Publications and Products

The Economics of Diabetes Mellitus:
An Annotated Bibliography

COSTS OF DIABETES

Direct and Indirect Costs

192

TITLE: The Burden of Diabetes in North Carolina. Kegler, M.; Lengerich, E.; Norman, M.; Sullivan, L.; Stoodt, G. North Carolina Medical Journal. 56(4): 141-144. April 1995.

OBJECTIVE: To assess the true burden of diabetes in North Carolina using prevalence, hospitalization rate, multiple-cause mortality, and economic cost figures.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Societal.

CONCLUSION: The burden of diabetes in North Carolina, as reflected by its listing on death certificates as the underlying cause, is underestimated. The total cost of diabetes in the state in 1990 was estimated to be $1.2 billion, 46 percent attributable to inpatient or outpatient care and 54 percent to indirect costs.

RECOMMENDATION: Regular collection, analysis, and reporting of data about diabetes in North Carolina is needed to enable decision makers to plan for and allocate scarce health resources.

ABSTRACT: The authors applied prevalence, hospitalization rate, multiple-cause mortality, and economic cost data to the calculation of the true burden of diabetes in North Carolina. Prevalence data, based on the Behavioral Risk Factor Surveillance System survey, indicated that 290,000 adult North Carolinians had diabetes (5.8 percent of the total population). Between 1980 and 1989, 41,676 North Carolina decedents had diabetes listed as a contributing cause of death, but it was listed as the underlying cause of death for only 9,771 decedents. Diabetes was mentioned as a diagnosis in 66,067 hospitalizations but as a primary cause for these hospitalizations in only 15 percent of cases. Charges for diabetes-related hospitalizations during 1988 to 1989 totaled $490 million, or $73 per resident. The total cost of diabetes in North Carolina during 1990, however, was estimated at $1.2 billion based on a Centers for Disease Control model. Approximately 46 percent of this total was attributed to inpatient and outpatient care ($574 million). The remaining 54 percent ($664 million) reflected indirect costs, including forgone wages due to short-term morbidity, long-term disability, and death. Because of the large burden of diabetes in North Carolina and the potential to prevent complications from diabetes, regular collection, analysis, and reporting of data about diabetes is needed so that decision makers can accurately plan for and allocate scarce health resources. 4 tables, 15 references.

193

TITLE: The Cost of Diabetes. Gerard, K.; Donaldson, C.; Maynard, A.K. Diabetic Medicine. 6(2): 164-170. March 1989.

OBJECTIVE: To estimate the cost of diabetes in England and Wales in 1984, and to outline problems in ascertaining the true costs of diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost-of-illness.
Perspective: Societal.

CONCLUSION: The total indirect and direct costs of diabetes in England and Wales ranged from £ 259.5 million to £ 602.5 million, depending on the rate of absenteeism from work.

RECOMMENDATION: The epidemiological base on which the cost of diabetes estimate was based must be improved by using prospective studies with long follow-up periods.

ABSTRACT: The authors estimated both direct and indirect costs of diabetes; direct costs were associated with preventing, detecting, and treating diabetes; indirect costs to lost productivity due to absenteeism, early retirement, and premature mortality. The cost of welfare effects attributable to deterioration in quality of life was not included. Total direct costs were estimated at £ 238.9 million, including £ 74.0 million and £ 86.0 million for inpatient treatment of diabetes as a primary and secondary diagnosis, respectively; £ 22.8 million for outpatient care; and £ 56.1 million for family practitioner services. Lost earnings were estimated at £ 20.6 million. Losses due to absenteeism varied based on the assumed rate of absenteeism relative to the normal population, ranging from £ 0.0 (absenteeism equal to that of the normal population) to £ 343.0 million (absenteeism equivalent to three times that of the normal population). Total costs associated with diabetes ranged from £ 259.5 million to £ 602.5 million. The authors state that existing data are deficient in quality (in particular, no distinction is drawn between patients treated by insulin and those not so treated). More detailed epidemiological data are needed to accurately assess the economic burden of diabetes. To set priorities, economic evaluation (e.g., cost-effectiveness, cost-benefit analysis) rather than cost-of-illness studies, such as this one, is needed. 5 tables, 22 references.

194

TITLE: Cost of Diabetes in France. Triomphe, A.; Flori, Y.A.; Costagliola, D.; Eschwege, E. Health Policy. 9(1): 39-48. February 1988.

OBJECTIVE: To determine direct costs for patients with type 1 and type 2 diabetes residing in Paris.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.

CONCLUSION: The direct cost for type 1 patients was 19 percent greater than the average direct cost of medical care for France's general population, while the direct cost of the type 2 group was somewhat below the national average.

RECOMMENDATION: None.

ABSTRACT: Of 109 patients studied, 27 had type 1 diabetes and 82 had type 2 diabetes. The type 1 group was 56 percent female and had an average age of 48 years; the type 2 group was 62 percent male and had an average age of 59 years. Mean duration of disease was 10 years for both groups. Costs (for hospitalization, laboratory studies, drugs, visits or consultations, sick leave, and other) for the type 1 group averaged 12,178 French francs (FFr); for the type 2 group, 6,908 FFr. Not including sick leave, mean costs were 7,711 FFr for the type 1 group and 5,892 FFr for the type 2 group, versus mean medical care costs of 6,462 FFr for the general French population. Two areas accounted for most of the non-sick leave costs: drugs (44.7 percent in the type 1 group and 34.3 percent in the type 2 group) and hospitalization (33.9 percent in the type 1 group and 40.1 percent in the type 2 group). In the general French population, drugs accounted for 17.9 percent of costs. The frequency of hospitalization was similar for the type 1 group and the general population. Compared with findings from a 1978 study, hospitalization was higher and visits or consultations lower. 5 tables, 13 references.

195

TITLE: Costs of Diabetes in Texas, 1992. Warner, D.C.; McCandless, R.R.; De Nino, L.A.; Cornell, J.E.; Pugh, J.A.; Marsh, G.M. Diabetes Care. 19(12): 1416-1419. December 1996.

OBJECTIVE: To estimate the direct and indirect costs of diabetes in Texas for 1992.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Societal.

CONCLUSION: Costs were conservatively estimated at $4.0 billion.

RECOMMENDATION: The databases consulted for this research study could assist state and local planning efforts in diabetes.

ABSTRACT: For this study, several billing databases were searched to find individuals with diabetes or hypoglycemia as a primary or secondary diagnosis. Unique individual identifiers were used to capture all incidents of health care use by persons with the diagnoses identified. For most direct medical costs, the authors relied on third party or provider databases, which included Texas Medicare and Medicaid; selected Veterans Affairs and public hospitals; the state's Department of Health, Rehabilitation Commission, and Commission for the Blind; a commercial pharmaceutical database; and a migrant/community health center. Costs were defined in terms of allowed charges or payments. Three categories of direct costs were estimated — clearly attributable to diabetes, clearly attributable plus probably attributable to diabetes, and all costs for people with diabetes. Survey data were used to estimate private insurance claims and associated out-of-pocket costs. For nursing home costs, the finding of a 1992 American Diabetes Association study on days attributable to diabetes was applied to Texas experience. National Health Interview Survey and U.S. Department of Labor data were used to estimate costs of short- and long-term disability. For the cost of premature mortality, 1992 Texas death records were searched for persons with diagnoses of diabetes or hypoglycemia; present value of lost future productivity was calculated with a discount rate of 3 percent. The authors estimated that $2.4 billion in indirect costs and $1.6 billion in direct costs were clearly or probably attributable to diabetes. Almost half of direct costs were paid by Medicare. 2 tables, 14 references.

196

TITLE: The Cost of Diabetic Foot Problems. Ward, J.D. PharmacoEconomics. 8 (Supplement 1):55-57. 1995.

OBJECTIVE: To discuss the prevalence and incidence of diabetic foot problems, their economic cost, and strategies for improvement.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Review of studies.
Perspective: Health care system.

CONCLUSION: Vigilant identification of those at risk for diabetic foot problems, supported by education, care, and appropriate clinical services, will dramatically lower the incidence of foot ulceration and its associated costs.

RECOMMENDATION: All diabetes clinic services should provide a foot care specialist.

ABSTRACT: Peripheral vascular disease, neuropathy, and social factors are important precursors of foot ulceration and amputation. In 50 percent of cases of the diabetic foot, vascular disease and neuropathy will both be present. The authors report their impression that few patients with diabetes who have foot problems are employed. According to a British study (Robinson et al. 1990), hospital admissions with peripheral vascular disease and neuropathy accounted for 20.8 percent of total bed days associated with diabetes in a recent period. Estimated total cost for these admissions was £ 12.9 million. According to Caddick et al. (1994), there is a very significant correlation between residing in an economically deprived area and being admitted to the hospital for diabetes; it is 2.5 times more common for those from deprived areas to be admitted. The author of the present article cites studies showing that provision of a foot care specialist reduces the incidence of diabetic foot ulceration and recommends offering a regular session for diabetic foot problems with a team consisting of a physician, diabetes specialist nurse, chiropodist (podiatrist), and orthotist. The author states that there is no reason that the objective included in the St. Vincent Declaration (1990) of reducing amputation in diabetes by 50 percent over 5 years cannot be achieved. In the United Kingdom, local diabetes services advisory groups have been set up in every health district to oversee, organize, and improve the provision of diabetes care. 17 references.

197

TITLE: The Cost of Diabetic Neuropathy. Ward, J.D. PharmacoEconomics. 8 (Supplement 1):52-54. 1995.

OBJECTIVE: To discuss diabetic neuropathy and its economic implications.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Review of studies.
Perspective: Health care system.

CONCLUSION: Assuming that up to 20 percent of patients in a diabetic clinic will have neuropathic problems, numb feet, predisposition to ulcers, or impotence, a great deal of time should be spent with such patients (much more than currently available in many clinics).

RECOMMENDATION: Measures of quality of life need to be developed to prove the efficacy of treatments in patients with diabetic neuropathy.

ABSTRACT: No available economic studies actually state the economic price of diabetic neuropathy, a term with no precise definition. The author suggests defining diabetic neuropathy as a state of nerve damage leading to unpleasant symptoms mainly in the distal lower limbs that will lead the patient to consult a physician or the detection of physical signs indicating severely damaged nerves likely to result in clinical problems (diabetic foot ulceration). This definition includes 20 percent of patients; many epidemiological studies quote a prevalence of 28 to 38 percent. The most common syndrome of diabetic neuropathy is chronic sensory neuropathy of insidious onset (83 percent of cases). Autonomic neuropathy is relatively rare. Among men with diabetes aged 55 or over, 50 percent are impotent — the cause is primarily neuropathic, with vascular sclerosis and psychological factors also contributing. Although the role of neuropathic disease in developing foot ulceration is well recognized, the great amount of suffering among those with other forms of neuropathy is not. As a consequence, these other forms of neuropathy are not recognized as costly. 15 references.

198

TITLE: The Cost of Education (Editorial). Siddons, H. Diabetic Medicine. 11(3): 239240. April 1994.

OBJECTIVE: To point out the benefit of education for patients with diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Societal.

CONCLUSION: Not educating patients with diabetes is very expensive.

RECOMMENDATION: Convince health service authorities that diabetes education is worth the investment.

ABSTRACT: The editorialist points out that the 1993 Diabetes Control and Complications Trial in the United States clearly demonstrated that intensive education and good blood glucose control can improve outcomes for people with diabetes. She notes the conflict between the Diabetes Control and Complications Trial recommendation of 20 minutes of education monthly for each patient with the British Diabetic Association's recommendation of one diabetes nurse per thousand patients. She also points out that nurse specialists are considered expensive and notes that many people are looking to practice nurses to take on this role. Practice nurses, however, are under pressure and their time is often used inappropriately. The editorialist states that it is well documented that education can reduce the cost of complications and hospital admissions for persons with diabetes. She notes that the cost of not educating patients is extremely high — about £ 300 for an overnight stay at the Manchester Royal Infirmary (before treatment) and £ 5,000 for inpatient treatment alone for a below-knee amputation at that institution. The health service is currently driven by financial implications, but this may not be bad. The authorities must be convinced that it is worthwhile to invest in diabetes. 2 references.

199

TITLE: The Cost of Insulin-Dependent Diabetes Mellitus (IDDM) in England and Wales. Gray, A.; Fenn, P.; McGuire, A. Diabetic Medicine. 12(12): 1068-1076. December 1995.

OBJECTIVE: To estimate the direct and indirect costs of type 1 diabetes in England and Wales in 1992 using a cost-of-illness approach.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Epidemiological cohort model.
Methodology: Cost-of-illness.
Perspective: Societal.

CONCLUSION: The estimated direct medical costs related to type 1 diabetes were £ 95.6 million; indirect costs (lost earnings) were estimated at £ 113 million.

RECOMMENDATION: Cost-of-illness studies such as this one should provide a framework for cost-effectiveness analyses to determine appropriate levels of expenditures for diabetes treatment programs.

ABSTRACT: The authors performed a cost-of-illness study of type 1 diabetes in England and Wales in 1992 using an incidence-based approach. The total number of persons with type 1 diabetes in 1992 was estimated at 93,581. Using available information on mortality risk, 2,014 deaths were estimated per year from type 1 diabetes for this imaginary cohort, resulting in an annual loss of 50,993 life-years and 23,691 potential working years. Estimated direct costs were £ 95.6 million; per sensitivity analysis, the range was £ 77 million to £ 113 million, depending on disease incidence. About half ( £ 47.3 million) of these costs were those directly attributable to diabetes. Almost half of those costs were associated with routine insulin maintenance, including glucose monitoring and routine outpatient care. The second most important direct expenditure category was renal complications, which accounted for £ 30.2 million. Most of these expenses ( £ 26.5 million) were for renal replacement therapy. Vascular complications accounted for nearly 5 percent of direct costs; ophthalmic and neurological complications each accounted for about 1 percent. Costs of social security payments attributable to type 1 diabetes were estimated at £ 11 million annually. Potential earnings losses (discounted at 6 percent per year) due to premature mortality were estimated at £ 113 million annually. 6 tables, 33 references.

200

TITLE: The Cost of Obesity: The US Perspective. Wolf, A.M.; Colditz, G.A. PharmacoEconomics. 5 (Supplement 1): 34-37. 1994.

OBJECTIVE: To estimate the economic impact of obesity in the United States by examining the economic costs attributable to obesity for major chronic disorders, including type 2 diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: The estimated direct cost of obesity-associated disease in 1990 was $45.8 billion; the indirect cost was estimated to be $23.0 billion. The cost of type 2 diabetes attributable to obesity was 12.7 billion dollars ($8.835 billion for direct costs and $3.890 billion for indirect costs associated with lost productivity resulting from excess mortality).

RECOMMENDATION: Programs aimed at preventing weight gain in children and adults may help contain rapidly rising health care costs.

ABSTRACT: The authors estimated the economic impact of obesity (body mass index $ 27.8 kg/m2 for men or 27.3 kg/m2 for women) in 1990 U.S. dollars. Direct costs were derived from estimates for five major chronic disorders (type 2 diabetes, gallbladder disease, cardiovascular disease, cancer, and musculoskeletal disease) for which obesity increases risk. The authors estimated the percentage of costs attributable to obesity for these disorders to be 57, 30, 19, 2.3, and 10, respectively. For diabetes, direct costs included routine care, cost of complications, and costs from an increased prevalence of other conditions (e.g., peripheral vascular disease, cerebrovascular disease). The authors used a published estimate of the 1980 U.S. health care expenditure for type 2 diabetes as the basis for making the cost estimate for this disorder. They estimated that, for 1990, direct costs for type 2 diabetes that were attributable to obesity were $8.8 billion. Indirect costs in this study were divided by causes: (a) lost productivity as a result of obesity-related illness and (b) excess mortality. The authors estimated that 52.59 million work days were lost in 1988 because of obesity-related disease; they estimated the cost of this lost productivity to be $4.06 billion. They did not, however, allocate this cost to the five diseases studied. Costs of lost productivity from excess mortality for the five diseases were estimated at $18.94 billion, including $3.89 billion for type 2 diabetes.

201

TITLE: Costs of Insulin-Dependent Diabetes Mellitus. Simell, T.T.; Sintonen, H.; Hahl, J.; Simell, O.G. PharmacoEconomics. 9(1): 24-38. January 1996.

OBJECTIVE: To review the costs to the individual and society of initial treatment, follow-up, and late treatment for type 1 diabetes; to assess the potential for decreasing these costs.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Review of studies.
Perspective: Societal

CONCLUSION: The costs of type 1 diabetes peak at the diagnosis of the disease and again with the development of complications. Intensive patient education, shortened hospitalizations for initial treatment, and maintenance of good metabolic control and mental health are the primary means of containing or reducing costs associated with type 1 diabetes until primary prevention is available.

RECOMMENDATION: None.

ABSTRACT: The authors review the epidemiology and the direct and indirect costs of type 1 diabetes. Their review considers total costs as well as costs associated with three clinical stages: initial treatment, follow-up after initial treatment, and late treatment. The incidence and prevalence of type 1 diabetes vary, with the highest rates occurring in Scandinavia and the lowest rates in the Far East. Costs are difficult to ascertain because there are no prospective or longitudinal studies of the total long-term costs of type 1 diabetes. Available data suggest that the highest costs of type 1 diabetes are associated with initial and late treatment stages of the disease. The mean length of initial hospitalization at the clinical onset of type 1 diabetes ranges from 2 days in the United States to 6 weeks in northern and eastern Europe. A Finnish study found that shortening the initial hospitalization from 23 to 9 days had no effect on patient outcome at 2 years. Costs accumulate slowly after initial treatment; for children with type 1 diabetes in one study, total costs of self-care over a 2-year period were less than the cost of 1 inpatient day at a university hospital. Late-phase treatment is characterized by a rapid increase in costs because of complications, including retinopathy requiring laser treatment; renal disease, requiring dialysis; cardiovascular disease, which is responsible for more than 20 percent of deaths among type 1 patients; and neuropathy, which can lead to amputation. The estimated cost to provide tight metabolic control in type 1 patients in the Diabetes Control and Complications Trial was $3,700 (U.S.) per year, which would increase current control costs by 120 percent. However, the authors assert that the potential long-term personal and economic savings of continuing tight control and reducing complications are enormous. 121 references.

202

TITLE: Costs of Temporary and Permanent Dis-ability Induced by Diabetes. Olivera, E.M.; Duhalde, E.P.; Gagliardino, J.J. Diabetes Care. 14 (7): 593596. July 1991.

OBJECTIVE: To evaluate indirect costs of diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Retrospective.
Methodology: Cost analysis.
Perspective: Societal.

CONCLUSION: Persons with diabetes but without complications and a control group of persons without diabetes had similar rates of absenteeism and yearly cost. In contrast, the costs of overall absenteeism and permanent disability were much higher in persons with diabetes and chronic complications.

RECOMMENDATION: Secondary prevention of diabetic complications might be an optimal approach to reducing the burden of diabetes.

ABSTRACT: From 1984 to 1986, the cost of temporary disability for patients with diabetes was studied in a group of La Plata University (Argentina) employees (n = 42 for both control group and group with diabetes); the cost of permanent disability for patients with diabetes was studied in a larger group (n = 2,763) of Buenos Aires government employees. The La Plata University group with diabetes was divided into those without complications (31 percent) and those with chronic complications (69 percent). Complications consisted of macroangiopathy (42 percent peripheral, 22 percent coronary, and 5 percent cerebral), peripheral neuropathy (33 percent), retinopathy (25 percent), and kidney lesions (11 percent). The number and causes of working days lost during the calendar year were determined for each individual. For temporary disability, average days per year lost in patients without complications did not differ significantly from the number for the age- and sex-matched control group. However, patients with chronic complications had a considerably higher rate (p < 0.05) of days lost than the control group or patients with diabetes without complications. The costs of permanent disability for the government employees were estimated by calculating the expected number of years to retirement age for each early retiree. Work production loss was discounted at a 6 percent rate. Average work production lost for 115 patients disabled by diabetes was 11 years (n = 115; 48 women, 67 men). At an annual cost of $23,660 (U.S. dollars), the total cost for these employees was $2,720,900. 3 tables, 17 references.

203

TITLE: Diabetes. Vaughan, J.P.; Gilson, L.; Mills, A. In: Disease Control Priorities in Developing Countries. Jamison, D.T; Mosley, W.H.; Measham, A.R.; Bobadilla, J.L.; eds. New York: Oxford University Press. 1993. pp. 561-576.

OBJECTIVE: To summarize information for developing countries on the incidence and development over time of diabetes as well as the direct and indirect costs of this disease.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Literature review.
Perspective: Societal.

CONCLUSION: Resource allocation to improve the diagnosis and treatment of diabetes in developing countries needs to be fitted to the public policy and needs of individual countries.

RECOMMENDATION: Further research is needed to assess the incidence and prevalence of type 1 and type 2 diabetes in developing countries, to evaluate possible interventions to prevent type 2 diabetes, and to assess case management and financing strategies.

ABSTRACT: The authors review the epidemiology and economic burden of diabetes worldwide and especially in developing countries. Detection and reporting of new cases of diabetes depends on the availability and use of health care services, or on the result of large-scale population-based surveys. The authors briefly review the evidence for genetic and environmental factors in the development of type 1 and type 2 diabetes. The incidence of type 1 diabetes may be rising, but it is considered a rare disease in most developing countries and the very poor database makes the situation there uncertain. The incidence of type 2 diabetes has risen in the United States, Singapore, and Taipei, Taiwan; for Africa and Latin America, there are fewer data on incidence of this disease. Poor epidemiologic data and the failure to separate type 2 from type 1 diabetes makes it difficult to determine indirect costs in developing countries. In most of these countries, direct costs for diabetes are likely to be low. Reducing the incidence of new cases requires primary prevention strategies, and reducing complications requires early detection and improved case management. Prevention needs to include modification of behavior, improvement in health services and health education, and government regulation. However, the success of attempts to modify behavior is debated, and improvement in health services is expensive, especially for developing countries. The authors suggest strategies for developing countries based on both incidence of diabetes and income. Further research is needed to assess the incidence and prevalence of type 1 and type 2 diabetes in developing countries, to conduct studies of possible interventions for preventing type 2 diabetes (including broad-based noncommunicable disease control), to assess case management, and to consider financing mechanisms.

204

TITLE: Diabetes Care in a UK Health Region: Activity, Facilities and Costs. Alexander, W.D. South East Thames Diabetes Physicians Group. Diabetic Medicine. 5(6): 577-581. September 1988.

OBJECTIVE: To compare facilities and staffing for diabetes care in the Southeast Thames region with recommendations from the Royal College of Physicians and the British Diabetic Association, and to assess the cost of diabetes care for the region.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.

CONCLUSION: Total cost of diabetes was estimated at £ 21.6 million per year for the South-east Thames region. Physician staffing, patient education, and an ineffectual system of care have resulted in an unacceptably high hospitalization rate for uncomplicated diabetes and metabolic imbalances.

RECOMMENDATION: Regional and District Health Authorities need to give priority to establishing effective, formal diabetes care strategies.

ABSTRACT: The authors reported facilities and services available to patients with diabetes mellitus in 15 districts of the Southeast Thames Regional Health Authority and estimated the basic hospitalization costs for patients with diabetes based on 1985 Hospital Activity Analysis data. Cost estimates were £ 100 per day for inpatient care and £ 20 per visit for outpatient care, and estimates for insulin, sulphonylurea, and modified diet were £ 165, £ 80, and £ 13 per patient per year, respectively. Using recommendations from the Royal College of Physicians and the British Diabetic Association as a standard, mean deficits (with ranges) in the number of consultant physicians, consultant sessions per week, and nurse specialists were 0.9 (0 to 2.2), 6.6 (0 to 15), and 3.1 (0.4 to 6.0), respectively. Many districts provided no retinopathy screening, education program, computerized records, secretaries, or diabetes day unit. Of 11,857 hospitalizations for patients with diabetes, 4,185 had diabetes given as the principal cause; mean stay for all cases was 4.4 days. Among stays where diabetes was the principal cause, 76 percent had no mention of complications or were for ketoacidosis or coma, and 7.6 percent were for peripheral circulatory disorders (mean stay of 35.3 days) or neurologic complications. Estimated total out-patient costs were £ 4,489,320 ( £ 1,432,000 for visits, £ 1,772,100 for insulin, £ 1,145,600 for sulphonylurea, £ 139,620 for dietary intervention, and £ 3,057,320 for supplies); estimated inpatient costs, £ 17,074,080 ( £ 6,026,400 for patients admitted primarily for diabetes). Estimated excess inpatient costs for all patients with diabetes, using hospitalization costs of the whole population as a comparison, were £ 13,460,800. Investment of approximately 11 percent of excess costs would bring staffing to recommended levels and improve chiropody and dietetic services. The authors recommend that all regional and district health authorities adopt a strategic plan for diabetes care. 4 tables, 12 references.

205

TITLE: Diabetes Mellitus in Egypt. Arab, M. World Health Statistics Quarterly (Rapport Trimestriel de Statistiques Sanitaires Mondiales). 45(4): 334-337. 1992.

OBJECTIVE: To report findings from a series of surveys of the epidemiology of diabetes mellitus in Egypt; to discuss the cost of diabetes, its interaction with other health problems, and health care delivery in that country.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost-of-illness.
Perspective: Societal.

CONCLUSION: The direct and indirect costs of diabetes mellitus constitute a great burden on the economy of Egypt; a national program cannot provide free care for all Egyptians with diabetes.

RECOMMENDATION: Improvements in health care delivery in Egypt are needed to reduce the morbidity and mortality of diabetes mellitus.

ABSTRACT: The author reports that average prevalence of diabetes mellitus in Egypt is 4.3 percent, with distinct geographical variations. The incidence of type 1 diabetes per 100,000 in children below age 15 was found to be 8.3 in an urban population and 7.6 in a rural population. Socioeconomic factors, including religion, may have an impact on the prevalence of diabetes in Egypt. For example, Islam calls for healthy behaviors, including the limitation of food intake to meet one's real needs and encouragement of exercise. Schistosomiasis, a parasitic disease, affects an estimated 20 million Egyptians; this disorder affects metabolism and may be reflected as hypoglycemia or hyperglycemia or as insulin resistance or other problems. Total 1990 direct costs for treatment of diabetes were approximately $74.3 million (U.S. dollars). Average monthly costs for medication were $1.12 for oral hypoglycemic agents, $2.44 for insulin, and $2.90 for glucose self-monitoring supplies. Laboratory assays cost $22.20 per year, and private physician fees ranged from $9.00 to $60.00 per year. There are substantial government subsidies for basic medication. The costs of treatment for complications of diabetes were much higher, ranging from $2.90 per month for hypertension to $555.00 per month for advanced nephropathy. Basic hospital costs ranged from $0.88 to $3.80 per day for uncomplicated cases. Indirect costs of diabetes mellitus were estimated at $11.8 million per year for the effects of absenteeism on productivity; indirect costs due to premature death were not estimated. 2 tables, 7 references, French summary.

206

TITLE: Diabetes - The Cost of Illness and the Cost of Control: An Estimate for Sweden 1978. Jonsson, B. Acta Medica Scandinavica. 671 (Supplement): 19-27. 1983.

OBJECTIVE: To estimate the economic cost of diabetes mellitus in Sweden in 1978, and to address the question of possible savings from improved metabolic control.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost-of-illness.
Perspective: Societal.

CONCLUSION: Total direct and indirect costs for diabetes mellitus in Sweden in 1978 were estimated at 568 million and 749 million SEK, respectively. Management and control costs were estimated at 313 million SEK, while the cost of treatment of complications was estimated to be 1 billion SEK. There are great potential benefits to be expected from improved metabolic control.

RECOMMENDATION: Cost-beneficial improvements in management and control of diabetes should be undertaken to lessen the economic burden of diabetes.

ABSTRACT: For this study, direct costs for inpatients were estimated by multiplying number of hospital bed days for patients with a principal diagnosis of diabetes by average cost per bed day. Outpatient costs were developed from estimates of physician visits; cost estimates took into account differences between general practice and hospital outpatient departments. The cost of drugs was estimated from data on drug sales. Indirect costs were estimated separately for short-term illness, permanent disability, and premature death. Direct costs were estimated at 568 million SEK, including 358 million and 210 million SEK for institutional and noninstitutional care, respectively. Persons aged 65 years and older accounted for more than 50 percent of the direct costs and nearly 66 percent of the institutional costs. Total indirect costs were estimated at 749 million SEK, including 134 million SEK from short-term illness, 438 million SEK from permanent disability, and 176 million SEK from premature mortality. Lost production due to permanent disability accounted for 60 percent of the indirect costs and one-third of the total costs. Complications were more costly than management or control: The latter costs were estimated at 313 million SEK; estimated costs associated with complications were 1 billion SEK, including 255 million SEK in direct costs and 749 million SEK in indirect costs. Models that estimate potential savings from improved metabolic control are discussed. Simulations show that even with modest assumptions for beneficial outcomes, investment in improved control of diabetes will be cost beneficial. Limiting cases to those with a primary diagnosis of diabetes leads to an underestimate of the economic consequences of this disease. 2 figures, 14 tables, 18 references.

207

TITLE: Direct and Indirect Costs of Cardiovascular and Cerebrovascular Complications of Type II Diabetes. MacLeod, K.M.; Tooke, J.E. PharmacoEconomics. 8(Supplement 1): 46-51. 1995.

OBJECTIVE: To review the literature on the economic impact of macrovascular complications on diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Review of studies.
Perspective: Societal.

CONCLUSION: Much of the cost of type 2 diabetes is attributable to accelerated atherosclerosis and macroangiopathy.

RECOMMENDATION: None.

ABSTRACT: The authors review various studies on the economic impact of diabetes. In a U.S. study (Rendell et al. 1993), persons with diabetes accounted for 3.1 percent of the study group but 8.3 percent of overall health care charges. Adjusted odds ratios for various diagnoses for those with diabetes (versus those without) were 3.32 for ischemic heart disease, 3.14 for peripheral vascular disease, 2.83 for hypertension, and 2.26 for cerebrovascular disease. In a Finnish study (Aro et al. 1994), 50.7 percent of those with diabetes and 12.4 percent of a control population were hospitalized annually. In a Danish study of elderly patients (Damsgaard et al. 1987), macro-vascular disorders (cardiac, cerebrovascular, and peripheral vascular disease) accounted for 87.4 percent of bed days used for diabetic complications. In a U.S. study reported in the same year, Jacobs et al. found the risk of acute myocardial infarction and chronic ischemic heart disease to be 8.3 and 7.2 times as great among patients with diabetes as it was in the general population. In that study, which used patient records identified from the 1987 U.S. National Hospital Discharge Survey, cardiovascular complications accounted for 74 percent of hospitalization costs for the treatment of the late complications of diabetes. Huse et al. (1986) found that circulatory disorders (hypertension, ischemic heart disease, and cerebrovascular disease) accounted for 33.3 percent of U.S. direct costs for type 2 disease in 1986. These authors estimated the national economic cost of type 2 diabetes to be $19.8 billion for that year, with macrovascular disease accounting for 47 percent. 1 figure, 4 tables, 17 references.

208

TITLE: Direct and Indirect Costs of Diabetes in Minnesota in 1988. Roesler, J.; Walseth, J.; Bishop, D. Minnesota Department of Health, Minnesota Diabetes Surveillance Project. 18 pp. September 1990.

OBJECTIVE: To assess the total economic burden of diabetes in Minnesota in 1988.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost-of-illness.
Perspective: Societal.

CONCLUSION: Diabetes represented a significant economic burden in Minnesota in 1988, with total costs estimated at more than $300 million.

RECOMMENDATION: To reduce the economic burden of diabetes; interventions designed to prevent the complications of diabetes should be expanded.

ABSTRACT: The authors estimated costs in Minnesota for diabetes by applying the cost estimates from a study by the Center for Economic Studies in Medicine for the American Diabetes Association (1988) to the state's population with diabetes. For 1988, the incidence and prevalence of diabetes in Minnesota were estimated at 9,695 and 87,109 cases, respectively. A total of 1,119 deaths were attributed to diabetes. Total direct costs were estimated at $189.4 million. Of this amount, $154.0 million was for hospital-related costs and $11.9 million for nursing home costs; outpatient costs were estimated at $23.5 million. Estimates of components of hospital costs included $30.5 million for diabetes, $75.6 million for its chronic complications, $7.3 million for increased intensity of care, $36.1 million for additional length of stay, and $4.4 million for physician visits to inpatients. Indirect costs were estimated at $112.2 million; those associated with short-term morbidity were estimated at $2.0 million; with long-term disability, $44.6 million. Lost earnings due to premature death were estimated at $65.5 million. Total costs were therefore estimated at $301.5 million, of which direct costs accounted for 62.8 percent and indirect costs, 37.2 percent. 12 tables, 23 references.

209

TITLE: Direct and Indirect Costs of Diabetes in the United States in 1987. Center for Economic Studies in Medicine, Pracon Incorporated. American Diabetes Association, Alexandria, VA. 20 pp. 1988.

OBJECTIVE: To update previous estimates of the incidence and prevalence of diabetes in the United States; to measure health care use for diabetes in 1987 and associated direct costs; and to estimate for that year morbidity and mortality resulting from diabetes and their associated indirect costs.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: Diabetes represents a significant economic burden in the United States, with total economic costs estimated at $20.4 billion for 1987.

RECOMMENDATION: As most of the costs of diabetes are due to its chronic complications, programs designed to increase the early diagnosis and treatment of this disorder and enhance accessibility to care should be expanded with a goal of preventing complications.

ABSTRACT: The authors used the human capital approach of D.P. Rice et al. (1985) to estimate costs associated with diabetes. Data sources for the study included several government surveys: the National Health Interview Survey, National Hospital Discharge Survey, National Nursing Home Survey, and National Ambulatory Medical Care Survey. Prices were specified in 1987 dollars. There were an estimated 6.51 million patients with diabetes in the United States in 1987, and 564,868 cases of diabetes were newly diagnosed during the year. Total costs of diabetes in 1987 were estimated at $20.4 billion. Direct costs, which accounted for 47.1 percent of the total, were composed of $7.9 billion in hospitalization and nursing home expenses and $1.7 billion for outpatient care. Indirect costs, 52.9 percent of the total, were composed of $142 million attributable to short-term morbidity, $3.14 billion attributable to long-term disability, and $7.5 billion associated with premature mortality. 2 exhibits, 12 tables, 33 references, 1 appendix.

210

TITLE: Direct and Indirect Costs of Diabetes in the United States in 1992. Medical Technology and Practice Patterns Institute, Washington, DC. American Diabetes Association, Alexandria, VA. 27 pp. 1993.

OBJECTIVE: To report prevalence and incidence estimates for diabetes; to estimate health care use and expenditures for treating diabetes; and to estimate morbidity and mortality costs as well as associated indirect costs for diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: The total economic cost of diabetes in the United States in 1992 was estimated to be $91.8 billion.

RECOMMENDATION: The economic burden of diabetes may be reduced by programs aimed at increasing the use of preventive and diagnostic services, improving patient education, and increasing access to maintenance therapies and screening for complications.

ABSTRACT: The authors used a prevalence-based and human capital approach to estimate costs associated with diabetes. Data sources used to determine the prevalence of diabetes and the use of health services included the Centers for Disease Control, the National Hospital Discharge Survey, the National Nursing Home Survey, the National Ambulatory Care Survey, and the National Medical Expenditure Survey. Costs were expressed in 1992 dollars. In that year, an estimated 7.2 million persons had diagnosed diabetes. Total costs (direct plus indirect) for diabetes in 1992 were estimated at $91.8 billion. Direct costs accounted for 49.2 percent ($45.2 billion), with hospital, nursing home, and outpatient costs estimated at $37.2, $1.8, and $6.2 billion, respectively. (Outpatient costs included $2.9 billion in hospital outpatient visits, $1.2 billion in prescription drugs, $1.0 billion in physician visits, and $0.5 billion for medical equipment.) Indirect costs, which comprised 50.8 percent ($46.6 billion) of total costs, included $8.5 billion for short-term morbidity, $11.2 billion for long-term morbidity, and $27.0 billion for premature mortality. Persons aged 45 to 64 years accounted for 62.3 percent of total mortality costs (the value of productivity foregone because of diabetes). The authors note that although 40.5 percent of all expenditures for diabetes were attributed to inpatient hospital care, only 0.03 percent of expenditures went toward diet/nutrition counseling and only 1.0 percent to diagnostic testing. They describe several ways in which the direct medical costs were likely to be underestimated in the report. 17 tables, 40 references, 1 appendix.

211

TITLE: Economic Consequences of Diabetes Mellitus in the U.S. in 1997. Alexandria, VA: American Diabetes Association, 1998.

OBJECTIVE: To estimate both direct medical and indirect costs attributable to diabetes and to calculate total and per capita expenditures for people with and without diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: Estimated direct medical and in-direct costs were $44 billion and $54 billion, respectively.

RECOMMENDATION: None.

ABSTRACT: The authors used prevalence-based cost-of-illness methods and data from national health care expenditure surveys to determine costs attributable to diabetes. Direct medical costs were calculated as both the medical expenditures attributable to diabetes (i.e., cost from excess prevalence of both diabetes-related chronic complications [e.g., neurological disease, renal disease] and general medical conditions [e.g., liver disease, respiratory failure, malignant neoplasms] and total cost of all services for people with diabetes). Indirect costs were those related to foregone earnings because of disability and diabetes-attributable mortality. The 1987 National Medical Expenditure Survey was used to estimate mean expenditures for encounters where the primary diagnosis was diabetes; this survey was also used to estimate diabetes prevalence as well as chronic complications and general medical conditions. For 1997, 2.3 million hospital discharges were attributed to diabetes (287.2 per 1,000 people with diabetes). Of an estimated $44.1 billion in direct medical expenditures attributable to diabetes in that year, $27.5 billion was for inpatient care, $10.9 billion for outpatient services and home health care, $5.5 billion for nursing home care, and $0.2 billion for hospice care. Of the total direct expenditures, treating diabetes and its acute metabolic complications accounted for 17.4 percent; chronic complications, 26.8 percent. Of an estimated $54.1 billion in attributable indirect costs, premature mortality accounted for $17.0 billion and disability, $37.1 billion. Total medical expenditures for people with diabetes were estimated at $77.7 billion ($10,071 per capita), versus $540.6 billion ($2,669 per capita) for people without diabetes. 44 references.

212

TITLE: Economic Cost of Diabetes Mellitus: Minnesota, 1988. Minnesota Department of Health. Centers for Disease Control. Morbidity and Mortality Weekly Report (MMWR). 40(14): 229-231. April 12, 1991.

OBJECTIVE: To describe the economic impact of diabetes mellitus on the state of Minnesota for the year 1988.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: During 1988 in Minnesota, direct and indirect costs for diabetes mellitus approximated $189 million and $112 million, respectively. Chronic complications of diabetes mellitus accounted for more than half of the hospital days and cost more than $75 million.

RECOMMENDATION: To reduce the costs of diabetes mellitus, effort should be directed at preventing its chronic complications.

ABSTRACT: The author summarizes an analysis prepared by the Minnesota Diabetes Surveillance Project that estimated the economic impact of diabetes mellitus on Minnesota in 1988. Prevalence of diabetes mellitus was obtained from a previous population-based study; national sources were used to estimate hospitalizations, physician visits, nursing home stays, laboratory tests, outpatient care, and disability compensation. In 1988, the total cost of diabetes mellitus in the state was $301 million. The direct cost of diabetes mellitus, including diagnosis, treatment, hospitalizations, nursing home care, and outpatient care, was $189 million. Chronic complications accounted for 78,304 hospital days, more than half of total hospital days for diabetes mellitus. Hospitalizations for chronic complications of diabetes mellitus cost more than $75 million, and the indirect cost of diabetes mellitus was calculated at $112 million, with $2 million, $44.6 million, and $65.5 million attributable to short-term morbidity, long-term disability, and mortality, respectively. Prevention of some of the chronic complications of diabetes mellitus should result in a major reduction in its cost. Minnesota has developed a multifaceted plan to reduce the morbidity and disability that result from lower extremity amputations, diabetic eye disease, uncontrolled hypertension, and adverse pregnancy outcomes; the project includes measuring the burden of diabetes mellitus, implementing a statewide plan, and monitoring that plan's impact. 1 table, 10 references.

213

TITLE: The Economic Cost of Obesity: The French Situation. Levy, E.; Levy, P.; Le Pen, C.; Basdevant, A. International Journal of Obesity and Related Metabolic Disorders. 19(11): 788-792. November 1995.

OBJECTIVE: To estimate for France the direct and indirect costs of obesity as well as excess mortality from this condition.

CATEGORY: Expert opinion.

CONCLUSION: The economic cost of obesity in France is approximately 2 percent of all health care costs.

RECOMMENDATION: Prospective studies of the costs of obesity are needed that will measure relative risks adjusted for age, sex, socioeconomic conditions, and country-specific epidemiologic data.

ABSTRACT: The authors used a prevalence-based analysis of 1992 data to estimate the direct and indirect costs of obesity in France. Indirect costs were based on lost production due to obesity as indicated by Health Insurance System reimbursements for obesity-related causes. The prevalence of obesity (body mass index 27 Kg/m2 or greater) and morbid obesity (body mass index 30 Kg/m2 or greater) in the population aged 20 to 49 was 16.7 and 6.2 percent, respectively. Direct costs, based on prevalence of obesity and the relative risk of an obese person contracting a disease, were calculated for obesity, hypertension, myocardial infarction, angina pectoris, stroke, venous thrombosis, type 2 diabetes, hyperlipidemia, gout, osteoarthritis, gall bladder disease, colorectal cancer, breast cancer, genitourinary cancer, and hip fractures (this last disorder was considered cost-saving). The lowest estimates for relative risk were used. Total direct costs were French francs 66 billion or approximately 11 percent of French Health Care System costs, 20 percent of which were for type 2 diabetes health services. Obesity and related diseases accounted for 65, 25, and 10 percent of drug, physician, and laboratory expenditures, respectively. Total indirect costs were FF 575 million, versus FF 3.5 billion from all causes. A savings of FF 190 million was found because of fewer hip fractures. One-year mortality from obesity was 180,000 persons. Total direct costs of obesity amounted to about 2 percent of French Health Care System costs, a finding that accords with results of studies in other western countries. This estimate is conservative, in part because not all obesity-related diseases were included and the lowest values for relative risk were used. Prospective studies of the costs of obesity are needed that will measure relative risks adjusted for age, sex, socioeconomic conditions, and epidemiology for each country. 3 tables, 42 references.

214

TITLE: The Economic Costs of Non-Insulin-Dependent Diabetes Mellitus. Huse, D.; Oster, G.; Kilen, A.; Lacey, M.; Colditz, G. Journal of the American Medical Association (JAMA). 262(19): 2708-2713. November 17, 1989.

OBJECTIVE: To estimate the 1986 costs of morbidity and mortality attributable to type 2 diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: The estimated total economic cost of type 2 diabetes in 1986 was $19.8 billion, including $11.6 billion in health care expenditures and $8.2 billion in foregone productivity. Of type 2 diabetes-related health care expenditures, 59 percent were attributable to diabetes or its complications, and 41 percent to an excess prevalence of related conditions, for the most part circulatory disorders.

RECOMMENDATION: The significance of the public health and clinical problems associated with type 2 diabetes-related conditions must be recognized.

ABSTRACT: The authors estimated the total economic burden of type 2 diabetes using conventional prevalence-based cost-of-illness techniques and data from government surveys. The total economic cost of type 2 diabetes in 1986 was $19.8 billion, including $11.6 billion in health care expenditures and $8.2 billion in foregone productivity due to morbidity or mortality. Approximately $4.8 billion of these health care costs were attributable to treatment of other diseases, principally circulatory disorders ($3.8 billion) that were attributable to type 2 diabetes. Men younger than 65 years accounted for $7.0 billion in economic costs; women aged 65 years and older, $5.9 billion; men over 65, $3.5 billion; and women under 60, $3.4 billion. Per case, annual health care expenditures attributable to type 2 diabetes ranged from $1,274 among men younger than 65 years to $3,078 among women 65 aged years and older. In addition to the significant economic burden of type 2 diabetes, the human toll of type 2 diabetes was staggering: 144,000 premature deaths were attributable to this disorder in 1986, representing 6.8 percent of all U.S. mortality and a loss of 1,445,000 years of life. This mortality estimate is more than four times greater than the previously reported estimate of 35,000 deaths due to all kinds of diabetes. Of the 144,000 deaths, cardiovascular disease was the cause in 124,000. 3 figures, 5 tables, 33 references.

215

TITLE: Economic Costs of Obesity. Colditz, G.A. American Journal of Clinical Nutrition. 55(2 Suppl): 503S-507S. February 1992.

OBJECTIVE: To estimate costs attributable to obesity for type 2 diabetes, hypertension, cardiovascular disease, gall bladder disease, colon cancer, and breast cancer.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: Estimated 1986 costs for the six diseases studied totaled $39.3 billion, representing 5.5 percent of the cost of illness in that year.

RECOMMENDATION: Programs that target weight gain avoidance in middle and older age should be implemented.

ABSTRACT: For this prevalence-based estimate of the economic cost of obesity (a body mass index $ 27.8 kg/m2 for men or 27.3 kg/m2 for women), the author includes both direct and indirect costs. A discount rate of 4 percent was applied to indirect costs. The cost for type 2 diabetes included routine care for uncomplicated illness, costs related to morbidity and mortality from complications, and costs from excess prevalence of other diseases (e.g., renal disorders). A published estimate of the 1980 cost of type 2 diabetes was adjusted to 1986; direct costs for the latter year were estimated at $11.6 billion; indirect costs, at $8.2 billion. Using information from the Nurses' Health Study, the author estimated that 57 percent of these costs ($11.3 billion) were attributable to obesity. Total estimated 1986 costs attributable to obesity for the 6 diseases studied were $39.3 billion. If the costs of obesity related to musculoskeletal disorders (about 50 percent of total costs for those problems) were included, another $17 billion would be added to this total. 5 figures, 36 references.

216

TITLE: Economic Evaluations of Type II Diabetes. Leese, B. PharmacoEconomics. 8 (Supplement 1): 23-27. 1995.

OBJECTIVE: To discuss the cost-of-illness methodology as an approach to studying the costs of diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Review of studies.
Perspective: Societal.

CONCLUSION: Existing studies highlight the difficulties involved in comparing diabetes costs between countries with different health care systems and the importance of defining the study population and the sources of costs. Relatively few studies make suggestions about how their cost estimates might be used.

RECOMMENDATION: None.

ABSTRACT: Most economic analyses of diabetes mellitus have been descriptive and used a cost-of-illness methodology. Evaluative studies, in contrast, use cost-effectiveness, cost-benefit, or cost-utility methodologies. Cost-of-illness studies bring together three kinds of costs — direct, indirect, and intangible — to measure the economic burden on society of a disease. These studies are usually based on disease prevalence, less often on its incidence. Incidence-based studies, which show where cost savings could be obtained from changes in treatment, are preferable but harder to perform. Direct costs are the easiest of the three kinds of costs to collect, but relevant costs may be excluded, or these costs may be incorrectly estimated or valued inappropriately. Indirect costs, which are the subject of considerable debate, measure the cost to society of illness, disability, and premature mortality. Intangible costs, which are more subjective, have rarely been considered in cost-of-illness studies because of difficulties in assigning costs to factors such as stress, pain, and anxiety. The valuation of life necessary to derive indirect costs is the main area of debate about the cost-of-illness methodology. One way of valuing life is to consider market earnings; here the human capital method is the most widely used approach. However, this method is biased (e.g., toward the unemployed). Willingness to pay is another approach, but it has found little acceptance. The friction method is a third approach, in which only production losses during the period required to replace the sick worker are costed. The most serious criticism of cost-of-illness studies is that they do not indicate where resources should be devoted to a particular disease. The author reviews various studies of the direct costs of type 2 diabetes and notes that most studies of the costs of diabetes have not distinguished between type 1 and type 2. 37 references.

217

TITLE: Economic Impact of Diabetes. Entmacher, P.S.; Sinnock, P.; Bostic, E.; Harris, M.I. In: Diabetes in America: National Diabetes Data Group. Chapter 32. U.S. Department of Health and Human Services, 1984.

OBJECTIVE: To review estimates of the direct and indirect costs of diabetes in the United States.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost-of-illness.
Perspective: Societal.

CONCLUSION: Diabetes represents a significant economic burden in the United States, with the estimated total cost of diabetes in 1984 equaling $13.75 billion.

RECOMMENDATION: None.

ABSTRACT: The authors discuss the direct and indirect costs of diabetes in the United States. Using the human capital approach of Rice (1966) and data derived from surveys of the National Center for Health Statistics, they estimate that the cost of diabetes rose from $2.6 billion in 1969 to $13.75 billion in 1984. They state that this increase was largely due to inflation and the increasing number of persons with diabetes. In 1984, estimated direct costs for medical and health care services were $7.4 billion; estimated indirect costs attributable to disability were $4.4 billion and $1.9 billion to premature death. An estimated $3.5 billion in direct costs were attributable to care in short-stay hospitals. The increase in total expenditures for diabetes from 1969 to 1984 was proportionate to the rise in overall health care expenditures during that time. The proportion of the total cost of diabetes attributable to direct medical care expenditures rose during the time period from 38 percent to 54 percent; the prevalence of diagnosed diabetes rose from 3.2 million in 1969 to 5.5 million in 1980 (no figures were given for 1984). In a 1977 national survey, 99 percent of persons with diabetes had a direct medical care expense versus 87 percent for those without diabetes. Per capita expenses for persons with diabetes who had an expense were 2.7 times higher than those for persons without diabetes. 8 tables, 2 appendices, 10 references.

218

TITLE: Economic Impact of Diabetes. Javitt, J.C.; Chiang, Y-P. In: Diabetes in America: National Diabetes Data Group, 2nd Edition. Chapter 30, 11 pages. 1995.

OBJECTIVE: To review studies of the economic costs of diabetes and examine methodological differences between published analyses.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost of illness.
Perspective: Societal.

CONCLUSION: The economic costs of diabetes are substantial, with estimates of total costs as high as $92 billion per year. However, comparisons between the estimates of this study and others are made difficult by several methodological issues.

RECOMMENDATION: As health care expenses escalate, the cost of treating diabetes should be considered from the perspective of transfer payments and tax losses.

ABSTRACT: The authors review recent published estimates of the economic cost of diabetes in the United States, update and synthesize them, and discuss methodological differences between these studies. Most studies of the economic costs of diabetes conducted in the 1980s used the human capital approach to valuing human life and relied on prevalence-based annual cost estimates. Methodological issues discussed by the authors include the attribution of cost to underlying conditions, valuing human life and health, and estimating volume of medical services; variation in these methodologies makes comparisons between studies difficult. In four 1980 studies, estimates (in 1990 dollars) of the total direct cost of diabetes ranged from $9.3 to $13.8 billion; indirect cost estimates ranged from $7.9 to $15.8 billion. Hospitalization accounted for most of the direct costs in all four studies. The American Diabetes Association published a study for the year 1992 in which it estimated direct medical costs of diabetes and its complications at $45.2 billion and indirect costs due to disability and premature death at $46.6 billion, for a total economic cost of $91.8 billion. This total, more than three times as great as the highest of the four earlier studies described, in part reflects increases in medical care costs above the inflation index used as well as more intensive and expensive technology. Important differences in the methodology of the 1992 analysis were the inclusion of cost components (e.g., home health visits, glucose monitors) not included in other studies, more comprehensive ascertainment of services due to systemic comorbid conditions attributed to diabetes, and morbidity databases that better capture contributing causes of death. The study attributed 344,914 deaths to diabetes in 1992 (80,339 were attributed to the disease in an analysis for 1987). Studies indicate that patients with diabetes use a much higher proportion of medical services and incur much higher expenses than do persons without diabetes 5 tables, 22 references.

219

TITLE: Economic Impact of Type II Diabetes Mellitus. Krosnick, A. Primary Care: Clinics in Office Practice. 15(2): 423-432. June 1988.

OBJECTIVE: To document the direct and indirect costs associated with type 2 diabetes in the United States.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Review of studies.
Perspective: Societal.

CONCLUSION: For 1988, total costs associated with type 2 diabetes were estimated at $20 billion: $10.5 billion for direct costs due to diagnosis and treatment and $9.5 billion for indirect costs associated with disability and mortality.

RECOMMENDATION: A concerted national effort is needed to address the morbidity, mortality, and economic burden of type 2 diabetes.

ABSTRACT: The author reviews recent studies of the direct and indirect costs associated with type 2 diabetes in the United States. Type 2 diabetes represents 85 to 90 percent of all diabetes cases; an estimated 50 percent of persons with type 2 diabetes are undiagnosed and/or untreated. In addition, the prevalence of diagnosed type 2 diabetes is rising steadily. In 1983, diabetes accounted for 7.2 percent of all hospitalizations nationally, up from 4.6 percent in 1971. In 1980, patients with diabetes made approximately 13 million to 16 million physician visits and 3 million additional contacts in clinics or emergency rooms or by telephone. Patients with diabetes currently average approximately $1,200 in pharmacy purchases annually, which is 3 to 8 times higher per year than persons without diabetes. Nursing home care for patients with diabetes is estimated to cost between $5 and $6 billion annually; the estimated total cost for hospital care in 1980 was $2.2 to $6.5 billion. Diabetes is the leading cause of blindness for persons aged 20 to 74; the annual cost of blindness due to diabetes is estimated at $75 million for lost income and welfare benefits. 1 table, 16 references.

220

TITLE: Financing the Care of Diabetes Mellitus in the U.S.: Background, Problems, and Challenges. Bran-some, E.D. Diabetes Care. 15(Supplement 1): 1-5. March 1992.

OBJECTIVE: To review the concerns of the diabetes community that led to the Second National Conference on Financing the Care of Diabetes Mellitus in the 1990s and to review topics included in that conference.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Commentary.
Methodology: None.
Perspective: Health care system.

CONCLUSION: Papers presented at the conference focused on the demographics of diabetes in the United States, standards of care, economic issues, and related topics.

RECOMMENDATION: None.

ABSTRACT: The author reports that the concerns that led to the conference included inadequate reimbursement for outpatient care interventions, below-standard quality of care of patients with diabetes, and confusion among patients and health care professionals about reimbursement changes. The author reviews the issues of the demographics of diabetes, the economic impact of diabetes, quality care, and reimbursement; he also briefly discusses papers presented at the conference. Approximately 6.6 percent of the population has diabetes. The American Diabetes Association estimated direct costs of diabetes care in 1987 to be about $9.6 billion, of which almost $7 billion was for inpatient care in acute short-stay hospitals. Indirect costs associated with short-term morbidity, long-term disability, and premature mortality were estimated at $10.8 billion. Reimbursement remains inadequate for preventive services and patient education, and there is evidence that limited access to care contributes to premature mortality in young adults with type 1 diabetes. The diabetes community has had difficulty convincing third parties that preventive care is cost effective. Quality assurance is emerging as a concern of accrediting organizations and the Health Care Financing Administration. Reimbursement procedures relevant to managed care programs and the Medicare Revised Fee Schedule are addressed. 2 figures, 3 tables, 12 references.

221

TITLE: The Indirect Costs of Morbidity in Type II Diabetic Patients. Persson, U. PharmacoEconomics. 8(Supplement 1):28-32. 1995.

OBJECTIVE: To estimate the indirect costs of type 2 diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Societal.

CONCLUSION: Excess indirect costs (value of lost production) averaged $7,000 (US dollars) per person with diabetes.

RECOMMENDATION: Estimates of the economic burden of diabetes must take into account the excess risk of diabetes-related illness, not just the costs of diseases that are related to diabetes.

ABSTRACT: The author reports estimates of indirect costs developed by Olsson et al. (1994). The population of Vetlanda, a town in southern Sweden, served as the study site; the age- and sex-adjusted prevalence of diabetes was 3.0 percent. The records of all diabetes patients aged 20 to 64 years (n = 285) were included (general retirement age in Sweden is 65). Of this group, 139 were treated with insulin (all 86 patients with type 1 and 53 of the 199 patients with type 2). To estimate the excess prevalence of sick days and early retirement in the study population, data for the 285 study patients were compared with data for the town's entire population aged 20 to 64 (including the study patients). The expected number of sick days for the study population was determined by multiplying annual sick days for each sex and age group in the general population by the number of insulin-treated and noninsulin-treated men and women with diabetes. Similar calculations were carried out for permanent disability (premature retirement). The cost of production losses was measured by multiplying lost time by average wages plus social security contributions, in 10-year age and sex groups. In an analysis adjusted by age and sex, insulin-treated men had 75 percent more sick days and insulin-treated women 31 percent had more sick days than expected. Among those not treated by insulin, sick days were near the expected number among women and slightly below expected among men. For subjects aged 40 years and above, the number of permanently disabled individuals was three times higher than expected for diabetic men. The annual per-patient excess cost of lost production due to short-term illness and permanent disability was estimated to be SEK 40,000 (US $7,000). Permanent disability among patients aged 40 to 64 years accounted for 92 percent of these costs. 3 tables, 6 references.

222

TITLE: Profiling Overweight Patients in the U.S. Navy: Health Conditions and Costs. Hoiberg, A.; McNally, M.S. Military Medicine. 156(2): 76-82. February 1991.

OBJECTIVE: To identify the health conditions in a sample of Navy men hospitalized in 1974 to 1984; to compare these disorders with those reported in the literature; and to examine obesity-related costs in terms of days hospitalized and career outcomes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Societal.

CONCLUSION: Conditions common among the obese patients included hypertension, alcoholism, and diabetes.

RECOMMENDATION: Assign overweight individuals to a weight reduction program at an early stage of their obesity.

ABSTRACT: Researchers studied 518 enlisted Navy men with a primary diagnosis of obesity and 1,092 Navy men with a secondary or additional diagnosis of obesity who had hospital stays in 1974 to 1984. A 10 percent sample of male Navy patients who had not been diagnosed as obese served as the comparison group. The most common diagnoses in the obese group were hypertension, alcoholism, diabetes mellitus, respiratory system symptoms, and chronic ischemic heart disease. In the comparison group, alcoholism, hernia, personality disorders, internal derangement of joint, and other cellulitis/abscess were the most common. For 20.2 percent of hospital days in the obese group, obesity was listed as the primary diagnosis. At $200 per day, the cost of treatment over the 11-year period would have been $2,115,000. About 25 percent of separations and retirements in the obese group were attributed to a disability (primarily diabetes mellitus), to chronic ischemic heart disease, or to hypertension. According to the authors, the study supports implementation of weight reduction programs. With stricter adherence to assigning overweight personnel to weight reduction programs, they assert that the Navy Medical Department can expect to reduce costs associated with obesity-related hospitalizations and physical disability separations. 5 tables, 19 references.

223

TITLE: Social and Economic Costs of Diabetes: An Estimate for 1979. Platt, W.G.; Sudovar, S.G. Pracon Incorporated, Washington, D.C. Home Health Care Group, Ames Division, Miles Laboratories, Inc. 20 pp. 1979.

OBJECTIVE: To provide detailed estimates of the direct and indirect costs associated with diabetes mellitus in the United States in 1979.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost-of-illness.
Perspective: Societal.

CONCLUSION: The authors estimated that 8.3 million persons had diabetes in 1979 and that they generated total costs (direct and indirect) of $15.7 billion in that year.

RECOMMENDATION: None.

ABSTRACT: The authors derived cost estimates for 1979 from earlier governmental and commercial statistics. The authors estimated there were 8.26 million persons with diabetes in the United States in 1979, 1.94 million with type 1 diabetes, 2.91 million with type 2 diabetes controlled by medication, 1.62 million with type 2 diabetes controlled by diet, and 1.78 million with undiagnosed diabetes. Total costs generated by these patients were estimated at $15.7 billion. Direct economic costs accounted for 36 percent of this total, with $1.95 billion associated with short-stay hospital and nursing home costs, $3.28 billion associated with noninstitutional costs, and $0.41 billion associated with complications. Indirect costs due to morbidity, mortality, and complications were estimated to be $10.03 billion, 64 percent of total costs. Morbidity (measured in work disruption) accounted for most of the indirect costs. Total costs for the 1.94 million patients with type 1 diabetes were estimated to be $4.8 billion. Direct costs accounted for 37 percent and indirect costs for 63 percent of total costs for these patients, for whom estimated per capita costs were $2,453. 7 figures, 5 tables, 4 appendices, 10 references.

224

TITLE: The SocioEconomic Cost of Diabetic Complications in France. Triomphe, A. Diabetic Medicine. 8 Symposium: S30-S32. 1991.

OBJECTIVE: To measure the direct cost of diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.

CONCLUSION: Average per-patient medical costs (in French francs [FFr]) in 1984 were 12,178 for type 1 diabetes and 6,908 for type 2 diabetes.

RECOMMENDATION: None.

ABSTRACT: An evaluation of the direct costs of diabetes was conducted in 1984. One hundred and nine sample patients with diabetes who resided in Paris were assessed; direct costs (e.g., home visits, laboratory tests, drugs, hospitalization, paramedical services) were analyzed. The average value (in FFr) of the total medical services used was 12,178 for patients with type 1 diabetes, 6,908 for patients with type 2 diabetes. Omitting sick leave, the figures were 7,711 for type 1 and 5,892 for type 2 diabetes; the average cost for a sample of the general French population was 6,462. The higher figure for type 1 (versus type 2 diabetes) patients was principally due to higher drug costs among type 1 patients. Hospitalization accounted for 34 percent of medical costs incurred by type 1 patients, 40 percent by type 2 diabetes patients, and 51 percent by patients in the national sample. The costs of diabetic complications were not assessed in this study because they were not well documented. Reports from the United States and Sweden indicate that hospitalization accounted for a relatively higher percentage of medical costs for patients with diabetes in those countries than in this study. 4 tables, 10 references.

225

TITLE: Structure and Costs of Health Care of Diabetic Patients in Finland. Kangas, T.; Aro, S.; Koivisto, V.A.; Salinto, M.; Laakso, M.; Reunanen, A. Diabetes Care. 19(5): 494-497. May 1996.

OBJECTIVE: To examine the structure and direct costs of health care for Finnish patients with drug-treated diabetes.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.

CONCLUSION: Direct medical costs (including insulin) for these patients accounted for 6 percent of all health care budgets in Finland in 1989 and 13 percent of total inpatient hospital days; direct costs per patient ($3,941) were three times higher than for patients without diabetes ($1,323).

RECOMMENDATION: Treatment strategies for diabetes should emphasize high-quality ambulatory care to improve blood glucose control, thereby reducing both the personal and economic burden of this disease.

ABSTRACT: Outpatient care was evaluated by analyzing 30,266 questionnaires completed by patients with diabetes who obtained their medications through pharmacies during a 7-week period in 1989. Data on hospital inpatient care were derived by linking data from the Hospital Discharge Register covering the years 1987 to 1989 and the Central Drug Register to identification numbers assigned to each Finnish citizen. Analysis of the questionnaires showed that 31 percent of survey respondents were treated by insulin alone, 63 percent by oral medication alone, and 6 percent by a combination of the two. The mean number of physician visits for diabetes among respondents was 3.6. During the study period, patients with diabetes used an average of 1.5 million inpatient hospital days, which represented 13 percent of the total hospital days in Finland. Total direct cost of health care for patients with diabetes averaged $3,941 per year. Of this, 80.8 percent was for hospital inpatient care, 8.8 percent for diabetes medications, 8.0 percent for ambulatory care, 2.1 percent for self-care equipment, and 0.3 percent for medical rehabilitation. The value of self-care was estimated at up to 48,282 working months per year, equivalent to $980 (U.S. dollars) per patient per year; however, this cost was not included in the calculation of ambulatory costs. The direct costs of health care for patients with diabetes represented 5.8 percent of the total health care costs in Finland in 1989 and were three times higher than the average costs of care for patients without diabetes. 1 figure, 1 table, 22 references.

226

TITLE: Systematic Care of Diabetic Patients in One General Practice: How Much Does It Cost? Koperski, M. British Journal of General Practice. 42(362): 370-372. September 1992.

OBJECTIVE: To assess the cost of conducting a monthly health promotion session focused on diabetes within a general practice consisting of seven physicians, two trainees, and two nurses.

CATEGORY: Cost of diabetes (direct and indirect).

Type of Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.

CONCLUSION: The costs of running a diabetic care day in a London general practice were greater than remuneration.

RECOMMENDATION: An adequate remuneration package for general practitioners could improve care for many patients with diabetes.

ABSTRACT: The author analyzed the direct costs for a London, England, general practice of seven physicians, two trainees, and two nurses for monthly "diabetic days" that emphasized care of patients with diabetes. To calculate the cost of physician time, the number of targeted consultations in 1987 was multiplied by the scheduled consultation time, then divided by the total number of office hours booked for 1 year for a full-time general practitioner. The result was multiplied by £ 31,105, the average net remuneration established by the Doctors and Dentists Review Body in 1989. The hours worked by nurses and clerical staff on diabetic days were divided by the total hours worked per year, then multiplied by their annual salary plus employer's contributions. Without family health service authority reimbursement, costs for 77 patients to make a total of 117 visits on diabetic days were £ 1,340.97 for physicians, £ 1,214.93 for nurses, £ 357.66 for clerical staff, £ 41.32 for 16.8 square meters of building space, £ 45.10 for stationery, and £ 1,465.66 for dietitian and chiropodist consultation, for a total of £ 4,465.66. With family health service authority reimbursement, costs for nurses and clerical staff were reduced to £ 330.01 and £ 97.13, respectively, lowering the total cost to £ 1,854.53. Thus, the cost per patient visit was £ 38.17 to the National Health Service and £ 15.85 to the practice. Diabetic days earn a practice £ 1,080.00 annually (1991 prices) through a health promotion incentive, well below the cost to the practice surveyed. Furthermore, there is no payment for setting up these clinics. The present system encourages low-quality, low-cost care. 1 table, 18 references.

Back to Economics of Diabetes Mellitus Indexes

Top of page

Privacy Policy | Accessibility

Home | About the Program | Site Map | Contact Us

CDC Home | Search | Health Topics A-Z

United States Department of Health and Human Services
Centers for Disease Control and Prevention
National Center for Chronic Disease Prevention and Health Promotion
Division of Diabetes Translation