VOL. 18 NO. 2 Spring 2000


It's Time to Change the Paradigm for Delivering Diabetes Care

Reviewed by Mayer B. Davidson, MD

Levetan CS, Passaro MD, Jablonski KA, Ratner RE: Effect of physician specialty on outcomes in diabetic ketoacidosis. Diabetes Care 22:1790-95, 1999.

Objective. The primary objective of this study was to investigate hospital charges and length of stay (LOS) for patients with diabetic ketoacidosis (DKA) according to the specialty of their primary care provider.

Design and Methods. All patients discharged with the principal ICD-9 code diagnosis of DKA between 1 January 1993, and 31 July 1996 were evaluated. Chart review by diabetes educators was supervised by an investigator who did not admit patients to this large, urban teaching hospital during the study period and was unfamiliar with the attending physicians or their areas of specialty. Specialty determination was obtained from hospital credentials. Case mix was established by two different methodologies. In addition to evaluations of the primary objective, demographic factors, severity of illness, laboratory data, and readmission rates were also compared.

Results. A total of 260 patients were identified, of whom 257 had medical records available for evaluation. Fifteen endocrinologists were the primary care providers for 33 admissions (13%), and 92 physicians who did not have specialty training in diabetes or endocrinology were the primary care providers for 224 admissions (87%). Patients cared for by generalists and endocrinologists had a similar case mix and severity of DKA.The age-adjusted mean LOS for patients of generalists was 4.9 days (95% CI 4.5–5.4) and the mean LOS for patients of endocrinologists was 3.3 days (2.6–4.2) Mean hospital charges differed (P < 0.001), with an age- and sex-adjusted mean for patients of generalists of $10,109 ($9,151–$11,166) and a mean for patients of endocrinologists of $5,463 ($4,179–$7,141). The additional charges incurred by generalists were due in part to patients undergoing more procedures.

No differences in diabetes-related complications occurred during admission, but the endocrinologist-treated group had a lower readmission rate for DKA after 30 days following discharge than the generalist-treated group (2% vs. 6%) (P = 0.03).

Conclusions. Endocrinologists provide more cost-effective care than do generalists when serving as primary care providers for patients hospitalized with DKA.

In this well done study, Levetan and colleagues have shown that hospitalized patients with DKA cared for by endocrinologists had significantly shorter LOS (3.3 vs. 4.9 days), fewer hospital procedures (24% vs. 49% of patients), lower hospital charges ($5,463 vs. $10,109) and fewer readmissions for DKA after 30 days following discharge (2% vs. 6%) than those cared for by generalists.

The authors went to great lengths to avoid bias by having the charts abstracted by diabetes educators who were not aware of the purpose of the study. An investigator who did not admit patients to that hospital supervised their chart review and was unfamiliar with the physicians who took care of the DKA patients reviewed in the study. A case-mix analysis was done by two separate methods to evaluate for differences in comorbidities in the patients cared for by the endocrinologists and the generalists. No differences in the comorbidities of the two groups of patients were found. Furthermore, there were no differences in the ages, years of clinical practice, or percentage of physicians with primary board certification between the endocrinologists and generalists. Thus, these impressive differences in the outcome of the patients cared for by these two groups of physicians are real and important.

The results of this article confirm similar studies. These same authors1 compared the LOS of diabetic patients discharged with a principal diagnosis of DKA, hyperosmolar state, or uncontrolled diabetes who were managed by an internist alone, consulted on by an endocrinology fellow under the supervision of an attending endocrinologist, or consulted on by a multidisciplinary team consisting of the endocrinology fellow, the attending endocrinologist, a diabetes educator, and a registered dietitian. The LOS for each of the three groups of patients was 8.2 days, 5.5 days, and 3.6 days, respectively. Interestingly, the LOS for patients seen by the endocrinology fellow and attending 48 h or more after admission were not different from the LOS for patients managed by an internist only. It was 4.5 days if patients were seen within 48 h of admission. All consultations by the multidisciplinary diabetes team occurred within 48 h of admission.

Koproski et al.2 carried out a randomized trial comparing the effects of a diabetes team (a diabetes nurse educator and an endocrinologist) on the LOS and readmission rates of diabetic patients with a primary diagnosis of uncontrolled diabetes. The median LOS was 2 days shorter for patients visited by the diabetes team (5.5 vs. 7.5 days). Fifteen percent of the patients seen by the diabetes team were readmitted within 3 months of discharge compared to 32% of those who were not exposed to the diabetes team. This difference persisted at 6 months (although the data were not shown). These studies on cost and LOS confirm earlier studies.3-5

Apparently, there is another inpatient area in which attention to diabetes is sorely lacking. Levetan et al.6 found that the diagnosis of diabetes was not considered in hospitalized patients previously undiagnosed whose glucose concentrations exceeded 200 mg/dl while in the hospital. The values ranged from 202 to 503 mg/dl, with a mean of 299 mg/dl. Two-thirds of the patients had two or more values exceeding 200 mg/dl. There was no physician documentation that diabetes or even hyperglycemia was present in two-thirds of the medical records (even though slightly over half of the patients received sliding scale regular insulin coverage). Diabetes was mentioned as a possible diagnosis in approximately 5% of the charts, and the percentage of patients receiving definitive therapy for this diagnosis was slightly less. Given the long delay (7–10 years) before the diagnosis of diabetes is made in people with asymptomatic hyperglycemia7 and the fact that 20% of these individuals will have retinopathy when finally diagnosed,8 these physicians certainly missed a golden opportunity to initiate therapy to delay the devastating microvascular complications of this disease.9-11

The paradigm for inpatient diabetes care has the near-term potential to change. The recognition of different skill sets and experience for delivering inpatient and outpatient care has led to the creation of a new specialist, the "hospitalist" or "intensivist."12 These physicians provide inpatient care only, and emerging data suggest that hospital outcomes are better.13 Although these specialists are not specifically trained in endocrinology, their continued exposure to the acute metabolic problems of diabetes should enable them not only to treat these problems efficiently, but also to diagnose patients correctly (if necessary) and to initiate appropriate therapy in concert with patients' primary care physicians.

The vast majority of diabetes care, of course, occurs in an office or outpatient setting. The American Diabetes Association has provided evidence-based guidelines (Table 1), which, if met, would greatly reduce the devastating morbidity and mortality caused by diabetes. Sadly, diabetes care in both fee-for-service and health maintenance organization (HMO) settings does not even come close to meeting these guidelines,14 and improvement is slow in coming.

Table 1. American Diabetes Association Guidelines for Good Diabetes Care
Frequency Goal Action Required
1. Glycosylated Hb
Every 6 months if <action-required level; every 3 months if greater <7%
<1% above;         upper limit of normal
>2% above;       upper limit of  normal
2. LDL cholesterol
   CAD absent
   CAD present
Yearly or more often as necessary <130 mg/dl
<100 mg/dl
>160 mg/dl
>130 mg/dl
3.  TG Yearly or more often as necessary <200 mg/dl >400mg/dl

4. Renal profile: yearly or more often as necessary
  a) Dipstick for proteinuria
    (1) if positive and confirmed, ACE inhibitor unless contraindicated; serum   creatinine every 6 months;
    (2) if dipstick negative, evaluation for microalbuminuria;
       (a) if positive and confirmed, ACE inhibitor unless contraindicated.

5. Blood pressure: minimum every 6 months (or more often as necessary) as long as target level of <130/85 mmHg met.

6. Visits: minimum every 6 months as long as all action required levels not exceeded and all target levels met; otherwise a contact at least every 3 months.

7. Eye exam: yearly dilated funduscopic exam in all diabetic patients except type 1 diabetic patients within 5 years of diagnosis.

8. Foot exams: minimum every 6 months, or more often as necessary.

9. Weight: minimum every 6 months.  

10. Smoking assessment: yearly; if current smoker, counseling or referral for cessation.

11. Low-dose aspirin in patients at high risk for CAD.

ACE, angiotensin-converting enzyme; CAD, coronary artery disease; Hb, hemoglobin; LDL, low-density lipoprotein; TG, triglyceride

Diabetes care is easier to evaluate in HMOs because of their organizational structure. It was systematically evaluated for 1993 in the second largest HMO in California in 353 diabetic patients scattered throughout the state and was found deficient.15 For instance, there were no measurements of glycosylated hemoglobin (HbA1c) levels in 56%, fasting plasma glucose concentrations in 65%, total cholesterol concentrations in 44%, fasting triglyceride concentrations in 50%, high-density lipoprotein and low-density lipoprotein cholesterol levels in 69%, and neither a serum creatinine concentration nor a dipstick urinary protein in 50%. Nearly 40% of patients in whom HbA1c levels were measured had values exceeding 10%. Only 22% were referred to an ophthalmologist for a dilated eye exam, and only 6% had a documented foot exam in the chart.

A recent survey of diabetes care for 1997 in 4,747 diabetes patients in 11 California HMOs showed similar results for HbA1c and eye exam outcomes, the only two outcomes that were evaluated (personal communication).

There are several reasons for the persistent poor level of diabetes care. Most patients are asymptomatic and do not bring any complaints to their physicians. Therefore, the multiple facets of diabetes care mostly consist of measures to prevent its complications, i.e., effectively treating hyperglycemia, hypertension, and hyperlipidemia; evaluating and treating microalbuminuria; examining the feet; and ensuring that routine dilated eye exams are carried out. Since patients have few complaints, physicians must take the initiative for the appropriate evaluations, treatment, and follow-up of these preventive measures. This takes time—a commodity in very short supply in today's medical care structure.

Often, a primary care physician has only 10–15 min to devote to a patient's visit, and diabetes is only one of the problems that must be dealt with. This was borne out by a study at Duke University,16 in which primary care physicians had on a computer screen in front of them during visits which guidelines were met and which ones had not been met. In spite of the availability of this critical information, compliance with the guidelines (which had been approved by the primary care physicians before the study was started) was achieved in only 32% of the patients. When queried about the difficulty in meeting the guidelines, the physicians noted the complexity of diabetes care and, most importantly, the lack of time.

Since it is extremely unlikely that this limitation of the delivery of good diabetes care will change, at least in the near future, it is time to change our paradigm. Diabetes care delivered by specially trained nurses17-20 or pharmacists21 following specific protocols and supervised by a physician is significantly better than that in appropriate control groups receiving their diabetes care in the usual structure. Although there are start-up costs involved, these can be mostly balanced in the short term by savings from avoidable hospitalizations.22 Large savings will start to accrue after approximately 5 years because of lessened complications.23

Diabetes is too devastating and costly24 to leave its care to our present approach, which has not been able to deal effectively with this common affliction. We need to "think outside of the box" and change the paradigms of inpatient1-5,13,14 and outpatient18-22 diabetes care to approaches with proven effectiveness. As unsettling as these changes may be to many, proceeding along our present path is not likely to change the situation any time soon, if ever.


1Levetan CS, Salas JR, Wilets IF, Zumoff B: Impact of endocrine and diabetes team consultation on hospital length of stay for patients with diabetes. Am J Med 99:22-28, 1995.

2Koproski J, Pretto A, Poretsky L: Effects of an intervention by a diabetes team in hospitalized patients with diabetes. Diabetes Care 20:1553-56, 1997.

3May ME, Young C, King J: Resource utilization in treatment of diabetic ketoacidosis in adults. Am J Med Sci 306:287-94, 1993.

4Edelstein EL, Cesta TG: Nursing case management: an innovative model of care for hospitalized patients with diabetes. Diabetes Educ 19:517-21, 1993.

5Feddersen E, Lockwood DH: An inpatient diabetes educator's impact on length of hospital stay. Diabetes Educ 20:125-28, 1994.

6Levetan CS, Passaro M, Jablonski K, Kass M, Ratner RE: Unrecognized diabetes among hospitalized patients. Diabetes Care 21:246-49, 1998.

7Harris MI, Klein R, Welborn TA, Knuiman MW: Onset of NIDDM occurs at least 4 to 7 years before clinical diagnosis. Diabetes Care 15:815-19, 1992.

8Harris MI: Undiagnosed NIDDM: clinical and public health issues. Diabetes Care 16:642-52, 1993.

9Ohkubo Y, Kishikawa H, Araki E, Takao M, Isami S, Motoyoshi S, Kojima Y, Furuyoshi N, Shichiri M: Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res Clin Pract 28:103-17, 1995.

10Davidson MB: Importance of control in type 2 diabetes mellitus. Endocrine Pract 3:145-52, 1997.

11U.K. Prospective Diabetes Study (UKPDS) Group: Intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352:837-53, 1998.

12Wachter RM, Goldman L: The emerging role of hospitalists in the American health care system. N Engl J Med 335:514-17, 1996.

13Diamond HS, Goldberg E, Janosky JE: The effect of full-time faculty hospitalists on the efficiency of care at the community teaching hospital. Ann Intern Med 129:197-203, 1998.

14Davidson MB: Diabetes care in health maintenance organization and fee-for-service settings. Dis Manage Health Outcomes 4:189-97, 1997.

15Peters AL, Legorreta AP, Ossorio RC, Davidson MB: Quality of outpatient care provided to diabetic patients. Diabetes Care 19:601-606, 1995.

16Lobach DF, Hammond WE: Computerized decision support based on a clinical practice guideline improves compliance with care standards. Am J Med 102:89-98, 1997.

17Peters AL, Davidson MD: Application of a diabetes managed care program: the feasibility of using nurses and a computer system to provide effective care. Diabetes Care 21:1037-43, 1998.

18Legorreta AP, Peters AL, Ossorio RC, Lopez RJ, Jatulis D, Davidson MB: Effect of a comprehensive nurse-managed diabetes program: an HMO prospective study. Am J Man Care 2:1024-30, 1996.

19Aubert RE, Herman WH, Waters J, Moore W, Sutton D, Peterson BL, Bailey CM, Koplan JP: Nurse case management to improve glycemic control in diabetic patients in a health maintenance organization: a randomized controlled trial. Ann Intern Med 129:605-12, 1998.

20Sadur GN, Moline N, Costa M, Michalik D, Mendolowitz D, Roller S, Watson R, Swain BE, Selby JV, Javoski WC: Diabetes management in a health maintenance organization: efficacy of care management using cluster visits. Diabetes Care 22:2011-17, 1999.

21Davidson MB, Karlan VJ, Hair TL: Effect of a pharmacist-managed diabetes care program in a free medical clinic. Am J Med Qual July/Aug 2000. (In press)

22Peters AL, Davidson MB, Ossorio RC: Management of patients with diabetes by nurses with support of subspecialists. HMO Pract 9:8-13, 1995.

23Herman WH, Eastman RC: The effects of treatment on the direct costs of diabetes. Diabetes Care 21:C19-24, 1998.

24American Diabetes Association: Economic consequences of diabetes mellitus in the U.S. in 1997. Diabetes Care 21:296-309, 1998.


The secretarial skills of Willie Nelson are gratefully acknowledged. Dr. Davidson is supported by National Institutes of Health grant #5U01 DK54047.

Mayer B. Davidson, MD, is director of the Clinical Trials Unit at Charles R. Drew University and a professor of medicine at the UCLA School of Medicine in Los Angeles.

Copyright 2000American Diabetes Association
Updated 4/00
For ADA Related Issues contact

For Technical Issues contact