Advertisement
JOGC

A Retrospective Analysis of Stillbirth Epidemiology and Risk Factors Among First Nations and Non-First Nations Pregnancies in Alberta From 2000 to 2009

      Abstract

      Objective

      Using a large administrative dataset we examined stillbirth epidemiology in First Nations and non-First Nations pregnancies in Alberta, focusing on previously unexplored longitudinal trends.

      Methods

      We undertook a retrospective analysis of de-identified data from 426 945 delivery records for the years 2000 to 2009. Age-adjusted prevalence of antepartum and intrapartum stillbirth were calculated and compared by ethnicity, as were longitudinal changes via average annual percent change analyses. Risk factors were explored via multivariable logistic regression analysis.

      Results

      Overall age-adjusted prevalence of antepartum and intrapartum stillbirth was significantly higher (P<0.001) in First Nations pregnancies than in non-First Nations pregnancies, and prevalence remained stable over time in both groups. Pre-existing diabetes was a strong predictor of stillbirth.

      Conclusion

      Stillbirth prevalence remains higher in First Nations pregnancies than in non-First Nations. Improved awareness of pre-existing diabetes and effective interventions are needed in First Nations women to decrease stillbirth risk.

      Résumé

      Objectif

      En utilisant un important ensemble de données administratives, nous nous sommes penchés sur l’épidémiologie de la mortinaissance dans le cas des grossesses chez des femmes issues ou non des Premières Nations en Alberta; nous nous sommes alors centrés sur des tendances longitudinales qui n’avaient pas déjà été explorées.

      Méthodes

      Nous avons mené une analyse rétrospective de données anonymisées issues de 426 945 dossiers d’accouchement pour la période 2000-2009. La prévalence (corrigée en fonction de l’âge) de la mortinaissance antepartum et intrapartum a été calculée et comparée en fonction de l’origine ethnique, tout comme les modifications longitudinales l’ont été par l’intermédiaire d’analyses des modifications annuelles moyennes en pourcentage. Les facteurs de risque ont été explorés par analyse de régression logistique multivariée.

      Résultats

      La prévalence (corrigée en fonction de l’âge) globale de la mortinaissance antepartum et intrapartum était considérablement plus élevée (P<0,001) dans le cadre des grossesses chez des femmes issues des Premières Nations que dans celui des grossesses chez des femmes n’étant pas issues des Premières Nations; cette prévalence est demeurée stable avec le temps au sein des deux groupes. Le diabète préexistant constituait un important facteur prédictif pour ce qui est de la mortinaissance.

      Conclusion

      La prévalence de la mortinaissance demeure plus élevée chez les femmes des Premières Nations. Pour en venir à abaisser le risque de mortinaissance chez celles-ci, nous devons nous efforcer d’améliorer la détection du diabète préexistant et de mettre en œuvre des interventions efficaces.

      Key Words

      INTRODUCTION

      Globally, Indigenous women have been shown to be at elevated risk of adverse pregnancy outcomes, including stillbirths which are 68% more likely to occur in Indigenous than in non-Indigenous women according to a systematic review and meta-analysis.
      • Shah P.R.
      • Zao J.
      • Al-Wassia H.
      • Shah V.
      Knowledge Synthesis Group on Determinants of Preterm/LBW Births. Pregnancy and neonatal outcomes of Aboriginal women: a systematic review and meta-analysis.
      In Canada, this has sparked new research into stillbirth among First Nations women, with Auger et al.
      • Auger N.
      • Park A.L.
      • Zoungrana H.
      • McHugh N.G.L.
      • Luo Z.C.
      Rates of stillbirth by gestational age and cause in Inuit and First Nations populations in Quebec.
      pointing toward diabetes (driven by obesity) as a major cause. The purpose of this study was to gain a better understanding of the extent of the problem of stillbirths and the relationship of stillbirth with diabetes and other potential pregnancy-related factors. In addition, we sought to quantify longitudinal trends in stillbirth prevalence, because these have yet to be rigorously explored in First Nations populations. It was our hope that such knowledge will better inform the type and timing of pregnancy intervention, ultimately leading to a lowered risk of stillbirth.

      METHODS

      We conducted a secondary analysis of data previously obtained.
      • Oster R.T.
      • King M.
      • Morrish D.W.
      • Mayan M.J.
      • Toth E.L.
      Diabetes in pregnancy among First Nations women in Alberta.
      Briefly, we obtained anonymized data for all provincial delivery records for the years 2000 to 2009 from the Alberta Perinatal Health Program (APHP). The year 2000 was the earliest in which fully populated data were available for the entire province. The APHP comprehensively collects perinatal data from the provincial delivery record for all hospital births and registered midwife-attended home births. Delivery record information is recorded by a health care provider (usually a nurse) when a woman presents for delivery. This information is obtained from prenatal records and/or from the patient.
      Stillbirth was defined in the APHP dataset as the birth (after20 weeks or attaining a weight of500 g) of a fetus that had died in utero. Stillbirths were further classified as antepartum and intrapartum. During the time period of the obtained data, the diagnosis of gestational diabetes mellitus followed Canadian Diabetes Association guidelines.
      • Canadian Diabetes Association
      Canadian Diabetes Association 2008 clinical practice guidelines for the prevention and management of diabetes in Canada.
      A diagnosis of pre-existing diabetes was made from an assessment of patient history, clinical chart records, and medication records (insulin or oral hypoglycemic agents). In addition to stillbirth and diabetes, other maternal risk factors explored included age, parity, ethnicity, pre-existing hypertension, diabetes retinopathy, proteinuria, pre-gestational weight, anemia, multiple gestation, illicit drug dependence, smoking use, alcohol use, history of abortion, history of preterm infant, history of neonatal death, history of stillbirth, history of Caesarean section, history of fetal anomaly, history of small for gestational age neotate, and history of large for gestational age neonate. A detailed description of all of the included variables is provided in our previous publication.
      • Oster R.T.
      • King M.
      • Morrish D.W.
      • Mayan M.J.
      • Toth E.L.
      Diabetes in pregnancy among First Nations women in Alberta.
      For each variable included in the analysis, data were complete or nearly complete (available for 97% to 100% of pregnancies). A total of 426 945 delivery records with valid stillbirth data (gestation20 weeks) were used.
      The data were sent to Alberta Health, matched via the personal health number, and First Nations ethnicity was ascertained. Three unique and distinct populations of Canadian Indigenous people are recognized in Canada: First Nations, Métis, and Inuit. Under the federal Indian Act of Canada, First Nations and Inuit individuals whose nations have engaged in treaties are accorded “Registered Indian” or treaty status. The Alberta Health Care Insurance Plan Central Stakeholder Registry file includes an identifier for such individuals. Thus, women delivering in Alberta with a First Nations identifier (First Nations or Inuit) were considered “First Nations,” whereas all other women (including First Nations individuals without treaty status and Métis individuals) were considered “non-First Nations” because they could not be identified. Although we recognize that these groupings are imperfect, we and other researchers are limited to the federal definitions when working with large administrative datasets. In Alberta, there are approximately 116 670 First Nations people (19 945 of these are non-registered) and 96 865 Métis people, representing approximately 3.3% and 2.7% of the total population, respectively.
      • Statistics Canada
      National Households Survey, 2011.
      Very few Inuit people (approximately 1600) reside in Alberta.
      Annual age-adjusted prevalence values and 95% confidence intervals of stillbirth (antepartum and intrapartum) were compared by ethnicity using chi-square analysis. For the prevalence calculations, the denominator was the total number of pregnancies in which data for that variable were available for the specific group of interest. The numerator was the total number of pregnancies in which the criteria for the variable of interest were met, for the specific group of interest. Average annual percentage change values and 95% confidence intervals in stillbirth prevalence over time were compared by ethnicity using parallelism tests.
      • Kim H.J.
      • Fay M.P.
      • Binbing Y.
      • Barrett M.J.
      • Feuer E.J.
      Comparability of segmented line regression models.
      Statistical modelling (purposeful) using multivariable logistic regression was used to evaluate the relationships between stillbirth, diabetes in pregnancy (both gestational diabetes and pre-existing diabetes, separately), ethnicity and other possible explanatory variables. Odds ratios and 95% confidence intervals were calculated. Briefly, independent variables with a P value<0.20 in the univariate linear regression analysis were fitted in a multivariable model. Variables with a P value>0.05 were removed, and the potential confounding effect of each variable was assessed. The linear assumptions of continuous variables and potential interaction effects were assessed. The Hosmer-Lemeshow test was used to determine the model goodness-of-fit.
      The Human Research Ethics Board of the University of Alberta approved the research.

      RESULTS

      The overall age-adjusted prevalence of antepartum stillbirth was significantly higher in First Nations pregnancies (1.30%; 95% CI 1.27% to 1.34%) than in non-First Nations pregnancies (0.46%; 95% CI 0.44% to 0.48%) (P<0.001). Similarly, the overall age-adjusted prevalence of intrapartum stillbirth was significantly higher in First Nations pregnancies (0.39%; 95% CI 0.37% to 0.41%) than in non-First Nations pregnancies (0.22%; 95% CI 0.20% to 0.23%) (P<0.001). The age-adjusted prevalence of stillbirth remained stable over time in both populations (Figure 1, Figure 2). The average annual percentage change for antepartum stillbirth was 1.63 (95% CI −6.79 to 10.81) among First Nations women and 0.47 (95% CI −1.11 to 2.07) among non-First Nations women, with no between-group differences (P=0.475). The average annual percentage change for intrapartum stillbirth was 0.37 (95% CI −8.33 to 9.90) among First Nations women and −1.13 (95% CI −5.15 to 3.05) among non-First Nations women, and again no between-group differences were noted (P=0.694).
      Figure thumbnail gr1
      Figure 1Age-adjusted prevalence of antepartum stillbirth over time in Alberta by ethnicity
      Figure thumbnail gr2
      Figure 2Age-adjusted prevalence of intrapartum stillbirth over time in Alberta by ethnicity
      In multivariable analysis (Table), First Nations ethnicity was an independent predictor of antepartum and intrapartum stillbirth. For First Nations women specifically, potentially manageable risk factors for antepartum stillbirth included pre-existing diabetes and drug dependence. Pre-existing diabetes was also a risk factor for intrapartum stillbirth among First Nations women.

      DISCUSSION

      Our study confirms an increased prevalence and risk of stillbirth in First Nations populations compared with the general population. Although it is encouraging that the prevalence of stillbirth is not increasing, the prevalence of stillbirth decreased in high-income nations from 0.39% to 0.20% between 1995 and 2009,
      • Cousens S.
      • Blencowe H.
      • Stanton C.
      • Chou D.
      • Ahmed S.
      • Steinhardt L.
      • et al.
      National, regional, and worldwide estimates of stillbirth rates in 2009 with trends since 1995: a systematic analysis.
      and Alberta’s performance may still be suboptimal. In a study of First Nations women in Quebec by Auger et al., predictors of stillbirth included diabetes (OR 15.30) and hypertension (OR 2.25), leading the authors to speculate that obesity was the major underlying cause.
      • Auger N.
      • Park A.L.
      • Zoungrana H.
      • McHugh N.G.L.
      • Luo Z.C.
      Rates of stillbirth by gestational age and cause in Inuit and First Nations populations in Quebec.
      In our study both hypertension and weight91 kg did not remain significant predictors in our final adjusted models. This suggests that although hypertension, diabetes, and obesity are no doubt linked pathophysiologically and are often comorbidities, it is primarily diabetes that is contributing to stillbirths in First Nations women. Moreover, it is pre-existing diabetes and not gestational diabetes that health care providers should be concerned about regarding stillbirth risk, a finding substantiated by data from First Nations women in Manitoba.
      • Aljohani N.
      • Rempel B.M.
      • Ludwig S.
      • Morris M.
      • Cheang M.
      • Murray R.
      • et al.
      Impact of diabetes on maternal-fetal outcomes in Manitoba: relationship with ethnic and environmental factors.
      It is noteworthy that type 2 diabetes has a higher incidence and an increasing prevalence in young First Nations women.
      • Oster R.T.
      • Johnson J.A.
      • Balko S.U.
      • Svenson L.W.
      • Toth E.L.
      Increasing rates of diabetes amongst status Aboriginal youth in Alberta.
      • Oster R.T.
      • Johnson J.A.
      • Hemmelgarn B.R.
      • King M.
      • Balko S.U.
      • Svenson L.W.
      • et al.
      Recent epidemiologic trends of diabetes mellitus among status Aboriginal adults.
      Tabled 1Significant adjusted multivariate predictors of antepartum and intrapartum stillbirths among Albertan women by ethnicity
      All Women (n=426 945)First Nations Women (n=28 296)
      OR (95% CI) for antepartum stillbirthOR (95% CI) for intrapartum stillbirthOR (95% CI) for antepartum stillbirthOR (95% CI) for intrapartum stillbirth
      Age35
      Compared with age 18 to 34
      1.58 (1.42 to 1.76)1.37 (1.16 to 1.63)1.67 (1.15 to 2.42)1.78 (1.03 to 3.24)
      Parity (0)
      Compared with parity=1 or 2
      1.37 (1.24 to 1.51)1.27 (1.10 to 1.46)1.47 (1.07 to 2.04)
      Parity (≥3)
      Compared with parity=1 or 2
      1.40 (1.22 to 1.62)0.65 (0.50 to 0.86)1.50 (1.10 to 2.04)
      Multiple gestation2.84 (2.44 to 3.31)1.80 (1.37 to 2.35)
      Ethnicity1.70 (1.47 to 1.97)1.62 (1.29 to 2.03)NANA
      Pre-existing hypertension2.46 (1.83 to 3.29)1.95 (1.16 to 3.27)
      Gestational diabetes0.32 (0.18 to 0.57)
      Pre-existing diabetes2.11 (1.48 to 3.00)2.38 (1.42 to 3.99)2.13 (1.07 to 4.63)3.20 (1.14 to 8.97)
      Pre-gestational weight91 kg0.62 (0.47 to 0.82)
      Anemia1.84 (1.29 to 2.64)2.26 (1.35 to 3.78)
      Drug dependent2.04 (1.59 to 2.62)2.12 (1.46 to 3.07)
      Any smoking1.29 (1.16 to 1.44)
      History of abortion1.52 (1.30 to 1.78)2.35 (1.92 to 2.88)1.71 (1.19 to 2.46)2.90 (1.73 to 4.87)
      History of neonatal death1.74 (1.20 to 2.52)2.84 (1.47 to 5.53)
      History of stillbirth3.85 (3.10 to 4.78)3.83 (2.68 to 5.46)3.06 (1.88 to 4.98)2.89 (1.30 to 6.41)
      History of Caesarean section0.62 (0.48 to 0.80)
      History of fetal anomaly1.80 (1.27 to 2.54)2.84 (1.82 to 4.44)
      History of large for gestational age neonate0.58 (0.34 to 0.98)
      * Compared with age 18 to 34
      Compared with parity=1 or 2
      As non-registered First Nations individuals and Métis individuals could not be identified and were included in the non-First Nations population group, our results cannot be generalized to these groups and the observed ethnic (First Nations vs. non-First Nations) disparities may be underestimated. Whether recognized pre-existing diabetes cases were type 1 diabetes or type 2 diabetes could not be discerned from the administrative data. Research is needed to determine the contribution of type 1 and type 2 diabetes specifically, but we speculate that diabetes control is more important than diabetes type with respect to stillbirth. The contribution of other potential contextual predictors to the logistic regression models such as health care access, lifestyle, social environment, and socioeconomic status (which have been shown to be associated with stillbirth)
      • Flenady V.
      • Koopmans L.
      • Middleton P.
      • Frøen J.F.
      • Smith G.C.
      • Gibbons K.
      • et al.
      Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis.
      could not be assessed. We speculate that ethnic disparities in these factors likely contributed to the increased risk of stillbirth in the First Nations women.

      CONCLUSION

      Our study is the first to explore the age-adjusted prevalence of stillbirth longitudinally in a First Nations population in Canada. Although the prevalence of stillbirth has remained stable in Alberta over 10 years, it remains persistently higher in First Nations pregnancies than in the general population. This may be due in part to the higher incidence and increasing prevalence of type 2 diabetes among young First Nations women. Increased early awareness of pre-existing diabetes could lead to better recognition and management and decreased numbers of stillbirths.

      ACKNOWLEDGEMENTS

      This study was funded by the Canadian Institutes for Health Research, Institute of Aboriginal Peoples Health.
      This study is based on data provided by the Alberta Perinatal Health Program and Alberta Health. The interpretation and conclusions contained herein are those of the researchers and do not necessarily represent those of the Alberta Perinatal Health Program, Alberta Health, or the Government of Alberta. Neither the Alberta Perinatal Health Program, nor Alberta Health, nor the Government of Alberta expresses any opinion in relation to this study.

      REFERENCES

        • Shah P.R.
        • Zao J.
        • Al-Wassia H.
        • Shah V.
        Knowledge Synthesis Group on Determinants of Preterm/LBW Births. Pregnancy and neonatal outcomes of Aboriginal women: a systematic review and meta-analysis.
        Womens Health Issues. 2011; 21: 28-39
        • Auger N.
        • Park A.L.
        • Zoungrana H.
        • McHugh N.G.L.
        • Luo Z.C.
        Rates of stillbirth by gestational age and cause in Inuit and First Nations populations in Quebec.
        CMAJ. 2013; 185: E256-E262
        • Oster R.T.
        • King M.
        • Morrish D.W.
        • Mayan M.J.
        • Toth E.L.
        Diabetes in pregnancy among First Nations women in Alberta.
        Canada: a retrospective analysis. BMC Pregnancy Childbirth. 2014; 14: 136
        • Canadian Diabetes Association
        Canadian Diabetes Association 2008 clinical practice guidelines for the prevention and management of diabetes in Canada.
        Can J Diabetes. 2008; 32: S1-S201
        • Statistics Canada
        National Households Survey, 2011.
        Aboriginal peoples in Canada: First Nations people. Métis, and Inuit, Ottawa: Statistics Canada2011 (Available at:) (Accessed October 28, 2014)
        • Kim H.J.
        • Fay M.P.
        • Binbing Y.
        • Barrett M.J.
        • Feuer E.J.
        Comparability of segmented line regression models.
        Biometrics. 2004; 60: 1005-1014
        • Cousens S.
        • Blencowe H.
        • Stanton C.
        • Chou D.
        • Ahmed S.
        • Steinhardt L.
        • et al.
        National, regional, and worldwide estimates of stillbirth rates in 2009 with trends since 1995: a systematic analysis.
        Lancet. 2011; 377: 1319-1330
        • Aljohani N.
        • Rempel B.M.
        • Ludwig S.
        • Morris M.
        • Cheang M.
        • Murray R.
        • et al.
        Impact of diabetes on maternal-fetal outcomes in Manitoba: relationship with ethnic and environmental factors.
        Clin Invest Med. 2008; 31: E338-E345
        • Oster R.T.
        • Johnson J.A.
        • Balko S.U.
        • Svenson L.W.
        • Toth E.L.
        Increasing rates of diabetes amongst status Aboriginal youth in Alberta.
        Canada. Int J Circumpolar Health. 2012; 71: 1-7
        • Oster R.T.
        • Johnson J.A.
        • Hemmelgarn B.R.
        • King M.
        • Balko S.U.
        • Svenson L.W.
        • et al.
        Recent epidemiologic trends of diabetes mellitus among status Aboriginal adults.
        CMAJ. 2011; 183: E803-E808
        • Flenady V.
        • Koopmans L.
        • Middleton P.
        • Frøen J.F.
        • Smith G.C.
        • Gibbons K.
        • et al.
        Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis.
        Lancet. 2011; 377: 1331-1340