Institute of Health Metrics and Evaluation and the global burden of disease

By Nicole Bond Edwards

Population health matters

If we want to save as many lives as possible, where should we focus our efforts? As professionals in the health industry dedicated to enhancing the well-being of others, there are several critical questions that we must address, including these:

• What is causing people to live in suboptimal health in our local communities?
• Where should we be focusing to improve maternal health outcomes globally?
• How do we evaluate the impact and effectiveness of a health intervention?
• Which modifiable risks are contributing to morbidity and mortality and how is it changing over time?
• Which health outcomes will require increased attention by the year 2050?
• In what ways can impartial data bolster trust with our key stakeholders?

These pressing questions are at the forefront of discussions among policymakers, business leaders, philanthropists, and decision-makers across diverse sectors and on a global scale.

To address these questions, detailed, comprehensive, and timely reporting on population health by underlying causes of disability and premature death is of the upmost importance. This data plays a crucial role in driving strong evidence-based decisions that aim to tackle the complexities of disease and injury burden over time and across age groups, sexes, and locations.

In an effort to answer the call these questions pose, the Institute of Health Metrics and Evaluation (IHME) was formed, and the creation of the Global Burden of Diseases, Injuries, and Risk Factors study (GBD) was ignited. IHME is a population health research institution housed at the University of Washington and is a global center of excellence in health data and analytics, whereas the GBD – a flagship research initiative led by IHME – represents the largest scientific endeavour aimed at measuring trends and levels in population health. IHME’s Client Services Unit, operating as a social enterprise within the university, serves to facilitate the utilization of this research within the private sector.

The four pillars of the GBD

One of the goals of the GBD is to provide a comprehensive quantification of health loss in every country and over time, so that health systems can be improved and disparities can be addressed. Supported by the contributions of more than 11,000 global collaborators, this work produces numerous health measures, including disease-specific incidence, prevalence and mortality estimates. Additionally, it offers summary burden measures that illuminate the level of disability associated with various health conditions, as well as the healthy life lost due to premature mortality and living in suboptimal health.

The GBD also generates overall estimates of life expectancy and health-adjusted life expectancy are also produced, enabling a better understanding of how increased longevity could potentially lead to more time spent struggling with ill health. Further, the study provides metrics to quantify the disease burden that is attributable to modifiable risk factors, including behavioural, metabolic and environmental risks.

There are four tenets of GBD data:

1. Comprehensive: Estimates are available for every combination of location, demographic group, year, metric and condition. When input sources are highly inconsistent or data is sparce, a best estimate is still produced, along with measures of uncertainty.
2. Comparable: Consistent methods and metrics facilitate comparisons across time and space, and within and between populations of interest.
3. Internally consistent: Our hierarchical approach to categorizing all health risks and outcomes assures that our estimates are collectively exhaustive and mutually exclusive.
4. Transparent: Every estimate is provided with an uncertainty range and all methods are vetted and published in peer-reviewed literature.

With such an ambitious agenda and scope, there are limitations that are faced by the GBD project. As reported, GBD measures the health of the world’s population. Without additional analysis, it would not provide direct insights for an insured, hospitalized or more targeted population.

The extensive range of data sources, each with its own potential biases, coupled with gaps in data for specific locations and years, complicates the analysis of single outcomes in GBD or tracking person-level data over time. Rather, the strength lies in the breadth of the estimates produced. The strength (or lack) of the evidence for many potential risk-outcome relationships is also a limitation.

“GBD follows a rules-based approach to synthesizing this evidence. This means that GBD findings may differ from assessments that place more weight on expert views than on data-driven evidence.”

Actuarial uses for GBD data

With research as comprehensive as GBD, there are a number of areas where it can be used to inform assumption setting, improve models, and generally strengthen the data used and results produced in actuarial use cases. Some examples of the ways actuaries have used GBD data include but are not limited to the following categories:

Assumption setting and risk categorization:

• Using incidence and prevalence metrics for assumptions-setting for critical illness policy.
• Using cause-specific years-of-life-lost metrics that measure premature mortality attributable to a health condition to support risk categorization for life insurance applicants.

Benchmarking and analysis:

• Benchmarking internal or industry data to quantify differences between population of interest and the general population.
• Understanding drivers of mortality, enabling better decision-making and more accurate rate setting, especially in projections for mortality changes into the future.

Business strategy and forecasting:

• Informing employee health strategy and wellness programs by identifying the risk factors and conditions contributing most to potential health loss.
• Using forecasted health scenarios for disease burden, mortality and longevity to inform the future projections of business outcomes.
• Quantifying expectations for new markets, based on differences in GBD metric between countries, combined with experience data for existing markets.
• Quantifying expectations for new products or changes in underwriting, informed by the GBD metrics for conditions that are relevant to the new features.

Global assumption setting:

• For countries without credible country-specific mortality tables or mortality improvement tables, using GBD life tables for general assumptions setting.

Customized guidance and personalized tables:

• Setting customized individual health guidance based on reported risk factor exposures and diagnoses.
• Creating personalized mortality tables based on individual risk factor exposures.

In addition to actuarial uses for this case, the real-world applications are far reaching, from life sciences, biotechnology, and medical devices, to supply chain management, technology and beyond. IHME equips decision makers all over the world with data-driven insights that drive confident, strategic decision making.

This article reflects the opinion of the author and does not represent an official statement of the CIA.