The Effects of Participating in Diabetes Disease Management Programs for Individuals with Type I and Type II Diabetes on Clinical and Economic Outcomes

The Effects of Participating in Diabetes Disease Management Programs (DDMPs) for Individuals with Type I and Type II Diabetes on Clinical and Economic Outcomes as Compared to Participants that Receive Standard and Usual Care: A Review of the Literature

Joe Molinaro

Spring 2006

University of Medicine and Dentistry of New Jersey

Evidence-Based Literature Review; IDST 6400

Dr. Craig Scanlan

According to the American Diabetes Association, 20.8 million Americans have diabetes and in 2005 alone, 1.5 million new cases were diagnosed. Not only is this disease considered the fifth leading cause of death in the United States, it is related to retinopathy, neuropathy, nephropathy, and cardiovascular diseases (Wolfangel, Brixner, and Jacobs, 2004). Consequently, $132 billion dollars was devoted to the care of people with diabetes in 2002 (Wolfangel et al., 2004); in other words, $1 out of every $10 healthcare dollars is spent treating diabetes (American Diabetes Association, 2002). Given that diabetes affects 7% of the population in the United States while inviting onset of other co-morbidities (American Diabetes Association, 2002; Wolfangel et al., 2004), diabetes and its medical consequences presents a pressing concern for healthcare today.

Sixty-three percent of people with diagnosed diabetes do not meet the 7% HbA1c standard set forth by the American Diabetes Association (Saydah, Fradkin, and Cowie, 2004). According to the Diabetic Control and Complications Trial Research Group (DCCT) (1993) and the UK Prospective Diabetes Study Group (UKPDS) (1998), glucose control that improves HbA1c will delay the onset or slow the progression of long-term diabetic complications. In addition, studies consistently demonstrate that glycemic control and cost have an inverse relationship. Simply stated, the poorer the glycemic control, the more expensive the care of diabetes and its comorbidities (Herman, and Eastmer, 1998). As a result, diabetes is the third most costly health condition in the United States (Jacobs, 2004).

Diabetes Disease Management Programs (DDMPs) have been purported to improve the quality of care for patients with diabetes while reducing costs. DDMPs are defined as the “coordination of all health care services across a continuum of care” (Sidorov et al., 2000, p. 1218). The benefits of DDMPs promise improving glycemic control while lowering health care costs (Sidorov et al., 2000).

The purpose of this review is to evaluate the effects of participating in DDMPs for individuals with Type I and Type II diabetes on clinical and economic outcomes as compared to the clinical and economic outcomes of individuals with Type I and Type II diabetes who receive standard and usual care.

Methods

A review of the literature was conducted to identify studies of the effectiveness of DDMPs on improving short and long-term clinical outcomes or reducing costs for the patient, the insurer, or the nation. An OVID MEDLINE database search was conducted for 1996 to February 2006. Also, an Evidenced-Based Medicine-only review was conducted using Cochrane DSR, ACP Journal Club, DARE, and CCTR to prevent reduplication. The following medical subject headings were used to search the database: Diabetes Mellitus, Type I, or Diabetes complications, or Diabetes Mellitus, Type II were mapped to their disease state MESH terms. The following program terms were mapped to their MESH terms critical pathways or “delivery of healthcare” or disease management or patient care team or physician’s practice patterns or point-of-care systems or preventive medicine. Boolean “AND” was used to combine the disease state terms with the program terms. A final search for cost was conducted using the following mapped MESH terms: costs and cost analysis or cost allocation or cost-benefit analysis or “cost control” or “cost of illness” or “cost sharing” or health care costs or direct service costs or drug costs or employer health costs or hospital costs or health expenditures or economics, pharmaceuticals or “fees and charges.” Boolean “AND” was used to combine the disease state and program terms with the cost terms. This search resulted in 50 articles.

The articles were then manually screened for relevance and attainability. Hand searches of bibliographies from relevant articles and reviews were also performed, which identified three additional studies. Of the 53 articles retrieved, 16 articles were relevant to this review. The articles were either kept or eliminated due to inclusion and exclusion criteria. The inclusion criteria included the following components: Type I and Type II diabetics, participation in a DDMP compared to non-participation, clinical and/or economic outcomes, randomized, prospective, or retrospective studies, and written in English language. The exclusion criteria included the following factors: non-diabetic samples, a lack of intervention comparison, non-reporting of clinical and/or economic outcomes, any study that does not meet the inclusion criteria, and non-English full text studies were excluded even if an English translation of the abstract was provided. Twelve studies met the inclusion criteria.

Literature Review

As reflected in Table 1, the 12 reviewed studies included 71,521 Type I and Type II diabetic patients. This review evaluated clinical outcomes in terms of improved blood glucose (HbA1c), improved cardio-vascular risks, reduced retinopathy, reduced nephropathy, and reduced neuropathy. Cost will be measured as overall cost per individual over a certain amount of time, the overall cost for the group in the study, the overall reduction in costs due to hospital stays, and the overall national cost savings.

Clinical Outcome

HbA1c.

As reflected in Table 1, the effect of DDMPs on patients’ blood glucose levels were measured in each of the 12 studies. Standard care was provided to patients in the comparison groups. Some studies included only Type II diabetics, while other studies included both Type I and Type II diabetics. Additionally, some studies included co-morbidities while others excluded co-morbidities. As expected, individuals who participated in DDMPs in 11 out of 12 studies showed a statistically significant improvement in HbA1c levels as compared to individuals who participated in non-DDMPs. Out of the four DDMP, randomized, and controlled trials, three of four studies showed a significant reduction in HbA1c levels compared to its respective standard-of-care group. One of the randomized, controlled trials, the Litaker et al. (2003) study, had a small but statistically significant improvement in HbA1c levels. On the other hand, the Williams et al. (2005) study, which included the largest sample size of the randomized, controlled trials, showed neither an improvement nor a decline in HbA1c in the DDMP group as compared to the standard-of-care group. The data suggested that the patients who participated in the DDMP group attended physician appointments more regularly than the standard-of-care patients. However, one of the threats to validity in the Williams et al. (2005) study was the standard-of-care group may have been indirectly exposed to the DDMP. Although inconclusive and in need of further study, Williams et al. (2005) postulated that the DDMP alone might eventually improve glycemic control. Accordingly, in a one-year cohort study of 5,332 patients, 91.3% of the participants reduced their HbA1c to an average of 7.41% (SD=1.76) from 8.76% (Sidorov et al., 2000). A reduction in HbA1c was also measured in the Steffens (2000) study which evaluated 10,000 diabetic HMO members after a one-year time frame. Although the sample sizes of both studies were significantly larger than the randomized, controlled studies, the Sidorov et al. (2000) and the Steffens (2000) studies both lacked rigorous control and were vulnerable to confounding variables. However, the patients in the DDMP group reduced their mean HbA1c from an average of 8.2% to 7.7%. “Because this reduction occurred across the entire population – not just in newly diagnosed diabetic patients in whom new treatment regimens were being initiated – that result is impressive” (Steffens, 2000, p. S702).

One of the threats to validity that may have reduced the DDMP’s impact on HbA1c is that most of the studies evaluated the HbA1c impact over only a one-year time period. In addition, the DDMP group was only exposed to the DDMP treatment for a one-year time period. For instance, the Litaker et al. (2003) study, which extended data collection over an additional one-year period after the DDMP was completed, found that HbA1c reduced back to its pre-treatment baseline. Consequently, the initial improvement in HbA1c that was demonstrated over the one- year study period was then reduced back to the baseline for the DDMP group. These results highlight the importance of extending the study period so that the long-lasting effects of DDMPs on HbA1c can be evaluated. Then, studies can further investigate the impact on HbA1c when participants are exposed to DDMPs over an extended time period.

Diabetes-Related Health Complications.

As shown in Table 1, nine studies investigated the effects of the DDMP on co-morbid diseases, increased testing rates, and the reduction or delay of the development of long-term diabetes complications, such as retinopathy, nephropathy, and neuropathy. Overall, four studies showed that the DDMP groups prevented or delayed diabetes-related health complications as compared to the control group. According to the results of the randomized and controlled Trento et al. (2000) study, retinopathy progressed less rapidly in the DDMP group as compared to the standard-of-care group. The findings of the Gozzoli, Palmer, Brandt, and Spinas (2001) study showed that in the sample of 1,256 diabetic patients, the DDMP group significantly reduced myocardial infarctions, stroke, amputation, kidney transplantation, and hypoglycemia as compared to the standard-of-care group. This evidence suggests that exposure and participation in DDMPs correlate with statistically significant reductions in long-term diabetes complications. However, data collection was limited to a one-year study period. Therefore, it is possible that long-term complications presented after the one-year period. Clearly, research methodologies that integrate a longitudinal design may produce more valid results in the study of long-term diabetes-related health complications.

Table 1: Overview of studies addressing the clinical outcome effects of DDMPs.

Author, Year	Participants	Interventions	Study Design	Outcomes
Litaker et al.(2003)	157 patients with established mild or moderate hypertension; non-insulin dependent Type II diabetes; and have no presentation of organ complications	Nurse/Doctor team approach which focused on disease management programs, patient education, and regular glucose monitoring.	Randomized Controlled	Increases in HDL and a small but significant improvement in decreases in HbA1c. There were no differences in BP or total Cholesterol between the two groups.
Middleton (2003)	30 patients with type II diabetes. Many had significant co- morbidities	Patients who were in a DDMP with a case manager	Randomized Controlled	Statistical significant improvement in the mean HbA1c .The mean risk for developing complications was significantly reduced.
Trento et al. (2002).	112 Type II diabetics, age < 80, greater than 1 year attendance in clinic	Interactive group follow-up visits for Type II Diabetic Care	Randomized Controlled	Reduction in HbA1c. Increase in HDL Increase QoL. Retinopathy progressed less with group.
Williams et al. (2005)	232 patients with type II diabetes. HbA1c reading 1 point or more above accepted range. Responsible for managing their own diabetes. Have greater than 1-year life expectancy.	Activation Intervention	Randomized Controlled	No significant differences in HbA1c.
Gozzoli et al. (2001)	1256 diabetic patients	Participation in a DDMP which included Educational programs, Nephropathy screening and retinopathy screening.	Systemic Review	Improved HbA1c. Reduced MI, Stroke, Amputation, Kidney transplantation and hypoglycemia.
Espinet et al. (2005)	7993 diabetic patients	Participation in a DDMP	Prospective Cohort	Those who participated had a statistically significant higher rate for HbA1c testing and control rate, LDL-C screening and level rate and eye exam rate.
Licciardone et al. (1997)	2,359 patients with Type I Diabetes (188) and Type II Diabetes (2171). The mean age was 42.3 years for the Type I diabetics and 54.8 years for the Type II diabetics. The majority reported no co-morbid conditions.	Participation in a DDMP	Prospective Cohort	Reduction in HbA1c Reduced work/school days
Sidorov et al. (2000)	5332 patients with both type 1 and type 2 diabetes	Diabetes disease management program (DDMP)	Prospective Cohort	At 1 year follow-up 91.3% of the participants reduced their HbA1c to 7.41%.
Snyder et al. (2003)	422 diabetic patients	Participation in a DDMP	Prospective Cohort	HbA1c values dropped from 8.89% to7.88%. Increase of HbA1c, retinal and foot exams.
Steffens (2000)	10000 diabetic patients who are members of the John Deere Health Care	Participation in an integrated DDMP	Prospective Cohort	HbA1c and microalbuminuria improved by approximately 20%. There was a greater long-term potential to reduce diabetes related health complications.
Tang et al. (2005)	136 diabetic patients who had already participated at least 5 times before in a health management plan. Many had DM complications	Nurse directed DDMP	Prospective Cohort	90% maintained or improved their HbA1c after participating in a DDMP. This study suggests that this DDMP was effective in preventing or delaying DM complications.
Villagra & Ahmed (2004)	43,492 with diabetes who were fully insured by a HMO or POS plan and their plan covered the diabetes disease management program.	Diabetes disease management program (DDMP)	Nonrandom quasi-experimental methods	Statistical significant improvement for retinal exams, mocroalbumin testing, lipid testing and tobacco use. HbA1c and prescription ACE and ARBs were trending positively.

The findings of Espinet, Osmick, Ahmed, and Villagra (2005), Snyder, Malaskovitz, Griego, and Flatt (2003), and Villagra & Ahmed (2004) suggested that the DDMP group had a significant increase in diabetes-related testing, screening, and exams. The findings further purported that the more a patient practices preventive testing, screening, and exams, the more these patients may delay the development or progression of diabetes-related health complications. However, the Espinet et al. (2005), Snyder et al. (2003), and Villagra & Ahmed (2004) results were subject to selection biases that may have threatened the validity of their findings. Volunteer biases were introduced because the patients volunteered to participate in the studies. Therefore, compliance alone, instead of or in addition to the DDMP, may have accounted for the significant increase in participation of diabetes-related testing, screening, and exams.

Additionally, Trento et al. (2002) and Litaker et al. (2003) evaluated cardiovascular risks and found an increase in HDL cholesterol for the DDMP groups as compared to the standard-of-care groups. However, the Williams et al. (2005) study found no significant differences between the DDMP and the standard-of-care groups in blood pressure and total cholesterol.

Economic Outcomes

According to Table 2, only eight of the 12 studies addressed cost. Direct cost for the DDMP, cost savings per patient, inpatient costs, and overall national healthcare costs were evaluated.

Direct cost for the DDMP.

Three randomized, controlled studies investigated the direct cost of developing and implementing a DDMP. All three studies showed that the initial costs for the one-year trial were higher for the DDMP groups as compared to the standard-of-care groups. The Trento et al. (2002), Litaker et al. (2003), and Middleton (2003) studies noted that although initial cost increases were reported, overall long-term cost reduction is expected. However, data supporting this assertion were neither researched nor reported.

Cost Savings per patient:

Three studies showed that through a DDMP, patients and insurers reduced health care spending as compared to the standard-of-care group. According to Snyder et al. (2003), in a study of 422 diabetic patients, participants in the DDMP group had a cost savings of $268.63 per patient per month. The Steffens (2000) study, which consisted of 10,000 patients, found a $56 per patient, per month decrease in cost with the DDMP group as compared to the standard-of-care group.

Table 2: Overview of studies addressing the economic outcome effects of DDMPs.

Author, Year	Participants	Interventions	Study Design	Outcomes
Litaker et al. (2003)	157 patients with established mild or moderate hypertension; non-insulin dependent Type II diabetes; and have no presentation of organ complications	Nurse/Doctor team approach which focused on disease management programs, patient education, and regular glucose monitoring.	Randomized Controlled	Even though initial costs were higher for 1 year in DDMP group, the author states that eventually the improved glycemic control would impact the cost of care through prevented complications, hospitalizations, or more labor-intensive interventions.
Middleton (2003)	30 patients with type II diabetes. Many had significant co- morbidities	Patients who were in a DDMP with a case manager	Randomized Controlled	This data did not prove that through a DDMP which improves glycemic control overall healthcare costs are reduced. However, it does suggest that further investigation on cost effectiveness is worthy.
Trento et al. (2002)	112 Type II diabetics, age < 80, greater than 1 year attendance in clinic	Interactive group follow-up visits for Type II Diabetic Care	Randomized Controlled	Cost for DDMP was $91 dollars more per patient; however, study suggests that there will be an overall long-term reduction in cost. Each QoL point gained cost $2.00.
Gozzoli et al. (2001)	1256 diabetic patients	Participation in a DDMP which included Educational programs, Nephropathy screening and retinopathy screening.	Systemic Review	The average annual for Type II diabetics in Switzerland is CHF 1.82 billion without intervention and CHF 1.62 billion with DDMP. When preventing diabetes complications, the initial costs become savings after about 3-4 years. This would average about CHF 194 million per year.
Licciardone et al. (1997)	2,359 patients with Type I Diabetes (188) and Type II Diabetes (2171). The mean age was 42.3 years for the Type I diabetics and 54.8 years for the Type II diabetics. The majority in both groups reported no co-morbid conditions.	Participation in a DDMP	Prospective Cohort	This study suggests that improved glycemic control may decrease indirect costs related to diabetes.
Snyder et al. (2003)	422 diabetic patients	Participation in a DDMP	Prospective Cohort	36% drop in inpatient costs. Total medical costs were reduced by 26.8%. This decrease amounts to $268.63 per patient per month.
Steffens (2000)	10000 diabetic patients who are members of the John Deere Health Care	Participation in an integrated DDMP	Prospective Cohort	An overall reduction in cost occurred. This was from a 22% reduction in hospital admission rates, a 34% reduction in hospital length of stay and a 12% reduction per member per month. This decrease amounts to $56 per patient per month,
Villagra & Ahmed (2004)	43,492 with diabetes who were fully insured by a HMO or POS plan and their plan covered the diabetes disease management program.	Diabetes disease management program (DDMP)	Nonrandom quasi-experimental methods	Average cost was reduced by 8.1% in the DDMP group. Costs rose in pharmacy and other direct costs associated with the intervention. Overall cost was reduced by $26 per patient.

Inpatient Costs.

Two studies showed that patients in a DDMP had reduced inpatient hospital costs. Snyder et al. (2003) and Steffens (2000) reported reductions in inpatient costs for the DDMP group as compared to standard-of-care group by 36% and 34% respectively.

National Yearly Cost Savings.

Although no data reported a national healthcare cost reduction in the United States, Gozzoli et al. (2001) found that the average annual cost for Type II diabetic care in Switzerland was 1.82 billion Swiss Francs ($1.4 billion) without a DDMP, and 1.62 billion Swiss Francs ($1.3 billion) with a DDMP. According to Gozzoli et al. (2001), this national cost reduction trend surfaced after the three- to four-year time period, averaging 194 million Swiss Francs ($152 million) per year (Gozzoli et al., 2001).

Discussion

One of the crucial concerns that most of these studies neglected to tease out includes that Type I diabetic patients may differ from Type II diabetic patients. Although these disease-states share the same name, the interventions, physiological implications, and disease-related consequences can differ. For instance, studies have consistently demonstrated that Type I diabetic patients experience more hospitalizations than Type II diabetic patients. Consequently, grouping together Type I and Type II diabetic patients in the samples, as did most of the studies reviewed, may muddy results relating to clinical and economic outcomes of DDMPs. Although Trento et al. (2002), Litaker et al. (2003), Williams et al. (2005), and Middleton (2003) included only Type II diabetic participants, none of the studies included only Type I diabetic patients. It is also possible that the type of DDMP may need to vary among Type I and Type II patients to maximize both clinical and economic benefits.

Another significant unresolved area includes long-term economic implications of the DDMPs. Most of the studies had only a one-year duration. For economic outcomes related to onset of long-term complications to be evaluated, a longer study period is needed. The longer study period is also relevant when evaluating clinical outcomes in terms of HbA1c for the DDMP group. Also, continuing research related to compliance for the DDMP over five-, ten-, or longer time periods may yield differing clinical and economic outcome results.

The studies reviewed were vulnerable to contamination biases. For instance, the Trento et al. (2002) study selected participants from one clinic and the Steffens (2000) study selected participants from only the John Deere health care plan. The subjects in each standard-of-care group may have been unintentionally exposed to the DDMP. Consequently, differences between the clinical and economic outcomes of the groups may be minimized or skewed. In addition, this selection method introduced co-intervention biases. The research designs did not control the extent to which the patients may receive other diabetic education and information during the same time period that the DDMP was introduced. In the future, selecting participants from various geographic locations that participate in DDMP in various managed care plans may reduce the impact of biases related to selecting participants from one clinic or one health care plan.

Future research directions also need to standardize the DDMP studied. In this review, some of the DDMPs included group interventions, nurse-practitioner interventions, and intervention education. Standardizing the DDMP studied would allow for more valid results. In addition, in this review, some of the control group interventions included video education, continuing the existing treatment of the patient, and evaluating the pre-DDMP baseline. Each of these methods for the control group invited extensive variability. Standardizing the standard of care for the control group would allow for more valid results.

In summary, the findings of this literature review suggest that as compared to standard-of-care, participation in a DDMP tends to reduce HbA1c over a one-year period. In addition, the findings of this literature review suggest that over a one-year time period, retinopathy, myocardial infarctions, stroke, amputation, kidney transplantation, and hypoglycemic were reduced and HDL cholesterol was increased in the DDMP groups as compared to the standard-of-care groups.

This literature review found conflicting results in terms of the initial direct costs for implementing DDMPs. Trento et al. (2002), Litaker et al. (2003), and Middleton (2003) found initial cost increases while Snyder et al. (2003) and Steffens (2000) found initial and monthly cost reductions for the DDMP groups. However, because neither the DDMP nor the standard-of-care treatments were standardized across the studies reviewed, the variation in direct cost is expected because the variations in the DDMP and standard-of-care interventions were not controlled.

The overall findings of this literature review suggest that DDMPs hold the promise of improving both clinical and economic outcomes for patients with diabetes. Future research that separates Type I diabetic participants from Type II diabetic participants, evaluates the effects of the DDMP over a longer time period, implements the DDMP over a longer time period, and standardizes the DDMP invention and the standard-of-care intervention might lead to clinical and economic benefits for the approximately 20.8 million people in our country that face the daily demands of living with diabetes.

References

American Diabetes Association. (2006). Standards of medical care in diabetes - 2006. Diabetes Care, 29(Suppl1), S4-S42.

Blonde, L. (2000). Disease management approaches to type 2 diabetes. Managed Care, 9(8), 18-23.

Diabetes Control Complications Research Group (DCCT), (1993).The Effect of Intensive Treatment of Diabetes on the Development and Progression of Long-Term Complications in Insulin Dependent Diabetes Mellitus. New England Journal Of Medicine. 329, 977-986.

Espinet, L., Osmick M., Ahmed T., and Villagra, V. (2005). A cohort study of the impact of a national disease management program on HEDIS diabetes outcomes. Disease Management, 8(2), 86-92.

Gozzoli, V., Palmer A., Brandt A., and Spinas, G. A. (2001). Economic and clinical impact of alternative disease management strategies for secondary prevention in type 2 diabetes in the Swiss setting. Swiss Medical Weekly, 131, 303-310.

Herman, W., Eastmer, D. (1998) The effects of treatment on the direct costs of diabetes. Diabetes Care, 21(S3), C19-24.

Jacobs, M (2004).The Employer's Perspectives on Type 2 Diabetes. Supplement to Managed Care. 13, 5, 17-21.

Licciardone, J. C., Kotsanos J. G., Brinkman-Kaplan V., Cooper, T., Jordan, J., and Wishner, K. (1997). Resource utilization and work or school loss reported. The American Journal of Managed Care, 3(5), 777-782.

Litaker, D., Mion L. C., Planavsky L., Kippes, C., Mehta, N., and Frolkis, J. (2003). Physician - nurse practitioner teams in chronic disease management: the impact on costs, clinical effectiveness, and patients' perception of care. Journal of Interprofessional Care, 17(3), 223-237.

Middleton, J. (2003). The effect of case management on glycemic control in patients with type 2 diabetes. The Case Manager, 14(6), 43-47.

Saydah, S., Fradkin, J., and Cowie, C., (2004).Poor Control of Risks Factors for Vascular Disease Among Adults with Previously Diagnosed Diabetes. JAMA. 291, 335-342.

Sidorov, J., Gabbay R., Harris R., Shull, R., Girolami, S., Tomcavage, J., Starkey, R., and Hughes, R. (2000). Disease management for diabetes mellitus: impact on hemoglobin A1c. The American Journal of Managed Care, 6(11), 1217-1226.

Snyder, J., Malaskovitz J., Griego J., and Flatt, K. (2003). Quality improvement and cost reduction realized by a purchaser through diabetes disease management. Disease Management, 6(4), 233-41.

Steffens, B. (2000). Cost-effective management of type 2 Diabetes: providing quality care in a cost constrained environment. The American Journal of Managed Care, 6(13), S697-S703.

Tang, P., Yuan, W., & Tseng H. (2005). Clinical follow-up study on diabetes patients participating in a health management plan. Journal of Nursing Research, 13(4), 253-262.

Trento, M., Passera P., Bajardi M., Tomalino, M., Grassi, G., Borgo, E., Donnola, C., Cavallo, F., Bondonio, P. and Porta, M. (2002). Lifestyle intervention by group care prevents deterioration of Type II diabetes: 4-year randomized controlled clinical trial. Diabetologia, 45, 1231-1239.

UK Prospective Diabetes Study Group (UKPDS), (1998).Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). The Lancet. 352, 9131, 837-853.

Villagra, V., & Ahmed, T. (2004). Effectiveness of a disease management program for patients with diabetes. Health Affairs, 23(4), 255-266.

Williams, G., McGregor H., Zeldman A., Freedman, Z., Deci, E., and Elder, D. (2005). Promoting glycemic control through diabetes self-management: evaluating a patient activation intervention. Patient Education and Counseling, 56, 28-34.

Wolfangel, P., Brixner, D., & Jacobs M. (2004). The rationale for early, aggressive treatment of type 2 diabetes: the cost ramifications of improved health outcomes. Supplement to Managed Care, 13(5), 1-24.

Zinman, B. (1997). Translating the diabetes control and complications trial (DCCT) into clinical practice: overcoming the barriers. Diabetologia, 40, S88-S90.

Appendix

Original Research Search History:

Database: Ovid MEDLINE(R) <1996 to February Week 3 2006>

Search Strategy:

-----------------------------------------------------------------------------

1 Diabetes Mellitus, Type 1/ or Diabetes Complications/ or Diabetes Mellitus, Type 2/ (47574)

2 critical pathways/ or "delivery of health care"/ or disease management/ or patient care team/ or physician's practice patterns/ or point-of-care systems/ (53694)

3 Preventive Medicine/ (2033)

4 2 or 3 (55563)

5 "costs and cost analysis"/ or "cost allocation"/ or cost-benefit analysis/ or "cost control"/ or "cost of illness"/ or "cost sharing"/ or health care costs/ or direct service costs/ or drug costs/ or employer health costs/ or hospital costs/ or health expenditures/ or economics, pharmaceutical/ or "fees and charges"/ (59408)

6 1 and 4 (629)

7 5 and 6 (50)

8 from 7 keep 2,4-8,10-27,30-38,40-42,44-49 (42)