Managed Care Interface

Decreasing Cost and Increasing Patient Satisfaction: The Implementation of a Cancer Disease Management Program

William D. Kirsh, DO, MPH, Rick Lee , MHA

In order to address growing costs among patients with cancer, two firms developed a disease management program relying heavily on case management techniques. Herein, the authors report on the success of this program, both in terms of costs and patient satisfaction.

In response to growing costs and the eroding effect of conventional medical management strategies to control them, Foundation Health (FH) has adopted a series of database-focused, case management interventions that are delivered through a coordinated, specialty health care system. Key to this model is the integration of onsite partnerships and technology.
      In early 1997, FH engaged Quality Oncology (QO) to help craft a solution to the problem of its double-digit expenditure growth associated with cancer care. This problem was of particular importance because of its growing Medicare risk population, and the relatively high prevalence of cancer in the elderly. The program was introduced in June 1997 for about 110,000 commercial members and 23,000 Medicare risk members. The collaborators in this joint program sought to test a number of assumptions about disease management and its specific application to cancer.

PROGRAM FEATURES

The first step in this program was to visit with PCPs and oncologists in order to implement the new network. Cancer disease management had been introduced in South Florida in 1994. Resistance from doctors to the FH program was anticipated, addressed, and ultimately overcome by proving the value of the initiative.
      When the program was initiated in June 1997, QO provided an onsite oncology nurse at FH's administrative office to help with daily problem solving. The company also offered telephone-accessible oncology nurses who, as care managers, helped empower patients by providing direction, reliable information, and resources.
      A third feature of the program, a full-time oncologist who interfaced with the provider community, was introduced in the Spring of 1998. Since all FH-contracted doctors were required to submit their cancer treatment plans to QO, conversations between QO's full-time oncologist and the community doctors were necessary for approximately one in every 10 cases. The QO oncologist's presence helped to precipitate systemic change throughout the service region, as QO's cancer treatment guidelines were implemented in an effort to combat inappropriate practice variation. Table I illustrates the various program differences.
      There was resistance from the provider community to the disease manager's presence. However, FH convinced providers to work with QO by requiring QO involvement as part of payment.

PROGRAM RESULTS

The data from the first year's effort, which spanned from July 1, 1997 to June 30, 1998, were recently unveiled. Noteworthy gains were achieved in all key measurable areas. All first-year program results were compared with the same time period in the year previous to the program's inception, using FH's claims as the source for baseline in- dices. For the commercial population, using average cost per case as a standard for financial impact, FH spent $9,241 per case in the first program year, a reduction of $4,756 (34%) from the previous year's experience (Figure 1).

 

TABLE 1: PROGRAM DIFFERENCES BETWEEN GENERAL VERSUS SPECIALIZED CANCER DISEASE MANAGEMENT   Generalized DM System Specialized DM System Care managers Generalist RNs Oncology RNs Medical director Internal medicine Medical, surgical, or  radiation oncology Treatment plan Fragmented or  nonexistent Evidence-based  by tumor site Access for patients Little clinical direction  during working hours Continuous telephone  access to RNs Treatment diagnostics Excessive;  inappropriate,  or duplicative Coordinated, efficient End-of-life treatment Are hospice use;  often in last week  of life 4-6 wk of hospice care;  majority of cancer  deaths occured in  hospice DM = Disease management, RNs = registered nurses

 

	Figure 1. Commercial case cost reduction

The results for the Medicare population (7% reduction) were not nearly as impressive. However, when commercial results are combined with the Medicare risk results, the cancer disease management program netted a 19% reduction in cancer claims per case for 2,473 cases in the first 12-month period. Total net cost avoidance to FH was greater than $4 million (Figure 2).

 

	Figure 2. Total plan savings for program year 1.

As one often sees in health care, patients with cancer fit the Pareto model, with a small minority of the patients responsible for the overwhelming majority of the costs. A total of 34% of the commercial patients with cancer, who were referred to QO for care management, consumed 77% of the cancer expenditures and 80% of the cancer inpatient days. Patient identification efforts were, for unknown reasons, not as successful with the Medicare risk population, in which 29% of patients used 58% of the cancer dollars (Figure 3). Foundation staff reorientation and communication efforts were stepped up in year 2 of the program to address this issue.

 

	Figure 3. Pareto metrics. The percentage of cancer cases indicate those managed by Quality Oncology. Other percents indicate proportions of total. ER = Emergency room.

 

	Figure 4. Average cost per patient with cancer per year, active and follow-up patients with cancer: Baseline versus program year 1.

As seen in Figure 4, the percentage of active patients changed from 49% of all patients with cancer in the baseline period to 46% of all patients with cancer in program year 1. Active patients provide the best estimate of severity in the cancer population, because they are receiving radiation, hormonal treatments, or chemotherapy during the study period. Surgical patients and institutionalized patients with cancer are also considered "active."
       Care managers promoted the utilization of the full continuum of care, such as home health and hospice programs, that are also offered through vendor/partners onsite at FH. On a per-member per-month (PMPM) basis, FH reduced its inpatient costs by 40%, from $3.35 to $2.00 for the commercial population. Care managers are also accessible to patients during weekends and weekdays, 24 hours a day. Forty-nine percent more dollars were allocated on a PMPM basis to home health for the commercial population, and 108% for the Medicare risk members compared with the baseline period (Figure 5).

 

	Figure 5. Per member per month (PMPM) cost by site of care. Baseline cost PMPM = $6.19, program year 1 cost PMPM = $5.34.

Many families often misunderstand the benefits of hospice services. As a result, this benefit is often underutilized. In program year 1, 96 FH members used hospice, representing 17% of the patients with active cancer under QO's management. A total of 58% of members who died from cancer expired in hospice. However, unlike most of the data cited, death is not easily captured from claims data. Therefore, the baseline ratio of deaths in hospice could not be obtained when the baseline information was built from FH's claims data.
      Chemotherapy is one component of cancer cost that is often misperceived by HMO managers to be much higher than it is. However, it represents a significant portion of income to medical oncologists.² In fact, studies by QO have confirmed that chemotherapy is generally a modest proportion, between 4% and 8%, of HMO spending on treatment of cancer. For FH, the commercial PMPM spent on chemotherapy in the baseline period was $0.24, or about 3.9% of the health plan's baseline cancer costs. For the Medicare risk population, chemotherapy cost $2.13 PMPM during the baseline period, or 5.8% of Medicare risk cancer costs. In the first year of the program for the commercial population, these costs rose 26% to $0.31 PMPM; for the Medicare risk population, the cancer chemotherapy expenses rose 13% to $2.42 PMPM. Once again, the prevalence increases of 31% for commercial and 22% for Medicare risk account for all of the PMPM increases in chemotherapy.
      Quality Oncology maintains a cancer claims database for 1997-1998 comprised of 5.2 million lives drawn from 21 payers in the United States. This cancer repository provides a range and median for common cancer utilization and cost metrics. From the repository data, similar southeastern HMOs with a similar cancer prevalence rate (6.8 cases/1,000 lives) had a 43% higher PMPM expenditure compared with FH's program experience (Figure 6). For the Medicare risk population, firms with a similar prevalence rate (71.4 cases/1,000 lives) had a 40% higher PMPM (Figure 7).

 

	Figure 6. Commercial case and cost comparison, adults. PMPM = Per member per month; QO = Quality Oncology.

	Figure 7. Medicare risk case and cost comparison, adults. PMPM = Per member per month; QO = Quality Oncology.

Finally, the most important test of a value-added program is the degree of satisfaction experienced by the members. Using a five-point Likert Scale (5 = strongly agree, 1 = strongly disagree), a 37% response rate from FH patients recorded a 4.5 average score for the Ii-question, cancer-specific survey (Figure 8). In comments recorded by the patients, QO's care managers were likened to "guardian angels" and other superhuman allies.

DISCUSSION

Would trained, experienced oncology nurses exert more influence with cancer patients and oncologists than general utilization management personnel? Quality Oncology firmly believes that oncology physicians and nurses, by speaking the language and understanding cancer's epidemiology, make community oncology more accountable and manageable.

 

	Figure 8. Patient satisfaction results. CM = Care manager. Note: Average score for all 11 questions was 4.5, and the response rate was 123 out of 332 surveys sent (37%).

Is it possible to say "no" to treatment requests like bone-marrow transplant in the highly charged world of cancer? Preventing questionable bone-marrow transplants remains a dicey proposition at best. Quality Oncology's initiative did not challenge most requests, electing to stay off the front page of the newspapers and out of the crosshairs of the trial lawyers.
      Is the reputed wide variation in practice occurring in cancer treatment the result of a lack of oversight or an exhaustive search for solutions?⁴ By forcing compliance with cancer-specific clinical treatment guidelines, QO reduced treatment variation.
      What are the likely savings associated with the disease management approach, and are these savings sustainable over time? It remains to be seen whether or not the 19% blended case cost reduction will be sustained in the remaining two years of the contract. Certain routinely tracked indices, such as hospital bed days, suggest continued cost improvement.
      How should disease management programs be introduced to at-risk provider groups? Quality Oncology has shared these first year results with risk-taking medical groups. Their initial and predictable resistance has faded in the wake of financial results that benefit their bottom lines.
      The management of chronic disease states will continue to challenge health care organizations. As policy makers look to the managed care industry to solve the Medicare solvency problem, it is inevitable that finding solutions for chronic conditions will be a critical part of the answer. Both FH and QO believe this cancer disease management program offers some valuable lessons for other payers:

• Coordinate specialty care through knowledgeable specialists who can communicate with their peers and are empowered to make appropriate decisions.
• Seek ways to involve patients, so they can enhance the outcomes of disease management programs.
• Acknowledge the Pareto principle and focus resources on those who will consume the vast majority of the medical costs.
• Find ways to create long-term, win-win partnerships with specialists.
• Be proactive; the problem of chronic care will only worsen as the baby boomers age.
• The structure and dynamics of the medical delivery system are changing rapidly. The benefits of managed care for all customers must be validated and proven. The analyses presented here are meant to demonstrate measurable improvements in cost, use of a continuum of care, avoidance of unnecessary hospital use, reduction in practice pattern variation, and patient satisfaction.

REFERENCES
1. Kurowski B: Cancer carve outs, specialty networks, and disease management: A review of their evolution, effectiveness, and prognosis. American Journal of Managed Cave 1998;4(suppl):71-89 .
2. Smyth AC: Reimbursement issues for the oncologist. Oncology Xeimbursementl993;1:1-4.
3. Lee FC: Disease management in the treatment of cancer. Medical Interface 199S8(12):126-131.
4. Hewitt M, Simone J: Ensuring Quality Cancer Care. Washington, DC, National Cancer Policy Board, Institute of Medicine, National Academy Press, 1999; 10.

Address for correspondence/reprints: Rick Lee, MHA, Chief Executive Officer, Quality Oncology, Inc., 1430 Spring Hill Road, Suite 106, McLean, Virginia 22102.

Mr. Lee is CEO of Quality Oncology, Inc., in Sunrise, Florida. At the time this paper was written, Dr. Kirsh was CEO, Foundation Health, a division of Foundation Health Systems, also located in Sunrise, Florida.

Reprinted from Managed Care Interface August 1999. Copyright 1999 by Managed Care Interface.