Chronic non‐malignant pain patients and health economic consequences

At a time when resources for health care are limited there is a growing concern not only about the clinical effectiveness of treatment but also the cost-effectiveness. There is no question that chronic pain syndromes are prevalent and costly. The expenditures for chronic pain, in addition to actual health care, include welfare and disability payments, lost tax revenue, and lost productivity. Despite the explosion in knowledge of the anatomy, physiology, and biochemistry of pain and the proliferation of treatments available, pain continues to be a problem for a significant number of people with diverse chronic pain conditions. Rehabilitation oriented programs provided at multidisciplinary pain centers (MPCs) were developed in the 1960s to treat patients with recalcitrant pain problems. Although initially welcomed, many third-party payers (e.g., insurance companies, governmental agencies) have become less enamored and have challenged the clinical benefits and more notably the cost-effectiveness of these programs. To those who must provide payment, “multidisciplinary” has become a synonym for expensive. The assumption seems to be that involvement of multiple health care disciplines is labor intensive and will inevitably be less efficient and, hence, more costly than alternative approaches. Moreover, these third-party payers question the adequacy of the outcome criteria used to support the effectiveness of MPCs. That is, the majority of the outcome studies have reported on the benefits of treatment on patients' self-reports of pain reduction, improvement of mood, and changes in beliefs and attitudes. In their view, these outcomes may be desirable but they are less important than functional outcomes. Their primary concern is whether patients treated at MPCs and who report less pain demonstrate concomitant reductions in health care consumption, return to gainful employment, and require less disability compensation. In one sense the skeptics of MPCs are correct. Chronic pain patients do consume a substantial amount of the finite health resources. Moreover, early studies of outcomes at MPCs neglected to pay attention to important economic outcomes. What skeptics fail to realize, however, is that MPCs are not alone in inadequately assessing the economic impact of chronic pain treatments. When other expensive treatments for chronic pain, such as surgeries, pharmacological interventions, neuroaugmentative procedures, and physical modalities have been studied, researchers have also typically relied on patient self-reports of pain and satisfaction with treatment as outcome measures instead of cost-effectiveness. Third-party payers seem to be asking the wrong question when they challenge MPCs as to their cost-effectiveness. The question is not whether MPCs are cost-effective and clinically effective but compared to what alternatives. As Flor et al. (1992) observed in their meta-analysis of 65 MPC treatment outcome studies, the mean duration of pain for patients treated at these facilities is over 7 years, with the majority already consuming significant analgesic medication, and having an average of 1.7 surgical procedures for pain prior to treatment at the MPC. Crook et al. (1986) compared people with chronic pain in the community and those referred to MPCs and noted that the patients treated at MPC reported constant pain, high levels of emotional distress, work-related injuries, significantly lower levels of education, higher levels of health care utilization, opioid use, functional impairment, and negative attitudes about the future. Coupled with the long duration of pain noted earlier the fact is that most patients seeking treatment at MPCs have extremely recalcitrant pain problems accompanied by many other functional and emotional difficulties. Taken together, it is impressive that MPCs are able to produce any positive benefits. The primary aim of the Thomsen et al. study is to address concerns regarding the economic consequences of treatment at an MPC. In this study both the clinical and cost-effectiveness of a set of consecutive patients treated at an MPC in Denmark were evaluated. The authors made a concerted effort to make use of social records to obtain data on health care costs along with disability and welfare costs for a period of 6 months prior to entry on a 4-month waiting list, at the time of treatment entry, during the 10 months of treatment, and at a 9-month follow up after termination. The results of the study confirm that rehabilitation program significantly reduce pain severity and improved patients' ratings of their general health following treatment and at follow-up. They report, however, that there was no significant reduction in costs for health care. At first the failure to reduce significantly costs might be discouraging, however, careful examination of the results leads to a more positive conclusion. I will analyze the cost data relying on the information provided in Tables 5 and 6 of the Thomsen et al. paper. In the 6 months prior to treatment the average patient spent 292 EUR/month for health care. I used this period rather than the 4-month waiting list period to calculate saving because it is possible that patients on a waiting list for treatment at an MPC may have been offered somewhat less treatment. The expenditures for health care in the period prior to placement on the waiting list may be more representative of actual monthly health care costs. During the 10-month treatment at the MPC, including the estimated costs of the MPC, the patients averaged 323 EUR/month. In the 9 months following treatment the patients averaged 202 EUR/month (a difference of 90 EUR/month less than prior to treatment). This reflects a 31.1% reduction in health care consumption following treatment! The authors note that the monthly expenditures for health care were consistent in the 4-month base line period compared to the 6 months prior to treatment and suggest that this indicates stability in health care consumption. The authors do not report on the age of the patients included in their study but if we accepted the data aggregated across 65 studies by Flor et al. (1992) the mean age would be 44 years old. Next we can calculate the savings that would be accrued over a time period. I will use 19 years for illustrative purposes, as this would allow us to follow pain sufferers until they were age 65 on average. Based on the figures reported by Thomsen et al. the average anticipated annual savings in health care would be 1080 EUR/patient. If we multiply the annual rate by 19 years, the expected savings in health care expenditures for each patient would be 20,520 EUR. In this study, the number of patients available at the follow-up was 122. Using the 19-year figure above, the estimated health care savings for patients treated at the Danish MPC would be 2,503,440 EUR. Of course, we have to subtract the cost of the treatment for the 131 patients—(110 EUR/patient) 14,410/EUR. Even after subtracting the costs of the treatment at the MPC, the health care savings would exceed 2.4 million/EUR. Furthermore, Thomsen et al. did find significant reductions in social transfers (welfare benefits, sickness benefit, and pensions). In the 6 months baseline, the average patient received 459 EUR/month, whereas in the 9-month follow-up period patients received 172 EUR, a reduction of 63%. Once again we can consider the reductions in social transfer for 19 years. Working through the arithmetic, the total reduction would amount to 65,436 EUR/patient treated. Using the patients available at follow-up (122), the figure equals 7,983,192 EUR. Combining the savings in health care and social transfers, indicates that the average savings/patient treated at an MPC over a period of 19 years would be equivalent to 85,956 EUR and for the 122 and subtracting the cost of treatment for 131 patients the calculation would lead to substantial savings in excess of 10 million/EUR. There are some important cautions about generalizing from the data that Thomsen et al. report. Most notably, there was a high refusal rate among those invited to participate and a significant loss of subjects to follow up. The results would be less impressive, although significant cost saving would likely still be accrued, if the author had used intent-to-treat analysis. A significant limitation, which the authors note, is the lack of a control group to compare to those who were treated in the MPC. The difficult problem is identifying an adequate control group. Perhaps the best that might be accomplished would be to compare patients treated at an MPC to a group receiving “standard care”. Other problems include lack of randomization and blinding, both of which are difficult in rehabilitation settings. A final point to revisit is the issue that I raised earlier about asking the wrong question. This study was designed to determine health economic costs of treatment at an MPC. I suggest that a more appropriate question would be how cost-effective is treatment at an MPC compared to existing alternatives such as surgery, pharmacotherapy, neuroaugmentative interventions, somatic modalities, or complementary methods? Other studies suggest that, in the long run, MPCs are indeed more cost-effective than alternative treatments for patients with recalcitrant pain (Flor et al., 1992). Of course, there are problems with extrapolations of the type that I calculated above. The data available, however, do paint a very positive picture regarding the potential health economic benefits of treatment of a very difficult population of patients at an MPC. I commend the authors of this study for addressing important health economic issues. There is a tremendous need for outcome studies that address both clinical effectiveness and cost-effectiveness of countless treatments for many pain syndromes. The use of social and health registry data is a particularly useful strategy to acquire important outcome data that are potentially less biased than self-reports. I do not mean to disparage the importance of patient perceptions, satisfaction, and quality of life but rather acknowledge that health economic data are also critical in evaluating treatments.