Rethinking Measurement of Movement-Evoked Pain with Digital Technology

Abstract

Movement-evoked pain (MEP) may be a useful metric for phenotyping musculoskeletal pain conditions. However, there is significant disagreement over operationalization, and no studies have assessed stability of MEP over time. Fitbit and Ecological Momentary Assessment (EMA) data were collected from adults with moderate-to-severe chronic pain schedule to receive lumbar/thoracolumbar fusion surgery (N=114). On average, participants provided 323 hours of Fitbit data and 74 EMA surveys (84% completion rate). To mimic task-based assessment of MEP using the 6-minute walk test, EMA pain ratings completed within 3 hours of walking at a speed ≥70spm for at least 6 minutes were extracted. Of the full sample, 91 individuals (80%) had any instances of pain ratings following 6-minute activity bouts (Median=6, SD=11). Post-activity pain scores exhibited good within-person consistency (ICC=.76). However, between-person differences in average pain accounted for >70% of the variance in post-activity pain. MEP change scores defined as the difference between post-activity and pre-activity pain scores had poor reliability (ICC = .08). MEP change scores were not associated with average pain or factors related to the uncontrolled nature of digital assessment (e.g., activity amount, time from activity to pain report). However, MEP change scores tended to be lower when the preceding pain rating was elevated (β = -7.96, 95% Credible Interval: -9.28, -6.66), suggesting ceiling effects. Small effects of time of day and prior activity were also observed, which could contaminate MEP assessed in the lab or clinic. Continued development of digital methodologies for assessing MEP is recommended.

Keywords: Movement-evoked pain; digital assessment; digital phenotyping; ecological momentary assessment; passive sensing.

Figure 1.

Schematic overview of traditional versus ecological movement-evoked pain assessment (A) and time-varying predictors of ecological MEP observations (B). EMA = ecological momentary assessment; PA = physical activity.

Figure 2.

Variability in step quantiles (A), heart rate at step count thresholds (B), and movement-evoked pain ratings (C). Panel A shows between-person variability in step counts associated with the 50^th, 75^th, 90^th, and 100^th quantiles during non-sedentary minutes (step count ≥ 10 steps per minute). 50^th quantile refers to the individual’s average (non-sedentary) step count; 100^th quantile refers to the individual’s maximum steps per minute. Panel B shows person-level average heart rate associated with various step count thresholds (spm = steps per minute). Dots indicate person-level average heart rate for each threshold, across all available 6-minute activity bouts. For each bout, percent of maximum heart rate was calculated as observed average heart rate divided by the individual’s maximum heart rate (208 – 0.7*Age). Panel C shows person-level variability in post-activity pain recorded within 3 hours of an activity bout (defined as at least 6 consecutive minutes walking at a speed ≥ 70 steps per minute). Grey vertical lines are observed post-activity pain ratings. Shaded areas are person-level probabilities (darker = higher probability).

Figure 3.

Predictors of post-activity pain ratings. Panel A shows the between-persons association of average pain across all EMAs (x-axis) with average post-activity pain (y-axis). Panels B-F show the effect of time-varying covariates (x-axis) on within-person variability in post-activity pain ratings (y-axis). In Panels B-F, post-activity pain is person mean centered such that 0 = average for the individual. PA = physical activity. Coefficients and standard errors are available in Table S2.

Figure 4.

Predictors of MEP change scores. Panel A shows the between-persons association of average pain across all EMAs (x-axis) with average MEP change score (y-axis). Panels B-F show the effect of time-varying covariates (x-axis) on within-person variability in MEP change scores (y-axis). In Panels B-F, MEP change score is person mean centered such that 0 = average for the individual. PA = physical activity. Coefficients and standard errors are available in Table S4.

Figure 5.

Panel A is a histogram describing step counts prior to all EMA pain observations marked as indexing pain at rest (PAR). EMA pain observations were marked as indexing PAR if they were not preceded a 6+ minute activity bout, based on the step count threshold (here, 70spm). Panel B is a scatter plot of step counts (x) and PAR scores (y). Blue dots = severe pain (≥80 out of 100). Panels C and D pertain to lag-1 pain observations marked as indexing PAR in the main analysis.