Reclassifying causes of obstetric death in Mexico: a repeated cross-sectional study

Bull World Health Organ 2016;94:362–369B | doi: http://dx.doi.org/10.2471/BLT.15.163360 Research 362 Reclassifying causes of obstetric death in Mexico: a repeated cross- sectional study Margaret C Hogan,a Biani Saavedra-Avendano,b Blair G Darney,b Luis M Torres-Palacios,c Ana L Rhenals-Osorio,c Bertha L Vázquez Sierra,c Patricia N Soliz-Sánchez,d Emmanuela Gakidoua & Rafael Lozanob Introduction Maternal mortality – defined as the death of a woman during pregnancy, childbirth or in the 42 days after delivery – is used as an outcome measure for any health system. The indicator of maternal mortality is the maternal mortality ratio (MMR), which is defined as the number of maternal deaths per 100 000 live births. MMR is a problematic indicator to measure, due to the relative few maternal deaths.1–5 It is even more challenging to measure cause-specific maternal mortality, since available methods have either low sensitivity or specificity.6 Therefore, accurate estimates of cause-specific maternal mortality are often not available in many countries.7 The International statistical classification of diseases and related health problems, 10th revision (ICD-10)8 manual divides the causes of maternal mortality into three broad categories: direct obstetric deaths (resulting from obstetric complications of the pregnant state); indirect obstetric deaths (resulting from a disease, often pre-existing and aggravated by the physiologic ef- fects of the pregnancy); and late maternal death (death between 42 days and one year post-obstetric event).8,9 Sequelae, which is maternal deaths that occur one year or more after delivery, is included in the late maternal death category. Indirect maternal deaths represent on average 20% of a country’s overall maternal mortality,10–12 but this proportion varies considerably across settings. Particularly indirect and late maternal causes are likely to be misclassified as non-maternal deaths.6,13,14 Even within countries with a very high quality vital registration system, there is a wide variation in the fraction of maternal deaths attributable to indirect causes; estimates range from zero to more than half of all reported maternal deaths.13–22 In recent decades, Mexico has improved its measurement of maternal mortality, but problems of underreporting and misreporting in its vital statistics systems still exist.23,24 In re- sponse, a new strategy was undertaken by the government in 2002 that aimed to identify all maternal deaths in Mexico using an approach of intentional search, review and reclassification of maternal deaths. A new procedure, referred to as Búsqueda intencionada y reclasificación de muertes maternas (BIRMM) was put in place,25,26 which provides a mechanism for a compre- hensive examination of maternal deaths in Mexico. There has been a rapid increase in obesity,27,28 diabetes, hypertension, and hypercholesterolaemia in Mexico,29,30 which puts women of reproductive age at higher risk for pre-existing hypertensive disorders and diabetes mellitus. The epidemiologic transition from communicable to noncommunicable diseases has implications for maternal health. The purpose of this study is to describe the reclassified and newly identified maternal deaths – especially indirect obstetric deaths – identified through the BIRMM process. We also compared sociodemographic characteristics at individual and municipality level of the women who died from direct and indirect causes. Methods We conducted a repeated cross-sectional study using the 2006– 2013 BIRMM data set, which includes all deaths in women Objective To describe causes of maternal mortality in Mexico over eight years, with particular attention to indirect obstetric deaths and socioeconomic disparities. Methods We conducted a repeated cross-sectional study using the 2006–2013 Búsqueda intencionada y reclasificación de muertes maternas (BIRMM) data set. We used frequencies to describe new cases, cause distributions and the reclassification of maternal mortality cases by the BIRMM process. We used statistical tests to analyse differences in sociodemographic characteristics between direct and indirect deaths and differences in the proportion of overall direct and indirect deaths, by year and by municipality poverty level. Findings A total of 9043 maternal deaths were subjected to the review process. There was a 13% increase (from 7829 to 9043) in overall identified maternal deaths and a threefold increase in the proportion of maternal deaths classified as late maternal deaths (from 2.1% to 6.9%). Over the study period direct obstetric deaths declined, while there was no change in deaths from indirect obstetric causes. Direct deaths were concentrated in women who lived in the poorest municipalities. When compared to those dying of direct causes, women dying of indirect causes had fewer pregnancies and were slightly younger, better educated and more likely to live in wealthier municipalities. Conclusion The BIRMM is one approach to correct maternal death statistics in settings with poor resources. The approach could help the health system to rethink its strategy to reduce maternal deaths from indirect obstetric causes, including prevention of unwanted pregnancies and improvement of antenatal and post-obstetric care. a University of Washington, Seattle, United States of America (USA). b National Institute of Public Health, Av. Universidad 655, Col. Santa Maria Ahucatitlan, 62100, Cuernavaca, Morelos, Mexico. c Secretaría de Salud, Mexico City, Mexico. d Pan American Health Organization, Washington, USA. Correspondence to Rafael Lozano (email: rafael.lozano@insp.mx). (Submitted: 25 August 2015 – Revised version received: 10 December 2015 – Accepted: 11 December 2015 ) Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360 363 Research Indirect and direct obstetric deaths in MexicoMargaret C Hogan et al. of reproductive ages (10–54 years). The aim was to identify miscoded maternal deaths. We investigated the underlying cause of death for those who had been assigned to a subset of 46 ICD-10 codes that we suspected of being maternal deaths but did not have maternal codes from the ICD-10 O chapter.25 (The list of the 46 ICD codes are available from the corresponding author). In addition to these 46 codes, we also investigated: deaths that were assigned maternal codes found in the ICD-10 O chapter (Table 1); deaths with suspicious or incomplete codes; deaths with complications but without a valid underlying cause of death; and all death certificates where the pregnancy checkbox had been ticked. The BIRMM review process collates maternal mortality and sociodemo- graphic data from available mortality data sources including death certificates, medical records, verbal autopsy records and confidential enquiry or autopsy re- ports. (The distribution of information contributed by each data source is avail- able from the corresponding author). The collated information was reviewed by two independent coders. In the event of a dis- agreement when assigning the appropri- ate code, a senior reviewer examined all documents and assigned the final cause. Full details of the BIRMM protocol have been published elsewhere.25 We extracted the following data from the BIRMM: individual-level sociodemo- graphic data (age, marital status, educa- tion level, number of pregnancies and number of prenatal visits during the last pregnancy); care-related information on each maternal death (place of care where first complication was documented and where skilled birth attendant care was provided); and place of death. The pro- portions of missing data for each year for the care-related information are available from the corresponding author. We used cause-of-death data as reported before and after the intentional review process. In addition, we used the 2010 municipality- level development index31 as a measure of community level socioeconomic status. The development index data were grouped into quintiles and merged into the individual level BIRMM data record according to the residence municipality of the deceased. To calculate the MMR, live births data for 2006–2013 were obtained from the General Directorate of Health Information.32 This study was approved by the ethics committee of the National Institute of Public Health, Mexico. Analysis Table 1 presents the ICD-10 categories and associated sub-categories and titles for codes used for deaths identified as indirect obstetric deaths, as listed in the manual. The list includes all codes that constitute the internationally agreed definition of indirect maternal death. Maternal deaths related to hu- man immunodeficiency virus (HIV) infections were treated as special cases due to evolving ICD-10 coding rules. HIV-related deaths coded to the spe- cific HIV ICD codes (B20-B24) and the recently created HIV ICD code (O98.7) were included and recoded as indirect maternal deaths. New and recoded indirect maternal deaths, their cause distribution and reclassification by development index quintile were examined using tabula- tions. We used descriptive statistics to examine differences in direct and indirect cause, before and after the re- view process and across the eight years (2006–2013). Sociodemographic char- acteristics and health service use were examined and tested using t-test and χ2 tests for differences in the proportion and means of direct and indirect deaths. Maternal mortality cause was analysed by municipality-level development index. All analyses were done in Stata version 13.1 (StataCorp. LP, College Sta- tion, United States of America). Results The total number of deaths in women of reproductive age during the study period obtained from the BIRMM was 357 446, of which 9043 deaths (2.5%) were sub- jected to the maternal death cause review and reclassification process.33 Table 2 summarizes the findings from the reclassification process. Before the review 7829 deaths were classified as maternal deaths. Of the deaths coded as non-maternal, we reclassified 1214 (13.4%) to maternal deaths. The number of late maternal deaths increased from 192 (2.1%) before the review to 628 (6.9%) after the review, representing over threefold increase in the proportion of late maternal deaths. There was a 6.8% (from 2099 to 2243 deaths) and 11.4% (from 5538 to 6172 deaths) increase in deaths categorized as indirect and direct, respectively. Deaths that were categorized the same before and after the review were, 85.5% (5281/6172) of di- rect, 68.8% (1544/2243) of indirect and 28.3 (178/628) of late maternal deaths. In the poorest municipalities, 14.7% (694) of maternal deaths were recoded from non-maternal to maternal, while in the wealthiest municipalities this figure was 8.1% (108). An additional 14.8% (30) of all direct deaths in the wealthi- est quintile were reclassified to indirect deaths (Table 2). Table 3 (avai lable at : http:// w w w . w h o . i n t / b u l l e t i n / v o l - umes/94/5/15-1633560) shows the original codes of the deaths recoded to indirect maternal deaths. Many indirect maternal deaths were originally misclas- sified as infectious and parasitic diseases (96), diseases of the circulatory system (94), diseases of the respiratory system (70) and neoplasms (56). In addition, there was miscoding within the mater- nal chapter of the ICD. For instance, 40 deaths were originally assigned to O10–O16 (oedema, proteinuria and hypertensive disorders in pregnancy, childbirth and the puerperium); 46 deaths were assigned to O21, O23–O31, O34 (other complications of pregnancy) and 46 more maternal deaths were as- signed with the codes O89–O92 (other complications during the puerperium). Two hundred and twenty indirect mater- nal deaths were coded as direct obstetric causes before the correction. Table 1. ICD-10 codes for indirect causes of maternal deaths8 Category Sub-category Title O10 O10.0–O10.9 Pre-existing hypertension complicating pregnancy, childbirth and the puerperium O24 O24.0–O24.9 Diabetes mellitus during pregnancy O98 O98.0–O98.9 Maternal infectious and parasitic diseases classifiable elsewhere but complicating pregnancy, childbirth and the puerperium O99 O99.0–O99.8 Other maternal diseases classifiable elsewhere but complicating pregnancy, childbirth and the puerperium ICD: International classification of diseases and related health problems, 10th revision. Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360364 Research Indirect and direct obstetric deaths in Mexico Margaret C Hogan et al. For maternal deaths categorized as direct, there was a declining trend in MMR between 2006 and 2013, from 46.4 to 32.1 deaths per 100 000 live births. There was no change for indirect ma- ternal deaths. MMR for indirect deaths was 12.2 deaths per 100 000 live births in 2006 and 13.3 deaths per 100 000 live births in 2013. There was a peak in 2009, mainly due to the Influenza A (H1N1) epidemic, which is known to increase the risk of hospitalization, severe ill- ness and death in pregnant women.34,35 The trends for both direct and indirect causes were similar before and after re- classifying the causes of deaths (Fig. 1). Comparison of sociodemographic characteristics and health system use showed that women who died of indirect maternal causes had fewer pregnancies, were slightly younger and were better educated than women dying of direct causes. The deceased women were also more likely to have delivered, received care for their first complication and died in Instituto Mexicano del Seguro Social facilities, which are employment-based insurance-affiliated facilities. Skilled birth attendants were more likely to have been present for the mothers who died of direct causes, but this could be due to the fact that women who died of indirect causes died before delivery (Table 4; available at: http://www.who. int/bulletin/volumes/94/5/15-1633560). For direct causes, poorer munici- palities had a higher MMR, but also a slightly higher ratio of maternal indirect deaths (Fig. 2). Fig. 3 and Fig. 4 show the MMR by direct and indirect causes between the wealthiest and poorest quintiles by year. Between 2006 and 2013, direct causes of maternal deaths among women residing in the poorest municipalities have nearly halved, go- ing down from 119.1 to 72.7 deaths per 100 000 live births (Fig. 3). The decline in MMR due to direct causes of maternal death in the wealthiest municipalities was 23.5%, going down from 35.2 to 26.9 deaths per 100 000 live births (Fig. 3). For each year and in both poorest and wealthiest quintiles, MMRs for indirect death causes were lower than the MMRs for direct causes. It is only in 2008 and 2010 that the poorest quintile had a sta- tistically significant higher MMR due to indirect causes than the wealthiest quin- tile (Fig. 4). In 2009, there was a peak in indirect deaths among the wealthiest municipalities, presumably due to the H1N1 epidemic (Fig. 4). Table 2. Summary of the intentional search and review outcomes of maternal deaths, by municipality-level development index quintiles, Mexico, 2006–2013 Cause of death After review Cause of death, no. (%) Total, no. (%)Indirecta Directb Late maternalc All quintiles Before review Non-maternald 479 (21.3) 365 (5.9) 370 (58.9) 1214 (13.4) Indirecta 1544 (68.8)e 514 (8.3) 41 (6.5) 2099 (23.2) Directb 218 (9.7) 5281 (85.5)e 39 (6.2) 5538 (61.2) Late maternalc 2 (0.1) 12 (0.2) 178 (28.3)e 192 (2.1) Total 2243 (100.0) 6172 (100.0) 628 (100.0) 9043 (100.0) Quintile 1 (poorest) Before review Non-maternald 280 (21.1) 193 (6.4) 221 (57.7) 694 (14.7) Indirecta 921 (69.4)e 303 (10.1) 22 (5.7) 1246 (26.4) Directb 124 (9.3) 2507 (83.3)e 23 (6.0) 2654 (56.2) Late maternalc 2 (0.2) 5 (0.2) 117 (30.5)e 124 (2.6) Total 1327 (100.0) 3008 (100.0) 382 (100.0) 4718 (100.0) Quintile 5 (wealthiest) Before review Non-maternald 38 (18.8) 36 (3.3) 34 (64.1) 108 (8.1) Indirecta 134 (66.3)e 48 (4.4) 5 (9.4) 187(13.9) Directb 30 (14.8) 1008 (92.1)e 1 (1.9) 1039 (77.0) Late maternalc 0 (0.0) 2 (0.2) 13 (24.5)e 15 (1.1) Total 202 (100.0) 1094 (100.0) 53 (100.0) 1350 (100.0) a Indirect is defined as maternal deaths resulting from a disease, often pre-existing and aggravated by the physiologic effects of the pregnancy. b Direct is defined as maternal deaths resulting from obstetric complications during pregnancy. c Late maternal is defined as death between 42 days and one year post-obstetric event. d All other maternal deaths not coded to direct, indirect or later maternal. e Concordant pair before and after review. Fig. 1. Direct and indirect maternal deaths before and after the review process, Mexico, 2006–2013 No . o f d ea th s p er 1 00 0 00 li ve b irt hs 50 45 40 35 30 25 20 15 10 5 0 Year Direct - after review Indirect - after review Indirect - before review 95% CI Direct - before review 2006 2007 2008 2009 2010 2011 2012 2013 CI: confidence interval. Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360 365 Research Indirect and direct obstetric deaths in MexicoMargaret C Hogan et al. Discussion This paper presents the results of a review and reclassification of causes of maternal death in Mexico, for the pe- riod 2006 to 2013. The identified 13% increase in the number of maternal deaths after the BIRMM review sug- gests that this type of exercise is one approach to correct misclassification of maternal cause-of-death data. There is progress in Mexico towards achiev- ing a reliable assessment of the causes of death in women of reproductive age, with the aim of generating im- proved data on the causes of maternal mortality. Our findings show that MMR from direct maternal deaths has been declin- ing between 2006 and 2013; it nearly halved among the poorest women. However, there was no such change in indirect deaths. Given the increase in the burden of noncommunicable diseases and associated risk factors,30 indirect deaths may continue to account for an increasing proportion of maternal deaths. Despite the overall decline in MMR from direct deaths, socioeconom- ic disparities among the poor persist. However, for indirect deaths, women residing in both poor and wealthy areas are affected. These findings are consis- tent with other studies that examined the obstetric death transition from direct to indirect causes.36 Similar projects to the BIRMM have been implemented in other settings as well; however there is a wide variation in the level of maternal death underreport- ing between settings. Correction factors for the number of maternal deaths range from 1.9 in north-east Brazil37,38 to 3.2 in Menoufia, Egypt.39 In high-income countries the correction factor ranges between 0.9 and 2.2.40 This observed variation limits our ability to generalize the extent of underreporting or misclas- sification across settings. Our review also highlighted the im- portance of the maternal death review to explain an unusual disease pattern, such as the 2009 H1N1 epidemic in Mexico. This type of temporary change in the pattern of maternal deaths has also been observed in Rwanda and South Africa.41 Such increases should be anticipated during an epidemic. The feasibility of a project that focuses on reclassifying maternal death causes in a given country is dependent on the level of maternal death misclas- sification or under-coding. In countries with relatively few maternal deaths, ex- tensive review of suspected cases may be feasible. For example, the reproductive age mortality study (RAMOS) investi- gates all reported deaths in women of reproductive age.9,42 In our study, we applied a cost-saving approach which uses expert opinion to identify a subset of maternal death cases.26 Our approach may be more appropriate for countries with poor resources. Most maternal health interventions are timed around the delivery period of the pregnancy and they focus mainly on skilled birth attendance or emergency obstetric care.43 These interventions have an impact on direct deaths and subsequently on the reduction of overall maternal mortality. However, except in cases where complications arise during labour, indirect deaths may not be avert- ed through these delivery-focused in- terventions.41 To reduce indirect deaths, obstetricians and other health-care per- sonnel interacting with pregnant women Fig. 2. Direct and indirect maternal deaths by municipality-level development index, Mexico, 2006–2013 95% CI No . o f m at er na l d ea th s pe r 1 00 0 00 li ve b irt hs 110 100 90 80 70 60 50 40 30 20 10 0 Quintile 1-poorest 2 3 4 5-wealthiest Indirect Direct CI: confidence interval. Note: Pooled number of indirect and direct deaths per 100 000 live births, by quintile of municipality-level development index, 2006–2013. Fig. 3. Direct maternal deaths for the wealthiest and poorest quintile, Mexico, 2006–2013 95% CI Year 2006 2007 2008 2009 2010 2011 2012 2013 Q1-poorest Q5-wealthiest No . o f m at er na l d ea th s p er 1 00 0 00 li ve b irt hs 140 120 100 80 60 40 20 0 CI: confidence interval; Q: quintile. Note: Number of direct deaths per 100 000 live births, by year, for the wealthiest and poorest quintile of the municipality-level development index. Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360366 Research Indirect and direct obstetric deaths in Mexico Margaret C Hogan et al. صخلم تاعاطقلا ةددعتم ةرركتم ةسارد :كيسكلما في ةدلاولاب ةطبترلما ةافولا بابسأ فينصت ةداعإ ىدم لىع كيسكلما في تاهملأا ينب ةافولا بابسأ فصو ضرغلا ةتجانلاو ةدلاولاب ةطبترلما ةافولا لىع مماتهلاا زيكرت عم ،تاونس نيماث .ةيداصتقلاا ةيعماتجلاا قراوفلا لىعو ،ةشرابم يرغ بابسأ نع  مادختساب  ةددعتم  تاعاطق  لمشت  ةرركتم  ةسارد  انيرجأ  ةقيرطلا Búsqueda intencionada y في  ةرفوتلما  تانايبلا  ةعوممج during the postpartum period need to be trained to treat the entire woman and not just her pregnancy.44,45 Planning for such training requires the development, dissemination and adoption of clinical guidelines.46 Effective implementation of such guidelines requires collaboration and the establishment of referral systems between specialities that deal with the major causes of indirect deaths.41 For example obstetricians need to be able to effectively communicate with chronic disease specialists regarding at-risk cases.41 There is a need for additional sur- veillance of pregnant women to identify at-risk pregnancies to respond to them appropriately. A health education pro- gramme focused on addressing indirect obstetric death risk factors, particularly for women with pre-existing conditions, is needed as well.46 Access to and use of effective contraception and safe abortion remains a key strategy to reduce maternal mortality worldwide.47 The Mexican Ministry of Health, recognizing the need for quality and state-level maternal mortality estimates, used its authority to ensure cooperation from states for the BIRMM project. While integration of the BIRMM as part of the vital statistics system treats the maternal mortality review as a core public health function,48 there is no separate specific budget for it from the ministry of health. The project is being implemented as a non-routine activity, subsidized by committed individuals and interested groups of public health practitioners.49 Annual results from the BIRMM are used to adjust official estimates of MMR. The adoption of electronic death registration would allow real-time reporting and validation of suspected maternal deaths. This study has several limitations. First, the review process did not target all deaths in women of reproductive age, but a subset identified based on the registered cause of death. While the codes used to identify cases capture most misclassified or miscoded mater- nal deaths, there may be some missed and not investigated.23,49 Second, the review process relies on the availability and quality of additional mortality data sources beyond the death certificate. For some deaths, the additional available information to make a reclassification decision was quite limited. Third, the role of improved ascertainment and catego- rization must not be overlooked when examining the trends in this analysis. Since the same search procedures have been used each year, it seems unlikely that the observed increase in indirect deaths would be solely due to improved ascertainment. Fourth, a few causes of death, considered to be indirect obstetric causes such as ICD-10 code O26.6 (liver disorders), are subsumed within chapters of the ICD broadly considered for direct maternal deaths. Full ICD-10 codes to the 4-digit level were not available for all deaths so the broader 3-digit categories were used and these cases were consid- ered to be of questionable quality. This means that some rare causes of indirect deaths may have been grouped with direct causes, and if so, the analysis may have underestimated the contribution of indirect deaths. Fifth, missing data in the covariates is another limitation. Sixth, when examining socioeconomic dispari- ties, the use of the municipality-level de- velopment index may hide within group differences. It is likely that the individual women dying of maternal deaths across all quintiles of the development index are the poorest women within those municipalities. The area-level nature of this component of the analysis does not allow for commentary about how an individual’s access to resources affects her risk of maternal death. Conclusion This study presents a useful strategy towards achieving a relatively complete and accurate assessment of the causes of maternal mortality in a country with complete vital registration. It provides useful lessons for other countries look- ing to improve maternal mortality mea- surement and highlights the importance of developing an appropriate health system response to address indirect maternal deaths. ■ Acknowledgements The authors thank Jaime Sepulveda, Julie Rajaratnam, Steven Goodreau and the Mexican Ministry of Health. Competing interests: None declared. Fig. 4. Indirect maternal deaths for the wealthiest and poorest quintile, Mexico, 2006–2013 95% CI Year Q1-poorest Q5-wealthiest No . o f m at er na l d ea th s p er 1 00 0 00 li ve b irt hs 35 30 25 20 15 10 5 0 2006 2007 2008 2009 2010 2011 2012 2013 CI: confidence interval; Q: quintile. Note: Number of indirect deaths per 100 000 live births, by year, for the wealthiest and poorest quintile of the municipality-level development index. Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360 367 Research Indirect and direct obstetric deaths in MexicoMargaret C Hogan et al. )reclasificación de muertes maternas )BIRMM .2013 ىتح 2006 نم ةترفلل تلاالحا فصول تلاالحا روهظ راركت تلادعم انمدختساو تلااح فينصت ةداـعلإو ،اهروهظ بابسأ راشتناو ،ةديدلجا اهنمضتي يتلا تاءارجلإاب ةناعتسلااب كلذو ،تاهملأا ينب ةافولا .BIRMM في قورفلا ليلحتل ةيئاصحلإا تارابتخلاا انمدختساو بابسأ نع ةتجانلا ةافولا تلااح ينب ةيعماتجلاا ةيناكسلا صئاصلخا تلادعم في قورفلاو ،ةشرابم يرغ بابسأ نع ةتجانلا كلتو ةشرابم بابسأ نع ةتجانلا كلتو ةشرابم بابسأ نع ةتجانلا ةافولا تلااح .ميلقلإا في رقفلا ىوتسمو ماعلل اًقفو كلذو ،ماع هجوب ةشرابم يرغ ةلاح 9043 لياجمإب تاهملأا ينب ةافولا تلااح تعضخ جئاتنلا 7829 نم( % 13 ةبسنب ةداـيز ترهظ .ةعجارلما تاءارــ جلإ هجوب تاهملأا ينب ةافولا تلااح نم هفاشتكا مت مايف )9043 لىإ ينب ةافولا لدعم في فاعضأ ةثلاث رادقمب ةدايز ترهظ ماك ،ماع تاترف في تاهملأا ينب تعقو ةافولل تلااحك ةفنصلماو تاهملأا تعجارت ةساردلا ةترف ىدم لىعو .)% 6.9 لىإ % 2.1 نم( ةرخأتم لم مانيب ،ةشرابم بابسأ نع ةتجانلاو ةدلاولاب ةطبترلما تايفولا دادعأ قلعتت ةشرابم يرغ بابسأ نع ةتجانلا تايفولا دادعأ في يريغت يأ رهظي نمم ءاسنلا في ةشرابم بابسأ نع ةتجانلا تايفولا تزكرتو .ةدلاولاب بابسلأ تايفوتلما ءاسنلا نأ ينبت دقو .اًرقف دشلأا ميلاقلأا في نشعي دق – ةشرابم يرغ بابسلأ تايفوتلما كئلوأب ًةنراقم – ةشرابم يرغ ردق لىع نلصحو ،اًنس رغصأ نكو ،اًددع لقأ لحم براتج نضخ عافتراب زيمتت ميلاقأ في ةشيعلما لماتحا نيهدل داز ماك ،ميلعتلا نم بركأ .يدالما ىوتسلما حيحصتل ةبسانلما بيلاسلأا دحأ BIRMM لثمي جاتنتسلاا رقتفت يتلا تائيبلا في تاهملأا ينب ةافولا نع ةيئاصحلإا تانايبلا لىع ينمئاقلل ةدعاسلما بولسلأا كلذ رفوي دقو .ةيفاكلا دراولما لىإ ضفلخ ميهدل ةعبتلما ةيجيتاترسلاا في رظنلا ةداعلإ يحصلا ماظنلا قلعتت ةشرابم يرغ بابسأ نع ةتجانلاو تاهملأا ينب ةافولا لدعم اهيف بوغرلما يرغ لملحا تلااح نم ةياقولا كلذ في ماب ،ةدلاولاب .ةدلاولا دعب مايفو لملحا ةترف في ةم َّدقلما ةياعرلا ينستحو 摘要 墨西哥境内孕产妇死亡的原因重新分类 ： 重复性横断面研究 目的 旨在说明八年多内墨西哥境内孕产妇死亡的原 因，重点关注间接孕产妇死亡人数和社会经济方面的 差异。 方 法 我 们 采 用 2006 至 2013 年 的 Búsqueda in tenc ionada y r ec l a s i f i c ac ión de muer te s maternas (BIRMM) 数据集开展了一项重复横断面研 究， 并且利用频率分析，通过 BIRMM 方法说明新增 案例、原因分布以及孕产妇死亡案例的重新分类情况。 我们采用统计测试，按照年份和地区贫困等级分析了 直接与间接死亡人数的社会人口特性差异，以及总体 直接死亡人数与间接死亡人数的比例差异。 结 果 研 究 过 程 中， 我 们 共 查 阅 了 9043 个 孕 产 妇 死亡案例。 确认的产妇死亡人数总体增长了 13% （从 7829 例增至 9043 例），归类为孕产妇后期死亡 人数的孕产妇死亡人数的比例增至三倍（从 2.1% 增 至 6.9%）。 研究期间，直接孕产妇死亡人数有所降低， 但是间接孕产妇死亡人数保持不变。 直接死亡人数集 中于居住在最贫困地区的女性。 与死于直接原因的女 性相比，死于间接原因的女性生育次数更少、年龄更 为年轻、受教育程度更高、并且更有可能居住在更富 裕的地区。。 结论 BIRMM 是一种在资源贫乏的情况下更正孕产妇 死亡统计数据的方法。 该方法可以帮助卫生系统重新 思考其降低间接孕产妇原因造成的孕产妇死亡人数的 策略。这些策略包括预防意外怀孕以及改善产前和产 后护理。 Résumé Reclassification des causes obstétricales de décès au Mexique: une étude transversale répétée Objectif Décrire les causes de mortalité maternelle au Mexique sur une période de huit ans, en s’intéressant particulièrement aux décès dus à des causes obstétricales indirectes ainsi qu’aux disparités socioéconomiques. Méthodes Nous avons réalisé une étude transversale répétée à l’aide des données de l’outil Búsqueda intencionada y reclasificación de muertes maternas (BIRMM) 2006–2013. Nous nous sommes basés sur les fréquences pour décrire les nouveaux cas, la répartition des causes et la reclassification des cas de mortalité maternelle selon la procédure BIRMM. Nous avons utilisé des tests statistiques pour analyser les différences, sur le plan des caractéristiques sociodémographiques, entre les décès pour cause directe et indirecte, ainsi que les différences dans la proportion globale de décès pour cause directe et indirecte, par année et par niveau de pauvreté des municipalités. Résultats Un total de 9043 décès maternels a été examiné. Nous avons constaté une augmentation de 13% du nombre global de décès maternels identifiés (de 7829 à 9043) et la proportion de décès maternels classés comme décès maternels tardifs a été multipliée par trois (passant de 2,1% à 6,9%). Sur la période étudiée, le nombre de décès dus à des causes obstétricales directes a diminué, mais il n’a été observé aucun changement dans le nombre de décès dus à des causes obstétricales indirectes. Les décès pour cause directe étaient concentrés chez les femmes vivant dans les municipalités les plus pauvres. En comparaison des femmes décédées de causes directes, les femmes décédées de causes indirectes avaient eu moins de grossesses et étaient légèrement plus jeunes, plus instruites et plus susceptibles de vivre dans des municipalités plus riches. Conclusion La procédure BIRMM est une approche qui permet de rectifier les statistiques sur les décès maternels dans les zones ayant peu de ressources. Cette approche pourrait permettre de repenser la stratégie du système de santé en vue de réduire le nombre de décès maternels dus à des causes obstétricales indirectes, avec notamment la prévention des grossesses non désirées et l’amélioration des soins prénataux et post-obstétricaux. Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360368 Research Indirect and direct obstetric deaths in Mexico Margaret C Hogan et al. Резюме Повторная классификация причин акушерских смертей в Мексике: повторное одномоментное поперечное исследование Цель Описать причины материнской смертности в Мексике за восемь лет, уделяя особое внимание акушерским смертям по непрямым причинам, а также из-за социально-экономического неравенства. Методы Было проведено повторное одномоментное поперечное исследование с использованием набора данных Búsqueda intencionada y reclasificación de muertes maternas (BIRMM) 2006– 2013 гг. Для описания новых случаев, распределения причин и повторной классификации случаев материнской смертности с помощью процедуры BIRMM использовались данные о частотности. С помощью статистических критериев были проанализированы различия в социально-демографических характеристиках рожениц, умерших по прямым и непрямым причинам, и различия в доле, которую эти смерти составляли в общем количестве смертей по прямым и непрямым причинам, в зависимости от года и уровня бедности муниципального образования. Результаты В ходе исследования было изучено 9043 случая смертей среди матерей. Общее количество определенных материнских смертей увеличилось на 13% (с 7829 до 9043), а доля материнских смертей, классифицированных как поздняя материнская смерть, увеличилась в три раза (с 2,1 до 6,9%). На протяжении исследования количество акушерских смертей по прямым причинам снизилось, однако количество смертей по непрямым акушерским причинам не изменилось. Смерть по прямым причинам в большинстве случаев наступала среди женщин, проживавших в беднейших муниципальных образованиях. По сравнению с роженицами, умершими по прямым причинам, женщины, чья смерть наступила по непрямым причинам, имели меньше беременностей и были несколько младше, лучше образованны и с большей вероятностью проживали в более зажиточных поселениях. Вывод Процедура BIRMM представляет собой один из подходов к получению правильной статистики по материнской смертности в условиях ограниченности ресурсов. С помощью этого подхода специалисты системы здравоохранения могли бы пересмотреть свою стратегию по сокращению количества материнских смертей по непрямым акушерским причинам, включающую предупреждение нежелательной беременности и повышение качества дородовой и послеродовой помощи. Resumen Reclasificación de las causas de muerte obstétrica en México: un estudio transversal repetido Objetivo Describir las causas de la mortalidad materna en México durante ocho años, con especial atención a las muertes obstétricas indirectas y a las desigualdades socioeconómicas. Métodos Se realizó un estudio transversal repetido utilizando el conjunto de datos Búsqueda intencionada y reclasificación de muertes maternas (BIRMM) de 2006-2013. Mediante el proceso de la BIRMM, se utilizaron frecuencias para describir nuevos casos, las distribuciones de las causas y la reclasificación de los casos de mortalidad materna. Se utilizaron pruebas estadísticas para analizar las diferencias en cuanto a las características sociodemográficas entre las muertes directas e indirectas y las diferencias en el porcentaje del total de muertes directas e indirectas, por año y por nivel de pobreza de los municipios. Resultados Un total de 9 043 muertes maternas fueron sujetas al proceso de revisión. En general, el número de muertes maternas identificadas aumentó un 13% (de 7 829 a 9 043) y el porcentaje de muertes maternas clasificadas como muertes maternas tardías se triplicó (de un 2,1% a un 6,9%). Durante el periodo de estudio, las muertes obstétricas directas se redujeron, mientras que las muertes derivadas de causas obstétricas indirectas se mantuvieron sin cambios. Las muertes directas se concentraron en mujeres que vivían en los municipios más pobres. En comparación con aquellas que fallecían por causas directas, las mujeres que fallecían por causas indirectas tuvieron menos embarazos y eran ligeramente más jóvenes, tenían una mejor educación y más posibilidades de vivir en municipios más adinerados. Conclusión La BIRMM es un enfoque cuyo objetivo es corregir las estadísticas de mortalidad materna en entornos con pocos recursos. El enfoque podría ayudar al sistema sanitario a reconsiderar su estrategia para reducir las muertes maternas por causas obstétricas indirectas, incluyendo la prevención de embarazos no deseados y la mejora del cuidado prenatal y postobstétrico. References 1. Campbell OM. Measuring progress in safe motherhood programmes: uses and limitations of health outcome indicators. In: Marge B, Ravindran TS, editors. Safe motherhood initiatives: critical issues. Oxford: Blackwell Science; 1999. pp. 31–42. 2. AbouZahr C, Wardlaw T. Maternal mortality at the end of a decade: signs of progress? Bull World Health Organ. 2001;79(6):561–8.http://www. ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&li st_uids=11436479&dopt=Abstract PMID: 11436479 3. Ronsmans C. How can we monitor progress towards improved maternal health. Stud Health Services Org Policy. 2001;17:313–38. 4. Yazbeck AS. Challenges in measuring maternal mortality. Lancet. 2007 Oct 13;370(9595):1291–2 doi: http://dx.doi.org/10.1016/S0140-6736(07)61553-0 PMID: 17933633 5. Graham WJ, Foster LB, Davidson L, Hauke E, Campbell OM. Measuring progress in reducing maternal mortality. Best Pract Res Clin Obstet Gynaecol. 2008 Jun;22(3):425–45. doi: http://dx.doi.org/10.1016/j.bpobgyn.2007.12.001 PMID: 18308640 6. Ronsmans C, Graham WJ; Lancet Maternal Survival Series steering group. Maternal mortality: who, when, where, and why. Lancet. 2006 Sep 30;368(9542):1189–200. doi: http://dx.doi.org/10.1016/S0140-6736(06)69380-X PMID: 17011946 7. Indicators for monitoring the Millennium Development Goals. Definitions, rationale, concepts and sources. New York: United Nations Development Group; 2003. 8. International statistical classification of diseases and related health problems, 10th revision, edition 2010. Geneva: World Health Organization; 2011. 9. Graham WJ, Ahmed S, Stanton C, Abou-Zahr C, Campbell OM. Measuring maternal mortality: an overview of opportunities and options for developing countries. BMC Med. 2008;6(1):12. doi: http://dx.doi.org/10.1186/1741-7015-6-12 PMID: 18503716 10. Tinker AG, Koblinsky MA. Making motherhood safe. [World Bank discussion papers]. Washington: World Bank; 1993. doi: http://dx.doi.org/10.1596/0-8213- 2468-3 Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360 369 Research Indirect and direct obstetric deaths in MexicoMargaret C Hogan et al. 11. Nour NM. An introduction to maternal mortality. Rev Obstet Gynecol. 2008 Spring;1(2):77–81.http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve &db=PubMed&list_uids=18769668&dopt=Abstract PMID: 18769668 12. Chowdhury ME, Ahmed A, Kalim N, Koblinsky M. Causes of maternal mortality decline in Matlab, Bangladesh. J Health Popul Nutr. 2009 Apr;27(2):108–23. doi: http://dx.doi.org/10.3329/jhpn.v27i2.3325 PMID: 19489410 13. AbouZahr C. Global burden of maternal death and disability. Br Med Bull. 2003;67(1):1–11. doi: http://dx.doi.org/10.1093/bmb/ldg015 PMID: 14711750 14. Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PFA. WHO analysis of causes of maternal death: a systematic review. Lancet. 2006 Apr 1;367(9516):1066–74. doi: http://dx.doi.org/10.1016/S0140-6736(06)68397-9 PMID: 16581405 15. Hogan MC, Foreman KJ, Naghavi M, Ahn SY, Wang M, Makela SM, et al. Maternal mortality for 181 countries, 1980–2008: a systematic analysis of progress towards Millennium Development Goal 5. Lancet. 2010 May 8;375(9726):1609–23. doi: http://dx.doi.org/10.1016/S0140-6736(10)60518-1 PMID: 20382417 16. Berg CJ, Callaghan WM, Henderson Z, Syverson C. Pregnancy-Related Mortality in the United States, 1998 to 2005. Obstet Gynecol. 2011 May;117(5):1230. doi: http://dx.doi.org/10.1097/AOG.0b013e31821769ed PMID: 21705929 17. Cantwell R, Clutton-Brock T, Cooper G, Dawson A, Drife J, Garrod D, et al. Saving Mothers’ Lives: Reviewing maternal deaths to make motherhood safer: 2006–2008. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG. 2011 Mar;118 Suppl 1:1–203.http:// www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&li st_uids=21356004&dopt=Abstract PMID: 21356004 18. Bouvier-Colle MH, Mohangoo AD, Gissler M, Novak-Antolic Z, Vutuc C, Szamotulska K, et al.; Euro-Peristat Scientific Committee. What about the mothers? An analysis of maternal mortality and morbidity in perinatal health surveillance systems in Europe. BJOG. 2012 Jun;119(7):880–9, discussion 890. doi: http:// dx.doi.org/10.1111/j.1471-0528.2012.03330.x PMID: 22571748 19. Halim A, Utz B, Biswas A, Rahman F, van den Broek N. Cause of and contributing factors to maternal deaths; a cross-sectional study using verbal autopsy in four districts in Bangladesh. BJOG. 2014 Sep;121 Suppl 4:86–94. doi: http://dx.doi. org/10.1111/1471-0528.13010 PMID: 25236640 20. Hodorogea S, Friptu V. The Moldovan experience of maternal death reviews. BJOG. 2014 Sep;121 Suppl 4:81–5. doi: http://dx.doi.org/10.1111/1471-0528.12945 PMID: 25236639 21. Kassebaum NJ, Bertozzi-Villa A, Coggeshall MS, Shackelford KA, Steiner C, Heuton KR, et al. Global, regional, and national levels and causes of maternal mortality during 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014 Sep 13;384(9947):980–1004. doi: http://dx.doi.org/10.1016/ S0140-6736(14)60696-6 PMID: 24797575 22. 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 Jun;89(6):762–8. doi: http://dx.doi. org/10.3109/00016341003657876 PMID: 20450247 23. Freyermuth-Enciso GC-ER, Cárdenas-Elizalde R. Evaluación del subregistro de la mortalidad materna en Los Altos de Chiapas mediante las estrategias RAMOS y RAMOS modificada. Salud Publica Mex. 2009;5(1):450–7. Spanish. doi: http:// dx.doi.org/10.1590/S0036-36342009000600002 24. Lozano R. La carga de la enfermedad y las desigualdades en salud de las mujeres en Mexico. Genero y Salud en cifras. 2012;10(1). Spanish. 25. Secretaría de Salud. Búsqueda intencionada y reclasificación de muertes maternas en México. Informe 2010. Mexico City: Dirección General de Información en Salud Secretaría de Salud México; 2012. Spanish. 26. Torres LM, Rhenals AL, Jiménez A, Ramírez-Villalobos D, Urióstegui R, Piña M, et al. [Intentional search and reclassification of maternal deaths in Mexico: The effect on the distribution of causes]. Salud Publica Mex. 2014 Jul-Aug;56(4):333–47. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&li st_uids=25604173&dopt=Abstract Spanish PMID: 25604173 27. Uauy R, Albala C, Kain J. Obesity trends in Latin America: transiting from under- to overweight. J Nutr. 2001 Mar;131(3):893S–9S.http://www. ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&li st_uids=11238781&dopt=Abstract PMID: 11238781 28. Barquera S, Campos-Nonato I, Hernández-Barrera L, Pedroza A, Rivera-Dommarco JA. [Prevalence of obesity in Mexican adults 2000–2012]. Salud Publica Mex. 2013;55 Suppl 2:S151–60. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cm d=Retrieve&db=PubMed&list_uids=24626691&dopt=Abstract Spanish. PMID: 24626691 29. Stevens G, Dias RH, Thomas KJ, Rivera JA, Carvalho N, Barquera S, et al. Characterizing the epidemiological transition in Mexico: national and subnational burden of diseases, injuries, and risk factors. PLoS Med. 2008 Jun 17;5(6):e125. doi: http://dx.doi.org/10.1371/journal.pmed.0050125 PMID: 18563960 30. Global Burden of Disease Study 2013 Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015 Aug 22;386(9995):743–800. doi: http://dx.doi.org/10.1016/S0140-6736(15)60692-4 PMID: 26063472 31. Anexo C: Metodología de estimación del índice de marginación por localidad. Mexico City: Consejo Nacional de Poblacion; 2011. Spanish. 32. Births Mexico. Mexico City: General Directorate of Health Information, Ministry of Health; 2014. 33. Mortality Mexico. Mexico City: National Institute of Statistics and Geography; 2015. 34. Mosby LG, Rasmussen SA, Jamieson DJ. 2009 pandemic influenza A (H1N1) in pregnancy: a systematic review of the literature. Am J Obstet Gynecol. 2011 Jul;205(1):10–8. doi: http://dx.doi.org/10.1016/j.ajog.2010.12.033 PMID: 21345415 35. Mortalidad materna en Mexico durante 2009. El efecto de las infecciones respiratorias agudas (neumonía e influenza). Mexico City: Secretaría de Salud; 2009. Spanish. 36. Chaves SdaC, Cecatti JG, Carroli G, Lumbiganon P, Hogue CJ, Mori R, et al. Obstetric transition in the World Health Organization Multicountry Survey on Maternal and Newborn Health: exploring pathways for maternal mortality reduction. Rev Panam Salud Publica. 2015 May;37(4-5):203–10. PMID: 26208186 37. Alves SV. Maternal mortality in Pernambuco, Brazil: what has changed in ten years? Reprod Health Matters. 2007 Nov;15(30):134–44. doi: http://dx.doi. org/10.1016/S0968-8080(07)30326-1 PMID: 17938078 38. Victora CG, Aquino EML, do Carmo Leal M, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. Lancet. 2011 May 28;377(9780):1863–76. doi: http://dx.doi.org/10.1016/S0140-6736(11)60138-4 PMID: 21561656 39. Grubb GS, Fortney JA, Saleh S, Gadalla S, el-Baz A, Feldblum P, et al. A comparison of two cause-of-death classification systems for deaths among women of reproductive age in Menoufia, Egypt. Int J Epidemiol. 1988 Jun;17(2):385–91. doi: http://dx.doi.org/10.1093/ije/17.2.385 PMID: 3403135 40. Hill K, Thomas K, AbouZahr C, Walker N, Say L, Inoue M, et al.; Maternal Mortality Working Group. Estimates of maternal mortality worldwide between 1990 and 2005: an assessment of available data. Lancet. 2007 Oct 13;370(9595):1311–9. doi: http://dx.doi.org/10.1016/S0140-6736(07)61572-4 PMID: 17933645 41. Cross S, Bell JS, Graham WJ. What you count is what you target: the implications of maternal death classification for tracking progress towards reducing maternal mortality in developing countries. Bull World Health Organ. 2010 Feb;88(2):147– 53. doi: http://dx.doi.org/10.2471/BLT.09.063537 PMID: 20428372 42. Campbell O, Gipson R, Issa AH, Matta N, El Deeb B, El Mohandes A, et al. National maternal mortality ratio in Egypt halved between 1992–93 and 2000. Bull World Health Organ. 2005 Jun;83(6):462–71. PMID: 15976898 43. Souza JP, Gülmezoglu AM, Vogel J, Carroli G, Lumbiganon P, Qureshi Z, et al. Moving beyond essential interventions for reduction of maternal mortality (the WHO Multicountry Survey on Maternal and Newborn Health): a cross-sectional study. Lancet. 2013 May 18;381(9879):1747–55. doi: http://dx.doi.org/10.1016/ S0140-6736(13)60686-8 PMID: 23683641 44. Johns N, Lozano R, Wulf S. What’s killing mothers and what’s killing women? Contextualising maternal conditions in the post-MDG era using the Global Burden of Disease (GBD) Study 2010. Lancet. 2013;381 S68:1. doi: http://dx.doi. org/10.1016/S0140-6736(13)61322-7 PMID: 23290953 45. Langer A, Horton R, Chalamilla G. A manifesto for maternal health post-2015. Lancet. 2013 Feb 23;381(9867):601–2. doi: http://dx.doi.org/10.1016/S0140- 6736(13)60259-7 PMID: 23439088 46. McCaw-Binns A, Alexander SF, Lindo JL, Escoffery C, Spence K, Lewis-Bell K, et al. Epidemiologic transition in maternal mortality and morbidity: new challenges for Jamaica. Int J Gynaecol Obstet. 2007 Mar;96(3):226–32. doi: http://dx.doi. org/10.1016/j.ijgo.2006.12.002 PMID: 17306270 47. Ahmed S, Li Q, Liu L, Tsui AO. Maternal deaths averted by contraceptive use: an analysis of 172 countries. Lancet. 2012 Jul 14;380(9837):111–25. doi: http://dx.doi. org/10.1016/S0140-6736(12)60478-4 PMID: 22784531 48. Berg CJ. From identification and review to action–maternal mortality review in the United States. Semin Perinatol. 2012 Feb;36(1):7–13. doi: http://dx.doi. org/10.1053/j.semperi.2011.09.003 PMID: 22280859 49. Lozano R, Torres-Palacios LM, Soliz PN. [Comments on the article “Evaluation of maternal mortality under-reporting in the heights of Chiapas using the RAMOS and modified RAMOS strategies” by Graciela Freyermuth et al.]. Salud Publica Mex. 2010 Sep-Oct;52(5):381–3.Spanish. doi: http://dx.doi.org/10.1590/S0036- 36342010000500001 PMID: 21049668 Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360 369A Research Indirect and direct obstetric deaths in MexicoMargaret C Hogan et al. Table 3. Cause of death reclassified as indirect maternal deaths, Mexico, 2006–2013 ICD-10 block8 Title No. of recorded cases (%) (n = 699) Non-maternal A15–B49 Infectious and parasitic diseases 96 (13.7) C00–C14 Malignant neoplasms of lip, oral cavity and pharynx 1 (0.1) C15–C96, D10–D48 Other malignant, benign and uncertain or unknown behaviour neoplasms 56 (8.0) D00–D09 In situ neoplasms (carcinoma) 1 (0.1) D55–D89 Diseases of blood and blood-forming organs 8 (1.1) E00–E90 Endocrine, nutritional and metabolic diseases 32 (4.6) F50–F59 Behavioural syndromes associated with physiological disturbances and physical factors 1 (0.1) G00–G99 Diseases of the nervous system 21 (3.0) I00–I99 Diseases of the circulatory system 94 (13.4) J00–J99 Diseases of the respiratory system 70 (10.0) K00–K93 Diseases of the digestive system 37 (5.3) L00–L08 Infections of the skin and subcutaneous tissue 1 (0.1) M00–M99 Diseases of the musculoskeletal system and connective tissue 7 (1.0) N00–N99 Diseases of the genitourinary system 9 (1.3) Q00–Q99 Congenital malformations, deformations and chromosomal abnormalities 20 (2.9) R10–R19 Symptoms and signs involving the digestive system and abdomen 1 (0.1) R50–R69 General symptoms and signs 7 (1.0) R95–R99 Ill-defined and unknown causes of mortality 1 (0.1) V01–Y98 External causes 16 (2.3) Maternal O01 Hydatidiform mole 1(0.1) O02–O08 Other pregnancy with abortive outcome 9 (1.3) O10–O16 Oedema, proteinuria and hypertensive disorders in pregnancy, childbirth and the puerperium 40 (5.7) O20, O45–O46, O67 Premature separation of placenta and other haemorrhage of pregnancy or birth 12 (1.7) O21, O23–O31, O34 Other complications of pregnancy 46 (6.6) O22,O87 Other maternal disorders predominantly related to pregnancy 2 (0.3) O35–O43, O68–O69 Fetal distress and other complications of pregnancy or birth 16 (2.3) O47–O48, O60–O75 Prolonged pregnancy, other complications of labour 9 (1.3) O72 Postpartum haemorrhage 6 (0.9) O88 Obstetric embolism 12 (1.7) O89–O92 Other complications during the puerperium 46 (6.6) O96–O97 Late and sequelae maternal death 2 (0.3) A34, O85–O86 Obstetrical tetanus, complications predominantly related to the peurperium 19 (2.7) Bull World Health Organ 2016;94:362–369B| doi: http://dx.doi.org/10.2471/BLT.15.163360369B Research Indirect and direct obstetric deaths in Mexico Margaret C Hogan et al. Table 4. Characteristics of indirect and direct maternal deaths, Mexico, 2006–2013 Characteristics Cause of death Indirect Direct Number of pregnancies (mean, SD) 2.45 (0.04) 3.03 (0.03)*,a Age (mean, SD) 27.40 (0.14) 28.70 (0.09)*,a Number of prenatal visits (mean, SD) 3.26 (0.07) 2.86 (0.04)*,a Marital status, n 2208 6022 Single, n (proportion) 394 (0.18) 900 (0.15) Common law, divorced or widowed, n (proportion) 869 (0.39) 2415 (0.40) Married, no. (%) 945 (0.43) 2707 (0.45)*,b Education, n 2131 5807 Primary, no. (%) 849 (0.40) 2660 (0.46) Secondary, no. (%) 726 (0.34) 1851 (0.32) High school or more, no. (%) 557 (0.26) 1296 (0.22)*,b Place of death, n4 1732 4331 Secretaria de Salud, no. (%) 903 (0.52) 2143 (0.49) IMSS/ISSSTE/SEDENA, no. (%) 511 (0.30) 1010 (0.23) Private medical unit, no. (%) 92 (0.05) 410 (0.09) Home, street, other, no. (%) 226 (0.13) 768 (0.18)*,b Place of delivery, n 1657 5329 Secretaria de Salud, no. (%) 901 (0.54) 2603 (0.49) IMSS/ISSSTE/SEDENA, no. (%) 491 (0.30) 1048 (0.20) Private medical unit, no. (%) 150 (0.09) 833 (0.16) Home, street, other, no. (%) 115 (0.07) 845 (0.16)*,b Place of care for first complication, n 1875 5129 Secretaria de Salud, no. (%) 1044 (0.56) 2864 (0.56) IMSS/ISSSTE/SEDENA, no. (%) 533 (0.28) 1069 (0.21) Private medical unit, no. (%) 285 (0.15) 1104 (0.22) Home, street, other, no. (%) 13 (0.01) 92 (0.02)*,b Development index, (municipality), n 2237 6145 Quintile 1 – Poorest, no. (%) 202 (0.09) 1094 (0.18)*,b Quintile 2, no. (%) 194 (0.09) 666 (0.11) Quintile 3, no. (%) 197 (0.09) 604 (0.10) Quintile 4, no. (%) 317 (0.14) 773 (0.13) Quintile 5 – Wealthiest, no. (%) 1327 (0.599 3008 (0.49) Skilled birth attendant, n 1910 5477 Doctor, no. (%) 1361 (0.71) 4267 (0.78) Nurse/auxiliary/midwife, no. (%) 43 (0.02) 460 (0.08) Relative/other, n (proportion) 20 (0.01) 231 (0.04) No delivery, n (proportion) 486 (0.25) 519 (0.09) IMSS: Instituto Mexicano del Seguro Social, ISSSTE: Instituto de Seguridad y Servicios Sociales de los Trabajadores del Estado, SEDENA: Secretaría de la Defensa Nacional, SD: standard deviation. * P < 0.05; **P < 0.01. a Tested using differences in mean b Tested using an overall Pearson χ2.