Analysis of the causes of maternal death in China*

Buletin of the World Health Organizaion, 66 (3): 387-390 (1988) © World Health Organization 1988 Analysis of the causes of maternal death in China* ZHANG LINGMEI1 & DING HUI2 Data were analysed on maternal mortality for 1984 in 287 cities, districts, and counties in 21 provinces, municipalities, and autonomous regions of China. The total population covered was 177.55 million, and during the study period there were 2 483 269 live births and 1211 matemal deaths, corresponding to a maternal mortality rate of 48.8 per 1() 000. The main cause ofmaternal death was obstetric haemorrhage, followed, in order, by cardiac diseases, toxaemia ofpregnancy, hepatic disease, puerperal infection, and amnioticfluid embolism. The health care measures received by the mothers who died are analysed, and methods of reducing the maternal mortality rate are proposed. Maternal mortality rate is an important index for determining the socioeconomic and cultural status, general health care, as well as the mother and child health care level of a country or an area. In order to ascertain the maternal mortality rates in various areas of China and to analyse the factors that cause such deaths, the Maternal Death Investigation Cooperative Group of China was set up in May 1984. The group covers 21 provinces, municipalities, and autonomous regions in the six principal administrative districts of the country. The maternal mortality rate was investi- gated by the group using a unified protocol, and pro- posals were made on improving the health care of mothers and infants in China. MATERIALS AND METHODS Method ofinvestigation The investigation was carried out in areas that were typical of each province, municipality, and auton- omous region, and included urban and rural locations in both flat and mountainous parts ofthe country. The mother and child health workers in the various areas organized the details of the investigation and ensured that a uniform method was used to collect the data. All the data were sent to the Beijing Municipal Ma- ternal Health Institute for analysis and the results reported to the Matemal Death Investigation Cooper- ative Group. * Results of a study of 21 provinces, municipalities, and autonomous regions, carried out by the Maternal Death Investi- gation Cooperative Group of China. 1 Director, Beijing Municipal Maternal Health Institute, Beijing, China. Requests for reprints should be sent to this author. 2 Visiting Doctor, Beijing Municipal Maternal Health Institute, Beijing, China. Study subjects Maternal deaths among residents registered in the study areas were investigated using the following definition of maternal death stipulated by WHO and the International Federation of Gynaecology and Obstetrics (FIGO): -Death ofa woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and the site ofthe pregnancy, from any cause related to or aggravated by the pregnancy or its management, but not from accidental or incidental causes. Study areas The investigation was carried out in the following 21 provinces, municipalities or autonomous regions of China: -Beijing, Shanghai, Tianjin, Heilongjiang, Liao- ning, Hebei, Shanxi, Neimenggu, Sha'anxi, Gansu, Ningxia, Jiangsu, Zhejiang, Jiangxi, Fujian, Hubei, Hunan, Guangdong, Sichuan, Yunnan, and Henan. Altogether 287 cities, districts, and counties were selected as typical areas, covering a population of 177.55 million. A preliminary analysis of the data on matemal deaths is presented here. A total of 2 483 269 live births and 1211 maternal deaths occurred in 1984 in the study areas. RESULTS General Distribution ofresidence areas. A total of46.7% of the maternal deaths occurred among women who 4893 -387 388 ZHANG LINGMEI & DING HUI lived on the plains, while 43.4% were from moun- tainous or semi-mountainous areas and the remainder from other areas. Economic status. The average family income per person per month of 49.7% of the women who died was less than 20 yuan, that of 39.2% ranged from 20 to 50 yuan, while that of only 8.3% was greater than 50 yuan (1US$= 3.71 yuan). Educational level. A total of 29.2% of the women who died were illiterate, 35.4% had attended only primary school, and 34.4% junior middle school or above. Maternal mortality rate The maternal mortality rate per 100 000 live births was calculated using the relationship: Number of maternal deaths per annum X 10 Total number of live births per annum The rate in the areas investigated was 48.8 per 100 000 in 1984-the rate being 24.8 per 100 000 in urban areas and 58.7 per 100 000 in rural areas (Table 1). The mortality rates differed widely depending on geographical area. For example in Ningxia, Gansu, Henan, and Shanxi the rates were greater than 100 per 100 000; those in Sha'anxi, Yunnan, Hunan, Neimenggu, and Sichuan, 50 per 100 000 to 100 per 100 000; those in Beijing, Tianjin, Hebei, Liaoning, Heilohgjiang, Zhejiang, Hubei, Guangdong, Jiangxi, Jiangsu, and Fujian, 20 per 100 000 to 50 per 100 000; and those in Shanghai, less than 20 per 100 000. In contrast, from 1975 to 1985 the levels of maternal mortality (per 100 000 live births) in some other countries in east and south Asia were as follows:a Greater than 100: Burma, India, Indonesia, Pakistan, and Viet Nam. 50-100: Philippines, Sri Lanka, and Thailand. 20-50: Democratic People's Republic of Korea, Republic of Korea, and Malaysia. Less than 20: Australia, Hong Kong, Japan, New Zealand, and Singapore. Both the results of the present study and the above- mentioned maternal mortality rates in Asia indicate that such rates are closely related to economic and cultural factors as well as to health care facilities. a Maternal mortality rates, a tabulation of available infor- mation, 2nd edition. Unpublished document, WHO/FHE/86.3. Table 1. Maternal mortality rates in the areas studied in 21 provinces, municipalities, and autonomous regions of China, 1984 Matemal mortality rate Area (per 100 000) Beijing 30.7 Shanghai 17.7 Tianjin 38.9 Heilongjiang 31.9 Liaoning 34.8 Hebei 48.5 Shanxi 101.7 Neimenggu 76.1 Sha'anxi 69.5 Sichuan 77.1 Gansu 101.1 Ningxia 108.2 Jiangsu 30.5 Zhejiang 43.8 Henan 106.6 Hunan 89.4 Hubei 45.0 Jiangxi 46.2 Fujian 42.7 Guangdong 40.1 Yunnan 86.3 Total 48.8 Analysis of causes ofmaternal deaths A total of 60.2% of the causes of deaths were re- ported by hospitals and 39.8% by medical personnel at the primary health care level. Among the hospital diagnoses, only 4.3 % were based on autopsy. Obstetric haemorrhage was the principal cause of death and was responsible for 45.4% of the 1211 maternal deaths in the study. This was followed by cardiac disease (11.1 %), pregnancy-induced hypertension (10.6%), puerperal infection (6.5%), hepatic diseases (4.4%), and amniotic fluid embolism (4.2%). Together, these causes constituted 82.2% of maternal deaths (Table 2). Sequence ofcauses ofmaternal mortalities in urban and rural areas. In urban areas the first six causes of maternal deaths, in decreasing order of importance, were: pregnancy-induced hypertension, obstetric haemorrhage, cardiac disease, hepatic diseases, amniotic fluid embolism, and puerperal infection. In contrast, in rural areas the principal causes of such CAUSES OF MATERNAL DEATH IN CHINA Table 2. Cause and distribution of maternal deaths in the areas studied in 21 provinces, municipalities, and autonomous regions of China, 1984 Cause of death No. of deaths Obstetric haemorrhage 550 (45.4)' Retained placenta 212 (17.5) Uterine inertia 125 (10.3) Ruptured uterus 68 (5.6) Placenta praevia 41 (3.4) Injury of birth canal 33 (2.7) Inversion of uterus 23 (1.9) Abruptio placentae 21 (1.7) Blood coagulation dysfunction 13 (1.1) Other haemorrhages 14 (1.2) Cardiac disease 134 (11.1) Pregnancy induced hypertension 128 (10.6) Puerperal infection 79 (6.5) Hepatic diseases 54 (4.5) Amniotic fluid embolism 51 (4.2) Ectopic pregnancy 23 (1.9) Other obstetric causes 22 (1.8) Infectious diseases 34 (2.8) Blood diseases 21 (1.7) Gastro-intestinal tract diseases 22 (1.8) Respiratory diseases 18 (1.5) Circulatory diseases 12 (1.0) Urological diseases 8 (0.7) Endocrine and metabolic diseases 7 (0.6) Cancer 12 (1.0) Miscellaneous 11 (0.9) Unknown 25 (2.1) Total 1211 (100.0) ' Figures in parentheses are percentaes. deaths followed the order: obstetric haemorrhage, heart disease, pregnancy-induced hypertension, puerperal infection, amniotic fluid embolism, and hepatic diseases. Obstetric haemorrhage accounted for two-thirds of the deaths from the six main causes in rural areas, but in urban areas pregnancy-induced hypertension and obstetric haemorrhage together were responsible for half the deaths (Fig. 1). Comparison ofdeaths due to direct and indirect ob- stetric causes. Direct obstetric causes accounted for 853 and indirect obstetric causes for 358 of the 1211 maternal deaths (ratio of direct to indirect deaths, 2.4 to 1). Health care received by the women who died Antenatal examination. Of the 1211 women who died, 764 (63%) had undergone at least one antenatal examination, and of these 37% had had one to two, 33% had had three to four, and 28.4% had had more than five examinations. The mean number of exami- nations was 3.5 ± 1.9. Place ofdelivery. Of the 1211 deaths recorded, the place of delivery was known in 978 instances. More than half (50.5%) of these deliveries were at home, while 46.9% were in regional hospitals. Place ofdeath. In the urban areas, a total of 82.0% of maternal deaths occurred in hospital. In contrast, in rural areas, a large proportion of the deaths were at home (31.8%) or on the way to hospital (14.8%). Attendance at delivery. Of women who died from one of the six principal causes of death, 27 were attended during birth by traditional birth attendants or gave birth without assistance from any health worker. DISCUSSION Maternal mortality rate is an important indicator for evaluating both socioeconomic and cultural con- ditions, the level of mother and child health care, as well as providing information for use in making pol- icy decisions on establishing health services. The results of the present study indicate that in China the focus for improvements in mother and child health care should be on rural areas, remote peripheral dis- tricts, and parts of the country where there is a large proportion of national minorities. U ~40 3. Ci M URBAN 3RURAL I io/ /6 Fig. 1. Distribution of the six principal causes of maternal deaths in 21 provinces, municipalities, and autonomous regions of China, 1984 389 ZHANG LINGMEI & DING HUI The results of the analysis of the causes of maternal deaths and their circumstances can be used to estab- lish priorities for the provision of mother and child health care and clinical services. Obstetric haemor- rhage was the main cause of maternal death in both urban and rural areas, the proportion of such deaths being considerably higher in rural areas. Also, in these areas it is significant that 31.8% of maternal deaths occurred at home and that 14.8% happened while on the way to hospital. In rural areas of China emphasis should therefore be placed on improving the coverage and service provided by trained health workers at deliveries, on increasing the proportion of deliveries in hospitals, and on providing women who experience obstetric emergencies at home with the life-saving care they require. In contrast, emphasis in the cities should be placed on improving the quality of obstetric services. In many instances, the survey indicated that the health care provided to mothers was far from ad- equate, particularly in the rural and more remote areas. Efforts must therefore be made to enlarge and improve the mother and child health care network, increase the number ofpersonnel engaged in maternal and child health and obstetric care, and improve the quality of the service provided. RESUME ANALYSE DES CAUSES DE MORTALITt MATERNELLE EN CHINE Le taux de mortalite matemelle est un indicateur impor- tant de la situation socio-economique et culturelle, ainsi que du niveau g6n6ral des soins de sante dans un pays ou une' region, et plus particulierement de la qualite des soins de sante maternelle et infantile. Le groupe cooperatif d'enquete sur la mortalite maternelle en Chine, cree en mai 1984 pour analyser la situation dans ce domaine, couvre 21 provinces, municipalites et regions autonomes des six principaux districts administratifs du pays. L'etude des causes de mortalite maternelle qui fait l'objet du present article a ete men&e par le groupe a la fois dans des regions urbaines et dans des regions rurales du pays et a port6 sur une population globale de 177,5 millions d'habitants. La definition de la mort maternelle utilisee aux fins de l'etude est celle qui a ete stipulee par 1'OMS et la Fed6ration inter- nationale de Gynecologie et d'Obstetrique. En 1984 il y a eu au total dans ces regions 2 483 269 naissances vivantes et 1211 morts maternelles, ce qui correspond a un taux de mortalite de 48,8 pour 100 000. L'hemorragie obstetricale a 6te la principale cause de mort maternelle, tant en milieu urbain qu'en milieu rural, mais son importance relative a ete beaucoup plus grande en milieu rural. D'autre part, dans les regions rurales, 32% des deces se sont produits I domicile et 15% pendant le transport a l'h6pital. Dans ces regions, les efforts devraient viser a ameliorer la couverture et la qualite des services fournis lors de l'accouchement par des agents de sante formes a cette tache, I accroitre la proportion des accouche- ments en milieu hospitalier et I fournir aux femmes qui sont victimes de complications obstetricales a domicile les soins vitaux dont elles ont besoin. Dans les villes, l'accent devrait etre mis sur l'amelioration de la qualite des services obstetricaux. 390