Journal of Back and Musculoskeletal Rehabilitation
ELSEVIER
Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
Guidelines for program evaluation in chronic non-malignant pain management Steven H. Sanders*a, Karen S. Rucker b , Karen O. AndersonC, R. Norman Hardend , Karl W. Jackson e , Peter J. Vicente f , Rollin M. Gallagherg a Pain
and Health Management Center, 3855 Pleasant Hill Road No. 460, Duluth, OA 30136, USA hVirginia Commonwealth University, Richmond, VA, USA CMemorial Hospital Southwest Pain Center, Houston, TX, USA d Rehabilitation Institute of Chicago Pain Center, Chicago, IL, USA C Community Hospitals Indianapolis Rehabilitation Center for Pain, Indianapolis, IN, USA fOood Samaritan Hospital Pain Rehabilitation Center, Cincinnati, OH, USA g SUNY Comprehensive Pain and Rehabilitation Center, East Setauket, NY, USA
Abstract The current article offers guidelines to systematically evaluate programs which treat chronic non-malignant pain syndrome patients. The guidelines represent a basic program evaluation strategy and include specific recommendations and choices on measurement-assessment tools based upon available research literature and common clinical practice. They are based on evaluation by objectives, which include the program's ability to reduce the misuse of medications, increase physical function, increase productive activity at home, work and socially, improve overall mood, reduce subjective pain intensity, reduce the use of healthcare, when applicable, achieve equitable case settlement, and minimize pain treatment program cost without compromising quality of care. The method and timing of assessing each of these objectives are delineated with an emphasis on using reliable, valid measures which can be applied effectively within a clinical setting. The guidelines also advocate patient and staff satisfaction assessment, thus offering a fully integrated program evaluation system which can measure effectiveness and allow ongoing improvement in care. Keywords: Chronic pain; Program evaluation; Guidelines; Pain rehabilitation
* Corresponding author. 1053-8127/96/$15.00 @ 1996 Elsevier Science Ireland Ltd. All rights reserved. PI I:S 1 053-8127(96)00187-X
20
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
1. Introduction
Given the climate in health care, delineation of practical guidelines to assess program effectiveness is critical and essential for maintaining and improving the overall quality of treatment for pain patients. The current guidelines are intended to offer a systematic and uniform approach to evaluating the effectiveness of treatment programs aimed at helping chronic nonmalignant pain syndrome (CPS) patients [1]. CPS patients are defined here as presenting with, ' ... persistent pain that mayor may not be consistent with physical findings and at least two of the following conditions: (a) progressive deterioration in ability to function at home, socially, and at work; (b) progressive increase in request and usage of medications or invasive medical procedures; (c) mood disturbance; and Cd) significant anger and hostility.' [2]. These guidelines represent a distillation of scientific research and recommendations from the Commission on the Evaluation of Pain [3], other more recent 'Blue Ribbon' panels of experts such as the ongoing work of the American Academy of Pain Medicine (AAPM) Committee to Establish Uniform Outcome Measures for pain treatment programs [4], and the current authors. The guidelines are meant to be used as a complimentary document to those published by the Commission on Accreditation of RehabilitatiOI! Facilities (CARF) [5], primarily in interdisciplinary, organized treatment programs for CPS patients. They can also be used in settings for higher functioning chronic non-malignant pain patients that have similar treatment goals. Likewise, the current guidelines are applicable across a number of different pain problems (e.g., low back, headache, upper extremity, etc.) and both inpatient and outpatient treatment settings. Thus, they represent a basic program evaluation strategy. This includes specific recommendations and choices on measure-assessment tools to use based upon available research literature and common clinical practice.
2. Goals for program evaluation For a program evaluation system to be valuable it must meet certain fundamental goals. These include using empirically and statistically reliable and valid measures to provide accurate and useful information about a treatment program's ability to significantly improve patients' conditions and function within an acceptable level of time and cost. Likewise, the system must provide a clear description of the types and characteristics of patients treated and the time frame for the evaluation. Furthermore, the system must be practical, as well as feasible across a variety of settings, program intensities, and emphasis (i.e. research to clinical). It should allow the integration-combination of information across various treatment programs into a common database network. Finally, it should provide useful information regarding patient and professional staff satisfaction levels. To the extent possible, the current guidelines make every effort to incorporate and achieve the preceding goals. 3. Program evaluation by objectives In keeping with recent practice guidelines [1], CARF [5], and the AAPM Uniform Outcome Measures Committee [4], it is recommended that programs be evaluated by eight basic outcome objectives. These include the: (A) (B) (C) (D) (E) (F) (G) (lI)
reduction in the misuse of medications (if present); increase in physical function; increase in productive activity at home, socially; and/or at work; improvement in overall mood; reduction in patients' subjective pain intensity; reduction in use of and expenditures for health care; where applicable, achievement of case settlement; and minimizing of pain treatment program cost without compromising quality of care.
S.H. Sanders et at. / Journal of Back and Musculoskeletal Rehabilitation 7 (J 996) 19-25
Objectives A-E can be categorized as clinical - patient health status outcome objectives, while P-H are economic-cost outcome objectives. All eight may not apply to a given patient or program depending upon the nature of the patients treated and the appropriateness and pre-treatment level for a given objective. If an objective is not appropriate for a given program (e.g. headache center treating mainly migraine patients in need of more medication usage to maintain function), its deletion needs to be fully justified. However, whenever possible and appropriate, it is recommended that evaluation include these eight outcome objectives. Likewise, the treatment program may wish to include more than these eight objectives in their overall evaluative model. If the resources are available, this goal is highly desirable. 4. Measurement recommendations by outcome objective This section offers specific recommendations regarding which measures to use in quantifying the eight outcome objectives outlined in the preceding section. As already noted, measures have been selected based upon their reliability and validity level, availability and usage in common practice, and their ease and feasibility of application in clinical practice. Thus, decisions regarding measures were not based simply upon the degree of sophistication and/or reliability and validity. For a measure to be recommended, it must also be feasible and applicable within a clinical setting. Therefore, such variables as time to complete a measure, educational level needed to understand, and scoring complexity were all factored into the choices made. A fundamental guiding principle was that whenever possible the measures recommended could be widely applied and readily employed in clinical practice. To do otherwise would significantly jeopardize the utility of these guidelines at the basic practice level, and thus not achieve widespread application and usage. When two or more measures had equal credibility (as just defined) all were recommended, with the limitation
21
that not more than three measures were recommended for anyone outcome objective. If two measures appeared equally credible but one had been more commonly or extensively applied to CPS patients based upon existing literature, the one with more direct application to CPS patients was recommended. Given the preceding qualifications, the following measures are recommended for use to assess the outcomes of the 8 program objectives. 4.1. Patient medication usage 4.1.1. Measures It is recommended that patient medication us-
age be assessed and quantified using the Medication Quantification Scale (MQS) as described by Masters-Steedman et al. [6]. To date, this is the most comprehensive method for quantifying medication usage in CPS patients across drug types and dosage levels. 4.1.2. Measurement administration
Information about medication usage needs to be obtained verbally from the patient, with corroboration from family, pharmacy records, medical records and/or insurance records whenever possible. Information about drugs used daily, weekly, and monthly should be obtained and included in the MQS. 4.2. Physical activity, capacity, and general function 4.2.1. Measure It is recommended that the Sickness Impact
Profile (SIP) [7] be used as a basic measure of physical activity and general functioning. The SIP is a broad measure of function across 12 areas. These include ambulation, body care, mobility, emotional behavior, social, alertness, communication, work, sleep, eating, home management, and recreation. If due to time and cost issues the complete SIP can not be consistently administered to patients, the SIP Roland Scale [8-11] may be substituted. This scale is a 'short form' of the SIP. Whenever possible, however, the complete SIP should be used. As an alternative to the SIP, the Medical Out-
22
S.H. Sanders et at. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
come Study (MOS) short form 36 (SF-36) [12-13] may be considered. This measure of general function has been recommended by the AAPM Committee on Uniform Outcome Measures and contains subscales focusing on general level of physical functioning. While it does not have the more extensive application of the SIP, it apparently is being used in a number of research protocols with low back pain patients [14], and has solid reliability and validity. The Multidimensional Pain Inventory (MPI) [15] is recommended as a second optional measure of general function to the SIP. The increased complexity in scoring this measure may make it less desirable for some. It is recommended that an estimation of actual time standing and walking (up time) also be obtained by patient self-monitoring using diary data as outlined by Fordyce [16] and extensively reported in the clinical literature. When possible and appropriate (e.g. low back pain) it is recommended that a measure of physical capacity be taken. Specifically, it is recommended that isokinetic and isoinertial lifting capacity be assessed [17]. These measures will only be applicable in those situations where programs have access to isokinetic lifting devices such as Cybex or Lido systems. Thus, it is recommended as an optional measure if available. 4.2.2. Measure administration
The SIP, MOS-SF-36, and MPI have specific instructions and protocols outlined in their respective descriptive articles which need to be followed. These measures are typically given in written form on an individual basis. The up-time diary measure has been described in detail by Fordyce [16], as well as in the research and clinical literature. Protocols for the optional isokinetic and isoinertial lifting are also readily available in literature. It is recommended that the protocol developed by Mayer et al. [18,19] be followed whenever possible.
4.3. Productivity at home, socially, and at work 4.3.1. Measure It is recommended that the SIP be used to
assess basic productivity of chronic pain patients. As an alternative the MOS-SF-36 or MPI can be
considered, however, the SIP provides separate subs cales specifically for home, social and the work environment. Likewise, it is recommended that the work status of each patient be obtained (e.g., currently working full-time, part-time, not working due to injury-pain, unemployed but able to work, not working and on workers' compensation, not working and on social security disability, not working and on private short-term or longterm disability insurance, not working and on no workers' compensation-disability income, retired, working in the home). 4.3.2. Measure administration
As noted, the questionnaire measures have their own standard protocols that need to be followed. Questions about work status should be administered individually to the patient in written or verbal form and confirmed by family and/or employers when ever possible. 4.4. General mood and well being 4.4.1. Measure
In addition to the subscales contained in the SIP, MOS-SF-36, and the MPI which provide some basic data on emotional-mood status, it is recommended that the long or short forms of the Beck Depression Inventory (BDI) [20,21] be used to specifically assess depression.
4.4.2. Measure administration
The SIP, MOS-SF-36, MPI, and long or short forms of the BDI all have standardized written instructions for administration and should be given in written form whenever possible. 4.5. Patient subjective pain intensity 4.5.1. Measure It is recommended that the Visual Analog Scale
(VAS) [22] be used to assess subjective pain intensity. Specifically, the VAS scale using a horizontal 10-cm line as depicted in the Department of Health and Human Services clinical practice guidelines for acute pain management [23] is re-
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
commended. For consistency, it is recommended that the measure be scored using a 0-10.0 scale. A more sophisticated version of the VAS has recently been introduced by Price et al. [24] which uses a mechanical slide rule device to depict the VAS scale. This would be an acceptable alternative to the VAS line technique, using the same 0-10.0 quantification scale.
4.5.2. Measure administration It is recommended that VAS subjective pain intensity scale be administered individually for a given patient, with only one VAS scale depicted on the page if the analog line method is used. The scale needs to include some brief written instructions describing the measure as a rating of their pain intensity from 'no pain' to 'pain as bad as it could possibly be'. The patient should be instructed that they are to make a vertical (up and down) mark somewhere along the line which they feel represents the average or typical pain intensity level over the last 24 h. 4.6. Health care usage 4.6.1. Measure As of this publication there was no well tested and widely used measurement instrument for health care usage applied to chronic pain patients. Thus, no standardized instrument can be recommended. However, it is recommended that the nature and extent of health usage by CPS patients be evaluated. Specifically, patients need to be asked to report the number and frequency of contacts with physicians for their pain problem, any ER visits for pain, and number of days hospitalized for treatment of their pain problem within the last 90 days. This information can be quantified in frequency counts and reported as such. 4.6.2. Measure administration The frequency counts for health care contact need to be obtained from the patient directly, with corroboration from family, physicians, and hospital records whenever possible. Likewise, if a third party carrier is involved and will cooperate, information needs to be requested from the third party if the patient will agree.
23
4. 7. Case settlement
4.7.1. Measure In those patients with a liability claim either through workers' compensation, private disability insurance or auto accident insurance, it is recommended that information about the status of the case and whether it has been settled be obtained whenever possible. This could be in the form of a simple 'status of your case' question scored as either settled or not settled in a binary fashion. 4.7.2. Measure administration It is recommended that the case settlement question be administered in written form if possible and corroborated by a third party when appropriate. 4.8. Pain treatment costs 4.8.1. Measure It is recommended that each program include the total cost of treatment for a given chronic pain patient. This should include service, materials, lodging, meals, and transportation costs for the direct care of the patient. 4.8.2. Measure administration Cost data can be gathered by the program administrator. 5. Timing of measurement It is recommended that measures for Objectives A-F be obtained at least at the beginning and end of initial treatment. If at all possible, these measures also need to be obtained at 3 months following the end of initial treatment. Likewise, contingent on patient availability, these data need to be obtained at 6 and 12 months following the end of treatment whenever possible. Pre- and post-treatment measures should be taken while the patient is actually present. Follow-up assessment should also be done in person, however, in those cases where it is not possible for the patient to return, information can be gathered by mail or phone interview where appropriate. When phone interview is used, measures for
24
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
Objectives A and B (at least up time), C (at least return to work), D (at least the BDl), E (verbal rating of pain on 0-10 scale), and F should be included. Measures for Objectives G and H need to be obtained at least at the end of treatment and at 3-month follow-up. Whenever possible, 6- and 12-month follow-up data for Objectives G and H need to be gathered. 6. Patient and staff satisfaction
6.1. Patient satisfaction It is recommended that patient satisfaction be evaluated at the end of active treatment and 3-month follow-up. While there are no standardized instruments currently in use, Chapman et al. [25] have proposed a preliminary scale which is recommended for consideration. The scale developed by Chapman et al. used a likert-type rating for a number of categories and treatment services. Regardless of the specific measure used, it is recommended that it assess at least patient satisfaction with overall treatment and individual treatment modalities-areas.
6.2. Staff satisfaction It is recommended that staff satisfaction within the pain treatment program be evaluated on a quarterly to annual basis [26]. Again, there are no standardized and widely used instruments within the chronic pain management area to recommend at this point. Whatever measure is used, it should provide quantifiable information regarding both professional and support staff satisfaction with patient mix, work environment, administrative and clinical support, treatment content and efficacy, and professional development-advancement.
7. Patient demographic data It is recommended that program evaluation include at least the following basic demographic and clinical characteristic data for patients served: age, sex, educational level, marital status, pain type-site using the International Association for
the Study of Pain (IASP) classification [27], duration of chronic pain, and number of pain related surgeries. This information can be quantified as means, medians, or percentages based upon the nature of the data. 8. Data sampling and reporting It is recommended that program outcome measures be applied to as many patients as possible. Follow-up data needs to be gathered on at least 10% of those patients treated in a given 12-month period. If only a sample of patients participate, they should be randomly chosen from the larger patient population treated. In no case should the number of patients participating in the program evaluation sample be less than 25. A larger sample size would be desirable if conducting controlled outcome research. However, for the purposes of program evaluation, the 10% and 25-patient sample minimums recommended, if truly randomized, should provide a meaningful estimate of treatment results that can be realistically gathered by most clinically based programs. Program evaluation data should be summarized on an annual basis and shared with all members of the support and professional staff, as well as appropriate individuals outside the actual program.
9. Recommendation update-review Given the evolutionary process of these recommendations, it is recommended that they be reviewed and updated at least every 2 years. This update should consider other, new measures of the various objectives that meet inclusion criteria. Examples of measures to consider in the future might include those supported by the on going research project of the AAPM Uniform Outcome Measures Committee [4] and the Multiperspective Multidimensional Pain Assessment Protocol [28], as well as others. References [1] Sanders SH, Rucker KS, Anderson KO, Harden RN, Jackson KW, Vicente PI, Gallagher RM. Clinical prac-
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
[2] [3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
tice guidelines for non-malignant pain syndrome patients. J Back Musculoskel Rehabil 1995;5:115-120. Sanders SH, Brena SF. Pain centers: what consumers want to know. APS Bull 1995;5:8-11. Commission on the Evaluation of Pain. Report of the Commission on the Evaluation of Pain. Washington, DC: US Department of HHS, 1986. Marcus N, ed. AAPM Uniform Outcome Measures Committee Report. Skokie, It: American Academy of Pain Medicine, 1994. CARF. Program Evaluation in Chronic Pain Management Programs. Tuson, AZ: Commission on Accreditation of Rehabilitation Facilities, 1987. Masters-Steedman S, Middaugh SJ, Kee WG, Stier-Carson D, Harden RN, Miller NC. Chronic-pain medications: equivalence levels and method of quantifying usage. Clin J Pain 1992;8:204-214. Bergner M, Bobbitt R, Carter W et al. The sickness impact profile: development and final revision of a health status measure. Med Care 1981 ;19:787-805. Deyo RA. Comparative validity of the sickness impact profile and shorter scales for functional assessment in low-back pain. Spine 1986;11 :951-954. Jensen MP, Strom SE, Turner JA, Romano JM. Validity of the sickness impact profile roland scale as a measure of dysfunction in chronic pain patients. Pain 1992;50: 157 -162. Roland M, Morris R. A study of the natural history of back pain part 1. Development of a reliable and sensitive measure of disability in low-back pain. Spine 1983;8:141-144. Stratford P, Solomon P, Binkley J, Finch E, Gill C. Sensitivity of sickness impact profile items to measure change over time in a low-back pain patient group. Spine 1993;18:1723-1727. Ware JE, Sherbourne CD. The MOS 36-item short-form health sUlvey (SF-36): I. Conceptual framework and item selection. Med Care 1992;30:473-483. Stewart A, Ware JE. Measuring Function and Well Being: The Medical Outcomes Study Approach. Durham, NC: Duke University Press, 1992. Patrick DL, Deyo RA, Atlas SJ, Singer DE, Chapin A, Keller RB. Assessing health-related quality of life in patients with sciatica. Spine 1995;20:1899-1909. Kerns RD, Turk DC, Rudy TE. The Westhaven-Yale
[16] [17]
[18]
[19]
[20]
[21) [22]
[23]
[24]
[25]
[26]
[27]
[28]
25
multidimensional pain inventory (WHYMPI). Pain 1985;23:345-356. Fordyce WE. Behavioral Methods for Chronic Pain and Illness. St. Louis, MO: Mosby Company, 1976. Curtis L, Mayer TG, Gatchel RJ. Physical progress and residual impairment quantification after functional restoration: Part III: lsokinetic and isoinertial lifting capacity. Spine 1994;19:401-405. Kishino N, Mayer T, Gatchel RJ et al. Quantification of lumbar function 4: Isometric andisokinetic stimulation in control subjects and low back pain patients. Spine 1985;10:921-927. Mayer T, Barnes D, Kishino N et al. Progressive isoinertial lifting evaluation: I. A standardized protocol and normative data base. Spine 1988;13:993·-997. Beck AT, Rial WY, Rickles K. Short form of depression inventory: cross validation. Psychol Rep 1974;34:1184-1186. Beck AT, Steer RA. Beck Depression Inventory a Manual. San Antonio, CA: Harcourt Brace Jovanovich, 1987. Huskisson EC. Visual analog scales. In: R. Melzack, ed. Pain Measurement and Assessment. New York: Raven Press, 1983;33-37. AHCPR. Clinical Practice Guidelines: Acute Pain Management: Operative or Medical Procedures and Trauma Rockville, MD: US Department, HHS 1992, pp. 116. Price DD, Bush FM, Long S, Hawkins SW. A comparison of pain measurement characteristics of mechanical visual analogue and simple numerical rating scales. Pain 1994;56:217-226. Chapman SL, Jameson RN, Sanders SH. Patient satisfaction and cost effectiveness: A comparative analysis across two interdisciplinary pain centers. Poster presented at the Annual Scientific Meeting of the American Pain Society, New Orleans, LA, 1991. Pawlicki RE. Achieving excellence: Staff satisfaction plays a role in program evaluation. APS Bull 1993;3:14-15. Merskey H, Bogduk N, eds. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms, 2nd ed. Seattle: lASP Press, 1994. Rucker KS, Metzler HM. Predicting subsequent employment status of SSA disability applicants with chronic pain. Clin J Pain 1995;11 :22-35.
ELSEVIER
Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
Guidelines for program evaluation in chronic non-malignant pain management Steven H. Sanders*a, Karen S. Rucker b , Karen O. AndersonC, R. Norman Hardend , Karl W. Jackson e , Peter J. Vicente f , Rollin M. Gallagherg a Pain
and Health Management Center, 3855 Pleasant Hill Road No. 460, Duluth, OA 30136, USA hVirginia Commonwealth University, Richmond, VA, USA CMemorial Hospital Southwest Pain Center, Houston, TX, USA d Rehabilitation Institute of Chicago Pain Center, Chicago, IL, USA C Community Hospitals Indianapolis Rehabilitation Center for Pain, Indianapolis, IN, USA fOood Samaritan Hospital Pain Rehabilitation Center, Cincinnati, OH, USA g SUNY Comprehensive Pain and Rehabilitation Center, East Setauket, NY, USA
Abstract The current article offers guidelines to systematically evaluate programs which treat chronic non-malignant pain syndrome patients. The guidelines represent a basic program evaluation strategy and include specific recommendations and choices on measurement-assessment tools based upon available research literature and common clinical practice. They are based on evaluation by objectives, which include the program's ability to reduce the misuse of medications, increase physical function, increase productive activity at home, work and socially, improve overall mood, reduce subjective pain intensity, reduce the use of healthcare, when applicable, achieve equitable case settlement, and minimize pain treatment program cost without compromising quality of care. The method and timing of assessing each of these objectives are delineated with an emphasis on using reliable, valid measures which can be applied effectively within a clinical setting. The guidelines also advocate patient and staff satisfaction assessment, thus offering a fully integrated program evaluation system which can measure effectiveness and allow ongoing improvement in care. Keywords: Chronic pain; Program evaluation; Guidelines; Pain rehabilitation
* Corresponding author. 1053-8127/96/$15.00 @ 1996 Elsevier Science Ireland Ltd. All rights reserved. PI I:S 1 053-8127(96)00187-X
20
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
1. Introduction
Given the climate in health care, delineation of practical guidelines to assess program effectiveness is critical and essential for maintaining and improving the overall quality of treatment for pain patients. The current guidelines are intended to offer a systematic and uniform approach to evaluating the effectiveness of treatment programs aimed at helping chronic nonmalignant pain syndrome (CPS) patients [1]. CPS patients are defined here as presenting with, ' ... persistent pain that mayor may not be consistent with physical findings and at least two of the following conditions: (a) progressive deterioration in ability to function at home, socially, and at work; (b) progressive increase in request and usage of medications or invasive medical procedures; (c) mood disturbance; and Cd) significant anger and hostility.' [2]. These guidelines represent a distillation of scientific research and recommendations from the Commission on the Evaluation of Pain [3], other more recent 'Blue Ribbon' panels of experts such as the ongoing work of the American Academy of Pain Medicine (AAPM) Committee to Establish Uniform Outcome Measures for pain treatment programs [4], and the current authors. The guidelines are meant to be used as a complimentary document to those published by the Commission on Accreditation of RehabilitatiOI! Facilities (CARF) [5], primarily in interdisciplinary, organized treatment programs for CPS patients. They can also be used in settings for higher functioning chronic non-malignant pain patients that have similar treatment goals. Likewise, the current guidelines are applicable across a number of different pain problems (e.g., low back, headache, upper extremity, etc.) and both inpatient and outpatient treatment settings. Thus, they represent a basic program evaluation strategy. This includes specific recommendations and choices on measure-assessment tools to use based upon available research literature and common clinical practice.
2. Goals for program evaluation For a program evaluation system to be valuable it must meet certain fundamental goals. These include using empirically and statistically reliable and valid measures to provide accurate and useful information about a treatment program's ability to significantly improve patients' conditions and function within an acceptable level of time and cost. Likewise, the system must provide a clear description of the types and characteristics of patients treated and the time frame for the evaluation. Furthermore, the system must be practical, as well as feasible across a variety of settings, program intensities, and emphasis (i.e. research to clinical). It should allow the integration-combination of information across various treatment programs into a common database network. Finally, it should provide useful information regarding patient and professional staff satisfaction levels. To the extent possible, the current guidelines make every effort to incorporate and achieve the preceding goals. 3. Program evaluation by objectives In keeping with recent practice guidelines [1], CARF [5], and the AAPM Uniform Outcome Measures Committee [4], it is recommended that programs be evaluated by eight basic outcome objectives. These include the: (A) (B) (C) (D) (E) (F) (G) (lI)
reduction in the misuse of medications (if present); increase in physical function; increase in productive activity at home, socially; and/or at work; improvement in overall mood; reduction in patients' subjective pain intensity; reduction in use of and expenditures for health care; where applicable, achievement of case settlement; and minimizing of pain treatment program cost without compromising quality of care.
S.H. Sanders et at. / Journal of Back and Musculoskeletal Rehabilitation 7 (J 996) 19-25
Objectives A-E can be categorized as clinical - patient health status outcome objectives, while P-H are economic-cost outcome objectives. All eight may not apply to a given patient or program depending upon the nature of the patients treated and the appropriateness and pre-treatment level for a given objective. If an objective is not appropriate for a given program (e.g. headache center treating mainly migraine patients in need of more medication usage to maintain function), its deletion needs to be fully justified. However, whenever possible and appropriate, it is recommended that evaluation include these eight outcome objectives. Likewise, the treatment program may wish to include more than these eight objectives in their overall evaluative model. If the resources are available, this goal is highly desirable. 4. Measurement recommendations by outcome objective This section offers specific recommendations regarding which measures to use in quantifying the eight outcome objectives outlined in the preceding section. As already noted, measures have been selected based upon their reliability and validity level, availability and usage in common practice, and their ease and feasibility of application in clinical practice. Thus, decisions regarding measures were not based simply upon the degree of sophistication and/or reliability and validity. For a measure to be recommended, it must also be feasible and applicable within a clinical setting. Therefore, such variables as time to complete a measure, educational level needed to understand, and scoring complexity were all factored into the choices made. A fundamental guiding principle was that whenever possible the measures recommended could be widely applied and readily employed in clinical practice. To do otherwise would significantly jeopardize the utility of these guidelines at the basic practice level, and thus not achieve widespread application and usage. When two or more measures had equal credibility (as just defined) all were recommended, with the limitation
21
that not more than three measures were recommended for anyone outcome objective. If two measures appeared equally credible but one had been more commonly or extensively applied to CPS patients based upon existing literature, the one with more direct application to CPS patients was recommended. Given the preceding qualifications, the following measures are recommended for use to assess the outcomes of the 8 program objectives. 4.1. Patient medication usage 4.1.1. Measures It is recommended that patient medication us-
age be assessed and quantified using the Medication Quantification Scale (MQS) as described by Masters-Steedman et al. [6]. To date, this is the most comprehensive method for quantifying medication usage in CPS patients across drug types and dosage levels. 4.1.2. Measurement administration
Information about medication usage needs to be obtained verbally from the patient, with corroboration from family, pharmacy records, medical records and/or insurance records whenever possible. Information about drugs used daily, weekly, and monthly should be obtained and included in the MQS. 4.2. Physical activity, capacity, and general function 4.2.1. Measure It is recommended that the Sickness Impact
Profile (SIP) [7] be used as a basic measure of physical activity and general functioning. The SIP is a broad measure of function across 12 areas. These include ambulation, body care, mobility, emotional behavior, social, alertness, communication, work, sleep, eating, home management, and recreation. If due to time and cost issues the complete SIP can not be consistently administered to patients, the SIP Roland Scale [8-11] may be substituted. This scale is a 'short form' of the SIP. Whenever possible, however, the complete SIP should be used. As an alternative to the SIP, the Medical Out-
22
S.H. Sanders et at. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
come Study (MOS) short form 36 (SF-36) [12-13] may be considered. This measure of general function has been recommended by the AAPM Committee on Uniform Outcome Measures and contains subscales focusing on general level of physical functioning. While it does not have the more extensive application of the SIP, it apparently is being used in a number of research protocols with low back pain patients [14], and has solid reliability and validity. The Multidimensional Pain Inventory (MPI) [15] is recommended as a second optional measure of general function to the SIP. The increased complexity in scoring this measure may make it less desirable for some. It is recommended that an estimation of actual time standing and walking (up time) also be obtained by patient self-monitoring using diary data as outlined by Fordyce [16] and extensively reported in the clinical literature. When possible and appropriate (e.g. low back pain) it is recommended that a measure of physical capacity be taken. Specifically, it is recommended that isokinetic and isoinertial lifting capacity be assessed [17]. These measures will only be applicable in those situations where programs have access to isokinetic lifting devices such as Cybex or Lido systems. Thus, it is recommended as an optional measure if available. 4.2.2. Measure administration
The SIP, MOS-SF-36, and MPI have specific instructions and protocols outlined in their respective descriptive articles which need to be followed. These measures are typically given in written form on an individual basis. The up-time diary measure has been described in detail by Fordyce [16], as well as in the research and clinical literature. Protocols for the optional isokinetic and isoinertial lifting are also readily available in literature. It is recommended that the protocol developed by Mayer et al. [18,19] be followed whenever possible.
4.3. Productivity at home, socially, and at work 4.3.1. Measure It is recommended that the SIP be used to
assess basic productivity of chronic pain patients. As an alternative the MOS-SF-36 or MPI can be
considered, however, the SIP provides separate subs cales specifically for home, social and the work environment. Likewise, it is recommended that the work status of each patient be obtained (e.g., currently working full-time, part-time, not working due to injury-pain, unemployed but able to work, not working and on workers' compensation, not working and on social security disability, not working and on private short-term or longterm disability insurance, not working and on no workers' compensation-disability income, retired, working in the home). 4.3.2. Measure administration
As noted, the questionnaire measures have their own standard protocols that need to be followed. Questions about work status should be administered individually to the patient in written or verbal form and confirmed by family and/or employers when ever possible. 4.4. General mood and well being 4.4.1. Measure
In addition to the subscales contained in the SIP, MOS-SF-36, and the MPI which provide some basic data on emotional-mood status, it is recommended that the long or short forms of the Beck Depression Inventory (BDI) [20,21] be used to specifically assess depression.
4.4.2. Measure administration
The SIP, MOS-SF-36, MPI, and long or short forms of the BDI all have standardized written instructions for administration and should be given in written form whenever possible. 4.5. Patient subjective pain intensity 4.5.1. Measure It is recommended that the Visual Analog Scale
(VAS) [22] be used to assess subjective pain intensity. Specifically, the VAS scale using a horizontal 10-cm line as depicted in the Department of Health and Human Services clinical practice guidelines for acute pain management [23] is re-
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
commended. For consistency, it is recommended that the measure be scored using a 0-10.0 scale. A more sophisticated version of the VAS has recently been introduced by Price et al. [24] which uses a mechanical slide rule device to depict the VAS scale. This would be an acceptable alternative to the VAS line technique, using the same 0-10.0 quantification scale.
4.5.2. Measure administration It is recommended that VAS subjective pain intensity scale be administered individually for a given patient, with only one VAS scale depicted on the page if the analog line method is used. The scale needs to include some brief written instructions describing the measure as a rating of their pain intensity from 'no pain' to 'pain as bad as it could possibly be'. The patient should be instructed that they are to make a vertical (up and down) mark somewhere along the line which they feel represents the average or typical pain intensity level over the last 24 h. 4.6. Health care usage 4.6.1. Measure As of this publication there was no well tested and widely used measurement instrument for health care usage applied to chronic pain patients. Thus, no standardized instrument can be recommended. However, it is recommended that the nature and extent of health usage by CPS patients be evaluated. Specifically, patients need to be asked to report the number and frequency of contacts with physicians for their pain problem, any ER visits for pain, and number of days hospitalized for treatment of their pain problem within the last 90 days. This information can be quantified in frequency counts and reported as such. 4.6.2. Measure administration The frequency counts for health care contact need to be obtained from the patient directly, with corroboration from family, physicians, and hospital records whenever possible. Likewise, if a third party carrier is involved and will cooperate, information needs to be requested from the third party if the patient will agree.
23
4. 7. Case settlement
4.7.1. Measure In those patients with a liability claim either through workers' compensation, private disability insurance or auto accident insurance, it is recommended that information about the status of the case and whether it has been settled be obtained whenever possible. This could be in the form of a simple 'status of your case' question scored as either settled or not settled in a binary fashion. 4.7.2. Measure administration It is recommended that the case settlement question be administered in written form if possible and corroborated by a third party when appropriate. 4.8. Pain treatment costs 4.8.1. Measure It is recommended that each program include the total cost of treatment for a given chronic pain patient. This should include service, materials, lodging, meals, and transportation costs for the direct care of the patient. 4.8.2. Measure administration Cost data can be gathered by the program administrator. 5. Timing of measurement It is recommended that measures for Objectives A-F be obtained at least at the beginning and end of initial treatment. If at all possible, these measures also need to be obtained at 3 months following the end of initial treatment. Likewise, contingent on patient availability, these data need to be obtained at 6 and 12 months following the end of treatment whenever possible. Pre- and post-treatment measures should be taken while the patient is actually present. Follow-up assessment should also be done in person, however, in those cases where it is not possible for the patient to return, information can be gathered by mail or phone interview where appropriate. When phone interview is used, measures for
24
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
Objectives A and B (at least up time), C (at least return to work), D (at least the BDl), E (verbal rating of pain on 0-10 scale), and F should be included. Measures for Objectives G and H need to be obtained at least at the end of treatment and at 3-month follow-up. Whenever possible, 6- and 12-month follow-up data for Objectives G and H need to be gathered. 6. Patient and staff satisfaction
6.1. Patient satisfaction It is recommended that patient satisfaction be evaluated at the end of active treatment and 3-month follow-up. While there are no standardized instruments currently in use, Chapman et al. [25] have proposed a preliminary scale which is recommended for consideration. The scale developed by Chapman et al. used a likert-type rating for a number of categories and treatment services. Regardless of the specific measure used, it is recommended that it assess at least patient satisfaction with overall treatment and individual treatment modalities-areas.
6.2. Staff satisfaction It is recommended that staff satisfaction within the pain treatment program be evaluated on a quarterly to annual basis [26]. Again, there are no standardized and widely used instruments within the chronic pain management area to recommend at this point. Whatever measure is used, it should provide quantifiable information regarding both professional and support staff satisfaction with patient mix, work environment, administrative and clinical support, treatment content and efficacy, and professional development-advancement.
7. Patient demographic data It is recommended that program evaluation include at least the following basic demographic and clinical characteristic data for patients served: age, sex, educational level, marital status, pain type-site using the International Association for
the Study of Pain (IASP) classification [27], duration of chronic pain, and number of pain related surgeries. This information can be quantified as means, medians, or percentages based upon the nature of the data. 8. Data sampling and reporting It is recommended that program outcome measures be applied to as many patients as possible. Follow-up data needs to be gathered on at least 10% of those patients treated in a given 12-month period. If only a sample of patients participate, they should be randomly chosen from the larger patient population treated. In no case should the number of patients participating in the program evaluation sample be less than 25. A larger sample size would be desirable if conducting controlled outcome research. However, for the purposes of program evaluation, the 10% and 25-patient sample minimums recommended, if truly randomized, should provide a meaningful estimate of treatment results that can be realistically gathered by most clinically based programs. Program evaluation data should be summarized on an annual basis and shared with all members of the support and professional staff, as well as appropriate individuals outside the actual program.
9. Recommendation update-review Given the evolutionary process of these recommendations, it is recommended that they be reviewed and updated at least every 2 years. This update should consider other, new measures of the various objectives that meet inclusion criteria. Examples of measures to consider in the future might include those supported by the on going research project of the AAPM Uniform Outcome Measures Committee [4] and the Multiperspective Multidimensional Pain Assessment Protocol [28], as well as others. References [1] Sanders SH, Rucker KS, Anderson KO, Harden RN, Jackson KW, Vicente PI, Gallagher RM. Clinical prac-
S.H. Sanders et al. / Journal of Back and Musculoskeletal Rehabilitation 7 (1996) 19-25
[2] [3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
tice guidelines for non-malignant pain syndrome patients. J Back Musculoskel Rehabil 1995;5:115-120. Sanders SH, Brena SF. Pain centers: what consumers want to know. APS Bull 1995;5:8-11. Commission on the Evaluation of Pain. Report of the Commission on the Evaluation of Pain. Washington, DC: US Department of HHS, 1986. Marcus N, ed. AAPM Uniform Outcome Measures Committee Report. Skokie, It: American Academy of Pain Medicine, 1994. CARF. Program Evaluation in Chronic Pain Management Programs. Tuson, AZ: Commission on Accreditation of Rehabilitation Facilities, 1987. Masters-Steedman S, Middaugh SJ, Kee WG, Stier-Carson D, Harden RN, Miller NC. Chronic-pain medications: equivalence levels and method of quantifying usage. Clin J Pain 1992;8:204-214. Bergner M, Bobbitt R, Carter W et al. The sickness impact profile: development and final revision of a health status measure. Med Care 1981 ;19:787-805. Deyo RA. Comparative validity of the sickness impact profile and shorter scales for functional assessment in low-back pain. Spine 1986;11 :951-954. Jensen MP, Strom SE, Turner JA, Romano JM. Validity of the sickness impact profile roland scale as a measure of dysfunction in chronic pain patients. Pain 1992;50: 157 -162. Roland M, Morris R. A study of the natural history of back pain part 1. Development of a reliable and sensitive measure of disability in low-back pain. Spine 1983;8:141-144. Stratford P, Solomon P, Binkley J, Finch E, Gill C. Sensitivity of sickness impact profile items to measure change over time in a low-back pain patient group. Spine 1993;18:1723-1727. Ware JE, Sherbourne CD. The MOS 36-item short-form health sUlvey (SF-36): I. Conceptual framework and item selection. Med Care 1992;30:473-483. Stewart A, Ware JE. Measuring Function and Well Being: The Medical Outcomes Study Approach. Durham, NC: Duke University Press, 1992. Patrick DL, Deyo RA, Atlas SJ, Singer DE, Chapin A, Keller RB. Assessing health-related quality of life in patients with sciatica. Spine 1995;20:1899-1909. Kerns RD, Turk DC, Rudy TE. The Westhaven-Yale
[16] [17]
[18]
[19]
[20]
[21) [22]
[23]
[24]
[25]
[26]
[27]
[28]
25
multidimensional pain inventory (WHYMPI). Pain 1985;23:345-356. Fordyce WE. Behavioral Methods for Chronic Pain and Illness. St. Louis, MO: Mosby Company, 1976. Curtis L, Mayer TG, Gatchel RJ. Physical progress and residual impairment quantification after functional restoration: Part III: lsokinetic and isoinertial lifting capacity. Spine 1994;19:401-405. Kishino N, Mayer T, Gatchel RJ et al. Quantification of lumbar function 4: Isometric andisokinetic stimulation in control subjects and low back pain patients. Spine 1985;10:921-927. Mayer T, Barnes D, Kishino N et al. Progressive isoinertial lifting evaluation: I. A standardized protocol and normative data base. Spine 1988;13:993·-997. Beck AT, Rial WY, Rickles K. Short form of depression inventory: cross validation. Psychol Rep 1974;34:1184-1186. Beck AT, Steer RA. Beck Depression Inventory a Manual. San Antonio, CA: Harcourt Brace Jovanovich, 1987. Huskisson EC. Visual analog scales. In: R. Melzack, ed. Pain Measurement and Assessment. New York: Raven Press, 1983;33-37. AHCPR. Clinical Practice Guidelines: Acute Pain Management: Operative or Medical Procedures and Trauma Rockville, MD: US Department, HHS 1992, pp. 116. Price DD, Bush FM, Long S, Hawkins SW. A comparison of pain measurement characteristics of mechanical visual analogue and simple numerical rating scales. Pain 1994;56:217-226. Chapman SL, Jameson RN, Sanders SH. Patient satisfaction and cost effectiveness: A comparative analysis across two interdisciplinary pain centers. Poster presented at the Annual Scientific Meeting of the American Pain Society, New Orleans, LA, 1991. Pawlicki RE. Achieving excellence: Staff satisfaction plays a role in program evaluation. APS Bull 1993;3:14-15. Merskey H, Bogduk N, eds. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms, 2nd ed. Seattle: lASP Press, 1994. Rucker KS, Metzler HM. Predicting subsequent employment status of SSA disability applicants with chronic pain. Clin J Pain 1995;11 :22-35.
