Is There a Business Case for Diabetes Disease Management?

Evidence-Based Diabetes Management, March/April, Volume 19, Issue SP2

Disease management programs, particularly for diabetes care, have been part of managed care benefit offerings for roughly 20 years. They are offered by nearly every health plan, whether commercial, Medicare, or Medicaid. Most experts would acknowledgethat disease management programs can improve the health of patients with chronic disease. However, little reliable, reproducible information is available to answer the question of whether these programs are cost saving.

Back to the Beginning

With the evidence in hand that tight glycemic control can prevent long-term complications of diabetes mellitus1,2 and the ability to measure and monitora person’s blood glucose levels at a moment’s notice, a new medical concept and burgeoning industry spawned in just a few years. In 1991, the Boston Consulting Group characterized this as disease state management,3 and described it as a new way for the pharmaceutical industry to prove (and enhance) the value of its medications. Soon after, pharmaceutical companies like Pfizer and Zeneca Labs began their own disease management programs,3 shopping them to the health plans and insurers.

The health plans realized that the disease management process represented a workable way to improve clinical outcomes and health status over an entire population of patients and codified a way to broadly conduct and promote secondary prevention. Furthermore, disease management programs were acknowledged to be a type of quality improvement (QI) process, which helped managed care plans fulfill QI requirements for accreditation from the National Committee for Quality Assurance. However, they had a choice of whether to contract with these pharmaceutical company—sponsored providers, work with independent start-up disease management organizations (eg, Diabetes Treatment Centers of America [which later became Healthways] or Cardiac Solutions [which later became Healthways]), or start their own internal programs.3

Over time, pharmaceutical company programs became burdened with concerns regarding conflict of interest and were soon outpaced by independent vendors and health plan—sponsored programs. The concept, in any case, took off. By 2006, disease management programs were part of every managed care organization’s portfolio of benefits fits: a survey by the Boston Consulting Group revealed that 96% of health plans had disease management programs.4 In 2010, one source reported that 21% of patients with at least 1 chronic disease were in a disease management program.5 Today, the domestic disease management market is $2.1 billion, but should grow significantly in 2014 (once health reform increases the size of the market for covered eligibles).6

The Poster Child of Disease Management

Although disease management can be applied to many chronic diseases, diabetes has been a prime target. Diabetes mellitus is highly prevalent, it results in tremendous costs (a significant portion of which are preventable), and it can be easily measured. The programs’ interventions also were well suited for QI: patients can be tracked, their pharmacy claims can be assessed for adherence, and their outcomes can be quantified, in terms of glycated hemoglobin (A1C) levels, clinical events, and use of medical resources (including emergency department [ED] and hospitalization).

According to the Care Continuum Alliance (formerly the Disease Management Association of America), the principal components of disease management programs include7:

• A method of identifying populations requiring secondary prevention

• Standards of care according to evidence-based practice guidelines

• Collaborative practice models to include physician and ancillary healthcare providers, including case managers and nurses

• Patient education on how to selfmanage their condition and change unhealthy behaviors

• A process for measuring outcomes, evaluating the effect of interventions, and managing the program

• Routine reporting and feedback loop to providers (which include communication with patients, physicians, health plan, and ancillary providers, and practice profiling)

Generally, patient interactions are through nurse call centers and are focused on health education and reminders.

Clear Economic Outcomes Are Elusive

These components were common to most first-generation disease management programs, and though they sounded like potent tools to contain costs and improve outcomes, their effect on outcomes was not generally impressive. Even when quality measures were demonstrated to improve, the savings did not necessarily follow.

A retrospective study of 2 years of claims data from an integrated healthcare system found that patients with diabetes enrolled in their disease management program had average claims of $394.62 per patient per month compared with $502.48 per patient per month for those with diabetes but not enrolled in the program.8 According to the researchers, the system saw a return on investment (ROI) of 2.23:1 (that is, $2.23 of savings accrued for every $1.00 invested); others question this result based on the cost calculation.9 Furthermore, this magnitude of savings could not be demonstrated by others. This was accompanied by modest—but statistically significant—reductions in hospital admissions and inpatient days.

In 2003, a study of HealthPartners’ diabetes disease management program determined that accrued savings over 10 years in this program would amount to only $75 per patient with diabetes.10 An analysis of Independence Health Association’s diabetes disease management program showed that it cost more money than it saved.10

Linden11 studied historical data of several chronic disease management programs. He performed an analysis (based on the number needed to decrease, as opposed to number needed to treat) to calculate the decrease in hospital visits (ED visits and hospital admissions) necessary to provide an ROI in the program. This analysis revealed that a decrease in admissions of 10% to 30% would be needed just to cover the costs of the program.11

It would seem then that the cost savings related to diabetes disease management programs are lower than expected, and the ROI is very much in question. A number of reasons for this may exist.12

Patients Identified Too Late. The use of predictive modeling programs that were available for use in first-generation programs did identify patients with type 2 diabetes who were at risk of increased expenditures and adverse health events, but that was because they had already occurred. As mentioned earlier, disease management programs were designed as secondary prevention programs, meaning that eligible patients were already subject to repeat ED visits or hospitalizations, and were already being treated (perhaps not optimally). As a result, these algorithms did not do a good job detecting patients who had high glycemic levels who had not yet incurred high costs (their primary, costly event, such as inpatient admission).

Comorbid Disease States. Patients with diabetes are at high risk for multiple comorbidities, including chronic kidney disease, hypertension, coronary heart disease, stroke, and others. A first-generation disease management program, which focused on the diabetic condition only, may improve some outcomes but not necessarily the comorbid clinical status. More recent disease management programs and population health models are patient-focused, rather than disease-focused, and are likelier to address the other related comorbidities.

Dependence on Call Center Management. The use of nurse call centers to deliver education and reminders to members is intuitively a positive aspect of disease management programs, but without other integral components, call center management has not been shown conclusively to significantly influence member education or to change members’ behavior. Much educational information is already available on the web. Furthermore, the physical distance between the call center and the patient creates a barrier to familiarity, trust, and continuity of communication.

Member Behavioral Self-Change. The program’s success depends on patients’ long-term adherence to the medication and exercise regimens prescribed by their doctors. This often requires incentives of some type. In first-generation programs, health plans did not necessarily coordinate disease management activities with formulary access (eg, to improve patient access to appropriate medications through lower cost sharing). This is also not a universal feature of today’s programs (coordination is particularly difficult in this respect with external program vendors).

Member Turnover. In addition to these factors, discontinuity in program participation may also pose a significant barrier. Related to annual turnover in health plans, the patient may be disenrolled in their previous plan’s disease management program, having to enroll anew with their new plan’s provider.

First- Versus Second-Generation Disease Management Programs

The greatest evolutionary differences in first-generation disease management with today’s programs are (1) the move away from a single-disease focus to a patient focus, which could address multiple disease comorbidities in 1 program; (2) the improved use of data and access to information through the medical record; and (3) the addition of incentives to induce patients to adhere to the medical regimen and providers to offer quality care.

Diabetes is not truly separable from its related disorders; obesity often leads to insulin resistance, which damages pancreatic beta cells, and results in type 2 diabetes.13 Patients with diabetes have high risks of hypertension, coronary disease, and cerebrovascular disease.14 Therefore, disease management programs that account for hese risk factors take a more holistic approach than firstgeneration programs that attempted to focus on improving glycemic levels only.

The quality, comprehensiveness, and interoperativity of electronic health records are considerably better today than in the mid-1990s. Today’s information technology systems allow for better integration of pharmacy claims data, medical records, and laboratory data, which enables greater analysis and facilitates improved care coordination. This is critical to the concept of patient-centered medical homes and accountable care organizations.

The current generation of disease management programs have gone beyond a call-center focus, and have experimented with offering inducements to patients to adhere to their edical regimen. This can be found in valuebased insurance design approaches,15 which may include reduced copays for medications for those in disease management programs or discounts for gym memberships.

Health coaching is also a popular addition to disease management programs today, as is a multidisciplinary approach to care teams16—hence the broader scope of “care management” programs, which may rely on case managers in addition to physicians as central coordinating personnel.

The Business Case for Diabetes Disease Management

There is little doubt that following evidence-based diabetes management Manageguidelines can demonstrate clinical benefits. The real question was whether those clinical benefits were great enough to translate into a strong economic argument. In a meta-analysis of randomized controlled trials of diabetes disease management programs, the authors found that disease management programs had the effect of improving A1C levels by 0.51% on average compared with usual care.17 If one agrees with the findings of the United Kingdom Prospective Diabetes Study, which linked a 1.0 point reduction in A1C with a 37% reduction in the risk of microvascular complications,3 the potential would seem to exist for significant cost savings years down the road.

From a societal viewpoint, this would make sense, if these efforts can be consistently applied over time. However, the high turnover rates at health plans alluded to earlier mean that a member with poorly controlled type 2 diabetes may move from one plan with extensive disease management program abilities to another with more modest abilities (if the patient is enrolled in the program at all).

Had the federal government decided that disease or care management programs were administrative costs for the purposes of calculating medicalloss ratios as part of health reform, this would have spurred a ompletely different business care argument (that is, can we even afford disease-management activities?).

From a health system standpoint, not only does the question of keeping patients in the program gnaw at medical executives and chief financial officers, but the very issue of ROI, which is one of the foundations of the business case for disease management, is not settled. Certainly, long-term savings associated with fewer complications in patients with superior care for diabetes would not be seen in the short term— a more likely scenario based on health plan turnover rates. However, some evidence does exist that short-term savings may result, not from lower complication rates, but from lower service utilization.

According to a study by RAND,18 investigations of ROI for disease management programs vary greatly in their scientific rigor, and as a result, costeffectiveness and ROI cannot be stated with any certainty.

In a review of the available evidence, Goetzel et al19 stated, “For studies reporting costs and enefits, a $0.70 ROI was calculated (lower than a break even). On balance, these studies point to the potential for diabetes DM programs to break even, if reatment costs are well managed.”

Regardless of the ROI, perhaps diseasemanagement is simply the “right thing to do.” Clinically, it does appear to moderately improve glycemic levels, and it may spur patients with diabetes mellitus to think of their health in the wider terms of cardiovascular risk, kidney health, and obesity. A study from the University of Michigan found that a comprehensive diabetes disease management program not only obtained a 0.6 point reduction in A1C level, but also reduced mean systolic blood pressure by 3 mm Hg and low-density lipoprotein concentrations by 18 mg/dL.20

Does ROI Matter?

EBDM

With more than 90% of health plans and insurers maintaining diabetes disease or care management programs as part of the benefit offerings, and with less-than-convincing evidence of ROI in these programs, it seems that something else is at work. Could it be that the competition for health plan members trumps any cost savings calculation for a program most implicitly believe is beneficial? With so many plans offering disease management benefits, and members and employers coming to expect them in their benefit packages, perhaps chief financial officers merely see disease management programming as a cost of doing business.

Payer PerspectiveInterview With Geoffrey Mwaungulu, MD

EBDM: How long have you been personally involved in disease management programs?

Dr Mwaungulu: That would be more than 10 years now, starting with my work at Health Alliance Plan.

EBDM: Over that time, what has been your overall perception of disease management in the major chronic diseases (ie, diabetes, cardiovascular, asthma)?

Dr Mwaungulu: Early on, our disease management programs were fairly “light touch,” and they really weren’t all that effective. We tended to just send educational items to the members we identified. We just weren’t sophisticated in terms of our approach. And we often separated the diseases rather than aggregating them depending on the patient. We have also moved more toward managing the whole population in our messaging and interventions.

EBDM: There might be a specific disease management program for diabetes, another for cardiovascular, and another for asthma?

Dr Mwaungulu: Yes. More recently, our understanding of disease management and thesophistication of the approach to disease management improved. We now understand the importance of changing member behavior with regard to their chronic illnesses. As a result, the approach has included a bit more focused engagement with the member. For example, the case manager will do interviews using some standardized approaches that help members understand how they can change their “unhealthy” behaviors.

EBDM: Does this mean more frequent contact with the patient?

Dr Mwaungulu: More frequent and continuing contact with the patient. If the patient has 2 or more chronic conditions, they are addressed at the same time. This may involve, for example, managing the hypertension in the member with diabetes. We also understood along the way that we had to tier the members according to not just the chronicity of the condition but the severity of their illness. The interventions and the intensity of the interventions would reflect that.

EBDM: You had mentioned that the case manager generally focuses on patient contact or engagement. How does this person interface with the patient’s primary care physician (PCP)?

Dr Mwaungulu: If the case manager, at the disease management level, finds issues that he or she believes should be brought to the attention of the PCP, the case manager has the option of reaching out directly to the physician, which they ordinarily wouldn’t do unless an issue requires urgent attention. More commonly, the case manager will tell the member to talk to his or her PCP about this particular issue, and then follow up with the member. Beyond disease management—into episodic case management—the PCP must be involved more directly to closely monitor the patient’s condition.

EBDM: Does your health plan currently use an external disease management provider?

Dr Mwaungulu: We have a layering of these things. Let me confine the discussion to the Medicare Advantage population first. We use an external vendor for the Medicare Advantage population for 4 target diseases: diabetes, congestive heart failure, coronary artery disease, and chronic obstructive pulmonary disease. Diabetes represents the biggest population, by far. These days, in trying to engage the members to improve their health and outcomes, we meet with the vendor on a regular basis to try to close gaps in quality of care. In the case of diabetes, the vendor really has to know the patient’s blood sugar levels, his or her hemoglobin A1C concentration, the cholesterol levels, whether the patient has had a kidney function test, whether the patient has had a follow-up eye exam, whether they have their blood pressure controlled, and whether they are on the right medication. They are responsible for doing a lot more today than in the past to improve the health outcomes of these members.

EBDM: This must be quite a challenge in terms of information technology. If you’re using an external provider, you must have some way to share the patient data.

Dr Mwaungulu: We have started sharing claims data with the vendor, so they get an understanding about what types of services are being done. In a few instances, the vendor actually has a link to members’ glucometers, in which glycemic concentration data can be uploaded by the vendor. They can communicate with the PCP if they see a problem.

EBDM: I understand that for the Medicare population at your health plan, the plan identifies members at high risk for complications and gives the information to the disease management vendor. Is this also the case for the commercial population?

Dr Mwaungulu: Yes, we identify which members are candidates and send the information on to our internal disease management staff. We can’t force members to participate in the disease management program. It’s up to the vendor to engage as many of these patients as possible.

EBDM: On the commercial side, do you link program participation with reduced copays or cost sharing?

Dr Mwaungulu: I don’t believe we offer any of these types of incentives to participate. I have seen other plans where copays have been reduced or waived. In those plans, if the member uses formulary-approved products, the copays can be reduced to generic levels. This encourages the member to be more adherent with the medication regimen.

EBDM: What’s your impression of any cost savings or return on investment with any disease management programs with which you’ve been involved?

Dr Mwaungulu: The return on investment tends to be small with any disease management program. If you can get a 1:1 return, I believe you’re still doing a great service for your membership. It is more than just saving the maximum amount of money—it is about improving outcomes. If you improve outcomes in general, you are probably going to see some savings. However, I don’t believe that disease management has traditionally saved money.

EBDM: For patients with diabetes, in particular, in what areas do you think disease management programs excel?

Dr Mwaungulu: In terms of outcomes, I think we’re seeing fewer people getting complications of diabetes, maybe less kidney disease over time. If you have regular eye checkups, there will be fewer cases of blindness. If you get your blood sugars controlled, you will probably see less peripheral vascular disease and more manageable coronary artery disease in the long run.

EBDM: Yes, in the long run. In the face of high health plan turnover, how do we achieve these outcomes? From the health plan’s perspective, what is the shortterm benefit?

Dr Mwaungulu: That has been one of the arguments against disease management. However, if the health plan works to optimize these outcomes today, they can receive some immediate financial benefits, in terms of improved HEDIS measures, which are part of the reimbursement formula for the Medicare Star program. For example, the Medicare Star program is interested in whether a patient with diabetes is taking their angiotensinconverting enzyme inhibitor or angiotensin-receptor blocker. These are part of the quality metrics we report to Medicare—resulting in a short-term benefit for the health plan in diabetes disease management programs. I also believe there is greater membership retention when they feel the health plan cares about them. In other words, the member turnover may be less than in the past.

EBDM: What do you think the next generation of disease management programs (or care management programs) will look like? How will they differ from those today?

Dr Mwaungulu: There will be more member engagement, showing them how to take better care of themselves. This will require even more case manager involvement with the member to emphasize educational approaches even more than in the past. There will also be more involvement by the members’ physicians, especially as more of them become patient-centered medical home—accredited and more them join accountable care organizations. For the health plans, these newer ways of delivering care will allow for more population management.

Dr Mwaungulu is medical director at Blue Cross Blue Shield of Michigan.Funding Source: None.

Author Disclosures: Mr Mehr reports receiving payment for involvement in the preparation of this article.

Authorship Information: Concept and design;acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content.References

1. The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Research Group. Retinopathy and nephropathy in patients with type 1 diabetes four years after a trial of intensive therapy. N Engl J Med. 2000;10;342(6):381-389.

2. Stratton IM, Adler AI, Neil HA, et al. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ. 2000;321(7258):405-412.

3. Todd WA. Disease Management: A Look Back and Ahead. International Disease Management Alliance; 2009.

4. Matheson D, Wilkins A, Psacharopoulos D. Realizing the Promise of Disease Management: Payer Trends and Opportunities in the United States. Boston, MA: Boston Consulting Group;February 2006.

5. Kalsekar I, Record S, Nesnidal K, et al. National estimates of enrollment in disease management programs in the United States:an analysis of the National Ambulatory Medical Care Survey data. Popul Health Manag.2010;13(4):183-188.

6. 10th Annual Report on the Disease Management ment and Wellness Industries. Disease Management Purchasing Consortium and the Institute for Health and Productivity Management,Boston, MA, and Scottsdale, AZ, 2012.

7. Care Continuum Alliance (CCA) definition of disease management. www.carecontinuum alliance.org/dm_definition.asp. Accessed November 19, 2012.

8. Sidorov J, Shull R, Tomcavage J, et al. Does diabetes disease management save money and improve outcomes? a report of simultaneous short-term savings and quality improvement associated with an HMO-sponsored disease management program among patients fulfilling HEDIS criteria. Diabetes Care. 2002;25(4):684-689.

9. Goetzel RA, Ozmintkowski RJ, Villagra VG, et al. Return on investment in disease management: a review. Healthc Financ Rev. 2005;26(4): 1-19. 10. Beaulieu ND, Cutler DM, Ho KE, et al. The business case for diabetes disease management at two managed care organizations: a case study of ealthPartners and Independent Health Association. The Commonwealth Fund, April 2003 (http:cmwf.org). Accessed November 26, 2012.

11. Linden A. What will it take for disease management to demonstrate a return on investment? new perspectives on an old theme. Am J Manag Care. 2006;12(4):217-222.

12. Lessons from first generation disease management programs. Society for Human Resource Management. www.shrm.org/hrdisciplines/benefits/articles /pages/dmlessons.aspx. Published September 22, 2011. Accessed November 19, 2012.

13. Pi-Sunyer S. The medical risks of obesity. Postgrad Med. 2009;121(6):21-33.

14. Diabetes, heart disease, and stroke. National Diabetes Information Clearinghouse 2005. http://diabetes.niddk.nih.gov/dm/pubs/stroke/index.aspx0. Accessed November 20,2012.

15. Tzeel AL. Diabetes benefit management:evolving strategies for payers. Am J Manag Care. 2011;17(suppl 14):S384-S390.

16. Bratcher CR, Bello E. Traditional or centralized models of diabetes care: the multidisciplinary diabetes team approach. J Fam Pract. 2011;60(11 suppl):S6-S11.

17. Pimouguet C, Le Goff M, Thiebaut R, et al. Effectiveness of disease-management programs for improving diabetes care: a meta analysis. CMAJ. 2011;183(2):e115-e127.

18. Mattke S, Seid M, Ma S. Evidence for the effect of disease management: is $1 billiona year a good investment? Am J Manag Care. 2007;13(12):670-676.

19. Goetzel RA, Ozmintkowski RJ, Villagra VG, et al. Return on investment in disease management: a review. Healthc Financ Rev. 2005;26(4):1-19.

20. McEwen LN, Hsiao VC, Nota-Kirby, et al. Effect of a managed care disease management program on diabetes care. Am J Manag Care. 2009;15(9):575-580.