Population health metrics
Objectives
By the end of the session, students should understand:
• Purpose of different health indicators
• Main health indicators used in public / population health
• Selected features and findings of the Global Burden of Disease
(GBD) programme in terms of:
• Health outcomes used in the GBD
• Conditions accounting for most ill health burden
• Risk factors responsible for ill health
• Absolute vs. relative inequalities of health
Steps to improve health
Steps to improve health
• Measure the outcome of interest / track changes in health and
disease
• Identify / understand causal factors
• Assess what health and social programmes will provide greatest gains
for healthy life
• Design and implement interventions
• Evaluate the interventions (measure the outcome of interest)
Population health metrics
• Indicators used to measure, estimate or quantify:
• health and disease in population(s)
• risk factors / determinants
• attribution of ill health to risk factors / determinants
• Sometimes “big brush” picture
• Largely based on routinely available data
• Often combining data from various sources
• Crucial for policy decisions, priority setting, design of interventions,
evaluation of interventions
“Conventional” measures
• Prevalence of a disease / exposure
• Incidence of a disease
• Mortality
• all causes vs. cause-specific rates
• all ages vs. age-specific rates
• Life expectancy
• At birth
• At specific age
Life expectancy: assumptions
• ‘Period life tables’ unrealistically assume mortality will stay the same in
the future
• Life expectancy has been growing at around 3 months a year for
decades, corresponding to the annual risk of death reducing at about 2%
per year.
• ‘cohort life tables’ make various projections about whether these trends
will continue in the future:
• the 'central projection' says girls born now have a life expectancy of 94, with a
median and mode of around 100
• men have a life expectancy of 91, with a median and mode of around 96.
• Under the ‘high' projections, with the possibly implausible assumption that the
increases continue at the same rate in the future, children born today will on
average live more than 100 years.
65
67
69
71
73
75
77
79
81
83
85
2020201020001990198019701960
Life expectancy Czech Republic 1960-2020
Summary measures
Building a summary measure from
partial measures of health
From Field & Gold, 1998
Life paths
Survival and health curves
Healthy life expectancy (HALE)
• Healthy life expectancy, or health-adjusted life expectancy (HALE)
measures the number of years that a person at a given age can expect
to live in good health, accounting for mortality and disability
• = the average number of years that a newborn can expect to live in
"full health"—in other words, not hampered by disabling illnesses or
injuries
• Summarises mortality and non-fatal outcomes in a single measure of
average population health
• Can compare health between countries or measure changes over
time
• Can inform policy questions dependent on how morbidity changes as
mortality decreases
Morbidity compression (rectangularisation )
Compression of morbidity
hypothesis – Fries, 1998
Healthy life expectancy at birth by country, 2010
The Lancet 2012 380, 2144-2162 A – male HALE ; B – female HALE
Global Burden of Diseases (GBD)
• “global descriptive epidemiology”
• systematic, scientific effort to quantify the comparative magnitude of
health loss due to diseases, injuries, and risk factors by
• age
• sex
• geographies
• specific points in time
• compare the effects of different diseases that kill people prematurely
and cause ill health and disability
• GBD was established by WHO, WB and Harvard in 1996, currently led by
the Institute for Health Metrics and Evaluation (IHME), University of
Washington
GBD metrics
• all-cause mortality
• deaths by cause
• years of life lost (YLLs)
• years lived with disability (YLDs)
• disability adjusted life years (DALYs) for
• Comprehensive list (2013)
• 291 causes of diseases and injuries
• 67 risk factors
The disability-adjusted life years (DALY)
• A summary measure which combines time lost through
premature death and time lived in states of less than optimal
health, loosely referred to as “disability”.
• One DALY can be thought of as one lost year of ‘healthy’ life.
• Combines the years of life lost through
• premature death
• years of healthy life lost through disability.
• DALYs lost = Years of healthy life lost.
• If everybody in a population lived to 80 completely healthy,
then there would be zero DALYs
Calculating DALYs
1. Classify all disease and conditions into 107 categories in 3
groups and assign into 7 disability classes weighted from 0
(perfect health) to 1 (death).
2. Assign all deaths to a category by age, sex and region
3. Calculate years of life lost per death
4. Estimate all cases of disability by age, sex , region, severity
(disability class) and duration (years of healthy life lost) until
remission or death
5. Combine all deaths and disability losses by cause, age, sex and
region
6. Allow a discount rate of 3% so future years of healthy life are
valued at progressively lower levels
7. Weight years of life lost at different relative values-less for
children and aged, more for adults (maximum value at age 25)
8. Sum all DALYS to obtain the Global Burden of Disease
Examples of mean disability weights (2004)
•AIDS 0.505
•Infertility 0.180
•TB 0.272
•Blindness 0.600
•Diabetes 0.015
•Depression 0.399
•Alzheimer’s 0.666
•Angina 0.141
•Deafness 0.234
YLL and YLD composition of total DALYs by region, 2010
Source: The Global Burden of Disease: Generating Evidence, Guiding Policy (GBD 2010)
Trends in DALYs from 1990 to 2015 by cause
(A) – Global DALYs; (B) – Crude DALY rates; (C) – Age-standardised DALY rates
The Lancet 2016 388, 1603-1658
Incorporating risk factors
• Comparative risk assessment (CRA) approach, developed by
Murray and Lopez (Epidemiol., 1999)
• Conceptual framework for population risk assessment across
risks and over time
• Evaluates how much of the burden of disease observed in a
given year can be attributed to past exposure to a risk factor
• Attributable burden is estimated by comparing observed
health outcomes to those that would have been observed
had a counterfactual level of exposure occurred in the past
• Different risks lead to different health outcomes →
separate assessments undertaken for specific risk-outcome
pairs
WHO (2002).
Global proportion of all-cause DALYs attributable to risk
factor clusters and their overlaps by age for both sexes, 2013
Summary
• Different health indicators have different purpose
• Conventional indicators remain primary sources of data
• GBD metrics extremely influential
• DALY widely used but complex and sometimes difficult to
interpret