The health transition in the Third World: The health transition in the Third World John Powles March 2007


Picture of India: Picture of India


Mortality decline in (to-day’s) low and middle income countries (the ‘Third World’): Mortality decline in (to-day’s) low and middle income countries (the ‘Third World’) India as an example Why? Data – censuses since 1880 ‘Central’ Decline slower than East Asia But Faster than Africa


India: the demographic transition since late C19: India: the demographic transition since late C19


Life expectancy in India since the 1880s*: Life expectancy in India since the 1880s* * estimates to the 1950s are for decades and do not show short term deviations


Episodes of catastrophic mortality in India since the C18: Episodes of catastrophic mortality in India since the C18


Life expectancy in India since the 1880s*: Life expectancy in India since the 1880s* * estimates to the 1950s are for decades and do not show short term deviations 1940s


Survival trends in childhood, India late C19 to 1940s: Survival trends in childhood, India late C19 to 1940s


Survival trends in adulthood, India late C19 to 1940s: Survival trends in adulthood, India late C19 to 1940s


Survival trends in adulthood, India late C19 to 1940s: Survival trends in adulthood, India late C19 to 1940s NB: 5 out of 6 15yr olds died before reaching 65 Mortality risks were NOT concentrated in childhood.


Survival trends in childhood, India since late C19: Survival trends in childhood, India since late C19


Survival trends in adulthood, India since late C19: Survival trends in adulthood, India since late C19


Severity of pre-transition mortality: Severity of pre-transition mortality Except in what are necessarily periods of transition, the death rate approximates the birth rate. 2 patterns before modern period: Very high birth ≈ death rates ‘Sub-maximal’ birth ≈ death rates


Very high birth ≈ death rates: Very high birth ≈ death rates ‘High pressure’ regimes Eg ‘Poor agrarian’ societies India, 1900; China 1930 Needed full use of reproductive potential TFR 6-7 Universal early marriage ‘Shortened’ birth intervals


‘Sub-maximal’ birth ≈ death rates: ‘Sub-maximal’ birth ≈ death rates ‘Low pressure’ regimes Reproductive potential moderated 2 types


1. ‘Natural’ moderation of fertility: 1. ‘Natural’ moderation of fertility Eg Hunter-gatherers Ie pre-agrarian economies (and great apes) Universal early marriage ‘Long’ birth intervals (4-5 years) TFR 4-5 e0 ≈ 35


2. ‘Institutional’ moderation of fertility: 2. ‘Institutional’ moderation of fertility Eg 1. ‘European marriage pattern’ Marriage delayed til mid 20s (Variable) proportion never married (15-40%) TFR 4-5 e0 ≈ 35 2. Other family systems eg Qing China Universal early marriage, but Control of marital ‘fertility’ Incl by female infanticide


Four survival patterns and transitions between them: Four survival patterns and transitions between them


Slide19: So the ‘bad old days’ Were VERY bad But not so OLD


Why has health improved?: Why has health improved? Increased knowledge and changed outlook Better health Deliberate attempts to control disease ‘Narrow’ interpretation = ‘interventions’


Intervention: Malaria control: Intervention: Malaria control Previously a leading direct and indirect cause of death in many countries eg Sri Lanka, Mauritius Spraying houses with residual DDT killed mosquitos and broke the chain of transmission Introduction of spraying campaigns coincided with massive falls in death rates eg Sri Lanka 1945-1955


Sri Lanka: Mortality decline after WWII: Sri Lanka: Mortality decline after WWII


Sri Lanka: Prevalence of malaria before the decline: Sri Lanka: Prevalence of malaria before the decline


Sri Lanka: Estimating the contribution of malaria control to mortality decline: Sri Lanka: Estimating the contribution of malaria control to mortality decline Before WWII mortality levels by district correlated strongly with the spleen rate After malaria control Mortality declined by 50% Mortality levels across districts become homogeneous


Sri Lanka: Estimating the contribution of malaria control to mortality decline: Sri Lanka: Estimating the contribution of malaria control to mortality decline Statistical model Malaria control only contributed about ¼ of the total decline.


Intervention: Management of diarrhoea: Intervention: Management of diarrhoea


Sachets of oral rehydration salts: Sachets of oral rehydration salts


“Tenting" of the skin over the right side of the child's abdomen caused by the pinch of the examiner's fingers seen withdrawing to the right. Tenting is seen when a child has lost at least 5% of body water.: “Tenting" of the skin over the right side of the child's abdomen caused by the pinch of the examiner's fingers seen withdrawing to the right. Tenting is seen when a child has lost at least 5% of body water.


But does ORT really exemplify an ‘intervention’ ?: But does ORT really exemplify an ‘intervention’ ? ORT is administered within households (ie not by professionals) and its wide diffusion has depended on expanded health consciousness


Two explanations: Two explanations Increased knowledge and changed outlook Better health ‘development’


Narrow view of ‘development’ as increased income: Narrow view of ‘development’ as increased income Income More food Better nutrition Resistance to infection Better survival


Relative risk of dying in the next 6 months by % of the Harvard weight for age norm: Relative risk of dying in the next 6 months by % of the Harvard weight for age norm Indian infants aged 1 to 36 months


Slide40: However net nutrition depends on both Food and Burden of infection


The burden of infection on infants and children in poor populations: The burden of infection on infants and children in poor populations One estimate: 160 days ill / year incl 3-4 bouts of diarrhoea (enough to halve growth potential) 4-5 respiratory infections 1 or more specific infections eg measles In 25% of cases measles causes loss of body weight > 10%


Field studies of poor agrarian populations with high burdens of infection: Field studies of poor agrarian populations with high burdens of infection


Slide43: 1978


Slide47: 3 boys born Feb 1964 at 10 years of age Similar height to 7 year olds in US


Conclusion of Narangwal Study, Punjab, India, 1968-72: Conclusion of Narangwal Study, Punjab, India, 1968-72 'If [the infection-malnutrition-infection] sequence moves rapidly, the child dies - although neither malnutrition nor infections by themselves would have caused death.'


Ie. Net nutritional status is a critical determinant of survival, but: Ie. Net nutritional status is a critical determinant of survival, but It depends not only on Food consumed But also on Burden of infection … and this is subject to social (institutional) influences Eg literacy of the mother cleanliness


The changing relationship between life expectancy and income (Preston et al): The changing relationship between life expectancy and income (Preston et al)


Which aspects of socio-economic development most strongly predict trends in child mortality?: Which aspects of socio-economic development most strongly predict trends in child mortality?


How does knowledge gained at school increase child survival?: How does knowledge gained at school increase child survival? 1. Practical knowledge 2. Less fatalism 3. More able to deal with modern world 4. 'Changes not only the educated but also the attitudes of others to them' (Caldwell) the wisdom of the school is able to displace the wisdom of the old


Lessons: health improvement in low income countries: Lessons: health improvement in low income countries Depends less on professional ‘interventions’, and income Depends more on capacity to use knowledge social organisation supportive of this capacity


The nature of ‘technical progress’: The nature of ‘technical progress’


Previous models for the production of health: Previous models for the production of health Technical progress Common to all


Slide65: Suppose ‘technical progress’ is thought of instead as each country’s distinctive capacity to make use of the international stock of knowledge ie that the coefficient on time (in statistical models) is allowed to vary by country


Slide66: Ie increasing capacity to im- prove health with given level of other measured inputs


Conclusion of Jamison et al: Conclusion of Jamison et al ‘…Even in a period of rapid economic growth income changes can account for only a modest fraction of the changes in infant mortality in most countries.’


Interdependent nature of health determinants: Interdependent nature of health determinants


Some developments at the supra-national level: Some developments at the supra-national level


Commission on Macroeconomics and Health (WHO) 2001: Commission on Macroeconomics and Health (WHO) 2001 Previous idea: Wealthy -> Healthy Now Healthy -> Wealthy


Reversals: 1. HIV: Reversals: 1. HIV


Slide77: Life expectancy: early 1990s 56 2001 37


Reversals: 2. Adult mortality in the former Soviet countries: Reversals: 2. Adult mortality in the former Soviet countries


Slide80: Copyright ©2003 BMJ Publishing Group Ltd. Men, T. et al. BMJ 2003;327:964 Fig 2 Observed and expected mortality in young and middle aged Russian adults 1991-2001. Data for men from 8 317 789 observed, 6 175 768 assuming constant rate from 1991, and 5 311 486 assuming same decrease as in Czech Republic. For women numbers were 3 699 717, 3 074 790, and 2 672 962 Observed 8,317,789 Expected 6,175,768 Excess 2.1 m Observed 3,699,717 Expected 3,074,790 Excess 0.6 m M+F 2.7 m excess deaths


Summary: Summary There have been dramatic improvements in survival in low and middle income countries during the 20th century Simple explanations of this success (eg ‘disease control’, ‘economic development’) are inadequate Most fundamental is the advance of knowledge and the ways in which societies equip themselves to make use of that knowledge eg by universal schooling Oral rehydration treatment for diarrhoea illustrates how scientific advance, economic capacity and institutional change (female schooling) can combine to lower child mortality The 2 major reversals in global health trends are HIV, especially in sub-saharan Africa, and rising adult mortality in Russia and other former Soviet countries. HIV prevalence is falling in countries such as Uganda and Kenya but there is, as yet, no clear sign that Russia’s public health catastrophe is under control.