Using mobile sensing data to assess stress: Associations with perceived and lifetime stress, mental health, sleep, and inflammation

Abstract

Background Although stress is a risk factor for mental and physical health problems, it can be difficult to assess, especially on a continual, non-invasive basis. Mobile sensing data, which are continuously collected from naturalistic smartphone use, may estimate exposure to acute and chronic stressors that have health-damaging effects. This initial validation study validated a mobile-sensing collection tool against assessments of perceived and lifetime stress, mental health, sleep duration, and inflammation. Methods Participants were 25 well-characterized healthy young adults ( Mage = 20.64 years, SD = 2.74; 13 men, 12 women). We collected affective text language use with a custom smartphone keyboard. We assessed participants’ perceived and lifetime stress, depression and anxiety levels, sleep duration, and basal inflammatory activity (i.e. salivary C-reactive protein and interleukin-1β). Results Three measures of affective language (i.e. total positive words, total negative words, and total affective words) were strongly associated with lifetime stress exposure, and total negative words typed was related to fewer hours slept (all large effect sizes: r = 0.50 – 0.78). Total positive words, total negative words, and total affective words typed were also associated with higher perceived stress and lower salivary C-reactive protein levels (medium effect sizes; r = 0.22 – 0.32). Conclusions Data from this initial longitudinal validation study suggest that total and affective text use may be useful mobile sensing measures insofar as they are associated with several other stress, mental health, behavioral, and biological outcomes. This tool may thus help identify individuals at increased risk for stress-related health problems.

Opportunities for Smartphone Sensing in E-Health Research: A Narrative Review

“…The majority of articles (59%) used smartphone sensing to infer mental health conditions. While six articles examined overall mental health [ 13 , 24 , 25 , 26 , 27 , 68 ], other studies examined specific factors such as mood [ 15 , 16 , 17 , 18 , 19 , 20 , 21 , 69 , 70 , 71 , 72 , 73 , 74 , 75 ] and stress [ 76 , 77 , 78 ]. Additionally, studies also examined specific mental health conditions such as depression [ 7 , 8 , 67 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 ], schizophrenia [ 90 , 91 , 92 , 93 ], and bipolar disorder [ 94 , 95 ].…”

“…Besides these conventional types of sensor and software features, several used less common approaches to infer health conditions. For example, two studies used custom keyboards to collect keystroke/keypress data—one to analyze the sentiment of typed text [ 78 ], and the other to infer alcohol intoxication (greater duration between consecutive keypresses, more text deletions and insertions) [ 62 ]. One study used the barometer sensor for activity recognition and found that it reduced the misclassification of stair-climbing/descending activity [ 11 ].…”

“…They collected limited samples to infer higher-level attributes such as amplitude levels [ 69 , 84 ] and presence/absence of voice [ 3 , 84 , 86 ]. Similarly, the study using keyboard presses only collected every third word and did not collect any passwords/credit card information [ 78 ]. Such approaches should be considered to minimize privacy of the users, which is a key factor for user participation and acceptance [ 99 , 107 ].…”

Cumulative lifetime acute stressor exposure interacts with reward responsiveness to predict longitudinal increases in depression severity in adolescence

“…A second possibility involves the immune system. Greater lifetime stressor exposure has been found to predict higher levels of inflammatory activity (Byrne et al, 2021 ), and a core biobehavioral effect of inflammation involves the reduction of reward-related neural signaling and behavior (Felger et al, 2016 ; Slavich, 2020 , 2022 ). Additional research is needed to investigate these and other biological pathways that may link greater lifetime stressor exposure with reward-related neural function and depression.…”

Reintroducing the Effortless Assessment Research System (EARS)

“…In addition, our data scientists derive location and travel features from raw GPS data, including but not limited to time spent at home, time spent traveling, and the number of travel events per day. Following the sentiment analysis approach of Byrne and colleagues [ 3 ], which showed a relationship between EARS-typed text and stress, our data scientists offer typed text features that measure positive sentiment and negative sentiment words. We also offer features that measure first-person pronoun use and absolutist language, which are related to mental health, especially indices of depression and suicide risk [ 4 , 5 ].…”