Social Inequalities in Perinatal Mortality in Belo Horizonte, Brazil: The ...

7 downloads 1248 Views 109KB Size Report
We examined the contribution of hospital type and quality of care to perinatal mortality rates in .... analyses were conducted with the software programs Epi Info 6.0 ..... Lawn J, Shibuya K, Stein C. No cry at birth: global estimates of intrapartum ...
 RESEARCH AND PRACTICE 

Social Inequalities in Perinatal Mortality in Belo Horizonte, Brazil: The Role of Hospital Care | Sônia Lansky, MD, PhD, Elisabeth França, MD, PhD, and Ichiro Kawachi, MD, PhD

Brazil has a persistently high infant mortality rate (22.5 deaths/1000 live births in 2003)1 that disproportionately affects the disadvantaged population. Most of the country’s infant and perinatal deaths are because of conditions originating during the perinatal period that are considered preventable through access to quality health care. Although there are important regional disparities, most births (97%) take place in hospitals, and 77% are assisted by doctors. A high proportion of perinatal and infant deaths occur within the first hours after birth (30% in the first 24 hours of life), which suggests the importance of the level of hospital care. Brazil’s Universal Public Health System (Sistema Único de Saúde, or SUS), which covers the medical expenses of almost 80% of the country’s population, relies on private hospitals contracted to SUS (37%), as well as hospitals run by the philanthropic sector (27%) and the government (36%).2 Private hospitals not contracted to SUS (non-SUS) provide care for the remaining minority who can afford private health insurance or direct payment. Consequently, there is a clear association between socioeconomic status and the type of health facility used. Hospital type can therefore be a marker for socioeconomic status,3,4 and it can also be an indicator of health care quality.5 Socioeconomic disparities in the quality of hospital care may in turn explain perinatal mortality differentials. Few studies have examined socioeconomic inequalities in perinatal mortality in Brazil, however, and quality of hospital care has not yet been systematically assessed. We analyzed the role of hospital quality at the time of delivery and birth and its contribution to the high perinatal mortality rates in the city of Belo Horizonte. Situated in the more developed southeast region of Brazil, Belo Horizonte is the country’s fourth largest city, with 2.2 million inhabitants. We focused on the differential in perinatal mortality rate between hospital categories (SUS vs non-SUS

Objectives. We examined the contribution of hospital type and quality of care to perinatal mortality rates in the city of Belo Horizonte, Brazil. Methods. We used a cohort study of all births (40953) and perinatal deaths (826) in Belo Horizonte in1999. After adjusting for maternal education and birthweight, we compared mortality rates according to hospital category—defined by a hospital’s relation to the national Universal Public Health System (SUS)—and quality of care. We used the Wigglesworth Classification to examine perinatal deaths. Results. After we controlled for birthweight and maternal education, the highest perinatal death rates were observed in private and philanthropic SUScontracted hospitals (relative to private, non-SUS-contracted hospitals). Hospital quality was also directly associated with perinatal death rates. Mortality rates were especially high for normal-birthweight babies born in private SUS-contracted hospitals. Intrapartum asphyxia was the leading cause of preventable death. Conclusions. In a class-segregated health care system, such as Brazil’s, disparities in quality of care between SUS-contracted and non-SUS-contracted hospitals contribute to the unacceptably high rates of perinatal mortality. (Am J Public Health. 2007;97:867–873. doi:10.2105/AJPH.2005.075986)

hospitals) and quality of hospital care. Our ultimate goal was to provide public health policymakers with information that can guide the planning and implementation of measures to improve the health care system and reduce disparities in infant and perinatal mortality.

METHODS This study is based on a 1999 cohort study involving surveillance of all births (n = 40 953) and perinatal deaths (n = 826) in the city of Belo Horizonte.6 Perinatal deaths comprise fetal deaths, defined as all stillbirths with birthweights of 500 g or more or gestation age of 22 weeks or more, and early neonatal deaths, defined as all infant deaths up to 7 days of life in which the infant weighed 500 g or more at birth or had a gestational age of 22 weeks or more.7 For our analysis, data were collected by hospital chart review and linkage of individual records to the National Live Birth Information System and the National Death Information System, yielding 775 perinatal deaths in 27 hospitals. Information gathered by 1 of the authors (S. L.) and by trained medical students included

May 2007, Vol 97, No. 5 | American Journal of Public Health

maternal education (from birth and death certificates), birthweight (from chart review when available or from birth or death certificates), cause of death (from charts and review of birth and death certificates; deaths were categorized by the Wigglesworth Classification),8 and hospital category (from death certificates and chart review for perinatal deaths and from birth certificates). Each hospital was categorized according to its relation to the SUS system and by its quality of care. There were 20 hospitals contracted to SUS (hereafter called SUS hospitals; 12 private SUS hospitals, 4 philanthropic SUS hospitals, and 4 public SUS hospitals) and 7 private non-SUS hospitals. Quality assessment was conducted only in Belo Horizonte hospitals (n = 24); each hospital received a standardized score of 0 to 2000 (assigned by Costa et al.9), which related to its structural ability to assist the mother and the baby.9 Ten hospitals were scored 1000 or lower, indicating that they lacked the conditions for such basic health care as neonatal resuscitation (low quality); 7 hospitals were scored between 1001 and 1500 (intermediate quality), while 7 were

Lansky et al. | Peer Reviewed | Research and Practice | 867

 RESEARCH AND PRACTICE 

TABLE 1—Distribution of Births and Perinatal Deaths by Selected Variables: Belo Horizonte, Brazil, 1999 Variable

Births, no. (%)

Deaths, no. (%)

Perinatal Mortality Rate

Rate Ratio (95% CI)

Birthplace Hospital Other Missing data

40 075 (97.9) 85 (0.2) 793 (1.9)

753 (97.2) 19 (2.5) 0 (0.0)

18.7 223.5 ...

1.0 11.9 (7.8, 17.8) ...

Total

40 953 (100.0)

775 (99.6)

18.9

1.0 (0.9, 1.1)

818 (2.0) 3648 (8.9) 36 487 (89.1)

382 (50.7) 177 (23.5) 194 (25.8)

466.9 48.5 5.3

Total

40 953 (100.0)

753 (100.0)

18.3

3.5 (3.0, 4.1)

Pregnancy Singleton Multiple Missing data

40 093 (97.9) 838 (2.0) 22 (0.1)

678 (90.1) 56 (7.4) 19 (2.5)

16.9 66.8 ...

1.0 4.0 (3.0, 5.1) ...

Total

40 953 (100.0)

753 (100.0)

18.3

1.1 (1.0, 1.2)

Delivery Cesarean Vaginal Missing data

17 002 (41.6) 23 922 (58.3) 29 (0.1)

252 (32.5) 497 (62.0) 4 (5.5)

14.8 20.7 ...

1.0 1.4 (1.2, 1.6) ...

Total

40 953 (100.0)

753 (100.0)

18.4

1.2 (1.1, 1.4)

514 (1.3) 21 536 (52.6) 11 338 (27.7) 5616 (13.7) 1949 (4.8)

40 (5.3) 358 (47.5) 122 (16.2) 74 (9.8) 159 (21.2)

77.8 16.6 10.8 13.2 ...

7.2 (5.1, 10.2) 1.5 (1.3, 1.9) 1.0 1.2 (0.9, 1.6) ...

40 953 (100.0)

753 (100.0)

18.4

1.7 (1.4, 2.1)

8970 (21.9) 16 194 (39.5) 8816 (21.6) 6180 (15.1) 793 (1.9)

95 (12.6) 220 (29.2) 166 (22.1) 272 (36.1) 0 (0.0)

10.6 13.6 18.8 44.0 ...

1.0 1.3 (1.0, 1.6) 1.8 (1.4, 2.3) 4.2 (3.3, 5.2) ...

40 953 (100.0)

753 (100.0)

18.4

1.7 (1.4, 2.1)

18 206 (44.5) 10 558 (25.7) 9688 (23.7) 2501 (6.1)

260 (35.3) 239 (32.4) 238 (32.3) 0 (0.0)

14.3 22.6 24.6 ...

0.6 (0.5, 0.7) 0.9 (0.8, 1.1) 1.0 ...

40 953 (100.0)

737 (100.0)

18.0

1.3 (1.1, 1.4)

a

Birthweight, g 500–1499 1500–2499 ≥ 2500

Maternal education, y 1500 adequate quality.

868 | Research and Practice | Peer Reviewed | Lansky et al.

scored above 1500 (adequate quality). Further details of the development and validation of the scoring system have been described previously.9 Maternal education (< 4 years, 4–7 years, 8–11 years, or ≥ 12 years) was used as an indicator of socioeconomic status. Using the Wigglesworth system,8 we classified the causes of perinatal death as antepartum, severe congenital malformation, immaturity (i.e., gestational period less than 37 weeks), intrapartum asphyxia, and other specific causes. We analyzed perinatal death rates according to hospital category, adjusting for 2 major confounders: maternal education and birthweight. Multivariable regression analysis was carried out to determine the association between hospital category and perinatal death. We excluded 231 (29.8%) antepartum deaths (those that happened before the onset of labor) because hospital obstetric care during labor could not affect birth outcomes in these cases. We also excluded nonhospital births (n = 85 [0.2%]) and deaths (n = 19 [2.5%]) and 3 deaths (0.4%) that took place in nonmaternity hospitals. In the case of newborn transfers between hospitals (17 of the deaths [2.2%]), death was attributed to the hospital of birth. Data entry, processing, and analyses were conducted with the software programs Epi Info 6.0 (Centers for Disease Control and Prevention, Atlanta, Ga) and Stata version 8 (StataCorp LP, College Station, Tex).

RESULTS The vast majority of the births took place in hospitals, although 19 (2.5%) of the deaths occurred outside a hospital, either at home, on the streets, during transfer to a hospital, or at another health facility (Table 1). This information could be ascertained only by hospital chart surveillance, because the babies’ birth and death certificates were registered as if they were born and had died in the hospitals. A total of 10.9% of all live births were low birthweight (