Abstract
OBJECTIVE:
To assess mortality in a birth cohort followed between 1982 and 2006 and its associated factors.
METHODS:
In 1982, all of the 5914 children born in hospitals in the city of Pelotas, Southern Brazil, were identified and followed up prospectively. Between 1982 and 1987, deaths were identified through regular visits to hospitals, cemeteries and death registries. As of 1987, death data were obtained through the Mortality Information System. The studied variables were: gender, color of mother, mother's schooling rate, family income, weight at birth, weight and height per age. Poisson regression was used to estimate the relative mortality risk.
RESULTS:
Between 1982 and 2006 there were 288 deaths. The infant mortality coefficient was 36 deaths/1 000 live births; and in the age brackets 1-4, years, 5-14 years and 15-24 years the mortality rates were, respectively, 14.4, 4.1 and 5.4 deaths for every 1 000 live births at the beginning of the period. In all age brackets, mortality was higher for individuals from low-income families, with a relative risk of 2.89 (95% CI: 2.08; 4.03) when comparing the first and third terciles after control for gender and skin color. Low weight at birth and height-for-age and weigh-for-height deficits were found to be associated to a higher mortality rate until age 4, but not after that age.
CONCLUSIONS:
The effects of social inequalities during childhood can be felt until the beginning of adult life, but birth weight and childhood nutritional status do not have a long-lasting effect on mortality rates for adolescents or young adults.
Keywords: Mortality, Risk Factors, Socioeconomic Factors, Cohort Studies, Brazil
INTRODUCTION
Infant mortality is inversely related to family income8 and weight at birth.13 Low socioeconomic level during childhood can also have long-term effects, such as higher cardiovascular disease mortality.4, 12
In most of the studies that have analyzed the effects of poverty on childhood and adult mortality, the data on socioeconomic status were obtained retrospectively, which may have lead to an underestimation of the magnitude of the effect measures.4 The Pelotas birth cohort study enables the assessment of the effect on mortality of weight at birth, socioeconomic and nutritional status during childhood, based on data gathered prospectively.
The goal of this study is to assess mortality of birth cohorts Born in 1982 and followed up until 2006, and to assess associated factor.
METHODS
In 1982, all the 5 914 children Born in all the three martenity hospitals in the city of Pelotas (Southern Brazil) were examined and their mothers interviewed. The babies represented 99.2% of the total births in Pelotas, in Rio Grande do Sul. In addition to perinatal information, the data used in this study were gathered in 1984 and 1986, when the average age of the children was 19.4 and 43.1 months, respectively. The visits were preceded by census-bureau visits covering the entire city area, and aimed at locating all the children born in 1982. Methodological details on the birth cohort and follow-ups have been published in another paper. 1
In all of the follow-ups during infancy, the mother or guardian answered a standardized and pre-coded questionnaire. In follow-ups during adolescence, one questionnaire was applied to the teenager and another to the mother or guardian. In the study carried out with the conscripts and in the 2004-5 follow-up, only the subject was interviewed.
From the beginning of the study, in order to identify cohort deaths, all the hospitals, cemeteries, death registries and the Regional Secretariat for Health were visited. It was found that, as of 1987, the civil registry was registering all the deaths occurred in the city, therefore, the other sources of data were no longer consulted. The causes of death were classified according to the Portuguese language edition of the International Statistical Classification of Diseases and Related Health Problems - ICD, versions 9 and 10.10
Cumulative incidence of mortality was estimated for the following periods: infancy; pre-school; five to 14 years, and 15 to 24 years. The infant mortality coefficient was based on the number of deaths children under one year of age per one thousand, and the denominator was the total number of live births. The cumulative incidence of mortality in the pre-school period was estimated from the number of deaths between one and four years of age per 1 000 children that reached their first year of age. Infant deaths and children lost during the follow-up study before turning one year old were excluded. To estimate the cumulative incidence of mortality between ages five and 14 and ages 15 to 24, deaths and follow-up losses of individuals younger than the age-brackets concerned were disregarded, in other words, individuals younger than five years of age for mortality analysis of the 5 to 14 group and individuals younger than 15 for deaths between 15 and 24 years.
The independent variables assessed were: sex, skin color of mother (observed by the interviewer); mother's schooling in years at the time of delivery; family income at birth (tertiles); weight at birth; weight and height for age, converted into z scores according to World Health Organization standards.16
Poisson regression was used to estimate relative risk of mortality during the 24-year study period.
Informed oral consent was given by the adult responsible for the children during the 1982-1986 stages of the study. This form of consent was a common practice at the time because there was no Ethics Committee at the Universidade Federal de Pelotas. In more recent stages, the Ethics in Research Committee, which is recognized by the National Ethics in Research Committee (Conselho Nacional de Ética em Pesquisa - CONEP), gave its approval, and informed consent was obtained through written statements given by the interviewees.
RESULTS
Between 1982 and 2006, 288 cohort deaths were identified. Table 1 shows that during infancy, the main causes of death were medical conditions originated during the perinatal period (42.8% of deaths) and infectious and parasitic diseases. In the pre-school period, parasitic and infectious diseases (31% of deaths) and respiratory diseases (24.1%) were the prevailing cause of the death. Between age five and 14, there was a small number of deaths and there was no prevailing cause of death. Whereas, in the 15 to 24 age group, external causes were responsible for approximately two out of every three deaths.
Table 1.
Mortality rate according to cause of death. Pelotas, Southern Brazil, 1982 – 2006.
| Cause of death | Death during infancy |
Death between 1-4 years |
Death between 5-14 years |
Death between 15-24 years |
|---|---|---|---|---|
| n (%) | n (%) | n (%) | n (%) | |
| Some parasitic and infectious diseases | 43 (20.0%) | 9 (31.0%) | 2 (10.5%) | 1 (4.0%) |
| Neoplasias | 0 | 0 | 2 (10.5%) | 2 (8.0%) |
| Endocrinal, nutritional and metabolic diseases | 0 | 0 | 2 (10.5%) | 0 |
| Diseases of the nervous system | 0 | 0 | 1 (5.3%) | 1 (4,0%) |
| Diseases of the circulatory system | 0 | 0 | 4 (21,1%) | 4 (16.0%) |
| Diseases of the respiratory system | 26 (12.1%) | 7 (24.1%) | 1 (5.3%) | 0 |
| Diseases of the digestive system | 0 | 0 | 1 (5.3%) | 0 |
| Conditions originated in the perinatal period | 92 (42.8%) | 0 | 0 | 0 |
| Congenital malformations, deformities and chromosome anomalies |
26 (12.1%) | 2 (6.9%) | 0 | 0 |
| Ill-defined | 23 (10.7%) | 4 (13.8%) | 0 | 1 (4.0%) |
| External causes | 0 | 0 | 3 (15.8%) | 16 (64.0%) |
| Other | 5 (2.3%) | 7 (24.1%) | 1 (5.3%) | 0 |
|
| ||||
| Total | 215 | 29 | 19 | 25 |
Table 2 shows mortality coefficients for the following groups: infants, pre-school, five to 14 years, and 15 to 24 years according to socioeconomic variables, demographics, weight at birth, weight and height for age. During infancy, mortality rates were higher for poor children born to black mothers, and with low birth weight. Among children between age one and four, low socioeconomic status at birth, low weight at birth, low height at birth and low weight-for-height at two years of age were associated with increased risk of death. In this last analysis, only deaths occurring between age two and four years and 11 months were considered. Among individuals aged five to 14, the only variable associated with mortality was mother's schooling rate. Among those aged 15 to 24, the male sex and low family income at birth were associated to a higher mortality rate.
Table 2.
Cumulative incidence of mortality (per 1 000) according to demographic and socioeconomic variables, condition at birth and breastfeeding period. Pelotas, Southern Brazil, 1982-2006.
| Variable | Death during Infancy |
Death between 1-4 years |
Death between 5-14 years |
Death between 15-24 years |
|---|---|---|---|---|
| Sex | p=0.29* | p=0.99* | p=0.53* | p=0.04* |
| Male | 38.9 | 5.4 | 4.8 | 7.7 |
| Female | 33.4 | 5.4 | 3.2 | 2.8 |
| Mother's skin color | p<0.001* | p=0.10* | p=0.51* | p=0.04* |
| White | 32.2 | 4.5 | 3.7 | 4.2 |
| Black or Mixed | 55.7 | 9.4 | 6.0 | 10.9 |
| Mother's level of schooling (years) | p<0.001** | p=0.001** | p=0.04* | p=0.41** |
| 0 – 4 | 53.1 | 10.3 | 1.3 | 6.0 |
| 5 – 8 | 35.4 | 4.0 | 6.5 | 6.1 |
| ≥ 9 | 16.1 | 1.5 | 3.5 | 3.5 |
| Family income at birth | p<0.001** | p=0.03** | p=0.93* | p=0.02** |
| 1st tertile | 61.6 | 8.7 | 4.0 | 9.4 |
| 2nd tertile | 30.8 | 4.3 | 3.7 | 3.7 |
| 3rd tertile | 16.7 | 3.3 | 4.5 | 3.3 |
| Weight at birth (g) – 3 groups | p<0.001** | p=0.02** | p=0.39* | p=0.86** |
| < 2500 | 207.9 | 12.7 | 0.0 | 6.0 |
| 2500 – 2999 | 35.2 | 7.1 | 5.4 | 5.5 |
| ≥ 3000 | 13.1 | 4.0 | 4.0 | 5.3 |
| Height-for-age at 2 years (Z scores) | p=0.001** | p=0.12* | p=0.72* | |
| 1st tertile | 73 | 3.5 | 5.0 | |
| 2nd tertile | 1.8 | 2.1 | 6.4 | |
| 3rd tertile | 0.6 | 6.9 | 4.2 | |
| Weight-for-height at 2 years of age (Z scores) | p=0.002** | p=0.80* | p=0.73** | |
| 1st tertile | 6.7 | 4.3 | 5.7 | |
| 2nd tertile | 2.5 | 5.0 | 5.0 | |
| 3rd tertile | 0.6 | 3.4 | 4.8 | |
| Height-for-age at 4 years of age (Z scores) | p=0.67* | p=0.12** | ||
| 1st tertile | 4.3 | 7.3 | ||
| 2nd tertile | 2.9 | 7.2 | ||
| 3rd tertile | 4.9 | 2.8 | ||
| Weight-for-height at 4 years of age (Z scores) | p=0.58** | p=0.61** | ||
| 1st tertile | 2.9 | 2.9 | ||
| 2nd tertile | 5.0 | 5.0 | ||
| 3rd tertile | 4.2 | 4.2 | ||
|
| ||||
| Total | 36.3 | 5.4 | 4.1 | 5.4 |
Categorical
Test for linear trend
Table 3 shows findings on the effects of sex, skin color and family income at birth on mortality rates during the 24-year period of study, the increased risk of death among children born to black or mixed mothers was mediated by socioeconomic variables, because after control for family income and sex, the relative risk of mortality decreased from 1.82 (95% CI: 1.42; 2.33) to 1.21 (95% CI: 0.93; 1.58). On the other hand, even after control for sex and skin color of father, the mortality risk was higher among those born to low-income families in the first tertile.
Table 3.
Relative risk of mortality according to family income, mother's skin color and sex. Pelotas, Southern Brazil, 1982-2006.
| Variable | Relative Risk | |
|---|---|---|
| (95% IC) | ||
| Crude | Adjusted* | |
| Sex | ||
| Male | 1.26 (1.00;1.59) | 1.25 (0.99;1.57) |
| Female | 1.00 (reference) | 1.00 (reference) |
| Mother's skin color | ||
| White | 1.00 (reference) | 1.00 (reference) |
| Black or Mixed | 1.82 (1.42;2.33) | 1.21 (0.93;1.58) |
| Family income at birth |
||
| 1st tertile | 3.07 (2.25;4.19) | 2.89 (2.08;4.03) |
| 2nd tertile | 1.58 (1.12;2.23) | 1.58 (1.11;2.24) |
| 3rd tertile | 1.00 (reference) | 1.00 (reference) |
Adjusted for: gender, family income and color of skin.
Deaths due to external factors were more frequent in males whose mothers' were black or mixed and who were born to low-income families in the first tertile. After control for family income at birth, mother's schooling rate and sex, the relative risk of death due to external causes among individuals born to black or mixed mothers decreased from 5.10 (95% CI: 2.08;12.51) to 3.77 (95% CI: 1.11;12.74). The effect of low family income at birth remained significant after control for sex and skin color of mother. (Table 4)
Table 4.
External causes mortality rate according to demographic and socioeconomic variables and weight at birth. Pelotas, Southern Brazil, 1982-2006.
| Variable | Mortality coefficient External Causes (10 000) |
Relative risk | |
|---|---|---|---|
| (95% IC) | |||
| Crude | Adjusted | ||
| Sex | p<0.001* | ||
| Male | 3.6 | 15.7 (2.10;117.4) | 15.3 (2.04;115.4) |
| Female | 0.2 | 1.00 (reference) | 1.00 (reference) |
| Mother's skin color | p<0.001* | ||
| White | 1.2 | 1.00 (reference) | 1.00 (reference) |
| Black or Mixed | 6.0 | 5.10 (2.08;12.51) | 3.77 (1.11;12.74) |
| Family income at birth | p=0.02** | ||
| 1st tertile | 3.7 | 3.80 (1.06;13.60) | 1.79 (0.19;16.51) |
| 2nd tertile | 1.5 | 1.59 (0.38;6.62) | 1.07 (0.17;6.74) |
| 3rd tertile | 1.0 | 1.00 (reference) | 1.00 (reference) |
| Weight at birth (grams) | p=0.38* | ||
| < 2500 | 3.0 | 1.29 (0.30;5.60) | 1.23 (0.28;5.46) |
| 2500 – 2999 | 0.9 | 0.39 (0.09;1.71) | 0.40 (0.09;1.71) |
| ≥ 3000 | 2.3 | 1.00 (reference) | 1.00 (reference) |
|
| |||
| Total | 2.1 | ||
Categorical
Linear trend test
DISCUSSION
This study enabled the assessment of the effects of early determinants (socioeconomic and nutritional) on mortality, from infancy to early adulthood. The prospective nature of this study makes it less subject to information bias. Studies that assess mortality at different stages of life are relevant, because they provide cues on the effect of exposure at different moments in the life cycle.
The low socioeconomic level in infancy was associated to higher mortality rates during the whole period of the study, whereas, during infancy, low income was associated to a higher risk of death due to infectious diseases and perinatal causes.14 During adolescence and the beginning of adult life, a lower socioeconomic level at birth was associated to deaths due to external causes.
Mortality rate due to external causes was also higher among children of black or mixed mothers; however, family income at birth was only able to explain part of this association. The higher mortality rate due to external factors among children born to black or mixed mothers corroborates national data. In Brazil, between 1996 and 2000, homicide mortality rates were higher among Blacks, mainly during adolescence and beginning of adult life, which mirrors the racial inequality in Brazil where violence predominates among young black males.
Low weight at birth was associated to a higher mortality rate only in the first four years of life. This was not found by Kajantie et al,6 who reported, in a study carried out in Finland a correlation between low weight at birth and higher mortality in adult life. Studies analyzing mortality due to cardiovascular diseases also reported a higher mortality rate among individuals born underweight.3,7 The follow-up period may be a possible explanation for these differences, because in Finland, individuals were followed-up until age 74,6 whereas in this study, follow-up was carried out until age 24, which is an age at which cardiovascular diseases do not yet represent an important cause of morbidity and mortality.
Overweight in the first years of life has been considered an early exposure that can have long-term effects on the occurrence of cardiovascular diseases2 or its risk factors, such as dyslipidemy,9 high blood pressure5 and obesity.15 In poor populations, sub-nutrition has been associated to short-term mortality.11 The findings of this study confirm the negative effect of poor nutrition on mortality up to age five, but no long-term association between nutritional status – due to deficit or excess – and mortality was found. The lack of correlation with excess weight can also be due to the shorter period of follow up.
In conclusion, the effect of social inequalities during infancy are observed at least until the beginning of adult life, whereas weight at birth and nutritional status during infancy did not have any long lasting effects on mortality rates for adolescents and young adults.
Acknowledgments
The 1982 birth cohort study is currently supported by the Wellcome Trust initiative entitled Major Awards for Latin America on Health Consequences of Population Change. Previous phases of the study were supported by the International Development Research Center, The World Health Organization, Overseas Development Administration, European Union, National Support Program for Centers of Excellence (PRONEX), the Brazilian National Research Council (CNPq) and Brazilian Ministry of Health.
Footnotes
This article is based on data from the study “Pelotas birth cohort, 1982” conducted by Postgraduate Program in Epidemiology at Universidade Federal de Pelotas.
The authors declare that there are no conflicts of interest.
REFERENCES
- 1.Barros FC, Victora CG, Horta BL, Gigante DP. Metodologia do estudo da coorte de nascimentos de 1982 a 2004-5, Pelotas, RS. Rev Saude Publica. 2008;42(Supl 2):7–15. doi: 10.1590/s0034-89102008000900003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Eriksson JG, Forsén T, Tuomilehto J, Osmond C, Barker DJ. Early growth and coronary heart disease in later life: longitudinal study. BMJ. 2001;322(7292):949–53. doi: 10.1136/bmj.322.7292.949. DOI: 10.1136/bmj.322.7292.949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Forsén T, Eriksson JG, Tuomilehto J, Osmond C, Barker DJ. Growth in utero and during childhood among women who develop coronary heart disease: longitudinal study. BMJ. 1999;319(7222):1403–7. doi: 10.1136/bmj.319.7222.1403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Galobardes B, Lynch JW, Davey Smith G. Childhood socioeconomic circumstances and cause-specific mortality in adulthood: systematic review and interpretation. Epidemiol Rev. 2004;26:7–21. doi: 10.1093/epirev/mxh008. DOI: 10.1093/epirev/mxh008. [DOI] [PubMed] [Google Scholar]
- 5.Horta BL, Barros FC, Victora CG, Cole TJ. Early and late growth and blood pressure in adolescence. J Epidemiol Community Health. 2003;57(3):226–30. doi: 10.1136/jech.57.3.226. DOI: 10.1136/jech.57.3.226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kajantie E, Osmond C, Barker DJ, Forsén T, Phillips DI, Eriksson JG. Size at birth as a predictor of mortality in adulthood: a follow-up of 350 000 person-years. Int J Epidemiol. 2005;34(3):655–63. doi: 10.1093/ije/dyi048. DOI: 10.1093/ije/dyi048. [DOI] [PubMed] [Google Scholar]
- 7.Leon DA, Lithell HO, Vâgerö D, Koupilová I, Mohsen R, Berglund L, et al. Reduced fetal growth rate and increased risk of death from ischaemic heart disease: cohort study of 15 000 Swedish men and women born 1915-29. BMJ. 1998;317(7153):241–5. doi: 10.1136/bmj.317.7153.241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Menezes AM, Hallal PC, Santos IS, Victora CG, Barros FC. Infant mortality in Pelotas, Brazil: a comparison of risk factors in two birth cohorts. Rev Panam Salud Publica. 2005;18(6):439–46. doi: 10.1590/s1020-49892005001000007. DOI: 10.1590/S1020-49892005001000007. [DOI] [PubMed] [Google Scholar]
- 9.Miura K, Nakagawa H, Tabata M, Morikawa Y, Nishijo M, Kagamimori S. Birth weight, childhood growth, and cardiovascular disease risk factors in Japanese aged 20 years. Am J Epidemiol. 2001;153(8):783–9. doi: 10.1093/aje/153.8.783. DOI: 10.1093/aje/153.8.783. [DOI] [PubMed] [Google Scholar]
- 10.Organização Mundial da Saúde . Manual da classificação estatística internacional de doenças, lesões e causas de óbito. São Paulo: 1980. Revisão 1975. [Google Scholar]
- 11.Pelletier DL, Frongillo EA, Jr, Schroeder DG, Habicht JP. The effects of malnutrition on child mortality in developing countries. Bull World Health Organ. 1995;73(4):443–8. [PMC free article] [PubMed] [Google Scholar]
- 12.Power C, Hyppönen E, Smith GD. Socioeconomic position in childhood and early adult life and risk of mortality: a prospective study of the mothers of the 1958 British birth cohort. Am J Public Health. 2005;95(8):1396–402. doi: 10.2105/AJPH.2004.047340. DOI: 10.2105/AJPH.2004.047340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Victora CG, Barros FC, Vaughan JP, Teixeira AM. Birthweight and infant mortality: a longitudinal study of 5914 Brazilian children. Int J Epidemiol. 1987;16(2):239–45. doi: 10.1093/ije/16.2.239. DOI: 10.1093/ije/16.2.239. [DOI] [PubMed] [Google Scholar]
- 14.Victora CG, Barros FC, Vaughan JP. Epidemiologia da Desigualdade. Hucitec; São Paulo: 2006. [Google Scholar]
- 15.Victora CG, Sibbritt D, Horta BL, Lima RC, Cole TJ, Wells J. Weight gain in childhood and body composition at 18 years of age in Brazilian males. Acta Paediatrica (Oslo) 2007;96:296–300. doi: 10.1111/j.1651-2227.2007.00110.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.WHO Multicentre Growth Reference Study Group WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl. 2006;450:76–85. doi: 10.1111/j.1651-2227.2006.tb02378.x. [DOI] [PubMed] [Google Scholar]