OUP user menu

Reduction in Medication Costs for Patients with Chronic Nonmalignant Pain Completing a Pain Rehabilitation Program: A Prospective Analysis of Admission, Discharge, and 6-Month Follow-Up Medication Costs

Julie L. Cunningham PharmD, Jeffrey D. Rome MD, Jennifer L. Kerkvliet MA, Cynthia O. Townsend PhD
DOI: http://dx.doi.org/10.1111/j.1526-4637.2009.00582.x 787-796 First published online: 1 July 2009


Objective. Chronic nonmalignant pain (CNMP) is both a prevalent and a costly health problem in our society. Pain rehabilitation programs have been shown to provide cost-effective treatment. A treatment goal for some rehabilitation programs is reduction in the use of pain-related medication. Medication costs savings from pain rehabilitation programs have not been analyzed in previous studies.

Design. This prospective cohort study of 186 patients with CNMP addresses the costs of medications at admission to a 3-week outpatient pain rehabilitation program, at discharge, and at 6-month follow-up. Medication use was determined through a detailed pharmacist interview with patients at admission and discharge. Patients were sent questionnaires 6 months after program completion, which obtained current medication information.

Results. Statistically significant medication cost savings were seen for program completers at discharge and at 6-month follow-up (P < 0.05). The mean (standard deviation) daily prescription medication cost reduction from admission to discharge was $9.31 ($12.70) using the average wholesale price of medications. From the original study cohort, 121 patients completed the 6-month follow-up survey. The mean daily prescription medication cost savings from admission to 6-month follow-up was $6.68 ($14.40).

Conclusion. Patients benefited from significant medication cost savings at the completion of the 3-week outpatient pain rehabilitation program and maintained significant savings after 6 months. This study adds to the current literature on the economic value of comprehensive pain rehabilitation programs.

  • Chronic Pain
  • Medication
  • Pain Rehabilitation
  • Cost
  • Pharmacist


The need for adequate recognition and treatment of pain has been highly visible in recent years. In 2000, the United States government launched the Decade of Pain Control and Research, followed in 2001 by pain management standards from the Joint Commission on Accreditation of Healthcare Organizations, which state that every patient has a right to have his or her pain assessed and treated [1]. These initiatives have clearly focused the attention of health care providers on the assessment and treatment of pain. Chronic nonmalignant pain (CNMP) remains a prevalent and costly problem in society, with the annual estimated health care costs of chronic pain in the United States exceeding $70 billion [2]. Research in specific chronic pain diagnoses such as fibromyalgia has found costs to be more than twice that of control patient groups [3,4].

Efforts to adequately treat pain have led to an increase in the use of analgesic medications. A recently published study utilizing the National Ambulatory Medical Care Survey found an increase in opioid prescribing by primary care physicians during the 1992–2001 study period. The prevalence of primary care visits where an opioid was prescribed increased from a low of 41 per 1,000 visits in 1992–1993 to 59 per 1,000 in 2000–2001 [5]. Despite the increased use of opioids, research continues to identify the inadequate treatment of pain in the community [6].

Pharmacotherapy is one of many modalities of treatment available for chronic pain. Breivik et al. found that two-thirds of patients with CNMP reported taking prescription pain medications [7]. Medications alone rarely eliminate CNMP, but the result may be a reduction in pain severity. Furthermore, these patients commonly have comorbid medical conditions that contribute to the use of polypharmacy. Patients with chronic pain are regularly prescribed agents to manage the side effects of treatment itself. Examples include laxatives, antacids, antihypertensives, diuretics, and psychostimulants, among others. Polypharmacy in the treatment of patients with CNMP can lead to dissatisfaction on the part of both providers and patients. It also increases the risk of adverse effects, drug interactions, and is associated with increased cost. Additionally, pharmacotherapy alone does not specifically address the decline in social and occupational functioning associated with CNMP. Efficacy studies of analgesic agents in patients with CNMP have generally not provided long-term follow-up, and therefore the ability to extrapolate to long-term outcomes is limited [8–11].

By 2015, total prescription drug costs are projected to more than double to $446 billion from $219 billion (estimated) in 2006. Drug costs are expected to increase from 10% in 2004 to 11% in 2015, as a percentage of overall health care expenditures [12]. According to data published by NDCHealth and posted on the RxList internet site, the medication most frequently prescribed in 2005 was the analgesic combination hydrocodone/acetaminophen, as has been the case since 2000 [13]. The next most frequently prescribed medication in 2005 was atorvastatin, which had nearly 40% fewer prescriptions written. Additionally, both celecoxib and fentanyl transdermal patches were in the top 25 in 2004 drug sales ($3,302,000 and $2,083,000, respectively) [7].

Regarding chronic pain, there is limited information available on the cost of prescription medications for the symptomatic treatment of this condition [14–17].

Pain rehabilitation programs offer treatment for patients who report high levels of pain severity, emotional distress, and compromised functioning. Treatment goals in pain rehabilitation programs address functional restoration and commonly incorporate a reduction in the use of pain-related medications. Substantial evidence supports the clinical benefit and cost-effectiveness of rehabilitative treatment for persons with CNMP [18,19]. Investigation of patients with CNMP in pain rehabilitation programs has shown the clinical benefit of improved physical and emotional functioning while discontinuing opioid analgesics, but has not addressed the pharmaceutical cost benefit of this approach [20,21]. Longitudinal follow-up demonstrated that the improvement was maintained 6 months following treatment [22]. Other studies have found a correlation between the acceptance of chronic pain and a decrease in medication consumption [23]. Tollison et al. reported sustained discontinuation of opioids in more than 65% of patients 1 year after rehabilitation [21].

The purpose of this study was to examine the direct cost savings associated with a reduction in medication use by patients with CNMP following the completion of a 3-week outpatient pain rehabilitation program and to determine if cost savings were sustained. The costs of all oral and transdermal prescription medications were considered in this study (not merely analgesics and other types of medications used specifically for CNMP).

Materials and Methods

Study Design

This prospective, nonrandomized, pre-post longitudinal study compared patient daily prescription medication costs at admission, discharge, and 6 months following a pain rehabilitation program. Patients were also categorized by primary chronic pain complaint to evaluate for differences between various pain disorders. Informed consent was obtained from each patient at admission. This study was approved by the Mayo Foundation Institutional Review Board.

Pain Rehabilitation Treatment Program

The Mayo Comprehensive Pain Rehabilitation Center (PRC) is an intensive 3-week outpatient multidisciplinary rehabilitation program designed to assist patients with CNMP to improve adaptation to pain and related symptoms. Patients considered eligible for admission to this program have persistent pain and substantial functional disability. Rehabilitative treatment in this setting is based on a cognitive–behavioral model incorporating physical reconditioning; occupational therapy for job, leisure, and home activities; relaxation training and biofeedback; stress management, pain management training (e.g., activity moderation, elimination of pain behaviors); chemical health education; and family participation. The primary goal of the program is functional restoration.

During the course of the program, patients receive education on the use of medications for CNMP, including information on efficacy, side effects, drug interactions, and withdrawal effects. Discontinuation of all analgesic medications (including opioids) and certain adjuvant medications used for CNMP (e.g., benzodiazepines, muscle relaxants) is one of the treatment goals. Withdrawal from opioids is gradual, structured on a time-contingent basis, and monitored closely, with the end point of discontinuation several days before program completion. During the process of opioid withdrawal, other analgesics or adjuvant medications for pain control are not utilized. Some patients however, are prescribed clonidine for a limited period of time (typically 7 to 14 days) to assist with the physiologic signs and symptoms of withdrawal.

Measures of Treatment Outcomes

At admission, discharge, and 6-month post-treatment, study patients completed questionnaires measuring pain severity and interference with daily activities, level of function, life control, depression, and pain catastrophizing. The Multidimensional Pain Inventory (MPI) is a widely used self-report assessment of psychosocial functioning in patients with chronic pain [24]. The inventory contains 52 items subdivided into 13 scales assessing the impact of pain on patients' lives. In the current study, the subscales utilized to assess the effectiveness of treatment included measures of pain severity, personal control in daily life functioning, and general activity level. The MPI has been found to be psychometrically adequate [25,26].

The SF-36 Health Survey (SF-36) [27] measures eight constituents of quality of life. Questionnaire items are scored on a 2-point, 3-point, 5-point, or 6-point Likert system with higher scores signifying more favorable health status. This study utilized the physical functioning and general health perception subscales of the overall measure. The reliability and validity of the SF-36 is well documented [27,28].

The Center for Epidemiologic Studies Depression Scale (CES-D) [29] is a self-report measure of depressive symptomology. Scores range from 0 to 60, with higher scores indicating increased depressive symptomology. A cutoff score of 16 has been suggested as an indicator of minor depressive symptoms [30] and a score of 27 has been used to identify patients with major depression [31]. The measure has been found to have satisfactory psychometric properties [32].

Finally, to determine the level of pain-related cognitions and emotions the catastrophizing subscale of the Coping Strategies Questionnaire (CSQ) was utilized [33]. The six items of this subscale are scored on a 6-point Likert scale, ranging from 0 (“not at all”) to 6 (“always”). High levels of catastrophizing have been associated with greater disability, a heightened pain experience, and greater utilization of health care services. [34]. The psychometric properties of this subscale have been found acceptable [35].


From May 31, 2005 to February 1, 2006, 236 patients were consecutively admitted to the PRC program in Rochester, Minnesota, of which 223 patients were eligible for study enrollment. Excluded patients included nine who were less than 18 years of age, two who did not have a CNMP diagnosis, two who were on intrathecal pain pumps. Of the 223 eligible patients, 15 patients refused consent for study participation. Two hundred and eight patients were then enrolled in the study, and of these 22 patients (10.6%) did not complete the program. The mean (standard deviation [SD]) length of stay of the 22 noncompleters was 7.0 (3.6) days in the program. The most common reasons for noncompletion were acute medical illness or discrepant expectations of the program, each reported by 27.3% (N = 6) of the noncompleters. Analysis of variance (ANOVA) and chi-square analyses showed no differences between completers and non-completers regarding gender, age, education, marital status, state residency, pain duration, or admission medication costs.


Demographic data, clinical information, and a medication profile were obtained from patients' medical records and a pharmacist interview on admission and discharge from the PRC. Additionally, patients were sent questionnaires 6 months after program completion, which obtained current medication information.

A detailed pharmacist interview with the patient and review of medication bottles were completed at admission and discharge to determine the specific dosage form of each medication. Costs were calculated for prescription oral dosage forms and transdermal patches based on the number used per day. Medications that were taken “as needed” were calculated based on the average number used per week, as reported by the patient. For example, if a patient reported taking a medication two times per week, the cost of the number of pill(s) was multiplied by 2 and divided by 7 for the average daily cost. The costs of over-the-counter (OTC) products and of topical creams/ointments, inhalation, ophthalmic, parenteral and intrathecal prescription medications were not tabulated in this study.


Prescription medication costs were based on the Average Wholesale Price (AWP), utilizing the lowest price where various generic alternatives were available. All medication prices were determined at the end of the study to avoid the confounding factor of price changes during the course of the study. Cost savings are calculated as daily costs and extrapolated to monthly and annual medication costs. Monthly costs were based on a 30-day month and were multiplied by 12 for annual costs.

Statistical Analyses

One-way ANOVAs were used to evaluate medication cost reductions associated with demographic and clinical variables. To assess for differences between patients completing the 6-month follow-up survey and those who did not, one-way ANOVAs and chi-square analyses were used. Paired t-test analyses were used to determine differences between medication costs and measures of treatment outcome at admission, discharge, and follow-up. P ≤ 0.05 was used to denote statistical significance. ANOVA and chi-square analyses were used to evaluate differences between the PRC program completers and noncompleters regarding demographic and clinical variables. Data analysis was performed using SPSS software, version 14.0 (SPSS, Inc., Chicago, IL, USA).


Study Population

The study cohort consisted of 186 patients who completed the 3-week outpatient program. Demographic and clinical data for the study cohort are provided in Table 1. At admission, 90 patients (48.4%) were taking an opioid medication on a daily basis, with an additional 11 patients taking opioid medications at least one time per week. The mean (SD) daily oral morphine equivalent dose of daily opioid users was 99.6 mg (133.6 mg) (range, 2.5–656 mg), with 29 (28.7%) on a morphine equivalent dosage of 100 mg or greater per day.

View this table:
Table 1

Characteristics of study patients admitted to the Mayo Comprehensive Pain Rehabilitation Center from May 2005 to February 2006 (N = 186)

CharacteristicMean (SD)/Percentage
Mean age (years) (SD)46.4 (14.1)
Gender (%)
Ethnicity (%)
 Other 4.3
Marital status (%)
Mean education (years) (SD)14.7 (3.0)
Residence (%)
 Minnesota residents40.9
 Non-Minnesota residents59.1
Primary pain diagnosis (%)
 Low back24.7
 Chronic headaches11.3
Mean pain duration (years) (SD) 9.5 (10.0)*
Daily opioid use at admission (%)48.4
Currently working for wages (%)28
  • * N = 185.

  • SD = standard deviation.

Medication Cost Analysis

The mean (SD) daily medication cost on admission was $23.13 ($19.81) for the study cohort of 186 patients who completed the 3-week program. The mean (SD) daily medication cost at discharge was $13.82 ($14.06). The mean daily medication cost reduction from admission to discharge was $9.31 ($12.70) (Table 2). This represents a 40.3% reduction in daily medication expense. No differences were found in admission medication costs between those who completed the program and those who did not (P > 0.05) (Table 3).

View this table:
Table 2

Total daily medication cost savings after a rehabilitation program

NMean ($)Standard Deviation ($)Minimum ($)Maximum ($)
Admission medication total daily cost18623.1319.81 0.00157.40
Dismissal medication total daily cost18613.8214.06 0.00120.35
Total daily medication cost savings from admission to discharge186 9.3112.70–4.48 64.43
View this table:
Table 3

Comparison of total daily admission medication cost between program completers and noncompleters

NMean ($)Standard Deviation ($)Minimum ($)Maximum ($)
Noncompleters 2227.5419.544.84 75.54

Seventy-eight percent (78%) of patients in the study had medication cost savings at program discharge (Figure 1). There were no differences in cost savings by site of pain, age, years of education, duration of pain, or marital status. However, significant differences were found in cost savings with regard to gender. Males experienced a greater cost savings, averaging $13.95 ($16.35) per day compared with females saving $8.23 ($11.50) per day (P < 0.05). State of residence also reflected significant differences, with non-Minnesota residents having greater admission medication costs, $26.39 ($22.08) and greater cost savings $11.48 ($13.96) than Minnesota residents, with admission medication costs of $18.40 ($14.87) and costs savings of $6.20 ($9.92) (P < 0.05).

Figure 1

Daily medication cost changes at discharge from a multidisciplinary pain program.

Statistical analyses showed significant differences in medication costs from admission to discharge for the following medication classes: opioids (P < 0.001), anticonvulsants (P < 0.001), antimigraine medications (P < 0.05), benzodiazepines (P < 0.001), stimulants (P < 0.05), muscle relaxants (P < 0.001), nonsteroidal anti-inflammatory drugs (NSAIDs) (P < 0.001), and sleep aids (P < 0.01) (Table 4). There was no significant difference in cost from admission to discharge for patients taking antidepressants.

View this table:
Table 4

Medication total daily cost by drug class upon admission and dismissal from pain rehabilitation program*

Medication ClassesAdmission (N = 186)Dismissal (N = 186)
No.Total Daily Cost ($)% of Total Admission Medication CostNo.Total Daily Cost ($)P Value of Cost Comparison
Muscle relaxants3394.252.238.00<0.001
Sleep aids70119.882.87164.59<0.01
  • * Sum of numbers do not equal the study cohort total due to patients taking multiple classes of medications. Sum of percentages do not equal 100 due to additional prescription medication use.

  • NSAIDs = nonsteroidal anti-inflammatory drugs; NS = not statistically different.

The analyses of admission and dismissal subscale scores from the MPI, SF-36, CES-D, and CSQ are found in Table 5. The mean differences showed significant reductions in pain severity (P < 0.001), depressive symptoms (P < 0.001), and pain catastrophizing (P < 0.001). Significant increases were seen in life control (P < 0.001), general activity (P < 0.001), health perception (P < 0.001), and physical functioning (P < 0.001).

View this table:
Table 5

Mean scores and outcome variables at admission and dismissal (N = 186)

MeasureAdmissionDismissalNP value
Pain severity (MPI)47.51 9.7636.5612.69183<0.001
Perceived life control (MPI)45.68 7.7157.13 6.88182<0.001
General activity level (MPI)52.17 8.9657.44 9.51183<0.001
Health perception (SF-36)42.4622.8559.7122.21178<0.001
Physical functioning (SF-36)42.0023.2067.3023.09178<0.001
Pain catastrophizing (CSQ)11.66 6.12 4.60 4.94183<0.001
  • SD = standard deviation; MPI = Multidimensional Pain Inventory (T scores); SF-36 = SF-36 Health Survey (T scores); CES-D = Center for Epidemiological Studies—Depression scale; CSQ = Coping Strategies Questionnaire.

Longitudinal Analysis

Longitudinal follow-up was conducted at 6-month post-treatment for the study cohort of 186 patients who completed the rehabilitation program. A total of 121 patients (65.1%) of the study population (N = 186) responded. There were no differences between patients who completed the longitudinal follow-up and those that did not with regard to gender, ethnicity, marital status, pain duration, admission medication cost, or dismissal medication cost. Age and education were found to be significantly different (P < 0.05), with older patients and those with more years of education more likely to respond to the follow-up questionnaire.

Table 6 provides medication costs at follow-up. The mean (SD) daily medication costs for the follow-up study population (N = 121) were: $22.30 ($17.63) at admission; $14.36 ($12.75) at discharge; and $15.62 ($15.00) at follow-up. Paired t-tests show there were significant differences from admission to both subsequent time points (P < 0.001). Overall, from admission to 6-month follow-up, patients completing the questionnaire showed a mean daily medication cost savings of $6.68 ($14.40), a reduction of 30.0%.

View this table:
Table 6

Cost information for patients completing 6-month follow-up questionnaire

NMean ($)Standard Deviation ($)Minimum ($)Maximum ($)
Admission medication total daily cost12122.3017.63 0.0079.61
Dismissal medication total daily cost12114.3612.75 0.0071.75
Total daily medication cost savings from admission to dismissal121 7.9411.58−4.4864.43
Six-month medication total daily cost12115.6215.00 0.0067.73
Cost savings from admission to 6-month follow-up121 6.6814.40−19.6867.51

OTC medication cost analyses were not addressed in this study. However, patients were evaluated for the use of OTC medications at admission, discharge, and 6-month follow-up to assess whether OTC medications were substituted for prescription medications. At admission, the mean number of oral OTC medications per patient was 2.46 (2.43) (range of 0 to 13 different OTC medications per day). At discharge, this number had dropped to a mean of 1.61 (1.62) OTC medications daily (range from 0 to 7 per day). Overall, 95.7% of patients had a decrease or no change in their use of OTC medications from admission to discharge. At 6-month follow-up (N = 121), the mean number of OTC medications taken was 1.48 (1.87) (range from 0 to 9 different medications). From admission to 6-month follow-up, 82.6% of patients reported having a decrease or no change in their use of OTC medications.

A more detailed review of OTC analgesic use revealed patients were taking a mean of 1.63 (3.50) analgesic OTC tablets per day at admission (range 0 to 24 tablets per day). At discharge, this had decreased to a mean of 0.27 (1.08) analgesic OTC tablets (range 0 to 8 tablets per day). All but one patient decreased or had no change in their use of OTC analgesic medication from admission to discharge. At 6-month follow-up, the mean OTC analgesic tablets used per day was 0.33 (0.96) (range of 0 to 6 tablets per day). Eighty-nine percent (89.3%) of patients had a decrease or no change in OTC analgesic use from admission to 6-month follow-up.

Analyses of the admission and 6-month post-treatment subscale scores from the MPI, SF-36, CES-D, and CSQ for 121 patients completing the longitudinal follow-up questionnaire are shown in Table 7. The mean differences showed sustained significant reductions in pain severity (P < 0.001), depressive symptoms (P < 0.001), and pain catastrophizing (P < 0.001). Significant increases remained in life control (P < 0.001), general activity (P < 0.001), health perception (P < 0.001), and physical functioning (P < 0.001).

View this table:
Table 7

Mean scores and outcome variables at admission and 6-month post-treatment for program completers (N = 121)

MeasureAdmissionSix-Month Post-TreatmentNP value
Pain severity (MPI)46.45 9.6835.0512.43121<0.001
Perceived life control (MPI)46.17 7.7457.90 6.26120<0.001
General activity level (MPI)52.28 9.0957.28 9.18121<0.001
Health perception (SF-36)42.6722.2560.8221.89119<0.001
Physical functioning (SF-36)42.5723.4369.3721.24119<0.001
CES-D27.0412.5313.51 9.52121<0.001
Pain catastrophizing (CSQ)10.86 6.11 3.74 4.20121<0.001
  • SD = standard deviation; MPI = Multidimensional Pain Inventory (T scores); SF-36 = SF-36 Health Survey (T scores); CES-D = Center for Epidemiological Studies—Depression scale; CSQ = Coping Strategies Questionnaire.


In the United States, health care expenditures per capita in 2004 were estimated at $6,280 [36]. Prescription medications account for approximately 10% of this amount [6]. Although medication costs comprise a relatively small portion of the health care dollar, for medication insurance programs and to the individual patient, this expense can be quite significant. Research has shown that physicians commonly are not familiar with direct pharmacy costs [37,38].

Pain rehabilitation programs have been shown to reduce health care costs for patients with CNMP who are known to be high utilizers of medical services. Several studies have reported reductions in pain-related clinic visits, surgical interventions, hospitalizations, emergency room visits, decreased disability claims, and overall medical costs among the benefits of rehabilitative intervention [18,39–42]. Pain rehabilitation programs vary considerably in duration, service intensity (hours per day; days per week), health care personnel providing treatment, and setting (inpatient, day-hospital, outpatient). Thus, the cost of rehabilitative treatment encompasses a wide range. The cost for the Mayo Comprehensive PRC, an intensive, outpatient interdisciplinary 3-week program is approximately $20,000, which is comparable with other similar programs in the nation.

This present study provides additional information on pharmaceutical expenses associated with CNMP and cost savings following a comprehensive pain program. In this study cohort, the mean daily cost of oral and transdermal prescription medications was $23.13 per day at the time of admission to the pain rehabilitation program. This translates to $693.90 per month (30 days) and $8,326.80 per year using the AWP, a national average of list prices charged by wholesalers to pharmacies, to determine medication costs.

We found a significant reduction in expenditures for prescription medications from admission to completion of a 3-week pain rehabilitation program. The mean medication cost savings were $9.31 per patient per day. Of particular importance, this study also demonstrates that the pharmaceutical cost benefit of pain rehabilitation is sustained following discharge. From the time of admission to 6-month follow-up of the study cohort who completed the post-treatment questionnaire, the mean medication cost savings was $6.68 per patient per day. At this rate, the estimated annual medication savings would be $2,404.80. In addition to other potential health care utilization savings attributable to pain rehabilitation, medication savings alone would cover program costs in just over 8 years, were these medication changes to be sustained.

Saving money by reducing or eliminating medications in patients with CNMP would not be appropriate or relevant if this were accompanied by worsening pain, functional deterioration, or increased emotional distress. However, this study finds rehabilitative treatment incorporating analgesic withdrawal results in a variety of beneficial outcomes including a reduction in pain severity, improved physical and emotional functioning, and enhanced health perception. Importantly, these benefits appear to be sustained over time. Similar treatment outcomes have been collaborated by other published results of pain rehabilitation programs [20,22,43].

Gender appeared to have a significant effect on the pharmaceutical cost benefit of rehabilitation. Although no gender differences in medication costs were seen on admission, males demonstrated greater savings (per patient per day) at the completion of the program ($13.95 compared with $8.23 for females [P < 0.05]). There were no differences between the number of males and females on opioid medications at admission (54.3% vs 47.7%, respectively); however, male patients were on a significantly higher mean (SD) morphine equivalent dosage, 191.4 (209.9) mg vs 75.6 (91.6) mg (P < 0.05). State of residence also affected cost savings, with non-Minnesota residents deriving greater benefit (mean daily cost savings of $11.48 [$13.96] than Minnesota residents $6.20 [$9.92]). Sixty-five percent of non-Minnesota residents compared with 40.8% of Minnesota residents were taking opioids on admission (P = 0.001). Thus, the greater cost savings by out-of-state residents and male participants appear attributable, at least in part, to the discontinuation of opioid analgesics.

The analysis of various classes of medications revealed statistically significant cost savings from admission to discharge (Table 4). These classes included opioids, anticonvulsants, antimigraine medications, benzodiazepines, stimulants, muscle relaxants, NSAIDs, and sleep aids. No significant difference was noted for antidepressants. These findings appear to represent the discontinuation of expensive pharmaceuticals such as controlled-release oral opioids, transdermal fentanyl, and some of the newer anticonvulsants. The lack of significant cost savings for antidepressants is a reflection of the depressive symptomatology commonly associated with chronic pain disorders. Depression is a focus of treatment in this pain rehabilitation program, and antidepressant pharmacotherapy was introduced or augmented in many cases, therefore a cost savings in the antidepressant category was not present.

This study did not include the costs of OTC medications. However, amounts and patterns of OTC use were evaluated at admission, discharge, and at 6-month follow-up. A substantial majority of patients in the study reported a decrease or no change in OTC medication use from admission to both subsequent time points (95.7% and 82.6%, respectively), indicating patients did not substitute OTC analgesics for prescription medications. This also suggests potentially greater cost saving to patients who often pay out-of-pocket for OTC medications in addition to prescription copayments.

This study has several important limitations. The applicability of these findings to other populations of patients with CNMP may be limited because of the selection of the study sample. Patients were admitted to this pain rehabilitation program with the understanding that they would be expected to discontinue analgesic and certain adjuvant medications used for chronic pain. Their motivation and resources for this undertaking may differentiate them from patients presenting to a primary care clinic or other medical settings [44]. Patients in this study reported a long duration of illness (mean duration of CNMP: 9.5 years). As such, this population may represent a distinct subset of chronic pain sufferers and therefore the results may not be reproducible in other settings. Additional research is needed to determine if similar cost savings can be obtained in other rehabilitative programs.

The 6-month follow-up data was obtained from a questionnaire to which 65.1% of the study group responded. Those returning the questionnaire may represent a biased subset of patients who perceive themselves to be more successful with rehabilitative therapy following program completion. However, as previously noted, the two groups (questionnaire responders and nonresponders) were similar in many respects, including medication costs at admission and discharge. Group differences in age and education have an uncertain relationship to ongoing adherence to medication recommendations.

AWP was used to determine prescription medication costs. The AWP is generally an overestimate of the price paid by pharmacies, as it does not reflect common discounts given to large purchasers. However, the AWP may underestimate cash costs. Retail cash costs were not used as they vary significantly from pharmacy to pharmacy and would not reflect a standard representative cost. Cost data for the 6-month follow-up analysis was obtained by patient self-report, without a pharmacist interview or other source of verification. Specific tablet sizes were not always indicated and in these instances, the closest tablet size was selected for any new medications reported. These factors may affect the accuracy of medication cost calculations at the time of 6-month follow-up.


There is substantial evidence in the medical literature that pain rehabilitation is a clinically effective and cost-beneficial treatment for persons with CNMP. This report documents that patients completing a 3-week pain rehabilitation program, during which multiple medications were discontinued (including opioid analgesics, nonsteroidal anti-inflammatory agents, anticonvulsants, benzodiazepines, muscle relaxants, and hypnotics, among others), were able to achieve a significant and sustained reduction in pharmaceutical costs. The findings from this study indicate that pain rehabilitation programs can provide significant savings on pharmaceutical expenditures for patients, and thus for health care systems and the health insurance industry.


We would like express appreciation to the numerous care providers and staff whose commitment and support made this endeavor possible, including: Kelli Lewandowski, PharmD; Erin Dale, PharmD; Jenna LaRock, PharmD; Tiffany Berg, MS; Polly Copeman; Sara Burns; and Yuko Voss.


  • This research was conducted at the Mayo Comprehensive Pain Rehabilitation Center, Mayo Clinic, Rochester, Minnesota, USA.


View Abstract