Severe maternal morbidity among immigrant women in the Netherlands: patients' perspectives

Abstract

This 2006 study investigated ethnicity-related factors contributing to sub-standard maternity care and the effects on severe maternal morbidity among immigrant women in the Netherlands. In-depth interviews were carried out with 40 immigrant and 10 native Dutch women. The immigrant women reported that health care providers often paid insufficient attention to their pregnancy-related complaints, especially in cases of pre-eclampsia. They also reported delays in receiving information about diagnosis and treatment. Obstetricians who reviewed 20 of these cases judged sub-standard care to have played a role in the development of complications in 16 of them. The women themselves had problems identifying medically significant complications, presenting their complaints to health care providers effectively, and taking an active role as patients. Even highly educated migrant women showed low health literacy skills in their interaction with doctors. Patients' perspectives are valuable as one of the tools to evaluate the quality of maternity care. Communication by maternal health professionals can be improved through more sensitivity to social factors that affect immigrant women's health problems. Women with limited health literacy should be empowered through education about danger signs in pregnancy and information about preferences and policies in obstetrics in the Netherlands. They should also be invited to participate in medical decision-making.

Résumé

Cette étude de 2006 s'est penchée sur les facteurs liés à l'ethnicité qui contribuent à des soins obstétricaux inférieurs à la norme et les conséquences sur la morbidité maternelle sévère chez les immigrantes aux Pays-Bas. Des entretiens approfondis ont été menés avec 40 immigrantes et 10 Néerlandaises de souche. Les immigrantes ont indiqué que, souvent, les prestataires de soins de santé n'accordaient pas suffisamment d'attention à leurs problèmes liés à la grossesse, en particulier dans les cas de pré-éclampsie. Elles ont aussi parlé de retards pour recevoir l'information sur le diagnostic et le traitement. Les obstétriciens qui ont examiné 20 des cas ont jugé que les soins inférieurs avaient joué un rôle dans l'apparition de complications dans 16 d'entre eux. Les femmes elles-mêmes peinaient à identifier des complications médicalement significatives, à présenter efficacement leurs doléances aux prestataires de santé et à jouer un rôle actif comme patientes. Même des migrantes très instruites se montraient peu informées en matière de santé dans leurs interactions avec les médecins. La perspective des patientes est précieuse comme l'un des outils d'évaluation de la qualité des soins obstétricaux. Les professionnels de la santé maternelle peuvent améliorer leur communication en étant plus sensibles aux facteurs sociaux qui touchent les problèmes de santé des immigrantes. Les femmes peu au fait des questions de santé devraient apprendre à reconnaître les signes de danger pendant la grossesse et être informées des préférences et des politiques obstétricales aux Pays-Bas. Elles devraient aussi être invitées à participer aux décisions médicales.

Resumen

En este estudio realizado en 2006 se investigaron los factores relacionados con etnia que contribuyen a la atención obstétrica de calidad inferior y a los efectos en morbilidad materna grave entre mujeres inmigrantes en Holanda. Se realizaron entrevistas a profundidad con 40 mujeres inmigrantes y 10 holandesas. Las inmigrantes informaron que los profesionales de la salud a menudo prestaban insuficiente atención a sus quejas referentes al embarazo, especialmente en casos de preeclampsia. Además, hablaron de demoras en recibir información sobre el diagnóstico y el tratamiento. Los obstetras que revisaron 20 de estos casos determinaron que la atención de calidad inferior contribuyó al desarrollo de complicaciones en 16 de ellos. Las mujeres tuvieron problemas identificando complicaciones de significancia médica, presentando sus quejas eficazmente al personal de salud y asumiendo un rol activo como pacientes. Incluso las inmigrantes muy cultas demostraron un bajo nivel de comprensión sobre la salud en su interacción con los médicos. Las perspectivas de las pacientes son valiosas como una de las herramientas para evaluar la calidad de la atención obstétrica. La comunicación por parte de los profesionales obstétricos puede mejorarse mediante una mayor sensibilidad a los factores sociales que afectan los problemas de salud de las inmigrantes. Las mujeres con limitada comprensión de la salud deben ser empoderadas con educación sobre los signos de alarma en el embarazo y con información sobre las preferencias y políticas de obstetricia en Holanda. Además, se recomienda invitarlas a participar en la toma de decisiones médicas.

Key Words:

• maternal morbidity
• ethnicity
• provider–patient relations
• quality of care
• health literacy
• Netherlands

Immigrant women in the Netherlands experience more serious complications during pregnancy and childbirth than native Dutch women, according to LEMMoN, a countrywide registration study (2004–2006).¹ We conducted a qualitative study of patients' perspectives to gain more insight into ethnicity-related factors contributing to sub-standard maternity care and to explore the possible relationship between sub-standard care and severe maternal morbidity among immigrant women. Sub-standard care is defined as care that falls below standards that clinicians consider should have been achieved given the circumstances.¹

Methodology

Forty immigrant and ten native Dutch women with severe maternal morbidity were interviewed. They were among 2,552 women (incidence 0.71 per 1000 deliveries) in the LEMMoN study.¹ ‘Country of birth’ was used as indicator to identify women as immigrant women: a woman is of immigrant descent if she was born abroad and at least one of her parents was born abroad (first generation), or if she was born in the Netherlands with at least one of her parents born abroad (second generation).²

Obstetricians from 18 hospitals in the western part of the country, a representative sample of all Dutch hospitals in terms of volume, level of care, and number of complications registered in the LEMMoN study, were asked to approach all immigrant women who met the LEMMoN criteria to invite them to participate. Between February and August 2006, 42 women were referred to the main researcher for participation. Two immigrant women could not be reached due to incorrect contact details. Three hospitals identified 15 native Dutch women from their LEMMoN study sample within a period of ten weeks. Native Dutch was defined as a person whose parents were both born in the Netherlands. The main researcher selected ten of them based on different levels of education.

All women gave written informed consent. Forty-six of the interviews were conducted in the homes of the interviewees between two and six weeks after discharge from hospital, and four in hospital because of their relatively long hospitalisation. All 50 women were eager to tell their stories. Interviews were recorded and transcribed.

The women were asked about their perspectives on the development of their medical complication and on the received health care, with particular attention to self-diagnosis, health care-seeking behaviour, presentation of complaints, recognition of complaints by health care providers, and communication with them. Husbands, partners, relatives and friends who had been involved during the complication participated in almost all interviews and added their perspectives on what had happened. Sometimes they also acted as interpreters, as they had done in previous contacts with the health care providers. These lay interpreters all had a good command of Dutch. Eight women with a substantial language barrier were asked if they wanted a professional interpreter. Only one woman said yes. During the interview, however, a female friend insisted on translating, as she had done in the hospital, thereby silencing the professional interpreter.

Twenty of the 40 interviews with immigrant women were selected for additional analysis, carried out by four obstetricians, who were selected based on their extensive experience of maternal morbidity audits and availability. The selection criterion for this analysis was mention by the patient of delay in receiving care as a significant factor in the deterioration of her health. The interviews were randomly divided among the four obstetricians. They were asked to assess the patients' perspectives in comparison to the women's physicians' perspectives, as obtained from the medical files. All hospitals and persons involved were anonymised. The information collected from the native Dutch women served as a source of verification that the findings were related to ethnicity.

The thematic analysis of the interviews focused on the question of where things went wrong, what health care providers did wrong, and why and how things went wrong from the women's perspectives. Following the method of grounded theory,³ sensitising concepts – such as the nature of the doctor–patient relationship, sources of social support, the language spoken, understanding of health and medical information – were identified. The use of the software Kwalitan supported the analysis.

The study design was approved by the Medical Ethical Committee of Leiden University Medical Center. Confidentiality has been kept by not using the real names of the interviewees.

Findings

Characteristics of the women

The 40 immigrant women fell into two groups, according to their migration history. The first group consisted of 22 women from large migrant groups, and included nine Moroccan, seven Turkish, four Surinamese, and two Dutch Caribbean women. Their first-generation male family members migrated to the Netherlands between 1965 and 1974, mainly as “guest labourers” and migrants from former Dutch colonies. Through chain migration and family reunion, their ethnic groups have grown into multigenerational communities with social networks embedded in the social structures of Dutch society. Women in these communities have access to collectively shared experiences of maternal health care in the Netherlands. Except for the Surinamese women, the women in this group were from a low socioeconomic position, the position of most large migrant groups in Dutch society. All the Turkish women in this study except one had a socially dependent position; they spoke only Turkish and depended on Dutch-speaking relatives when visiting health facilities. The Surinamese women, on the other hand, had a relatively strong social status because of their better education, job skills, and economic independence.

Those in the second group were mostly first generation migrants and had lived in the Netherlands for a much shorter time. This group consisted of 18 women from smaller migrant groups from Eastern Europe and Middle East, Asian and sub-Saharan African countries. Apart from those who were political refugees and expatriates, the migration motives in the second group were similar to the larger migrant groups. However, nine of the 18 had lived in the Netherlands for less than five years and therefore had a much more limited experience of Dutch society and Dutch maternal health care. However, this second group had a much higher level of education; only four of them had had only a primary school education. Ten of them spoke good English.

Twenty-one of the 50 women in our study had been admitted to intensive care because of a life-threatening obstetric complication, of whom some had more than one complication. The course and consequences of these complications could thus have had a far-reaching and traumatic impact on the women. In some cases, their babies had died or survived with an uncertain prognosis as well. The stories of Yasemin and Ayumi illustrate some of the experiences that the women went through during and after a complication. Yasemin, a Turkish woman, belongs to the first migrant group, and Ayumi, a Japanese woman, to the second. Their stories represent the most important differences between the two groups in terms of social background and its influence on the ways in which they dealt with severe complications.

• Yasemin's story

Yasemin (age 35) came to the Netherlands 11 years ago to marry her Turkish husband. She had completed primary school only and does not speak Dutch. Her husband translates. During her pregnancy Yasemin felt tired and dizzy, and suffered from headaches and hyperventilation. The midwife referred her to the obstetrician three times because of hypertension. The obstetrician did not find anything abnormal and told her not to worry. Her General Practitioner attributed her complaints to increased stress resulting from quarrels with her mother-in-law, who lived next door. At 40 weeks of pregnancy Yasemin gave birth in hospital, assisted by her midwife, and returned home two hours later. At home she developed terrible headaches, pain throughout her whole body, and dizziness. The midwife again referred her to the hospital because of high blood pressure. Yasemin had an eclamptic seizure while waiting for a doctor to examine her. Immediate action was then taken. She was given intravenous medication to prevent further convulsions and to lower her blood pressure. She recovered quickly, and after a few days she was discharged.

• Ayumi's story

Ayumi (age 33) was a university-educated Japanese woman and spoke reasonable English. Ayumi was 26 weeks pregnant with twins when she came to the Netherlands as an expatriate together with her husband. She complained to her obstetrician several times about headaches and considerable weight gain. Her blood pressure, which was usually low, rose and she suffered from fluid retention. She was worried and asked the doctor for clarification about the diagnosis and whether a caesarean section should be performed. The obstetrician told her that everything was normal and planned a vaginal delivery. When Ayumi presented with similar complaints at 38 weeks, she was admitted to hospital with pre-eclampsia. Labour was induced and when it didn't progress, caesarean section was performed. Healthy twins were born. One hour later Ayumi suddenly started to bleed heavily from her womb because of uterine atonia. She was immediately operated upon again but kept bleeding heavily. Hysterectomy was performed in order to save her life. Ayumi received 35 units of blood and remained in the intensive care unit for three weeks. Finally, after 30 days, she was able to leave the hospital. She came home with many questions about what had happened and depressed.

Immigrant women's experiences

Health care providers underestimated complaints

Both Yasemin and Ayumi believed that their midwives and obstetricians did not pay proper attention to the complaints they expressed during pregnancy.

“I worried that I had gained too much weight. I worried that I was swelling. I worried that I had headaches. But they said:‘No problem.’ Why they waited so long? Why they tried to do natural delivery for two babies?… The uterus becomes bigger… so big that it could not shrink.” (Ayumi)

“Each time she came for a check-up, she expressed that she was constantly tired and dizzy and that her head was pounding. It bothers her that they did not pay any attention to her complaints.” (Yasemin, husband translates)

A number of the other immigrant women, from both groups, also thought that their complaints had not been taken seriously by health care providers, and that this might have been the reason why complications developed or were aggravated. Soriya (an Afghan refugee) and Ayse (a Turkish marriage migrant) reported that during induced labour extreme and abnormal pain was interpreted as normal birth pains. Both women later turned out to have sustained uterine rupture. Hanan, a Moroccan second generation immigrant, phoned the midwife several times at 36 weeks of gestation because she felt contractions. The midwife interpreted Hanan's complaints as Braxton Hicks contractions (infrequent, irregular, painless but uncomfortable uterine contractions that can occur before labour), and only came to her home after the baby had been born without skilled attendance. According to Hanan, the midwife panicked because of the baby's unexpected birth and pulled the placenta out with too much force; she considered this to be the cause of her having a major obstetric haemorrhage. Sun-Hi, a marriage migrant from South Korea, had to wait five hours in hospital after a normal vaginal birth before being operated on for a large vulvar haematoma, even though she had frequently complained about steadily increasing pain. She eventually developed major obstetric haemorrhage as well.

Lack of patient information

We asked what doctors could do better next time. Many of the immigrant women said they had received little information from health care providers about diagnosis, treatment options, and risks before, during or after their complication occurred.

“If someone had told me what was going on, I would have felt okay… The most important thing about that obstetrician was that he had not told us everything in detail.” (Sun-Hi)

“I think it is better that they give clearer information.” (Ayumi)

The more educated immigrant women had collected general information about pregnancy and birth via social contacts (family members, colleagues at work), books and journals, but especially from the Internet. Ayumi, for instance, was well informed about the risks of twin pregnancies and interacted with health care providers equipped with that knowledge. She wanted to have a say in the medical decisions, and if she had understood the doctor's reasoning, she could have participated in choosing the interventions.

“It feels like: that is my birth… But it is like: what does the doctor want to do? It is not like what I want to do. The doctor wants to do natural delivery… I much prefer natural delivery. But if they had said: ‘Your chest is so big, your blood pressure high, we can go to c-section now. But if you want to try natural delivery…?’” (Ayumi)

Yasemin and her husband, in contrast, did not have any knowledge about the early symptoms of pre-eclampsia. They also had difficulties understanding the medical information about her complication. Even so, they valued provision of patient information. They felt it meant they were taken seriously and respected as persons. However, they underestimated its importance as support for self-determination in decision-making.

Daily stress factors among the women

Chronic stress factors seem to have had a negative influence on the early identification of symptoms by the women themselves and their health care providers. The socioeconomically deprived women in particular talked about extensive problems with income, housing, residence permits, childrearing, war trauma, homesickness, chronic ill-health and relationship problems with the fathers of their children and family members. Pregnancies were often unplanned and unwanted. The women were often not fully aware of physical changes and signs of pregnancy-related morbidity and blamed it on non-pregnancy-related stress.

“[Yasemin] thinks all Turkish women who deliver have problems with their mothers-in-law. That is perhaps what caused the problems during pregnancy.” (Husband translates)

Family members and health care providers too attributed women's complaints to stress stemming from various preoccupations. Livine, a young woman from the Netherlands Antilles, arranged a secret adoption for her unwanted baby with the help of her midwife, because of her psychosocial and economic problems. Neither she nor the midwife paid much attention to the nausea, vomiting, dizziness, sleepiness and exhaustion she suffered in the last week of her pregnancy because they attributed them to the stress of hiding the pregnancy. Immediately after delivery, Livine was diagnosed with acute fatty liver of pregnancy.

Lack of knowledge of danger signs during pregnancy

Few of the women had any knowledge of the danger signs during pregnancy. That is why the immigrant women often did not interpret their complaints as early signs of complications, especially in cases of pre-eclampsia.

Many had read general information about health issues relating to uncomplicated pregnancy. They had received brochures from the midwife or had performed their own searches in magazines and on Internet sites. They said the information they had read treated pregnancy and childbirth as positive events, accompanied by minor physical discomfort. A few of the well-educated women with high-risk pregnancies, like Ayumi – including those with twin pregnancies, placenta praevia, and essential hypertension – had obtained information from professional websites.

Female relatives and friends with children were seldom consulted about health problems during pregnancy. The women preferred to ignore disaster scenarios affecting pregnancy and birth, as was the case with Remzu, who had fled from Chechnya four years ago and who suffered from a delayed diagnosis of pre-eclampsia.

“I did read something about complications, but I thought, that is not for me. I always thought positive.” (Remzu)

Language barriers

Language barriers with health care providers are perhaps the most obvious cause of exclusion from good quality care and from positive health outcomes. In particular, women with the lowest levels of education had difficulties searching for effective care, clear presentation of complaints and understanding information given about diagnosis and treatment. Alma, who had fled seven years ago from civil war in Bosnia, failed to explain to her midwife that she had had an epileptic seizure during a bombardment of her native village. She had had little education and did not know the word for epilepsy. During delivery Alma experienced another seizure.

Husbands and others did not always translate complaints and information well. Deniz, a Turkish marriage migrant, who did not speak Dutch, had an extremely quick progression of labour after induction with very frequent, strong contractions. According to Deniz, the obstetrician did not take her complaints, translated by her sister-in-law, seriously. He told them that the onset of labour would take some time. Subsequently, Deniz refused instructions to turn onto her side following decreasing fetal heart rates because she felt an irresistable urge to push. Her sister-in-law started to panic and failed to translate why she refused. When the nurse pushed Deniz onto her side, the baby was born. Deniz sustained cervical rupture resulting in major obstetric haemorrhage.

Some of the women who spoke little or no Dutch did not experience this as a problem. Remzu said that she had always succeeded in outlining her complaints to caregivers in broken Dutch. But she also admitted that there was a language barrier.

“I understand almost 80%. Not all the words, of course, but what is necessary, what is important.” (Remzu)

Some women, such as Yasemin, whose husband had to translate everything, seemed aware of the negative consequences of language barriers for communication. When asked whether her complaints were perhaps not taken seriously because she could not present them herself, she said:

“How well can you translate me, what I am feeling? Perhaps I would have said it better.” (Husband translates)

In no case was a professional interpreter arranged during antenatal visits or hospital admission.

Communication skills and patient attitudes

The immigrant women were often very modest about asking for medical attention or information about diagnosis and treatment. When Remzu was 29 weeks pregnant she developed serious headaches and oedema. Despite the fact that she became sicker and sicker, she waited a week before consulting her general practitioner because it was Christmas. She was then admitted to hospital immediately with severe pre-eclampsia.

“I thought: it is a holiday, I do not want to be a problem for someone. I will try to go Monday or Tuesday after the holidays. But I think now: why did I wait? Why didn't I phone immediately?” (Remzu)

Julita was a well-educated, English-speaking woman from the Philippines, who had come to the Netherlands as a marriage migrant seven years before. She waited obediently for ten hours for an operation to relieve a vulvar haematoma, though it was increasing in size. Julita was told several times by the nurses that the operating theatre was still busy. She did not ask a nurse or doctor for clarification because she accepted waiting as “normal”. After the operation she needed six units of blood because of major obstetric haemorrhage.

Ayumi had access to the basic resources needed for open and equal communication with her obstetrician. She was informed about the risks of twin pregnancies and could communicate in English. However, she did not understand his reasoning. Neither did she succeed in putting her concerns onto the obstetrician's agenda.

“I should have challenged the doctor… I was maybe too obedient, what they told me… But I even showed my record [of very low blood pressure]. I thought I did everything I could. But they still said ‘no problem’.”

The well-educated women from the smaller migrant groups saw themselves as critical consumers of care in the Netherlands because of their knowledge of medicalised maternity care at home. They expected intervention as soon as danger signs of maternal morbidity developed. They were surprised by the conservative birthing policies of Dutch obstetric professionals, but did not succeed in convincing them of their own doubts about intervention decisions. Julita, for example, asked the doctor for a caesarean section because she had scoliosis:

“They did not really explain to me why I could not have a c-section. What they said was I could have a vaginal delivery… So, of course, they know better. So all I needed to do was just trust them.” (Julita)

The modest responses of many of the immigrant women were more to do with trust based on the expertise ascribed to doctors in general, and less with a relationship of trust based on their understanding of the doctor's reasoning in their own case. The women may, however, have been conscious of their minority status in Dutch society and afraid of discrimination, despite the fact that only a few of them spoke about this explicitly.

“They [doctors] really hated us. I have to challenge the doctor. But I was new here. I try to respect the way things are here.” (Ayumi)

Native Dutch women's experiences

The native Dutch women interviewed had a more engaged, pro-active attitude in interactions with health care providers than immigrant women. They were also more satisfied with the professional attention they received for their complaints and the information they were given before, during and after their complication. Both resulted in shared decision-making about diagnosis and treatment options.

“The doctor himself didn't give much information. If I asked for it, he explained very well… I had simply bought a book about twins. In fact this gave me a lot of information which I had confirmed by the doctor now and again. When I finally had to stay [in hospital] because of my blood pressure, he [the doctor] asked: ‘How do you look at it – wait or not?’ I wanted to make it to week 38. So I said: ‘I want to wait another week.’ The doctor replied: ‘Okay, come back next week, and then we'll talk about it again’. And when I eventually developed pre-eclampsia the doctor came to me again and said: ‘Now I am in favour of getting the babies out.' I replied: ‘Yes, I agree.’” (Karin, intermediate vocationally educated)

Native Dutch women understood better how, where and why medical problems had developed. Within their reconstructions of the development and course of their complication they could reproduce medical information coherently, including those with lower levels of education. However, despite their satisfaction with the obstetric care they received, some of them also referred to delays, which in their opinion had contributed to their complications. Delays were related to the late start of appropriate medical treatment and insufficient cooperation between obstetric professionals, such as late referral to a higher level of care.

Quite some persuasion by them too could be needed to convince a physician of the seriousness of health problems. Laura was referred to an emergency unit by her GP six days after delivery with a painful swollen leg and persistent heavy vaginal blood loss. Her husband had to convince the physician on duty of the seriousness of these problems. Three operations were needed to stop the bleeding and remove the remaining placenta. Miep, on the other hand, whose two previous deliveries were problematic due to slow progress and placenta accreta, experienced the same problems a third time, plus major post-partum haemorrhage. She felt that her anxiety was well understood, that she had been invited to play an active role in decision-making in the emergency unit and was cared for very well.

Evaluation by four obstetricians

Sub-standard care was categorised as inadequate antenatal visits and/or delays in diagnosis, management of treatment, referral to a higher level and organisation of health care. The four obstetricians judged sub-standard care to have played a role in the development of complications in 16 of the 20 cases they reviewed, which occurred despite women's complaints. In five of the 16 cases, more than one type of delay was identified. In eight cases, delays in diagnosing pre-eclampsia were considered to be present, e.g. with Yasemin. Other delays occurred in diagnosing uterine rupture (2), preterm labour (1) and ketoacidotic diabetic coma (1). Appropriate management was started too late in cases of major obstetric haemorrhage (3) and uterine rupture (2). In three cases (two with pre-eclampsia) midwives had referred the woman to the obstetrician too late. In one case the obstetrician should have referred a woman who had uncontrollable haemorrhage sooner to a university hospital. In one case, the obstetrician judged that antenatal visits were not performed adequately by the midwife, resulting in pre-eclampsia remaining unnoticed. In two cases there were transport delays, which aggravated complications.

The obstetricians found that the patients' behaviour had directly contributed to sub-standard care in five cases. In two cases language barriers may have resulted in inadequate communication, which in turn contributed to delays in diagnosing pre-eclampsia. The case of Yasemin was one of the two. Reading Yasemin's own story one may get the impression that the midwife acted adequately. She referred her three times to the obstetrician because of high blood pressure. However, according to the evaluating obstetrician, there was insufficient medical attention first by the midwife and later by the obstetrician for Yasemin's high blood pressure after birth, though she did complain of it. This resulted in delayed diagnosis, delayed start of treatment and eventually the eclamptic fit. Yasemin and her husband concurred. In another case, the obstetrician judged that a woman had missed an extra blood pressure check at a diagnostically crucial moment and another woman waited too long before consulting a General Practitioner or midwife when she had significant complaints. The obstetricians also judged deficits in communication with six patients regarding diagnosis and treatment, although they did not think this contributed to sub-standard care.

In only four cases was no sub-standard care observed. Medical check-ups during pregnancy were done properly, in accordance with the guidelines. There were no delays in diagnosis, management and referral and complications could not have been prevented by more appropriate attention. These cases involved major obstetric haemorrhage after placenta abruptio and after uterine atonia. One woman had problems with hypotension and angina pectoris during delivery and was observed in the intensive care unit. Ayumi was among this group; her haemorrhage was due to uterine atonia and had no relation to hypertension or other delays, though she had voiced the opinion that rising blood pressure had increased the risk of eclampsia and why she thought one twin was born prematurely. Miscommunication between Ayumi and her obstetrician seemed clear, though not due to language barriers.

Discussion

Patient-centred care focuses on cooperation and communication between patient and provider in a process of shared decision-making. This assumes that patients have the knowledge and confidence to do so and that such a partnership reduces the risk of complications. In contrast to the pro-active attitude of the native Dutch women, the reticent mindset of many of the immigrant women hindered effective involvement in getting needed care and in decisions about that care. Often substandard care as identified by the obstetricians was related to poor doctor–patient interaction.

Previous research has shown that immigrant patients in the Netherlands talk less and show more agreement with native Dutch GPs than native Dutch women. Moreover, they give less information about lifestyle and psychosocial issues. The active, task-oriented behaviour of physicians was similar with both groups of women, but the physicians were more empathic and more involved with the native Dutch patients.⁴

Immigrant patients' cultural and gendered expectations vis-à-vis formal authoritative relationships in their home countries, as well as fear of exclusion from care due to discrimination, merit further investigation and understanding.

Health literacy

With the move towards a more “consumer-centric” health care system, health literacy becomes even more of an issue. Health literacy is “the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions”. It comprises a complex set of reading, listening, analytical, and decision-making skills, and the ability to apply these skills to health problems.⁵ In our study even highly educated women from the smaller migrant groups, who presented themselves as critical health consumers, showed low health literacy skills. They did understand basic health information, but lacked the appropriate communication strategies to interact adequately with caregivers, and vice versa as regards those same caregivers. This is one of the most important findings of our study. Research is still dominated by a paradigm in which ethnic health inequities as a result of low health literacy are correlated with factors such as general illiteracy, low levels of education, low income, and a socially disadvantaged minority position in society among patients, which was not what we found.

Our study supports the LEMMoN study finding that immigrant women in small migrant groups are most at risk for severe maternal morbidity, though we have shown that all immigrant women are more at risk. One of the explanations for the differing level of risk between migrants could be that social support from an extensive ethnic network, available among the large migrant groups, helps to prevent delays that lead to maternal complications worsening.⁶

Competent maternity care for immigrant women

There is a need for maternal health care providers to be sensitive to the needs and sociocultural contexts of immigrant women, which may differ across different migrant groups. Cultural awareness among medical professionals is increasingly being promoted to improve communication in health care. However, it would be a mistake if providers started approaching pregnant women only from a culturally deterministic angle, which would underestimate differences in educational levels and social status, differences between the generations, and different individual patients' experiences in the Netherlands.

From our study it is clear that maternity care for immigrant women can be improved through early diagnosis of maternal morbidity, in particular pre-eclampsia. Focused questioning of complaints, even if they are vague, and attention to individual circumstances are needed. Where language barriers exist, professional interpreters should be brought in.

Not all patients wish to be involved in decision-making about their care, however. Much depends on the nature of the care – elective, acute or chronic – as well as whether or not patients have the motivation and ability to be involved and informed. Women participating in our study, who were in a life-threatening situation, needed competent diagnosis and decisive action by obstetric professionals far more than patient-centred participation in the consumer sense.

Improved education for immigrant women about signs and symptoms of complications during pregnancy, especially pre-eclampsia, would go far towards overcoming ethnic differences in maternal morbidity. But this education needs to take account of professional preferences in the delivery of maternity care in the Netherlands, which includes high levels of home birth and shared decision-making within a government-supported consumer care mode. A controversial issue in the Netherlands is how a policy for reducing maternal morbidity can succeed within an approach to pregnancy and childbirth as natural events that can take place at home. The caesarean delivery rate of 15% is among the lowest in the western world, and the home delivery rate of approximately 25% is exceptionally high for a Western country. Reducing the high incidence of pre-eclampsia within such a model creates a particular challenge for Dutch obstetrics.

Amsterdam, Netherlands, 2007

Inclusion of patients' perspectives in evaluating standard of care

Good medical care is an issue both of high quality medical practice and good provider–patient communication. Our study shows that the patient's evaluation of what went right and wrong in cases of serious maternal morbidity may or may not be the same as the providers'. For this reason, we recommend that patient perspectives are an essential element of and should be integrated into the medical evaluation of the quality of maternity care.

Acknowledgements

The authors thank all patients and obstetricians involved in this study for their cooperation, especially JIP de Vries, JJ Duvekot and J Beltman, and J van den Hoven for the pilot study.