- Email: [email protected]
(354) Estimating Minimally Important Differences for the PROMIS® Pain Interferences Scales Using Three Clinical Trials
Abstracts
(352) Older adults’ identification and response to adverse analgesic drug related events D McDonald, S Coughlin, and C Jin; University of Connecticut, Storrs, CT
Older adults (age 65 or greater) who take analgesics for chronic pain are at increased risk for adverse drug events (ADEs). Patient reports of adverse drug events might be helpful to incorporate into analgesic safety efforts. The purpose of this descriptive pilot survey was to examine how older adults self-identify analgesic ADEs, and what actions they take in response to analgesic ADEs. Twenty-two community dwelling older adults with chronic pain lasting six months or longer who reported an analgesic ADE associated with their chronic pain management were interviewed and asked to describe their analgesic related ADE. Written responses were independently content analyzed by two raters, with disagreement discussed to consensus. Nineteen opioids were reported by 15, 11 NSAIDs were reported by 8, and acetaminophen was reported by 2 older adults as associated with an ADE. Gastrointestinal ADEs were most common with upset stomach (31.8%) most frequent. Neurological ADEs were also common but more varied with dizziness (27.3%) and headache (13.6%) reported most frequently along with 13 other symptoms such as hallucinations and depression. Cardiovascular symptoms, dysrhythmia, was reported by only one older adult. A total of 54.5% responded to their ADE by contacting their physician. Three (13.6%) went to the emergency department. A total of 36.4% stopped taking their ADE associated analgesic, 22.7% started taking a different analgesic, and 22.7% started prophylaxis. Three (13.6%) continued their ADE related analgesic. A total of 54.5% reported their symptoms subsided, but 13.6% reported their symptoms remained. Three (13.6%) older adults reported their pain also remained. A significant number of older adults with chronic pain self-manage their analgesic related ADE without contacting their primary care provider. Analgesic related ADE prevention and management should be discussed during primary care visits to reduce ADEs and enhance pain management outcomes for older adults with chronic pain.
(353) Excess hours spent on health and health care management among older U.S. adults with disabling chronic pain M Janevic, S McLaughlin, A Heapy, and J Piette; University of Michigan, Ann Arbor, MI
The detrimental effects of chronic pain on functioning and wellbeing among older adults are well-documented. However, no prior investigation has assessed the excess time that individuals with chronic pain spend on managing their health and interacting with the health care system. The Health and Retirement Study (HRS) is a biannual panel study of a nationally-representative sample of U.S. adults over age 50. We used data from 5,423 HRS respondents who completed a supplement to the HRS in 2015, the Consumption and Activities Mail Survey. Respondents were classified as having disabling chronic pain if they indicated in a survey the prior year that: they were ‘‘often troubled’’ by either ‘‘moderate’’ or ‘‘severe’’ pain ‘‘most of the time’’ and that their pain limited their activities. To estimate the excess time burden for health care, we summed the hours/month that respondents reported spending on each of the following activities: ‘‘treating or managing a medical condition,’’ ‘‘seeing doctors, nurses, therapists, or other health care providers about your own health,’’ or ‘‘going to the pharmacy’’ and ‘‘paying or managing medical bills.’’ Results indicate that individuals with disabling chronic pain (n=1077) spent, on average, more than twice as much time on health-related activities (mean: 36 vs. 14 hours/month; median 11 vs. 4 hours; Mann-Whitney z=-16.7, p=.000) as those without disabling pain (n=4326). In multivariable models controlling for potential confounders (number of chronic health conditions, depressive symptoms, age, and education), disabling chronic pain remained a significant predictor of health management hours/month (b=15.5; t=7.36, p=.000). Findings suggest that disabling chronic pain is associated with a substantial time burden for health management activities. Future studies using daily life data collection strategies such as ecological momentary assessment may elucidate the reasons for this excess time burden and its effects on experienced – i.e., measured in real time – well-being.
The Journal of Pain
S63
(354) Estimating Minimally Important Differences for the PROMISâ Pain Interferences Scales Using Three Clinical Trials C Chen, K Kroenke, T Stump, J Kean, J Carpenter, E Kreb, M Bair, T Damush, and P Monahan; Indiana University, Indianapolis, IN The minimally important difference (MID) refers to the smallest meaningful difference that carries implications for patients’ care. MIDs are necessary to guide the interpretation of patient-reported outcome tools in research and clinical practice. The Patient Reported Outcomes Measurement Information System (PROMISâ) pain interference scales have been validated across diverse samples. However, to improve the scales’ interpretability, information about their MIDs is still warranted. The purpose of this study was to estimate MIDs for the PROMISâ pain interference scales, including the 6-item Pain Short Form, and the 4-, 6-, and 8- item pain scales used on the PROMIS profiles. Data were from three randomized controlled trials totaling 759 patients. The three samples respectively consisted of patients with chronic low back pain (n =261), chronic musculoskeletal pain (n =240), and a history of stroke (n =258). Participants completed PROMISâ pain interference scales and anchor measures (e.g., the Brief Pain Inventory and global ratings of change) at baseline and follow-up assessment 3 or 6 months later. Anchor- and distribution-based approaches were used to estimate MIDs. Anchor-based MIDs were established by mapping the PROMISâ pain interference scores to the established anchor measures. Standard error of measurement (SEM) was used as a distribution-based MID estimate. MID estimates based on anchor- and distribution-based approaches will be reported. MID estimates can enhance the interpretability of PROMIS pain interference scales. They can be used to evaluate treatment effects in research and clinical care and to provide a basis for power calculations in clinical trials.
(355) Measuring clarity, relevance, and usefulness of HEAL and PROMIS measures in pain treatment through interviews with patients and their healthcare providers J Slutsky, C Greco, C McFarland, N Dodds, K Johnston, R Glick, M Schneider, J Janjic, N Kelly, N Morone, C Adams, S Lawrence, and P Pilkonis; University of Pittsburgh Medical Center, Pittsburgh, PA Patient engagement is particularly important in pain treatment. Clinicians may enhance positive outcomes by listening to and valuing patients’ perspectives on nonspecific factors, such as treatment expectations, and patient-provider relationships. The Healing Encounters and Attitudes Lists (HEAL) are patient-reported measures developed using PROMIS methodology that evaluate such factors in treatment. Validation studies of HEAL support its concurrent and predictive validity. This study investigates patient and provider views of the clinical utility of the HEAL and PROMIS measures. Participants receiving a new treatment for chronic pain completed demographic information and computerized adaptive test (CAT) versions of HEAL and PROMIS measures. We interviewed 46 patients and 11 of their clinicians. The patient sample consisted primarily of women (65%), with 31% African American and 2% Hispanic. The mean age was 49 (SD = 16), and 35% had less than a 4-year college degree. During the individual interviews, patients and clinicians rated the clarity, relevance, and usefulness of HEAL and PROMIS measures. Item clarity was rated consistently high, ranging from 3.4 to 3.9 (on a 0 to 4 scale). Patients endorsed HEAL Treatment Expectancy, Patient-Provider Connection, Healthcare Environment and Positive Outlook measures as highly relevant to their treatment (means ranging from 3.2 to 3.6). Patients also felt these HEAL measures provided useful information to providers (means ranging from 3.0 to 3.4). For PROMIS, pain intensity was the highest rated (relevance and usefulness – M=3.8, SD=.46, equal respectively) and anxiety was the lowest (relevance – M=3.2, SD=.99, usefulness – M=3.2, SD=.95). Clinicians’ ranking of HEAL and PROMIS domains’ importance was identical to those of patients. These findings support the suitability and importance of assessing both nonspecific factors and pain outcomes in the context of treatment. Supported by a PCORI Methods award.
(356) A Center for Patients with Severe Persistent Pain after Groin Hernia Repair: Five-years of Nationwide Experience E Kjær Jensen, T
Ringsted, H
Kehlet, and M
Werner;
Multidisciplinary Pain Center, Neuroscience Center, Copenhagen University Hospitals, Copenhagen, Denmark
Rigshospitalet,
(352) Older adults’ identification and response to adverse analgesic drug related events D McDonald, S Coughlin, and C Jin; University of Connecticut, Storrs, CT
Older adults (age 65 or greater) who take analgesics for chronic pain are at increased risk for adverse drug events (ADEs). Patient reports of adverse drug events might be helpful to incorporate into analgesic safety efforts. The purpose of this descriptive pilot survey was to examine how older adults self-identify analgesic ADEs, and what actions they take in response to analgesic ADEs. Twenty-two community dwelling older adults with chronic pain lasting six months or longer who reported an analgesic ADE associated with their chronic pain management were interviewed and asked to describe their analgesic related ADE. Written responses were independently content analyzed by two raters, with disagreement discussed to consensus. Nineteen opioids were reported by 15, 11 NSAIDs were reported by 8, and acetaminophen was reported by 2 older adults as associated with an ADE. Gastrointestinal ADEs were most common with upset stomach (31.8%) most frequent. Neurological ADEs were also common but more varied with dizziness (27.3%) and headache (13.6%) reported most frequently along with 13 other symptoms such as hallucinations and depression. Cardiovascular symptoms, dysrhythmia, was reported by only one older adult. A total of 54.5% responded to their ADE by contacting their physician. Three (13.6%) went to the emergency department. A total of 36.4% stopped taking their ADE associated analgesic, 22.7% started taking a different analgesic, and 22.7% started prophylaxis. Three (13.6%) continued their ADE related analgesic. A total of 54.5% reported their symptoms subsided, but 13.6% reported their symptoms remained. Three (13.6%) older adults reported their pain also remained. A significant number of older adults with chronic pain self-manage their analgesic related ADE without contacting their primary care provider. Analgesic related ADE prevention and management should be discussed during primary care visits to reduce ADEs and enhance pain management outcomes for older adults with chronic pain.
(353) Excess hours spent on health and health care management among older U.S. adults with disabling chronic pain M Janevic, S McLaughlin, A Heapy, and J Piette; University of Michigan, Ann Arbor, MI
The detrimental effects of chronic pain on functioning and wellbeing among older adults are well-documented. However, no prior investigation has assessed the excess time that individuals with chronic pain spend on managing their health and interacting with the health care system. The Health and Retirement Study (HRS) is a biannual panel study of a nationally-representative sample of U.S. adults over age 50. We used data from 5,423 HRS respondents who completed a supplement to the HRS in 2015, the Consumption and Activities Mail Survey. Respondents were classified as having disabling chronic pain if they indicated in a survey the prior year that: they were ‘‘often troubled’’ by either ‘‘moderate’’ or ‘‘severe’’ pain ‘‘most of the time’’ and that their pain limited their activities. To estimate the excess time burden for health care, we summed the hours/month that respondents reported spending on each of the following activities: ‘‘treating or managing a medical condition,’’ ‘‘seeing doctors, nurses, therapists, or other health care providers about your own health,’’ or ‘‘going to the pharmacy’’ and ‘‘paying or managing medical bills.’’ Results indicate that individuals with disabling chronic pain (n=1077) spent, on average, more than twice as much time on health-related activities (mean: 36 vs. 14 hours/month; median 11 vs. 4 hours; Mann-Whitney z=-16.7, p=.000) as those without disabling pain (n=4326). In multivariable models controlling for potential confounders (number of chronic health conditions, depressive symptoms, age, and education), disabling chronic pain remained a significant predictor of health management hours/month (b=15.5; t=7.36, p=.000). Findings suggest that disabling chronic pain is associated with a substantial time burden for health management activities. Future studies using daily life data collection strategies such as ecological momentary assessment may elucidate the reasons for this excess time burden and its effects on experienced – i.e., measured in real time – well-being.
The Journal of Pain
S63
(354) Estimating Minimally Important Differences for the PROMISâ Pain Interferences Scales Using Three Clinical Trials C Chen, K Kroenke, T Stump, J Kean, J Carpenter, E Kreb, M Bair, T Damush, and P Monahan; Indiana University, Indianapolis, IN The minimally important difference (MID) refers to the smallest meaningful difference that carries implications for patients’ care. MIDs are necessary to guide the interpretation of patient-reported outcome tools in research and clinical practice. The Patient Reported Outcomes Measurement Information System (PROMISâ) pain interference scales have been validated across diverse samples. However, to improve the scales’ interpretability, information about their MIDs is still warranted. The purpose of this study was to estimate MIDs for the PROMISâ pain interference scales, including the 6-item Pain Short Form, and the 4-, 6-, and 8- item pain scales used on the PROMIS profiles. Data were from three randomized controlled trials totaling 759 patients. The three samples respectively consisted of patients with chronic low back pain (n =261), chronic musculoskeletal pain (n =240), and a history of stroke (n =258). Participants completed PROMISâ pain interference scales and anchor measures (e.g., the Brief Pain Inventory and global ratings of change) at baseline and follow-up assessment 3 or 6 months later. Anchor- and distribution-based approaches were used to estimate MIDs. Anchor-based MIDs were established by mapping the PROMISâ pain interference scores to the established anchor measures. Standard error of measurement (SEM) was used as a distribution-based MID estimate. MID estimates based on anchor- and distribution-based approaches will be reported. MID estimates can enhance the interpretability of PROMIS pain interference scales. They can be used to evaluate treatment effects in research and clinical care and to provide a basis for power calculations in clinical trials.
(355) Measuring clarity, relevance, and usefulness of HEAL and PROMIS measures in pain treatment through interviews with patients and their healthcare providers J Slutsky, C Greco, C McFarland, N Dodds, K Johnston, R Glick, M Schneider, J Janjic, N Kelly, N Morone, C Adams, S Lawrence, and P Pilkonis; University of Pittsburgh Medical Center, Pittsburgh, PA Patient engagement is particularly important in pain treatment. Clinicians may enhance positive outcomes by listening to and valuing patients’ perspectives on nonspecific factors, such as treatment expectations, and patient-provider relationships. The Healing Encounters and Attitudes Lists (HEAL) are patient-reported measures developed using PROMIS methodology that evaluate such factors in treatment. Validation studies of HEAL support its concurrent and predictive validity. This study investigates patient and provider views of the clinical utility of the HEAL and PROMIS measures. Participants receiving a new treatment for chronic pain completed demographic information and computerized adaptive test (CAT) versions of HEAL and PROMIS measures. We interviewed 46 patients and 11 of their clinicians. The patient sample consisted primarily of women (65%), with 31% African American and 2% Hispanic. The mean age was 49 (SD = 16), and 35% had less than a 4-year college degree. During the individual interviews, patients and clinicians rated the clarity, relevance, and usefulness of HEAL and PROMIS measures. Item clarity was rated consistently high, ranging from 3.4 to 3.9 (on a 0 to 4 scale). Patients endorsed HEAL Treatment Expectancy, Patient-Provider Connection, Healthcare Environment and Positive Outlook measures as highly relevant to their treatment (means ranging from 3.2 to 3.6). Patients also felt these HEAL measures provided useful information to providers (means ranging from 3.0 to 3.4). For PROMIS, pain intensity was the highest rated (relevance and usefulness – M=3.8, SD=.46, equal respectively) and anxiety was the lowest (relevance – M=3.2, SD=.99, usefulness – M=3.2, SD=.95). Clinicians’ ranking of HEAL and PROMIS domains’ importance was identical to those of patients. These findings support the suitability and importance of assessing both nonspecific factors and pain outcomes in the context of treatment. Supported by a PCORI Methods award.
(356) A Center for Patients with Severe Persistent Pain after Groin Hernia Repair: Five-years of Nationwide Experience E Kjær Jensen, T
Ringsted, H
Kehlet, and M
Werner;
Multidisciplinary Pain Center, Neuroscience Center, Copenhagen University Hospitals, Copenhagen, Denmark
Rigshospitalet,