for each of the areas was guided by Kaiser’s criterion (eigenvalue of 1) and Cattell’s scree test by inspecting the scree plots. Further, Horn’s parallel analysis²¹ was used to compare eigenvalues in each of the areas (antenatal, intrapartum, postpartum/newborn care) in the present sample with randomly generated data sets of the same size, where only those eigenvalues exceeding the corresponding values from the data sets were retained. All items in the questionnaire with an eigenvalue of 1 and a loading >0.40 were included. Cronbach alpha coefficients were thereafter calculated for each component to assess reliability. All items in each component were thereafter summed to form subscales. The subscales in all areas were analyzed in relation to students’ background characteristics through analysis of variance¹⁸.

Principal component analyses

Antenatal care

Different numbers of components were investigated in each of the areas until the best solution was found. The KMO value of antenatal care was 0.895 and Bartlett’s test of sphericity <0.001, suggesting good factorability. The initial rotated solution comprised three components. Horn’s parallel analysis, however, only confirmed the solution of retaining two components. The two components’ solution explained 40.08% of the variance. The first component covered nine items and was labelled: ‘Performing routine midwifery assessments during pregnancy’. The second component included eight items and was called: ‘Identifying fetal and maternal risk factors and educating parents’. The first component included common antenatal health check-ups, such as measuring the fundal height and listen to fetal heartbeat, while the second component included identifying medical and obstetrical health risks. Cronbach alpha coefficients calculated for each domain were 0.848, and 0.827, respectively. An overview of the included items, the pattern and structure matrix are shown in Table 2.

Intrapartum care

The KMO value for intrapartum care was 0.905 and Bartlett’s test <0.001. Three components were identified, and the KMO test indicated that the assumed samples were adequate and had good factorability. Horn’s parallel analysis confirmed the solution of retaining three components. This solution explained 42.21% of the variance. Cronbach alpha coefficients calculated for each domain were 0.917, 0.809 and 0.777, respectively. Table 3 presents the pattern matrix and structure matrix of the PCA for intrapartum care. The three components in intrapartum care were labelled: 1) ‘Managing physiological birth and providing support’, 2) ‘Managing complicated labor and birth’, and 3) ‘Identifying and providing lifesaving measures’. The first component comprised 19 items related to common midwifery tasks during an uncomplicated labor and birth, such as measuring uterine contractions, supporting women and assessing the progress of labor. The second component with 10 items involved hands-on manoeuvres during complicated labor and birth, as well as birth interventions such as performing an episiotomy. The third component with three items were about treatment of life-threatening conditions.

Postnatal and newborn care

The same procedure was performed for postnatal and newborn care. In the first attempt to perform a PCA, items related to only postnatal care were included in the model. It was suggested to retain two components, where one component only comprised one item (support to bereaved families). We also tried a forced one-component solution, but then this item did not load at all. Therefore, the items in postnatal and newborn care were collapsed, and the analysis revealed a solution of two components. The KMO value was 0.941 and Bartlett’s test of sphericity p<0.001, suggesting good factorability. A total of 50.88% of the variance was explained by the model. The two-component solution was confirmed by the parallel analysis. The two components were labelled: 1) ‘Managing uncomplicated postpartum and newborn care and educating parents’, and ‘Managing complicated postpartum and newborn care’. Cronbach alpha values for the two components were 0.883 and 0.858 (Table 4).

The first component with 18 items comprised routine postnatal assessments, information and education as well as initiating breastfeeding. Ten items came from the postnatal section [information and documentation (7), routine maternal check-ups (2), breastfeeding initiation (1)] and eight items from the newborn section [routine newborn care (5), information and documentation (3)].

The second component contained 11 items relating to identifying and acting on postnatal complications and supporting parents with special needs, such as giving birth to twins or triplets. In the second component eight of the 11 items came from the newborn section [identifying and acting on illness or emergency needs of the baby (6), and supporting parents (2)] and the rest of the items were from the postnatal section [identifying complications in mother or baby (2) and supporting bereaved parents (1)].

In the next step, the subscales formed from the components in all the three areas (antenatal, intrapartum and postpartum/newborn care) were analyzed in relation to students’ background characteristics (age, having children, civil status, number of births attended, length of experience as a nurse, medical faculty or not). Only age showed an association with certain components, with younger students (<31 years) reporting higher confidence in ‘Performing routine midwifery assessments during pregnancy’ compared to older students (≥31 years) [mean: 27.43 (SD=3.59) vs mean: 26.01 (SD=3.97), p=0.004]. Similarly, younger students showed higher confidence in the subscale ‘Managing physiological birth and providing support’ [mean: 65.41 (SD=8.25) vs 62.85 (SD=7.89), p=0.01], and finally in the subscale ‘Managing complicated labor and birth’ [mean: 20.37 (SD=5.38) vs mean: 19.05 (SD=4.80), p=0.046]. In the other subscales, there were no statistically significant associations with students’ background characteristics.

Antenatal care

Compared with the international studies that used the COMICE questionnaire, some similarities were found. The wording of subscales/components require fantasy and intuition. The two components in antenatal care were labelled: 1) ‘Performing routine midwifery assessments during pregnancy’, and 2) ‘Identifying fetal and maternal risk factors and educating parents’. In the original study from India¹⁰ where 633 students participated, three components were identified in antenatal care and labelled: 1) Assess maternal and fetal health, 2) Identify antepartum risks, and 3) Provide counselling and health education to women and families. The last two components are fairly similar to our second component. Midwifery might not be a strong contributor in the Indian study, where the midwifery education is incorporated with the nursing education and only 8% of the sample wanted to work within the field of midwifery after completed education¹⁰. The results from the study with 1407 participating students from seven African countries, where there are three types of midwifery education also provided components with a semantic similar wording as in the present study¹³. For antenatal care three components were identified in the African study: 1) Identify fetal and maternal risk factors, 2) Educate parents, manage and document emergent complications, and 3) Physical assessment and nutrition¹³.

Intrapartum care

The three components of intrapartum care identified in the present study were labelled: 1) ‘Managing physiological birth and providing support’, 2) ‘Managing complicated labor and birth’, and 3) ‘Identifying and providing lifesaving measures’. In the Indian as well as the African studies^10,14, intrapartum care was divided into two parts: the first and second stage of labor, and the third stage of labor. In both studies, the labelling of the components captured the physiological stages as well as the complicated stages, which is quite similar to the components in the present study. In both these studies, there were different types of education, and the result showed that students in direct entry programs and those with post-nursing education were more likely to be highly confident compared to students in integrated nurse-midwifery education^10,14. There were also some differences between the two studies. Students in the Indian study reported low levels of confidence; diploma students were 2–4 times more likely to have high confidence compared to students at the Bachelor’s level¹⁰. In the African study, Bachelor’s students and those within direct entry programs were more likely to be confident in intrapartum care¹⁴.

Postnatal and newborn care

In the area of postnatal and newborn care, the two components were labelled: 1) ‘Managing uncomplicated postpartum and newborn care and educating parents’ and 2) ‘Managing complicated postpartum and newborn care’. Both the Indian¹⁰ and the African studies¹⁵ divided postnatal and newborn care. The wording of the components was quite similar and described basic postnatal care, education of parents and postnatal complications^10,15. For newborn care, there was one component in the African study¹⁵ that differed from the Indian¹⁵ as well as the present study. One component comprised newborn care when the mother was infected with HIV. This component is country-specific due to the high prevalence of HIV in African countries²². Those particular items about HIV were omitted in the Swedish questionnaire, as HIV infections are not as common in Sweden.

Students’ background characteristics

We found no major differences between the students’ background characteristics and the components, with the exception of age, where younger students seemed to be more confidence compared to their older counterparts. In the Indian study, the 633 included students were much younger compared to the Swedish sample, with a median age of 21 years¹¹. In the African studies^13-15, nearly half of the 1407 students were aged 18–25 years. The median age of 31 years in the present study is therefore difficult to compare between the countries. There are indications that younger individuals might be more confident than older individuals in certain contexts^23,24, but other factors such as the quality of the education or individual learning styles might be more important. However, it has been suggested to the government²⁵ that Swedish midwifery education should change into a direct entry program; as such, the country would have a younger midwifery population that could serve society for a longer time, as they would not have to take the detour through the three-year long nursing education.

Strengths and limitations

This study is compromised by its cross-sectional design and the self-reported nature of the Swedish COMICE questionnaire, which limits the generalizability. Another limitation is the time span since the original study was conducted. The scope of practice and the midwifery education are, however, the same as they were during the data collection. The stepwise procedure used to translate and culturally transform the original COMICE into the Swedish language inspired by Brislin¹⁷ was a strength, even if it also resulted in the reduction of items that made no sense in a Swedish educational context. We believe that the essence of the COMICE is kept in the Swedish version. The comparison with the studies performed in India and seven African countries showed semantic similarities but are difficult to compare with the Swedish sample. In both India and the African countries, there are a variety of different types of midwifery education, and the length of the programs differs. Midwifery students’ confidence seemed to be higher, at least in the African studies^13-15, compared to students in the Swedish sample, and it is possible that cultural norms could make students give socially desirable answers as shown²⁶. Another strength of the study is the national representativeness, with all midwifery programs participating and the 76% response rate²⁷. The Indian and the African studies comprised much larger sample sizes. Nevertheless, the recommended ratio of items to participants was sufficient. Nunnally²⁸ recommends a ratio of 1:10, while Tabachnick and Fidell²⁹ suggest that 1:5 is enough. Translated into the present study, the number of items ranged from 17–32 in the areas studied, thus they are still within the recommended range. The Cronbach alpha values for the seven identified components ranged from 0.777 to 0.917, suggesting good reliability, as six out of seven components yielded a Cronbach alpha value of more than 0.80, which is regarded as preferable¹⁸.