Volume 39, Issue 2 , Pages 259-267, February 2010
Retrospective Assessment of Cancer Pain Management in an Outpatient Palliative Radiotherapy Clinic Using the Pain Management Index
Article Outline
Abstract
Context
The Pain Management Index (PMI) is a simple index linking the usual severity of cancer pain with the category of medication prescribed to treat it. Medication categories are derived from the World Health Organization's “analgesic ladder” approach to cancer pain, and the PMI is an indicator of the extent to which the medication prescribed corresponds to the recommended categories for mild, moderate, and severe pain.
Objectives
The aim of this study was to assess prevalence of inadequate pain management in an outpatient palliative radiotherapy clinic using the PMI.
Methods
All patients with bone metastases referred for palliative radiotherapy from 1999 to 2006 were retrospectively analyzed for patient-rated pain scores (0–10 scale) and analgesic consumption. Pain scores were assigned 0, 1, 2, and 3 when patients reported no pain (0), mild (1–4), moderate (5–6), or severe pain (7–10), respectively. Analgesic scores of 0, 1, 2, and 3 were assigned when patients were prescribed no pain medication, nonopioids, “weak” opioids, and “strong” opioids, respectively. The PMI score was calculated by subtracting the pain score from the analgesic score. A negative PMI score was considered an indicator of potentially inadequate pain management by the prescriber. Descriptive statistics, Pearson's r correlation, and univariate and multivariate logistic regression analysis were used to determine the relationship of PMI over time, and the relationship with predictive factors.
Results
One thousand patients were included from January 1999 to December 2006. A negative PMI was calculated for 25.8% of patients at initial consultation. Prevalence of negative PMI significantly increased over years (P
<0.0001). Higher Karnofsky Performance Status (P<0.0001) and breast primary cancer site (P<0.0001) were significantly associated with negative PMI after adjusting for year variable.
Conclusion
Despite publication of numerous cancer pain management guidelines, undermedication appears to be a persistent problem for patients with painful bone metastases referred for radiotherapy.
Key Words: Bone metastases, cancer pain, Pain Management Index
Introduction
The most recurrent and debilitating symptom of metastatic cancer is pain, and this can manifest in the early stages of the disease.1, 2 While more than half of cancer patients report experiencing some level of pain,2, 3, 4, 5, 6, 7, 8 this can increase to three-quarters of patients when advanced cancer is included. Despite a general consensus among health care professionals that 90% of cancer patients can obtain sufficient pain relief with analgesics, many patients do not have adequate pain relief as a result of inadequate treatment.9, 10 In fact, 17%–70% of cancer patients have reported experiencing severe levels of pain.2, 3, 6, 7, 8, 9, 11, 12, 13 Consequently, this may negatively impact a patient's quality of life.2, 3
The World Health Organization (WHO) developed guidelines for the treatment of cancer pain in 1986 (revised in 1996), which were aimed at decreasing the prevalence of inadequate analgesia.14 The guidelines include suggestions about the type of analgesic that should be prescribed in the setting of pain that is generally mild, moderate, or severe. For mild pain or at the initial outbreak of pain, patients should receive at least a nonsteroidal anti-inflammatory drug (NSAID) or acetaminophen. If the pain persists or increases to a moderate level, the patient should be prescribed a so-called “weak” opioid (e.g., codeine). If the pain is severe or inadequately treated by “weak” opioids, a so-called “strong” opioid (e.g., morphine, hydromorphone, oxycodone, fentanyl) is recommended.14
The Pain Management Index (PMI) was developed by Cleeland et al.2 as a measure of treatment adequacy for cancer pain. It is claimed to be a conservative measure based on the Agency for Health Care Policy and Research (AHCPR) and WHO guidelines. A review by Deandrea et al.15 reported on studies that measured undermedication within the cancer population. Of the published studies, 60% of publications used Cleeland's PMI tool to determine their results on undermedication.
The primary objective of our investigation was to use the PMI to calculate the prevalence of inadequate cancer pain management in a cohort of Canadian cancer patients who arrived for their first palliative radiotherapy consultation for metastatic bony pain, in a Rapid Response Radiotherapy Program (RRRP). The secondary objective was to compare our results to other reported PMI levels with other similar study populations globally.
Methods and Materials
Patient data were collected retrospectively from a prospective clinical database for patients who are referred to the RRRP. The RRRP is dedicated to provide timely access for palliative radiotherapy to patients in the Greater Toronto Area. Patients are referred to the program by community medical oncologists, palliative care physicians, and general practitioners, as well as by other oncologists and palliative care physicians from within the Sunnybrook Odette Cancer Centre. Inclusion criteria included patients with radiographic evidence of bone metastases, English speaking, and older than 18 years. Ethics approval was obtained from the hospital Research Ethics Board.
The database corresponded to the years 1999–2006, excluding 2002, as data were not captured for that year. The data extracted from these databases were patient age, sex, primary cancer site, worst pain score on the Brief Pain Inventory (BPI) or any other pain score from other measures (e.g., Edmonton Symptom Assessment Scale [ESAS]), and the type of analgesic medication prescribed to the patient before initial consultation.
A pain score of “0” defined an absence of pain, “1” was for mild pain, “2” for moderate pain, and “3” for severe pain. These pain scores corresponded to the ESAS and BPI worst pain score categorization of “0” as an absence of pain, “1–4” for mild pain, “5–6” for moderate pain, and “7–10” for severe pain.15, 16 A patient's analgesic score on the PMI was calculated based on the analgesic prescribed by the physician. No prescribed analgesic was scored as “0,” a nonopioid (i.e., NSAID or acetaminophen) was “1,” a “weak” opioid (e.g., codeine) was “2,” and a “strong” opioid (e.g., morphine, hydromorphone, oxycodone, fentanyl) was scored as “3”.17 Table 1 illustrates the analgesics and NSAIDs used by our patients, along with the assigned PMI score for each drug.
Table 1. List of Medications and Their Pain Management Index Scores
| Analgesic Type | Strength | Analgesic Score |
|---|---|---|
| Dexamethasone | None | 0 |
| Gabapentin | None | 0 |
| Lyrica® | None | 0 |
| Nortriptyline | None | 0 |
| Acetaminophen | NSAID | 1 |
| Advil® | NSAID | 1 |
| Arthrotec® | NSAID | 1 |
| Asa EC | NSAID | 1 |
| Celebrex® | NSAID | 1 |
| Extra strength Tylenol® | NSAID | 1 |
| Ibuprofen | NSAID | 1 |
| Ketamine | NSAID | 1 |
| Ketoprofen | NSAID | 1 |
| Naproxen | NSAID | 1 |
| Prednisone | NSAID | 1 |
| Toradol® | NSAID | 1 |
| Trilisate® | NSAID | 1 |
| Tylenol plain® | NSAID | 1 |
| Tylenol—ES® | NSAID | 1 |
| Codeine | Weak | 2 |
| Codeine Contin® | Weak | 2 |
| Propoxyphene | Weak | 2 |
| Talwin® | Weak | 2 |
| Tylenol #2® | Weak | 2 |
| Tylenol #3® | Weak | 2 |
| Tylenol #4® | Weak | 2 |
| Tramacet® | Weak | 2 |
| Demerol® | Strong | 3 |
| Dilaudid® | Strong | 3 |
| Duragesic® | Strong | 3 |
| Endocet® | Strong | 3 |
| Fentanyl | Strong | 3 |
| Hydrocodone Bitartrate/Hycodan® | Strong | 3 |
| Hydrocodone | Strong | 3 |
| Hydromorph Contin® | Strong | 3 |
| Hydromorphone | Strong | 3 |
| Hydromorphone HCL | Strong | 3 |
| Kadian® | Strong | 3 |
| M Eslon® | Strong | 3 |
| Morphine | Strong | 3 |
| MS CONTIN® | Strong | 3 |
| Oxycocet® | Strong | 3 |
| Oxycodone | Strong | 3 |
| Oxycontin® | Strong | 3 |
| Percocet® | Strong | 3 |
| Statex® | Strong | 3 |
| Methadone | Strong | 3 |
Only the patient's first visit was included and all subsequent visits during the study period were excluded. When patients had difficulty pinpointing a particular pain score and provided a range of scores, for example, 8–10 out of 10, only the worst pain score was used to calculate the PMI. For patients on multiple analgesic regimens of varying types, the most potent analgesic as per PMI definition was used for the calculation. The PMI was then determined by subtracting the worst pain score from the analgesic score.2 Table 2 describes the scoring system. Patients with negative PMI scores were classified as receiving inadequate analgesic treatment for their pain.
Table 2. Pain Management Index Scoring
| Analgesic Type | Intensity of Pain | |||
|---|---|---|---|---|
| None (0) | Mild (1) | Moderate (2) | Severe (3) | |
| No prescribed analgesic (0) | 0 | −1 | −2 | −3 |
| Nonopioid (1) | 1 | 0 | −1 | −2 |
| “Weak” opioid (2) | 2 | 1 | 0 | −1 |
| “Strong” opioid (3) | 3 | 2 | 1 | 0 |
Descriptive statistics for patient characteristics were provided as medians and ranges for continuous variables, e.g., age, and as proportions for categorical variables, e.g., primary cancer site. Pearson's r correlation coefficient was calculated between negative PMI at presentation and age, Karnofsky Performance Status (KPS), sex, and primary cancer site. Univariate logistic regression analysis of negative PMI was used to determine the relationship over time, and the relationship with the predictive factors after adjusting for “year” variable. Odds ratio (OR) and the 95% confidence intervals (CI) of OR were estimated. Furthermore, a backward stepwise selection procedure of multivariate logistic regression analysis was performed to search for any association between significant variables and negative PMI after adjusting for year variable.
Results
Between January 1999 and December 2006, 1,000 patients with bone metastases were seen in consultation at the RRRP. In the years 1999, 2000, 2001, 2003, 2004, 2005, and 2006, there were 223, 223, 159, 94, 124, 107, and 70 patients analyzed, respectively.
The ratio of males to females was approximately equal, with 565 (56.5%) males and 435 (43.5%) females. The median age of patients was 68 years (range 28–95). The most common primary cancer sites were the lung, breast, and prostate, affecting 24.7%, 24.3%, and 21.9% of patients, respectively. The median performance status of patients, as measured by the KPS score, was 70 (range 10–100). The demographic information can be found in Table 3.
Table 3. Patients Seen in the RRRP from January 1999 to December 2006 (n=1,000)
| Sex | |
| Male, n (%) | 565 (56.5) |
| Female, n (%) | 435 (43.5) |
| Age (years) | |
| Mean±SD | 66.8±11.8 |
| Median (range) | 68 (28–95) |
| Karnofsky Performance Status | |
| Mean±SD | 64.5±14.7 |
| Median (range) | 70 (10–100) |
| Pain score | |
| Mean±SD | 6.3±2.8 |
| Median (range) | 7 (0–10) |
| Primary cancer site, n (%) | |
| Breast | 243 (24.3) |
| Lung | 247 (24.7) |
| Prostate | 219 (21.9) |
| Gastrointestinal | 78 (7.8) |
| Others | 213 (21.3) |
At the initial RRRP consultation, 25.4% (254/1,000) of patients reported experiencing mild pain, 19.6% (196/1,000) reported moderate pain and 46.5% (465/1,000) reported experiencing severe pain (Fig. 1). Additionally, 14.4% (144/1,000) of patients were prescribed no pain medication when they arrived at the RRRP for a radiation consultation; 11.6% of patients (116/1,000) were prescribed nonopioids, 15.5% (155/1,000) “weak” opioids, and 58.5% (582/1,000) “strong” opioids on presentation (Fig. 2). Inadequate analgesic pain management, which was represented by a negative PMI score, was found in 25.8% (258/1,000) of all patients at initial consultation (Fig. 3, Table 4).
Fig. 1
Proportion of patients with mild, moderate, and severe pain at initial consultation and one-month, two-month, and three-month follow-up.
Fig. 2
Proportion of patients with no analgesics, nonopioids, “weak” opioids, and “strong” opioids at initial consultation and one-month, two-month, and three-month follow-up.
Fig. 3
Proportion of patients with negative PMI at initial consultation and at one-month, two-month, and three-month follow-up.
Table 4. PMI Scoring Results
| Analgesic Intake | No Pain (PMI score), # of patients | Mild Pain (PMI score), # of patients | Moderate Pain (PMI score), # of patients | Severe Pain (PMI score), # of patients | Total (# of patients), % of patients |
|---|---|---|---|---|---|
| 1999 | |||||
| No pain medication | (0) 11 | (−1) 12 | (−2) 6 | (−3) 8 | 37 (16.6) |
| Nonopioids | (1) 1 | (0) 5 | (−1) 2 | (−2) 5 | 13 (5.8) |
| Weak opioids | (2) 2 | (1) 10 | (0) 14 | (−1) 25 | 51 (22.9) |
| Strong opioids | (3) 3 | (2) 30 | (1) 18 | (0) 71 | 122 (54.7) |
| Total, n (%) | 17 (7.6) | 57 (25.6) | 40 (17.9) | 109 (48.9) | 223 |
| 2000 | |||||
| No pain medication | (0) 5 | (−1) 16 | (−2) 7 | (−3) 5 | 33 (14.8) |
| Nonopioids | (1) 1 | (0) 11 | (−1) 7 | (−2) 12 | 31 (13.9) |
| Weak opioids | (2) 4 | (1) 7 | (0) 6 | (−1) 11 | 28 (12.6) |
| Strong opioids | (3) 5 | (2) 33 | (1) 35 | (0) 58 | 131 (58.7) |
| Total, n (%) | 15 (6.7) | 67 (30.0) | 55 (24.7) | 86 (38.6) | 223 |
| 2001 | |||||
| No pain medication | (0) 2 | (−1) 11 | (−2) 1 | (−3) 2 | 16 (10.1) |
| Nonopioids | (1) 2 | (0) 4 | (−1) 3 | (−2) 4 | 13 (8.2) |
| Weak opioids | (2) 5 | (1) 7 | (0) 6 | (−1) 14 | 32 (20.1) |
| Strong opioids | (3) 4 | (2) 20 | (1) 28 | (0) 46 | 98 (61.6) |
| Total, n (%) | 13 (8.2) | 42 (26.4) | 38 (23.9) | 66 (41.5) | 159 |
| 2003 | |||||
| No pain medication | (0) 1 | (−1) 1 | (−2) 0 | (−3) 0 | 2 (2.1) |
| Nonopioids | (1) 0 | (0) 3 | (−1) 5 | (−2) 11 | 19 (20.2) |
| Weak opioids | (2) 0 | (1) 0 | (0) 1 | (−1) 15 | 16 (17.0) |
| Strong opioids | (3) 0 | (2) 3 | (1) 9 | (0) 45 | 57 (60.7) |
| Total, n (%) | 1 (1.1) | 7 (7.4) | 15 (16.0) | 71 (75.5) | 94 |
| 2004 | |||||
| No pain medication | (0) 0 | (−1) 3 | (−2) 4 | (−3) 8 | 15 (12.1) |
| Nonopioids | (1) 0 | (0) 1 | (−1) 5 | (−2) 15 | 21 (16.9) |
| Weak opioids | (2) 0 | (1) 1 | (0) 2 | (−1) 9 | 12 (9.7) |
| Strong opioids | (3) 0 | (2) 8 | (1) 12 | (0) 56 | 76 (61.3) |
| Total, n (%) | 0 (0.0) | 13 (10.5) | 23 (18.5) | 88 (71.0) | 124 |
| 2005 | |||||
| No pain medication | (0) 0 | (−1) 4 | (−2) 3 | (−3) 11 | 18 (16.8) |
| Nonopioids | (1) 0 | (0) 5 | (−1) 2 | (−2) 4 | 11 (10.5) |
| Weak opioids | (2) 0 | (1) 5 | (0) 4 | (−1) 7 | 16 (15.0) |
| Strong opioids | (3) 0 | (2) 8 | (1) 12 | (0) 42 | 62 (57.9) |
| Total, n (%) | 0 (0.0) | 22 (20.6) | 21(19.6) | 64 (59.8) | 107 |
| 2006 | |||||
| No pain medication | (0) 0 | (−1) 5 | (−2) 6 | (−3) 10 | 21 (30.0) |
| Nonopioids | (1) 0 | (0) 1 | (−1) 3 | (−2) 3 | 7 (10.0) |
| Weak opioids | (2) 0 | (1) 2 | (0) 0 | (−1) 4 | 6 (8.6) |
| Strong opioids | (3) 0 | (2) 8 | (1) 1 | (0) 27 | 36 (51.4) |
| Total, n (%) | 0 (0.0) | 16 (22.8) | 10 (14.3) | 44 (62.9) | 70 |
| Cumulative total, n (%) | 46 (4.6) | 224 (22.4) | 202 (20.2) | 528 (52.8) | 1,000 (100) |
PMI and Patient Demographics
Using a univariate logistic regression model of negative PMI, the relationship with demographic factors was performed. When we adjusted for year, higher performance status (i.e., higher KPS [P<0.0001; OR=1.03]) was highly related to negative PMI. Furthermore, patients with breast primary cancer site had 2.09 times higher probabilities of experiencing inadequate pain management than patients with other primary cancer histologies (P<0.0001; OR=2.09). Other variables were not significant in the analysis. A backward stepwise selection procedure of multivariate logistic regression analysis also confirmed only two covariates to be significantly associated with negative PMI after adjusting for year variable, namely, KPS and breast primary cancer site.
PMI from 1999 to 2006
The proportions of patients with negative PMI scores, representing inadequate pain management, for the years 1999, 2000, 2001, 2003, 2004, 2005, and 2006 were 14.8% (33/223), 26% (58/223), 22% (35/159), 34% (32/94), 35% (44/124), 29% (31/107), and 44% (31/70), respectively.
The prevalence of negative PMI significantly increased over years (P<0.0001), as evaluated using univariate logistic regression analysis. Therefore, the proportion of patients experiencing inadequate analgesic pain management at initial consultation at the RRRP significantly increased from 1999 to 2006.
PMI at Follow-Up
At one-month follow-up (FU1), we were able to gather data from 52% of the initial patient sample. At this time, 39.7% of these patients reported experiencing mild levels of pain, 15.2% reported moderate pain, and 19.4% reported experiencing severe pain. In terms of analgesic intake, 15.4% of patients were prescribed no pain medication, 12.4% were prescribed nonopioids, 15.2% were prescribed “weak” opioids, and 56.6% were prescribed “strong” opioids.
At the two-month follow-up (FU2), we were able to gather data from 42% of the initial patient population. At this time, mild, moderate, and severe pain levels were reported by 42.1%, 14.5%, and 14.7% of patients, respectively. The analgesic intake pattern included 15.2% of patients who were prescribed no pain medication, 12.3% nonopioids, 17.6% “weak” opioids, and 54.7% “strong” opioids.
At the three-month follow-up (FU3), data were available from 36% of the initial sample of patients. Mild, moderate, and severe pain levels were reported by 36.2%, 15.1%, and 18.5% of patients, respectively. In terms of analgesic consumption, 16.8% of patients were prescribed no pain medication, 12.3% were prescribed nonopioids, 18.5% were prescribed “weak” opioids, and 52.2% were prescribed “strong” opioids.
The proportion of patients with mild, moderate, and severe pain at initial consultation, FU1, FU2, and FU3 can be seen in Fig. 1. The proportion of patients with no analgesic treatment, nonopioid, weak opioid, and strong opioid prescriptions at initial consultation, FU1, FU2, and FU3 is represented in Fig. 2.
The proportion of patients with inadequate analgesic pain management, as reported by the PMI score, was 17.53% at FU1, 16.19% at FU2, and 15.45% at FU3 (Fig. 3). Because of the high attrition rates, we did not compare the PMI at baseline with those at follow-ups.
Discussion
The prevalence of inadequate analgesic regimens for pain associated with bone metastases in the RRRP has been previously reported. Kirou-Mauro et al.18 developed a definition of undermedication for the patients who experienced moderate or severe levels of pain and were prescribed either no medication, nonopioid, or “weak” opioid analgesic medication. Forty percent (29%–48%) of patients were inadequately treated for their pain symptoms over the seven years. Although the data from Kirou-Mauro et al.18 are consistent with the literature,15 the prevalence of undermedication is difficult to compare against other published results because the data were not assessed using the same measurement tool. This is because de Wit et al.,19 for example, found a wide variation in the estimates (16%–91%) of undermedicated patients within the same sample depending on the type of outcome measure and tool used. Accordingly, future management strategies to address undermedication across similar study populations in a homogeneous manner may become challenging when pain management assessment tools vary. Consequently, to fairly compare and analyze the estimate of the prevalence of inadequate pain management in cancer patients, a common measure should be used.
The present study investigated the prevalence of inadequate analgesic treatment for pain associated with bone metastases using the PMI. In our retrospective analysis of 1,000 patients with bone metastases at initial consultation at the RRRP, we found that 25.8% of patients were experiencing inadequate analgesic pain management. Specifically, the proportion of patients with inadequate analgesic pain management significantly increased between the years 1999 to 2006, with 18.9% of patients having a negative PMI score in 1999 and 44.3% of patients experiencing negative PMI scores in 2006. A potential explanation for this result may be that a change in patients' and physicians' awareness and attitudes toward pain and analgesic management over the years may have influenced the PMI scores over time.
We also found a strong negative association between PMI and performance status (P<0.001; OR=1.03). This finding is consistent with the published literature.2, 5, 8, 9, 20, 21 In a recent review of undertreatment by Deandrea et al.,15 they found 26 studies using the PMI, of which nine analyzed the relationship between performance status and PMI. Four of the nine publications confirmed the same relationship as our results, which is that patients who have good performance status (e.g., KPS>60) may clinically appear less ill and also may be judged to have lower pain scores and thus considered to require less potent analgesics. If so, this observation may have clinical relevance. Pain management should be dictated by a patient's pain score rather than making a subjective assessment of pain intensity based on performance status alone.
Although the primary cancer site has been found to be unrelated to undermedication,21 our study determined that patients with breast histology were more likely to be undertreated for their pain. Although we did not find sex to be related to prevalence of negative PMI scores, the relationship between breast primary cancer and inadequate pain management could be indicative of a sex relationship. Although our sample size is the largest in the literature to report on this, the results within the literature remain controversial. For example, Donovan et al.22 found female patients significantly more likely to receive inadequate analgesic pain management, as measured by the PMI, than their male counterparts. However, Deandrea et al.15 summarized that although 10 of 14 studies examined sex as a predictor of PMI, only two of the 10 found this variable to affect PMI.
Our center has a relatively lower prevalence of negative PMI scores (25.8%), that is, inadequate analgesic pain management, when compared with other countries. Additionally, our results are derived from the largest reported homogeneous sample of cancer patients with metastases to the bone. For example, a review of 26 studies on undertreatment revealed that 43% of cancer patients present with a negative PMI where sample sizes varied from 39 to 905 patients from the years 1999–2007.15 Specifically, in the United States, Cleeland et al.2 found that 42% of patients (n=1308) were undermedicated. Similar proportions were found in the European countries; in France, Larue et al.8 found that 57.5% of patients (n=605) were undermedicated; in Germany, the proportion was 44% (n=561);4 and in the Netherlands, it was 42% (n=1,429).4
The prevalence of undermedication for cancer pain may be highest in Asian countries; in China, 67% of patients (n=147) were undermedicated,13 whereas in India, the proportion was 79% (n=200).23 Although these international publications on PMI used cancer patients in their samples, the specific type of cancer population and the settings for each study were not consistent. Thus, a true comparison of negative PMI between countries is limiting. Furthermore, there may be socioeconomic reasons why pain medications may not have been used for patients. For example, relative to India, in Canada and other developed countries, the population is relatively affluent; access to doctors and prescription drugs, and social programs to provide drugs for underprivileged patients are better; and a culture exists where taking pain medications is not perceived negatively24. Additionally, in many countries, morphine and other analgesics are not available, might be very expensive, or there may be reluctance for physicians to prescribe and patients to use opioids because of their known side effects.
There are several limitations of the present study. One main reason for referral is uncontrolled bone pain, whereas other reasons include poor tolerance or side effects to analgesics, or fracture. Those with asymptomatic or well-controlled bone metastases are not likely to be referred to this clinic and would consequently not be included in our sample; these samples would be interesting to include in a future analysis. Furthermore, all of our patients were diagnosed with bone metastases. Previous studies have found that the presence of metastases is a predictor of better analgesic pain management.8, 21 As such, our sample may have produced a more conservative measure of the proportion of inadequate pain management, as compared with all cancer patients. Another noted limitation was that the correlation between PMI and the referral pattern was not possible, as this information was not captured. Furthermore, an expected limitation of this study was the number of patients who were lost to follow-up.
There also are limitations of the measurement tool used. The PMI is defined as a conservative measure of pain management adequacy. It captures pain management at one point in time, rather than over a period of time, and it does not take into account analgesic dose, schedule, route of administration, or adjuvant therapies. In addition, the use of adjuvant medication and the patient's adherence to medication is not recorded, and there is not always a consistent cut-off for the definition for mild, moderate, and severe pain.4, 5, 6, 25, 26 As a result, patients with positive PMI scores may have inadequate pain management. One study found that 21% of patients with positive PMI scores had unrelieved severe peak pain.12 An additional limitation of the PMI is that the categorization of the pain does not account for absolute pain levels but instead measures what patients believe is tolerable.19
To address some of these potential shortcomings, the original PMI2 has been modified.25, 26, 27, 28, 29, 30, 31, 32, 33 Ward et al.25 modified the PMI to account for a patient's adherence to medication by recording the pain medications patients actually used rather than recording what was prescribed by a clinician. Additionally, a patient's least pain score is incorporated into the equation and a score of 2 is considered to be desirable. de Wit et al.26 modified the PMI tool to be even more comprehensive by incorporating not only the worst pain score but also the present and average pain score. They created what is known as a tolerable pain score as an attempt to individualize each PMI score and attempted to incorporate the dose of medication by standardizing each medication to an oral morphine equivalent dose.
Despite these aforementioned modified PMI measures, there is still an opportunity to further improve this tool. For example, perhaps accounting for pain complexity, capturing a PMI measure over time, and including other forms of pain medication and nondrug therapies may be meaningful to incorporate in the scoring system. Thus, further studies are needed to compare these modified tools and determine which is most reliable and simplest to use in a clinic setting. Additionally, a greater sample size aimed at capturing patients along the course of the disease trajectory to determine where the breakdown in pain management occurs is required to provide a broader picture of undermedication in Canada.
Acknowledgment
The authors thank Stacy Lue for secretarial assistance.
References
- Prevalence and management of pain in Italian patients with advanced non-small-cell lung cancer. Br J Cancer. 2004;90:2288–2296
- Pain and its treatment in outpatients with metastatic cancer. N Engl J Med. 1994;330:592–596
- . Pain prevalence in hospitalized patients in a German university teaching hospital. J Pain Symptom Manage. 2005;29:498–506
- High prevalence of pain in patients with cancer in a large population-based study in the Netherlands. Pain. 2007;132:312–320
- Adequacy of cancer pain management in a Japanese cancer hospital. Jpn J Clin Oncol. 2004;34:37–42
- Minority cancer patients' and their providers' pain management attitudes and practice. Cancer. 2000;88:1929–1938
- . Pain and treatment of pain in minority patients with cancer: the Eastern Cooperative Oncology Group Minority Outpatient Pain Study. Ann Intern Med. 1997;127:813–816
- . Multicentre study of cancer pain and its treatment in France. Br Med J. 1995;310:1034–1037
- Development of a cancer pain assessment tool in Korea: a validation study of a Korean version of the Brief Pain Inventory. Oncology. 2004;66:439–444
- Russian Brief Pain Inventory: validation and application in cancer pain. J Pain Symptom Manage. 2008;35:95–102
- . The effects of analgesic prescription and patient adherence on pain in a Dutch outpatient cancer population. J Pain Symptom Manage. 2007;34:523–531
- . Prevalence and management of cancer pain in South Africa. Pain. 2001;94:75–84
- . The Chinese version of the Brief Pain Inventory (BPI-C): its development and use in a study of cancer pain. Pain. 1996;67:407–416
- . The WHO analgesic ladder for cancer pain control, twenty years of use. How much pain relief does one get from using it?. Support Care Cancer. 2006;14:1086–1093
- . Prevalence of undertreatment in cancer pain. A review of published literature. Ann Oncol. 2008;19:1985–1991
- Mild, moderate, or severe pain categorization by patients with cancer with bone metastases. J Palliat Med. 2007;10:1338–1346
- . What should be the optimal cutpoints for mild, moderate and severe?. J Palliat Med. 2007;10:1338–1346
- Has pain management in cancer patients with bone metastases improved? A seven-year review at an outpatient palliative radiotherapy clinic. J Pain Symptom Manage. 2009;37:77–84
- Empirical comparison of commonly used measures to evaluate pain treatment in cancer patients with chronic pain. J Clin Oncol. 1999;17:1280–1287
- Prospective patient-based assessment of effectiveness of palliative radiotherapy for bone metastases. Radiother Oncol. 2001;61:77–82
- Multicenter study of pain and its management in patients with advanced cancer in Korea. J Pain Symptom Manage. 2003;25:430–437
- . Sex differences in the adequacy of pain management among patients referred to a multidisciplinary cancer pain clinic. J Pain Symptom Manage. 2008;36:167–172
- . Assessment of cancer pain in North India: the validation of the Hindi Brief Pain Inventory—BPI-H. J Pain Symptom Manage. 1999;17:21–41
- . A Canadian survey of issues in cancer pain management. J Pain Symptom Manage. 1997;14:332–342
- . The impact on quality of life of patient-related barriers to pain management. Res Nurs Health. 1998;21:405–413
- The Amsterdam Pain Management Index compared to eight frequently used outcome measures to evaluate the adequacy of pain treatment in cancer patients with chronic pain. Pain. 2001;91:339–349
- . Improving cancer pain management through patient and family education. J Pain Symptom Manage. 2003;25:344–356
- . Cancer pain management in prisons: a survey of primary care practitioners and inmates. J Pain Symptom Manage. 2005;29:466–473
- . The prevalence and severity of cancer pain: a study of newly-diagnosed cancer patients in Taiwan. J Pain Symptom Manage. 1998;15:285–293
- Greek Pain Inventory: validation and utility in cancer pain. Oncology. 2001;60:35–42
- Pain control in ambulatory cancer patients—can we do better?. J Pain Symptom Manage. 2003;26:716–722
- Pain and its treatment in patients with cancer in Korea. Oncology. 2003;64:237–244
- . An assessment of methods used to evaluate the adequacy of cancer pain management. J Pain Symptom Manage. 2006;32:581–588
This work was supported by the Michael and Karyn Goldstein Cancer Research Fund.
PII: S0885-3924(09)01128-2
doi:10.1016/j.jpainsymman.2009.07.005
Crown Copyright © 2010. Published by Elsevier Inc. All rights reserved.
Volume 39, Issue 2 , Pages 259-267, February 2010
