Pain interference may be an important link between pain severity, impairment, and self-reported disability in participants with wrist/hand pain

Predicting Clinical Outcomes at the Toronto General Hospital Transitional Pain Service via the Manage My Pain App: Machine Learning Approach

Background

Chronic pain is a complex condition that affects more than a quarter of people worldwide. The development and progression of chronic pain are unique to each individual due to the contribution of interacting biological, psychological, and social factors. The subjective nature of the experience of chronic pain can make its clinical assessment and prognosis challenging. Personalized digital health apps, such as Manage My Pain (MMP), are popular pain self-tracking tools that can also be leveraged by clinicians to support patients. Recent advances in machine learning technologies open an opportunity to use data collected in pain apps to make predictions about a patient's prognosis.

Objective

This study applies machine learning methods using real-world user data from the MMP app to predict clinically significant improvements in pain-related outcomes among patients at the Toronto General Hospital Transitional Pain Service.

Methods

Information entered into the MMP app by 160 Transitional Pain Service patients over a 1-month period, including profile information, pain records, daily reflections, and clinical questionnaire responses, was used to extract 245 relevant variables, referred to as features, for use in a machine learning model. The machine learning model was developed using logistic regression with recursive feature elimination to predict clinically significant improvements in pain-related pain interference, assessed by the PROMIS Pain Interference 8a v1.0 questionnaire. The model was tuned and the important features were selected using the 10-fold cross-validation method. Leave-one-out cross-validation was used to test the model's performance.

Results

The model predicted patient improvement in pain interference with 79% accuracy and an area under the receiver operating characteristic curve of 0.82. It showed balanced class accuracies between improved and nonimproved patients, with a sensitivity of 0.76 and a specificity of 0.82. Feature importance analysis indicated that all MMP app data, not just clinical questionnaire responses, were key to classifying patient improvement.

Conclusions

This study demonstrates that data from a digital health app can be integrated with clinical questionnaire responses in a machine learning model to effectively predict which chronic pain patients will show clinically significant improvement. The findings emphasize the potential of machine learning methods in real-world clinical settings to improve personalized treatment plans and patient outcomes.

Gender Discrimination is Associated with Greater Chronic Pain Interference Among Women

Pain disparities between men and women are found in multiple domains; women have been shown to experience greater pain intensity, pain disability, and risk for chronic pain. While often ascribed to biological differences, recent research has demonstrated the significance of social determinants of gendered pain disparities. Gender discrimination is one factor that disproportionally affects women and has been associated with adverse health outcomes, yet has received less attention in pain research. Discrimination is intrusive and stressful, and may exacerbate the extent to which chronic pain interferes with life. Prior work has shown that among women, general experiences of discrimination are indirectly associated with pain interference through perceived stress. However, the direct relationship between gender discrimination specifically and pain interference has not been explored. Here, using data from the Midlife in the United States national survey, we first assessed the relationship between daily experiences of discrimination due to any aspect of identity and pain interference in those with chronic pain. We further explored whether discrimination due to gender specifically was associated with pain interference among women. Results indicated that daily discrimination was associated with greater pain interference within the whole sample; however, within-group analyses found that this relationship was only significant for women, and not men. Exploring further within women only, discrimination due to gender predicted greater pain interference, controlling for health-related covariates. These findings support recent calls for probes into the role of discrimination on health outcomes and suggests that experiencing discrimination contributes to disruption of life and pain disparities. PERSPECTIVE: The findings presented here advance our understanding of the harmful impact of discrimination on pain outcomes, broadening its scope by providing evidence regarding the association between gender discrimination and pain interference. Considering known pain disparities between men and women, we discuss potential insight into mechanisms contributing to this burden.

Comparative responsiveness of the PROMIS-29 and SF-36 instruments in individuals with chronic musculoskeletal pain

BACKGROUND

Chronic musculoskeletal pain (CMP) impacts Health-Related Quality of Life (HRQoL). Several questionnaires are widely used for evaluating the HRQoL, such as Patient Reported Outcomes Measurement Information System 29-item Health Profile (PROMIS-29) and 36-Item Short Form Health Survey (SF-36). This study aimed to assess and compare the responsiveness of PROMIS-29 and SF-36 in individuals with CMP.

METHODS

The prospective study collected data from 215 patients with CMP. The participants completed both questionnaires at baseline and a 4-week follow-up with the global perceived effect (GPE) scale. The internal and external responsiveness methods evaluated the responsiveness of the instruments. The effect size (ES) and standardized response mean (SRM) assessed the internal responsiveness. External responsiveness was measured by Spearman's correlation coefficients (r) and area under the receiver operating curve (AUC). The responsiveness of the PROMIS-29 and SF-36 were compared in similar constructs.

RESULTS

The PROMIS-29 demonstrated responsiveness as moderate in fatigue (ES = -0.79, SRM = -0.64), moderate to large in pain interference (ES = -0.94, SRM = -0.64), and large effect in pain intensity subscale (ES = -1.25, SRM = -1.16). The SF-36 bodily pain exhibited moderate responsiveness (ES = 0.73, SRM = 0.60). Spearman's correlation showed moderate between GPE and change scores of PROMIS-29 physical function (r = 0.31), fatigue (r = -0.36), pain intensity (r = 0.45), and weak between GPE and changes score of SF-36 (r = 0.15-0.26). Besides, the AUC of the PROMIS-29 fatigue, and pain intensity scale were 0.701 and 0.725, respectively, indicating superior discriminative ability.

CONCLUSIONS

The PROMIS-29 and SF-36 were the most responsive in assessing the pain domain for physical health and PROMIS-29 fatigue domain for mental health among individuals with CMP. The PROMIS-29 demonstrated superior results compared to the SF-36. Other domains related to HRQoL could be considered in other assessment instruments.

Symptom characteristics, perceived causal attributions, and contextual factors influencing self-care behaviors: An ecological daily assessment study of adults with chronic illness

OBJECTIVE

Insights into how symptoms influence self-care can guide patient education and improve symptom control. This study examined symptom characteristics, causal attributions, and contextual factors influencing self-care of adults with arthritis, asthma, chronic obstructive pulmonary disease, diabetes, or heart failure.

METHODS

Adults (n = 81) with a symptomatic chronic illness participated in a longitudinal observational study. Using Ecological Daily Assessment, participants described one symptom twice daily for two weeks, rating its frequency, severity, bothersomeness, duration, causes, and self-care.

RESULTS

The most frequent symptoms were fatigue and shortness of breath. Pain, fatigue, and joint stiffness were the most severe and bothersome. Most participants engaged in active self-care, but those with fatigue and pain engaged in passive self-care (i.e., rest or do nothing), especially when symptoms were infrequent, mild, somewhat bothersome, and fleeting. In people using passive self-care, thoughts, feelings, and the desire to conceal symptoms from others interfered with self-care.

CONCLUSION

Most adults with a chronic illness take an active role in managing their symptoms but some conceal or ignore symptoms until the frequency, severity, bothersomeness, or duration increases.

PRACTICE IMPLICATIONS

When patients report symptoms, asking about self-care behaviors may reveal inaction or ineffective approaches. A discussion of active self-care options may improve symptom control.

Enhancing Chronic Non-Cancer Pain Management: A Systematic Review of Mindfulness Therapies and Guided Imagery Interventions

Background and Objectives: There has been an increasing interest in the use of non-pharmacological approaches for the multidimensional treatment of chronic pain. The aim of this systematic review was to assess the effectiveness of mindfulness-based therapies and Guided Imagery (GI) interventions in managing chronic non-cancer pain and related outcomes. Materials and Methods: Searching three electronic databases (Web of Science, PubMed, and Scopus) and following the PRISMA guidelines, a systematic review was performed on Randomized Controlled Trials (RCTs) and pilot RCTs investigating mindfulness or GI interventions in adult patients with chronic non-cancer pain. The Cochrane Risk of Bias Tool was utilized to assess the quality of the evidence, with outcomes encompassing pain intensity, opioid consumption, and non-sensorial dimensions of pain. Results: Twenty-six trials met the inclusion criteria, with most of them exhibiting a moderate to high risk of bias. A wide diversity of chronic pain types were under analysis. Amongst the mindfulness interventions, and besides the classical programs, Mindfulness-Oriented Recovery Enhancement (MORE) emerges as an approach that improves interoception. Six trials demonstrated that mindfulness techniques resulted in a significant reduction in pain intensity, and three trials also reported significant outcomes with GI. Evidence supports a significant improvement in non-sensory dimensions of pain in ten trials using mindfulness and in two trials involving GI. Significant effects on opioid consumption were reported in four mindfulness-based trials, whereas one study involving GI found a small effect with that variable. Conclusions: This study supports the evidence of benefits of both mindfulness techniques and GI interventions in the management of chronic non-cancer pain. Regarding the various mindfulness interventions, a specific emphasis on the positive results of MORE should be highlighted. Future studies should focus on specific pain types, explore different durations of the mindfulness and GI interventions, and evaluate emotion-related outcomes.

Evaluating Factors Associated With Patient-Perceived Applicability of the QuickDASH Survey in Reflecting Their Goals of Care Prior to Hand Surgery

PURPOSE

Patient-reported outcome measures intend to capture patients' perspectives on their health status. However, the patient-perceived applicability of many of these patient-reported outcome measures is unknown. We hypothesized that patients experiencing greater upper extremity disability and greater pain interference would be more likely to report that the Quick Disabilities of the Arm, Shoulder, and Hand questionnaire (QuickDASH) survey content is responsive to their daily lives and goals in seeking surgical care.

METHODS

Adult preoperative hand surgery patients at a single tertiary academic center were recruited prospectively. QuickDASH, Patient-Reported Outcomes Measurement Information System Pain Interference computerized-adaptive-testing, and the Godin Leisure-Time Exercise Questionnaire (GLTEQ)-a validated adult physical activity level metric-data were collected. The following two Likert response questions were also asked: question (1) "How applicable is the above questionnaire to your treatment goals for your upper extremity condition?" and question (2) "How applicable is the overall questionnaire to your daily life?" Multivariable binary logistic regression was performed to define the factors associated with patients reporting that the survey was "very applicable."

RESULTS

Of the 133 included patients, the mean age was 49 ± 18 years, 40% were women, and the mean GLTEQ score was 54.1 ± 5.5 (consistent with a high level of activity). For questions 1 and 2, 32% and 29% of the patients reported that QuickDASH was "very applicable," respectively. The multivariable model demonstrated that for every 10-point increase in QuickDASH, there was a 45% to 49% greater odds of respondents reporting that the survey was "very applicable," and for every 5-point increase in pain interference computerized-adaptive-testing, the odds increased by 55% to 70%. No association with GLTEQ was observed.

CONCLUSIONS

Patients with greater upper extremity disability and pain interference were more likely to find the QuickDASH content to be applicable to their daily lives and goals in seeking surgical care.

CLINICAL RELEVANCE

These findings suggest that QuickDASH may not be an optimal instrument when evaluating upper extremity function in cohorts with mild disability and low pain interference.

Hand therapists' knowledge and practice-related beliefs about pain science: A survey study

INTRODUCTION

Identifying hand therapists' knowledge and beliefs about pain can illuminate familiarity with modern pain science within hand therapy.

PURPOSE OF THE STUDY

The primary aim was to identify hand therapists' knowledge of pain neurophysiology. Secondary purposes were to explore demographic variation in knowledge, describe practice-related beliefs about pain science, and explore associations between knowledge and beliefs.

STUDY DESIGN

Cross-sectional descriptive survey study.

METHODS

An electronic survey, including the Revised Neurophysiology of Pain Questionnaire (R-NPQ) and Likert-type questions about practice-related beliefs, was distributed to American Society of Hand Therapists members.

RESULTS

Data from 305 survey responses were analyzed. R-NPQ accuracy ranged from 42% to 100%, with a mean of 75% (9/12 ± 1.5). Certified hand therapists scored, on average, 0.8 points lower than their noncertified peers. Participants with a doctoral degree scored 0.7 or 0.6 points higher, respectively, than those with a bachelor's or master's degree. Objective knowledge of pain neurophysiology was positively associated with perceived knowledge of pain science (ρ = .31, P < .001). Associations between R-NPQ and perceived importance of knowing pain science; confidence in pain-related evaluation, treatment, and education; and frequency of incorporating pain science principles into practice were small but statistically significant (ρ = .12-.25, P = <.001-.04).

CONCLUSIONS

Although hand therapists recognized the importance of knowing pain science, they had objective and subjective limitations in that knowledge. Specific errors in their R-NPQ responses suggest misconceptions related to the modern differentiation between nociception and pain. Blurring of these constructs may relate to participants' self-reported practice emphasis on acute versus chronic conditions. Future studies should explore knowledge, attitudes, and beliefs about pain beyond R-NPQ scores to understand variation in practice and training needs.

Depressive symptoms and health-related quality of life in a heterogeneous psychiatric sample: conditional indirect effects of pain severity and interference

BACKGROUND

Few studies have examined clinically relevant mechanisms that underlie the association between two important indices of recovery- depression severity and health-related quality of life (HRQOL) in psychiatric outpatients. This study aimed to explicate the roles of pain interference and pain severity as mediating and moderating mechanisms in the relationship between depressive symptoms and HRQOL.

METHODS

Data from 290 outpatients diagnosed with schizophrenia (n = 102), depressive (n = 98), and anxiety (n = 90) disorders were examined. Participants completed a set of questionnaires that queried their sociodemographic statuses, current pain severity and interference levels, depression severity levels, and HRQOL. Subsequently, mediation and moderation analyses were conducted.

RESULTS

Analyses revealed that pain interference fully mediated the relationship between depressive symptoms and physical (34% of the total effect) but not mental HRQOL. At high pain levels (+ 1 SD from mean), depressive symptoms may interfere with physical quality of life through pain interference, but this was not present at low pain levels (- 1 SD from mean).

CONCLUSIONS

Prolonged pain symptoms could negatively influence psychiatric recovery beyond the physical aspect of HRQOL. These results thus imply a need to detect and manage severe physical pain complaints at the acute stage in psychiatric outpatients.