Philippine mortality in changing times

Cabigon, Josefina V

Philippine mortality in changing times

Date

1990

Authors

Cabigon, Josefina V

Abstract

Hand in hand with an increasing recognition of mortality as one of the most important demographic processes interacting with development in the Philippines, there is more awareness of the need for identification of major health issues and analysis of the relevant available data. This thesis responds to the need by undertaking an extensive appraisal of the mortality conditions of the Philippines. Given the inadequacy of most statistics in the Philippines, the thesis includes the assessment of socioeconomic and health-related statistics, registered deaths, census data and the 1983 National Demographic Survey (NDS). This critical evaluation reveals that: (1) Published socioeconomic and healthrelated statistics are useful for the analysis of covariates of mortality and the construction of a composite index of development at the provincial level; (2) Techniques applied to obtain estimates of the level of completeness of death registration are largely consistent and can be used to correct the registered deaths for overregistration or underregistration of deaths, and to generate life tables for 1960, 1970, 1975 and 1980 at the provincial, regional and national levels; (3) Age-reporting is more or less accurate in the 1960 to 1980 censuses, particularly at ages below 55 years; the age-heaping found at ages 55 years and above is minimized by grouping into five-year standard intervals; (4) There are some problems in the classification of marital status in the 1975 census; (5) There are no indications of marked peculiarities or measurement errors of the 1960,1970 and 1980 data used in the regression analyses; (6) The extent of omission of births by women aged 45-49 years and misplacement of dates of births by women aged 40-44 years is not so great as to seriously distort the birth histories of the 1983 NDS; and (7) There is a close fit between the generated life tables and the United Nations Latin American model life tables at all ages and both with the United Nations Latin American and Coale-Demeny model life tables at childhood ages for the nation as a whole and all regions except Mindanao. The life tables produced are used, along with the 1978 Philippines Fertility Survey and the 1983 NDS, to re-examine mortality trends. The long-term rapid pace of mortality reduction before 1960 decelerated for males during 1960-1970, slightly hastened during 1970-1975 and slackened again during 1975-1980; the same trend occurred for women but with an acceleration during 1975-1980. The level of infant and child mortality has plateaued since 1960. There is evidence of some coincidence between mortality and economic growth trends. The changes in socioeconomic and health-related variables are quantified and related to changes in the structure of mortality relationships through estimations of a number of equations based upon data from the periods 1960-1970, 1960-1980 and 1970-1980. These yield no significant changes during 1960-1970, the decade of retardation of economic growth and mortality decline. During 1960-1980, there were significant changes related to the effects of road density, but such changes did not mean fast improvements in this indicator nationwide, but some improvement in selected areas, or sharply reduced intensity of exogenous factors, thus worsening or maintaining the gap between the affluent and non-affluent areas. During 1970-1980, changes in the effects of exogenous factors omitted in the regression equations, such as health programs and improvements in personal health practices, are more dominant than changes in the effects of road density. Such changes appear to have benefited females but not males as seen from the accelerating female mortality decline but decelerating male mortality decline during the 1975-1980 period. The excess male mortality over time, especially among the single and widowers, is consistent with findings of other analysts that all causes mortality increased clearly in the Philippine male population but decreased in the female population during the period 1964-1976. Examination of the relationship between socioeconomic and health-related variables and overall male and female survival and infant mortality at the provincial level of analysis using multiple regression analysis has shown that the factors strongly associated with overall and male survival, are road density, motor vehicles, safe drinking water supply and toilet sanitation; in the case of female survival, the factors are road density, motor vehicles, electricity, literacy, safe drinking water supply, toilet sanitation and use of midwife; and in the case of infant mortality, they are road density, motor vehicles, literacy and use of physician. It is evident that education, as measured by literacy, is positively and inversely related to female life expectancy and infant mortality, respectively, but not to male life expectancy. Education and other development variables dominate over health variables in affecting female life expectancy and infant mortality. While this appears to be the case for male life expectancy in early years, health factors seem to be of equal importance or of dominance in influencing life expectancy in recent years. Estimates of the relative impact of a variety of socioeconomic, demographic, health relatedenvironmental and proximate factors upon the risk of neonatal, post-neonatal, overall infant and child death using log-linear models, reveal net independent effects of maternal education, preceding birth interval, birth order and maternal age at childbirth on neonatal mortality; of household income, maternal education, preceding birth interval, birth order, maternal age at birth of child, time period of birth, housing quality and toilet sanitation on post-neonatal and child mortality; and of almost the same seven predictors, with housing quality replaced by source of drinking water on overall infant mortality. While the demographic variables are more dominant during infancy, the socioeconomic and health related-environmental factors are prominent at later childhood ages. Breastfeeding, immunization and supplemental feeding are found important, but their measured effects are attenuated through circular causation and censoring problems, which are not accounted for in the analysis. There are no significant net independent effects of prenatal care, place of maternal delivery and birth attendance on child mortality, at any age. On the whole, while other countries have recorded remarkable successes in reducing mortality, the Philippines, faced with an ailing economy, has followed a deceleration in mortality decline. The varying patterns of effects of socioeconomic, health-related and demographic factors on the components of child mortality, the differing trend in and patterns of influence of socioeconomic and health-related factors on male and female mortality and the lack of substantial structural and developmental changes shine a spotlight on the reasons of this retardation.