Global burden of Diabetes Prevalence and projections Peter
Global burden of Diabetes: Prevalence and projections Peter H. Bennett, M. B. , F. R. C. P. National Institute of Diabetes and Digestive and Kidney Diseases, Phoenix, Arizona, U. S. A. Belllagio, March 16 th 2004
Outline • • • Prevalence of Diabetes Estimates of future prevalence (projections) Mortality attributable to diabetes Effect of the epidemic on rates of complications Projections for ESRD
Newsweek, September 4, 2000 Time, September 4, 2000
Global Projections for the Diabetes Epidemic: 2000 -2030 (in millions) EU 17. 8 25. 1 41% NA 19. 7 33. 9 72% LAC 13. 3 33. 0 248% Wild, S et al. : Global prevalence of diabetes: Estimates for 2000 and projections for 2030 Diabetes Care 2004 In press SSA 7. 1 18. 6 261% MEC 20. 1 52. 8 263% China 20. 8 42. 3 204% India 31. 7 79. 4 251% World 2000 = 171 million 2030 = 366 million Increase 213% A+NZ 1. 2 2. 0 65%
Estimated Number of People with Diabetes in 2000 and 2030 (and % change) +176% +26% -13% Wild, S et al. : Global prevalence of diabetes: Estimates for 2000 and projections for 2030 Diabetes Care 2004. In press
Estimated Number of People with Diabetes in 2000 and 2030 (and % change) +247% +308% +189% Wild, S et al. : Global prevalence of diabetes: Estimates for 2000 and projections for 2030 Diabetes Care 2004 In press
Estimated Number of People with Diabetes in 2000 and 2030 +212% +242% +75% Wild, S et al. : Global prevalence of diabetes: Estimates for 2000 and projections for 2030 Diabetes Care 2004 In press
Are these projections realistic? Based on: 1990 s estimates of diabetes prevalence Demographic projections Assume constant (current) age-sex specific prevalence of diabetes
Changing Prevalence of Type 2 Diabetes
No of persons with Physician-diagnosed Diabetes in USA by year Data from the US National Health Interview Survey: http: //www. cdc. gov/diabetes/statistics/prev/national/fig 1. htm
Diabetes in the United States 1958 -1993 Prevalence (%) > 65 years 55 -65 years 45 -54 years < 45 years Data from the US National Health Interview Survey
Diabetes in the United States 1980 -2000 Prevalence (%) 65 -74 years 75+years 45 -64 years < 45 years Data from the US National Health Interview Survey: http: //www. cdc. gov/diabetes/statistics/prev/national/fig 3. htm
Why were former projections inaccurate? • They were based on demographic changes (which are very predictable) • The epidemic of diabetes is driven by other factors (some or perhaps most of which may be reflected in occurrence of IGT and IFG)
Importance of IGT and IFG
Impaired Glucose Homeostasis IGT 'Impaired Glucose Tolerance’ (IGT) 2 h post-load Plasma Glucose of 140 -199 mg/dl [7. 8 -11. 0 mmol/l] and FPG <126 mg/dl [<7. 0 mmol/l] IFG 'Impaired Fasting Glucose' (IFG) FPG of 100 -125 mg/dl [5. 6 -<7. 0 mmol/l]
Incidence of Diabetes in people with Impaired Glucose Homeostasis
Prevalence and Incidence of Diabetes (%) in Impaired Glucose Homeostasis Prevalence of Impaired Glucose Homeostasis Hoorn 5 -year Cumulative Incidence of DM 19. 4 29. 7 19. 3 24. 5 15. 1 24. 8 19. 3 ? 25% ? 50 -75 y Mauritius 25 -74 y Pima Indians 15+y Nhanes III 40 -74 y
Projections of Numbers with Diabetes* among persons aged 40 -74 years in USA (Millions) *Based on diabetes incidence (5% / year) in the 20% of persons with impaired glucose homeostasis; 300, 000 diabetes deaths / year; Net increase = c. 700, 000 cases of diabetes / year
Proportion of new cases of Diabetes (%) in persons with NGT and Impaired Glucose Homeostasis Proportion from persons with NGT Hoorn Proportion from those with IGT or IFG 38% 62% 40% 60% 44% 56% 50 -75 y Mauritius 25 -74 y Pima Indians 15+y Nhanes III 40 -74 y ? 40% ? ? 60% ?
Projections of Numbers with Diabetes* among persons aged 40 -74 years in USA (Millions) *Based on diabetes incidence (5% / year) in the 20% of persons with impaired glucose homeostasis; 40% of new cases from persons with NGT; 450, 000 diabetes deaths / year Net increase = c. 830, 000 cases of diabetes / year
Mortality attributable to diabetes
Number of Deaths with Diabetes as Underlying Cause of Death, United States, 1980 -1996. CDC, Diabetes surveillance, 1999
Excess Deaths attributable to Diabetes • Excess mortality =No of deaths in DMs – Expected no. if not DM • No of DM deaths = (RR of death in DM x No. with DM) • Expected No. = Mortality rate in non DMs*No. with DM • Mortality rate; RR of death in DM; No. with diabetes
Global Projections of Excess Deaths attributable to Diabetes (in 000’s) and percent of all deaths in year 2000 EUR 609. 0 6. 4% AMRA 251. 1 9. 75% AMRD 32. 2 6. 1% AMRB 194. 0 7. 6% Roglic et al. : Burden of mortality attributable to diabetes: Estimates for the year 2000. In preparation AFR 319. 2 2. 6% EMR 125. 3 7. 2% SEAR 1154. 1 6. 9% World In 2000 = 3. 164 million (5. 4% of all deaths) WPR 469. 3 4. 1%
Consequences of the Epidemic
Consequences of Epidemic • Increase in number with diabetes • Disproportionate increase in duration-related complications
Effect of Epidemic of Diabetes on Duration-related complications Cumulative No. of DM cases % with DM>15 y duration Prevalence of complications (%) among those with DM 50 40 30 20 150 Percent with complications No. of Cases 450 10 0 5 10 15 20 25 30 Time (years) 35 40 45 50 16 new cases/yr. Death occurs after 30 y DM; Complication incidence 50% after 15 y DM After 30 years No. of new cases equals no. of deaths.
Incidence of Type 2 Diabetes in Pima Indians in two time periods Time Period 1965 -75 1975 -85 60% increase 0 10 20 30 40 Incidence per 1000 person-years Age-sex adjusted rates 50
Age-standardized mortality from Ischemic Heart Disease in diabetic and non-diabetic Pima Indians aged 35+ years, (30 year follow-up) Mortality /1000 person-years 6 5 4 3 2 1 0 1965 -74 1975 -84 1985 -94 Non-diabetic 1965 -74 1975 -84 1985 -94 Diabetic
Renal disease
Frequency of Renal Failure among Pima Indians 25 Cases 20 Deaths Dialysis 15 10 5 0 1965 1970 1975 1980 Year 1985 1990 1994
Cases per 1000/year Incidence of Renal Failure 12 8 4 0 1965 -74 1975 -84 Time Period 1985 -94
Number of persons beginning treatment for ESRD attributable to diabetes in USA by year http: //www. cdc. gov/diabetes/statistics/esrd/Fig 1 Detl. htm
DIABETES IN END-STAGE RENAL FAILURE: AUSTRALIA 1980 - 2000 Number of Diabetics Number of New Patients Year of Entry
Crude and Age-adjusted incidence of ESRD related to diabetes in the United States 1984 -2001 Incidence/100, 000 Diabetic persons Crude rate Age-adjusted rate Data from the US Renal Data System: http: //www. cdc. gov/diabetes/statistics/esrd/fig 7. htm
Projections for number of patients with diabetes initiating ESRD treatment in USA* 2000 Number with Diabetes No with new ESDR 2030 (using WHO estimate) 18 million 31 million 2030 Revised projections 39 -48 million 40, 000 86 -105, 000 68, 000 * Assuming current incidence rates for initiating ESRD treatment remain constant
Summary • Based on demographic changes alone: The numbers of persons with diabetes in the world will more than double in the next 30 years • In developed countries they will increase by 3070% (mostly in older persons) • In developing countries they will increase by c. 250% (mostly in 45 -64 y age group) • These projections do not take into account any increase that is attributable to future increases in obesity
Summary • c. 3. 2 million (excess) deaths were attributable to diabetes in year 2000. • • In the USA this is 9% of all deaths – e. g. In USA c. 200, 000 excess deaths vs. 14, 500 for AIDs • The numbers of deaths attributable to diabetes in future years will increase especially in developing countries
Summary • Because of the current epidemic of diabetes, reflected in increasing age specific prevalence, the proportion of the diabetic population with complications will increase. • This will result in a greater relative increase in complications than in diabetes prevalence. • Because serious complications e. g. ESRD, typically develop after 15 -20 years duration, the incidence of ESRD due to diabetes will continue to increase for at least the next 20 years
Thank you
Age-adjusted incidence of ESRD related to diabetes in the United States 1984 -2001 Incidence/100, 000 Diabetic persons Amer. Indian male Amer. Indian female Black female Hispanic female White female Data from the US Renal Data System: http: //www. cdc. gov/diabetes/statistics/esrd/fig 5. htm
Are there better ways to predict future trends? • • • Incidence = No. of new cases/ period of time Mortality = No. of deaths/ period of time Prevalence=Cumul. incidence – Cumul. Mortality = Number of cases at a point in time Future no. of cases = No. of prevalent cases + no. of new cases- No. of deaths(among all cases)
- Slides: 50