Analysis of Direct Cost of Standard Compared with Intensive Insulin Treatment of Insulin-Dependent Diabetes Mellitus and Cost of Complications

Z. Stern and R. Levy
Acta Diabetol 33:48-52, 1996

Summary and Commentary by
Frank Vinicor, MD, MPH

Objective. To assess the direct costs of standard versus intensive insulin therapy, and the direct costs of diabetes complications under both therapies, for patients with type I diabetes.

Design. The authors developed an imaginary patient who contracted type I diabetes before age 20 years and was followed over a period of 35 years. A comparison of the direct costs of standard therapy versus intensive therapy was performed along with a comparison of the direct costs of both microvascular and macrovasular complications during the two treatment modes.

Intensive therapy was defined as insulin administered 3–4 times per day by injections or through an external pump, four blood glucose tests per day, monthly clinic visits including a glycohemoglobin test, and support from an educator and a diabetes specialist. Standard therapy was defined as 1–2 insulin injections per day, once daily SMBG, and physicians’ visits, including glycohemoglobin tests, 3–4 times per year.

Measurements. The authors measured the direct costs associated with each treatment and its complications. The direct costs of the complications were estimated using the probability according to which a patient would develop complications throughout the 35-year course of illness in two tracks, standard or intensified treatment.

Costs of insulin therapy, self-monitoring of blood glucose, outpatient visits, laboratory tests, and diabetes complications were used for the comparison of direct costs of treatment and direct costs of complications. An annual interest rate of 6% was used to adjust for the time span of 35 years.

Results. The annual direct cost of the standard insulin treatment in a typical type I diabetes patient was $1,184 (in 1995 prices). In comparison, the direct cost of intensified insulin treatment was $3,329 per year (in 1995 prices), about 3 times more, for a difference of $2,145 per patient-year. When the direct costs of basic treatment in both tracks for a 35-year period were compared, the cost of treatment in the standard track totaled approximately $41,000, while the cost of treatment in the intensified track amounted to approximately $116,000, a difference of about $75,000.

When the direct costs of complications were evaluated for standard treatment versus intensified treatment during a 35-year period, (based on figures compiled for an 8-year period), the investigators found that complications costs related to standard therapy were $288,000, while costs related to intensive therapy were $234,482; therefore intensive therapy entailed lower complications costs (by $53,520).

Thus, the total direct costs of the standard treatment were $329,400 versus total direct costs of $350,980 for the intensified treatment. The investigators report that, at an annual interest rate of 6%, intensive therapy supports lower complications costs (by $20,900) than the standard therapy, while the total cost of the standard therapy is $132,900 versus intensive therapy costs of $151,900—a difference of $19,000.

Conclusions. In view of the direct costs of intensified treatment, it is recommended that government, health-care institutions, and insurance providers initiate research programs and appoint appropriate committees to elaborate guidelines to identify, select, and treat the proper patients who will benefit from intensified therapy.

Commentary

During the past decade, three central and interrelated issues have dominated health-care discussions and decisions: cost, quality of care, and access.¹ Among these issues, cost—the economic aspect of health care—has assumed a growing and prominent seat at the health-care decision table.² This is particularly true for diabetes because ¹) the condition is very common and serious,^{3 2}) it is very expensive, with the direct costs of care for people with diabetes in considerable excess of the prevalence of this disease,^4-6 and ³) there is an emerging consensus that a large percentage of the burden (and thus costs) of diabetes can be reduced, if not prevented.⁷

While many questions have been raised regarding if, how, and how much economics should be a factor in health decisions,⁸ cost considerations remain paramount in individual, organizational, and societal perspectives on health. In performing and interpreting economic analyses, however, at least two aspects must be carefully considered: 1) the quality of the economic investigation,⁹ and 2) the underlying assumptions behind the analyses.¹⁰ Appropriate and helpful economic studies are complex, difficult, and challenging.⁹ And the basic assumptions underlying various analyses color the meaning of their results.¹⁰

Where, then, does this study by Stern and Levy fit in? Three general types of economic analyses exist: 1) cost-identification (What does Disease X cost the nation/hospital/HMO/etc.?), 2) cost-effectiveness (What are the results [in common clinical terms, such as decreased hospital days] of having spent $Y on Disease X?, and 3) cost-benefit (What are the results in dollars of having spent $Y on Disease X?).⁹

The investigation by Stern and Levy seems to compare two cost-identification efforts: 1) the cost of intensive treatment and 2) the cost of specific diabetes complications. The authors relied on the results of the Diabetes Control and Complications Trial (DCCT)¹¹ to determine the reduction in frequency of these complications with intensive treatment. Thus, Stern and Levy performed the most basic type of economic analyses, using the results of other studies to compare so-called “intensive” versus “standard” care over 35 years on economic outcomes. It is not clear from whose perspective the question is being asked.¹²

The authors determine that the increased expense of intensive treatment in type I diabetes is offset (approximately) by decreased costs associated with the diminished likelihood of specific diabetes complications when compared to standard treatment, which is itself associated with lower expenses for glycemic control but greater costs for complications. The authors conclude that “the regime of long-term intensified insulin treatment, although promising individual benefits, may have a relatively negative influence on the future total health-care costs.” (p. 50). This conclusion is not obvious to me.

How do Stern and Levy’s results, conclusions, and interpretations compare to similar investigations?

1. In general, most studies suggest that, at least for chronic disease, preventive interventions are at best cost-reasonable.¹³ In other words, few actions will actually save money with time, but most will not cost much over the long haul.¹⁴ In large part, this is due to the impact of “discounting” on chronic disease investments.¹² (Dis-counting reflects the fact that since people would rather have dollars now than in the future, one must increase the value to future dollars to overcome the desire to have them now. To do economic analysis, all future dollars must be discounted to allow comparison to present investments.) Thus, the approximately equal “grand totals” for intensive and standard treatment are neither surprising nor, in my judgment, do they justify the conclusion that intensive approaches are not “cost-beneficial.”

2. That “intensive approaches” are more expensive is not surprising,¹⁵ and is consistent with the “pay me now or pay me later” philosophy. In this analysis, the authors seem to periodically equate “intensive insulin” and “intensive treatment.” These are not the same. Indeed, in the DCCT, the amount of insulin used in the two study groups was quite similar.¹¹ Further, secondary DCCT investigations have suggested that factors other than insulin usage probably accounted for the improved glycemic control in the intensive therapy group.¹⁶

3. The authors suggest that careful “patient selection” is necessary for safe, effective, and economical “intensive glycemic treatment.” The concept of “risk stratification” (targeting the proper interventions to individuals who could benefit the most with the lowest risk and the most reasonable cost) has been recognized by the DCCT study group and others¹⁷ as important, both for complications associated with diabetes and for their associated economic impact. The development of a validated and useful risk stratification system for people with diabetes is a major opportunity for additional leadership by the diabetes community.

There are some limitations to this study that should be explicitly recognized. As indicated above, the quality and level of economic analyses were quite basic and perhaps not consistent with the usual requirements for examinations to determine the impact of cost factors.⁹ Second, “intensive approaches” to diabetes management encompass much, much more than insulin therapy.¹⁶ Third, it is surprising that cardiovascular disease was not more important in the authors’ economic analyses, given 35 years of follow-up, and the results of previous economic studies of diabetes.^4,5 Fourth, it is not clear why 6% was chosen as the discounting factor or why it had so little impact on the comparison between standard and intensive results, given that complications would be reduced many years later in life, when the value of the initial investment would be less. The issue of greatest concern, though, is the authors’ black-and-white view that someone is either in “poor control” or “ideal regulation.”¹⁷

Furthermore, the authors imply that these issues of possible risk and benefit of intensive diabetes management are relevant only to people with type I diabetes. In fact, most economic analyses are, and perhaps should be, performed from a societal perspective.¹² Thus, the impact of modest improvements in glycemic control in the much larger population with type II diabetes should be carefully considered before judgments are made about the economic justification of any diabetes preventive intervention.¹⁷

Finally, economic considerations should never be the only, or perhaps even the major, consideration for health decisions. After all, even if the study by Stern and Levy suggests that the bottom line after 35 years is similar between standard and intensive interventions, the improved quality of life for those people with type I diabetes not having to deal with blindness, amputations, and kidney failure has to count for something.

References

¹Swartz K, Brennan T: Integrated health care, capitated payment, and quality: the role of regulation. Ann Intern Med 124:442-48, 1996.

²Kassirer J: The new health care game. New Engl J Med 335:433, 1996.

³National Diabetes Data Group: Diabetes in America. Harris M, Ed. Washington, D.C., National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, NIH Publication No. 95-1468, 1995.

⁴Ray N, Wills S, Thamer M and the Medical Technology and Practice Patterns Institute: Direct and Indirect Costs of Diabetes in the United States in 1992. Alexandria Va., American Diabetes Association, 1993.

⁵Rubin R, Altman W, Mendelson D: Health care expenditures for people with diabetes mellitus. J Clin Endocrinol Metab 78:809A-F, 1994.

⁶Levatan C, Ratner R: The economic bottom line on preventive diabetes care. Practical Diabetology 14:10-19, 1995.

⁷Vinicor F: Is diabetes a public health disorder? Diabetes Care 17 (Suppl 1):22-27, 1994.

⁸Kassirer J: Managed care and the morality of the marketplace. New Engl J Med 333:50-52, 1995.

⁹Detsky A, Denton A: A clinician’s guide to cost-effectiveness analysis. Ann Intern Med 113:147-54, 1990.

¹⁰Reinhardt U: Economics: market-based health care rationing. J Am Med Assoc 275:1802-04, 1996.

¹¹The DCCT Research Group: The effect of intensive diabetes treatment on the development and progression of long-term complications in insulin dependent diabetes mellitus. New Engl J Med 329:977-986, 1993.

¹²Eisenberg J: Clinical economics: a guide to the economic analysis of clinical practice. J Am Med Assoc 262:2879-86, 1989.

¹³Laupacis A, Feeny D, Detsky A, Tugwell A: How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations. Can Med Assoc J 146:473-581, 1992.

¹⁴Russell L: The role of prevention in health reform. N Engl J Med 329:352-54, 1993.

¹⁵Herman W, Dasbach E, Songer T, Thompson D, Crofford OL: Assessing the impact of intensive insulin therapy on the health care system. Diabetes Reviews 2:384-88, 1994.

¹⁶Walker E: “. . . not just ‘more insulin’. . .” Diabetes Spectrum 6:220, 1993.

¹⁷Vinicor F: Barriers to the translation of the Diabetes Control and Complications Trial. Diabetes Reviews 2:371-83, 1994.

Frank Vinicor, MD, MPH, is director of the Division of Diabetes Translation at the Centers for Disease Control and Prevention in Atlanta, Ga.