SEARCH

SEARCH BY CITATION

Keywords:

  • Immigration;
  • maternal near-miss;
  • severe maternal morbidity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information

Objective

To determine if immigrant women from low-, middle- and high-income countries have an increased risk of severe maternal morbidity (near-miss) when they deliver in Sweden.

Design

Population register-based study.

Setting

Nationwide study including all singleton deliveries (≥28 weeks of gestation) between 1998 and 2007.

Population

Women with a near-miss event; all women with a singleton delivery ≥28 weeks of gestation during the same period acted as reference group.

Methods

Near-miss was defined by a combined clinical and management approach with use of International Classification of Diseases, 10th revision codes for severe maternal morbidity. A woman's country of origin was designated as low-, middle- or high-income according to the World Bank Classification of 2009. Unconditional logistic regression models were used in the analysis.

Main outcome measures

Maternal near-miss frequencies per 1000 deliveries and odds ratios with 95% confidence intervals.

Results

There were 914 474 deliveries during the study period and 2655 near-misses (2.9 per 1000 deliveries). In comparison to Swedish-born women, those from low-income countries had an increased risk of near-miss (odds ratio 2.3, 95% confidence interval 1.9–2.8) that was significant in all morbidity groups except for cardiovascular diseases and sepsis. Women from middle- and high-income countries showed no increased risk of near-miss.

Conclusions

Women from low-income countries have an increased risk of maternal near-miss morbidity compared with women born in Sweden. Although the rate is low it should alert healthcare providers.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information

In high-income countries maternal near-miss or severe acute maternal morbidity,[1] as well as maternal mortality,[2, 3] have been shown to be more common among women whose ethnic background differs from that of the native population. However, an association of immigration status with maternal outcome is not easily interpreted. Most countries today are multi-ethnic societies, although the definition of ethnicity varies greatly. Studies on maternal outcome have shown considerable differences between ethnic minority groups, ranging from a risk of severe acute maternal morbidity for Moroccan and Turkish women in the Netherlands that is on a par with the general population[4] to a six-fold increased risk of maternal mortality for “black women” in the UK.[5] In previous studies ethnicity was commonly based on either a woman's self-definition of ethnic affiliation,[6] or the country where she or her parents were born.[7] The latter is often further subdivided into local geographical regions.

Maternal near-miss is defined as ‘a woman who almost died but survived a complication that occurred during pregnancy, delivery or within 42 days of termination of pregnancy’.[8] A much larger number of women suffer from near-miss morbidity than maternal mortality, which gives increasing power of studies to investigate risk factors for the occurrence of disease and progression to death.[8]

In present-day Sweden 19% of all women of reproductive age have been born in another country.[9] The aim of our investigation was to assess whether an association existed between an immigrant woman's country of birth, stated as low-, middle- or high income, and a maternal near-miss experienced in Sweden. We used an immigrant women's country of birth and the World Bank division of countries (see Supporting Information, Table S1) into levels as we regarded this to be a more homogeneous classification than geographical regions.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information

Case identification

Maternal near-misses may be identified through clinical criteria, such as severe haemorrhage; management criteria, such as specific surgical procedures; or organ system dysfunction, such as vital organ failure. When possible, the World Health Organization recommends use of the latter, specifying objective parameters for defining near-miss events such as acute cyanosis, gasping or clotting failure.[8] However, such an approach demands prospective collection of information. Due to the nature of our data we used a combination of clinical and management criteria in our study.

Study population

By means of the Swedish Medical Birth Register (MBR), we were able to identify all singleton deliveries between 1998 and 2007 with a gestational age ≥28 weeks. Using the unique personal identification number assigned to each Swedish citizen at birth or upon immigration, we linked data from the MBR with the Swedish Hospital Discharge Register and the Swedish Education Register. The MBR includes information on approximately 98% of all births in Sweden and gives information on the mother, the course of pregnancy and delivery. Data are prospectively collected, beginning with a woman's first prenatal visit. The Swedish Hospital Discharge Register contains information on hospital admissions and discharges, diagnoses according to the International Classification of Diseases, 10th revision (ICD-10),[10] and coded surgical procedures according to the Nordic Medical Statistics Committee. The clinical diagnoses defined as maternal near-miss morbidities in our study were acute renal failure, liver failure, respiratory failure, shock, cerebrovascular failure, pulmonary embolism, eclampsia, severe pre-eclampsia (haemolysis, elevated liver enzymes, low platelet count [HELLP] and disseminated intravascular coagulation [DIC]), uterine rupture, sepsis, cardiac events and ruptured aneurysm of the aorta. The management criteria used were procedure codes for obstetric hysterectomy and obstetric laparotomy. The codes were grouped according to major organ systems into organ failure (acute renal, liver and respiratory failure); shock (cardiogenic, hypovolaemic, septic, or unspecified shock, as well as hysterectomy indicating severe bleeding); cerebrovascular events; and cardiovascular events (aortic dissection with rupture, and cardiac events, including heart failure, acute myocardial infarction and cardiac arrest). Pre-eclampsia (HELLP and DIC) and eclampsia were classified as one group, as was uterine rupture and obstetric laparotomy. The final diagnostic group was sepsis caused by any agent. The ICD-10 codes, procedure codes and near-miss categories are listed in the Supporting Information, Appendix S1. Cases of maternal mortality were identified through linkage with the Swedish Cause of Death Register.

Register data

Country of birth for all women was extracted from the Swedish Population Register. Women with an unknown country of origin were excluded (n = 480). In the case of women born outside Sweden, country of origin was categorised as low-, middle- or high-income, using the World Bank Country Classification of 1 July 2009 (see Supporting Information, Table S1).[11] A woman's characteristics were extracted from the MBR and the variables were classified into groups before being analysed. Maternal age at delivery was categorised into four groups: <25, 25–34, 35–44 and ≥45 years. Parity was classified as one, two to three and four or more deliveries. Body mass index (BMI) was calculated (weight [kg]/height [m2]) and women were grouped as lean (BMI <20), normal (BMI 20 to <25), overweight (BMI 25 to <30) or obese (BMI ≥30). Cigarette smoking during early pregnancy was divided into non-smoker or smoker. As immigrants in Sweden commonly live in large cities and coding traditions may differ between university hospitals and smaller hospitals, we grouped hospitals as district, county or university. Education is a well-known mediator for pregnancy outcome, so we used the Swedish Education Register to obtain information on each woman's highest level of education attained as of 31 December 2007.[12] Educational level was divided as follows: <9 years, 9–12 years (secondary education) and >12 years (tertiary education started). Civil and cohabiting/non-cohabiting status was extracted from the MBR.

Statistical analysis

The frequency of maternal near-miss was calculated per 1000 deliveries in total, and for each morbidity group. Logistic regression models were used for possible confounders. Age, parity, BMI, smoking and type of hospital in Model 1, with the addition of education and cohabiting status in Model 2. Women who had more than one event or condition were included as near-miss in each morbidity group. Because women with a previous near-miss event might be exposed to higher risk in a succeeding pregnancy, we calculated odds ratios (OR) and their 95% confidence intervals (95% CI) using clustered data by means of the generalised estimation equation method. Statistical Analysis System (SAS®, Cary NC, USA) software version 9.2 was used for statistical analysis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information

There were 914 474 deliveries during the study period, of which 2677 deliveries involved one or more severe maternal morbidity events as defined by our inclusion criteria. Twenty-two women in the latter group died between the date of delivery and 42 days postpartum, and were excluded from the analysis. The result was 2655 maternal near-misses (frequency 2.90 per 1000 deliveries). There were 189 women with more than one near-miss diagnosis, 176 women with two diagnoses, nine women with three, two women with four and two women with five, for a total 2863 near-miss events. The general characteristics for all women with a near-miss event are shown in Table 1. Multiparity (>4) was more common among women from low-income countries than in Swedish women (22.9% versus 4.7%), as was low education (37.6% versus 9.3%), defined as <9 years of schooling. A greater proportion of women from low-income countries were non-cohabiting. The various groups showed similar distribution with regard to maternal age and BMI. Women from low-income countries had a 2.5-fold increased risk (95% CI 2.1–2.9) of developing a maternal near-miss condition, compared with Swedish women (Table 2). The increased risk remained significant when adjusting for age, parity, BMI, smoking and type of hospital in Model 1 (OR 2.5, 95% CI 2.1–3.0), and when adding education and cohabiting status in Model 2 (OR 2.3 95% CI 1.9–2.8). Six out of ten women originating from low-income countries came from Somalia (n = 66, near-miss frequency 9.1 per 1000 deliveries), Ethiopia (n = 21, near-miss frequency 6.6 per 1000 deliveries) and Eritrea (n = 12, near-miss frequency 7.2 per 1000 deliveries). Women from middle- and high-income countries had no increased risk of near-miss morbidity.

Table 1. Description of background characteristics of maternal near-misses and frequency per 1000 deliveries at ≥28 weeks of gestation by economic level of country of origin, Sweden 1998–2007
 Low-income countriesMiddle-income countriesHigh-income countriesSweden
n % n % n % n %
Maternal age in years
<25414217.823 39421.127787.985 12011.4
25–3414 28261.465 01958.721 74461.7502 45867.4
35–44473920.422 14020.010 62530.1156 51321.0
≥451150.52560.21100.310390.1
Parity
1690929.745 49841.114 37940.8335 43445.0
2–311 03947.455 03749.718 04651.2374 55150.3
≥4533022.910 2749.328328.035 1454.7
Body Mass Index (kg/m2)
<20276311.910 7469.734919.962 7908.4
20 to <25847836.448 43043.716 36246.4359 15948.2
25 to <30521622.425 43022.9674319.1157 34721.1
≥30255711.098939.029818.469 3409.3
Missing426418.316 31014.7568016.196 49412.9
Smoking
Non-smoker21 01390.393 92384.828 42580.6624 07383.8
Smoker5082.299409.0382010.870 6069.5
Missing17577.569466.330128.550 4516.8
Type of hospital
University13 79859.353 19048.013 56338.5228 52630.7
County715430.743 62939.416 52246.9378 57050.8
District22039.513 26112.0478913.6132 05017.7
Missing1230.57290.73731.159840.8
Level of education
≤9 years874537.631 43228.4370110.569 0969.3
9–12 years867937.341 56037.511 72033.2359 36648.2
>12 years403217.334 10330.818 06351.2314 50542.2
Missing18227.837143.417735.021630.3
Cohabiting status
Cohabiting17 57375.597 43587.930 87087.6668 12989.7
Non-cohabiting409517.669886.316504.732 6154.4
Missing16106.963865.827377.844 3866.0
Total23 278100.0110 809100.035 257100.0745 130100.0
Table 2. Crude and adjusted odds ratios with 95% confidence intervals for near- miss among immigrants in Sweden, 1998–2007
Country groupTotalMaternal near-missFrequency per 1000 deliveriesCrudeModel 1aModel 2b
OR95% CIOR95% CIOR95% CI
  1. a

    Adjusted for maternal age, parity, BMI, smoking, and type of hospital.

  2. b

    Adjusted for maternal age, parity, BMI, smoking, and type of hospital, education and cohabiting status.

Low-income countries23 2781626.962.52.1–2.92.52.1–3.02.31.9–2.8
Middle-income countries110 8093132.821.00.9–1.11.00.9–1.21.00.9–1.2
High-income countries35 2571073.031.10.9–1.31.00.8–1.21.00.9–1.2
Sweden745 13020732.781.0 1.0   

Table 3 presents the diagnostic categorisation of maternal near-miss events stratified by country groups. When a woman has more than one diagnosis, each event or diagnosis is presented in the corresponding morbidity category. The most common diagnostic group was uterus rupture or obstetric laparotomy, or both (1.18 per 1000 deliveries), followed by pre-eclampsia (HELLP and DIC) and eclampsia (0.68 per 1000 deliveries), and cardiovascular diseases (0.43 per 1000 deliveries). In the bivariate analyses women from low-income countries had an increased risk of maternal near-miss events in all morbidity groups, compared with Swedish-born women, whereas the percentage distribution of morbidity categories was almost the same except for cardiovascular diseases. The excess risk remained significant after adjustment for background and socio-economic factors but not for cardiovascular diseases and sepsis.

Table 3. Frequency of near-miss events per 1000 deliveries by diagnostic groups, crude and adjusted odds ratios with 95% confidence intervals, Sweden 1998–2007
 Organ failureShockCerebrovascular diseasesCardiovascular diseasesPre-eclampsia, eclampsiaRuptureSepsis
  1. a

    Model 2 adjustment for age, parity, Body Mass Index, smoking, type of hospital, education and cohabiting status.

Low-income countries (n = 23 278)
n 1417101245699
Rate per 10000.60.730.430.521.932.960.39
Crude
OR4.442.532.521.153.02.532.44
95% CI2.54–7.761.55–4.151.32–4.800.65–2.052.21–4.071.95–3.281.24–4.81
Model 2a
OR4.31.982.420.963.032.521.94
95% CI 1.10–3.581.01–5.770.47–1.982.04–4.491.82–3.470.80–4.68
Middle-income countries (n = 110 809)
n 254421376811328
Rate per 10000.230.400.190.340.611.020.25
Crude
OR1.661.381.110.750.950.881.60
95% CI1.07–2.581.00–1.900.70–1.770.53–1.050.74–1.230.72–1.081.06–2.41
Model 2a
OR 1.231.060.620.970.911.77
95% CI 0.84–1.810.60–1.880.40–0.950.73–1.300.72–1.161.10–2.87
High-income countries (n = 35 257)
n 21561522519
Rate per 10000.060.430.170.430.621.450.26
Crude
OR0.421.471.000.950.971.261.61
95% CI0.10–1.700.87–2.490.44–2.270.57–1.600.63–1.480.94–1.700.82–3.18
Model 2a
OR 1.031.240.910.781.221.25
95% CI 0.55–1.960.54–2.880.50–1.650.45–1.350.87–1.700.51–3.08
Sweden (n = 745 130)
n 104216127336483847118
Rate per 10000.140.290.170.450.651.140.16
Total
n 1452921644006181080164
Rate per 10000.160.320.180.430.681.180.18

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information

Main findings

Our population-based study found that women born in low-income countries had an increased risk of near-miss morbidity, compared with those born in Sweden. The increased risk remained after adjustment for several well-known risk factors.

Strengths and limitations of the study

The major strength of this study is its large sample size, making it possible to examine relatively rare causes of maternal near-miss events. All hospital delivery units report to the Swedish MBR and the Swedish Hospital Discharge Register. As exposure data are obtained prospectively during a woman's pregnancy, this has little effect on risk estimates. A register study is dependent on the accuracy of the ICD-10 codes. A medical record study on maternal delivery diagnoses in the Swedish MBR showed that only 3.5% of the diagnoses were incorrect, doubtful or might have been replaced by a more suitable diagnosis.[13] In another study, 93% of the pregnancies coded as pre-eclampsia in MBR had the disease according to the individual records.[14] Only relying on register data lessens the clinical interpretation of the results. The WHO recommends prospective data collection with criteria based on organ system dysfunction when studying maternal near-miss.[8, 15] The same criteria, however, cannot be applied in register studies. Therefore, we used a mix of clinical diagnostic criteria and management criteria and concentrated only on severe endpoints. In so doing we hoped to approximate the concept of organ system dysfunction as far as possible.

In previous register studies on severe maternal morbidity the diagnosis of uterine rupture is sometimes omitted.[16] The severity of uterine rupture can vary from inconsequential to catastrophic. A US study on hospital discharge data that monitored uterine rupture found that the code was not applied consistently over time.[17] In other register studies on severe maternal morbidity uterine rupture is one of the diagnoses used as a criterion.[18, 19] As uterine rupture was the most common diagnosis in our study, and it is known that uterine rupture as a maternal near-miss criterion may be questioned, we repeated the analysis excluding the diagnosis group uterine rupture or obstetric laparotomy. This did not change our results: women from low-income countries still had an increased risk of near-miss of 2.20 after adjustment according to Model 2 (95% CI 1.73–2.81).

We did not know how long the foreign-born women in our study had lived in Sweden. Worse pregnancy outcomes have been reported for recent immigrants, compared with those who had lived in Switzerland longer and had reached a better social status.[20] Unfortunately, asylum seekers and illegal immigrants, who have been shown to be at particularly high risk of severe maternal morbidity in the Netherlands,[21] could not be included in our study. They do not have a personal identification number that is consistently used in the registers.

We only included singleton deliveries with a gestational age ≥28 weeks. Twin pregnancies are high-risk pregnancies and severe prematurity can be caused by serious illness of the mother. If multiple pregnancies and premature deliveries (≥22 to ≤28 weeks of gestation) had been included, then the maternal near-miss frequency would most likely have been higher.

To our knowledge this is the first study to investigate maternal near-misses among immigrants that seeks to establish a more homogeneous classification by applying a categorisation of country of origin based on economic status. In research, countries of origin are commonly classified by geographic areas. As a result countries experiencing great economic and public health differences may be grouped together. The increased risk of near-miss morbidity among women from low-income countries and the unchanged risk for women from middle-income countries that we found agrees with the results of previous studies. In the Dutch LEMMoN study, women from Morocco and Turkey had no increased risk for severe maternal morbidity, whereas a 3.5-fold increased risk was seen for sub-Saharan African women.[7] Hence division of native countries into low-, middle- and high-income provides a picture of obstetric vulnerability that might not be as clear when looking at individual countries or regions.

The total frequency of maternal near-miss that we found, 2.9 per 1000 deliveries, was relatively low compared with other studies. The problem of diverging case identification criteria is well known when comparing near-miss studies.[22] Studies applying less strict criteria for near-miss, such as transfusion for haemorrhage, generally report higher near-miss figures between 5.1 and 13.8 per 1000 deliveries.[16, 18, 23] In our study, haemorrhage was regarded severe enough to cause near-miss only in women where there was a diagnosis of shock or hysterectomy. In an Irish audit study of near-miss events based on organ system dysfunction criteria, the frequency was 1.4 per 1000 deliveries.[24] Our results, therefore, are in agreement with the WHO recommendations for using restrictive inclusion criteria.

Interpretation

This study shows an increased risk of near-miss morbidity among women from low-income countries compared with native Swedes, consistent with an excess risk for maternal deaths and other reproductive outcomes in other settings.[3, 25] There are several plausible mechanisms for this association. The role of immigrant “naturalisation” and a country's integration policy in relation to pregnancy outcomes have been examined in previous studies.[26] Immigration stress, the rupture of previous social networks, low socio-economic status, poor access to healthcare services, for example communication problems and discrimination within the health system alone may interact to produce poorer outcomes in immigrant than in native communities.[26]

Another explanatory model for the increased risk of maternal near-miss among women from low-income countries might be the higher morbidity in this group.[27-30]

The distribution by category did not differ between women from low-income countries and Swedish-born women, whereas the rates did. The interpretation of the excess risk by categories for the near-misses from low-income countries is complex. The excess rate of cerebrovascular diseases might be a result of premorbidity conditions. The excess rate of pre-eclampsia (HELLP, DIC), eclampsia and sepsis suggests a proneness for severity of pregnancy complications, while shock and rupture might be more interpreted in terms of quality of care. Substandard care has been shown to be disproportionally more common in cases of maternal and perinatal mortality among non-western immigrant women compared with indigenous maternal and perinatal deaths.[31-33] Hence the results in this study call for special attention on women from low-income countries in planning pregnancy and obstetric-care policies as well as an alertness to this group's proneness for most pregnancy and delivery complications in clinical practice. A theoretical model has been proposed to understand barriers to good care for immigrant women, including miscommunication and mistrust as important aspects.[34] Consistent use of professional interpreters, individual plans for obstetric care and the establishment of sociocultural sensitive guidelines are suggested to improve quality of care for immigrant women.[34] To further understand the web of causation behind the increased risk of near-miss morbidity in our study, audit studies of barriers and quality of care may be useful.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information

Our analysis showed a 2.3-fold increased risk of maternal near-miss among women from low-income countries, compared with Swedish-born women. Although the rate is low it should alert healthcare providers.

Disclosure of interests

The authors affirm that they have no conflicts of interest.

Contribution to authorship

BE had the original idea of the study and designed it together with ÅW and MR. ÅW and BH collected the data and did the statistical analysis. All authors participated in the analysis of data. ÅW wrote the final paper in collaboration with the other authors, and all authors approved the final version.

Details of ethics approval

This study was approved by the local ethics committee of Uppsala University on 13 January 2010 (Dnr 2009/394).

Funding

This work was supported by a grant from the Swedish Council for Working Life and Social Research (FAS 2007-2026) and by the Medical Faculty of Uppsala University.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information
  • 1
    Reime B, Janssen PA, Farris L, Borde T, Hellmers C, Myezwa H, et al. Maternal near-miss among women with a migrant background in Germany. Acta Obstet Gynecol Scand 2012;91:8249.
  • 2
    Schutte JM, de Jonge L, Schuitemaker NW, Santema JG, Steegers EA, van Roosmalen J. Indirect maternal mortality increases in the Netherlands. Acta Obstet Gynecol Scand 2010;89:7628.
  • 3
    Esscher A, Haglund B, Högberg U, Essén B. Excess mortality in women of reproductive age from low-income countries: a Swedish national register study. Eur J Public Health 2013;23:2749.
  • 4
    Jonkers M, Richters A, Zwart JJ, Öry F, van Roosmalen J. Severe maternal morbidity among immigrant women in the Netherlands: patients' perspectives. Reprod Health Matters 2011;19:14453.
  • 5
    The Confidential Enquiry into Maternal and Child Health (CEMACH). Saving Mother's Lives: reviewing maternal deaths to make motherhood safer, 2003–2005. The seventh report on confidential enquiries into maternal deaths in the United Kingdom. Lewis G, editor. London: CEMACH; 2007.
  • 6
    Knight M, Kurinczuk JJ, Spark P, Brocklehurst P. Inequalities in maternal health: national cohort study of ethnic variation in severe maternal morbidities. BMJ 2009;338:b542.
  • 7
    Zwart JJ, Jonkers MD, Richters A, Öry F, Bloemenkamp KW, Duvekort JJ, et al. Ethnic disparity in severre acute maternal morbidity: a nationwide cohort study in the Netherlands. Eur J Public Health 2011;21:22934.
  • 8
    Say L, Souza JP, Pattinson RC. Maternal near miss – towards a standard tool for monitoring quality of maternal health care. Best Pract Res Clin Obstet Gynaecol 2009;23:28796.
  • 9
    Number of Births, Deaths, Immigrants, Emigrants and Population Size. Year 2008–2010. Stockholm: Statistics Sweden. c.2008–2010 [http://www.ssd.scb.se/databaser/makro/]. Accessed 10 October 2011.
  • 10
    International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10). Geneva: WHO. c1992. [http://www.who.int/classifications/icd/icdonlineversions/en/index.html]. Accessed 15 September 2010.
  • 11
    Gross National Income. The World Bank. World Bank country classification. c2009. [http://go.worldbank.org/U9BK71A1JO]. Accessed 25 August 2009.
  • 12
    Luo ZC, Wilkins R, Kramer MS. Effect of neighbourhood income and maternal education on birth outcomes: a population-based study. CMAJ 2006;174:141522.
  • 13
    Källén B, Otterblad Olausson P. The Swedish Medical Birth Register – A Summary of Content and Quality. Report No. 2003-112-3. Stockholm: Centre for Epidemiology, The National Board of Health and Welfare; 2003.
  • 14
    Ros HS, Cnattingus S, Lipworth L. Comparison of risk factors for preeclampsia and gestational hypertension in a population-based cohort study. Am J Epidemiol 1998;147:106270.
  • 15
    Say L, Pattinson RC, Gulmezoglu AM. WHO systematic review of maternal morbidity and mortality: the prevalence of severe acute maternal morbidity (near miss). Reprod Health 2004;1:3.
  • 16
    Callaghan WM, Mackay AP, Berg CJ. Identification of severe maternal morbidity during delivery hospitalizations, United States, 1991–2003. Am J Obstet Gynecol 2008;199:133. e1–8.
  • 17
    Weiss J, Nannini A, Fogerty S, Sachs B, Frigoletto F, Roberts D, et al. Use of hospital discharge data to monitor uterine rupture- Massachusetts, 1990–1997. MMWR 2000;49:2458.
  • 18
    Joseph KS, Liu S, Rouleau J, Kirby RS, Kramer MS, Sauve R, et al. Severe maternal morbidity in Canada, 2003 to 2007: surveillance using routine hospitalization data and ICD-10CA codes. J Obstet Gynaecol Can 2010;32:83746.
  • 19
    Baskett TF, O'Connell CM. Severe obstetric maternal morbidity: a 15-year population-based study. J Obstet Gynaecol 2005;25:79.
  • 20
    Lehmann P, Mamboury C, Minder CE. Health and social inequities in Switzerland. Soc Sci Med 1990;31:36986.
  • 21
    Van Hanegem N, Miltenburg AS, Zwart JJ, Bloemenkamp KW, Van Roosmalen J. Severe acute maternal morbidity in asylum seekers: a two-year nationwide cohort study in the Netherlands. Acta Obstet Gynecol Scand 2011;90:10106.
  • 22
    Zhang WH, Alexander S, Bouvier-Colle MH, Macfarlane A. Incidence of severe pre-eclampsia, postpartum haemorrhage and sepsis as a surrogate marker for severe maternal morbidity in a European population-based study: the MOMS-B survey. BJOG 2005;112:8996.
  • 23
    Zwart JJ, Richters JM, Ory F, de Vries JI, Bloemenkamp KW, van Roosmalen J. Severe maternal morbidity during pregnancy, delivery and puerperium in the Netherlands: a nationwide population-based study of 371,000 pregnancies. BJOG 2008;115:84250.
  • 24
    Lynch CM, Sheridan C, Breathnach FM, Said S, Daly S, Byrne B. Near miss maternal morbidity. Ir Med J 2008;101:1346.
  • 25
    Hill AB. The Environment and disease: association or causation? Proc R Soc Med 1965;58:295300.
  • 26
    Bollini P, Pampallona S, Wanner P, Kupelnick B. Pregnancy outcome of migrant women and integration policy: a systematic review of the international literature. Soc Sci Med 2009;68:45261.
  • 27
    Farah MG, Tverdal A, Selmer R, Heldal E, Bjune G. Tuberculosis in Norway by country of birth, 1986–1999. Int J Tuberc Lung Dis 2003;7:2325.
  • 28
    Achonu C, Jamieson F, Behr MA, Lillebaek T, Khan K, Gardam M. Evidence for local transmission and reactivation of tuberculosis in the Toronto Somali community. Scand J Infect Dis 2006;38:77881.
  • 29
    Karthikeyan G, Mayosi BM. Is primary prevention of rheumatic fever the missing link in the control of rheumatic heart disease in Africa? Circulation 2009;120:70913.
  • 30
    UNAIDS/WHO. AIDS Eoidemic Update 2009. Geneva: UNAIDS/WHO; 2009.
  • 31
    van Roosmalen J, Schuitemaker NW, Brand R, van Dongen PW. Bennebroek Gravenhorst J. Substandard care in immigrant versus indigenous maternal deaths in The Netherlands. BJOG 2002;109:2123.
  • 32
    Philibert M, Deneux-Tharaux C, Bouvier-Colle MH. Can excess maternal mortality among women of foreign nationality be explained by suboptimal obstetric care? BJOG 2008;115:14118.
  • 33
    Essen B, Bödker B, Sjöberg NO, Langhoff-Roos J, Greisen G, Gudmundsson S, et al. Are some perinatal deaths in immigrant groups linked to suboptimal perinatal care services? BJOG 2002;109:67782.
  • 34
    Binder P, Johnsdotter S, Essén B. Conceptualising the prevention of adverse obstetric outcomes among immigrants using the ‘three delays’ framework in a high-income context. Soc Sci Med 2012;75:202836.
Commentary on ‘Increased risk of severe maternal morbidity (near-miss) among immigrant women in Sweden: a population register-based study’

In this nationwide study from Sweden, an increased risk of severe maternal morbidity in women who had migrated from low-income countries compared with native Swedish women is confirmed with an odds ratio of 2.3 (95% confidence interval 1.9–2.8). Although only based on registers, its nationwide aspect is important, because most studies addressing ‘maternal near-miss’ are facility-based and originate from tertiary-care centres (Tunçalp et al. BJOG 2012;119:653–61).

This study will be used to compare maternal morbidity in Sweden with that of other settings. This comparison is difficult, because case selection is at stake. One could ask why women with multiple pregnancies or very preterm births (<28 weeks of gestation) are excluded. In addition, women from an unknown country of origin are less likely native Swedish women, but are excluded from analysis.

Classification of severe maternal morbidity is complicated, as can also be judged from this paper: ‘The clinical diagnosis … were acute renal failure, …, respiratory failure, shock, cerebrovascular failure, …, severe (pre-)eclampsia, uterine rupture, …’. Both eclampsia and uterine rupture may manifest themselves with renal, respiratory and cerebrovascular failure. From audit of organ dysfunction only, no lessons can be learnt as to what we should do to prevent the events leading to the dysfunction. Audit of clinical criteria such as ‘eclampsia’, however, will allow us to answer relevant questions such as: was magnesium sulphate given in time to a woman suffering from severe pre-eclampsia in order to prevent eclampsia?

With the aim to develop standard definitions and uniform identification criteria, the World Health Organization developed a Maternal Near-Miss Approach, concentrating on women presenting with features of organ dysfunction. One of their goals is to enable comparisons across different settings, particularly including low-income countries. The near-miss classification system has been applied now in a number of countries including Brazil and South Africa. A recent study from Malawi, however, indicates that this approach may not be sufficient to detect severe maternal morbidities in low-income countries (Van den Akker et al. PLoS One 2013;8:e54805). In those places where near-misses occur most frequently, organ dysfunction is not easily detected. Moreover, efforts should be directed towards the prevention of organ dysfunction.

The World Health Organization Maternal Near-Miss Approach can be difficult to apply in low-income settings or poorly resourced health systems. For example, it includes the use of more than four units of blood for transfusion as a criterion for organ dysfunction, while at the same time the use of any blood product would suffice to include a woman as a case of severe maternal morbidity. More than four units of blood is often used as a criterion in studies from high-resource countries. This leads to the contradictory conclusion that haemorrhage has a low prevalence in resource-poor settings and seriously underestimates its devastating effects (Nelissen et al. PLoS One 2013;8:e61248).

The Swedish paper adds to the ongoing discussion about an appropriate approach to the study of severe maternal morbidity. The last word on this topic, however, has not yet been said.

Declaration of interests

No competing interests.

  • J van Roosmalen, T van den Akker

  • Department of Obstetrics, Leiden University Medical Centre, Leiden, the Netherlands

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
bjo12326-sup-0001-TableS1.pdfapplication/PDF51KTable S1. The World Bank Country Classification of July 1, 2009.
bjo12326-sup-0002-AppendixS1.pdfapplication/PDF12KAppendix S1. Diagnosis and procedure codes used for identifying maternal near-miss cases.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.