Performance Evaluation of a Supervised Machine Learning Pain Classification Model Developed by Neonatal Nurses.

BACKGROUND: Early-life pain is associated with adverse neurodevelopmental consequences; and current pain assessment practices are discontinuous, inconsistent, and highly dependent on nurses' availability. Furthermore, facial expressions in commonly used pain assessment tools are not associated with brain-based evidence of pain. PURPOSE: To develop and validate a machine learning (ML) model to classify pain. METHODS: In this retrospective validation study, using a human-centered design for Embedded Machine Learning Solutions approach and the Neonatal Facial Coding System (NFCS), 6 experienced neonatal intensive care unit (NICU) nurses labeled data from randomly assigned iCOPEvid (infant Classification Of Pain Expression video) sequences of 49 neonates undergoing heel lance. NFCS is the only observational pain assessment tool associated with brain-based evidence of pain. A standard 70% training and 30% testing split of the data was used to train and test several ML models. NICU nurses' interrater reliability was evaluated, and NICU nurses' area under the receiver operating characteristic curve (AUC) was compared with the ML models' AUC. RESULTS: Nurses weighted mean interrater reliability was 68% (63%-79%) for NFCS tasks, 77.7% (74%-83%) for pain intensity, and 48.6% (15%-59%) for frame and 78.4% (64%-100%) for video pain classification, with AUC of 0.68. The best performing ML model had 97.7% precision, 98% accuracy, 98.5% recall, and AUC of 0.98. IMPLICATIONS FOR PRACTICE AND RESEARCH: The pain classification ML model AUC far exceeded that of NICU nurses for identifying neonatal pain. These findings will inform the development of a continuous, unbiased, brain-based, nurse-in-the-loop Pain Recognition Automated Monitoring System (PRAMS) for neonates and infants.

Chronic pain: an update on burden, best practices, and new advances.

Chronic pain exerts an enormous personal and economic burden, affecting more than 30% of people worldwide according to some studies. Unlike acute pain, which carries survival value, chronic pain might be best considered to be a disease, with treatment (eg, to be active despite the pain) and psychological (eg, pain acceptance and optimism as goals) implications. Pain can be categorised as nociceptive (from tissue injury), neuropathic (from nerve injury), or nociplastic (from a sensitised nervous system), all of which affect work-up and treatment decisions at every level; however, in practice there is considerable overlap in the different types of pain mechanisms within and between patients, so many experts consider pain classification as a continuum. The biopsychosocial model of pain presents physical symptoms as the denouement of a dynamic interaction between biological, psychological, and social factors. Although it is widely known that pain can cause psychological distress and sleep problems, many medical practitioners do not realise that these associations are bidirectional. While predisposing factors and consequences of chronic pain are well known, the flipside is that factors promoting resilience, such as emotional support systems and good health, can promote healing and reduce pain chronification. Quality of life indicators and neuroplastic changes might also be reversible with adequate pain management. Clinical trials and guidelines typically recommend a personalised multimodal, interdisciplinary treatment approach, which might include pharmacotherapy, psychotherapy, integrative treatments, and invasive procedures.

Pain phenotypes classified by machine learning using electroencephalography features.

Pain is a multidimensional experience mediated by distributed neural networks in the brain. To study this phenomenon, EEGs were collected from 20 subjects with chronic lumbar radiculopathy, 20 age and gender matched healthy subjects, and 17 subjects with chronic lumbar pain scheduled to receive an implanted spinal cord stimulator. Analysis of power spectral density, coherence, and phase-amplitude coupling using conventional statistics showed that there were no significant differences between the radiculopathy and control groups after correcting for multiple comparisons. However, analysis of transient spectral events showed that there were differences between these two groups in terms of the number, power, and frequency-span of events in a low gamma band. Finally, we trained a binary support vector machine to classify radiculopathy versus healthy subjects, as well as a 3-way classifier for subjects in the 3 groups. Both classifiers performed significantly better than chance, indicating that EEG features contain relevant information pertaining to sensory states, and may be used to help distinguish between pain states when other clinical signs are inconclusive.

Initial classification of low back and leg pain based on objective functional testing: a pilot study of machine learning applied to diagnostics.

OBJECTIVE: The five-repetition sit-to-stand (5R-STS) test was designed to capture objective functional impairment and thus provided an adjunctive dimension in patient assessment. The clinical interpretability and confounders of the 5R-STS remain poorly understood. In clinical use, it became apparent that 5R-STS performance may differ between patients with lumbar disk herniation (LDH), lumbar spinal stenosis (LSS) with or without low-grade spondylolisthesis, and chronic low back pain (CLBP). We seek to evaluate the extent of diagnostic information contained within 5R-STS testing. METHODS: Patients were classified into gold standard diagnostic categories based on history, physical examination, and imaging. Crude and adjusted comparisons of 5R-STS performance were carried out among the three diagnostic categories. Subsequently, a machine learning algorithm was trained to classify patients into the three categories using only 5R-STS test time and patient age, gender, height, and weight. RESULTS: From two prospective studies, 262 patients were included. Significant differences in crude and adjusted test times were observed among the three diagnostic categories. At internal validation, classification accuracy was 96.2% (95% CI 87.099.5%). Classification sensitivity was 95.7%, 100%, and 100% for LDH, LSS, and CLBP, respectively. Similarly, classification specificity was 100%, 95.7%, and 100% for the three diagnostic categories. CONCLUSION: 5R-STS performance differs according to the etiology of back and leg pain, even after adjustment for demographic covariates. In combination with machine learning algorithms, OFI can be used to infer the etiology of spinal back and leg pain with accuracy comparable to other diagnostic tests used in clinical examination. These slides can be retrieved under Electronic Supplementary Material.

What Do Adult Surgical Patients Speaking Primarily Cantonese, Italian, or Portuguese Really Want to Know About Pain and Pain Management?

BACKGROUND: Surgical patients consider information about pain and pain management to be highly important (Apfelbaum, 2003). At the same time, evidence indicates that members of racial and ethnic minorities are more likely to experience inadequate pain management (Green, Anderson, Baker, Campbell, Decker, Fillingim, & Todd, 2003; Mossey, 2011). AIMS: This study investigated the needs of general day surgery patients who spoke primarily Cantonese, Italian, or Portuguese at home for information about postoperative pain. DESIGN: This was a mixed methods, descriptive study. SETTING: The day surgery unit of a large, quaternary care hospital in downtown Toronto. PARTIPANTS/SUBJECTS: Inclusion criteria were day patients who were at least 18 years of age or older and spoke primariy Cantonese, Italian or Portugues at home. and were able to read and write in their primary language. METHODS: Participants who had undergone a day surgery procedure completed a telephone information needs survey in their primary language (Cantonese, Italian, Portuguese) within 72 hours after discharge. Composite mean scores were calculated for each item. Chi-squared analyses were used to probe for intergroup differences and compare with English-fluent participants from phase 1 of this study (Kastanias, Denny, Robinson, Sabo, & Snaith, 2009). RESULTS: Sixty-three participants in total completed the survey: 21% Cantonese, 41% Italian, and 38% Portuguese. The mean age of the sample was 70 years old; 89% were born outside of Canada, and 52% were male. For the combined group, the average importance rating score range for the information items was 6.2-8.9 out of a possible score of 10. All items were rated as moderate (5-6 out of 10) to high (≥7out of 10) importance. Surgical subtype, health status, and age had no effect on the importance of any information item. There were no significant differences between the three language groups on any of the information items. This lack of difference may have been a result of a lack of power due to the small sample size of the individual language groupings. Overall, the top-ranked information items were "the plan for which drugs to take and when," "what I can do if I still have pain or side effects," and "side effects I was most likely to get." CONCLUSIONS: Similar to English-fluent participants (Kastanias et al., 2009), participants who primarily spoke either Cantonese, Italian, or Portuguese at home placed moderate to high importance on all of the information items. and neither surgical subtype, health status nor age had any effect on the importance of any item. The multilingual sample in this study placed more importance than English-fluent participants on information regarding help with paying for pain medication (p = .001) and the side effects they were most likely to experience (p < .05). Due to a paucity of literature in this area, further research is warranted. Results may assist with evaluating and improving current approaches to surgical patient pain management education.

African American Young Adults' Pain and Pain Reduction Strategies.

BACKGROUND: Effective acute pain management strategies are important for young adults in order to reduce risk for transition to chronic pain. AIM: To describe pain and pain self-management strategies used by African American young adults. DESIGN & SETTING: A national online cross-sectional survey design was used. PARTICIPANT/SUBJECTS: Ninety-four African Americans Qualtrics panelists ages 18-25 who reported previous experience with acute pain responded. Methods: Respondents completed the Brief Pain Inventory Short Form to describe their pain intensity, pain interference with function, pain self-management, and percent of relief obtained from their self-management. RESULTS: African American young adults reported pain primarily in the back (n = 22, 23.4%) and head (n = 19, 20.2%), with moderate pain intensity M = 4.5 (standard deviation [SD] = 1.79) and pain interference with function M = 4.6 (SD = 2.36). African American young adults described their worst pain in the last 24 hours as M = 5.7 (SD = 2.01), least pain as M = 3.4 (SD = 2.41), and average pain as M = 5.1 (SD = 2.09). They reported 61.3% pain relief from self-treatment. A total of 45 (47.9%) reported no pain self-management strategies. CONCLUSIONS: African American young adults report moderate levels of pain intensity and pain interference with function. A significant number report no pain self-management strategies. Focused pain assessment and education about efficacious pain self-management strategies, both pharmacological and complementary, could assist young African Americans to reduce their pain and risk of chronic pain in the future.

Classification of Pain Event Related Potential for Evaluation of Pain Perception Induced by Electrical Stimulation.

Variability in individual pain sensitivity is a major problem in pain assessment. There have been studies reported using pain-event related potential (pain-ERP) for evaluating pain perception. However, none of them has achieved high accuracy in estimating multiple pain perception levels. A major reason lies in the lack of investigation of feature extraction. The goal of this study is to assess four different pain perception levels through classification of pain-ERP, elicited by transcutaneous electrical stimulation on healthy subjects. Nonlinear methods: Higuchi's fractal dimension, Grassberger-Procaccia correlation dimension, with auto-correlation, and moving variance functions were introduced into the feature extraction. Fisher score was used to select the most discriminative channels and features. As a result, the correlation dimension with a moving variance without channel selection achieved the best accuracies of 100% for both the two-level and the three-level classification but degraded to 75% for the four-level classification. The best combined feature group is the variance-based one, which achieved accuracy of 87.5% and 100% for the four-level and three-level classification, respectively. Moreover, the features extracted from less than 20 trials could not achieve sensible accuracy, which makes it difficult for an instantaneous pain perception levels evaluation. These results show strong evidence on the possibility of objective pain assessment using nonlinear feature-based classification of pain-ERP.

Comparing the physical, psychological, social, and spiritual needs of patients with non-cancer and cancer diagnoses in a tertiary palliative care setting.

OBJECTIVE: The purpose was to describe the physical, psychological, social, and spiritual needs of patients with non-cancer serious illness diagnoses compared to those of patients with cancer. METHOD: We conducted a retrospective chart review of all patients with a non-cancer diagnosis admitted to a tertiary palliative care unit between January 2008 and December 2017 and compared their needs to those of a matched cohort of patients with cancer diagnoses. The prevalence of needs within the following four main concerns was recorded and the data analyzed using descriptive statistics and content analysis: •Physical: pain, dyspnea, fatigue, anorexia, edema, and delirium•Psychological: depression, anxiety, prognosis, and dignity•Social: caregiver burden, isolation, and financial•Spiritual: spiritual distress. RESULTS: The prevalence of the four main concerns was similar among patients with non-cancer and cancer diagnoses. Pain, nausea/vomiting, fatigue, and anorexia were more prevalent among patients with cancer. Dyspnea was more commonly the primary concern in patients with non-cancer diagnoses (39%), who also had a higher prevalence of anxiety and concerns about dignity. Spirituality was addressed more often in patients with cancer. SIGNIFICANCE OF RESULTS: The majority of patients admitted to tertiary palliative care settings have historically been those with cancer. The tertiary palliative care needs of patients with non-cancer diagnoses have not been well described, despite the increasing prevalence of this population. Our description of the palliative care needs of patients with non-cancer diagnoses will help guide future palliative care for the increasing population of patients with non-cancer serious illness diagnoses.

Classifying pain in transoral endoscopic thyroidectomy.

PURPOSES: Knowledge of visual analog scale (VAS) pain assessment for transoral endoscopic thyroidectomy vestibular approach (TOETVA) is limited. The purpose of this analysis was to classify the postoperative discomfort scores in patients undergoing TOETVA compared to open thyroidectomy. METHODS: Observational clinical study of patients who underwent thyroidectomy by VAS pain assessment from September 2016 to March 2017. Patients were stratified into two groups: patients eligible for TOETVA (Group TOETVA) and non-candidates for endoscopic intervention (open thyroidectomy approach-OTA). VAS was recorded in the recovery room, at 24 h, + 2, + 5, + 15, + 30, + 90 days, and 6 months after surgery. Pain assessment was stratified in VAS-lower lip, VAS-chin, VAS-jaw, VAS-anterior neck, VAS-cervical/back, VAS-swallowing, VAS-brushing, VAS-speaking, and VAS-shaving. Secondary outcome assessed were analgesic rescue dose, morbidity, operative notes, hospital stay, and histopathology. RESULTS: 41 TOETVA and 45 OTA constituted the analysis. There were differences between the TOETVA and OTA for age, gland volume, mean nodule diameter, coexistence thyroiditis, bilateral procedures, and use of drain. Operative time was longer in TOETVA. Results indicated that TOETVA was associated with reduced neck, cervical back, and swallowing VAS scores in the 24 h after surgery. Conversely, jaw and brushing teeth resulted in higher VAS score in TOETVA group. OTA patients never experienced lower lip or chin pain. The use of rescue analgesics did not differ between the two groups. CONCLUSIONS: VAS was used to measure treatment outcome in TOETVA. VAS scores achieved overall a minimal clinical importance difference from the two procedures. There appears to be both a short- and long-term different range of interpretations of pain between TOETVA and OTA.

Physiological Response to Opioids.

The clinical setting in which women's health physicians practice, whether as generalist, obstetricians and gynecologists, or subspecialists, dictates our frequent clinical interaction with "pain." Opioid-containing medications are frequently prescribed within our specialty as a means of immediate pain relief. Opioid-containing medication causes a deep physiological alteration of several systems resulting in potential harm to acute and chronic opioid users. This article includes a thorough system-based review of opioid-containing medications on physiological systems. Women's health providers should have an in-depth understanding of such reverberations on patients' wellbeing to maintain the safest level of care. A solid grasp of physiological repercussions of opioid use would encourage physicians to seek alternative treatment options. Such practice is essential in curbing the opioid epidemic our patients are facing.

The Differentiation Between Pain and Discomfort: A Concept Analysis of Discomfort.

BACKGROUND: Discomfort is a concept found in the literature, usually related to pain. Some sources do not distinguish between pain and discomfort. Others refer to different sources of discomfort, thereby leading to a lack of conceptual clarity. AIMS: The objective of this paper was to present a concept analysis of discomfort. Full-text articles published between 1970 and 2016 in English were used to inform the concept analysis. DESIGN: Articles were taken from CINAHL, Medline and PsycNET databases. METHODS: A total of 7,406 articles and 120 abstracts were identified for evaluation. After initial review, 42 articles were further analyzed. Two reviewers independently evaluated the selected publications using the Walker and Avant approach to concept analysis. RESULTS: Discomfort can be physical or psychological and is characterized by an unpleasant feeling resulting in a natural response of avoidance or reduction of the source of the discomfort. Pain is one of the causes for discomfort, but not every discomfort can be attributed to pain. It is identified by self-report or observation. Discomfort in noncommunicative patients is assessed and measured via behavioral expression, also used to describe pain and agitation, leading to discomfort being interpreted as pain in some conditions. CONCLUSIONS: A clarification of the concept of discomfort leads to a more accurate theoretical and operational definition. This clarification can help nurses to make more accurate nursing diagnoses and develop methods to measure discomfort in order to provide optimal quality of nursing care.

Common Causes of Pain in Systemic Sclerosis: Frequency, Severity, and Relationship to Disease Status, Depression, and Quality of Life.

BACKGROUND: In routine clinical practice, healthcare professionals draw little attention to pain in patients with systemic sclerosis (SSc). Pain has adverse effects on functional ability, social and emotional wellbeing. AIMS: This study aims to assess the frequency and severity of different types of pain in patients with SSc, and the relationship of pain with disease status, depression and quality of life. DESIGN: Consecutive patients with SSc were included in this cross-sectional study. Patients with previously diagnosed painful diseases or conditions (other rheumatic diseases, angina pectoris, neurological disorders, etc.) were excluded. SETTINGS: Patients, who visited our rheumatology outpatient clinic from February to November 2016, participated in this study. PARTICIPANTS/SUBJECTS: 42 consecutive patients with SSc (38 women and 4 men), mean age 56.5 years, mean disease duration 9.5 years, were included. METHODS: All patients filled in a questionnaire, to indicate the presence or absence of some predefined pain syndromes. Disease status was assessed using the Scleroderma Assessment Questionnaire (SAQ), symptoms of depression by the Beck's Depression Inventory (BDI), whilst the quality of life was evaluated using the EuroQol questionnaire. RESULTS: It was found that 92.9% of SSc patients suffer from different types of pain, and 45.2% of patients have pain every day. Joint pain was the most common type of pain, present in 78.6% of patients, followed by pain associated with Raynaud's phenomenon (69%), back pain (47.6%), headache (31%), chest pain (23.8%), odynophagia (21.4%) and painful digital ulcers (19%). Symptoms of neuropathic pain were noticed in 26.2% of patients. Severe joint pain, everyday pain and symptoms of neuropathic pain in SSc were associated with more severe disease and poorer quality of life. Pain related to Raynaud's phenomenon, digital ulcers, odynophagia and joint pain were associated with significant symptoms of depression. CONCLUSION: The majority of patients with SSc suffer from different types of pain. Pain is associated with more severe disease, depression and poor quality of life.

The Effectiveness of an Evidence-Based Pain Management Program on Pain Intensity and Chest Rehabilitation Improvement Among Chest Trauma Patients in a Thai Hospital.

BACKGROUND: Pain after thoracic injury has further profound impacts on patients resulting in increased length of hospital stay and hospital care cost, and decreased quality of life. Utilization of the cutting-edge evidence on pain management that fits with the individual care context is therefore important. AIM: To examine the effects of an evidenced-based pain management program on the worst pain intensity and lung vital capacity among acutely ill hospitalized chest trauma patients. DESIGN: A two-group repeated measures design. SETTINGS: trauma unit, a university hospital in southern Thailand. PARTICIPANTS/SUBJECTS: 42 chest trauma patients. METHODS: The study population included 42 chest trauma patients admitted to the trauma unit. Twenty-one eligible chest trauma patients were consecutively assigned into intervention and control groups. The impacts of the intervention on the level of the worst pain intensity and lung vital capacity were measured before implementation of the program and throughout the first 5 days of admission. RESULTS: The study found a significant reduction in the worst pain intensity and an increase in the lung vital capacity among chest trauma patients in the intervention group compared with the control group (p < .05). CONCLUSIONS: Use of a pain management program can be an effective, inexpensive, and low-risk intervention for the improvement of pain management and chest rehabilitation among chest trauma patients.

Pain Quality by Location in Outpatients with Cancer.

BACKGROUND: The McGill Pain Questionnaire (MPQ) pain quality descriptors have been analyzed to characterize the sensory, affective, and evaluative domains of pain, but have not been differentiated by pain location. AIM: To examine MPQ pain quality descriptors by pain location in outpatients with lung or prostate cancer. DESIGN: Cross sectional. SETTINGS: Eleven oncology clinics or patients' homes. SUBJECTS: 264 adult outpatients (80% male; mean age 62.2 ± 10.0 years, 85% White). METHODS: Subjects completed a 100 mm visual analogue scale of pain intensity and MPQ clinic or home visit, marking sites where they had pain on a body outline and circling from 78 verbal descriptors those that described their pain. A researcher noted next to the descriptor spontaneous comments about sites feeling like a selected word and queried the subjects about any other words to obtain the site(s). RESULTS: Pain quality descriptors were assigned to all 7 pain locations marked by ≥ 20% of 198 lung or 66 prostate cancer patients. Four pain locations were marked with pain quality descriptors significanlty (p < .05) more frequently for lung cancer (53% chest-aching, burning; 58% back-aching, stabbing; 48% head-aching, sharp; and 19% arms-aching, stabbing) than for prostate cancer, which had significantly more frequent pain locations in the abdomen (64%-aching, burning) and lower back/buttocks (55%-aching, burning). CONCLUSIONS: This type of pain characterization is innovative and has the potential to help implement targeted treatments for patients with cancer and other chronic pain conditions.

Spiritual Perspectives on Pain in Advanced Breast Cancer: A Scoping Review.

OBJECTIVES: The aim of the present review was to characterize how pain and spirituality have been conceptualized, assessed, and addressed and how these concepts may be related among women with advanced breast cancer. DESIGN: A scoping review was conducted including publications of various methodologies. DATA SOURCES: Searches were conducted in PubMed, CINAHL, PsycINFO, Cochrane Library, OpenGrey, OAIster, and a large university library database (published 2006-2018). REVIEW/ANALYSIS METHODS: Research questions and criteria were formulated at the outset, followed by identification of publications, charting data, and collating results. RESULTS: Forty-two publications met the inclusion criteria. Most (n = 33) focused exclusively on pain, five pain and spirituality, and four exclusively spirituality. Conceptual definitions were not explicitly provided but were implied. Most assessments used the 0-10 Numeric Rating Scale (pain) and qualitative methods (spirituality). Pain management primarily focused on radiotherapy and pharmaceuticals, and two publications identified spiritual interventions. No publications directly examined the impact of spirituality on pain. Findings of qualitative studies including both concepts suggest the potential value of spirituality as a mechanism to cope with pain. CONCLUSIONS: This review identified significant unmanaged pain in women with advanced breast cancer. Women identified dimensions of spirituality as important for coping with their disease. A gap in understanding spirituality and its potential influence on pain in this population was identified.

A Deep Neural Network-Based Pain Classifier Using a Photoplethysmography Signal.

Side effects occur when excessive or low doses of analgesics are administered compared to the required amount to mediate the pain induced during surgery. It is important to accurately assess the pain level of the patient during surgery. We proposed a pain classifier based on a deep belief network (DBN) using photoplethysmography (PPG). Our DBN learned about a complex nonlinear relationship between extracted PPG features and pain status based on the numeric rating scale (NRS). A bagging ensemble model was used to improve classification performance. The DBN classifier showed better classification results than multilayer perceptron neural network (MLPNN) and support vector machine (SVM) models. In addition, the classification performance was improved when the selective bagging model was applied compared with the use of each single model classifier. The pain classifier based on DBN using a selective bagging model can be helpful in developing a pain classification system.

Vital signs: Valid indicators to assess pain in intensive care unit patients? An observational, descriptive study.

Pain is a stressor for intensive care unit (ICU) patients, and inadequate pain assessment has been linked to increased morbidity and mortality. One hundred and twenty patients were evaluated during three periods: (T1) 1 min before, (T2) during, and (T3) 20 min after the nociceptive procedure. For each patient, data were obtained through at least two nociceptive procedures. Conscious patients' self-reports of pain were assessed using the Numerical Rating Scale and Visual Analog Scale. For unconscious patients, the Behavioral Pain Scale was used instead. Descriptive statistical methods, Friedman's test, and Spearman's rank correlation coefficient were used for the data analysis. Significant changes were observed in heart rate (HR), respiratory rate (RR), and peripheral oxygen saturation (SpO2 ) during nociceptive procedures. The HR, RR, and pain scores increased, while the SpO2 decreased. Positive correlation coefficients were observed between the pain intensity and HR and RR levels. According to our study findings, vital signs are not strong indicators for pain assessment in neurosurgery ICU patients. However, HR and RR can be used as cues when behavioral indicators are not valid in these unconscious patients.

Sleep and Pain in Mid- to Late-Life: An Exploration of Day-to-Day Pain Inconsistency.

OBJECTIVES: This study examined how different quantifications of pain (average vs. day-to-day inconsistency) are related to sleep in older adults beyond known predictors. METHODS: Baseline measures from the Active Adult Mentoring Project were used for secondary analyses. Participants included 82 adults in mid- to late-life. Depression was assessed with the BDI-II. Pain intensity was assessed over seven days on a 11-point Likert-scale, while sleep efficiency (SE), total sleep time (TST), and total wake time (TWT) were assessed using a self-report diary. RESULTS: Regression analyses revealed that pain inconsistency was associated with both SE and TWT and accounted for significant variance over age, gender, and depression. In contrast, average pain was not associated with SE, TST, or TWT. CONCLUSIONS: The findings indicate that pain inconsistency may be a more meaningful predictor of sleep disturbance than average pain level, suggesting that one's ability to regulate pain may be related to one's ability to engage in optimal sleep in mid- to late-life. CLINICAL IMPLICATIONS: Pain inconsistency appears to contribute more to sleep disturbance than average pain. Pain inconsistency in late-life warrants greater attention and may be an area of clinical intervention through activity-pacing or coping skills training.

Assessing the prognostic features of a pain classification system in advanced cancer patients.

PURPOSE: The Edmonton Classification System for Cancer Pain (ECS-CP) has been shown to predict pain management complexity based on five features: pain mechanism, incident pain, psychological distress, addictive behavior, and cognitive function. The main objective of our study was to explore the association between ECS-CP features and pain treatment outcomes among outpatients managed by a palliative care specialist-led interdisciplinary team. METHODS: Initial and follow-up clinical information of 386 eligible supportive care outpatients were retrospectively reviewed and analyzed. RESULTS: Between the initial consultation and the first follow-up visit, the median ESAS pain intensity improved from 6 to 4.5 (p < 0.0001) and the median total symptom distress score (0-100) improved from 38 to 31 (p < 0.0001). At baseline, patients with neuropathic pain (p < 0.001) and those with at least one ECS-CP feature (p = 0.006) used a higher number of adjuvant medications. At follow-up, patients with neuropathic pain were less likely to achieve their personalized pain goal (PPG) (29 vs 72%, p = 0.015). No statistically significant association was found between increasing sum of ECS-CP features and any of the pain treatment outcomes at follow-up. CONCLUSION: Neuropathy was found to be a poor prognostic feature in advanced cancer pain management. Increasing sum of ECS-CP features was not predictive of pain management complexity at the follow-up visit when pain was managed by a palliative medicine specialist. Further research is needed to further explore these observations.