Evaluating Wearable Electrodermal Activity as a Candidate Digital Biomarker of Occupational Well-Being: Longitudinal Real-World Study

75 Finnish knowledge workers wore a Nuanic EDA ring for two 14-day waves six months apart with 3-hourly mood (valence) prompts. Greater within-night variability of the ring's proprietary DNE arousal index went with higher burnout and lower engagement, and adding DNE plus valence significantly improved model fit; the authors stress the index is proprietary and not validated as a burnout measure.

Background

Wearable physiological sensing and mobile self-report methods offer new opportunities for monitoring occupational well-being in real-world settings. However, interpreting continuous physiological signals in relation to psychological states such as job burnout and work engagement remains challenging, particularly when using deployable systems that rely on proprietary signal processing.

Objective

This study examined a proprietary electrodermal activity (EDA)–derived arousal index as a candidate digital biomarker for research on occupational well-being. We examined associations between job burnout, work engagement, and sympathetic activation indexed by a device-derived EDA metric, and assessed whether combining physiological data with mobile-based ecological momentary assessment (mEMA) of valence provides complementary information.

Methods

A total of 75 knowledge workers participated, with sufficient paired physiological and mEMA data obtained from 57 participants in wave 1 and 51 in wave 2 (total observations=108). Participants were monitored continuously over 14 consecutive days at 2 time points 6 months apart. Autonomic arousal was operationalized using the device-derived double normalized electrodermal activity (DNE) index obtained from wearable EDA recordings. Valence was assessed using mEMA. Associations between physiological activation, valence, and occupational well-being outcomes were analyzed using generalized least squares models with a compound-symmetry covariance structure to account for repeated observations within participants.

Results

The wearable-derived DNE index showed a clear diurnal pattern, with higher values during daytime and lower values during nighttime. Greater within-nighttime DNE variability was associated with higher burnout (estimate=0.02, 95% CI 0.01-0.03; P P =.009). The daily DNE × valence interaction was also associated with engagement (estimate=−0.35, 95% CI −0.69 to −0.00; P =.047). Removing all DNE and valence variables significantly worsened model fit for both burnout ( χ ² ₁₀ =28.5; P =.002) and engagement ( χ ² ₁₀ =23.7; P =.009), supporting their combined explanatory value. Model-based profiles further showed that negative valence was associated with higher predicted burnout and lower predicted engagement than positive valence across both high- and low-arousal scenarios, although prediction intervals were wide and partially overlapping. Model comparisons showed that including physiological and mEMA-based variables improved explanatory power.

Conclusions

These findings provide practical evidence that continuous wearable EDA monitoring can be combined with mEMA in naturalistic occupational settings. Associations between the proprietary DNE index, valence, burnout, and work engagement provide preliminary convergent evidence supporting further evaluation of DNE as a candidate digital biomarker in occupational well-being research. Interpretation remains constrained by the proprietary signal-processing pipeline, and this study does not establish construct-specific validity, prospective predictive performance, screening or diagnostic validity, sensitivity to change, or individual-level monitoring usage.

Verbatim from the publisher's DOI record (opens in a new tab) · Abstract as deposited by the publisher with Crossref; © the authors/publisher

No results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.