Does model of care affect women's health and wellbeing in the perinatal period in Sweden?

Diverse aspects of care models have been shown to influence women's perception of pregnancy. One such model is the ‘continuity of care’ model, which is used in Sweden (Waldenström, 1998) and elsewhere around the world, including in the UK (Jenkins et al, 2014). In the Swedish context, the goal in providing continuity of maternal healthcare is to allow the same midwife to follow up with a pregnant woman during both the antenatal period and the postpartum visit.

Leahy-Warren et al (2021) reported that most women in the UK felt that they did not have a choice in the model or location of their maternity care, but felt they were involved enough in decision making, especially during birth. In Ireland, women who are able to access private maternity care have reported higher levels of choice and control than those who attend public maternity care, which has been shown to be the most influential factor in almost all choice and control measures (Leahy-Warren et al, 2021).

There are many determinants of maternal satisfaction, covering all dimensions of care, including structure, process and outcome, which affect how women perceive their general health (Srivastava et al, 2015). Structural elements may include a good physical environment, cleanliness and the availability of adequate human resources, medicines and supplies. Process determinants can encompass interpersonal behaviour, privacy, promptness, cognitive care, perceived provider competency and emotional support. Outcome-related determinants include the health status of the mother and newborn. Additionally, access, cost, socioeconomic status and reproductive history can influence perceived maternal satisfaction (Srivastava et al, 2015). Senarath et al (2006) reported that determinants of satisfaction include providing immediate mother-newborn contact, information after examination and counselling on family planning.

In Sweden, the majority of pregnancies are low-risk and are the domain of midwives (Swedish Society of Obstetrics and Gynecology, 2016). When examining factors that influence women's experiences and satisfaction with pregnancy and birth, self-rated health can be influenced by pregnancy complications (Floderus et al, 2008; Henderson et al, 2013; Henriksson et al, 2020).

Self-rated health is considered to be a simple and easy measure of general health (Bombak, 2013). For the purposes of this article, the definition of self-rated health was that used in Graviditetsregistret© (2018) (the Swedish pregnancy register) and earlier studies published based on data from the register. The register collects data based on the question ‘how do you perceive your general health?’ at three different time points during pregnancy and the puerperium. Women are asked this question at the registration visit (gestational week 8–12), about both their current health and their health in the 3 months before pregnancy, and are asked about their current health at the postpartum visit, at approximately 8–12 weeks postpartum.

Maternal healthcare in Sweden

At the start of the 1990s, Swedish municipalities were required to rapidly increase maternity healthcare provision, and so it became possible for private healthcare providers to establish themselves in the Swedish healthcare system. Private healthcare settings in Sweden provide similar services as government settings, with both models of care being funded by the Swedish tax system, providing care at no cost and following Swedish rules and regulations (Government Official Reports, 2015). Since the establishment of private care, women have the option to choose the antenatal healthcare provider that they prefer. However, not all municipalities offer private healthcare services (Table 1).

Antenatal care for pregnant women in Sweden has resulted in one of the lowest reported maternal and child mortality rates globally (Berg and Lundgren, 2010). This is also related to the free healthcare system, which is regulated by national recommendations, offered on a continuity basis by registered midwives at midwifery clinics in the community to all pregnant women (National Board of Health and Welfare, 2014). The recommendations state that women with low-risk pregnancies should be offered 7–9 planned visits to the antenatal clinic. Every visit is attended by a midwife, but women are guaranteed to encounter a maximum of two different midwives during their pregnancy (Petersson et al, 2014).

Aims

To provide the best possible support for parenthood, it is important to create an optimal psychosocial environment for families (Meng and Chen, 2014). Therefore, investigating self-rated health before, during and after pregnancy is important. Such research can be used to raise awareness of self-rated health among healthcare providers and mothers, promoting women's quality of life, as well as the bonding process between mother and child (Ertmann et al, 2021).

To the best of the authors' knowledge, no previous research has been published regarding how different models of maternity care affect self-rated health in women with low-risk pregnancies. Sweden has rigourous national registers that make high-quality data accessible for research (Graviditetsregistret, 2018). However, register data on self-rated health have only been used to a limited extent. The aim of the present study was to investigate self-rated health during the perinatal period in women with a low-risk pregnancy in relation to different models of care, public or private, in Sweden.

Methods

This study used a descriptive, retrospective design to review comprehensive register data for low-risk pregnancies between 2010 and 2015. Ethical approval and permission to use data from the register was provided in 2015; thus, the data for this period were analysed in the present study.

Inclusion criteria

Births by caesarean section, instrumental vaginal births and mothers with chronic disease (diabetes, chronic kidney disease, epilepsy, chronic hypertonia and cardiovascular disease) were excluded as not being low risk (Ny, 2007). To be classified as low-risk, women were required to have a body mass index of <30 at the start of their pregnancy, a normal obstetric history (no comorbidities) and be pregnant with one fetus. As the study involved the perinatal period, a low-risk pregnancy was defined as a pregnancy at term (gestation age: 37+0 to 41+6), resulting in spontaneous labour and the spontaneous birth of a baby of appropriate weight for the gestational age, where neither mother nor baby suffered complications during birth that required specialist care.

Data collection

The Swedish pregnancy register records information about a pregnant woman and her child from the initial registration visit at the antenatal clinic to the postpartum visit at 8–12 weeks postpartum, including birth and postpartum care (Graviditetsregistret, 2018).

In the reviewed time period (2010–2015), a total of 462 000 pregnancies were registered. Information about the pregnancy and care that is collected in the register is based on midwives' and physicians' documentation during pregnancy and birth. The records in the database consist largely of well-defined input variables, which have been validated repeatedly, showing an internal validity of 71.4–99.7% (Swedish Society of Obstetrics and Gynecology, 2020). The dataset was delivered without personal identification numbers or names.

All records included in the analysis contained documented self-rated health before, during and after pregnancy (n=320 430). Records with the answer ‘don't know’ were excluded, because this could either be given by the mother or the midwife. The final number of records included in the investigation was 167 523.

Sweden has 21 municipalities, 20 of which were represented in the data set. Only 15 municipalities offer both private and government healthcare for pregnant women (Table 1), but all complete data were included in the analysis regardless of healthcare model.

Data analysis

Descriptive analysis was used, followed by group comparisons and ordinal regression analysis. The data consisted only of categorical variables, and so the Pearson Chi-squared test was used to test for possible relationships between self-rated health and healthcare model (public or private). Background variables related to the organisation (number of clinic visits, number of midwives encountered, postpartum visit) and mothers' sociodemographic factors (age, parity, country of birth) were also included. To avoid the risk of mass-significance caused by multiple comparisons, a Bonferroni correction was used and differences with P≤0.001 were regarded as significant.

Ordinal regression analysis was run to investigate associations between self-rated health (before, during and after pregnancy) and the model of care (public or private, number of midwifery visits, number of midwives, postpartum visit). Self-rated health was categorised as very good, good, neither good nor bad or bad to allow for threshold comparisons in ordinal regression. The analysis was adjusted for age, parity and country of birth. Reference categories were chosen after verifying normal distribution.

Ethical considerations

Prior to obtaining access to the register data, ethical approval was granted by the Regional Ethical Review Board in Lund (approval number: 2016/234).

Results

The majority of pregnant women received care at government facilities (80.6%). Most were born in Sweden or other Scandinavian countries (84.1%), were >30 years old and multiparous (58.7%). Almost all (99.7%) women attended postpartum visits at midwife-led clinics. Further details are shown in Table 2.

The majority of women rated their health as very good or good before, during and after pregnancy, regardless of their model of care (Table 3). More women in the private sector rated their health as very good during the perinatal period than those enrolled in government care. Conversely, more women at government facilities rated their health as good compared to the private sector. Both private and government care had comparable results in terms of the number of women who rated their care as neither good nor bad or bad.

Results from multiple ordinal regression are presented in Tables 4–6. Primiparous women (odds ratio=0.81, P<0.001) and those who received private care (odds ratio=0.84, P<0.001) were significantly less likely to rate their health as bad before pregnancy. Women born outside Scandinavia (odds ratio=1.64, P<0.001), those aged <30 years (odds ratio=1.07–1.45, P<0.001) and >34 years (odds ratio=1.13, P<0.001) and women with 10 or more antenatal visits (odds ratio=1.24, P<0.001) were significantly more likely to rate their health as bad before pregnancy. Further details are shown in Table 4.

During pregnancy, primiparous women (odds ratio=0.52, P<0.001), those who had less than seven visits to midwives (odds ratio=0.96, P=0.044) and women who were cared for privately (odds ratio=0.84, P<0.001) were more likely to rate their health as good or very good. Several factors were associated with being more likely to rate health as bad. These were the number of midwives visited (>4: odds ratio=1.09, P<0.001), age (25–29 years old: odds ratio=1.03, P=0.027; <24 years: odds ratio=1.21, P<0.001) and country of birth (outside Sweden/Scandinavia: odds ratio=1.27, P<0.001). Further details are shown in Table 5.

Similar results were found for the postpartum period. Those born outside Sweden/Scandinavia (odds ratio=1.62, P<0.001), aged over 35 years (odds ratio=1.06, P<0.001), with >10 midwifery visits (odds ratio=1.21, P<0.001), with no postpartum visit (odds ratio=1.22, P<0.031) and who encountered >4 midwives (odds ratio=1.11, P<0.001) were more likely to rate their health as bad. Further details are shown in Table 6.

Discussion

The aim of the present study was to investigate self-rated health during the perinatal period in women with low-risk pregnancies in relation to different models of care in Sweden. The majority of women, regardless of model of care, rated their health as very good or good before, during and after pregnancy.

Women's perception of self-rated health was significantly different depending on the antenatal healthcare model they attended. However, looking at global research, a study from Kenya showed that women expressed satisfaction with the quality of services from admission until discharge in both public and private facilities, with no significant differences (Okumu and Oyugi, 2018). This calls for further studies on the association between satisfaction with healthcare services and self-rated health in different healthcare models. Baranowska et al (2020) reported that many predictors affect a woman's perception of and satisfaction with childbirth, regardless of whether they attended a public or private hospital; these predictors depended on the women's needs fulfillment during birth. Similarly, Kabakian-Khasholian et al (2017) examined women's satisfaction and perception of control in public hospitals in three Arab countries and found differences based on variation in and management of care. However, Jha et al (2017) examined postnatal Indian women's satisfaction with birth services at selected public health facilities in Chhattisgarh, India, and reported that there were gaps in the provision of care in certain areas, which may have impacted satisfaction with postpartum care.

In Sweden, all care offered in the perinatal period is free of charge, regardless of whether care is given by a private or government facility. The results of the present study should be interpreted with emphasis on equity, which is the long-standing basis for Swedish healthcare systems. Factors such as cost, convenience and accessibility influence service users' satisfaction and are therefore important for a successful maternity service (Swedish Society of Obstetrics and Gynecology, 2016). These services are grounded by the World Health Organization recommendations for balance between providing quality services and minimising harmful effects and adverse outcomes (Miller et al, 2016). This is important, as the outcomes of pregnancy and birth without complications may influence women's perception of their health (Cook and Loomis, 2012).

The results of the present study show that primiparous women and those cared for in the private sector were significantly less likely to rate their health as bad before pregnancy. Swedish healthcare pays particular attention to young mothers and parents by offering support groups during and after pregnancy and childbirth (Swedish Society of Obstetrics and Gynecology, 2016). Additionally, Sweden offers sexual and reproductive education for teenagers and young adults with the aim of decreasing sexually transmitted diseases and unwanted pregnancies (Swedish Association for Sexual Education, 2022). Furthermore, antenatal parental education is provided free of charge, offering women and their partners information about breastfeeding, childbirth, parenthood and the bonding process (Swedish Society of Obstetrics and Gynecology, 2016). This is perceived as a successful model with effective outcomes. However, the present study's results show that younger women had lower self-rated health compared to other groups, which was likewise shown in an earlier study in Sweden (Burström and Fredlund, 2001). It has also been shown that younger women may not use the available opportunity for parental education (Knowledge Center for Women's Health and Knowledge Center for Child Health Care, 2019).

Mothers should be able to build trusting relationships with midwives that make them feel safe and supported in the antenatal period; this process is encouraged if limited to one or two midwives. The present study shows a relationship between low self-rated health and encountering more than four midwives. It is possible that having several midwives limits the establishment of a deep relationship, which might lead to increased maternal stress (Sehhatie et al, 2014).

Continuity of caregiver means encountering the same care provider at each visit (Sandall et al, 2016), and requires the midwife to be a safe and knowledgeable individual (Jenkins et al, 2014). A Swedish study found that providing continuity in the form of a known midwife during birth led to higher satisfaction with intrapartum care among mothers (Hildingsson et al, 2020). Thus, when organising midwifery-led healthcare for pregnant women, it is important to consider limiting the number of midwives that a woman will encounter. There is evidence that a midwifery team that follows a mother during her pregnancy and birth could have an overall positive effect on the woman's perception of care, as well as her pregnancy and birth (Homer, 2016; Sandall et al, 2016).

In the present study, women born outside Sweden/Scandinavia rated their health lower than Scandinavian-born women before, during and after pregnancy. Private and government care is available to all women in Sweden during pregnancy, but immigrant women have been shown to underuse maternal healthcare services as compared to Swedish-born women and they attend later for their first visit (Petersson et al, 2014). This calls for attention from healthcare authorities and the public health sector to invest in the care of vulnerable groups. One method by which this can be done is the creation of multi-professional teams, including maternity healthcare providers, child healthcare providers, psychologists and social welfare agencies (Naidoo et al, 2019).

Living in a foreign country where healthcare is provided in an unfamiliar way may influence perception of health, regardless of the model of care provided, which may partly explain the present study's finding. Variables related to self-rated health, such as not being seen, not being treated with kindness and not having the possibility to be involved in care, are all factors that can affect women's self-rated health. In a South Korean population of immigrant women (Chae and Kang, 2021), women reported a high risk of poor self-rated health when they had unmet health needs, had been in the country for more than 15 years and had low life satisfaction. The findings of the present study highlight the need to shed light on foreign-born mothers' needs and expectations of the maternal healthcare in order to improve their self-rated health.

Swedish regulations for maternity care are the same regardless of whether accessing care privately or via a government institution, and care is always offered by midwives. The findings of the present review highlight differences in self-rated health between pregnant women attending these two models of care, and some associated sociodemographic factors. This calls for future studies that examine these differences in depth.

Strengths and weaknesses

One of the study's main strengths is the use of population comprehensive data to examine the role of organisational model in self-rated health of women in the perinatal period. This is one of the first studies of its kind in Sweden.

The study's main limitation is the increased risk for mass-significance, because of multiple comparisons and high sample size. Efforts were made to control for this by lowering the margin of error from 0.05 to 0.01 and 0.001. Nevertheless, the results showed several insignificant findings, which confirms the marginal effect of the sample size on the findings from regression analyses.

Conclusions

There is a gap in knowledge regarding self-rated health in pregnant women attending private and governmental health sectors in Sweden. There is a need for further studies to add to the current findings by investigating factors that influence women's choice of health facility. Vulnerable groups of women, such as younger and foreign-born women, need attention from healthcare authorities. On an organisational level, there remains a need to ensure that no individual woman encounters more than three midwives during the antenatal period.

Key points

• Measurement of self-rated health is an essential part of maternal healthcare in Sweden, measured at two key time points: the first antenatal visit and the postpartum visit, to evaluate experiences before conception, during pregnancy and postpartum.
• Self-rated health is an important indicator of a woman's quality of life, and may be affected by the maternity care model (whether public or private).
• Although in Sweden, the same maternal services are provided in both models, private healthcare settings led to higher self-rated health compared to government facilities.
• Further research is needed to explore factors affecting women's choice of facility.

CPD reflective questions

• How can self-rated health during the perinatal period be used to promote better quality of life for women and enhance the bonding process between mothers and newborns?
• How can accessibility be improved to support women's freedom of choice during the antenatal period?
• How can the number of midwifery visits be used to promote, enhance and maintain women's self-rated health, regardless of their background?