SD29 - Study of the Effects on Medicaid Costs and Services of Chronic Pain and Pain Management


Executive Summary:
Chronic pain is a widespread problem in the population of the Commonwealth, leading to an increase in the utilization of health care services. Many studies have shown that pain management practices can shorten hospital stays, improve outcomes, and reduce physician visits. The Virginia General Assembly, through their study of pain management, determined that the use of pain management would be beneficial to the Commonwealth in reducing costs of medical assistance. Senate Joint Resolution No. 368 mandated that the Department of Medical Assistance Services (DMAS) conduct a study to determine how pain resulting from an illness or injury affects Medicaid costs and services (See Appendix A). The legislation asked DMAS to produce findings and recommendations on patterns of treatment, treatment that appeared to shorten the duration of pain, and best outcomes. The Department of Medical Assistance Services contracted with the Department of Health Evaluation Sciences at the University of Virginia to conduct the study.

A comprehensive review of the most recent fifteen years of academic literature on pain and pain management was completed using MEDLINE. It was determined through the literature search that many of the articles on chronic, non-cancer pain management focused on back pain and headache. This finding, in conjunction with the legislative language of SJR 368 that states “…80 percent of physician visits are for pain complaints, …23 million Americans have back pain, and …24 million Americans have debilitating headaches…” prompted the study team to focus their efforts on two broad diagnostic groups affecting Medicaid recipients: back pain and headache. In order to better understand the economic and health benefits of appropriate pain management, the study team combined information gleaned from Medicaid claims data with a survey of health care providers on pain and pain management. These quantitative and qualitative research strategies built on prior activities of the Joint Subcommittee to Study the Commonwealth’s Current Laws and Policies Related to Chronic, Acute, and Cancer Pain Management in assessing the economic benefits of appropriate management of pain on various benefits programs.

Review of the Literature

The literature review showed that health care professionals have varied opinions about pain, pain education, and treatment modalities. Specific problems mentioned in the literature include:

• inadequate knowledge of analgesic pharmacology and pain therapy

• poor pain assessment

• concern about regulatory oversight

• fear of patient addiction

• concern about the side effects of analgesics

• concern about development of tolerance to analgesics

Patients often lack knowledge about pain and available treatment options. Specific problems mentioned in the literature include:

• reluctance to report pain

• reluctance to take pain medications

• inadequate educational tools

Methodology and Analysis of Medicaid Claims Data on Chronic Pain Management

The analysis of Virginia Medicaid claims data was designed to focus on two broadly defined diagnostic categories: chronic low back pain and chronic headache. Medicaid claims data from fiscal years 1995 and 1996 was obtained from DMAS for analysis. Post-1996 data were not available to the study team because of normal delays in processing claims data. No analysis was done on enrollees of Medicaid HMOs because HMO encounter data is not included in the claims database.

The goal of the legislation was to determine the effects on Medicaid costs and services of pain resulting from an illness or injury. Data studied included hospital admissions and length of stay, physician visits, pharmacy usage, and rehabilitation therapies.

The goal of this effort was to assess Virginia Medicaid costs for two diagnoses (back pain and headache), examine use of prescription drugs (including opioids), and to evaluate variations in treatment across regions and physician specialty. For the purpose of analysis, an episode of illness was defined as an initial medical care visit with a principal diagnosis for one of the two conditions under study and all Medicaid costs of medical care (including pharmaceuticals) for these patients within one year of the initial encounter.

The diagnostic screens identified 18,935 Medicaid patients with at least one medical encounter with a principal diagnosis of back pain. The 12-month cost for these patients was approximately $25.9 million including $12.95 GF. Sixteen million dollars of this amount went towards prescription drug charges with the remainder being paid for provider, hospital inpatient, and hospital outpatient services (including rehabilitation services). The overall mean cost per patient was $1,379. There were substantial and significant variations across geographic regions in total non-drug costs and total costs. Additional analyses showed that patients who received strong opioids at their initial visit consumed more resources than patients who did not. It is unclear whether these additional costs reflected more effective care or more seriously ill patients.

A total of 19,751 Medicaid patients who had at least one medical encounter for headache were identified. Overall, headache patients cost less and used fewer resources than back pain patients. The 12-month cost for headache patients was approximately $18.3 million. Expenses for prescriptions accounted for more than $13.3 million. Costs for provider, hospital inpatient, and hospital outpatient care (including rehabilitation services) totaled $5 million. There were also substantial and significant variations across geographic regions in total non-drug costs and total costs. As seen in back pain patients who received strong opioids at their first medical encounter, headache patients who received strong opioids consumed significantly more resources than patients who did not, and again, it is impossible to determine the overall benefits, if any, of the additional costs.

It should be noted that the analyses reported are subject to limitations inherent in large-scale data sets. For example, total drug costs were most likely overestimated because it was impossible to differentiate which pharmaceuticals were prescribed to treat only the back pain or headache diagnoses followed in this study. However, despite this limitation, we did find that most patients with diagnoses of back pain or headache seemed to respond fairly promptly to treatment (because the utilization slowed or stopped), but that the care for a small group of patients is long and expensive.

Methodology and Analysis of the Provider Survey

A survey was mailed to 798 health care providers who are eligible to be Medicaid providers in the Commonwealth: 354 went to primary care providers and 444 went to pain management specialists. A total of 168 usable surveys were returned during the study period, resulting in a response rate of 21%. Restrictions on the use of mailing lists prohibited follow-up contacts normally used to raise response rates. However, this response rate is typical for other surveys on pain and pain management involving physicians (Cherkin 1995, Wolff 1991, Schwartz 1989).

Analysis was completed on the total study sample and the two subsamples of primary care providers and specialists. The two subsamples did not differ significantly in their answers to the questions, but there was substantial variation within each subsample. This supports previous studies and surveys of caregiver management of pain treatment.

The survey respondents were 82% male, nearly half were in private practice, and 75% were younger than 50 years of age. Primary care providers treated more patients overall, but a smaller percentage of their patients reported noncancer pain than in the pain specialist practices. Nearly half of the caregivers described learning about pain through clinical experience, rather than formalized training. Almost three-quarters of the study sample received continuing medical education on pain, mostly in the form of informal lectures or written materials. A majority of the respondents stated that a patient’s insurance status sometimes or always affects their ability to use certain treatment modalities, especially psychiatric services and interventional procedures. In general, there was much variation in the answers to questions dealing with the prescription of opioids, especially for patients with non-cancer pain.

The provider survey affirmed what was found in the literature concerning health care providers and pain management: there is little consensus about treatment modalities (as well as what type of practitioner is best to treat pain). While our survey produced a moderately low response rate, it did reflect the variation across treatment modalities and practice specialties found in other studies.

Recommendations

Recommendation: Decrease the cost of treatment for back pain and headache by closer management, on the part of DMAS and Medicaid providers, of patients with records of high utilization (more than 10 visits to a health care provider per year - See Chapter 4). DMAS is currently developing a program that would provide disease management services to Medicaid patients with chronic conditions. It is possible that chronic pain could be identified as a chronic disease to be targeted by this program.

Recommendation: Examine more closely the reasons for significant differences in utilization and cost of health services in the geographic regions of the Commonwealth.

Recommendation: More education for health care professionals about pain and available treatment options, including prescription of opioids, is needed. Educational materials based on clinical guidelines, such as guidelines developed by the American Society of Anesthesiologists, could provide standardized treatment options. In addition, SJR No. 366 (1997) called for the study of pain management curricula in the medical schools in the Commonwealth; recommendations from this study could be incorporated to supplement existing education programs for medical students.

Recommendation: More education for patients and their nonprofessional caregivers about pain and available treatment options, including prescription of opioids, is needed.

Recommendation: Existing laws and regulations governing prescription of opioids and provisions for reimbursement for pain should be examined for undue restrictions whether real or perceived.

Concluding Remarks

The Medicaid population has unique and distinguishing characteristics from the rest of the population. Medicaid recipients tend to be adult Caucasian females or children. In addition, Medicaid prescription drug formularies tend to be less restrictive than those of other insurance programs. Because of this, recommendations based on analysis done on Medicaid claims data may not be generalizable to the population at large.

However, certain patterns did emerge that might be applicable to other segments of the population. The costs of treatment for back pain and headache in the Medicaid population in Virginia are large, totaling approximately $41.4 million, including drug costs. This figure does not include indirect costs, such as loss of productivity or lost work days. With respect to utilization, while most patients with a diagnosis of back pain or headache seek medical care on 10 or fewer occasions per year, the small number of patients who have more than 10 encounters per year account for more than half of the total direct costs related to these diseases.

Studies are needed to determine whether the different perspectives and attitudes of caregivers are associated with differences in the costs and outcomes of care. Unfortunately, we could not do this with the available data.

While the providers surveyed for this study reported receiving education on pain and pain management, we were not able to assess the adequacy of this knowledge. The academic literature shows that medical societies and many providers are concerned that health care providers are not trained in how to deal with pain, especially with chronic pain. In addition, health care providers are concerned about possible patient addiction, tolerance, and side effects to certain medications, and they are also worried about strict regulations and laws surrounding the prescription of controlled substances such as narcotics. More education for health care professionals and patients about pain and available treatment options, including prescription of opioids, is needed. Clinical guidelines may help to decrease variability in provider knowledge and attitudes about pain and pain management and may help to decrease variation in treatment modalities.

Laws and regulations governing prescription of opioids and provisions for reimbursement for pain impact patterns of treatment. The low priority given to chronic pain management (as opposed to treating acute episodes), the inadequate or inappropriate provisions for reimbursement for treatment, and restrictive regulation of controlled substances may obstruct the appropriate treatment of pain. Possible reimbursement for complementary treatment options (such as chiropracty or massage therapy), based on the experience of other states, should also be examined.