The Danish MAMAACT trial as an example for discussing evaluation methods in complex interventions

Helle Johnsen, Trine Damsted Rasmussen, Ulla Christensen, Mette Juhl, Sarah Fredsted Villadsen

Publikation: Konferencebidrag uden forlag/tidsskriftAbstraktForskningpeer review


Workshop abstract
Over recent years, immigration to Europe has risen significantly. This development has led to increasing birth rates by non-Western immigrant women. Maternity care systems are not yet adapted to this increased diversity of women. In Denmark, 13 % of all children are currently born by women originating from non-Western countries. Despite a publically funded antenatal care, non-Western immigrant women have lower utilization of Danish antenatal care compared to other women giving birth.
Many non-Western immigrant women have an increased prevalence of severe maternal morbidity and higher risks of maternal death, stillbirth and infant death compared to the majority populations. Poor health status at birth can impair the cognitive, sensory and motor development and lead to learning disabilities during school age, and thus reduce equality in life chances. Suboptimal care is a contributing factor to these ethnic disparities. Provision of appropriate antenatal care is pivotal to reduce these serious disparities and challenges to public health, however little is known about models of care, which can overcome these barriers.
The MAMAACT intervention was developed to increase response to symptoms of pregnancy complications among maternity care providers as well as non-Western immigrant women. The intervention consisted of postgraduate training of midwives in intercultural communication and health education materials (leaflet and an app) for non-Western immigrant women describing how to respond to warning signs during pregnancy in six different languages, The intervention was implemented at 10 out of 20 Danish maternity wards as part of a national trial from the year 2018 to 2019, potentially reaching 25.000 pregnant women, including 2500 of non-Western origin.
This workshop shares insight into innovative ways of providing antenatal care for non-Western immigrant women and will focus on methodological aspects of scientific evaluation of complex interventions for vulnerable populations. The workshop includes an introduction, three presentations, and a discussion with the audience.
Using quantitative data, the first presentation will give an overview of ethnic disparities in stillbirth and infant death in Denmark. The second presentation will present qualitative data from the implementation evaluation of the MAMAACT intervention with a focus on the interventions program theory, methodological considerations and evaluation results. Finally, the third presentation will illuminate how the concepts of Health Literacy and cultural health capital contribute to an understanding of mechanisms leading to ethnic disparities and how they can be used for evaluating the impact of the MAMAACT intervention. The final discussion will stimulate knowledge sharing between the participants on how to combine quantitative and qualitative insights in trials, how to build partnerships and recruit vulnerable populations in evaluations, and how to use sociological theory in evaluations.

Message 1
Reducing ethnic disparity in reproductive health requires maternity care systems to rethink their antenatal care services so they are better equipped to support non-Western immigrant women’s needs.

Message 2
Sharing knowledge on the use of mixed methods, recruitment of vulnerable populations, and the use of sociological theory can contribute to future approaches to evaluate complex interventions.

Chairperson 1 : Sarah Fredsted Villadsen

Chairperson 1 email :

Chairperson 2 : Helle Johnsen

Chairperson 2 email :

No. of abstracts : 3

Presentation 1

Differences in stillbirth and infant mortality by maternal country of birth and descent in Denmark

Trine Damsted Rasmussen

Author / co-authors
T Damsted Rasmussen, 1
SF Villadsen, 2
S Smith Jervelund, 3
AM Nybo Andersen, 1
1 - Section of Epidemiology, University of Copenhagen, Copenhagen, Denmark
2 - Section of Social Medicine, University of Copenhagen, Copenhagen, Denmark
3 - Section of Health Services Research, University of Copenhagen, Copenhagen, Denmark

International studies have reported inequalities in stillbirth and infant death among ethnic minority groups compared to the host population. The same tendencies have been reported in Denmark. We investigated differences in the risk of stillbirth and infant death among offspring of immigrants and descendants in Denmark, compared to the risk among women of Danish origin.
The population-based register study included all live births and stillbirths with gestational age ≥ 22 weeks delivered by women of Danish origin, immigrant women and descendants in Denmark in the period 2005-2016. The study population was restricted to deliveries by women of Danish origin and country groups of immigrants and descendants with more than 2000 deliveries during the study period (n=792 705).
Logistic regression analysis adjusted for year of birth was used to estimate odds ratios (OR) with 95 % confidence intervals (CI) of the association between maternal country of birth and country of descent and respectively stillbirth and infant mortality.
Immigrant mothers from Turkey, Iraq, Somalia, Pakistan, Afghanistan, Syria and Iran had a statistically significant elevated OR of stillbirth compared to women of Danish origin; adjusted OR`s ranging 1.45-2.93. Danish-born women with respectively Turkish (OR 1.44, 95 % CI 1.00-2.07) and Pakistani descent (OR 2.32, 95 % CI 1.50-3.60) had an increased risk of stillbirth similar to the one among immigrant women with the same origin. For infant death, we found increased OR´s among immigrant women from Turkey (OR 1.76, 95 % CI 1.28-2.40), Somalia (OR 1.84, 95 % CI 1.31-2.58), Lebanon (OR 1.63, 95 % CI 1.03-2.60) and Pakistan (OR 2.85, 95 % CI 2.05-3.96). Only women of Pakistani descent (2.22, 95 % CI 1.28-3.86) had a statistically significant increased risk of infant death.
These findings show substantial differences in stillbirth and infant death according to maternal country of birth and descent in Denmark during the first decades of the 21st century.

Presentation 2

What can we learn from the national implementation of The MAMAACT intervention?

Helle Johnsen

Author / co-authors
H Johnsen, 1 - 2
U Christensen, 1
M Juhl, 2
SF Villadsen, 1
1 - Section of Social Medicine, University of Copenhagen, Copenhagen, Denmark
2 - Midwifery Programme, University College Copenhagen, Copenhagen, Denmark

Initiatives to improve maternal and child health among non-Western immigrant populations in Western countries are still sparse. So far, interventions within antenatal care have primarily focused on doula support and group-based antenatal care.
A qualitative evaluation of the implementation of the intervention was undertaken to explore midwives and non-Western immigrant women’s attitudes towards and experiences of using the MAMAACT intervention. The evaluation also investigated key contextual factors impacting the interventions utilization. Data consisted of focus group interviews with midwives (n=9) and in-depth interviews with non-Western immigrant women (n=15), as well as observations of midwifery visits (n=42). Data were collected at five intervention sites across Denmark and it was analyzed using systematic text condensation.
Findings show that midwives found the MAMAACT intervention to be a relevant tool, which was easy to implement in antenatal care. Several organizational factors such as task load, interpreter services and lack of cooperation with general practitioners impacted the implementation of the intervention. Non-Western immigrant women found the leaflet and app to be useful tools in distinguishing between normal and abnormal pregnancy symptoms. The degree to which the intervention was used by the women varied between the participants, and women also used other sources of information during their pregnancy. Contextual factors such as lack of social network and material resources as well as language proficiency impacted how women responded to their pregnancy symptoms. Findings show that in addition to targeting the intervention to midwives and non-Western immigrant women’s informational needs, structural factors such as the organization of antenatal care and women’s socioeconomic conditions need to be addressed in order to reduce ethnic disparity in reproductive health.

Presentation 3

Ethnic disparity in health literacy related to interaction with the health care provider

Sarah Fredsted Villadsen

Author / co-authors
SF Villadsen, 1
1 - Section of Social Medicine, University of Copenhagen, Copenhagen, Denmark

With the current orientation towards patient-centered care, the health literacy level is an important predictor for the engagement in preventive health care services. Research has shown that adult immigrants have lower levels of health literacy compared to the host populations, however little is known about the health literacy of pregnant, immigrant women. The aim of this study was to analyze whether ethnic minority women found it more difficult to actively engage with their health professionals than ethnic Danish women did.
Using cross-sectional data we studied the difference in the Health Literacy Questionnaire (HLQ) domain on actively engagement with health care providers between non-Western born and ethnic Danish women (n=406). Pregnant women were recruited in 2016 from the largest maternity ward in Denmark. The mean health literacy level of actively engagement for ethnic Danish women was 4.15 and for non-Western born women, it was 3.97. In an adjusted model the non-Western born women had 1.15 (CI:-0.30; -0.01) lower levels of actively engagement compared to ethnic Danish women.
These results indicate that this domain of the HLQ seems to be a useful link in understanding ethnic inequality in birth outcomes and thus relevant in the effect evaluation of the MAMAACT study. Currently, the health care system in Denmark has not focused on intercultural competence. The lower levels of self-reported ability to engage with health care providers and the known perspectives of lower patient safety in immigrant groups calls for more research in the interactional dynamics. A theoretical analysis using the concept of cultural health capital could be interesting for assessing how health care providers and non-Western women might value certain attitudes and behaviors exchanged during the encounter differently, which could lead to imbalance as the health care provider has more influence in this setting.
StatusUdgivet - 2019


  • Sygdom, sundhedsvidenskab og sygepleje