Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 73 of the 73 references it cites were found in a registry or an index.

Personalized Machine Learning-Based Prediction of Wellbeing and Empathy in Healthcare Professionals

Open access

Abstract

Healthcare professionals are known to suffer from workplace stress and burnout, which can negatively affect their empathy for patients and quality of care. While existing research has identified factors associated with wellbeing and empathy in healthcare professionals, these efforts are typically focused on the group level, ignoring potentially important individual differences and implications for individualized intervention approaches. In the current study, we implemented N-of-1 personalized machine learning (PML) to predict wellbeing and empathy in healthcare professionals at the individual level, leveraging ecological momentary assessments (EMAs) and smartwatch wearable data. A total of 47 mood and lifestyle feature variables (relating to sleep, diet, exercise, and social connections) were collected daily for up to three months followed by applying eight supervised machine learning (ML) models in a PML pipeline to predict wellbeing and empathy separately. Predictive insight into the model architecture was obtained using Shapley statistics for each of the best-fit personalized models, ranking the importance of each feature for each participant. The best-fit model and top features varied across participants, with anxious mood (13/19) and depressed mood (10/19) being the top predictors in most models. Social connection was a top predictor for wellbeing in 9/12 participants but not for empathy models (1/7). Additionally, empathy and wellbeing were the top predictors of each other in 64% of cases. These findings highlight shared and individual features of wellbeing and empathy in healthcare professionals and suggest that a one-size-fits-all approach to addressing modifiable factors to improve wellbeing and empathy will likely be suboptimal. In the future, such personalized models may serve as actionable insights for healthcare professionals that lead to increased wellness and quality of patient care.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/s24082640

Cites

73 of 73 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Personalized machine learning pipeline flowchart adapted from [32]. This pipeline is run for each subject with a target variable set to wellbeing and empathy.
Figure 1 Personalized machine learning pipeline flowchart adapted from [32]. This pipeline is run for each subject with a target variable set to wellbeing and empathy.CC BY · PubMed Central
Figure 2 Results from wellbeing and empathy models. (A,B). Mean absolute percentage error (MAPE) of wellbeing and empathy models, respectively. Mean and std are shown from nested cross-validation for each subject and the average across all participants. The dashed line at 0.2 as a threshold for good MAPE [57]. (C,D). Combined histogram of all predicted values for wellbeing and empathy across all participants. Blue and red represents actual and predicted values respectively. (E,F). Heatmap of actual vs. predicted values for wellbeing and empathy.
Figure 2 Results from wellbeing and empathy models. (A,B). Mean absolute percentage error (MAPE) of wellbeing and empathy models, respectively. Mean and std are shown from nested cross-validation for each subject and the average across all participants. The dashed line at 0.2 as a threshold for good MAPE [57]. (C,D). Combined histogram of all predicted values for wellbeing and empathy across all participants. Blue and red represents actual and predicted values respectively. (E,F). Heatmap of actual vs. predicted values for wellbeing and empathy.CC BY · PubMed Central
Figure 3 Spearman rank correlation coefficients with 95% confidence interval bounds for predicted vs. actual wellbeing and empathy scores. Composite Spearman’s rho was also calculated using all subject’s data indicated by the “overall” marker. In total, 10/11 and 4/7 participants showed significant correlations between predicted and actual data for wellbeing and empathy, respectively.
Figure 3 Spearman rank correlation coefficients with 95% confidence interval bounds for predicted vs. actual wellbeing and empathy scores. Composite Spearman’s rho was also calculated using all subject’s data indicated by the “overall” marker. In total, 10/11 and 4/7 participants showed significant correlations between predicted and actual data for wellbeing and empathy, respectively.CC BY · PubMed Central
Figure 4 SHapley Additive Explainer (SHAP) plots across participants comparing wellbeing and empathy models. Rank feature importance and feature effects are shown. P28’s wellbeing slide shows timeshap [58] rank feature importance for their LSTM best-fit model. Individual effects of each feature on the dependent variable of wellbeing/empathy are depicted using the colored points to the right of the bar graph from low (blue) to high (pink) feature values relative to the outcome plotted on the x-axis (from low to high wellbeing/empathy going from left to right). For example, the P-6 Empathy Shapley plot shows that their top-ranking predictors of empathy are mild exercise over the past day followed by sleep satisfaction; since the color dots for mild exercise progress from pink to blue from left to right, lesser mild exercise supports greater patient empathy in this case but for sleep, greater sleep satisfaction (color dots progressing from blue to pink) supports greater patient empathy.
Figure 4 SHapley Additive Explainer (SHAP) plots across participants comparing wellbeing and empathy models. Rank feature importance and feature effects are shown. P28’s wellbeing slide shows timeshap [58] rank feature importance for their LSTM best-fit model. Individual effects of each feature on the dependent variable of wellbeing/empathy are depicted using the colored points to the right of the bar graph from low (blue) to high (pink) feature values relative to the outcome plotted on the x-axis (from low to high wellbeing/empathy going from left to right). For example, the P-6 Empathy Shapley plot shows that their top-ranking predictors of empathy are mild exercise over the past day followed by sleep satisfaction; since the color dots for mild exercise progress from pink to blue from left to right, lesser mild exercise supports greater patient empathy in this case but for sleep, greater sleep satisfaction (color dots progressing from blue to pink) supports greater patient empathy.CC BY · PubMed Central
Figure 5 Percentages of models with specific mood and domain variables as top predictors, depression, anxiety, social connection, sleep, diet, breath attention, and gratitude. The first row, Empathy–Wellbeing, indicates percentages of wellbeing models with empathy as a top predictor and percentages of empathy models with wellbeing as a top predictor. Wellbeing model percentages are out of 12 total models except for the empathy variable which is out of 7. Empathy models are out of 7.
Figure 5 Percentages of models with specific mood and domain variables as top predictors, depression, anxiety, social connection, sleep, diet, breath attention, and gratitude. The first row, Empathy–Wellbeing, indicates percentages of wellbeing models with empathy as a top predictor and percentages of empathy models with wellbeing as a top predictor. Wellbeing model percentages are out of 12 total models except for the empathy variable which is out of 7. Empathy models are out of 7.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

1 more claim about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 72 references

Research works printed but not linked: 2

  1. Misra, A., Ojeda, A., and Mishra, J. (2018). BrainE: A Digital Platform for Evaluating, Engaging and Enhancing Brain Function, University of California Press. Regents of the University of California Copyright SD2018-816.
  2. Ashforth, 1997, Burnout as a Process: Commentary on Cordes, Dougherty and Blum, J. Organ. Behav., 18, 703, 10.1002/(SICI)1099-1379(199711)18:6<703::AID-JOB847>3.0.CO;2-1 · DOI

Cited by

See all 22

Metadata

Authors
Nan, J., Herbert, M. S., Purpura, S., Henneken, A. N., Ramanathan, D., & Mishra, J.
Affiliations
  • University of California San Diego (US)
  • VA San Diego Healthcare System (US)
Corresponding author
Nan, Jason
Venue
Sensors
Publisher
MDPI AG
Published
2024-04-20
Volume
24
Issue
8
Pages
2640
ISSN
1424-8220 (ISSN-L)
OpenAlex
W4395002494
Cited by
22
References
72
Citations by year
2026: 10 · 2025: 10 · 2024: 2
FWCI
12.92
Type
journal article
Language
en
Open access
Locations
  • Sensors (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
Funders
  • T. Denny Sanford Institute for Empathy and Compassion
  • Hope for Depression Research Foundation
  • Stein Institute for Research on Aging
Topics
  • Health Sciences › Health Professions › General Health Professions › Healthcare professionals’ stress and burnout
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Social Sciences › Psychology › Social Psychology › Stress and Burnout Research
Keywords
healthcare workers, well-being, empathy, burnout, occupational stress, personalized machine learning, ecological momentary assessment, mood, social connectedness, sleep, diet, physical activity, individual differences, Shapley value, quality of care
MeSH terms
Machine Learning, Wearable Electronic Devices, Adult, Empathy, Female, Health Personnel, Humans, Male, Middle Aged

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Personalized Machine Learning-Based Prediction of Wellbeing and Empathy in Healthcare Professionals. https://app.cochat.ai/papers/10.3390/s24082640
The paper
Nan, J., Herbert, M. S., Purpura, S., Henneken, A. N., Ramanathan, D., & Mishra, J. (2024). Personalized Machine Learning-Based Prediction of Wellbeing and Empathy in Healthcare Professionals. Sensors. https://doi.org/10.3390/s24082640

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.