|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Empirical Comparison of Commonly Used Measures to Evaluate Pain Treatment in Cancer Patients With Chronic PainFrom the Division of Psychosocial Research and Epidemiology, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, and Faculty of Psychology, University of Amsterdam, Amsterdam; Department of Medical Psychology and Psychotherapy, Erasmus University, Rotterdam, and Pain Expertise Center, University Hospital Rotterdam, Rotterdam; and Department of Nursing Science, University of Maastricht, Maastricht, the Netherlands. Address reprint requests to Frits van Dam, PhD, Division of Psychosocial Research and Epidemiology, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands; email RdeWit{at}nki.nl
PURPOSE: There is limited consensus about the most appropriate measures to evaluate the adequacy of pain treatment in cancer patients. There are no known studies describing commonly used measures to simultaneously evaluate the adequacy of cancer pain treatment. The purpose of this study was to compare measures, which are frequently reported in the literature, to assess the adequacy of pain treatment in cancer patients with chronic pain. This study was part of a randomized controlled trial. PATIENTS AND METHODS: In total, 313 cancer patients with a pain duration of at least 1 month were evaluated. After a baseline measure in the hospital, patients were followed up at 2, 4, and 8 weeks after discharge at home. Adequacy of cancer pain treatment was evaluated by means of four different types of outcome measures. The four types included three pain intensity markers based on patients' pain intensity, a pain relief scale, a patient satisfaction scale, and three pain management indexes that related patients' pain medication with pain intensity. RESULTS: The proportion of inadequately treated pain patients varied extremely. Depending on the outcome measure used, the percentage of inadequately treated patients ranged from 16% to 91%. The choice of measure, rather than pain treatment itself, determined the proportion of inadequacy. CONCLUSION: There is an urgent need for consensus about how to evaluate the effectiveness of pain treatment. Studies that evaluate adequacy of pain treatment should be interpreted with caution. Further research is necessary to elucidate the validity and reliability of outcome measures simultaneously.
MANY STUDIES SHOW that pain due to cancer is inadequately treated.1,2 The percentage of cancer patients with inadequate treatment of pain is reported to range widely from 7% to 90%.2-23 Heterogeneity of patient groups, treatment regimens, and research designs may, in part, explain the large variance in adequacy. The present study investigates another possible cause of differences in treatment outcomes, namely, the diversity of measures used to evaluate pain treatment. Little attention has been directed toward the measurement of outcomes of effective pain treatment. There are no known studies comparing commonly used measures to evaluate the adequacy of cancer pain treatment. Consequently, the lack of a gold standard to evaluate the efficacy of pain programs results in the use of various measures and makes comparison of pain studies difficult. Because pain is a multidimensional concept, a wide variety of dimensions might be used to assess the quality of pain treatment, such as physiologic, sensory, affective, cognitive, behavioral, and sociocultural dimensions. Although the importance of multidimensional measurement of cancer pain has been acknowledged, pain treatment is most frequently evaluated unidimensionally. However, the measures described in this study primarily represent the sensory and behavioral dimensions. In the literature, the typologic classification of commonly used outcome measures to evaluate pain treatment can be arranged into four groups. The first type of outcome measure that is frequently used to estimate treatment effects is based on patients' self-report of pain intensity.7-11,24,25 The measurement of pain intensity continues to be the most frequently used tool in monitoring pain. Several pain intensity scales are currently used to assess the pain, such as visual analog scales, numeric rating scales, and verbal rating scales.26-28 Zech et al showed that 23% of patients experienced severe pain or worse for more than 10% of the treatment period as measured on a verbal rating scale. In two studies, Grond et al reported that 12% and 3% of the cancer patients experienced severe pain or worse. To obtain an indication for adequate pain treatment, the American Pain Society (APS)29 and the Agency for Health Care Policy and Research (AHCPR)30 proposed a cutoff score of 5 or higher on a numeric rating scale to describe inadequate cancer pain treatment. This cutoff score is based on studies showing that higher pain scores interfere with daily functioning.2,6,12 Cleeland et al found that 62% of the advanced cancer patients rated their worst pain intensity as 5 or higher; Larue et al found that 69% reported a pain score of 5 or higher on Worst Pain Intensity, whereas 54% had a score of 5 or higher on the Average Pain Intensity Scale. Serlin et al reported that 70% to 90% of the patients rated their worst pain intensity as 5 or higher. The second type of frequently used outcome measure to evaluate pain treatment consists of estimating the change in pain intensity.31,32 With the Pain Relief Scale, a weighted judgment about the effect of the pain treatment is given. Patients are asked to compare themselves on a scale that starts for everyone at the same level of "no change" or "worsening pain" and ends at "complete relief."10,16-19 Studies showed complete pain relief in 87% to 93% of patients, whereas Goisis et al found partial or total pain control in 93% of patients. A third type of outcome measure to evaluate cancer pain treatment consists of measuring patient satisfaction. The Patient Satisfaction Scale measures how satisfied patients are with their treatment.33 The APS29,34 and the AHCPR30 proposed the Patient Satisfaction Scale as one of the criteria by which the success of pain guidelines should be investigated. Studies measuring patient satisfaction showed modest to extremely high proportions (59% to 93%) of patients who were satisfied with pain relief or cancer pain treatment or with the responsiveness of the staff.13-15 Finally, the fourth group of measures consists of pain management indexes, which relate the level of analgesics with the patient's reported level of pain.2 Three types of pain management indexes can be distinguished,3,5,6,20-23 namely, Cleeland's pain management index, in which patients' worst pain intensity is related to pain medication as prescribed by the clinician; Ward's pain management index, in which worst pain intensity is related to the pain medication as used by the patient; and Zelman's pain management index, which consists of a composite pain intensity score related to the pain medication prescribed. Cleeland et al found that 42% of the metastatic cancer patients were prescribed analgesics at less than optimal potency. Other studies that used one of the pain management indexes to evaluate pain treatment showed that 27% to 74% of patients received inadequate cancer pain treatment. It is unknown, however, whether the high variance of patients not receiving adequate pain treatment is due to differences in defining adequate versus inadequate pain medication, heterogeneity of patient groups, or differences in the quality of the pain treatment itself. Pain intensity markers, pain relief scales, patient satisfaction scales, and pain management indexes are used interchangeably for the evaluation of pain treatment,16 which results in substantial differences in reported adequacy of cancer pain treatment. When a pain relief scale or a patient satisfaction scale is used, cancer pain treatment is most frequently evaluated as adequate. However, when a pain intensity marker or pain management indexes are used, most studies show high percentages of inadequate cancer pain treatment. On the basis of the diversity of results, it seems that the commonly used groups of outcome measures cover different aspects of patients' pain experience. Given the lack of systematic use of well-established criteria to evaluate the adequacy of cancer pain treatment, we compared the commonly used outcome measures to evaluate pain treatment in a cancer pain population. The aim was to elucidate whether different measures lead to different results.
Design and Patient Population This study was part of a randomized controlled trial in which the effect of a pain education program was evaluated. Because the focus of the present study is on comparing frequently used measures to evaluate the adequacy of pain treatment in cancer patients, no description of the intervention is reported here. A detailed description of the pain education program and the way patients were randomized is given elsewhere.35 The protocol was approved by the research ethics committee of the hospital. Consenting patients were assigned to either the control group, in which regular pain treatment was provided, or to the experimental group, in which, in addition to the regular pain treatment, the pain education program was given. Both patient groups were randomly assigned after stratification for three variables: sex (male/female), age (< 60 years/  60 years), and metastatic sites (yes/no/unknown). After admission to the hospital, patients who provided informed consent participated in the study. Eligible patients had to meet the following inclusion criteria: pain related to cancer, cancer therapy, or illness; pain duration of at least 1 month; an expected hospital stay of at least 3 days; and life expectancy of at least 3 months after discharge from the hospital (as assessed by the physician). Patients were excluded if they were unable to read and speak Dutch, not accessible by telephone, or residing in a nursing or retirement home.
Procedure
Measures Pain intensity markers. Pain intensity markers are based on patients' pain intensity scores. Pain intensity was measured on an 11-point numeric rating scale that rated patients' present pain intensity, average pain intensity during the last week, and worst pain intensity.36 Pain intensity was assessed by an 11-point numeric rating scale ranging from 0 to 10, in which 0 represents "no pain" and 10 represents "pain as bad as you can imagine." The validity and reliability of the pain intensity scale are well established,27,37-39 and the scale has been found easy to administer, with very few patients unable to complete it. Patients with a pain score of 5 or higher are considered to have substantial pain,12 meaning that these patients are treated inadequately for their pain. In the present study, pain intensity markers were based on patients' present pain intensity (substantial present pain), average pain intensity (substantial average pain), and worst pain intensity (substantial worst pain). Pain relief. With the Pain Relief Scale, patients are asked, "To what extent is the pain relieved by the pain treatment?" The Pain Relief Scale is rated on a 6-point scale with the anchor points "complete relief" and "worsening of pain." Patient scores of 5 to 6 (no relief, worsening of pain) were considered indicators of inadequate treatment, and scores that ranged from 1 to 4 (complete relief, strong relief, moderate relief, and slight relief) were assessed as indicators of adequate treatment.31 Patient satisfaction. The Patient Satisfaction Scale rates how satisfied patients are with the pain treatment on a 5-point scale, from very satisfied (1), somewhat satisfied (2), neither satisfied nor dissatisfied (3), somewhat dissatisfied (4), to very dissatisfied (5). The scores are then classified into inadequate (4 to 5) or adequate (1 to 3) treatment. An evaluation of patient satisfaction with pain management is one component of a total quality assurance program on pain management recommended by the APS.14,40 Pain management indexes. Three pain management indexes were used (Table 1). Cleeland's pain management index2 is based upon the patient's level of worst pain intensity and categorized as 0 (no pain), 1 (1 to 3, mild pain), 2 (4 to 7, moderate pain), or 3 (8 to 10, severe pain). Then, the pain level is subtracted from the most potent level of analgesic drug therapies as prescribed by clinicians, scored as 0 (no analgesic drug), 1 (nonopioids), 2 (a weak opioid), or 3 (a strong opioid). Cleeland's index can range from -3 (a patient receiving no analgesic drugs [score 0] who is in severe pain [score 3]) to +3 (a patient receiving strong opioids [score 3] who is pain-free [score 0]). These scores are then dichotomized: negative scores indicate inadequate orders for analgesic drugs, and scores of 0 or higher are considered to be conservative indicators of acceptable treatment. A variation on Cleeland's index was developed by Ward et al.5,21 Ward's pain management index is fully based on Cleeland's index, except that it incorporates analgesics as used by a patient instead of analgesics as prescribed by the physician. A third pain management index was developed by Zelman et al.22 Zelman's index consists of a pain intensity score that is computed by averaging the patient's present pain intensity, average pain intensity, and worst pain intensity. The composite pain score is then categorized into 0 (no pain), 1 (1 to 3, mild pain), 2 (4 to 6, moderate pain), or 3 (7 to 10, severe pain) and related to medication used. Evidence of the validity of the pain management indexes has been demonstrated by Ward et al.
Statistical Methods
Of the 383 patients asked to participate in the study, 313 (82%) agreed. Patients who declined (n = 70) found the study too burdensome (n = 48), were not motivated (n = 15), or were too ill (n = 7). Patients who declined were significantly older than those who participated (P < .05), and women refused to participate more frequently than men (P < .05). Of the participating patients, 159 patients (51%) received the intervention and 154 patients (49%) did not. At follow-up, 89% of the patients completed the posttest at 2 weeks after discharge, 83% completed the posttests at 2 and 4 weeks after discharge, and 75% completed all three posttests up to 8 weeks after discharge. Dropouts were due to death (60%), being too ill (21%), experiencing the study as too burdensome to continue (14%), or lack of motivation (5%). Of all patients, 63% were women, and the mean age was 56 years (SD, 12). From the total group, 41% had primary school or lower vocational training, 33% had middle- to low-level secondary school or intermediate vocational training, and 26% had high-level secondary school training or were college or university graduates. For 33% of the patients, district nursing was planned after discharge. The major diagnoses were breast cancer (30%), cancer of the genitourinary organs (24%), and bone, connective tissue or skin cancer (14%). In total, 58% had metastases; the extent of disease was regional in 19% and local in 14%, and the tumor status was unknown or not applicable in 8%. Bone, liver, lung, and lymphoma metastases were reported most frequently. In 36% of the patients, chemotherapy was the main therapy, 16% received surgery, 13% received radiation therapy, 7% received hormonal therapy, and 7% received another therapy or the therapy was still not known at pretest. In 22% of the patients, no anticancer treatment was given. Pain characteristics are listed in Table 2. Pain in the abdominal and lower back regions was experienced most frequently, in 40% and 39% of patients, respectively. The mean pain duration was 14 months (SD, 33 months; range, 1 to 324 months), and Pain lasted 5 years or longer in 16 patients (4%). Pain was associated with direct tumor involvement in 77% and with the cancer therapy in 24%. Patients reported on average 1.8 pain locations. Seven percent of the patients reported that they experienced pain constantly, 73% reported that pain was intermittent without pain-free periods, 19% reported that pain was intermittent with pain-free periods, and 7% reported that pain was constant; the pattern of pain was unknown in 1%. At baseline, patients' mean present pain intensity score was 3.3 (SD, 2.3), the mean average pain intensity was 4.9 (SD, 2.1), and the mean worst pain intensity was 7.7 (SD, 2.1). At 2 weeks after discharge, patients' present pain intensity, average pain intensity, and worst pain intensity were significantly decreased and stayed fairly constant at 4 and 8 weeks after discharge.
Adequacy of Pain Treatment Results of the four types of outcome measures at pretest are shown in Fig 1. The percentage of patients who were treated inadequately ranged from 16% to 91%. The enormous difference in inadequacy of cancer pain treatment depended on the scale used. When substantial worst pain was measured, approximately nine out of every 10 patients were assessed as receiving inadequate pain treatment; in contrast, use of the Pain Relief Scale resulted in fewer than two out of every 10 patients being treated inadequately.
After discharge, the highest percentage of inadequate pain treatment was again shown by the Substantial Worst Pain Scale (60% to 91%), whereas use of the Pain Relief Scale resulted in the lowest percentages of inadequate cancer pain treatment (16% to 26%) (data not shown). It is evident that the choice of measure, rather than the pain treatment itself, largely determines the proportion of inadequately treated patients. The percentage of agreement, assessed by dichotomizing all measures and not corrected for chance, ranged from 23% to 96% at the four assessment points. The highest percentage of agreement was found between the three pain management indexes (80% to 96%), followed by agreement between the Pain Relief Scale and the Patient Satisfaction Scale (78% to 83%). Agreement between the other measures was extremely low to moderate. Except for the high agreement between the three pain management indexes, one measure cannot be substituted for another measure because they seem to measure different aspects. Because the percentage of agreement can be relatively high by chance alone when dichotomized data are analyzed, agreement between the scales was also calculated by kappa statistics. High agreement (kappa from 0.81 to 1.00) was found in only 3%, namely, between Cleeland's pain management index and Ward's. Substantial agreement (kappa from 0.61 to 0.80) was found in 11%, between Cleeland's index and Zelman's index, between Ward's index and Zelman's index, and between substantial present pain and substantial average pain. Moderate agreement (kappa of 0.41 to 0.60) was found in 11%, 39% showed fair agreement (kappa of 0.21 to 0.40), and 36% had slight or poor agreement (kappa < 0.20). To investigate whether one or more underlying constructs were emerging, all eight measures were subjected to factor analyses with oblique rotation (Table 3). An examination of the eigenvalues suggested a three-factor structure. Across the four assessment points, three factors accounted for 78% to 83% of the variance. At pretest, factor 1 accounted for 45% of variance. The first factor consisted of the three pain management indexes and substantial worst pain. Substantial worst pain, however, also loaded on the second factor. At pretest, factors 2 and 3 accounted for 20% and 13%, respectively. At 2, 4, and 8 weeks after discharge, the first factor consisted of the three pain management indexes, the second factor consisted of the Pain Relief Scale and the Patient Satisfaction Scale, and all pain intensity markers loaded on a third factor. It can be concluded that a degree of consistency in the results of the factor analyses was present, showing the same three factors at different time points.
Outcome measures are important to describe the effectiveness of the pain treatment. Because no single outcome measure has received widespread acceptance for the evaluation of cancer pain treatment, several outcome measures are described in the literature. We compared four types of outcome measures that are used frequently in the literature to evaluate pain treatment. Adequacy of cancer pain treatment can be based on either patient-oriented measures (eg, pain intensity markers and pain relief and patient satisfaction scales) or a pain score related to the pain medication as provided by clinicians (eg, pain management indexes). Our results show that the percentage of missing data was low to moderate; the pain management indexes had the highest percentages of missing data. In the hospital, some patients had difficulty reporting what pain medication was prescribed; in contrast, patients at home were better informed. When the different outcome measures were compared simultaneously, the proportion of chronic cancer pain patients treated inadequately was shown to range from 16% to 91%. This striking variance in adequacy shows that the choice of measure largely determines the proportion of inadequately treated pain patients, not the pain treatment itself. Except for the high agreement between the three pain management indexes, which could be expected because of some overlaps, agreement between all other measures was moderate to extremely low. Factor analyses revealed a high degree of association between the variables at the four assessment points. The starting point was four groups of outcome measures, but three factors were found. The three factors clearly emerged, explaining most of the variability in the different scales. The results provide support for the conclusion that the different types of outcome measures do not measure a single underlying construct. Each of the different measures seems to have its own advantages and disadvantages for the evaluation of pain treatment. Pain intensity markers, based on pain intensity scores, are most frequently used in pain research. The APS Quality of Care Committee34 has advocated the use of a value of 5 or greater on a numeric rating scale as a starting point that will elicit an immediate assessment of pain treatment. Although valuable when evaluating change over time or comparing groups, this measure has the limitation that a pain score of 5 or greater may have different meanings for different patients because the anchor points might be interpreted differently. To avoid this problem, the Pain Relief Scale can be used so patients can indicate whether the pain remained the same, was relieved, or worsened. However, this measure has the limitation that change in pain intensity alone is not a direct indication of adequacy.26 Pain treatment is evaluated as adequate when patients' pain is decreased, even when there is continuation of substantial pain. Consequently, the Pain Relief Scale seems to measure change over time rather than adequacy of pain treatment. At first glance, use of the Patient Satisfaction Scale seems to overcome this problem. However, this scale is subject to distortion because patients can be in severe pain and nevertheless assess their satisfaction with the pain treatment as high.34 Data from the present study show satisfaction ratings that are skewed toward the high end of the scale. A majority of patients (68%) who were somewhat or very satisfied reported moderate to severe pain at baseline. These results are in concurrence with other studies on patient satisfaction with pain treatment, underscoring the critical concern about the appropriateness of patient satisfaction as an outcome measure.14,15,36,42 Moreover, factors other than patients' reports of satisfaction with their pain treatment may be subject to social desirability: when patients do not know what quality of pain treatment to expect, they will not necessarily be dissatisfied when their pain is severe. This suggests the need for caution in using patient satisfaction alone to monitor quality of pain treatment. The fourth type of outcome measure consisted of relating patients' self-report of pain with the pain treatment itself. Ideally, when evaluating pain treatment, various aspects related to pain should be considered, such as diagnosis, tumor extension, cancer treatment, drug choice, dosage, dosing intervals, duration of treatment, adverse reactions, and clarity of instruction. Although this method of evaluating pain treatment has shown satisfactory results in some medical fields,43,44 Dorrepaal et al3,45 found low percentages of agreement between physician raters regarding the adequacy of cancer pain treatment. A more simplified method relating patients' self-report of pain with the pain treatment was developed by Cleeland et al.2 The pain management indexes evaluated in the present study are all based on the most potent analgesic. Consequently, less potent analgesics are not taken into account. Pain medication in advanced cancer patients, however, usually consists of more than one pain medication. Furthermore, all pain management indexes are based on the World Health Organization's analgesic ladder, which does not take into account the dose, schedule, or route of administration for the individual patient. A final disadvantage of the pain management indexes is that categorizing pain intensity into mild, moderate, and severe might not fully depict pain as a subjective experience because it does not take into account what the patient experiences to be tolerable. It is evident that the different measures generate different results. Although the AHCPR30 and the APS46 recommend using pain intensity markers and patient satisfaction as outcome measures, the results of the present study reveal an urgent need for identifying one or more valid and reliable outcome measures to evaluate pain treatment in cancer patients. So far, pain treatment is primarily evaluated unidimensionally. In this study, we compared outcome measures representing the sensory and behavioral dimensions. Because pain is a multidimensional concept, the physiologic, sensory, affective, cognitive, behavioral, and sociocultural dimensions should be integrated to evaluate pain treatment. Further psychometric analyses are needed to identify instruments to evaluate pain treatment in patients with cancer in a variety of situations. However, as long as the various outcomes are not consistent within one dimension, a unidimensional instrument must first be identified.
Supported by grant no. NKI 92-469 from the Dutch Cancer Society We thank the patients who participated in this study and extend special thanks to nurses of the Antoni van Leeuwenhoek Hospital, Amsterdam.
1. Bonica JJ: Treatment of cancer pain: Current status and future needs, in Fields HL, Dubner F, Cervero F (eds): Advances in Pain Research and Therapy. New York, NY, Raven Press, 1985, pp 589-616
2.
Cleeland CS, Gonin R, Hatfield AK, et al: Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 330:592-596, 1994 3. Dorrepaal KL: Pijn bij patienten met kanker [dissertation in Dutch]. Amsterdam, the Netherlands, Vrije Universiteit, 1989 4. Dalton JA: Nurses' perceptions of their pain assessment skills, pain management practices, and attitudes toward pain. Oncol Nurs Forum 16:225-231, 1989[Medline] 5. Ward SE, Hernandez L: Patient-related barriers to management of cancer pain in Puerto Rico. Pain 58:233-238, 1994[Medline]
6.
Larue F, Colleau SM, Brasseur L, et al: Multicentre study of cancer pain and its treatment in France. BMJ 310:1034-1037, 1995 7. Grond S, Zech D, Schug SA, et al: Validation of World Health Organization guidelines for cancer pain relief during the last days and hours of life. J Pain Symptom Manage 6:411-422, 1991[Medline] 8. Grond S, Zech D, Lynch J, et al: Validation of World Health Organization guidelines for pain relief in head and neck cancer: A prospective study. Ann Otol Rhinol Laryngol 102:342-348, 1993[Medline] 9. Ventafridda V, Tamburini M, Caraceni A, et al: A validation study of the WHO method for cancer pain relief. Cancer 59:850-856, 1987[Medline] 10. Walker VA, Hoskin PJ, Hanks GW, et al: Evaluation of WHO analgesic guidelines for cancer pain in a hospital-based palliative care unit. J Pain Symptom Manage 3:145-149, 1988[Medline] 11. Zech DF, Grond S, Lynch J, et al: Validation of World Health Organization guidelines for cancer pain relief: A 10-year prospective study. Pain 63:65-76, 1995[Medline] 12. Serlin RC, Mendoza TR, Nakamura Y, et al: When is cancer pain mild, moderate or severe? Grading pain severity by its interference with function. Pain 61:277-284, 1995[Medline] 13. Ward S, Gatwood J: Concerns about reporting pain and using analgesics: A comparison of persons with and without cancer. Cancer Nurs 17:200-206, 1994[Medline] 14. Miaskowski C, Nichols R, Brody R, et al: Assessment of patient satisfaction utilizing the American Pain Society's quality assurance standards on acute and cancer-related pain. Clin J Pain 9:5-11, 1994 15. Zhukovsky DS, Gorowski E, Hausdorff J, et al: Unmet analgesic needs in cancer patients. J Pain Symptom Manage 10:113-119, 1995[Medline]
16.
Jadad AR, Browman GP: The WHO analgesic ladder for cancer pain management: Stepping up the quality of its evaluation. JAMA 274:1870-1873, 1995 17. Takeda F: Results of field-testing in Japan of the WHO draft interim guidelines on relief of cancer pain. The Pain Clinic 1:83-89, 1986 18. Goisis A, Gorini M, Ratti R, et al: Application of a WHO protocol on medical therapy for oncologic pain in an internal medicine hospital. Tumori 75:470-472, 1989[Medline] 19. Syrjala KL, Chapman CR: Measurement of clinical pain: A review and integration of research findings, in Benedetti C (ed): Advances in Pain Research and Therapy. New York, NY, Raven Press, 1984 20. Lin CC, Ward SE: Patient-related barriers to cancer pain management in Taiwan. Cancer Nurs 18:16-22, 1995[Medline] 21. Ward SE, Goldberg N, Miller-McCauley V, et al: Patient-related barriers to management of cancer pain. Pain 52:319-324, 1993[Medline] 22. Zelman DC, Cleeland CS, Howland EW: Factors in appropriate pharmacological management of cancer pain: A cross-institutional investigation. Pain 4:S136, 1987 (suppl)
23.
Cleeland CS, Gonin R, Baez L, et al: Pain and treatment of pain in minority patients with cancer. Ann Intern Med 127:813-816, 1997 24. Jamison RN, Brown GK: Validation of hourly pain intensity profiles with chronic pain patients. Pain 45:123-128, 1991[Medline] 25. Jensen MP, Turner LR, Turner JA, et al: The use of multiple-item scales for pain intensity measurement in chronic pain patients. Pain 67:35-40, 1996[Medline] 26. De Conno F, Caraceni A, Gamba A, et al: Pain measurement in cancer patients: A comparison of six methods. Pain 57:161-166, 1994[Medline] 27. Jensen MP, Karoly P, Braver S: The measurement of clinical pain intensity: A comparison of six methods. Pain 27:117-126, 1986[Medline] 28. Price DD, Bush FM, Long S, et al: A comparison of pain measurement characteristics of mechanical visual analogue and simple numerical rating scales. Pain 56:217-226, 1994[Medline] 29. American Pain Society (APS): Principles of Analgesic Use in the Treatment of Acute Pain and Chronic Cancer Pain: A Concise Guide to Medical Practice. Skokie, IL, American Pain Society, 1992 30. Agency for Health Care Policy and Reserach (AHCPR): Management of cancer pain: Clinical practice guideline. Rockville, MD, AHCPR, 1994 31. Cleeland CS: Pain assessment in cancer, Osoba D (ed):Effect of Cancer on Quality of Life294-305Boca Raton, FL, CRC Press, 1991 32. Wallenstein SL: Scaling clinical pain and pain relief, Bromm B (ed):Pain Measurement in Man: Neurophysiological Correlates of Pain389-396Amsterdam, the Netherlands, Elsevier Science Publishers, 1984 33. Ware J, Snyder M, Wright W, et al: Defining and measuring patient satisfaction with care. Eval Prog Plan 6:247-263, 1983[Medline]
34.
American Pain Society Quality of Care Committee: Quality improvement guidelines for the treatment of acute pain and cancer pain. JAMA 274:1874-1880, 1995 35. De Wit R, Van Dam F, Zandbelt L, et al: A pain education program for chronic cancer patients: Follow-up results from a randomized controlled trial. Pain 73:55-69, 1997[Medline] 36. Ward SE, Gordon D: Application of the American Pain Society quality assurance standards. Pain 56:299-306, 1994[Medline] 37. McGuire DB: Measuring pain, in Frank-Stromborg M (ed): Instruments for Clinical Nursing. Norwalk, CT, 1988, pp 333-356 38. Kremer E, Atkinson JH Jr: Pain measurement: Construct validity of the affective dimension of the McGill Pain Questionnaire with chronic benign pain patients. Pain 11:93-100, 1981[Medline] 39. Syrjala KL: The measurement of pain, in McGuire DB, Yarbro CH (eds): Cancer Pain Management. Philadelphia, PA, WB Saunders, 1987, pp 133-150 40. Ward SE, Gordon DB: Patient satisfaction and pain severity as outcomes in pain management: A longitudinal view of one setting's experience. J Pain Symptom Manage 11:242-251, 1996[Medline] 41. Fleiss JL: Large sample standard errors of kappa and weighted kappa. Psycol Bull 72:323-327, 1969 42. Dietrick-Gallagher M, Polomano R, Carrick L: Pain as a quality management initiative. J Nurs Care Qual 9:30-42, 1994[Medline] 43. Hanlon JT, Schmader KE, Samsa GP, et al: A method for assessing drug therapy appropriateness. J Clin Epidemiol 45:1045-1051, 1992[Medline] 44. Chrischilles EA, Helling DK, Aschoff CR: Effect of clinical pharmacy services on the quality of family practice physician prescribing and medication costs. Drug Intell Clin Pharm 23:417-421, 1989 45. Dorrepaal KL, Aaronson NK, van Dam FS: Pain experience and pain management among hospitalized cancer patients: A clinical study. Cancer 63:593-598, 1989[Medline] 46. Max M: American Pain Society quality assurance standards for relief of acute pain and cancer pain, in Bond M, Charlton J, Woolf C (eds): Proceedings of the VIth World Congress on Pain. Amsterdam, the Netherlands, Elsevier, 1991, pp 186-189 Submitted October 13, 1998; accepted December 9, 1998.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 tests, and Student t tests. P < .05 was considered significant. Three approaches were taken to investigate the agreement between the measures. First, results were analyzed by means of overall agreement, ie, the proportion of judgments in which two dichotomized measures, uncorrected for change, agree. Second, results were analyzed by means of Cohen's kappa, which is a measure of agreement corrected for chance.
