outcome

These pages are best viewed with Netscape version 3.0 or higher or Internet Explorer version 3.0 or higher. When viewed with other browsers, some characters or attributes may not be rendered correctly.

If the goal of third-party payor, health maintenance, and managed care organizations is to provide cost-effective care to help patients with diabetes achieve desired glucose outcomes, there is evidence that dietitians and nurses can provide this care effectively. Dietitian-nurse teams, trained to provide basic and continuing diabetes care, with access not only to primary care providers but also to diabetes specialists, might be most effective. Having a team of dietitian and nurse, cross-trained, would give medical settings the advantage of having multidisciplinary skills and the ability to provide cost-effective care for patients with diabetes.

Outcomes and Cost-Effectiveness of Medical Nutrition Therapy for Non-Insulin-Dependent Diabetes Mellitus

Marion J. Franz, MD, RD, CDE,
Arlene Monk, RD, CDE,
Richard Bergenstal, MD, and
Roger Mazze, PhD

Efforts to define and document quality, systems, and outcomes of medical nutrition therapy (MNT) are important issues for dietitians and health-care organizations. However, quality care does not stand alone. Quality nutrition care and associated outcomes must also be evaluated against the costs necessary to deliver and achieve the desired outcomes. Cost savings and cost-effectiveness of MNT as part of acute and chronic disease treatment are issues that are also being addressed.

Medical professionals, including dietetic practitioners, must be skilled not only in providing high quality care that leads to desired health outcomes, but also in evaluating the costs associated with that care. Health-care expenditures and changes in the delivery of health care are two major forces driving and requiring that costs be studied. The objective is not always to reduce costs, but instead to ensure that financial and professional resources are used to improve overall health and the delivery of health care in a cost-effective manner. Providing no care at all would initially involve the least amount of monetary expenditure, but may not in the long term be the most cost-effective.¹

Evaluating and Quality and Costs of Medical Care
The process of evaluating care provided for patients began with quality assurance (QA). QA requires review criteria and is retrospective. To be more prospective, the focus shifted to identifying clinical indicators linked to total quality management or continuous quality improvement (CQI). CQI establishes protocols or standards for the optimal care process, measures performance, and requires further actions based on the findings. This leads naturally to the redefining of protocols or practice guidelines for improved outcomes. Practice guidelines differ from QA and CQI in that they are prospective, involve descriptors on how to provide care and are outcome-orientated, whereas QA and CQI are process- and review-oriented.²Because practice guidelines incorporate desired or expected outcomes, data related to these outcomes can be collected and analyzed to quantify the relationship between interventions, outcomes, and costs (see Figure 1).

While practice guidelines cannot achieve the goal of cost containment alone, they can make major contributions to it. Resolution of the cost issues also requires connecting values to costs.³ Various types of economic evaluations address these issues.

Types of Economic Evaluations
Economic evaluations differ in regard to how they measure consequences. Cost-of-illness and cost-benefit analyses measure consequences in monetary terms, whereas cost-effectiveness and cost-utility analyses measure consequences in terms of concrete expressions of desired health outcomes. Cost-effectiveness analysis, therefore, incorporates both costs and effect.^4-6

Cost-of-illness. Cost-of-illness computes lifetime costs and dollar consequences of an illness in current dollar values. The economic burden resulting from a chronic illness such as diabetes is of major importance in the allocation of health-care resources and in the evaluation of health research and treatment programs.

The purpose of cost-of-illness is to determine the economic impact of an illness by computing current costs and dollar consequences of an illness. Direct costs (medical care costs) plus indirect costs (dollar value of lost productivity) equal the total cost of an illness. For example, the total economic costs of diabetes in 1992 was reported to be $91.8 billion. Total direct costs, including all institutional costs and outpatient costs, were estimated at $45.2 billion. Total indirect costs, including lost productivity due to short-term and long-term morbidity as well as mortality, were estimated at $46.6 billion.⁷

Rubin and associates⁸ reported that in 1992, although people with diabetes constituted 4.5% of the U.S. population, they accounted for 14.6% of the total U.S. health-care expenditures ($105 billion). Health-care expenditures for people with diabetes constituted about one in seven health-care dollars spend in 1992.

In the 1992 report, 0.03% of all expenditures for diabetes were attributed to diet/nutrition counseling and 1.0% to diagnostic testing.⁷ A total of 1.0 million visits to a dietitian or nutritionist were reported due to diabetes. Estimates were based on the 1989 National Health Interview Survey (NHIS)—Diabetes Public Use File, which consists of data regarding dietary practices, health-care utilization, medications, and other health-care topics asked of each identified and self-confirmed adult with diabetes. Dietitian/nutritionist visit outpatient costs due to diabetes were estimated to be $31.9 million, or approximately $32 per visit. As noted, this is a very small percentage of total costs.

The report on costs summarizes the issue by making an important point. That is, an increase in preventive and diagnostic services as well as early aggressive intervention (including nutrition therapy) may well result in a significant decrease in total costs of diabetes and its sequelae.⁷

Cost-benefit. Cost-benefit analysis incorporates calculation in dollar terms of net benefit (cost savings) or net cost by subtracting cost from benefit. It can be used to determine whether the cost is worth the benefit by measuring both in the same units, and assesses whether the investment in a program is warranted.

Direct benefits of health care are expressed in dollars saved through fewer hospitalizations, physician visits, medical procedures, medications, or laboratory tests and in improved clinical outcomes that result in reduced health-care costs. Indirect benefits are those that may indirectly reduce the cost of disease to society by reducing morbidity and mortality rates, improving quality of life, and increasing productivity. Intangible benefits, such as the reduction of pain, suffering, and grief, are important benefits to those who experience them even though dollar values are generally not assigned.

Cost-effectiveness. Cost-effectiveness analysis incorporates both costs and effect. It compares two or more alternative means of achieving a desired outcome in terms of cost per outcome. Cost-effectiveness measures the cost of providing a service and the outcomes obtained and considers the possibility of obtaining improved outcomes in exchange for the use of more or different resources. Costs are expressed in dollar terms and outcomes in natural units (lives saved, years of lives saved, cases prevented, improved metabolic parameters).

The cost-effectiveness ratio compares the dollar costs from a program to outcomes in natural units with the dollar costs with alternative programs or without a program to outcomes in natural units. In other words, the cost effectiveness ratio = program A costs ÷ outcomes from program A compared to program B costs ÷ outcomes from program B.

There are several examples of cost-effectiveness studies related to diabetes. Scheffler and associates9 collected hospital charges and length-of-stay data on 102 women enrolled in the California Diabetes and Pregnancy Program (a program using intensive diabetes management preconception and early in pregnancy to improve pregnancy outcomes) with 218 control cases. Hos-pital charges were about 30% less for program participants, and days in the hospital were roughly 25% less. After adjusting for inflation and differences in charges across hospitals, $5.19 was saved for every dollar spent on the program.

Cost-utility. Cost-utility analysis is similar to cost-effectiveness analysis, but measures the psychosocial consequences of programs. It uses the same ratio as cost-effectiveness but compares two or more alternatives in terms of quality-adjusted outcomes. The denominator is quality-adjusted to reflect quality of life. Outcomes are expressed in natural units weighted by their quality.

Kaplan and associates¹⁰ reported on the effects of diet and exercise interventions on metabolic control and quality-of-life in non-insulin-dependent diabetes. Compared to a control group, at 18 months, the combination diet-and-exercise groups had achieved the greatest reduction in glycated hemoglobin and showed significant improvements on a general quality-of-life measure. The expected effect on health of the combination program was calculated to be 0.092 years of well-being for each participant. That is, the average patient in the combined treatment group received 0.092 units of benefits in comparison with the control group. Costs were estimated to be $1,000 per patient. The cost/utility ratio was $1,000/0.092 = $10,870/well year.

Medical Nutrition Therapy Costs
In order to conduct an economic evaluation, costs must first be determined. With increased concern about the escalating costs of health care, it is more important than ever that costs of providing nutrition services be documented. Costs need to be documented for budgeting purposes, cost control, and internal administrators; for negotiating third-party, managed care, and health maintenance organization reimbursement for services; for communicating with funders; and for governing boards and politicians.¹ Comparisons can be made between actual costs and predicted costs. Data can be used to communicate costs of predetermined outcomes for nutrition care and services provided.

An understanding of unit costs is essential as a basis for establishing fees for service and for negotiating reimbursement fees. While the fee is not necessarily equal to the actual cost, knowledge about actual costs is an important part of the fee-setting decision.

Determining and Reporting Costs
It is important to be clear about the purpose of cost finding. Knowing why the data is needed and how it will be used assures that the right information is collected and documented appropriately to fulfill the purpose of cost analysis.

Cost data is derived by answering the following questions: who does what, with or for whom, for how long, and at what costs?¹ Determining costs necessitates accounting for all costs, direct and indirect, that are necessary to produce or deliver a distinct nutrition service. Cost finding results in a per-patient unit cost. The cost per patient = direct costs + indirect costs ÷ the number of patients served.

Included in direct and indirect costs are other costs for which there may be no charge to the nutrition care but which must be calculated. If this is not done, the final result will underestimate the actual costs. Costs that are easy to overlook include the planning and documentation of interventions; staff planning and other committee meetings; continuing education; support staff time for scheduling and confirming visits, accessing charts, and billing; laboratory charges; room and building overhead; and supplies.¹

Direct costs are those charges in medical resource use that are required for or are attributable to the intervention being studied. Although not usually calculated, it may also include medical and nonmedical costs such as the cost borne by the patient in seeking care. Direct costs involved in providing nutrition care/service include personnel, salaries and benefits for both professional and support staff; clinic and office supplies; and education and counseling materials.

Indirect costs are resources that support nutrition care/service including overhead, space, maintenance, depreciation, central supplies, data processing, billing, bookkeeping, and laboratory fees. Indirect costs can also be the result of morbidity or mortality and cost-savings related to the avoidance of poor health.

One of the difficulties in conducting cost-effectiveness studies for nutrition therapy is determining legitimate costs and cost-savings to ascribe to nutrition interventions. It is clear the dietitian and support staff times, materials, and overhead are costs. It is debatable if the cost of laboratory tests (i.e., glycated hemoglobin and fasting plasma glucose values and lipid profiles) and use and changes in medications are costs that should be attributed to the nutrition intervention. An array of cost information allows users of data from cost documentation to use the costs appropriate for their needs.

Costs of care for the clinical management of patients with insulin-dependent diabetes mellitus (IDDM) in the Diabetes Control and Complications Trial (DCCT) have been reported.11 The annual cost of intensive therapy ($4,000 and $5,800/year for multiple daily injections and continuous subcutaneous insulin infusion, respectively) was approximately three times the cost of conventional therapy ($1,700/year). Although intensive therapy is more expensive than conventional therapy, it offers the hope of cost savings as a result of averted complications. Furthermore, the relative costs of intensive and conventional therapy may be different in settings where controlling costs is a priority.

Costs for nutrition therapy in the DCCT were also collected. For intensive therapy, dietitians averaged 3.5 hours/year with patients compared to approximately 1.25 hours for conventional therapy. Costs attributed to the dietitian would be approximately 2% of the total cost for intensive therapy/year with multiple daily injections and 1.5% of the total cost for intensive therapy/year with continuous subcutaneous insulin, or approximately 1.7% of the total cost for conventional therapy/year. Again, this is a small percentage of total costs for diabetes care.

Nutrition Practice Guidelines and Costs for NIDDM
The American Dietetic Association contracted with the International Diabetes Center to develop nutrition practice guidelines for non-insulin dependent diabetes mellitus (NIDDM) and to conduct a clinical trial and cost-effectiveness study of care based on nutrition practice guidelines compared to basic nutrition care.^12-13

These nutrition practice guidelines apply to people with newly diagnosed NIDDM or previously diagnosed NIDDM at the first visit to a dietitian for initial or ongoing nutrition therapy. The guidelines can be implemented for patients treated with nutrition therapy alone, nutrition therapy and oral glucose-lowering agents, or nutrition therapy and insulin. They provide a framework to assist the dietitian in the assessment, intervention (nutrition prescription, education, goal setting) and evaluation of outcomes for nutrition therapy.

The practice guidelines define minimal referral data necessary for clinical decision making and outcome criteria. Basic care is defined as one visit with the dietitian. Care that follows the nutrition practice guidelines consists of a series of visits with a dietitian. At the second follow-up visit, the dietitian uses blood glucose and hemoglobin A1c values to assess what has been accomplished with the nutrition interventions. If a patient has implemented the nutrition recommendations to the best of his or her ability and has not achieved the treatment goals, the dietitian should notify the physician and recommend that changes in medical management are needed (i.e., addition of medications or dose change). Ongoing nutrition care is recommended at 6-month to 1-year intervals for both basic and nutrition guideline care.¹⁴

A prospective, randomized, controlled clinical trial measuring the effect of the two levels of nutrition care on metabolic control and cost-effectiveness was conducted at diabetes centers in three states (Minnesota, Colorado, and Florida). Basic care (BC) consisted of a single visit with a dietitian; practice guideline care (PGC) involved an initial visit with a dietitian followed by two visits during the first 6 weeks of the study period. Data were collected at entry to the study and at 3 and 6 months. Results have been reported for 179 men and women aged 38–76 years; 85 assigned randomly to BC and 94 to PGC. This represents 72% of the 247 subjects enrolled.

A nonrandom comparison group (n = 62) of similar patients seen by primary care physicians with no contact with a dietitian, and who had hemoglobin A_1c laboratory tests during the same enrollment period as the study participants and again 6 ± 1 month later was identified. The intensification of the nutrition intervention was the primary variable between groups, with all other aspects of care as provided by physicians or nurses being similar.¹⁵

Glycemic control (hemoglobin A1c [HbA_1c] and fasting plasma glucose levels [FPG]), lipids (total cholesterol, LDL- and HDL-cholesterol, and triglycerides) were measured at entry and at 3- and 6-month intervals for BC and PGC groups. Weight (body mass index) and waist/hip ratios were measured at entry and at 6 months. Goal achievement and lifestyle changes (eating habits, food intake, and exercise), knowledge, and quality of life were also assessed.

The primary goal of each session was achievement of target blood glucose goals. Lipids and weight were secondary issues. A variety of strategies tailored to individual patients were used to meet these goals, including improved food choices, especially a decrease in calories and fat; spreading food throughout the day; moderate weight loss; and an increase in physical activities.

At 6 months, PGC resulted in significant improvement in FPG and HbA_1c, and BC resulted in significant improvement in HbA1c. PGC participants had a mean FPG level at 6 months that was 10.5% lower than at entry, and those in BC group had a 5.3% lower values. Mean HbA_1c levels in the PGC group decreased from 8.3% at entry to 7.4% at 6 months, and in the BC group from 8.3% to 7.6%. Among participants who had diabetes for longer than 6 months, those who received PGC had a significantly better HbA_1c level at 3 months compared to those receiving BC. In contrast, the comparison group showed no improvement in HbA_1c over a comparable 6 months.16 See Figure 2.

Cost data were collected and the cost was determined to be $112 ($95 without adding cost of an HbA_1c test) per patient for PGC care compared to $42 per patient for BC. Costs included dietitians’ salaries and benefits for time spent with patients, preparation, documentation, and committee times; support staff salaries and benefits; supplies and educational materials used; overhead; and, for the PGC group, the cost of the 6-week HbA_1c assay which the BC group did not have.

In the PGC group, 17 people had changes in medical therapy (recommended by the dietitian), which yielded an average 12-month cost savings prorated for all patients of $31.49. In contrast, in the BC group, 9 people had changes in medical therapy, for an average 12-month prorated cost savings of $3.13. An array of outcomes and costs was published.14 See Table 1.

Table 1. An array of nutrition visits, costs, outcomes, and cost-savings for two levels of medical nutrition therapy for non-insulin-dependent diabetes mellitus
Level of nutrition care^a	No. of visits	Mean contact time	Total costs^b	Per-patient costs^b	Mean chanage in fasting plasma glucose level^c	Mean change in HbA_1c	No. of changes in therapy	Mean cost saving due to changes in therapy
BG Group (n=85)	1	1h	$3,565.55	$41.95	-0.41 ± 2.74 mmol/L	-0.69 ± 1.67	9	$3.13
PGC group (n = 94)	3	2 1/2h	$10,534.33	$112.07	-1.07 ± 2.77 mmol/L	-0.93 ± 1.63	17	$31.49
^aBC = basic nutrition care; PGC = practice guidelines nutrition care. ^bIncremental costs for medical nutrition therapy as a component of diabetes care, expressed in 1993 dollars. ^cTo convert mmol/L glucose to mg/dL, multiply mmol/L by 18.0. To converg mg/dL glucose to mmol/L, multiply mg/dl by 0.0555. Gl;ucose of 6.0 mmol/L = 108 mg/dL. Reprinted with permission¹⁶

A difficult issue in conducting cost-effectiveness studies relates to the length of the outcome benefit from one nutrition intervention. We used cost-savings from medication changes recommended by the dietitian as cost savings and assumed the medication changes would be sustained for another 6 months. One could make the case that medication changes would last longer than 6 months.

Cost-Effectiveness of Nutrition Care Provided by Dietitians
Each unit of change in FPG level from entry to the 6-month follow-up was achieved with an investment of $5.75 by implementing BC or of $5.84 by implementing PGC (including cost for HbA1c assay). If net costs are considered (per-patient costs, cost savings due to therapy changes), the cost-effectiveness ratios were $5.32 for BC and $4.20 for PGC, assuming the medical changes in therapy are maintained for 12 months.

In relation to total costs of diabetes, nutrition interventions can be provided by experienced dietitians at very reasonable costs and result in significant improvements in metabolic control. When dietitians are involved in active decision-making about interventions (nutrition prescriptions, number of visits needed, medication changes), cost-effectiveness is enhanced.

A Cost-Effective Proposed Model for the Management of NIDDM
In our study,15 no improvement in glycated hemoglobin values were observed in the comparison group treated only by primary care physicians. Information from a national survey of primary care physicians indicates that glycemic control recommended for patients with diabetes is likely to be discordant with the glycemia values conducive to prevention of diabetes complications.¹⁷ Weiner and associates¹⁸ reviewed claims data for Medicare patients with diabetes treated in primary care practices and concluded that elderly patients with diabetes do not receive optimal care. Other evidence suggests this is true for patients with IDDM, as well.¹⁹

On the other hand, we were able to document an average decrease in HbA1c values of approximately 1% in patients treated with intensive nutrition interventions. Furthermore, it was apparent by 3 months whether nutrition therapy changes alone would lead to desired glucose outcomes. Peters and associates²⁰ reported similar improvements in glucose control in a program for managing patients with diabetes within a health maintenance organization using diabetes nurse specialists.

Pringle and associates²¹ assessed patient, doctor, practice, and process of care variables for their effect on glycemic control in diabetes mellitus. Glycemic control was related to the organization and process of care. In a multiple regression analysis using all variables shown to influence HbA1 values, only two variables were related to delivery of care. Access to a dietitian reduced the random HbA1 value by a mean of 1.06%, and the general practitioner having a special interest in diabetes reduced it by 0.86%.

One could speculate that if third-party payor, managed care, and health maintenance organizations wish to be cost-effective and to assist patients to achieve desired treatment outcomes, they might consider training teams of dietitians and nurses (case managers) to provide basic and continuing diabetes care following sets of predetermined practice guidelines such as Staged Diabetes Management²² or other predetermined medical practice guidelines.²³It is important these teams have access not only to primary care physicians, but also to diabetes specialists for advice or referral for patients unable to achieve glycemic control using practice guidelines and for the assessment and management of the long-term complications of diabetes. It should also be the responsibility of diabetes specialists to develop and update practice guidelines for diabetes. Team access to licensed psychologists or behavioral scientists is also important.

Although lack of dietitians and nurse educators is often used as a reason for not implementing team care, this is usually not the case. In a study to identify the characteristics of diabetes care delivered by primary care physicians in Minnesota and Wisconsin, Peterson²⁴ reported that 93% referred patients to a dietitian for nutrition education, whereas only 44% had access for referral to a certified diabetes educator. Even though nurses and dietitians specializing in diabetes may not be available, nurses and dietitians are available. They may not have the training necessary to provide diabetes care and, therefore, a system to train teams is essential.

Traditionally, dietitians and nurses have thought of their role as educators and not as clinicians. Yet based on the experiences of the DCCT, that role can change into one of also being clinicians providing cost-effective medical care.²⁵

Other barriers cited as reasons for not implementing comprehensive treatment involve financial reimbursement for care by practitioners who provide intensive management, such as dietitians and diabetes educators, and for self-care techniques, such as blood glucose monitoring. They also include access by physicians to allied health-care professionals and specialists in diabetes complications.¹⁹

Managing diabetes is a labor-intensive process, and physicians—both specialists and general practitioners—rarely have the time necessary to help patients with the lifestyle changes necessary to achieve desired glycemic outcomes. Although the studies have looked at outcomes from dietitians and nurses working independently,^15,20 combining as teams to take advantage of each professional’s skills could potentially lead to even better patient diabetes management outcomes.

Instead of having a staff of two or more nurses or two or more dietitians, having teams of dietitians and nurses, cross-trained, would allow medical settings the advantage of having experts in different areas that could provide better and more cost-effective care for patients with diabetes. Therefore, if third party payor, managed care, and health maintenance organizations are interested in improving glycemic control in patients with diabetes in a cost-effective manner, new approaches need to be considered and reimbursement for the necessary services needs to be provided.

References
¹Splett P, Caldwell M: Costing Nutrition Services. A Workbook. Washington, DC, MCH Clearinghouse, 1985.

²Franz MJ: Practice guidelines: actions to deliver desired outcomes. Diabetes Educ 19:185-89, 1993.

³Franz, MJ: Practice guidelines: road maps for nutrition care. J Am Diet Assoc 92:1136, 1992.

⁴Elixhauser A: The cost-effectiveness of preventive care for diabetes mellitus. Diabetes Spectrum 2:349-53, 1989.

⁵Eisenberg JM: Clinical economics: a guide to the economic analysis of clinical practices. JAMA 262:2879-86, 1989.

⁶Detsky AS, Naglie IG: Clinician’s guide to cost-effectiveness analysis. Ann Int Med 113:147-54, 1990.

⁷American Diabetes Association: Direct and Indirect Costs of Diabetes in the United States in 1992. Alexandria, VA, American Diabetes Association, 1993.

⁸Rubin RJ, Altman WM, Mendelson DN: Health care expenditures for people with diabetes mellitus, 1992. J Clin Endocrinol Metab 78:809A-809F, 1994.

⁹Scheffler RM, Feuchtbaum LB, Phibbs CS: Prevention: the cost-effectiveness of the California Diabetes and Pregnancy Program. Am J Public Health 82:168-75, 1992.

¹⁰Kaplan RM, Hartwell SL, Wilson DK, Wallace JP: Effects of diet and exercise interventions on control and quality of life in non-insulin-dependent diabetes mellitus. J Gen Intern Med 2:220-27, 1987.

¹¹The DCCT Research Group: Resource utilization and costs of care in the Diabetes Control and Complications Trial. Diabetes Care 18:1468-78, 1995.

¹²Franz MJ: Practice guidelines for nutrition care by dietetic practitioners for outpatients with non-insulin-dependent diabetes mellitus: consensus statement. J Am Diet Assoc 92:1136-39, 1992.

¹³Mazze RS, Franz MJ, Monk A, Cooper N, Barry B, Weaver T, McClain K, Upham P, Haugen D, Bergenstal R: Methodologies for field-testing and cost-effectiveness analysis. J Am Diet Assoc 92:1139-42, 1992.

¹⁴Monk A, Barry B, McClain K, Weaver T, Cooper M, Franz MJ: Practice guidelines for medical nutrition therapy provided by dietitians for people with non-insulin-dependent diabetes mellitus. J Am Diet Assoc 95:999-1006,1995.

¹⁵Franz MJ, Monk A, Barry B, McLain K, Weaver T, Cooper N, Upham P, Bergenstal R, Mazze RS: Effectiveness of medical nutrition therapy provided by dietitians in the management of non-insulin-dependent diabetes mellitus: a randomized, controlled clinical trial. J Am Diet Assoc 95:1009-17, 1995.

¹⁶Franz MJ, Splett PL, Monk A, Barry B, McClain K, Weaver T, Upham P, Bergenstal R, Mazze RS: Cost-effectiveness of medical nutrition therapy provided by dietitians for people with non-insulin dependent diabetes mellitus. J Am Diet Assoc 95:1018-24, 1995.

¹⁷Siebert C, Lipsett LF, Greenblatt J, Silverman RE: Survey of physician practice behaviors related to diabetes mellitus in the U.S. 1. Design and methods. Diabetes Care 42: 759-64, 1993.

¹⁸Weiner JP, Parente ST, Garnick DW, Fowles J, Lawthers AG, Palmer H: Variation in office-based quality. A claims-based profile of care provided to Medicare patients with diabetes. JAMA 273:1503-508, 1995.

¹⁹Harris ML, Eastman RC, Siebert C: The DCCT and medical care for diabetes in the U.S. Diabetes Care 17:761-64, 1994.

²⁰Peters AL, Davidson MB, Ossorio RC: Management of patients with diabetes by nurses with support of subspecialists. HMO Practice 9:8-13, 1995.

²¹Pringle M, Stewart-Evans C, Coupland C, Williams I, Allison S, Sterland J: Influences on control in diabetes mellitus: patient, doctor, practice, or delivery of care? Brit Med J 306:630-34, 1993.

²²Mazze RS, Etzwiler DD: Implications of the DCCT for national health care reform. Diabetes Reviews 2:256-62, 1994.

²³Institute for Clinical Systems Integration: Health Care Guidelines. Diabetes Mellitus. Institute for Clinical Systems Integration, Minneapolis, MN, 1996.

²⁴Peterson KA: Diabetes care by primary care physicians in Minnesota and Wisconsin. J Fam Pract 38:361-67, 1994.

²⁵Franz MJ, Callahan T, Castle G. Changing roles: educators and clinicians. Clinical Diabetes 12:53-54, 1994.

²⁶Monk A: Linking practice guidelines to outcomes. DCE On the Cutting Edge 16(4):10-12, 1995.

Marion J. Franz, MS, RD, CDE, is the director of nutrition and publications; Arlene Monk, RD, CDE, is a diabetes nutrition specialist and coordinator of the Community Diabetes Prevention Project; Richard Bergenstal, MD, is the senior vice president, diabetes care, and an adult endocrinologist; and Roger Mazze, PhD, is the senior vice president, research and development, of the International Diabetes Center, Institute for Research and Education, HealthSystems Minnesota, in Min-neapolis. Dr. Mazze is also a clinical professor in family practice at the University of Minnesota Medical School, in Minneapolis.