Library
APS Bulletin • Volume 10, Number 3, May/June 2000
>Managed Care Forum
APS Managed Care Forum on Pain
Bill McCarberg, MD (Chair), Managed Care Committee, American Pain Society; Stephen D. Lande, PhD (Consultant)
Authors’ note: Chronic pain is a common presenting symptom among patients seeking medical care and is often associated with high utilization rates and major costs for medical care and social services (Liu & Byrne,1995; Polatin & Mayer, 1996).
FIGURE 1. Total Annual Costs of Treatment by Chronic Condition at Group Health Cooperative of Puget Sound in 1992(Costs were calculated by multiplying the number of patients by the mean cost per patient. The chronic pain category includes back and neck pain, facial pain, and headache.) |
 |
Note. Adapted from Fishman P., Von Korff M., Lozano P., Hecht, J. (1997). Chronic care costs in managed care. Health Affairs, 16(3), 239247. |
Data indicate that during a 1-year period, 4.8% of the covered population of a 3.5 million-member managed care organization (MCO) was given at least one prescription pain medication (White & Dubroff, 1999). Chronic pain accounts for greater total annual costs than other chronic conditions, including heart disease, hypertension, and diabetes (Figure 1) (Fishman, Von Korff, Lozano, & Hecht, 1997). Such findings indicate that chronic pain is highly prevalent and has a major impact on patients’ utilization of healthcare services and on costs in managed care settings.
APS developed a Managed Care Forum on Pain in April 1999 to formalize its efforts to improve the quality of pain care in managed care settings. In conjunction with its 18th Annual Scientific Meeting in October 1999, APS conducted an interactive roundtable meeting, “Evaluating the Health Economic Impact of New Pain Therapies in Managed Care.” The managed care participants included seven medical directors and one pharmacy director from some of the largest health maintenance organizations (HMOs) in the United States (i.e., Aetna US Healthcare, United Healthcare, Prudential Healthcare, Independence Blue Cross, and Pacificare). Meeting participants discussed a broad range of issues that influence the management of pain in managed care, as well as methods and models for evaluating the health and economic impact of pain disorders and treatment. The following article contains highlights of four roundtable presentations.
New Drugs and Devices:
Fitting into the Continuum of Pain Care
Speaker: J. David Haddox, MD DDS, Medical Director, International Analgesics, Purdue Pharma, LP, Norwalk, CT
The treatment of chronic pain presents a complex challenge to physicians because it does not follow the standard biomedical model. Too often, patients undergo excessive and unproductive diagnostic testing that frequently is inappropriate, produces unsatisfactory outcomes, and increases the utilization and costs associated with chronic pain.
Prevalence of chronic pain
Chronic pain is highly prevalent and has a major impact on the healthcare system. A survey by Louis Harris & Associates (1996) reported that pain was the third leading cause of workplace absenteeism in 1995, with 14% of the population having taken sick days because of pain. In addition, 75% of the population used over-the-counter medications for pain and 35% took prescription pain medications.
The Michigan Chronic Pain Study (EPIC/MRA, 1997) found that 1 in 5 adults in the state had significant chronic pain problems. It also found that 70% of the study’s respondents continued to experience pain despite treatment and that 29% utilized emergency departments for pain (EPIC/ MRA). A poll conducted by Roper Starch Worldwide (1999) found that approximately half of the respondents with severe pain reported poor control of their pain.
Regulatory and legal trends
Recent regulatory initiatives, new accreditation standards, and legal precedents are likely to have a significant impact on pain care. For example, in May 1998, the Federation of State Medical Boards (1998) formulated guidelines for the use of controlled substances in the treatment of pain. In addition, the Joint Commission on Accreditation of Healthcare Organizations has outlined a new initiative that recognizes patients’ rights to appropriate management of their pain (Joint Commission, 1999). To optimize patient care and avoid litigation, MCOs must ensure that Joint Commision—accredited affiliates conform to this new standard. Both healthcare organizations and physicians can be held liable for improper pain management. In today’s regulatory environment, pain cannot be ignored, and MCOs may face liability for neglecting pain care.
A continuum of treatment
The concept of a treatment continuum that begins with an accurate and comprehensive diagnosis is key to pain management. To date, evidence-based guidelines are not available for the management of chronic pain. However, a large armamentarium of pain therapies encompassing a broad spectrum of drugs and devices—from nonsteroidal anti-inflammatory drugs (NSAIDs) to potent opioids, and from spinal cord stimulators to opioid pumps—are available to clinicians.
The recent introduction of selective cyclooxygenase (COX-2) inhibitors illustrates the balancing act that MCOs face when selecting appropriate therapy. For example, NSAIDs, which have low acquisition costs, are associated with potentially costly adverse events (e.g., 76,000 to 107,000 hospitalizations per year for gastrointestinal complications) (American Medical Association, 1997; Bjorkman, 1996). Conversely, the acquisition costs of COX-2 inhibitors are high, but the incidence and, ultimately, the cost of drug-related adverse events (e.g., gastropathy) and complications appear to be substantially lower.
Chronic pain is an ongoing condition with ongoing treatment costs that are magnified by the high prices of new drugs and devices. Physicians also unwittingly drive up costs, either because they lack access to treatment guidelines or because they fail to follow existing guidelines. These factors should compel managed care decision makers to support programs (e.g., pain guidelines and programs, disease management initiatives, better physician education) that improve the quality and control the costs of pain care.
Panel discussion
The managed care participants agreed on the need to avoid expensive and inappropriate diagnostic and treatment interventions. They supported the continuum-of-care approach, which begins with a careful diagnosis and then matches the appropriate intervention to the diagnosis. Using this approach, expensive interventions would be used only when other less expensive and less invasive options are inappropriate or have been unsuccessful. The challenge is to adapt existing pain guidelines and clinical pathways to forms that MCOs can use to help physicians improve the treatment of chronic pain.
The participants concluded that the profile of chronic pain in managed care would continue to increase because of the rising costs of medication and technology; regulatory demands, new accreditation standards, and legal precedents; and the large volume of patients seeking treatment for pain.
Health Economic Models for Evaluating Pain Treatments
Speaker: David B. Matchar, MD, Director, Center for Clinical Health Policy Research, Duke University, Durham, NC
The implementation of efficient decision-making processes that include, exclude, or regulate treatment is a fundamental challenge in medicine. Decisions regarding resource allocation can be based on several factors, including the burden of illness, effectiveness of treatment, or the cost-effectiveness of a given treatment compared to another more expensive treatment (i.e., Is the increased benefit worth the additional cost?). Traditional cost-effectiveness analyses involve the computation of an incremental cost-effectiveness ratio (ICER) calculated in terms of dollars per quality-adjusted life years saved, as illustrated in Figure 2.
FIGURE 2. The Formula for a Traditional Cost-Effectiveness Analysis |
|
Models as decision support tools
Cost-benefit analyses are increasingly based on modeling rather than traditional cost-effectiveness analyses, because comprehensive trials that compare the cost benefit of various interventions are extraordinarily expensive and complex. To facilitate the decision-making process, a structured process using existing information resources and decision support tools (such as computer-based models) should be developed. In the managed care setting, such models would assist in evaluating the impact of chronic pain and its treatment and could provide projections of the flows of cost and benefit based on the best available evidence tailored to local circumstances.
Computer-based models can use actual data from a MCO along with published research data to estimate the incidence and cost of pain disorders and the cost and impact of varying treatment strategies. These models can accept multiple input variables such as number of patients, age distribution, current drug utilization, and other parameters. The impact of various interventions can then be tested by modifying input variables and following the flow of costs and benefits.
A computer-based health economics model developed by the Duke Center for Clinical Health Policy Research was used to evaluate the impact of cancer pain and treatment (a flowchart of this model is represented in Figure 3). This model uses data from the National Cancer Institute’s surveillance, epidemiology, and end results (SEER) database, which provides information about the incidence of various types of cancer, along with other research data on the effectiveness of cancer pain management (Hankey, Ries, & Edwards, 1999; Wingo et al., 1999).
FIGURE 3. Flowchart of a Health Economics Model for Evaluating the Impact of Cancer Pain and Treatment |
 |
Note. Presented at the American Pain Society 18th Annual Scientific Meeting, October 2223, 1999, Ft. Lauderdale, FL. |
Demographic data from a MCO can be entered into the program, and the incidence of cancer pain for that population can be estimated. Accordingly, the model calculates the impact of various treatment strategies as a percentage of patients with cancer pain who are relieved of pain. The model is currently being enhanced to allow the user to trace relative costs of various pain treatment strategies.
Computer models are powerful and convenient tools for evaluating the impact of pain disorders and the treatment of pain. However, before using these types of decision-support tools, a global analysis of the evidence based on a comprehensive review of established databases should be conducted. From this tableau of information, the flow of costs and benefits can be determined and projected to a specific environment.
Panel discussion
The managed care participants debated several issues related to models as decision-support tools. They suggested that the effects of improved pain management on productivity in the workplace could be included in the models. Modeling could be used as a sales technique to demonstrate potential benefits to employers. Some participants observed that an intervention based on good cost-benefit analysis could yield “paper savings” without real cost savings, because fixed costs remain the same. However, the participants agreed that even cost-neutral interventions could improve patient and provider satisfaction. They concluded that plan members and physicians could spur the demand for improved pain management. All types of HMOs would be amenable to modeling; however, the fact that the data of a staff or group-model HMO is much more easily accessible to analysis than other types of HMOs makes it easier to measure change.
Cost Effectiveness of Chronic Pain Programs
Speaker: Bill McCarberg, MD, Director, Chronic Pain Management Program, Coordinator, Pain Services, Kaiser Permanente, San Diego, CA
A needs analysis of physicians at Kaiser Permanente, a large group-model HMO, was conducted in San Diego in 1985 and found that 8.9% of respondents would refer patients with chronic pain to a specialty pain clinic. In response, a chronic pain program was developed that offered primarily cognitive-behavioral group treatment.
A cost-effectiveness study of the chronic pain program was conducted (McCarberg & Wolf, 1999). Patients referred to the pain program were randomly assigned to either a cognitive-behavioral group treatment or a control group that received only a pain management workbook. The cognitive-behavioral group treatment included cognitive therapy, relaxation training, self-hypnosis, biofeedback, drug education, problem solving, and goal setting. Outcome measures included subjective ratings of pain severity, suffering due to pain, interference of pain with life functioning, and pain-related outpatient utilization. Projected cost savings and overall satisfaction with the program were also measured. Among patients referred to the program, the most common complaint was low-back pain (49.4%), followed by neck pain and headache. About half had multiple sites of pain, and the mean level of pain was severe (7.5 on a 10-point scale). Primary care physicians were the main source of referrals.
Results of intervention
At a 6-month follow up, the chronic pain program produced a statistically significant decrease (p < 0.01) in pain severity, suffering, and interference with life functioning. Among patients assigned to the program’s cognitive behavioral group treatment, there were significant reductions in utilization of consultation services, routine and after-hours visits, and physical therapy. Patients in the workbook control group also utilized fewer outpatient services, but the decrease in utilization was smaller. Significant reductions in pain-related utilization were further observed at 1 year following intervention.
Cost estimates based on Medicare charge data indicated that the total projected cost of pain-related outpatient utilization was $785 per patient in the year following the group intervention compared with $2,155 in the year prior. This amounted to projected savings of $1,370 per patient per year. When the cost of the program and savings from the control group were subtracted, the total net annual projected savings from the program was $852 per patient, or a total of $562,000 per year for the San Diego service area. Projected total net annual savings in the southern California region would be approximately $3.5 million. Approximately 80% of patients and 75% of providers reported being satisfied or very satisfied with the group pain program. These findings are a useful starting point toward estimating potential cost effectiveness at other MCOs. However, a critical issue is the validity of generalizing these data to independent practice association-model HMOs, which provide health care for the majority of patients enrolled in MCOs in this country.
Panel discussion
The initial discussion focused on the future of cognitive-behavioral pain programs in MCOs. Some participants were concerned about the costs of such programs and potential difficulties in reimbursement. Others were impressed by the relatively small utilization of resources and the potential cost savings.
If the cost-benefit of a chronic pain program were demonstrated, participants believed that staff model HMOs would be more likely than IPA-model HMOs to embrace this type of program. In the latter situation, providers of interventional pain management procedures may show resistance to cognitive-behavioral pain programs because of potential lost revenue to such programs.
Participants recognized that the profile of chronic pain in managed care would increase if supported by reproducible evidence from utilization data indicating high total annual costs. They suggested that an investigator—possibly funded by an outside source—could visit individual plans and collect and aggregate data if individual plans did not have the available resources to conduct such an analysis. Acceptance of a chronic pain program would depend on the cost and integrity of the program and evidence that improved outcomes translate into lower total costs.
Disease Management Programs for Pain
Speaker: David Hubbard, MD, Medical Director, MyoPoint Medical Group, Associate Clinical Professor of Neurology, University of California, San Diego, CA
Chronic pain is characterized by overutilization of resources, including urgent care visits, procedures, and medications. Claims data from the Chicago region of a national preferred provider organization (PPO) highlight the high costs of managing patients with musculoskeletal pain. The cost of managing these patients accounted for nearly 40% of all claims, and more than 90% of pain-related claims, over a 24-month period (Hubbard, 1999). The overwhelming majority (86%) of musculoskeletal pain patients had claims of less than $500 per year; however, 108 of these patients (2.0%) accounted for claims of $10,000 or more, and 22 of these patients (0.4%) had claims exceeding $25,000. These patients represent an obvious target for interventions to control costs.
Several factors complicate the management of musculoskeletal pain, including a large number of pain patients; patients scattered among primary care providers, specialists, and ancillary providers; wide variation in clinical practice; and ineffective treatments. However, misdiagnosis is probably the largest single problem in managing musculoskeletal pain.
A disease management program developed by MyoPoint Medical Group, a group of pain specialists providing disease management programs to at-risk medical groups and insurance plans, emphasizes accurate diagnosis and systematic management of pain. The program first helps MCOs to target pain patients with high utilization of drugs and services and uses a panel of specially trained neurologists to improve diagnosis of these patients. The program reduces practice variation through clinical practice guidelines based on diagnosis. When initial attempts at pain management have been unsuccessful, patients are referred to a “center of excellence,” which is a facility identified to provide “best practice” for treatment of a specified pain condition.
Results of pain management strategy
A study of this pain management approach was conducted at Palomar Pomerado Medical Associates (PPMA), a primary care IPA in San Diego covering 25,000 capitated lives. All non-emergent back and neck pain patients were referred to the program prior to referral to specialists or ancillary services. The program’s services include diagnostic screening, utilization management, clinical treatment, and case management, as well as patient and provider education.
The PPMA program resulted in substantial reductions in utilization and costs, compared to historical patterns (Hubbard, 1999). On an annualized basis, there was a 94% reduction in surgeries, hospital days, and anesthesia; an 83% decrease in MRIs for patients with a musculoskeletal pain diagnosis; and a total elimination of nerve block injections. The annualized costs of pain treatment for all patients referred to the program was roughly $53,500, compared to historic costs of more than $1.1 million per year. Patient satisfaction ratings conducted 4 months after discharge from the program indicated that 75% of patients were very satisfied with the treatment they received.
Musculoskeletal pain management is an enormous problem for MCOs. However, high-risk patients and high utilizers of care can be identified, and pathogenesis-based diagnosis, psychophysiologically-based care, and patient and provider education can dramatically reduce utilization rates.
Panel discussion
The managed care meeting participants commented that the musculoskeletal pain program appeared to be quite labor intensive, requiring time and resources to reeducate primary care physicians (PCPs) to make correct initial diagnoses.
To ensure that PCPs recognize when to refer patients to a specialty pain program (the so-called red flag moment), it was suggested that a simple, easy-to-use algorithm would be valuable. Ongoing physician education also was stressed. Discussion centered on a multi-pronged approach that would (a) provide physicians with tools to assist in differential diagnosis at the initial visit, (b) offer feedback from a clinical metric that would allow providers to compare their own prescribing and referral conduct with national standards or local physician behavior, and (c) use provider representatives to act as envoys from the health plan to distribute information to physicians on pain management initiatives.
Summary and conclusions
The prevalence of chronic pain and the high cost of pain diagnosis and treatment is attracting increased attention within MCOs. The demand for quality pain care is likely to expand based upon regulatory initiatives, new accreditation standards, and legal precedents. Taken together, these factors suggest that the profile of chronic pain in the managed care setting will continue to increase.
Acknowledgments
The American Pain Society gratefully acknowledges the generous support of the following companies, which provided an unrestricted educational grant in support of the Managed Care Forum on Pain: Knoll Pharmaceutical Company, Merck & Co, Inc.; Purdue Pharma, LP; Roxane Laboratories, Inc.; Arrow International, Inc.; Endo Pharmaceuticals, Inc.; Glaxo Wellcome, Inc.; and Medtronic, Inc.
References
American Medical Association (1997, July). Many people still unaware of harmful effects of some pain medications. Science News Update. [Online]. Available: Science News Update (http://www.amaassn.org/
sci-pubs/sci-news/1997/snr0716.htm#kimmey)
Bjorkman, D.J. (1996). Nonsteroidal anti-inflammatory drug-induced gastrointestinal injury. American Journal of Medicine 101(1A), S25S32.
EPIC/MRA. (1997, October 13). Michigan chronic pain survey. Lansing, MI: Author.
Federation of State Medical Boards of the United States, Inc. (1998). Model guidelines for the use of controlled substances for the treatment of pain. [Online]. Available: http://www.fsmb.org.
Fishman, P., Von Korff, M., Lozano, P., & Hecht, J. (1997). Chronic care costs in managed care. Health Affairs, 16, 239247.
Hankey, B.F., Ries, L.A., & Edwards, B.K. (1999). Surveillance, epidemiology, and end results: A national resource. Cancer Epidemiology Biomarkers Prev, 8(12), 11171121.
Hubbard, D.H. (1999). Disease management programs for musculoskeletal pain. Paper presented at the Managed Care Forum on Pain, American Pain Society 18th Annual Scientific Meeting, October 2223, 1999, Ft. Lauderdale, FL.
Joint Commission on Accreditation of Healthcare Organizations. (1999, August 10). Pain assessment and management standards. [Online]. Available: http://www.jcaho.org/standard/pm_frm.html.
Liu, A.C., & Byrne, E. (1995). Cost of care for ambulatory patients with low back pain. Journal of Family Practice, 40, 449455.
Louis Harris and Associates. (1996). Pain and absenteeism report. Rochester, NY: Author.
McCarberg, B., & Wolf, J. (1999). Chronic pain management in a health maintenance organization. Journal of Pain, 15(1), 5057.
Polatin, P.B., & Mayer, T.G. (1996). Occupational disorders and the management of chronic pain. Orthopedic Clinics of North America, 27, 881890.
Roper Starch Worldwide Inc. (1999). Chronic pain in America: Roadblocks to relief. [Online]. Available: http://www.ampainsoc.org/whatsnew/toc_road.htm.
White, J., & Dubroff, M. (1999). Utilization of drugs and services related to pain management. Paper presented at the Managed Care Forum on Pain, American Pain Society 18th Annual Scientific Meeting, October 2223, 1999, Ft. Lauderdale, FL.
Wingo, P.A., Reis, L.A., Giovino, G.A., Miller, D.S., Rosenberg, H.M., Shopland, D.R., Thun, M.J., & Edwards, B.K (1999). Annual report to the nation on the status of cancer, 19731996. Journal of the National Cancer Institute, 91, 675690.
