SlideShare a Scribd company logo
1 of 100
Global Burden of Disease Study – 2010

Dr. Rizwan S A, M.D.,
Outline of presentation
•
•
•
•
•
•
•
•
•

Beginnings
Prior studies
What is GBD?
Need for GBDS – 2010
Methodology of GBDS – 2010
Analytical components of GBDS – 2010
Five critical choices
Results overview
Articles published
Beginnings
• Original GBD 1990 Study was commissioned by the
World Bank in 1991 to provide a comprehensive
assessment of the burden of 107 diseases and
injuries and ten selected risk factors for the world
and eight major regions in 1990
• The disability-adjusted life year (DALY), a time-based
measure that combined years of life lost due to
premature mortality and years of life lost due to time
lived in health states less than ideal health, was
developed to assess the burden of disease.
Prior studies
• Global Burden of Disease Study 1990 – World bank
commissioned (for 131 diseases and injuries and 10
risk factors)
• Global Burden of Disease Study 2000 (GBD 2000)
• Results for 1999, 2000, 2001, and 2002 were published
in World Health Reports 1999-2003
• WHO published estimates for 2004
• WHO also issued updates for estimates in 2009
• National Burden of Disease studies in at least 37
countries
What is GBD ? - 1
• There are some 6 billion people in the world, and hundreds of
millions experience disease or injury each year.
• Taken as a whole, the combined pain, suffering, loss of
productivity and unrealised hopes and dreams are our world’s
burden of disease !
• In other words, the burden of disease is a measurement of
the gap between the current health of a population and an
ideal scenario where everyone completes their full life
expectancy in full health
• The Global Burden of Disease project attempts to measure
this total disease burden
What is GBD? - 2
• Comparison is at the heart of the GBD approach
• Comparison requires consistent definitions over
time and across place;
• Comparison across diseases for non-fatal health
outcomes requires a system of disability eights
that produce meaningful summary metrics across
hundreds or thousands of specific disease sequelae.
• metrics that quantify types of outcomes in
commensurable units.
What is GBD? - 3
• One needs estimates for all diseases, injuries, and
risk factors, not just those for which there are the
best data in a place
• Poorly studied health problems may in fact be
some of the most important in some places and
neglected because no one has collected good data
in the past
• All age-groups are included in the GBD so that
comparisons across age and sex groups are also a
key advantage
What is GBD? - 4
• GBD construct of the burden of disease is health
loss, not income or productivity loss
• Health loss does not capture all the key dimensions
of health events;
• GBD treats a unit of health loss as equally important
across individuals whether rich or poor, old or young,
male or female.
• Incorporate advances in epidemiology, demography,
statistics, and other quantitative disciplines.
• Completely revamped and improved tools.
What is GBD? - 5
• An approach to global descriptive epidemiology
• It is a systematic, scientific effort to quantify the comparative
magnitude of health loss due to diseases, injuries, and risk
factors by age, sex, and geographies for specific points in time
• Provides an opportunity to see the big picture, to compare
diseases, injuries, and risk factors, and to understand in a
given place, time, and age-sex group, what are the most
important contributors to health loss.
GBD 2010 Team
486 authors from 302 institutions in 50 countries

Seven institutions: IHME as the coordinating center, the University of Queensland, the
Harvard School of Public Health, the Johns Hopkins Bloomberg School of Public Health,
the University of Tokyo, Imperial College London, and the World Health Organization.
10
Key Activities
1. Produced valid, unbiased, and comparable estimates of
prevalence of disease and injury cases or episodes and
relevant disabling sequelae at the population level for the
GBD regions
2. Produced cause-specific mortality by region that collectively
sums to all-cause mortality estimates globally
3. Revised the health state severity weight system, a highly
debated component of past GBD studies that seeks to
measure health state severity
4. Produced estimates of YLDs, YLLs, death, and DALYs for
diseases, injuries, and risk factors for 21 regions for 20 age
groups and both sexes, for 1990, 2005, and 2010
Scope of the 2010 study
• 291 diseases and injuries
• 1,160 disabling sequelae of these diseases and
injuries
• 67 risk factors
• 20 age groups, two sexes
• 187 countries
• 1990-2010
• Improved methods for the estimation of health state
severity weights
12
Study objectives
• Two major objectives
– Revising figures for 1990 given new data and
improved techniques and generating new
estimates for 2005 along with projections for
2010.
– Developing various sets of tools, each tailored to a
specific audience, to standardize and broaden the
field of burden research and analysis.
GBD advantages - 1
1. Evidence-based evaluations
– standard measures to ensure that all conditions
receive systematic, objective analysis.

2. Changing awareness and validity of different
global health conditions
– visibility and legitimacy to conditions like
depression which cause great suffering with little
associated mortality, and to conditions like road
traffic accidents, formerly outside the scope of
public health.
GBD advantages - 2
3. Cost effectiveness analysis
– allows for lives in every part of the world to be
valued equally and creates a common unit of
currency for making decisions about the costs and
benefits of various health interventions.
– using standard units of measurement, such as
disability-adjusted life years (DALYs).

4. Engaging researchers, experts, and
policymakers
Need for GBDS (2010) - 1
• While various groups have published partial
updates of GBD rankings, there has not yet
been a comprehensive and systematic
revision.
• As a result, burden estimates today contain
some outdated, and often, inconsistent
information.
Need for GBDS (2010) - 2
• A core principle of the GBD is to facilitate unbiased
comparisons across diseases, injuries, and risk
factors.
• Disability weights are the mechanism through which
time spent with a disease or injury sequela is made
commensurate with other disease and injury
sequelae and with time lost due to premature
mortality.
• Disability weights had not been empirically revised
since the GBD 1990.
Need for GBDS (2010) - 3
• Demand for more detailed information and
estimation of the burden of disease from diarrhea,
pneumonia, and meningitis broken down by specific
pathogen
• Fourth, GBD 1990 and GBD 2000 results were based
on different datasets and methods, but many users
of the information have compared the results to
examine trends for key outcomes.
• A reassessment of 1990 burden using strictly
comparable methods to the current time period was
needed
Need for GBDS (2010) - 4
• Methods innovation across the spectrum from
mortality measurement, cause of death analysis, and
meta-regression for epidemiological synthesis
provided opportunities to modernize the GBD
approach
• Combined with a global movement towards
enhanced transparency and access to data, the
empirical basis for assessing the epidemiology of
diseases is much richer than in the past
Methods
Disease and injury causes, sequelae,
and health states
• Cause list has a hierarchical structure
• At each cause level, the list provides a set of mutually
exclusive and collectively exhaustive categories
• Added causes on the basis of three criteria:
potentially large, health policy interest, feasibility of
measurement.
• Four levels of diseases and injuries and a fifth level
for sequela,
Causes, sequelae, and health states - 1
• The four levels provide different degrees of
disaggregation of the causes:
• Three level 1 causes, 21 level 2 causes, 155 level 3
causes, and 112 level 4 causes
• Of the level 3 causes 129 have no level 4 causes below
them in the hierarchy;
• At the most disaggregated level there are, therefore,
241 mutually exclusive and collectively exhaustive
causes
• The three level 1 causes are communicable, maternal,
neonatal, and nutritional disorders; non-communicable
diseases; and injuries.
Causes, sequelae, and health states - 2
• List of sequelae is designed to capture the direct
consequences of disease or injury that are not
otherwise captured elsewhere in the cause list
• Number of sequelae evaluated has increased for all
three broad cause groups, leading to 1,160 in total
• A given outcome only appears once in the cause
sequela list to avoid double counting
Causes, sequelae, and health states - 3
• For the GBD 2010 an important innovation has been to
identify which disease sequelae are common across
conditions; these common disease sequelae are called
health states

• Shorter list of health states which facilitates the
disability weight measurement component of the study
• For each of the health states, a lay description was
developed for use in the empirical assessment of
disability weights
Disease and injury ranking list
• Because the leading causes of burden tend to have some influence
on the perception of disease control priorities, the choice of
aggregation is at once important
• Provides information at the second level cause (21 causes)
• A ranking list with 176 causes selected to distinguish and cluster
conditions together that may have programmatic or public health
significance
• Results in the tables are provided for all 291 causes; the ranking list is
used only for the leading cause figures
• The 176 causes along with the excluded residual categories are
also mutually exclusive and collectively exhaustive
Risks rather than specific diseases
• An alternative way of looking at ill health is to consider risks for disease
rather than diseases themselves
• A risk is defined as the probability of an adverse health outcome or a
factor that raises this probability
• A single risk factor may underlie several diseases; e.g. unsafe sex is a risk
for HIV/AIDS, many other sexually transmitted diseases and teenage
pregnancy
• It is often more useful to direct health interventions against risks rather
than specific diseases
• DALYs have been assigned to health risks as well as diseases
Risk factor list
• Comparative Risk Assessment 2000 list of 26 risk
factors and subtracted and added risk factors based
on consideration of four criteria:
• Convincing or probable evidence on relative risks
by cause, sufficient data to estimate exposure by
region, potential size of the attributable burden
of disease, and the extent to which burden is
modifiable
• A hierarchy with three levels
Risk factor list
• Level 1 risks
– Groups of risk factors that are related by mechanism,
biology, or potential policy intervention.

• Most risks are presented at level 2, but in some
cases, such as occupational carcinogens
calculations, are done at a third level
• The category diet and physical inactivity clusters
together some key behavioral risks. This category
has been substantially expanded to include 14
components of diet.
Risk factor list
• For some of the level 1 clusters of risk factors, there are
analytical challenges for computation related to risk factors
that operate through each other.
• The problem of mediation is particularly challenging for
the physiological risks. For example, BMI operates in part
through blood pressure and cholesterol changes.
• In this analysis, we have not been able to estimate this
level 1 risk factor, so that results are presented for only 67
of the 69 risks or clusters of risks
• As with diseases and injuries, we have identified a list of
43 risk factors for use in developing tables and figures of
leading causes of risk factors.
Regions
• Basis of two criteria
– Epidemiological homogeneity and geographic contiguity

• Maximize the between-region variance in outcomes and
minimize the within-region variance in outcomes
• The process was iterative based on examination of levels
of under-5 mortality and adult mortality (as measured by
the probability of death between ages 15 and 60)
• 21 regions were made.
• For some types of analysis, created seven super-regions
• For some Bayesian estimation procedures used in the
analysis, super-regions are useful in the presence of
sparse data.
Subgroups - 1
• Early neonatal, late neonatal, post-neonatal, 1-4,
5-9, 10-14, …..70-74, 75-79, and 80+
• In the demographic analysis of age-specific
mortality, estimated deaths and death rates for
age-groups up to 100.
• Due to limitations of data at older ages, however,
80+ used as the terminal age-group for the cause of
death and disease and injury sequelae analysis.
Subgroups - 2
• All estimates are produced separately for males and females
• Not been able to include estimation of fetal deaths; this is
an important area for future quantitative work.
• A critical dimension of the GBD 2010 is the addition of
multiple time periods assessed using strictly comparable data
and methods
• For three years - 1990, 2005, and 2010.
• All results for causes of death have in the computation been
estimated for single years as well
Analytical components
1.
2.
3.
4.
5.

6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.

Covariate database
Age-specific mortality rates
Cause of death database
Estimating causes of death
Ensuring consistency between cause-specific estimates and all-cause mortality
estimates
Disease sequelae epidemiology database
Estimating disease sequela prevalence, incidence, and duration
Impairment levels
Nature and external cause of injury analysis
Disability weights for 220 health states
Comorbidity simulation
Healthy life expectancy
Computation of YLLs, YLDs, and DALYs from diseases and injuries with uncertainty
Risk factor exposure database
Estimating the prevalence of exposure
Estimating relative risks for risk-disease pairs
Theoretical minimum risk exposure
Computation of attributable burden for each risk factor with uncertainty
Covariate database
• In settings where there are sparse data or conflicting data
from multiple sources
• This estimation task is facilitated by the use of appropriate
covariates that can be used as independent variables in the
modeling process
• A database of 84 covariate topic areas and 179 variants
of the covariates has been developed for this purpose
• Missing data in covariates have been addressed using spatialtemporal regression and Gaussian process regression (GPR)
Age-specific mortality rates
• Complete re-analysis of the evidence on levels of
age-specific mortality has been undertaken for
187 countries from 1970 to 2010
• This re-analysis takes advantage of improvements
in demographic estimation methods for various
types of data on child and adult mortality
• All available data on child and adult mortality
have been combined into a time series of best
estimates
Cause of death database
• Information on causes of death comes primarily from vital
registration systems and verbal autopsy studies
• IHME has systematically created a cause of death database
spanning the period 1980 to 2010, including mapping data
across various revisions and national variants of the ICD
• Algorithms have also been developed and applied to deal with
the problem of ill-defined coding in vital registration and
verbal autopsy data where deaths are assigned to unknown
causes, intermediate causes, or immediate causes of death
rather than the underlying cause of death
Estimating causes of death
• For 133 of 235 causes, cause of death ensemble modeling
(CODEm) used as the primary strategy to develop causespecific estimates;
• Three key aspects to this approach
– To develop and test a diverse set of plausible models instead
of a single analyst-preferred model;
– To objectively test the prediction performance of these
models by holding out some of the data from the analysis
and then comparing model prediction to the held out data
– In the CODEm approach, predictions for the GBD are based
on a pool of the best models.

• For some sub-causes, such as diarrhea etiologies,
HIV/AIDS, and for some causes that rarely cause death,
alternative methods have been used
Ensuring consistency between cause-specific
estimates and all-cause mortality estimates
• Cause of death estimation proceeds in two
stages.
– First, models are developed for each cause of
death in isolation
– Second, the estimates for each cause are
combined
into estimates
for all causes
simultaneously that sum to the demographic
estimates of all-cause mortality.

• This simple algorithm leads to greater change
in the estimates for more uncertain causes
Disease sequelae epidemiology
database
• To support estimation for each region of disease
sequelae incidence, prevalence, duration, remission,
and excess mortality, a database of available
published and unpublished data has been developed
Estimating disease sequela
prevalence, incidence, and duration
• Developing estimates for disease incidence,
prevalence, duration, and excess mortality estimates
for 21 regions for three time periods by age and sex, a
meta-regression tool was used (DisMod-MR)
• Dealing with selection bias intrinsic to some data
sources, such as hospital discharges
• For some causes where more complicated models
capturing more stages of disease progression are
necessary, such as HIV, more elaborate natural history
models were used, refined, or developed.
Impairment levels
• The impairment analysis has been undertaken for
vision loss, hearing loss, anemia, infertility, heart
failure, intellectual disability, and epilepsy
• In these cases, the quality and extent of data on the
overall impairment level have been determined to
be stronger than on disease specific impairments.
Nature and external cause of injury
analysis
• Dual nature of injury assessment. Each injury can be characterized by
the nature of injury such as a femur fracture or head trauma and
the external cause of injury such as a road injury or a fall.
• To assess disability, data is required on the frequency of the nature
of injury as well as follow-up data on reduced health functioning at
certain points in time after the injury.
• capture data on the matrix between the nature of injury
incidence and the external causes leading to these injuries
• first time included survey data on the incidence of external causes
and, perhaps more importantly, multiple sources of long-term
follow-up data to estimate the fraction of individuals with each nature
of injury progressing to permanent disability
Disability weights for 220 health states
• Disability weights are the key mechanism in the GBD approach through
which disease and injury sequelae are made comparable with each
other and with time lost due to premature mortality.
• we have switched to the construct of disability weights for health
states as assessed by the general public.
• To achieve this, simple lay descriptions of the 220 unique health states
across the 1,160 disease and injury sequelae were developed.
• Disability weights were assessed using general population surveys in
five countries to capture a diverse set of views and through an
Internet survey.
• Disability weights were elicited using pairwise comparisons where the
respondent is asked to identify which health state represents a higher
level of health.
Comorbidity simulation
•

Combinations of these sequelae change the likely health experience of
the individual.

•

Comorbidity is taken into account in three stages.
– First, for each age-sex group, a micro-simulation is developed to estimate the number of
individuals who will have combinations of disease conditions.
– Second, the combined disability weight for individuals with
estimated assuming a multiplicative model.

multiple

conditions

is

– Finally, the overall individual health loss captured in the disability weight is assigned to the
different sequelae experienced by an individual proportionate to the disability weights of
individual conditions on their own.

•

The micro-simulation process is repeated for each country-age-sex-year 1000
times to propagate uncertainty in the prevalences and disability weights into
the comorbidity estimates.
Healthy life expectancy
• Healthy life expectancy is a single summary metric of population
health that is not affected by population age structure
• It can be thought of as equivalent to the expectation of life at birth
in full health given the currently observed age-specific death rates
and prevalence of disease and injury sequelae
• Used the life tables developed in component 2 for each country in
1990 and 2010, the disability weights and the output of the
comorbidity simulations to estimate healthy life expectancy for 187
countries
Computation of YLLs, YLDs, and DALYs from
diseases and injuries with uncertainty
• Each quantity is estimated with uncertainty by
generating 1,000 draws from the posterior
distribution
• Given the large number of diseases and injuries and
their sequelae, this task is computationally intensive
Risk factor exposure database
• Assesses the attributable burden of disease using
four components:
– Database on risk factor exposure,
– Estimating the prevalence of risk factor exposure by age
and sex
– Relative risk for a specific disease,
– Counterfactual distribution of risk exposure called the
theoretical minimum risk distribution.
Estimating the prevalence of exposure
• Exposure estimates have been developed for many
risks using the GBD Bayesian meta-regression tool,
DisMod-MR
• Related Bayesian methods have been used in
published analyses for the metabolic risks
• For some risks, alternative modeling strategies have
been used
Estimating relative risks for riskdisease pairs
• For each risk-disease pair, relative risks for specific
disease outcomes have been taken from published
meta-analyses or updated meta-analyses that have
been undertaken as part of this study.
• Special analyses have been undertaken for water and
sanitation, all sources of particulates less than 2·5
microns, and the components of diet
Theoretical minimum risk exposure
• The tradition in risk factor epidemiology is to assess
attributable burden by estimating a counterfactual scenario
where exposure to a risk factor in the past is set to a
counterfactual distribution
• Following the framework for comparative risk assessment,
counterfactuals are set to the lowest risk distribution that has
been observed or is theoretically possible – the theoretical
minimum risk exposure distribution (TMRED)
Computation of attributable burden for
each risk factor with uncertainty
• Risk factor burden is assessed using the distributions
of burden for each disease associated with a risk
factor
• Uncertainty from the prevalence of exposure,
relative risks, and where appropriate the TMRED, are
all propagated into the final risk factor uncertainty
Five critical choices
•
•
•
•
•

New standard reference life table
Comorbidity
Incidence versus prevalence
Discounting for time
Age-weighting
New standard reference life table - 1
• The standard reference life table for the GBD 1990 was based
on the highest observed life expectancy at the time, Japanese
females with a life expectancy at birth close to 82·5 years
• Given the progress in extending life expectancy in the last 20
years, it is appropriate to extend the normative goal for life
expectancy
• Developed a new reference standard life table by identifying
the lowest observed death rate for any age-group in countries
with more than five million in population
New standard reference life table - 2
• At the Critical Ethical Choices for DALYs meeting, there was a
consensus that using a different standard for males and
females was inappropriate
• First, the empirically observed gap in life expectancy between
males and females continues to narrow
• Within high-income countries, the gaps between male and
female life expectancies especially for the lowest mortality
communities have also narrowed
• Second, there is no reason that society should have lower
aspirations for health for males than females
Comorbidity
• A year of life lived with a disabling sequela such as
blindness should contribute the same amount to the
burden of disease regardless of whether it occurs in
someone with another condition or not.
• Critical Ethical Choices for DALYs meeting viewed that
burden should be assessed individual by individual.
Individuals with the same functional health loss should be
treated as like regardless of whether that functional health
loss came from one or several contributing conditions.
Incidence versus prevalence
•
•

•

•

•
•

Two types of DALYs were calculated in the GBD 1990: prevalence DALYs and
incidence DALYs.
Murray argued that the incidence perspective had two advantages: mortality
rates used to compute YLLs are by definition incidence rates; using incidence
for YLDs means that both non-fatal and fatal outcomes are analyzed in a
consistent fashion.
several disadvantages - individuals living with reductions in health functioning
who need health and other social services may not be reflected in burden
estimates if incidence has declined substantially.
If the incidence of ischemic heart disease is designated to start with plaque
formation or detectable myocardial ischemia will profoundly change the number
of discounted YLDs. This arbitrariness to the construct of when a disease
process starts is not present in a prevalence perspective.
incorporation of comorbidity is more straightforward in a prevalence
approach than an incidence approach.
Incidence approach - information intensive and computationally challenging.
Discounting for time
• Anand and Hanson argued that there was no intrinsic
reason to value a year of health as less important
simply because it is in the future
• One cannot expect that the value of a year of healthy
life to society will be less in the future;
• There is no expectation of declining marginal utility
of years of healthy life
• Chosen to compute the base case of DALYs for the
with a discount rate of 0%
Age-weighting
• Some have argued that deaths at particularly young ages or
even fetal death represents less of a loss of health because
individuals do not have fully developed life plans
• “the well-being of some age groups is instrumental in making
society flourish; therefore, collectively we may be more
concerned with improving health status for individuals in
these age groups”
• We recognize, however, that there is still an important set of
issues about social priority that should be attached to
preventing fetal death compared to death after a live birth
that have not been addressed here
Results overview
The disability transition: progressive shift in burden
from premature mortality to years lived with disability
1990

2010

66
Articles published
• All available data on causes of death for 187 countries from
1980 to 2010 from vital registration, verbal autopsy, mortality
surveillance, censuses, surveys, hospitals, police records, and
mortuaries
• Six different modeling strategies to estimate cause-specific
mortality trends depending on the strength of the data
Methods
• For 133 causes and three special aggregates - Cause of Death Ensemble
model (CODEm) approach, which uses four families of statistical models
testing a large set of different models using different permutations of
covariates. Model ensembles were developed from these component
models. model performance was assessed with rigorous out-of-sample
testing of prediction error and the validity of 95% UIs.
• For 13 causes with low observed numbers of deaths - negative binomial
models with plausible covariates.
• For 27 causes for which death is rare - modelled the higher level and then
allocated deaths across component causes proportionately, estimated
from all available data in the database.
• For selected causes (African trypanosomiasis, congenital syphilis,
whooping cough, measles, typhoid and parathyroid, leishmaniasis, acute
hepatitis E, and HIV/AIDS), - natural history models based on information
on incidence, prevalence, and case-fatality.
Methods
• Separately estimated cause fractions by aetiology for
diarrhoea, lower respiratory infections, and
meningitis, as well as disaggregations by subcause
for chronic kidney disease, maternal disorders,
cirrhosis, and liver cancer.
• For deaths due to collective violence and natural
disasters, we used mortality shock regressions
Sources of data
• Identified 25 variants of cause of death reporting lists in use
from 1980 to 2010 across all sources of vital registration.
• Identified 2798 site-years of data from 130 countries between
1980 and 2010
• Verbal autopsy data were collected through sample
registration systems, demographic surveillance systems, or
surveys. identified 486 site-years of published and
unpublished verbal autopsy data across 66 countries, of which
10% were nationally representative.
• We identified 2715 site-years of cancer registry data across 93
countries.
Sources of data
• Police and crime reports are an important source of information for some
types of injuries, notably road injuries and interpersonal violence.
• The police reports used in this analysis were obtained from published
studies, national agencies, and institutional surveys such as the UN Crime
Trends survey and the WHO Global Status Report on Road Safety Survey.
• 1129 site-years of police reports from 122 countries
• 32 site-years of burial and mortuary data in 11 countries
• identified 1557 survey years with sibling history data, and a further 52
household survey years or census years of data covering 61 countries.
• We also identified 56 surveys or censuses with information on injury
mortality across 65 survey years or census years.
• identified eight countries with nationally representative maternal
mortality surveillance systems covering 83 site-years
• 21 site-years of data based on deaths in health facilities.
Results
• Population growth, increased average age of the
world’s population, and largely decreasing agespecific, sex-specific, and cause-specific death rates
combine to drive a broad shift from communicable,
maternal, neonatal, and nutritional causes towards
non-communicable diseases.
http://www.healthmetricsandevaluation.org/gbd/visualizations/gbd-2010-leading-causesand-risks-region-heat-map
Other articles
Thank you
Basics
What is DALY?
DALY= YLL + YLD
Years of Lost Life (due to premature mortality)
Years Lost to Disability (due to injury or illness)

•

The DALY is the internationally-accepted measure of death and disability and is increasingly cited as
a powerful tool for decision makers in international health.

•

It is the sum of the number of years lost to premature mortality and the number of productive
years lost to disability. One DALY is equivalent to one lost year of healthy life.

•

Calculation of DALYs is based on the assumption that everyone in the world has a right to the best
life expectancy in the world. The only differences in the rating of a death or disability should be due
to age and sex and not to income, culture, location or social class.

•

It’s attractiveness lies in the fact that it combines information about mortality and morbidity in a
single number. DALY allows the losses due to disability and the losses due to premature death to be
expressed in the same unit. Hence, DALYs facilitate comparisons of different health states or health
outcomes.
What is DALY?
Sample “DALY” calculations for a disease
Example A:
• 100,000 children are stricken for 1 week with a disease with a disability weighting*
of 0.3; 2% die at 1 year old.
• DALYs
= YLL + YLD
= (2000 X 80) + (100,000 X 7/365 X 0.3)
= 160,000 + 575
= 160,600

Example B:
• 100,000 adults are stricken for 2 years with a disease with disability weighting* of
0.6; 20% die at age 80 years.
•

DALYs

= YLL + YLD
= (20,000 x 2) + (100,000 x 2 x 0.6)
= 40,000 + 120,000
= 160,000

*There are STANDARDISED DISABILITY RATINGS for various conditions
e.g deafness= 0.33, Down syndrome=0.5, Diarrhoea=0.12
Why are DALYs important?
DALYs attempt to provide an appropriate, balanced attention to the effects of
non-fatal as well as fatal diseases on overall health. In the absence of such
assessments, conditions which cause decrements in function but not
mortality tend to be neglected.
DALYs help to inform debates on priorities for health service delivery, research
and planning. For example, DALYs can be used to:

• Compare the health of one population with another – and allow decision
makers to focus on health systems with the worst performance
• Compare the health of the same population at different points in time

• Compare the health of subgroups within a population - to identify health
inequalities
Disability weights, discounting and age
weighting
• Egalitarian principles are explicitly built into the DALY metric,
and the global burden of disease studies apply these to all
regions of the world.
• The studies use the same "ideal" life expectancy for all
population subgroups and exclude all non-health
characteristics (such as race, socioeconomic status or
occupation) apart from age and sex from consideration in
calculating lost years of healthy life.
• Most importantly, they use the same "disability weight" for
everyone living a year in a specified health state.
Disability weights, discounting and age
weighting
• Disability weights
– A disability weight is a weight factor that reflects the severity of
the disease on a scale from 0 (perfect health) to 1 (equivalent to
death). Years Lost due to Disability (YLD) are calculated by
multiplying the incident cases by duration and disability weight
for the condition
• Age weighting and discounting
– 3% discounting and non-uniform age weighting was used in the
original GBD 1990 study. These adjustments result in less weight
given to years lived at young and older ages
– The GBD 2001-2 study used 3% discounting but uniform age
weighting. GBD 2004 update used the original 3% discounting
and non-uniform age weighting
DALYs: Social values
1. How to compare years lost due to death with years
lived in poor health?

essential

 Disability Weights values between 0 and 1
2. Value of health year of life equal at all ages?
 age weights
3. Value of future years of life?
 discounting

http://www.who.int/healthinfo/global_burden_disease/daly_disability_weight/en/index.html

optional
Disability Weights
 In GBD, non-fatal consequences of diseases and injuries understood
as transitions through different ‘health states’
 YLD calculation requires aggregate assessments of the overall
decrements in health associated with particular health states 
disability weights
 DWs are measures of overall levels of health rather than contribution
of health to overall welfare

GBD1990 :
 DW elicited from panel of health professionals following explicit
protocol evaluating 22 indicator conditions in an intensive group
exercise with ‘deliberative phase’ using person trade-off (PTO)
method. Responses averaged across participants
New Disability Weight Project

The new DW will have a greater emphasis on paired
comparisons, anchored by time trade-off methods.
It also aims to engage members of the general community
(including those in developing countries) to a greater degree.

The DW project is being carried out in two stages:

 a community household survey in selected regions, and
 an online open-access survey.
New Disability Weight Project
1.

Community surveys
•
•
•

2.

(Pemba, Bangladesh, Indonesia, Peru, USA)
Cultural/SES diversity >> representativeness
Paired comparisons

Internet survey
•
•
•
•

the primary source of data for the final DW
will include random(?) selections from all ~230 sequelae
open to all interested in participating (open-access
includes a variety of measurements paired comparisons, ranking,
visual analogue scale, time trade-off, population equivalence) to
anchor the scale (paired comparisons, ranking, visual analogue scale,
time trade-off, population equivalence)

http://gbdsurvey.org/
Discounting
Discounting common practice in economic analyses.

Assumes that individuals value their health more now
than at some point in the future. So the further in the
future health loss occurs the more they are
discounted.
•

GBD 1990 used 3% discounting
Discounting
Why discount?
1.
Consistency with cost-effectiveness analyses
2.
Prevent giving ‘excessive’ weight to deaths at younger ages
Age weighting


Used to reflect a social preference that values a year
lived by young adult more highly than that of young
children or the elderly.

• Eg. An Australian survey found that respondents considered saving four
20-year olds as important as saving ten 60-year olds (Nord et al, 1996
and 1998)


Not related to productivity but ‘social’ role in caring
for the young and old
Age weighting
Arguments against:

1.6

•

•

relative value

1.2

Unacceptable on
equity grounds
Does not reflect actual
social values

0.8

But
• Everyone potentially
lives through every age
 not inequitable

0.4

0.0
0

20

40

60

age

80

100
Source: Murray and Lopez, 2006. Chapter 5
Terminology used GBDS
• Disability : Disability refers to any short-term or long-term health loss.
• Sequelae :In the GBD 2010 cause list there are 291 diseases and injuries,
of which 289 cause disability. In total, 1160 sequelae of these diseases and
injuries. For example, diabetic neuropathy is a sequela of diabetes
mellitus. To avoid double counting, a sequela can only be counted in the
cause list once even if the same outcome might be caused by more than
one disease.
• Health state :Across the 1160 sequelae, 220 unique health states. For
example, both malaria and hookworm have mild anaemia as a sequela.
Mild anaemia is a unique health state. The list of unique health states
serves two purposes: to allow assessment of the total burden of some
health states such as anaemia across various causes, and to simplify the
task of measuring disability weights for sequelae.
Terminology used GBDS
• Disability weights : A quantification of the severity of health loss
associated with the 220 unique health states on a scale from 0 to 1,
when 0 is commensurate with perfect health and 1 is
commensurate with death.
• Years lived with disability (YLDs) : For the GBD 2010, YLDs per
person from a sequela are equal to the prevalence of the sequela
multiplied by the disability weight for the health state associated
with that sequela. YLDs for a disease or injury are the sum of the
YLDs for each sequela associated with the disease or injury.
• Impairments : In the GBD 2010 we estimated the prevalence and
burden of several unique health states that are sequelae for
multiple diseases including anaemia, heart failure, vision loss,
seizures, hearing loss, infertility, and intellectual disability. These
are referred to as impairments.

More Related Content

What's hot

International health regulation
International health regulationInternational health regulation
International health regulationVenu Bolisetti
 
Introduction to epidemiology and it's measurements
Introduction to epidemiology and it's measurementsIntroduction to epidemiology and it's measurements
Introduction to epidemiology and it's measurementswrigveda
 
One Health: Concept and applications
One Health: Concept and applicationsOne Health: Concept and applications
One Health: Concept and applicationsILRI
 
International health agencies & current global health issues
International health agencies & current global health issuesInternational health agencies & current global health issues
International health agencies & current global health issuesAmrut Swami
 
Triad wheel web new
Triad wheel web newTriad wheel web new
Triad wheel web newarchana353
 
International health Regulations
International health RegulationsInternational health Regulations
International health Regulationsbitan sengupta
 
DYNAMICS OF DISEASE & DISEASE TRANSMISSION.ppt
DYNAMICS OF DISEASE & DISEASE TRANSMISSION.pptDYNAMICS OF DISEASE & DISEASE TRANSMISSION.ppt
DYNAMICS OF DISEASE & DISEASE TRANSMISSION.pptVanithadurai
 
Indian Healthcare System An Overiew
Indian Healthcare System An OveriewIndian Healthcare System An Overiew
Indian Healthcare System An Overiewdrdivyahm
 
International Health Regulations & PHEIC
International Health Regulations & PHEICInternational Health Regulations & PHEIC
International Health Regulations & PHEICVignesh Loganathan
 
Concept of efficiency,effectivenes,Role of health care financing
Concept of efficiency,effectivenes,Role of health care financingConcept of efficiency,effectivenes,Role of health care financing
Concept of efficiency,effectivenes,Role of health care financingSahdev Bishnoi
 
Health systems strengthening
Health systems strengtheningHealth systems strengthening
Health systems strengtheningJuan Seclen
 
Presentation1 phc
Presentation1 phcPresentation1 phc
Presentation1 phclakhanpal1
 
History of public health
History of public healthHistory of public health
History of public healthSuman Saurabh
 

What's hot (20)

International health regulation
International health regulationInternational health regulation
International health regulation
 
Introduction to epidemiology and it's measurements
Introduction to epidemiology and it's measurementsIntroduction to epidemiology and it's measurements
Introduction to epidemiology and it's measurements
 
Rise of public health
Rise of public healthRise of public health
Rise of public health
 
Burden of Disease using DALYs
Burden of Disease using DALYsBurden of Disease using DALYs
Burden of Disease using DALYs
 
One Health: Concept and applications
One Health: Concept and applicationsOne Health: Concept and applications
One Health: Concept and applications
 
International health agencies & current global health issues
International health agencies & current global health issuesInternational health agencies & current global health issues
International health agencies & current global health issues
 
Triad wheel web new
Triad wheel web newTriad wheel web new
Triad wheel web new
 
International health Regulations
International health RegulationsInternational health Regulations
International health Regulations
 
International health new
International health newInternational health new
International health new
 
Global Burden of Disease - Pakistan Presentation
Global Burden of Disease - Pakistan PresentationGlobal Burden of Disease - Pakistan Presentation
Global Burden of Disease - Pakistan Presentation
 
DYNAMICS OF DISEASE & DISEASE TRANSMISSION.ppt
DYNAMICS OF DISEASE & DISEASE TRANSMISSION.pptDYNAMICS OF DISEASE & DISEASE TRANSMISSION.ppt
DYNAMICS OF DISEASE & DISEASE TRANSMISSION.ppt
 
Indian Healthcare System An Overiew
Indian Healthcare System An OveriewIndian Healthcare System An Overiew
Indian Healthcare System An Overiew
 
Dalys
DalysDalys
Dalys
 
International Health Regulations & PHEIC
International Health Regulations & PHEICInternational Health Regulations & PHEIC
International Health Regulations & PHEIC
 
Concept of efficiency,effectivenes,Role of health care financing
Concept of efficiency,effectivenes,Role of health care financingConcept of efficiency,effectivenes,Role of health care financing
Concept of efficiency,effectivenes,Role of health care financing
 
Health systems strengthening
Health systems strengtheningHealth systems strengthening
Health systems strengthening
 
What is a health system?
What is a health system?What is a health system?
What is a health system?
 
Presentation1 phc
Presentation1 phcPresentation1 phc
Presentation1 phc
 
Global health
Global healthGlobal health
Global health
 
History of public health
History of public healthHistory of public health
History of public health
 

Viewers also liked

Global warming impact on health
Global warming impact on healthGlobal warming impact on health
Global warming impact on healthGulrukh Hashmi
 
The Effects of Global Warming on Health
The Effects of Global Warming on HealthThe Effects of Global Warming on Health
The Effects of Global Warming on HealthMomina Mateen
 
Global health risks_report_figures
Global health risks_report_figuresGlobal health risks_report_figures
Global health risks_report_figureskaminandos
 
Presentation1 (2)
Presentation1 (2)Presentation1 (2)
Presentation1 (2)Simer Basra
 
Global Burden of Disease - Big Data in Global Health
Global Burden of Disease - Big Data in Global HealthGlobal Burden of Disease - Big Data in Global Health
Global Burden of Disease - Big Data in Global HealthPeter Speyer
 
National mental health program
National mental health programNational mental health program
National mental health programJORRY POULOSE
 
National AIDS control program
National AIDS control programNational AIDS control program
National AIDS control programmigom doley
 
Sjogren’s syndrome
Sjogren’s syndromeSjogren’s syndrome
Sjogren’s syndromeWaseem Khan
 
Health Management Information & Evaluation system
Health Management Information & Evaluation systemHealth Management Information & Evaluation system
Health Management Information & Evaluation systemSujata Mohapatra
 
Homeopathy – A Holistic Approach to Your Disease By Dr. Shelly Sharma
Homeopathy – A Holistic Approach to Your Disease By  Dr. Shelly SharmaHomeopathy – A Holistic Approach to Your Disease By  Dr. Shelly Sharma
Homeopathy – A Holistic Approach to Your Disease By Dr. Shelly SharmaHealth Education Library for People
 
National cancer control program
National cancer control programNational cancer control program
National cancer control programJORRY POULOSE
 
Global health introduction
Global health introductionGlobal health introduction
Global health introductionDrZahid Khan
 
National cancercontrol programme power point presentation
National cancercontrol programme power point presentationNational cancercontrol programme power point presentation
National cancercontrol programme power point presentationSabeena Sasidharan
 
National mental-health-programme-community-health-nursing-ppt
National mental-health-programme-community-health-nursing-pptNational mental-health-programme-community-health-nursing-ppt
National mental-health-programme-community-health-nursing-pptSudeep Nath
 
enamel careis by khubaib ahmed shaikh
enamel careis by khubaib ahmed shaikhenamel careis by khubaib ahmed shaikh
enamel careis by khubaib ahmed shaikhkhubaib ahmed shaikh
 
Increasing Burden of NCD in Malaysia: Challenges in resource allocation
Increasing Burden of NCD in Malaysia: Challenges in resource allocationIncreasing Burden of NCD in Malaysia: Challenges in resource allocation
Increasing Burden of NCD in Malaysia: Challenges in resource allocationFeisul Mustapha
 
NATIONAL AIDS CONTROL PROGRAMME IN INDIA
NATIONAL AIDS CONTROL PROGRAMME IN INDIANATIONAL AIDS CONTROL PROGRAMME IN INDIA
NATIONAL AIDS CONTROL PROGRAMME IN INDIABharat Masal
 

Viewers also liked (20)

Global warming impact on health
Global warming impact on healthGlobal warming impact on health
Global warming impact on health
 
The Effects of Global Warming on Health
The Effects of Global Warming on HealthThe Effects of Global Warming on Health
The Effects of Global Warming on Health
 
Global health risks_report_figures
Global health risks_report_figuresGlobal health risks_report_figures
Global health risks_report_figures
 
Presentation1 (2)
Presentation1 (2)Presentation1 (2)
Presentation1 (2)
 
Healthy life expectancy for 187 countries, 1990 - 2010
Healthy life expectancy for 187 countries, 1990 - 2010Healthy life expectancy for 187 countries, 1990 - 2010
Healthy life expectancy for 187 countries, 1990 - 2010
 
Global Burden of Disease - Big Data in Global Health
Global Burden of Disease - Big Data in Global HealthGlobal Burden of Disease - Big Data in Global Health
Global Burden of Disease - Big Data in Global Health
 
National mental health program
National mental health programNational mental health program
National mental health program
 
National AIDS control program
National AIDS control programNational AIDS control program
National AIDS control program
 
National Cancer
National CancerNational Cancer
National Cancer
 
Sjogren’s syndrome
Sjogren’s syndromeSjogren’s syndrome
Sjogren’s syndrome
 
Health Management Information & Evaluation system
Health Management Information & Evaluation systemHealth Management Information & Evaluation system
Health Management Information & Evaluation system
 
Homeopathy – A Holistic Approach to Your Disease By Dr. Shelly Sharma
Homeopathy – A Holistic Approach to Your Disease By  Dr. Shelly SharmaHomeopathy – A Holistic Approach to Your Disease By  Dr. Shelly Sharma
Homeopathy – A Holistic Approach to Your Disease By Dr. Shelly Sharma
 
A comparative risk assessment of burden of disease and injury attributable to...
A comparative risk assessment of burden of disease and injury attributable to...A comparative risk assessment of burden of disease and injury attributable to...
A comparative risk assessment of burden of disease and injury attributable to...
 
National cancer control program
National cancer control programNational cancer control program
National cancer control program
 
Global health introduction
Global health introductionGlobal health introduction
Global health introduction
 
National cancercontrol programme power point presentation
National cancercontrol programme power point presentationNational cancercontrol programme power point presentation
National cancercontrol programme power point presentation
 
National mental-health-programme-community-health-nursing-ppt
National mental-health-programme-community-health-nursing-pptNational mental-health-programme-community-health-nursing-ppt
National mental-health-programme-community-health-nursing-ppt
 
enamel careis by khubaib ahmed shaikh
enamel careis by khubaib ahmed shaikhenamel careis by khubaib ahmed shaikh
enamel careis by khubaib ahmed shaikh
 
Increasing Burden of NCD in Malaysia: Challenges in resource allocation
Increasing Burden of NCD in Malaysia: Challenges in resource allocationIncreasing Burden of NCD in Malaysia: Challenges in resource allocation
Increasing Burden of NCD in Malaysia: Challenges in resource allocation
 
NATIONAL AIDS CONTROL PROGRAMME IN INDIA
NATIONAL AIDS CONTROL PROGRAMME IN INDIANATIONAL AIDS CONTROL PROGRAMME IN INDIA
NATIONAL AIDS CONTROL PROGRAMME IN INDIA
 

Similar to Global Burden of Disease Study - 2010

International health
International  healthInternational  health
International healthNursing Path
 
Practicum_Poster_LizWartella_ew2480
Practicum_Poster_LizWartella_ew2480Practicum_Poster_LizWartella_ew2480
Practicum_Poster_LizWartella_ew2480Liz Wartella
 
International health
International  healthInternational  health
International healthNursing Path
 
International health
International  healthInternational  health
International healthNursing Path
 
Risk factor surveillance of Non-communicable diseases
Risk factor surveillance of Non-communicable diseasesRisk factor surveillance of Non-communicable diseases
Risk factor surveillance of Non-communicable diseasesVineetha K
 
A Brief Introduction to Epidemiology
A Brief Introduction to EpidemiologyA Brief Introduction to Epidemiology
A Brief Introduction to EpidemiologyPaul Barratt
 
Lowitja institute
Lowitja instituteLowitja institute
Lowitja instituteBiteTheDust
 
7 community diagnosis
7 community diagnosis7 community diagnosis
7 community diagnosisAnup Kharde
 
Cost utility analysis
Cost utility analysisCost utility analysis
Cost utility analysisARUNAYESUDAS
 
Public Health (Part 1) Lecture B
Public Health (Part 1) Lecture BPublic Health (Part 1) Lecture B
Public Health (Part 1) Lecture BCMDLearning
 
2019 global health_script_challenges
2019 global health_script_challenges2019 global health_script_challenges
2019 global health_script_challengesYuZheNg49
 
Population Health - Data & Visualizations for Decision Making
Population Health - Data & Visualizations for Decision MakingPopulation Health - Data & Visualizations for Decision Making
Population Health - Data & Visualizations for Decision MakingPeter Speyer
 
7- Global Health Policies Programs 2020.pptx
7- Global Health Policies Programs 2020.pptx7- Global Health Policies Programs 2020.pptx
7- Global Health Policies Programs 2020.pptxkevincorrellkeith
 
Pro instrument and its types - pubrica
Pro instrument and its types  - pubricaPro instrument and its types  - pubrica
Pro instrument and its types - pubricaPubrica
 

Similar to Global Burden of Disease Study - 2010 (20)

GBD 2023.pptx
GBD 2023.pptxGBD 2023.pptx
GBD 2023.pptx
 
GBD.pptx
GBD.pptxGBD.pptx
GBD.pptx
 
International health
International  healthInternational  health
International health
 
Practicum_Poster_LizWartella_ew2480
Practicum_Poster_LizWartella_ew2480Practicum_Poster_LizWartella_ew2480
Practicum_Poster_LizWartella_ew2480
 
International health
International  healthInternational  health
International health
 
International health
International  healthInternational  health
International health
 
Risk factor surveillance of Non-communicable diseases
Risk factor surveillance of Non-communicable diseasesRisk factor surveillance of Non-communicable diseases
Risk factor surveillance of Non-communicable diseases
 
A Brief Introduction to Epidemiology
A Brief Introduction to EpidemiologyA Brief Introduction to Epidemiology
A Brief Introduction to Epidemiology
 
Lowitja institute
Lowitja instituteLowitja institute
Lowitja institute
 
7 community diagnosis
7 community diagnosis7 community diagnosis
7 community diagnosis
 
Cost utility analysis
Cost utility analysisCost utility analysis
Cost utility analysis
 
Public Health (Part 1) Lecture B
Public Health (Part 1) Lecture BPublic Health (Part 1) Lecture B
Public Health (Part 1) Lecture B
 
2019 global health_script_challenges
2019 global health_script_challenges2019 global health_script_challenges
2019 global health_script_challenges
 
Population Health - Data & Visualizations for Decision Making
Population Health - Data & Visualizations for Decision MakingPopulation Health - Data & Visualizations for Decision Making
Population Health - Data & Visualizations for Decision Making
 
slidebank331
slidebank331slidebank331
slidebank331
 
slidebank331slide
slidebank331slideslidebank331slide
slidebank331slide
 
7- Global Health Policies Programs 2020.pptx
7- Global Health Policies Programs 2020.pptx7- Global Health Policies Programs 2020.pptx
7- Global Health Policies Programs 2020.pptx
 
Health service determinants
Health service determinantsHealth service determinants
Health service determinants
 
The EQ-5D and Its Use Internationally
The EQ-5D and Its Use InternationallyThe EQ-5D and Its Use Internationally
The EQ-5D and Its Use Internationally
 
Pro instrument and its types - pubrica
Pro instrument and its types  - pubricaPro instrument and its types  - pubrica
Pro instrument and its types - pubrica
 

More from Rizwan S A

Introduction to scoping reviews
Introduction to scoping reviewsIntroduction to scoping reviews
Introduction to scoping reviewsRizwan S A
 
Sources of demographic data 2019
Sources of demographic data 2019Sources of demographic data 2019
Sources of demographic data 2019Rizwan S A
 
Effect sizes in meta-analysis
Effect sizes in meta-analysisEffect sizes in meta-analysis
Effect sizes in meta-analysisRizwan S A
 
Presenting the results of meta-analysis
Presenting the results of meta-analysisPresenting the results of meta-analysis
Presenting the results of meta-analysisRizwan S A
 
Heterogeneity in meta-analysis
Heterogeneity in meta-analysisHeterogeneity in meta-analysis
Heterogeneity in meta-analysisRizwan S A
 
Overview of the systematic review process
Overview of the systematic review processOverview of the systematic review process
Overview of the systematic review processRizwan S A
 
Biases in meta-analysis
Biases in meta-analysisBiases in meta-analysis
Biases in meta-analysisRizwan S A
 
Moderator analysis in meta-analysis
Moderator analysis in meta-analysisModerator analysis in meta-analysis
Moderator analysis in meta-analysisRizwan S A
 
Fixed-effect and random-effects models in meta-analysis
Fixed-effect and random-effects models in meta-analysisFixed-effect and random-effects models in meta-analysis
Fixed-effect and random-effects models in meta-analysisRizwan S A
 
Inverse variance method of meta-analysis and Cochran's Q
Inverse variance method of meta-analysis and Cochran's QInverse variance method of meta-analysis and Cochran's Q
Inverse variance method of meta-analysis and Cochran's QRizwan S A
 
Data extraction/coding and database structure in meta-analysis
Data extraction/coding and database structure in meta-analysisData extraction/coding and database structure in meta-analysis
Data extraction/coding and database structure in meta-analysisRizwan S A
 
Introduction & rationale for meta-analysis
Introduction & rationale for meta-analysisIntroduction & rationale for meta-analysis
Introduction & rationale for meta-analysisRizwan S A
 
Types of correlation coefficients
Types of correlation coefficientsTypes of correlation coefficients
Types of correlation coefficientsRizwan S A
 
Checking for normality (Normal distribution)
Checking for normality (Normal distribution)Checking for normality (Normal distribution)
Checking for normality (Normal distribution)Rizwan S A
 
Analysis of small datasets
Analysis of small datasetsAnalysis of small datasets
Analysis of small datasetsRizwan S A
 
A introduction to non-parametric tests
A introduction to non-parametric testsA introduction to non-parametric tests
A introduction to non-parametric testsRizwan S A
 
Kruskal Wallis test, Friedman test, Spearman Correlation
Kruskal Wallis test, Friedman test, Spearman CorrelationKruskal Wallis test, Friedman test, Spearman Correlation
Kruskal Wallis test, Friedman test, Spearman CorrelationRizwan S A
 
Kolmogorov Smirnov good-of-fit test
Kolmogorov Smirnov good-of-fit testKolmogorov Smirnov good-of-fit test
Kolmogorov Smirnov good-of-fit testRizwan S A
 
Mantel Haenszel methods in epidemiology (Stratification)
Mantel Haenszel methods in epidemiology (Stratification) Mantel Haenszel methods in epidemiology (Stratification)
Mantel Haenszel methods in epidemiology (Stratification) Rizwan S A
 
Use of checklists in critical appraisal of health literature
Use of checklists in critical appraisal of health literatureUse of checklists in critical appraisal of health literature
Use of checklists in critical appraisal of health literatureRizwan S A
 

More from Rizwan S A (20)

Introduction to scoping reviews
Introduction to scoping reviewsIntroduction to scoping reviews
Introduction to scoping reviews
 
Sources of demographic data 2019
Sources of demographic data 2019Sources of demographic data 2019
Sources of demographic data 2019
 
Effect sizes in meta-analysis
Effect sizes in meta-analysisEffect sizes in meta-analysis
Effect sizes in meta-analysis
 
Presenting the results of meta-analysis
Presenting the results of meta-analysisPresenting the results of meta-analysis
Presenting the results of meta-analysis
 
Heterogeneity in meta-analysis
Heterogeneity in meta-analysisHeterogeneity in meta-analysis
Heterogeneity in meta-analysis
 
Overview of the systematic review process
Overview of the systematic review processOverview of the systematic review process
Overview of the systematic review process
 
Biases in meta-analysis
Biases in meta-analysisBiases in meta-analysis
Biases in meta-analysis
 
Moderator analysis in meta-analysis
Moderator analysis in meta-analysisModerator analysis in meta-analysis
Moderator analysis in meta-analysis
 
Fixed-effect and random-effects models in meta-analysis
Fixed-effect and random-effects models in meta-analysisFixed-effect and random-effects models in meta-analysis
Fixed-effect and random-effects models in meta-analysis
 
Inverse variance method of meta-analysis and Cochran's Q
Inverse variance method of meta-analysis and Cochran's QInverse variance method of meta-analysis and Cochran's Q
Inverse variance method of meta-analysis and Cochran's Q
 
Data extraction/coding and database structure in meta-analysis
Data extraction/coding and database structure in meta-analysisData extraction/coding and database structure in meta-analysis
Data extraction/coding and database structure in meta-analysis
 
Introduction & rationale for meta-analysis
Introduction & rationale for meta-analysisIntroduction & rationale for meta-analysis
Introduction & rationale for meta-analysis
 
Types of correlation coefficients
Types of correlation coefficientsTypes of correlation coefficients
Types of correlation coefficients
 
Checking for normality (Normal distribution)
Checking for normality (Normal distribution)Checking for normality (Normal distribution)
Checking for normality (Normal distribution)
 
Analysis of small datasets
Analysis of small datasetsAnalysis of small datasets
Analysis of small datasets
 
A introduction to non-parametric tests
A introduction to non-parametric testsA introduction to non-parametric tests
A introduction to non-parametric tests
 
Kruskal Wallis test, Friedman test, Spearman Correlation
Kruskal Wallis test, Friedman test, Spearman CorrelationKruskal Wallis test, Friedman test, Spearman Correlation
Kruskal Wallis test, Friedman test, Spearman Correlation
 
Kolmogorov Smirnov good-of-fit test
Kolmogorov Smirnov good-of-fit testKolmogorov Smirnov good-of-fit test
Kolmogorov Smirnov good-of-fit test
 
Mantel Haenszel methods in epidemiology (Stratification)
Mantel Haenszel methods in epidemiology (Stratification) Mantel Haenszel methods in epidemiology (Stratification)
Mantel Haenszel methods in epidemiology (Stratification)
 
Use of checklists in critical appraisal of health literature
Use of checklists in critical appraisal of health literatureUse of checklists in critical appraisal of health literature
Use of checklists in critical appraisal of health literature
 

Recently uploaded

Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...
Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...
Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...narwatsonia7
 
Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...
Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...
Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...Nehru place Escorts
 
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment BookingCall Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment BookingNehru place Escorts
 
Call Girl Service Bidadi - For 7001305949 Cheap & Best with original Photos
Call Girl Service Bidadi - For 7001305949 Cheap & Best with original PhotosCall Girl Service Bidadi - For 7001305949 Cheap & Best with original Photos
Call Girl Service Bidadi - For 7001305949 Cheap & Best with original Photosnarwatsonia7
 
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...narwatsonia7
 
call girls in munirka DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️
call girls in munirka  DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️call girls in munirka  DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️
call girls in munirka DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️saminamagar
 
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...narwatsonia7
 
97111 47426 Call Girls In Delhi MUNIRKAA
97111 47426 Call Girls In Delhi MUNIRKAA97111 47426 Call Girls In Delhi MUNIRKAA
97111 47426 Call Girls In Delhi MUNIRKAAjennyeacort
 
Call Girl Koramangala | 7001305949 At Low Cost Cash Payment Booking
Call Girl Koramangala | 7001305949 At Low Cost Cash Payment BookingCall Girl Koramangala | 7001305949 At Low Cost Cash Payment Booking
Call Girl Koramangala | 7001305949 At Low Cost Cash Payment Bookingnarwatsonia7
 
Mumbai Call Girls Service 9910780858 Real Russian Girls Looking Models
Mumbai Call Girls Service 9910780858 Real Russian Girls Looking ModelsMumbai Call Girls Service 9910780858 Real Russian Girls Looking Models
Mumbai Call Girls Service 9910780858 Real Russian Girls Looking Modelssonalikaur4
 
Call Girls Hosur Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hosur Just Call 7001305949 Top Class Call Girl Service AvailableCall Girls Hosur Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hosur Just Call 7001305949 Top Class Call Girl Service Availablenarwatsonia7
 
Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...
Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...
Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...narwatsonia7
 
High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...
High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...
High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...narwatsonia7
 
Hemostasis Physiology and Clinical correlations by Dr Faiza.pdf
Hemostasis Physiology and Clinical correlations by Dr Faiza.pdfHemostasis Physiology and Clinical correlations by Dr Faiza.pdf
Hemostasis Physiology and Clinical correlations by Dr Faiza.pdfMedicoseAcademics
 
Call Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service AvailableCall Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service Availablenarwatsonia7
 
Hematology and Immunology - Leukocytes Functions
Hematology and Immunology - Leukocytes FunctionsHematology and Immunology - Leukocytes Functions
Hematology and Immunology - Leukocytes FunctionsMedicoseAcademics
 
9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi Ncr
9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi Ncr9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi Ncr
9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi NcrDelhi Call Girls
 
Glomerular Filtration and determinants of glomerular filtration .pptx
Glomerular Filtration and  determinants of glomerular filtration .pptxGlomerular Filtration and  determinants of glomerular filtration .pptx
Glomerular Filtration and determinants of glomerular filtration .pptxDr.Nusrat Tariq
 
Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...
Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...
Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...Ahmedabad Escorts
 
Call Girls Viman Nagar 7001305949 All Area Service COD available Any Time
Call Girls Viman Nagar 7001305949 All Area Service COD available Any TimeCall Girls Viman Nagar 7001305949 All Area Service COD available Any Time
Call Girls Viman Nagar 7001305949 All Area Service COD available Any Timevijaych2041
 

Recently uploaded (20)

Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...
Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...
Russian Call Girls Gunjur Mugalur Road : 7001305949 High Profile Model Escort...
 
Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...
Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...
Call Girls Service in Virugambakkam - 7001305949 | 24x7 Service Available Nea...
 
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment BookingCall Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
 
Call Girl Service Bidadi - For 7001305949 Cheap & Best with original Photos
Call Girl Service Bidadi - For 7001305949 Cheap & Best with original PhotosCall Girl Service Bidadi - For 7001305949 Cheap & Best with original Photos
Call Girl Service Bidadi - For 7001305949 Cheap & Best with original Photos
 
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
 
call girls in munirka DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️
call girls in munirka  DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️call girls in munirka  DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️
call girls in munirka DELHI 🔝 >༒9540349809 🔝 genuine Escort Service 🔝✔️✔️
 
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
 
97111 47426 Call Girls In Delhi MUNIRKAA
97111 47426 Call Girls In Delhi MUNIRKAA97111 47426 Call Girls In Delhi MUNIRKAA
97111 47426 Call Girls In Delhi MUNIRKAA
 
Call Girl Koramangala | 7001305949 At Low Cost Cash Payment Booking
Call Girl Koramangala | 7001305949 At Low Cost Cash Payment BookingCall Girl Koramangala | 7001305949 At Low Cost Cash Payment Booking
Call Girl Koramangala | 7001305949 At Low Cost Cash Payment Booking
 
Mumbai Call Girls Service 9910780858 Real Russian Girls Looking Models
Mumbai Call Girls Service 9910780858 Real Russian Girls Looking ModelsMumbai Call Girls Service 9910780858 Real Russian Girls Looking Models
Mumbai Call Girls Service 9910780858 Real Russian Girls Looking Models
 
Call Girls Hosur Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hosur Just Call 7001305949 Top Class Call Girl Service AvailableCall Girls Hosur Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hosur Just Call 7001305949 Top Class Call Girl Service Available
 
Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...
Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...
Call Girls Frazer Town Just Call 7001305949 Top Class Call Girl Service Avail...
 
High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...
High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...
High Profile Call Girls Kodigehalli - 7001305949 Escorts Service with Real Ph...
 
Hemostasis Physiology and Clinical correlations by Dr Faiza.pdf
Hemostasis Physiology and Clinical correlations by Dr Faiza.pdfHemostasis Physiology and Clinical correlations by Dr Faiza.pdf
Hemostasis Physiology and Clinical correlations by Dr Faiza.pdf
 
Call Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service AvailableCall Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Jayanagar Just Call 7001305949 Top Class Call Girl Service Available
 
Hematology and Immunology - Leukocytes Functions
Hematology and Immunology - Leukocytes FunctionsHematology and Immunology - Leukocytes Functions
Hematology and Immunology - Leukocytes Functions
 
9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi Ncr
9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi Ncr9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi Ncr
9873777170 Full Enjoy @24/7 Call Girls In North Avenue Delhi Ncr
 
Glomerular Filtration and determinants of glomerular filtration .pptx
Glomerular Filtration and  determinants of glomerular filtration .pptxGlomerular Filtration and  determinants of glomerular filtration .pptx
Glomerular Filtration and determinants of glomerular filtration .pptx
 
Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...
Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...
Air-Hostess Call Girls Madambakkam - Phone No 7001305949 For Ultimate Sexual ...
 
Call Girls Viman Nagar 7001305949 All Area Service COD available Any Time
Call Girls Viman Nagar 7001305949 All Area Service COD available Any TimeCall Girls Viman Nagar 7001305949 All Area Service COD available Any Time
Call Girls Viman Nagar 7001305949 All Area Service COD available Any Time
 

Global Burden of Disease Study - 2010

  • 1. Global Burden of Disease Study – 2010 Dr. Rizwan S A, M.D.,
  • 2. Outline of presentation • • • • • • • • • Beginnings Prior studies What is GBD? Need for GBDS – 2010 Methodology of GBDS – 2010 Analytical components of GBDS – 2010 Five critical choices Results overview Articles published
  • 3. Beginnings • Original GBD 1990 Study was commissioned by the World Bank in 1991 to provide a comprehensive assessment of the burden of 107 diseases and injuries and ten selected risk factors for the world and eight major regions in 1990 • The disability-adjusted life year (DALY), a time-based measure that combined years of life lost due to premature mortality and years of life lost due to time lived in health states less than ideal health, was developed to assess the burden of disease.
  • 4. Prior studies • Global Burden of Disease Study 1990 – World bank commissioned (for 131 diseases and injuries and 10 risk factors) • Global Burden of Disease Study 2000 (GBD 2000) • Results for 1999, 2000, 2001, and 2002 were published in World Health Reports 1999-2003 • WHO published estimates for 2004 • WHO also issued updates for estimates in 2009 • National Burden of Disease studies in at least 37 countries
  • 5. What is GBD ? - 1 • There are some 6 billion people in the world, and hundreds of millions experience disease or injury each year. • Taken as a whole, the combined pain, suffering, loss of productivity and unrealised hopes and dreams are our world’s burden of disease ! • In other words, the burden of disease is a measurement of the gap between the current health of a population and an ideal scenario where everyone completes their full life expectancy in full health • The Global Burden of Disease project attempts to measure this total disease burden
  • 6. What is GBD? - 2 • Comparison is at the heart of the GBD approach • Comparison requires consistent definitions over time and across place; • Comparison across diseases for non-fatal health outcomes requires a system of disability eights that produce meaningful summary metrics across hundreds or thousands of specific disease sequelae. • metrics that quantify types of outcomes in commensurable units.
  • 7. What is GBD? - 3 • One needs estimates for all diseases, injuries, and risk factors, not just those for which there are the best data in a place • Poorly studied health problems may in fact be some of the most important in some places and neglected because no one has collected good data in the past • All age-groups are included in the GBD so that comparisons across age and sex groups are also a key advantage
  • 8. What is GBD? - 4 • GBD construct of the burden of disease is health loss, not income or productivity loss • Health loss does not capture all the key dimensions of health events; • GBD treats a unit of health loss as equally important across individuals whether rich or poor, old or young, male or female. • Incorporate advances in epidemiology, demography, statistics, and other quantitative disciplines. • Completely revamped and improved tools.
  • 9. What is GBD? - 5 • An approach to global descriptive epidemiology • It is a systematic, scientific effort to quantify the comparative magnitude of health loss due to diseases, injuries, and risk factors by age, sex, and geographies for specific points in time • Provides an opportunity to see the big picture, to compare diseases, injuries, and risk factors, and to understand in a given place, time, and age-sex group, what are the most important contributors to health loss.
  • 10. GBD 2010 Team 486 authors from 302 institutions in 50 countries Seven institutions: IHME as the coordinating center, the University of Queensland, the Harvard School of Public Health, the Johns Hopkins Bloomberg School of Public Health, the University of Tokyo, Imperial College London, and the World Health Organization. 10
  • 11. Key Activities 1. Produced valid, unbiased, and comparable estimates of prevalence of disease and injury cases or episodes and relevant disabling sequelae at the population level for the GBD regions 2. Produced cause-specific mortality by region that collectively sums to all-cause mortality estimates globally 3. Revised the health state severity weight system, a highly debated component of past GBD studies that seeks to measure health state severity 4. Produced estimates of YLDs, YLLs, death, and DALYs for diseases, injuries, and risk factors for 21 regions for 20 age groups and both sexes, for 1990, 2005, and 2010
  • 12. Scope of the 2010 study • 291 diseases and injuries • 1,160 disabling sequelae of these diseases and injuries • 67 risk factors • 20 age groups, two sexes • 187 countries • 1990-2010 • Improved methods for the estimation of health state severity weights 12
  • 13. Study objectives • Two major objectives – Revising figures for 1990 given new data and improved techniques and generating new estimates for 2005 along with projections for 2010. – Developing various sets of tools, each tailored to a specific audience, to standardize and broaden the field of burden research and analysis.
  • 14. GBD advantages - 1 1. Evidence-based evaluations – standard measures to ensure that all conditions receive systematic, objective analysis. 2. Changing awareness and validity of different global health conditions – visibility and legitimacy to conditions like depression which cause great suffering with little associated mortality, and to conditions like road traffic accidents, formerly outside the scope of public health.
  • 15. GBD advantages - 2 3. Cost effectiveness analysis – allows for lives in every part of the world to be valued equally and creates a common unit of currency for making decisions about the costs and benefits of various health interventions. – using standard units of measurement, such as disability-adjusted life years (DALYs). 4. Engaging researchers, experts, and policymakers
  • 16. Need for GBDS (2010) - 1 • While various groups have published partial updates of GBD rankings, there has not yet been a comprehensive and systematic revision. • As a result, burden estimates today contain some outdated, and often, inconsistent information.
  • 17. Need for GBDS (2010) - 2 • A core principle of the GBD is to facilitate unbiased comparisons across diseases, injuries, and risk factors. • Disability weights are the mechanism through which time spent with a disease or injury sequela is made commensurate with other disease and injury sequelae and with time lost due to premature mortality. • Disability weights had not been empirically revised since the GBD 1990.
  • 18. Need for GBDS (2010) - 3 • Demand for more detailed information and estimation of the burden of disease from diarrhea, pneumonia, and meningitis broken down by specific pathogen • Fourth, GBD 1990 and GBD 2000 results were based on different datasets and methods, but many users of the information have compared the results to examine trends for key outcomes. • A reassessment of 1990 burden using strictly comparable methods to the current time period was needed
  • 19. Need for GBDS (2010) - 4 • Methods innovation across the spectrum from mortality measurement, cause of death analysis, and meta-regression for epidemiological synthesis provided opportunities to modernize the GBD approach • Combined with a global movement towards enhanced transparency and access to data, the empirical basis for assessing the epidemiology of diseases is much richer than in the past
  • 21. Disease and injury causes, sequelae, and health states • Cause list has a hierarchical structure • At each cause level, the list provides a set of mutually exclusive and collectively exhaustive categories • Added causes on the basis of three criteria: potentially large, health policy interest, feasibility of measurement. • Four levels of diseases and injuries and a fifth level for sequela,
  • 22. Causes, sequelae, and health states - 1 • The four levels provide different degrees of disaggregation of the causes: • Three level 1 causes, 21 level 2 causes, 155 level 3 causes, and 112 level 4 causes • Of the level 3 causes 129 have no level 4 causes below them in the hierarchy; • At the most disaggregated level there are, therefore, 241 mutually exclusive and collectively exhaustive causes • The three level 1 causes are communicable, maternal, neonatal, and nutritional disorders; non-communicable diseases; and injuries.
  • 23. Causes, sequelae, and health states - 2 • List of sequelae is designed to capture the direct consequences of disease or injury that are not otherwise captured elsewhere in the cause list • Number of sequelae evaluated has increased for all three broad cause groups, leading to 1,160 in total • A given outcome only appears once in the cause sequela list to avoid double counting
  • 24. Causes, sequelae, and health states - 3 • For the GBD 2010 an important innovation has been to identify which disease sequelae are common across conditions; these common disease sequelae are called health states • Shorter list of health states which facilitates the disability weight measurement component of the study • For each of the health states, a lay description was developed for use in the empirical assessment of disability weights
  • 25. Disease and injury ranking list • Because the leading causes of burden tend to have some influence on the perception of disease control priorities, the choice of aggregation is at once important • Provides information at the second level cause (21 causes) • A ranking list with 176 causes selected to distinguish and cluster conditions together that may have programmatic or public health significance • Results in the tables are provided for all 291 causes; the ranking list is used only for the leading cause figures • The 176 causes along with the excluded residual categories are also mutually exclusive and collectively exhaustive
  • 26. Risks rather than specific diseases • An alternative way of looking at ill health is to consider risks for disease rather than diseases themselves • A risk is defined as the probability of an adverse health outcome or a factor that raises this probability • A single risk factor may underlie several diseases; e.g. unsafe sex is a risk for HIV/AIDS, many other sexually transmitted diseases and teenage pregnancy • It is often more useful to direct health interventions against risks rather than specific diseases • DALYs have been assigned to health risks as well as diseases
  • 27. Risk factor list • Comparative Risk Assessment 2000 list of 26 risk factors and subtracted and added risk factors based on consideration of four criteria: • Convincing or probable evidence on relative risks by cause, sufficient data to estimate exposure by region, potential size of the attributable burden of disease, and the extent to which burden is modifiable • A hierarchy with three levels
  • 28. Risk factor list • Level 1 risks – Groups of risk factors that are related by mechanism, biology, or potential policy intervention. • Most risks are presented at level 2, but in some cases, such as occupational carcinogens calculations, are done at a third level • The category diet and physical inactivity clusters together some key behavioral risks. This category has been substantially expanded to include 14 components of diet.
  • 29. Risk factor list • For some of the level 1 clusters of risk factors, there are analytical challenges for computation related to risk factors that operate through each other. • The problem of mediation is particularly challenging for the physiological risks. For example, BMI operates in part through blood pressure and cholesterol changes. • In this analysis, we have not been able to estimate this level 1 risk factor, so that results are presented for only 67 of the 69 risks or clusters of risks • As with diseases and injuries, we have identified a list of 43 risk factors for use in developing tables and figures of leading causes of risk factors.
  • 30. Regions • Basis of two criteria – Epidemiological homogeneity and geographic contiguity • Maximize the between-region variance in outcomes and minimize the within-region variance in outcomes • The process was iterative based on examination of levels of under-5 mortality and adult mortality (as measured by the probability of death between ages 15 and 60) • 21 regions were made. • For some types of analysis, created seven super-regions • For some Bayesian estimation procedures used in the analysis, super-regions are useful in the presence of sparse data.
  • 31.
  • 32. Subgroups - 1 • Early neonatal, late neonatal, post-neonatal, 1-4, 5-9, 10-14, …..70-74, 75-79, and 80+ • In the demographic analysis of age-specific mortality, estimated deaths and death rates for age-groups up to 100. • Due to limitations of data at older ages, however, 80+ used as the terminal age-group for the cause of death and disease and injury sequelae analysis.
  • 33. Subgroups - 2 • All estimates are produced separately for males and females • Not been able to include estimation of fetal deaths; this is an important area for future quantitative work. • A critical dimension of the GBD 2010 is the addition of multiple time periods assessed using strictly comparable data and methods • For three years - 1990, 2005, and 2010. • All results for causes of death have in the computation been estimated for single years as well
  • 35. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. Covariate database Age-specific mortality rates Cause of death database Estimating causes of death Ensuring consistency between cause-specific estimates and all-cause mortality estimates Disease sequelae epidemiology database Estimating disease sequela prevalence, incidence, and duration Impairment levels Nature and external cause of injury analysis Disability weights for 220 health states Comorbidity simulation Healthy life expectancy Computation of YLLs, YLDs, and DALYs from diseases and injuries with uncertainty Risk factor exposure database Estimating the prevalence of exposure Estimating relative risks for risk-disease pairs Theoretical minimum risk exposure Computation of attributable burden for each risk factor with uncertainty
  • 36. Covariate database • In settings where there are sparse data or conflicting data from multiple sources • This estimation task is facilitated by the use of appropriate covariates that can be used as independent variables in the modeling process • A database of 84 covariate topic areas and 179 variants of the covariates has been developed for this purpose • Missing data in covariates have been addressed using spatialtemporal regression and Gaussian process regression (GPR)
  • 37. Age-specific mortality rates • Complete re-analysis of the evidence on levels of age-specific mortality has been undertaken for 187 countries from 1970 to 2010 • This re-analysis takes advantage of improvements in demographic estimation methods for various types of data on child and adult mortality • All available data on child and adult mortality have been combined into a time series of best estimates
  • 38. Cause of death database • Information on causes of death comes primarily from vital registration systems and verbal autopsy studies • IHME has systematically created a cause of death database spanning the period 1980 to 2010, including mapping data across various revisions and national variants of the ICD • Algorithms have also been developed and applied to deal with the problem of ill-defined coding in vital registration and verbal autopsy data where deaths are assigned to unknown causes, intermediate causes, or immediate causes of death rather than the underlying cause of death
  • 39. Estimating causes of death • For 133 of 235 causes, cause of death ensemble modeling (CODEm) used as the primary strategy to develop causespecific estimates; • Three key aspects to this approach – To develop and test a diverse set of plausible models instead of a single analyst-preferred model; – To objectively test the prediction performance of these models by holding out some of the data from the analysis and then comparing model prediction to the held out data – In the CODEm approach, predictions for the GBD are based on a pool of the best models. • For some sub-causes, such as diarrhea etiologies, HIV/AIDS, and for some causes that rarely cause death, alternative methods have been used
  • 40. Ensuring consistency between cause-specific estimates and all-cause mortality estimates • Cause of death estimation proceeds in two stages. – First, models are developed for each cause of death in isolation – Second, the estimates for each cause are combined into estimates for all causes simultaneously that sum to the demographic estimates of all-cause mortality. • This simple algorithm leads to greater change in the estimates for more uncertain causes
  • 41. Disease sequelae epidemiology database • To support estimation for each region of disease sequelae incidence, prevalence, duration, remission, and excess mortality, a database of available published and unpublished data has been developed
  • 42. Estimating disease sequela prevalence, incidence, and duration • Developing estimates for disease incidence, prevalence, duration, and excess mortality estimates for 21 regions for three time periods by age and sex, a meta-regression tool was used (DisMod-MR) • Dealing with selection bias intrinsic to some data sources, such as hospital discharges • For some causes where more complicated models capturing more stages of disease progression are necessary, such as HIV, more elaborate natural history models were used, refined, or developed.
  • 43. Impairment levels • The impairment analysis has been undertaken for vision loss, hearing loss, anemia, infertility, heart failure, intellectual disability, and epilepsy • In these cases, the quality and extent of data on the overall impairment level have been determined to be stronger than on disease specific impairments.
  • 44. Nature and external cause of injury analysis • Dual nature of injury assessment. Each injury can be characterized by the nature of injury such as a femur fracture or head trauma and the external cause of injury such as a road injury or a fall. • To assess disability, data is required on the frequency of the nature of injury as well as follow-up data on reduced health functioning at certain points in time after the injury. • capture data on the matrix between the nature of injury incidence and the external causes leading to these injuries • first time included survey data on the incidence of external causes and, perhaps more importantly, multiple sources of long-term follow-up data to estimate the fraction of individuals with each nature of injury progressing to permanent disability
  • 45. Disability weights for 220 health states • Disability weights are the key mechanism in the GBD approach through which disease and injury sequelae are made comparable with each other and with time lost due to premature mortality. • we have switched to the construct of disability weights for health states as assessed by the general public. • To achieve this, simple lay descriptions of the 220 unique health states across the 1,160 disease and injury sequelae were developed. • Disability weights were assessed using general population surveys in five countries to capture a diverse set of views and through an Internet survey. • Disability weights were elicited using pairwise comparisons where the respondent is asked to identify which health state represents a higher level of health.
  • 46. Comorbidity simulation • Combinations of these sequelae change the likely health experience of the individual. • Comorbidity is taken into account in three stages. – First, for each age-sex group, a micro-simulation is developed to estimate the number of individuals who will have combinations of disease conditions. – Second, the combined disability weight for individuals with estimated assuming a multiplicative model. multiple conditions is – Finally, the overall individual health loss captured in the disability weight is assigned to the different sequelae experienced by an individual proportionate to the disability weights of individual conditions on their own. • The micro-simulation process is repeated for each country-age-sex-year 1000 times to propagate uncertainty in the prevalences and disability weights into the comorbidity estimates.
  • 47. Healthy life expectancy • Healthy life expectancy is a single summary metric of population health that is not affected by population age structure • It can be thought of as equivalent to the expectation of life at birth in full health given the currently observed age-specific death rates and prevalence of disease and injury sequelae • Used the life tables developed in component 2 for each country in 1990 and 2010, the disability weights and the output of the comorbidity simulations to estimate healthy life expectancy for 187 countries
  • 48. Computation of YLLs, YLDs, and DALYs from diseases and injuries with uncertainty • Each quantity is estimated with uncertainty by generating 1,000 draws from the posterior distribution • Given the large number of diseases and injuries and their sequelae, this task is computationally intensive
  • 49. Risk factor exposure database • Assesses the attributable burden of disease using four components: – Database on risk factor exposure, – Estimating the prevalence of risk factor exposure by age and sex – Relative risk for a specific disease, – Counterfactual distribution of risk exposure called the theoretical minimum risk distribution.
  • 50. Estimating the prevalence of exposure • Exposure estimates have been developed for many risks using the GBD Bayesian meta-regression tool, DisMod-MR • Related Bayesian methods have been used in published analyses for the metabolic risks • For some risks, alternative modeling strategies have been used
  • 51. Estimating relative risks for riskdisease pairs • For each risk-disease pair, relative risks for specific disease outcomes have been taken from published meta-analyses or updated meta-analyses that have been undertaken as part of this study. • Special analyses have been undertaken for water and sanitation, all sources of particulates less than 2·5 microns, and the components of diet
  • 52. Theoretical minimum risk exposure • The tradition in risk factor epidemiology is to assess attributable burden by estimating a counterfactual scenario where exposure to a risk factor in the past is set to a counterfactual distribution • Following the framework for comparative risk assessment, counterfactuals are set to the lowest risk distribution that has been observed or is theoretically possible – the theoretical minimum risk exposure distribution (TMRED)
  • 53. Computation of attributable burden for each risk factor with uncertainty • Risk factor burden is assessed using the distributions of burden for each disease associated with a risk factor • Uncertainty from the prevalence of exposure, relative risks, and where appropriate the TMRED, are all propagated into the final risk factor uncertainty
  • 54. Five critical choices • • • • • New standard reference life table Comorbidity Incidence versus prevalence Discounting for time Age-weighting
  • 55. New standard reference life table - 1 • The standard reference life table for the GBD 1990 was based on the highest observed life expectancy at the time, Japanese females with a life expectancy at birth close to 82·5 years • Given the progress in extending life expectancy in the last 20 years, it is appropriate to extend the normative goal for life expectancy • Developed a new reference standard life table by identifying the lowest observed death rate for any age-group in countries with more than five million in population
  • 56.
  • 57. New standard reference life table - 2 • At the Critical Ethical Choices for DALYs meeting, there was a consensus that using a different standard for males and females was inappropriate • First, the empirically observed gap in life expectancy between males and females continues to narrow • Within high-income countries, the gaps between male and female life expectancies especially for the lowest mortality communities have also narrowed • Second, there is no reason that society should have lower aspirations for health for males than females
  • 58. Comorbidity • A year of life lived with a disabling sequela such as blindness should contribute the same amount to the burden of disease regardless of whether it occurs in someone with another condition or not. • Critical Ethical Choices for DALYs meeting viewed that burden should be assessed individual by individual. Individuals with the same functional health loss should be treated as like regardless of whether that functional health loss came from one or several contributing conditions.
  • 59. Incidence versus prevalence • • • • • • Two types of DALYs were calculated in the GBD 1990: prevalence DALYs and incidence DALYs. Murray argued that the incidence perspective had two advantages: mortality rates used to compute YLLs are by definition incidence rates; using incidence for YLDs means that both non-fatal and fatal outcomes are analyzed in a consistent fashion. several disadvantages - individuals living with reductions in health functioning who need health and other social services may not be reflected in burden estimates if incidence has declined substantially. If the incidence of ischemic heart disease is designated to start with plaque formation or detectable myocardial ischemia will profoundly change the number of discounted YLDs. This arbitrariness to the construct of when a disease process starts is not present in a prevalence perspective. incorporation of comorbidity is more straightforward in a prevalence approach than an incidence approach. Incidence approach - information intensive and computationally challenging.
  • 60. Discounting for time • Anand and Hanson argued that there was no intrinsic reason to value a year of health as less important simply because it is in the future • One cannot expect that the value of a year of healthy life to society will be less in the future; • There is no expectation of declining marginal utility of years of healthy life • Chosen to compute the base case of DALYs for the with a discount rate of 0%
  • 61. Age-weighting • Some have argued that deaths at particularly young ages or even fetal death represents less of a loss of health because individuals do not have fully developed life plans • “the well-being of some age groups is instrumental in making society flourish; therefore, collectively we may be more concerned with improving health status for individuals in these age groups” • We recognize, however, that there is still an important set of issues about social priority that should be attached to preventing fetal death compared to death after a live birth that have not been addressed here
  • 63.
  • 64.
  • 65.
  • 66. The disability transition: progressive shift in burden from premature mortality to years lived with disability 1990 2010 66
  • 68. • All available data on causes of death for 187 countries from 1980 to 2010 from vital registration, verbal autopsy, mortality surveillance, censuses, surveys, hospitals, police records, and mortuaries • Six different modeling strategies to estimate cause-specific mortality trends depending on the strength of the data
  • 69. Methods • For 133 causes and three special aggregates - Cause of Death Ensemble model (CODEm) approach, which uses four families of statistical models testing a large set of different models using different permutations of covariates. Model ensembles were developed from these component models. model performance was assessed with rigorous out-of-sample testing of prediction error and the validity of 95% UIs. • For 13 causes with low observed numbers of deaths - negative binomial models with plausible covariates. • For 27 causes for which death is rare - modelled the higher level and then allocated deaths across component causes proportionately, estimated from all available data in the database. • For selected causes (African trypanosomiasis, congenital syphilis, whooping cough, measles, typhoid and parathyroid, leishmaniasis, acute hepatitis E, and HIV/AIDS), - natural history models based on information on incidence, prevalence, and case-fatality.
  • 70. Methods • Separately estimated cause fractions by aetiology for diarrhoea, lower respiratory infections, and meningitis, as well as disaggregations by subcause for chronic kidney disease, maternal disorders, cirrhosis, and liver cancer. • For deaths due to collective violence and natural disasters, we used mortality shock regressions
  • 71. Sources of data • Identified 25 variants of cause of death reporting lists in use from 1980 to 2010 across all sources of vital registration. • Identified 2798 site-years of data from 130 countries between 1980 and 2010 • Verbal autopsy data were collected through sample registration systems, demographic surveillance systems, or surveys. identified 486 site-years of published and unpublished verbal autopsy data across 66 countries, of which 10% were nationally representative. • We identified 2715 site-years of cancer registry data across 93 countries.
  • 72. Sources of data • Police and crime reports are an important source of information for some types of injuries, notably road injuries and interpersonal violence. • The police reports used in this analysis were obtained from published studies, national agencies, and institutional surveys such as the UN Crime Trends survey and the WHO Global Status Report on Road Safety Survey. • 1129 site-years of police reports from 122 countries • 32 site-years of burial and mortuary data in 11 countries • identified 1557 survey years with sibling history data, and a further 52 household survey years or census years of data covering 61 countries. • We also identified 56 surveys or censuses with information on injury mortality across 65 survey years or census years. • identified eight countries with nationally representative maternal mortality surveillance systems covering 83 site-years • 21 site-years of data based on deaths in health facilities.
  • 73. Results • Population growth, increased average age of the world’s population, and largely decreasing agespecific, sex-specific, and cause-specific death rates combine to drive a broad shift from communicable, maternal, neonatal, and nutritional causes towards non-communicable diseases.
  • 74.
  • 75.
  • 76.
  • 77.
  • 80.
  • 81.
  • 84. What is DALY? DALY= YLL + YLD Years of Lost Life (due to premature mortality) Years Lost to Disability (due to injury or illness) • The DALY is the internationally-accepted measure of death and disability and is increasingly cited as a powerful tool for decision makers in international health. • It is the sum of the number of years lost to premature mortality and the number of productive years lost to disability. One DALY is equivalent to one lost year of healthy life. • Calculation of DALYs is based on the assumption that everyone in the world has a right to the best life expectancy in the world. The only differences in the rating of a death or disability should be due to age and sex and not to income, culture, location or social class. • It’s attractiveness lies in the fact that it combines information about mortality and morbidity in a single number. DALY allows the losses due to disability and the losses due to premature death to be expressed in the same unit. Hence, DALYs facilitate comparisons of different health states or health outcomes.
  • 86. Sample “DALY” calculations for a disease Example A: • 100,000 children are stricken for 1 week with a disease with a disability weighting* of 0.3; 2% die at 1 year old. • DALYs = YLL + YLD = (2000 X 80) + (100,000 X 7/365 X 0.3) = 160,000 + 575 = 160,600 Example B: • 100,000 adults are stricken for 2 years with a disease with disability weighting* of 0.6; 20% die at age 80 years. • DALYs = YLL + YLD = (20,000 x 2) + (100,000 x 2 x 0.6) = 40,000 + 120,000 = 160,000 *There are STANDARDISED DISABILITY RATINGS for various conditions e.g deafness= 0.33, Down syndrome=0.5, Diarrhoea=0.12
  • 87. Why are DALYs important? DALYs attempt to provide an appropriate, balanced attention to the effects of non-fatal as well as fatal diseases on overall health. In the absence of such assessments, conditions which cause decrements in function but not mortality tend to be neglected. DALYs help to inform debates on priorities for health service delivery, research and planning. For example, DALYs can be used to: • Compare the health of one population with another – and allow decision makers to focus on health systems with the worst performance • Compare the health of the same population at different points in time • Compare the health of subgroups within a population - to identify health inequalities
  • 88. Disability weights, discounting and age weighting • Egalitarian principles are explicitly built into the DALY metric, and the global burden of disease studies apply these to all regions of the world. • The studies use the same "ideal" life expectancy for all population subgroups and exclude all non-health characteristics (such as race, socioeconomic status or occupation) apart from age and sex from consideration in calculating lost years of healthy life. • Most importantly, they use the same "disability weight" for everyone living a year in a specified health state.
  • 89. Disability weights, discounting and age weighting • Disability weights – A disability weight is a weight factor that reflects the severity of the disease on a scale from 0 (perfect health) to 1 (equivalent to death). Years Lost due to Disability (YLD) are calculated by multiplying the incident cases by duration and disability weight for the condition • Age weighting and discounting – 3% discounting and non-uniform age weighting was used in the original GBD 1990 study. These adjustments result in less weight given to years lived at young and older ages – The GBD 2001-2 study used 3% discounting but uniform age weighting. GBD 2004 update used the original 3% discounting and non-uniform age weighting
  • 90. DALYs: Social values 1. How to compare years lost due to death with years lived in poor health? essential  Disability Weights values between 0 and 1 2. Value of health year of life equal at all ages?  age weights 3. Value of future years of life?  discounting http://www.who.int/healthinfo/global_burden_disease/daly_disability_weight/en/index.html optional
  • 91. Disability Weights  In GBD, non-fatal consequences of diseases and injuries understood as transitions through different ‘health states’  YLD calculation requires aggregate assessments of the overall decrements in health associated with particular health states  disability weights  DWs are measures of overall levels of health rather than contribution of health to overall welfare GBD1990 :  DW elicited from panel of health professionals following explicit protocol evaluating 22 indicator conditions in an intensive group exercise with ‘deliberative phase’ using person trade-off (PTO) method. Responses averaged across participants
  • 92. New Disability Weight Project The new DW will have a greater emphasis on paired comparisons, anchored by time trade-off methods. It also aims to engage members of the general community (including those in developing countries) to a greater degree. The DW project is being carried out in two stages:  a community household survey in selected regions, and  an online open-access survey.
  • 93. New Disability Weight Project 1. Community surveys • • • 2. (Pemba, Bangladesh, Indonesia, Peru, USA) Cultural/SES diversity >> representativeness Paired comparisons Internet survey • • • • the primary source of data for the final DW will include random(?) selections from all ~230 sequelae open to all interested in participating (open-access includes a variety of measurements paired comparisons, ranking, visual analogue scale, time trade-off, population equivalence) to anchor the scale (paired comparisons, ranking, visual analogue scale, time trade-off, population equivalence) http://gbdsurvey.org/
  • 94. Discounting Discounting common practice in economic analyses. Assumes that individuals value their health more now than at some point in the future. So the further in the future health loss occurs the more they are discounted. • GBD 1990 used 3% discounting
  • 95. Discounting Why discount? 1. Consistency with cost-effectiveness analyses 2. Prevent giving ‘excessive’ weight to deaths at younger ages
  • 96. Age weighting  Used to reflect a social preference that values a year lived by young adult more highly than that of young children or the elderly. • Eg. An Australian survey found that respondents considered saving four 20-year olds as important as saving ten 60-year olds (Nord et al, 1996 and 1998)  Not related to productivity but ‘social’ role in caring for the young and old
  • 97. Age weighting Arguments against: 1.6 • • relative value 1.2 Unacceptable on equity grounds Does not reflect actual social values 0.8 But • Everyone potentially lives through every age  not inequitable 0.4 0.0 0 20 40 60 age 80 100
  • 98. Source: Murray and Lopez, 2006. Chapter 5
  • 99. Terminology used GBDS • Disability : Disability refers to any short-term or long-term health loss. • Sequelae :In the GBD 2010 cause list there are 291 diseases and injuries, of which 289 cause disability. In total, 1160 sequelae of these diseases and injuries. For example, diabetic neuropathy is a sequela of diabetes mellitus. To avoid double counting, a sequela can only be counted in the cause list once even if the same outcome might be caused by more than one disease. • Health state :Across the 1160 sequelae, 220 unique health states. For example, both malaria and hookworm have mild anaemia as a sequela. Mild anaemia is a unique health state. The list of unique health states serves two purposes: to allow assessment of the total burden of some health states such as anaemia across various causes, and to simplify the task of measuring disability weights for sequelae.
  • 100. Terminology used GBDS • Disability weights : A quantification of the severity of health loss associated with the 220 unique health states on a scale from 0 to 1, when 0 is commensurate with perfect health and 1 is commensurate with death. • Years lived with disability (YLDs) : For the GBD 2010, YLDs per person from a sequela are equal to the prevalence of the sequela multiplied by the disability weight for the health state associated with that sequela. YLDs for a disease or injury are the sum of the YLDs for each sequela associated with the disease or injury. • Impairments : In the GBD 2010 we estimated the prevalence and burden of several unique health states that are sequelae for multiple diseases including anaemia, heart failure, vision loss, seizures, hearing loss, infertility, and intellectual disability. These are referred to as impairments.

Editor's Notes

  1. Summary measures of health do reflect some degree of social value. The measure used in the GBD study is the DALY, which includes a disabilty weight, or utility weight, for each disease and injury, that aims to reflect the loss of health for an individual. In the original GBD there are two other optional weightings based on social values that were applied to the DALYs : age weighting and discounting. These invited quite a lot of debate.
  2. In GBD, non-fatal consequences of diseases and injuries understood as transitions through different ‘health states’Years of life lost (YLDs) are based on relative decrements in health which are quantified using disability weights. So the important point here is that ‘disability’ in the GBD is defined as ‘within the skin’ or purely in terms of health of the individual.
  3. SummaryThe new DW will have a greater emphasis on paired comparisons, anchored by time trade-off methods. It also aims to engage members of the general community (including those in developing countries) to a greater degree compared to the 1996 iteration. The DW project is being carried out in two stages: a community household survey carried out in a limited number of sites, and an online open-access survey.  
  4. Stage 1. Household SurveysHousehold surveys have already been carried out in Pemba, Banglasdesh, Indonesia and Peru. They are also currently being carried out in the USA via telephone. These populations were chosen for cultural diversity (also we suspect for convenience). Approx 2,500 people were interviewed at each site with 108 of the ~230 sequelae covered. The group report that preliminary data at individual level is very noisy, however they feel it is relatively consistent at population level. A good degree of consistency was found across the first four sites. Preliminary data from the US (~700 completed so far) is also quite consistent with other sites.A few implausible results were identified (for example paraplegia was given a higher ranking than quadriplegia). Preliminary findings from the pilot household survey suggest that participants had difficulty with long lay descriptions. Josh suggests that too much info in the lay description distract from the most salient points. Also salutary qualifications (for example quadriplegia described ‘… but has no difficulty thinking or learning’) distracts from severity. Another issue identified was that disorders with a high level of stigma or cultural disapproval attached (for example substance-use disorders gained a higher rating. The example given was illicit drug use disorders and alcohol use disorders – no word on how other mental disorders rated.The DW group intend to use these findings to inform the development of internet surveys. Hence descriptions for the internet survey will be restricted to <35 words. They aim to use vocabulary that is simple and non-clinical. Also a range of descriptions will be included for substance-use disorders so a sensitivity analysis can be carried out to determine what difference certain words make to the ranking (for example substance / cannabis / medication).Stage 2. Online surveysThe next stage of surveys will be online and this will be considered the primary source of data for the final DW. The Internet survey will include all ~230 sequelae, will be open to all interested in participating (open-access), and will include a variety of methods of measurement. Three versions will be available, with three different goals: The first will replicate the household survey with the goal of validating findings across diverse cultures, levels of education and language - the internet survey is expected to capture samples with higher education and a greater proportion of English-speaking people. Version 2 will fill out responses for the remaining sequelae not covered in the household survey. Version 3 will include all ~230 sequelae and multiple methods of comparison (including paired comparisons, ranking, visual analogue scale, time trade-off, population equivalence) to anchor the scale.
  5. Changes in the discount rate do have an effect on the proportion of burden due to disability, on the age distribution of burden and on the distribution of burden by broad disorder group.Basically the higher the discounting rate, the less contribution disability makes to the overall burden, and the greater contribution of mortality. EG When discounting is set to 10%, YLDs account for more than 40% of burden. When zero discounting is used, YLDs account for less than 25% of burden. A low discount rate enhances the importance of burden from disease in children 0-4 years. (All from Murray and Lopez, 2006 p279)The consequence of not including discounting in new estimates will be to give greater prominence given to childhood disorders and for adults, greater emphasis on mortality relative to disability
  6. Used in GBD to reflect social preference to value year lived by young adult more highly than that of young children or the elderlyNot related to productivity but ‘social’ role in caring for the young and oldThe effects of using age weighting are SMALLER in magnitude than the effects of using a discount rate (Murray and Lopez, 2006 p279), but still have a not insignificant effect on burden of Mental disorders and illicit drug use disorders.
  7. Used in GBD to reflect social preference to value year lived by young adult more highly than that of young children or the elderlyNot related to productivity but ‘social’ role in caring for the young and oldThe effects of using age weighting are SMALLER in magnitude than the effects of using a discount rate (Murray and Lopez, 2006 p279), but still have a not insignificant effect on burden of Mental disorders and illicit drug use disorders.
  8. This graph shows results of a sensitivity analyses around the impact of NOT using age-weighting and discounting GBD1990 DALY calculations. Sensitivity analysis found that the most important effect of using zero discount rate and uniform age-weights is the substantial reduction in the proportion of burden attributable to neuro-psychiatric disorders (Murray and Lopez, 2006 Chapter 5 p280). The RANK ORDER for mental disorders and substance use DROP with the exclusion of both adjustments.