The Effect of Cost Sharing on Employees With Diabetes

Published Online: December 15, 2006
Sean Nicholson, PhD

Most employers are responding to rapidly growing healthcare costs by asking workers to pay more of the premium and by increasing prices (eg, deductibles and copayments) workers face at the point of care. In 1999, for example, 84% of people covered by employer-sponsored health insurance had a physician copayment of $10 or less, whereas 72% had physician copayments of $15 or more in 2005.1 Likewise, copayments for prescription drugs on the second (preferred branded) and third (nonpreferred branded) tiers have almost doubled between 2000 and 2005, rising from $13 to $25 and from $17 to $33, respectively. The objective of these changes is to encourage employees and their dependents to compare the benefits of medical care with the cost, to discourage the use of low-value medical services, and thereby to slow down the growth rate of health insurance premiums.

It is clear that increasing employee cost sharing achieves its primary objective of reducing costs.2,3 For people with certain health conditions, however, increasing employee cost sharing may have unintended consequences. Several recent studies show that patients with high blood pressure, elevated cholesterol levels, or diabetes mellitus (DM) are less likely to adhere to medical therapy when drug copayments rise.4-9 If the health of nonadherent patients worsens such that they require additional medical services, then reducing drug copayments may actually increase total medical expenditures. Furthermore, if one defines health costs broadly to include health-related absenteeism and reduced on-the-job productivity (often referred to as presenteeism) due to a health condition, direct medical costs may actually represent a small proportion of total health-related costs.10,11

At least 20 employers, including Pitney Bowes Inc and the University of Michigan, are adopting a contrarian approach.12-14 Rather than increasing copayments for all drug classes, these employers are selectively reducing copayments for conditions such as diabetes, asthma, and hypertension in which the clinical evidence suggests there is a strong connection between adherence to drug therapy and health. Unfortunately, there is no single study that performs a complete analysis of the financial effect of changing DM drug copayments on adherence, medical expenditures, absenteeism, and on-the-job productivity from the perspective of an employer, to my knowledge. The objective of this article is to perform such an analysis for DM by constructing a financial model that links together results from several different published studies. The Diabetes Outcomes Analyzer model, which is a disease-specific application of a general model proposed recently by Nicholson et al,15,16 focuses on DM because there are more studies examining the key links required for the financial model than for other diseases and because it is a fairly common and expensive condition.

Although this financial model can help employers design health benefits, one must be careful when interpreting the results. The model uses mean values (eg, the percentage change in pharmaceutical expenditures associated with a $1 change in the drug copayment) across several different studies and then links these values together to estimate the financial effect of drug copayments on total health-related costs. For pragmatic reasons, I assume that the employee populations and the examined interventions are similar to one another and are representative of employers nationally.


The baseline situation for an employer with 5000 workers is given in Table 1.2,7,10,11,14,17-24 The analysis focuses on workers only. Dependents and retirees could be incorporated, although the employer would not benefit from any changes in absenteeism or presenteeism. Each data element is described in column 1, the mean value based on the literature is given in column 2, and the sources or assumptions are reported in column 3. Using the National Health and Nutrition Examination Survey, Cowie et al17 reported the percentages of US employees between 1999 and 2002 by sex and by age category (20-39 years vs 40-64 years), as well as the percentages of employees (men and women combined) in these age categories who were diagnosed as having DM. Based on these data, an estimated 6.6% of the employed population has DM, or 332 workers for a typical employer with 5000 workers.

The patient populations, interventions, and results from several key studies are summarized in Table 2.1,7,14,18,25 Two studies examined the medication adherence rates for patients with DM. Among a population of 137 300 nonelderly employees and dependents insured by a manufacturing firm, Sokol et al18 found that 55% of the patients with DM adhered to their medication regimen. A patient was considered adherent if he or she received a supply of drugs for at least 80% of the year. Dor and Encinosa7 reported that 58% of patients with DM adhered to their physicians' recommended drug treatment among a population of 27 100 workers and retirees who received employer-based health insurance from 9 large firms. In the latter study, a patient was considered adherent if he or she had a full 3-month supply of drugs following the expiration of his or her first DM prescription, among patients who filled at least 1 DM prescription between June 1999 and September 2000. I averaged these 2 results and assumed that, before any drug copayments were changed, 56.5% of employees with DM adhered to their physicians' recommended drug therapy. For an employed population of 5000, this implies that 144 workers are not adherent among 332 workers with DM.

The model assumes that an employer's pharmaceutical and nonpharmaceutical expenses per employee with DM are $3922 and $10 776, respectively. These are the mean amounts reported by Sokol et al18 and by Garrett and Bluml.14 Garrett and Bluml studied a population of 256 patients with DM in Asheville, NC, who were covered by employer-sponsored health insurance (Table 2). The spending amounts reported in these 2 studies were adjusted from the period examined (1999 for Sokol et al18 and 2004 for Garrett and Bluml14) to 2006 values by using the overall US growth rate of pharmaceutical and nonpharmaceutical expenses. Specifically, I applied a 12.5% annual growth rate for pharmaceutical expenses between 1999 (the date of the study by Sokol et al18) and 2004; I applied a 9.1% annual growth rate for nonpharmaceutical expenses during the same period. The growth rates from 2004 (the date of the study by Garrett and Bluml14) to 2006 (8.2% for pharmaceutical expenses and 7.9% for nonpharmaceutical expenses) are the actual growth rates in the United States between 2003 and 2004.

Eight studies examined how often workers with DM were absent because of their health condition. Although the questions differed somewhat across the studies, the data were self-reported by surveyed employees, and employees were usually asked to distinguish health-related absences from overall absences. The numbers of annual self-reported absence days ranged from 1.2 days to 12.6 days across the 8 studies, with a mean of 4.4 days. Goetzel et al10,11 summarized 4 of these studies: the Medstat MarketScan Health and Productivity Management database that contained absenteeism information on 375 000 employees between 1997 and 1999 at 6 large corporations, the Employers Health Coalition, Inc,9 that surveyed 10 000 employees at 8 large employers in 1998 and 1999, the Work Productivity Short Inventory (Ozminkowski et al19) that evaluated 619 employees at a large telecommunications company, and the Bank One Worker Productivity Index (Burton et al20) that collected data from 1000 Bank One customer service operators in Illinois in 1995. To arrive at the overall mean of 4.4 days per year, the data from these 4 studies were combined with the data from 4 other studies on how frequently employees with DM were absent: Collins et al21 surveyed 6000 employees at Dow Chemical Company in 2004, Cranor et al22 and Cranor and Christiansen23 examined 164 patients with DM covered by employer-sponsored health insurance in Asheville, Wang et al24 examined an employer-based population, and Egede et al26 examined patients with DM in the general population.

PDF is available on the last page.


Recommended Reading

No Result Found