Putting Health Metrics into Practice Using DALYs for

Putting Health Metrics into Practice: Using DALYs for Programmatic Decision-Making Authors: Kim Longfield, Steven Chapman, Warren Stevens & David Jeffries Population Services International BACKGROUND THE DISABILITY-ADJUSTED LIFE YEAR In 2006, Population Services International (PSI) adopted the disabilityadjusted life year (DALY) as its key performance metric to inform programmatic decision-making and to estimate the health impact of its products, services, and behavior change interventions. The DALY was developed by the World Bank and WHO in 1993 to measure the burden of disease by region. It combines the value of lives saved and illness and disability prevented. The DALY allows PSI to: • Measure the impact of interventions in relation to the global burden of disease • Compare impact across health conditions and countries • Compare the cost-effectiveness of interventions and compare costs to other NGOs and global standards While international standards exist to calculate DALYs at the national level, PSI has pioneered the application of DALYs to individual health interventions. METHODS DALY model creation • PSI creates DALY models 1 for its interventions. Parameters for models include: epidemiology, demography (including sub-population level), product/service efficacy, utilization data (e. g. wastage, adherence), intra-country variance (e. g. weights for urban vs rural, risk stratifications, etc. ), and delivery mechanism • Coefficients are created for each country based on stochastic modeling DALY calculation • Each month, country offices report sales/distribution and service utilization figures • Survey data are used to estimate the impact of PSI’s communication activities on non-product behaviors and the use of non-PSI brands • (Coefficients * sales/distribution/service) + communication DALYs = estimate of health impact Cost-effectiveness • Annual expenditure and output activity data are used to calculate cost per DALY estimates across interventions, health areas and for each country DALYs Trend 2000 -2010, % growth from the prior year Millions 18 13% 16 13% 14 13% Child Survival 12 -7% 10 8 40% Malaria Control 41% Reproductive Health 29% 6 34% 4 25% 21% 2 32% HIV & TB 28% 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 PSI uses graphs like this to track DALYs averted. Health impact has grown each year at a rate of 13 -41% Pakistan National burden of disease measured in DALYs DALY Share of Communicable, maternal, perinatal (000 s) burden and nutrition conditions (CMPN) Perinatal conditions 6, 515 16. 06% ® Respiratory infections 3, 295 8. 12% ® Diarrheal diseases 2, 115 5. 21% ® 4. 49% 3. 57% ® Childhood-cluster diseases Tuberculosis 1, 820 1, 446 Maternal conditions 1, 411 3. 48% ® Nutritional deficiencies 1, 097 2. 71% ® 1. 72% 0. 70% 0. 31% 0. 26% ® 0. 25% ® STDs excluding HIV Meningitis Tropical-cluster diseases Intestinal nematode infections 698 286 126 105 HIV/AIDS 102 Hepatitis B 67 Malaria 66 Hepatitis C Dengue Leprosy Trachoma Other CMPN 26 20 19 4 1, 018 Total CMPN 20, 235 0. 16% ® 0. 16% 0. 06% 0. 05% 0. 01% 2. 51% 49. 89% Non-communicable diseases and injuries (NCD) Cervical cancer Respiratory cancer* Cardiovascular diseases Potential PSI interventions Contraception, clean delivery kit, iron folate, multivitamins Pneumonia pre-packaged therapy, zinc, community case management Water treatment, diarrhea treatment (oral rehydration salts & zinc) Directly observed therapy (short course) Contraception, clean delivery kit, iron folate, multivitamins Micronutrient & macronutrient supplementation, exclusive breastfeeding STI treatment kits & condoms Condoms, voluntary counseling & testing, male circumcision, PMTCT Insecticide-treated bed nets, malaria pre-packaged therapy, case management 0. 00% 20, 326 Total NCD 20, 326 Total all causes 40, 561 * Mouth, oropharynx, trachea, bronchus, lung cancers Source: WHO 2009 50. 11% 100. 00% CHALLENGES AND LIMITATIONS • Creating models is resource-intensive, and they must be updated as new evidence becomes available • Data – and especially high-quality data – can be lacking • When new epidemiological data become available, estimates of health impact change, which can be confusing to stakeholders • Sales and distribution figures are used as proxies for product use/behavior • Coefficients are not exact and accuracy depends on model parameters • Capturing health impact through communications and targeting has proven more challenging than expected NEXT STEPS 50. 11% Other NCD This graph depicts the proportion of disease burden averted due to current programs and identifies strategies that should either be scaled up or diversified to increase health impact Burden of Disease tables are used to identify which interventions to launch or scale up. They present country-specific burden of disease and rank the importance of each health area in which PSI works 1 Current models include: HIV, Family Planning, Maternal Health, Malaria, Diarrhea, Acute Respiratory Infection, Tuberculosis, Nutrition, STI Kit, Basic Care Package for PLWH, Naloxone, Male Circumcision (neonatal & adult), 6 communication models for HIV • Quantify coefficient uncertainty • Calculate cost-effectiveness in real time • Develop models to include co-morbidity and target populations at elevated risk for multiple adverse health outcomes • Develop tools to guide optimum investment strategies • Measure market strengthening and growth For a copy of this poster go to www. psi. org