Long-term Use of Wearable Health Technology by Chronic Pain Patients

OBJECTIVES

People living with chronic pain may use wearable health technology (WHT) in conjunction with an expert-directed pain management program for up to 1 year. WHT use may be associated with improvements in key patient outcomes.

METHODS

A 12-month study of WHT use among people with chronic pain was conducted, consisting of iPhone and Apple Watch applications to measure movement, sleep, and self-reported pain. Clinical outcomes among 105 patients enrolled in a multidisciplinary pain program that included WHT use were compared with 146 patients in the same program but without WHT, and to 161 patients receiving medical pain management without WHT.

RESULTS

Participants used the WHT on average 143.0 (SD: 117.6) out of 365 days. Mixed-effects models revealed participants who used WHT had decreases in depression scores (-7.83, P <0.01) and prescribed morphine milligram equivalents (-21.55, P =0.04) over 1 year. Control groups also showed decreases in depression scores (-5.08, P =0.01; -5.68, P <0.01) and morphine milligram equivalents (-18.67, P =0.01; -10.99, ns). The estimated slope of change among the WHT was not statistically different than control groups.

DISCUSSION

Patients who used WHT as part of their pain management program demonstrated a willingness to do so for extended periods of time despite living with chronic pain and other comorbidities. Data trends suggest that WHT use may positively impact depression and prescribed medication. Additional research is warranted to investigate the potential of WHT to improve the negative consequences of chronic pain.

User Engagement and Clinical Impact of the Manage My Pain App in Patients With Chronic Pain: A Real-World, Multi-site Trial

Background

Chronic pain imposes a large burden on individuals and society. A patient-centric digital chronic pain management app called Manage My Pain (MMP) can be used to enhance communication between providers and patients and promote self-management.

Objective

The purpose of this study was to evaluate the real-world engagement of patients in urban and rural settings in Ontario, Canada with the MMP app alongside their standard of care and assess the impact of its usage on clinical outcomes of pain and related mental health.

Methods

A total of 246 participants with chronic pain at a rural and 2 urban pain clinics were recruited into this prospective, open-label, exploratory study that compared the use of MMP, a digital health app for pain that incorporates validated questionnaires and provides patients with summarized reports of their progress in combination with standard care (app group), against data entered on paper-based questionnaires (nonapp group). Participants completed validated questionnaires on anxiety, depression, pain catastrophizing, satisfaction, and daily opioid consumption up to 4.5 months after the initial visit (short-term follow-up) and between 4.5 and 7 months after the initial visit (long-term follow-up). Engagement and clinical outcomes were compared between participants in the two groups.

Results

A total of 73.6% (181/246) of the participants agreed to use the app, with 63.4% (111/175) of them using it for at least one month. Individuals who used the app rated lower anxiety (reduction in Generalized Anxiety Disorder 7-item questionnaire score by 2.10 points, 95% CI –3.96 to –0.24) at short-term follow-up and had a greater reduction in pain catastrophizing (reduction in Pain Catastrophizing Scale score by 5.23 points, 95% CI –9.55 to –0.91) at long-term follow-up relative to patients with pain who did not engage with the MMP app.

Conclusions

The use of MMP by patients with chronic pain is associated with engagement and improvements in self-reported anxiety and pain catastrophizing. Further research is required to understand factors that impact continued engagement and clinical outcomes in patients with chronic pain.

Trial Registration

ClinicalTrials.gov NCT04762329; https://clinicaltrials.gov/ct2/show/NCT04762329

Interpretability and Class Imbalance in Prediction Models for Pain Volatility in Manage My Pain App Users: Analysis Using Feature Selection and Majority Voting Methods

Background

Pain volatility is an important factor in chronic pain experience and adaptation. Previously, we employed machine-learning methods to define and predict pain volatility levels from users of the Manage My Pain app. Reducing the number of features is important to help increase interpretability of such prediction models. Prediction results also need to be consolidated from multiple random subsamples to address the class imbalance issue.

Objective

This study aimed to: (1) increase the interpretability of previously developed pain volatility models by identifying the most important features that distinguish high from low volatility users; and (2) consolidate prediction results from models derived from multiple random subsamples while addressing the class imbalance issue.

Methods

A total of 132 features were extracted from the first month of app use to develop machine learning–based models for predicting pain volatility at the sixth month of app use. Three feature selection methods were applied to identify features that were significantly better predictors than other members of the large features set used for developing the prediction models: (1) Gini impurity criterion; (2) information gain criterion; and (3) Boruta. We then combined the three groups of important features determined by these algorithms to produce the final list of important features. Three machine learning methods were then employed to conduct prediction experiments using the selected important features: (1) logistic regression with ridge estimators; (2) logistic regression with least absolute shrinkage and selection operator; and (3) random forests. Multiple random under-sampling of the majority class was conducted to address class imbalance in the dataset. Subsequently, a majority voting approach was employed to consolidate prediction results from these multiple subsamples. The total number of users included in this study was 879, with a total number of 391,255 pain records.

Results

A threshold of 1.6 was established using clustering methods to differentiate between 2 classes: low volatility (n=694) and high volatility (n=185). The overall prediction accuracy is approximately 70% for both random forests and logistic regression models when using 132 features. Overall, 9 important features were identified using 3 feature selection methods. Of these 9 features, 2 are from the app use category and the other 7 are related to pain statistics. After consolidating models that were developed using random subsamples by majority voting, logistic regression models performed equally well using 132 or 9 features. Random forests performed better than logistic regression methods in predicting the high volatility class. The consolidated accuracy of random forests does not drop significantly (601/879; 68.4% vs 618/879; 70.3%) when only 9 important features are included in the prediction model.

Conclusions

We employed feature selection methods to identify important features in predicting future pain volatility. To address class imbalance, we consolidated models that were developed using multiple random subsamples by majority voting. Reducing the number of features did not result in a significant decrease in the consolidated prediction accuracy.

Outcome of a High-Frequency Transcutaneous Electrical Nerve Stimulator (hfTENS) Device for Low Back Pain: A Randomized Controlled Trial

OBJECTIVE

This controlled trial examined the benefit of a high-frequency transcutaneous electrical nerve stimulation (hfTENS) device (the Quell, NeuroMetrix, Inc., Waltham, MA, U.S.A.) for patients with chronic low back pain (CLBP).

METHODS

Thirty-five (n = 35) participants were randomly assigned to use the device each day for 3 months (experimental group) and were compared with 33 subjects without the device (treatment-as-usual control group). All patients were instructed to complete baseline questionnaires and were assessed on thresholds of pressure pain and mechanical temporal summation as part of standardized quantitative sensory testing (QST). The subjects also uploaded smartphone applications (apps) for tracking use of the hfTENS and for daily pain assessment. Each participant completed weekly phone interviews, was prompted to complete daily pain app assessments, and was asked to repeat the baseline questionnaires again after 6 weeks and 3 months.

RESULTS

Sixty percent of the subjects were female, 77.9% were Caucasian, and the average age was 46.2 years. Significant reductions in pain intensity (P < 0.01) and activity interference (P < 0.025) and significant improvements in pain catastrophizing (P < 0.025) were noted in the experimental group compared with the control group. No group differences were found on depression, anxiety, or pain-related disability. Older subjects with a longer duration of pain tended to use the hfTENS more often. Subjects who showed greater sensitivity based on QST results revealed increased use of the hfTENS (P < 0.025) and tended to believe that the hfTENS was more helpful in reducing their back pain, but these findings did not reach significance (P = 0.09).

CONCLUSION

These results suggest that hfTENS can have a moderate effect in reducing pain and improving quality of life in low back pain patients. Further trials designed to determine the mechanism of action of the hfTENS are needed.

Defining and Predicting Pain Volatility in Users of the Manage My Pain App: Analysis Using Data Mining and Machine Learning Methods

Background

Measuring and predicting pain volatility (fluctuation or variability in pain scores over time) can help improve pain management. Perceptions of pain and its consequent disabling effects are often heightened under the conditions of greater uncertainty and unpredictability associated with pain volatility.

Objective

This study aimed to use data mining and machine learning methods to (1) define a new measure of pain volatility and (2) predict future pain volatility levels from users of the pain management app, Manage My Pain, based on demographic, clinical, and app use features.

Methods

Pain volatility was defined as the mean of absolute changes between 2 consecutive self-reported pain severity scores within the observation periods. The k-means clustering algorithm was applied to users’ pain volatility scores at the first and sixth month of app use to establish a threshold discriminating low from high volatility classes. Subsequently, we extracted 130 demographic, clinical, and app usage features from the first month of app use to predict these 2 volatility classes at the sixth month of app use. Prediction models were developed using 4 methods: (1) logistic regression with ridge estimators; (2) logistic regression with Least Absolute Shrinkage and Selection Operator; (3) Random Forests; and (4) Support Vector Machines. Overall prediction accuracy and accuracy for both classes were calculated to compare the performance of the prediction models. Training and testing were conducted using 5-fold cross validation. A class imbalance issue was addressed using a random subsampling of the training dataset. Users with at least five pain records in both the predictor and outcome periods (N=782 users) are included in the analysis.

Results

k-means clustering algorithm was applied to pain volatility scores to establish a threshold of 1.6 to differentiate between low and high volatility classes. After validating the threshold using random subsamples, 2 classes were created: low volatility (n=611) and high volatility (n=171). In this class-imbalanced dataset, all 4 prediction models achieved 78.1% (611/782) to 79.0% (618/782) in overall accuracy. However, all models have a prediction accuracy of less than 18.7% (32/171) for the high volatility class. After addressing the class imbalance issue using random subsampling, results improved across all models for the high volatility class to greater than 59.6% (102/171). The prediction model based on Random Forests performs the best as it consistently achieves approximately 70% accuracy for both classes across 3 random subsamples.

Conclusions

We propose a novel method for measuring pain volatility. Cluster analysis was applied to divide users into subsets of low and high volatility classes. These classes were then predicted at the sixth month of app use with an acceptable degree of accuracy using machine learning methods based on the features extracted from demographic, clinical, and app use information from the first month.

Determining Pain Catastrophizing From Daily Pain App Assessment Data: Role of Computer-Based Classification

This study compared persons with chronic pain who consistently reported that their pain was worsening with those who reported that their pain was improving or remaining the same per daily assessment data from a smartphone pain app. All participants completed baseline measures and were asked to record their progress every day by answering whether their overall condition had improved, remained the same, or gotten worse (perceived change) on a visual analogue scale. One hundred forty-four individuals with chronic pain who successfully entered daily assessments were included. Those persons who were classified as worse showed significantly higher pain intensity scores, greater activity interference, higher disability and mood disturbance scores, and higher scores on the Pain Catastrophizing Scale both at baseline and after 3 months (P < .001). Repeated measures analyses and multilevel modeling of perceived change data over different time intervals of 20 assessments over 40 days, 10 assessments over 20 days, and 5 assessments over 10 days were examined. These analyses demonstrated that group classification of better, same, and worse could be reliably determined, even with as few as 5 assessments. These results support the use of innovative mobile health technology to identify individuals who are prone to catastrophize about their pain. Perspective: This study demonstrated that daily assessment of overall perceived change with a smartphone pain app was positively correlated with the Pain Catastrophizing Scale and capturing short-term daily assessment trends data using computer-based classification methods might be a future way to help to identify individuals who tend to catastrophize about their pain.

A Pilot Comparison of a Smartphone App With or Without 2-Way Messaging Among Chronic Pain Patients: Who Benefits From a Pain App?

OBJECTIVES

The overall aim of this study was to determine the effect of introducing a smartphone pain application (app), for both Android and iPhone devices that enables chronic pain patients to assess, monitor, and communicate their status to their providers.

METHODS

This study recruited 105 chronic pain patients to use a smartphone pain app and half of the patients (N=52) had 2-way messaging available through the app. All patients completed baseline measures and were asked to record their progress every day for 3 months, with the opportunity to continue for 6 months. All participants were supplied a Fitbit to track daily activity. Summary line graphs were posted to each of the patients' electronic medical records and physicians were notified of their patient's progress.

RESULTS

Ninety patients successfully downloaded the pain app. Average age of the participants was 47.1 (range, 18 to 72), 63.8% were female and 32.3% reported multiple pain sites. Adequate validity and reliability was found between the daily assessments and standardized questionnaires (r=0.50) and in repeated daily measures (pain, r=0.69; sleep, r=0.83). The app was found to be easily introduced and well tolerated. Those patients assigned to the 2-way messaging condition on average tended to use the app more and submit more daily assessments (95.6 vs. 71.6 entries), but differences between groups were not significant. Pain-app satisfaction ratings overall were high.

DISCUSSION

This study highlights some of the challenges and benefits in utilizing smartphone apps to manage chronic pain patients, and provides insight into those individuals who might benefit from mHealth technology.

Patterns of User Engagement With the Mobile App, Manage My Pain: Results of a Data Mining Investigation

BACKGROUND

Pain is one of the most prevalent health-related concerns and is among the top 3 most common reasons for seeking medical help. Scientific publications of data collected from pain tracking and monitoring apps are important to help consumers and healthcare professionals select the right app for their use.

OBJECTIVE

The main objectives of this paper were to (1) discover user engagement patterns of the pain management app, Manage My Pain, using data mining methods; and (2) identify the association between several attributes characterizing individual users and their levels of engagement.

METHODS

User engagement was defined by 2 key features of the app: longevity (number of days between the first and last pain record) and number of records. Users were divided into 5 user engagement clusters employing the k-means clustering algorithm. Each cluster was characterized by 6 attributes: gender, age, number of pain conditions, number of medications, pain severity, and opioid use. Z tests and chi-square tests were used for analyzing categorical attributes. Effects of gender and cluster on numerical attributes were analyzed using 2-way analysis of variances (ANOVAs) followed up by pairwise comparisons using Tukey honest significant difference (HSD).

RESULTS

The clustering process produced 5 clusters representing different levels of user engagement. The proportion of males and females was significantly different in 4 of the 5 clusters (all P ≤.03). The proportion of males was higher than females in users with relatively high longevity. Mean ages of users in 2 clusters with high longevity were higher than users from other 3 clusters (all P <.001). Overall, males were significantly older than females (P <.001). Across clusters, females reported more pain conditions than males (all P <.001). Users from highly engaged clusters reported taking more medication than less engaged users (all P <.001). Females reported taking a greater number of medications than males (P =.04). In 4 of 5 clusters, the percentage of males taking an opioid was significantly greater (all P ≤.05) than that of females. The proportion of males with mild pain was significantly higher than that of females in 3 clusters (all P ≤.008).

CONCLUSIONS

Although most users of the app reported being female, male users were more likely to be highly engaged in the app. Users in the most engaged clusters self-reported a higher number of pain conditions, a higher number of current medications, and a higher incidence of opioid usage. The high engagement by males in these clusters does not appear to be driven by pain severity which may, in part, be the case for females. Use of a mobile pain app may be relatively more attractive to highly-engaged males than highly-engaged females, and to those with relatively more complex chronic pain problems.