Disease Management & Health Outcomes

, Volume 11, Issue 8, pp 489–498

Improving the Health Status of US Working Adults with Type 2 Diabetes Mellitus

A Review


    • The William G. McGowan School of BusinessKing’s College
  • Bernard J. Healey
    • The William G. McGowan School of BusinessKing’s College
  • Joseph S. Coyne
    • Washington State University
    • Peking University
Leading Article

DOI: 10.2165/00115677-200311080-00002

Cite this article as:
Akinci, F., Healey, B.J. & Coyne, J.S. Dis-Manage-Health-Outcomes (2003) 11: 489. doi:10.2165/00115677-200311080-00002


Diabetes mellitus is the seventh leading cause of death (sixth leading cause of death by disease) in the US. Approximately 5.9% of the US population has diabetes and one-third of those with diabetes are unaware that they have the condition. Diabetes is the leading cause of adult blindness, end-stage renal disease, and non-traumatic lower extremity amputation.

The annual per-capita incremental cost of diabetes among employees compared with individuals without diabetes has been estimated at $US4410 (1998 values). Furthermore, more than 30% of the costs associated with diabetic employees are attributable to medically related work absences and disability, and this is estimated to cause a one-third reduction in earnings due to reduced workforce participation.

The incidence of diabetes and long term medical complications could be reduced through more effective diabetes education and patient self-management. Intensive management of diabetes can help workers remain productive, decrease costs associated with complications, and reduce associated costs for overtime. Policy complications from this review encourage employers and Medicare/Medicaid to invest in diabetes education and Wellness programs.

Diabetes mellitus is the seventh leading cause of death (sixth leading cause of death by disease) in the US. Approximately 5.9% of the US population has diabetes and one-third of those with diabetes are unaware that they have the condition.[1] According to Boyle et al.,[2] the number of people with diagnosed diabetes in the US is projected to increase by 165% between 2000 and 2050, from 11 to 29 million. The biggest percent increases will be among people aged 75 years and over (336%) and among Blacks (275%). However, the National Institutes of Health (NIH) has also reported a dramatic increase in type 2 diabetes prevalence among youth in the US.[3] A recent study conducted by researchers from the NIH found that many obese children and adolescents have impaired glucose tolerance, a condition that often appears before the development of type 2 diabetes.[4]

Individuals with diabetes are at higher risk of heart disease, stroke, high blood pressure, blindness, kidney disease, nervous system disease, amputations, dental disease, and complications of pregnancy.[5] Long-term complications in patients with poorly controlled diabetes have significant clinical and economic consequences. Several long-term studies, most notably the Diabetes Control and Complications Trial (DCCT)[6] and the United Kingdom Prospective Diabetes Study (UKPDS),[7] have shown that improved blood glucose control can delay the onset and progression of many diabetic complications, thereby resulting in avoidance of costs related to such complications.[8]

More than 90% of patients with diabetes have type 2 disease, which is often caused by lifestyle factors. These factors include poor diet, obesity, and physical inactivity. The Finnish Diabetes Prevention Study (DPS) Group found that good nutrition, weight reduction and increased physical activity reduced the incidence of type 2 diabetes by 58% in high-risk individuals.[9] These findings were supported by those of the Diabetes Prevention Program (DPP), which also found a 58% reduction in progression to diabetes after lifestyle intervention, and a 31% reduction after treatment with metformin, in high-risk individuals.[10]

In order to apply the findings of the DPS and the DPP, patients with diabetes must receive intensive education in the management of their condition. The workplace may be an appropriate setting in which to provide this education, given the high prevalence of diabetes in working adults, and the high costs of this disease to employers.[11] Wing et al.[12] have argued that the increased prevalence of diabetes will result in loss of employee productivity, increased health insurance costs and, perhaps, discrimination for employees with this chronic disease.

This article documents the economic burden of diabetes in the US workplace, including productivity costs associated with work loss and disability. It reviews literature on the impact of diabetes on employers, based on health services research rather than clinical research, and provides policy implications for both government and private payors, as well as employer groups.

Peer-reviewed publications were searched for articles on type 2 diabetes covering the following topics: (i) incidence rates among the US population and subsets by workplace; (ii) costs to US employers for care of employees with diabetes; and (iii) employer-based health management programs related to diabetes among employees. The articles selected for review were of two types: first, articles on diabetes costs and labor productivity impact on employers; and second, diabetes studies pertaining to employer interventions in collaboration with health plans. Assessment of the quality of the information contained in the literature was based on the following criteria: (i) ability to replicate the study findings; (ii) research methods as represented in the study design; and (iii) the significance of results and their implications for health policy.

1. Economic Impact of Diabetes in the Workplace

1.1 Diabetes Disease Costs

The costs associated with treating and managing diabetes are significant for the healthcare sector, businesses, and patients with the disease. Estimates from the literature clearly indicate that the cost of diabetes has increased substantially over the last 25 years. The American Diabetes Association (ADA) estimated total costs for diabetes in the US to be $US2.6 billion in the year 1969[13] and $US132 billion in the year 2002.[14]

Direct costs of a disease usually refer to those costs generated by the resources used in treating or coping with the disease, including expenditures for medical care and treatment of the illness (e.g. hospital care, physician services, nursing home care, prescription drugs, and other medical services). Indirect costs, on the other hand, address the potential productivity lost due to the disease. They include the societal costs of morbidity, disability, and premature mortality.

Direct costs of diabetes in the US were estimated to be $US91.8 billion in the year 2002.[14] These direct costs included the costs of managing diabetes itself, as well as the costs of complications and other medical conditions. According to the ADA, direct costs of diabetes in 2002 represented 19% of total personal healthcare expenditures in the US. Approximately $US40.3 billion was spent on inpatient hospital care and $US13.9 billion on nursing home care services. Patients aged over 65 years incurred 51.8% of total direct costs. Diabetes-related hospitalizations accounted for 16.9 million hospital days, and there were 62.6 million physician office visits made by patients with diabetes in the US in 2002. Cardiovascular disease was the most costly complication of diabetes, accounting for more than $US17.6 billion of the total direct costs of diabetes in 2002. Overall, medical expenditure incurred by patients with diabetes was estimated to be $US13 243 per capita, compared with $US2560 per capita for individuals without diabetes (see table I).
Table I

The economic burden of diabetes mellitus on the US healthcare system in 2002[14]

The indirect costs of diabetes related to absenteeism, disability, lost productivity, and premature mortality were estimated to be $US39.8 billion in the US in the year 2002.[14]

1.2 Cost Impact of Diabetes in the Workplace

The impact of diabetes in the workplace is also significant. In a recent study, Peele et al.[15] used health insurance billing data for approximately 1.3 million individuals enrolled in plans sponsored by 862 large, self-insured employers in the US to examine employer expenditures and consumer out-of-pocket payments for 20 937 diabetic patients. The researchers compared these expenditures with expenditures for patients with other chronic illnesses. Overall, diabetes was estimated to account for 6.5% of total health plan expenditures. In a study published in 1998, D’Allegro[16] reported the annual expense for routine management of diabetes to be $US3000 per person, not including costs for laboratory tests.

In addition to direct expenditures, the economic burden of diabetes on the workplace includes reduced workforce participation and productivity due to premature mortality, disability, diminished work effectiveness, and absences caused by medical service utilization.[17] Most of the estimates of the economic burden of diabetes in the US come from population-based surveys rather than actual employer payments from administrative data. According to the ADA,[14] diabetes accounted for a loss of approximately 88 million work days due to disability in the US in 2002. It was estimated that 176 475 cases of permanent disability were caused by diabetes, at a cost of $US7.5 billion. Controlling for age, men with diabetes were estimated to have 3.1 more lost work days and 7.9 more bed days per year, on average, than men without diabetes. Women with diabetes were estimated to have 0.6 more lost work days and 8.1 more bed days, on average, than women without diabetes.

1.3 Labor Productivity Impact of Diabetes in the Workplace

Existing literature suggests that people with diabetes are more likely to be unemployed and to have problems with their jobs compared with the general population. Individuals with diabetes are shown to face barriers in labor force participation, and therefore tend to have lower annual incomes.[18] According to a recent study by Valdmanis et al.,[18] the US unemployment rate for persons with diabetes was 16% compared with only 3% among persons without diabetes (p = 0.001). The researchers also reported that 71% of individuals with diabetes had an annual income of less than $US20 000 compared with 59% of the matched respondents (p = 0.007). In this study, ‘total disability days’ was used as a measure to assess the resource cost of diabetes as the time an individual spends away from his or her usual activities because of physical or mental disability, assuming the value of time lost. Individuals with diabetes were found to average 5.2 days of total disability compared with 1.3 days for matched respondents (p = 0.005). The researchers used the 1990–1995 Behavioral Risk Factor Survey results from Oklahoma, USA, to compare 425 persons with diabetes with 400 age-, sex-, and race/ethnicity-matched respondents without diabetes. In this study, data collection through a population survey with subsequent case matching enabled researchers to achieve significant results with a relatively small sample size.

In another study, Mayfield et al.[19] examined the rates and demographic determinants of work disability, hours worked per week, work-loss days, and wages among 1502 individuals with diabetes, using 1987 US National Medical Expenditures Survey (NMES) data. The authors defined work disability as “a self-report of having been unable to work because of illness or disability for ≥ 2 quarters in 1987”. Multiple regression analysis was used to estimate the number of work-loss days due to illness, hours worked per week, and wages. After adjusting for other demographic variables, the predicted work disability rate was estimated to be 25.6% for individuals with diabetes, compared with 7.8% for individuals without diabetes. Individuals with diabetes were found to have had significantly more work-loss days (∼ 2 more work-loss days each year) than did people without diabetes (p = 0.035). Among individuals with diabetes, men had more work-loss days than women; 5.9 days versus 5.4 days, respectively. Predicted mean earnings were reported to be significantly lower for individuals with diabetes at all ages, resulting in an estimated $US4.7 million loss in earnings in 1987 due to work disability. While omissions in the NMES data (i.e. exact cause of disability, duration, and type of diabetes) imposed some limitations on the analysis, this study provided the first reliable and comprehensive estimates of earning loss due to work disability for individuals with diabetes by demographic subgroups, based on data from an established population survey.

Similarly, Ng et al.[20] used data from the 1989 US National Health Interview Survey with a diabetes supplement to estimate the cost of productivity losses in the US attributable to diabetes. Multivariate regression analyses were used to estimate the association of selected independent variables (i.e. demographics, health, and diabetes status) with labor force participation, hours of work lost, and the economic value of lost work attributable to diabetes and its complications and duration. Controlling for factors such as age, sex, and health status, the presence of diabetes itself was shown to reduce employment by 3.5%, and the presence of complications reduced employment by 12% compared with the absence of complications. While for those individuals who were employed, having diabetes did not have a significant impact on hours worked, those who had complicated diabetes were estimated to work 3.2 days less every 2 weeks than those whose diabetes was without complications. The researchers concluded that “the net productivity costs of preventing complications once an individual has diabetes can be very significant, amounting to >$US3700–8700 per person per year, depending on the demographic group”. This study accounted for variations in days of work loss and associated lost productivity costs attributable to diabetes by including an important variable, the presence of complications, which has important implications for economic evaluations. However, despite the fact that the results were based on data from a large national survey, the analysis could not distinguish between long- and short-term disability and productivity losses associated with diabetes, and was not able to incorporate productivity losses caused by premature mortality associated with diabetes.

Using the 1994 Behavioral Risk Factor Surveillance System data on 83 566 US individuals aged 18–64 years, Yassin and colleagues[21] estimated the annual cost of disability among people with diabetes. After adjusting for relevant socioeconomic characteristics, they found that people with diabetes were more likely to stop working outside the home. The adjusted odds ratios for men and women were 3.1 (95% CI, 1.2–8.0) and 2.9 (95% CI, 1.0–8.8), respectively. The annual cost of disability among people with diabetes in the US was estimated at $US9.3 billion in 1994.

1.4 Cost and Labor Productivity Impact of Diabetes in the Workplace

In a more recent study, Ramsey et al.[17] examined the economic burden of diabetes from an employer’s perspective, considering direct costs and productivity costs associated with work loss and disability. Productivity costs in this study were based on ‘medically related absences’, including both sporadic work loss associated with use of medical services and extended work loss caused by disability. Using the 1998 claims data from a large, self-insured US Fortune 100 company, the researchers found that the employer’s average annual per capita costs were higher for all employees with diabetes than for employees without diabetes ($US7778 vs $US3367; p < 0.0001), yielding an incremental cost of $US4410 associated with diabetes. The authors concluded that more than 30% of the costs associated with diabetic employees were attributable to medically related work absences and disability. The mean annual number of work loss days for ambulatory visits or hospitalization was estimated to be 8 ± 13 days for diabetic employees versus 5 ± 11 days for employees without diabetes. This research extends the existing literature by using claims data from over 100 000 beneficiaries to evaluate the cost of diabetes from the perspective of a self-insured company. While the sample data used in this research appeared to be relatively representative of the US population in terms of diabetes prevalence, the analysis was subject to the usual limitations of claims data, such as lack of clinical detail and validation of its accuracy. In addition, the study sample excluded patients enrolled in capitated health plans and did not contain information to consider indirect costs of mortality.

1.5 Implications of Employer Impact Literature

What are the implications of these research findings for an average employer? Tim McDonald, a corporate health promotion manager at General Motors Corporation, states, “Corporate America is in a unique position to address some of the health problems that arise from diabetes. With people spending more than one-third of their waking hours on the job, employers can help raise awareness of proactive steps people can take to prevent the onset of diabetes”.[22] Indeed, the workplace offers an excellent opportunity to educate individuals about the health dangers associated with diabetes. There is clearly a need for cost-effective interventions for diabetes to prevent the unwanted effects of this condition. There is generally an agreement among experts that the key component of any company’s diabetes intervention should be the promotion of glycemic control among its employees in order to enhance the quality of life for employees with diabetes, reduce costs from hospitalization and absenteeism, and improve the health and productivity of the workforce.

The main goal of any type of diabetes care is to maintain blood glucose levels at or below normal levels as consistently as possible. An essential part of a diabetes management plan is assessment of glycosylated hemoglobin (HbA1c), which is a measure of average blood glucose levels over the past 2–3 months.[23] Many studies have demonstrated the relationship between sustained hyperglycemia and risk of microvascular complications. HbA1c has been established as both an assessment of glycemic control and a surrogate marker for risk of future complications. The ADA has set reasonable goals for glycemic control based on data from available intervention studies. Unacceptable HbA1c values are those higher than 8%, with the optimal desired value being less than 6%. ADA guidelines, however, recommend that the HbA1c goals for any patient must be ‘individualized in consultation between patient and primary healthcare provider’.[24]

The ADA[25] has proposed that standards of medical care for patients with diabetes be evaluated and continuously improved in order to provide ongoing quality care. These standards offer physicians guidelines that include evaluation of the comorbidities associated with diabetes and suggestions for diabetes self-management techniques for physicians and their patients.[25] An example of physician interaction with a patient with diabetes would include evaluation of the possibility of aspirin therapy for the patient. Prior to 1997, many physicians believed that aspirin therapy was contraindicated in patient with diabetes. However, in a recent study Rolka et al.[26] argued that the daily use of aspirin may prevent cardiovascular disease in patients with diabetes. Their data strongly suggested that most patients with diabetes have at least one risk factor for cardiovascular disease. Several other studies have also shown that aspirin is effective and well tolerated in patients with diabetes.[26]

The TRIAD Study Group[27] reported that while the complications of diabetes can also be reduced through better management of hypertension and lipid levels, and periodic screening for eye, kidney and foot disease, these strategies have not been implemented adequately in patient treatment protocols. Adhering to such strategies should increase quality of life for patients with diabetes as they grow older.

2. The Costs and Benefits of Employer- and Health Plan-Sponsored Diabetes Interventions

Klonoff and Schwartz[28] suggested that “Cost-savings or cost-effective interventions can prevent the economic impact of long-term complications, such as blindness, end-stage renal disease (ESRD), and lower extremity amputations (LEA), as well as short-term complications, such as hospitalization for poor glycemic control”. Is preventive care in diabetes a prudent allocation of society’s resources? Recently, a number of economic evaluation studies examined the economic efficiency of diabetes treatments and prevention programs. Economic evaluation techniques, such as cost-effectiveness, cost-utility, and cost-benefit analysis, typically examine the costs and benefits of a program, project, or intervention to inform decisions about allocation of scarce resources.

2.1 A Controlled Study of Diabetes Interventions

The impact of interventions aimed at improving glycemic control for patients with diabetes has been extensively studied in the literature, with four studies specifically examining impact on the workforce. Overall, any improvement in glycemic control has been shown to reduce the risk for diabetes-related complications.

Testa and Simonson[29] conducted a 12-week, randomized, double-blind, placebo-controlled study of 569 individuals with type 2 diabetes at 62 sites in the US. They showed that, compared with diabetic employees who did not control and lower their blood glucose levels, those employees who did improve their blood glucose control were more productive on the job (99% vs 87%) and were able to remain employed longer (97% vs 85%) than diabetic employees who did not control and lower their blood glucose levels. While employees with improved blood glucose control lowered their absenteeism rate by 1%, there was an 8% increase in absenteeism among diabetic employees with poor blood glucose control. The researchers estimated the lost earnings due to absenteeism at only $US24 per employee per month for male employees who improved their blood glucose control in comparison with $US115 for those with uncontrolled blood glucose. Finally, employees with improved blood glucose control reported fewer days of restricted activity and bed rest than those with poor blood glucose control. Lost earnings due to restricted activity were $US2660 per 1000 person-days for male employees with good blood glucose control versus $US4275 for those with uncontrolled blood glucose. For those restricted to bed rest, lost earnings were $US1539 per 1000 person-days for male employees compared with $US1843 for those with poor blood glucose control. While the long-term health benefits of good glycemic control in patients with diabetes are well documented, this research suggested that improved glycemic control of type 2 diabetes is associated with substantial short-term symptomatic, quality-of-life, and health economic benefits.

2.2 Diabetes Workplace Education Effectiveness

In another study, Burton and Connerty[30] evaluated a worksite-based patient education program for employees with diabetes. The goal of the program was to determine if such a program could result in improved diabetes control. A total of 53 employees participated in baseline laboratory testing and met monthly with a diabetes health educator. To encourage employee enrollment and continued program participation, various incentives were offered, such as a free glucose monitor, and a lunch box for each session with nutritional content consistent with ADA dietary recommendations. Laboratory tests were also provided at no cost to the participants. Of the 53 employees, 45 (85%) participated in repeat laboratory testing at 3 months. After 3 months of attending this worksite diabetes education program, employees with diabetes were found to have lowered their mean fasting blood glucose level from 198 to 180 mg/dL. Employees also reduced their mean HbA1c from 9% to 8.3% (p < 0.001).

Among the 36 employees who participated in repeat testing at 6 months, mean fasting blood glucose levels at baseline, 3 months, and 6 months were reported to be 193.4, 179.2, and 169 mg/dL, respectively.[11] These results represented a 7.3% decrease in fasting blood glucose after 3 months and a 12.2% decrease after 6 months. The mean HbA1c values at baseline, 3 months, and 6 months were 9.0%, 8.4% (p = 0.006), and 8.5% (p = 0.024), respectively.

Employee empowerment through education for improved diabetes self-care was the key to success in this research. Improved diabetes control within 3 months and maintained at 6 months is an important short-term benefit, but the study investigators could only speculate about the long-term benefits of fewer days of absenteeism, lower medical costs related to diabetes and complications, and ultimately enhanced worker productivity.

Trief et al.[31] examined 129 adults with insulin-requiring diabetes who were employed outside of the home to determine whether there was any association between the work environment and adequacy of metabolic control or the individuals’ adaptation to diabetes. They further studied the interactions between an individual’s life at work and ways of coping with diabetes. The authors used two work system measures to assess the social environment of work setting (i.e. The Work Environment Scale and the Work Apgar Scale), perceptions of support at the workplace, and two quality-of-life measures in their study (i.e. The Diabetes Quality of Life Scale and the Appraisal of Diabetes Scale). Glycemic control was assessed using HbA1c measurements. The results of the analyses indicated that neither of the work system measures was a significant predictor of HbA1c. With respect to psychological adaptation, supervisor support was found to be a significant predictor of positive appraisal and diabetes-related satisfaction. Involvement and coworker cohesion also predicted aspects of diabetes-related quality of life at work. The researchers concluded by emphasizing that worksite interventions should target not only employees with diabetes but also their supervisors and coworkers. The fact that more than 16% of study participants had hidden their diabetes may indicate that there is still a stigma associated with the disease. Researchers argue that fear of layoff or receiving fewer promotions may make employees who are open about their diabetes hesitant to request worksite interventions and supervisor support.

2.3 The Role of Managed Care in Improving Diabetes Health Outcomes

What is the role of managed care in improving the health status of adults with diabetes? Managed care has an intense interest in community wellness rather than community illness. Therefore, population-based medicine is growing in popularity among managed care plans. Diabetes represents a very expensive chronic disease that could be reduced in incidence and long-term medical complications through more effective diabetes education and patient self-management. Berg and Wadhwa[5] have argued that diabetes self-management has been proven as a successful intervention to save health insurance programs money and improve patient outcomes. Behavioral and lifestyle changes require individual motivation, which could occur after intense education and incentives for the patient with diabetes to manage their blood glucose levels. Similarly, Clark et al.[32] have argued that improvements in clinical outcomes for individuals with diabetes could occur in a managed care setting. Their study found that patient empowerment was the critical variable in patient compliance with physicians’ recommendations. More recently, a multicenter study[27] was undertaken to determine how the structure and organization of managed care systems influenced the process and outcomes of diabetes care. By linking information from health plans, provider groups, and patients across a wide range of models of care and diverse populations, this study aims to improve understanding of both the facilitators and barriers to diabetes care in managed care settings.

Several studies have examined the potential short-term clinical and economic benefits of improved glycemic control for patients with diabetes under managed care. In one of the health plan-sponsored diabetes interventions, Gilmer et al.[33] examined the inpatient and outpatient cost for enrollees with diabetes in a nonprofit health maintenance organization, HealthPartners, using administrative and laboratory data. The findings indicated that those members with higher HbA1c values had higher medical costs over a 3-year period. As can be seen from table II, expenses were reported to be significantly higher for each percentage point increase above an HbA1c of 7%. Costs were even higher for individuals with coronary heart disease and high blood pressure. While an overall increase in costs was demonstrated in this study, the specific components of the resource use associated with poor glycemic control were not discussed.
Table II

Diabetes treatment cost control flow chart[33]

Similarly, Menzin and colleagues[34] studied the inpatient admissions for short-term complications and associated costs for diabetes patients at the Fallon clinic, a multi-specialty group clinic in Worchester, Massachusetts, USA. They conducted a retrospective study using enrollment, medical claims, and clinical laboratory data for approximately 2500 adults with diabetes for the time frame 1994–1998. The researchers found that patients with poor glycemic control had more than double the number of inpatient admissions over a 3-year period compared with those with good glycemic control (31 vs 13 admissions per 100 patients, respectively; p < 0.01). The average adjusted per-patient charges for inpatient treatment among those with good control were approximately $US970 over the study period compared with $US3040 (costs based on a 4-year study period from 1 January 1994 to 30 June 1998) for those with poor glycemic control. There was an attempt to adjust for the severity of diabetes among the study participants by controlling for age, sex, several chronic diabetic complications and follow-up time. However, the authors acknowledged that the reduction in costs associated with improved glycemic control may have been overstated due to suboptimal statistical control. In addition, the database used for this study did not enable the researchers to distinguish acute hospital stays from admissions to skilled nursing facilities. To address this issue, the researchers conducted chart reviews based on inpatient stays from two randomly selected patient samples. The results of the chart review indicated that the number of inpatient stays for acute diabetic complications may have been underestimated, suggesting that the study findings may be conservative.

Sadur et al. reported on a randomized controlled trial among patients with diabetes at Kaiser Permanente’s center in Pleasanton, California, USA, who had either poor glycemic control (HbA1c > 8.5%) or no HbA1c test performed during the previous year.[35] In this study, intervention patients received multidisciplinary outpatient diabetes care management delivered by a diabetes nurse educator, a psychologist, a nutritionist, and a pharmacist in cluster visit settings of 10–18 patients/month for 6 months. After the intervention, the researchers found that HbA1c values declined by 1.3% in the intervention group versus 0.2% in the control group (p < 0.0001). Both hospital (p = 0.04) and outpatient healthcare resource utilization (p < 0.01) were found to be significantly lower for the intervention group after implementation of the program. Limitations of this study include the failure to obtain follow-up HbA1c values and questionnaires on approximately 16–25% of the study participants, respectively, and lack of information on the cost of the intervention itself. However, this study suggested that improved glycemic control may lead to an earlier reduction in healthcare utilization, which may offset costs of the intervention in the short-term.

In another study, Sidorov et al.[36] compared healthcare costs for patients who fulfilled health employer data and information set (HEDIS) criteria for diabetes and were in a US health maintenance organization-sponsored disease management program (n = 3118) with costs for those not in disease management (n = 3681). Based on 2-year data, per-member per-month paid claims averaged $US394.62 for program patients compared with $US502.48 for non-program patients (p < 0.05) [costs based on 2-year study period from 1 July 1999 to 30 June 2001]. Program participants were found to have lower inpatient healthcare use and emergency room visits. Program participants also achieved more favorable results for HbA1c testing as well as for lipid, eye, and kidney screening. Based on these results, the researchers concluded that disease management may result in savings for sponsored managed care organizations, and improvements in HEDIS measures are not necessarily associated with increased medical costs.

Berg and Wadhwa[5] examined the behavioral change and medical care utilization impact of a US health plan-sponsored diabetes disease management program in a community-based setting. They followed 127 patients with diabetes from baseline through 1 year. Poisson multivariate regression models were estimated for counts of inpatient, emergency department, physician evaluation and management, and facility visits, while controlling for potential confounders. Within 6 months, it was found that the number of participants having a HbA1c test increased by 44.9% (p < 0.001) and there was a 53.2% decrease in symptoms of hyperglycemia (p = 0.002). The results of the regression analysis showed that inpatient admissions decreased by 391 (p < 0.001) per 1000 enrollees for each group while controlling for age, length of membership, and the number of comorbid claims for congestive heart failure. The cost analysis concluded that there was a 4.34 : 1 return on the program.[5] Despite a relatively small sample size, this statistical model demonstrated many significant findings. The researchers acknowledged that since their model emphasized macrovascular risk reduction they were able to show a rapid return on the program, and that there is a need for further research to evaluate the duration of behavioral change and reductions in healthcare resource utilization for periods greater than 1 year.

Selecky[37] examined the implementation of a technology-enabled disease management program for diabetes in a mixed-model US health plan, the Fallon Community Health Plan. The results showed a 9% cost reduction for the population in the LifeMasters Diabetes Disease Management Program compared with a 3% cost reduction for the remainder of the Fallon membership. The net savings from the program were slightly greater than 5%, which were higher than expected savings. Fallon and it reported a compliance rate of 90% for daily vital signs entry in the diabetes disease management program. Such rankings and compliance rates could produce more significantly positive results than with less compliant populations or those in which the quality of care is lower, particularly those outside of a managed care population.

2.4 Self-Management and Long-Term Diabetes Interventions

An economic analysis of diabetes self-management programs that prevent hospitalizations was performed in nine empirical outpatient studies. Benefit-cost ratios reported in these studies ranged from 0.44 to 8.76. Thus, for every $US1.00 spent on self-management training, there was a net saving of $US0.44–8.76.[28]

The cost effectiveness of long-term diabetes interventions is well established by existing economic evaluation studies. The NIHlth[38] identified diabetes as the leading cause of new cases of blindness in adults aged 20–74 years. Javitt and Aiello[39] found that the cost of screening and treating eye disease in people with diabetes was $US3190 (1986 values) per quality-adjusted life-year (QALY) saved versus $US5100 for coronary artery bypass graft surgery or $US7650 for thyroid screening. The Centers for Disease Control and Prevention (CDC) Diabetes Cost-Effectiveness group recently reported the incremental cost-effectiveness ratio for intensive glycemic control to be $US41384 per QALY (1997 values). They further reported that this ratio increases with age at diagnosis from $US9614 per QALY for patients aged 25–34 years to $US2.1 million for patients aged 85–94 years.[40]

Type 2 diabetes is rapidly increasing in prevalence in the adolescent population, resulting in future workers developing this chronic disease before they even enter the work force. This compounds the problem of costs associated with diabetes because of the need for health insurance coverage for dependents with this disease, and the fact that these individuals will bring the disease with them when they become employed. Therefore, when utilizing cost-benefit analyses on any chronic disease it is very important to analyze the length of time the individual will have this particular disease.

Diabetes is also the leading cause of end-stage renal disease, accounting for approximately 40% of new cases.[38] The annual cost for treatment of end-stage renal disease exceeds $US2 billion, excluding reduced productivity and employment costs (1991 values).[41] Research by Rodby and colleagues[42] indicated that treatment with captopril resulted in absolute direct savings or benefits of $US32 550 per patient with type 1 diabetes over a lifetime and $US9900 per patient with type 2 diabetes (1995 values). Absolute savings in indirect costs were $US84 390 for type 1 diabetes and $US45 730 for type 2 diabetes. Finally, Elixhauser and colleagues[43] showed that providing preconception counseling and care (in addition to prenatal care) for women with known diabetes in order to avoid complications resulted in cost savings of $US1720 per participant (year of costing not provided). The authors concluded that for every dollar expended in the preconception program, $US1.86 was saved in direct expenses.

3. Discussion

Diabetes is a serious, costly disease that is rapidly increasing in prevalence in the US. It is the most common endocrine disorder of adults in the workforce,[11] with substantial direct and indirect costs for the employer paying for health insurance. Diabetes presents enormous potential problems for patients, the workplace and the healthcare system. These problems include management and costs of complications and loss of employee productivity.

Diabetes does not have to cause long-term complications if managed properly. This management should include workplace education programs designed to eliminate high-risk behaviors. Burton and Connerty[11,30] have shown that worksite diabetes education programs improve control of blood glucose levels and, therefore, should result in reduced direct and indirect costs for both the employer and the employee, reduced absenteeism and enhanced quality of life in the long term, provided that intervention takes place at an early stage of the disease, before the onset of complications.

3.1 Uncontrolled Diabetes as a Societal Issue

Control of diabetes has been shown to delay the onset and slow the progression of many complications, thereby resulting in avoidance of costs related to such complications. The cost of uncontrolled diabetes is a serious issue for society at large and for employers and their employees. Employer action to improve diabetes care and education will help their workers remain productive, decrease diabetes-related complications, and reduce associated costs over time. A dedicated effort and financial investment at the senior management level are essential to achieving these goals.

Employer- and/or health plan-sponsored diabetes management programs are one way that employers can help their employees better manage this chronic condition. These programs can not only raise awareness of the dangers of diabetes and the importance of controlling blood glucose levels but also stress the importance of diet and exercise, as well as other elements of a healthy lifestyle. An integrated information system that tracks diabetes-related medical claims and costs is crucial in enabling employers to identify the opportunities where costs and outcomes can be improved.

While most government payors and employers know about diabetes management programs, relatively few have actually implemented such programs. The prevailing perception is that these programs are too complicated to implement. Others await additional research studies demonstrating that these programs actually improve employee health and reduce costs. Some even believe that diabetes management should be the role of managed care plans, and that offering them would be nothing more than duplicating programs that are already in place. Healey et al.[44] argued that employers should not view diabetes education and intervention as an additional cost but instead see these measures as a critical and strategic value-added component of human resources management.

For many years, the CDC has worked with the National Business Coalition on Health, the Washington Business Group on Health, the Employers Managed Health Care Association, and other US organizations to explore ways of improving the health of employees with diabetes, promoting health messages, and managing this condition in the workplace. The CDC also developed specific guidelines for employers to help them assist their employees with the management of diabetes.[8] These include developing a supportive work environment so that employees with diabetes feel comfortable adopting and performing behaviors that promote good diabetes control; providing encouragement and opportunities for all employees to adopt healthier lifestyles that reduce the risk of chronic diseases; coordinating all corporate diabetes control efforts to make them more efficient as well as accountable within the organization; and demanding the highest quality medical care for people who are dealing with diabetes.

Further studies are required to examine how implementing comprehensive diabetes disease management programs can be paid for through healthcare cost savings in patient populations or employees. Studies need to explore the incentive structure necessary for implementing such health management programs. The ADA believes that health insurance plans should pay for both the medical treatment of diabetes and self-management education programs that meet accepted standards, such as the American Diabetes Education’s National Standards for Diabetes Self-Management Education Programs. The ADA has further argued that employers purchasing healthcare benefits for their employees should insist on the inclusion of self-management education as well as diabetes-related medications and supplies in the scope of services covered.[45]

3.2 Policy Implications

There are a number of policy implications for government agencies and employers to reduce the clinical and economic impact of diabetes and its complications in the workplace. Early intervention can prevent or delay the onset of type 2 diabetes.[9,10] Furthermore, it can reduce direct and indirect costs in those with overt diabetes.[29] The benefits to employer groups and government-sponsored patient populations, such as Medicare or Medicaid, appear to be significant.

It is worth underscoring the significance of the studies discussed in this review by offering the following policy recommendations. Employers, in collaboration with their insurers, should consider investing in diabetes education and wellness programs. More explicitly, the following directions in health policy should be considered:

  • private insurers should offer incentives through discounts on the annual premium to employers who offer an approved employee education program

  • Medicare payors should offer incentives to former employers of employees now on pensions who are eligible for and are receiving Medicare payments for healthcare coverage

  • Medicaid payors should offer incentives to providers who accept Medicaid patients and offer these patients diabetes education programs

  • successful programs should be documented and shared among employers and health plans across the US.

Through additional research and the efforts of employers and health insurance plans, we can move closer to making the desired improvements in health status from workplace interventions for diabetes.


The authors wish to acknowledge the bibliographical and editorial contributions of Ms Jill Royston, WSU Health Policy and Administration Department Program Assistant and the bibliographical development and refinement work by Mr Dan Simonsen, Washington State University Graduate Student in the Health Policy and Administration Department. The authors received no funding for this manuscript and do not have any conflicts of interest directly relevant to the content of this review.

Copyright information

© Adis Data Information BV 2003