Background
Use of intensive longitudinal methods (e.g. ecological momentary assessment, passive sensing) and machine learning (ML) models to predict risk for depression and suicide has increased in recent years. However, these studies often vary considerably in length, ML methods used, and sources of data. The present study examined predictive accuracy for depression and suicidal ideation (SI) as a function of time, comparing different combinations of ML methods and data sources.
Methods
Participants were 2459 first-year training physicians (55.1% female; 52.5% White) who were provided with Fitbit wearable devices and assessed daily for mood. Linear [elastic net regression (ENR)] and non-linear (random forest) ML algorithms were used to predict depression and SI at the first-quarter follow-up assessment, using two sets of variables (daily mood features only, daily mood features + passive-sensing features). To assess accuracy over time, models were estimated iteratively for each of the first 92 days of internship, using data available up to that point in time.
Results
ENRs using only the daily mood features generally had the best accuracy for predicting mental health outcomes, and predictive accuracy within 1 standard error of the full 92 day models was attained by weeks 7–8. Depression at 92 days could be predicted accurately (area under the curve >0.70) after only 14 days of data collection.
Conclusions
Simpler ML methods may outperform more complex methods until passive-sensing features become better specified. For intensive longitudinal studies, there may be limited predictive value in collecting data for more than 2 months.
The Benefits and Barriers model of NSSI postulates that engagement in NSSI is positively reinforced by potent benefits, however there are a host of barriers to engagement, any one of which is salient enough to prevent engagement. It is possible that individual differences in sensation seeking, a trait that describes optimal level of positive reinforcement, may alter the balance between the benefits and barriers of engagement in NSSI. There are significant associations between engagement in NSSI and sensation seeking in college undergraduates, a population with disproportionately high rates of NSSI. However, it is unclear whether these traits play a similar role in adolescents. We expected that higher levels of sensation seeking would positively relate to any NSSI history, lifetime frequency of NSSI, and earlier age at onset of NSSI among a sample of 200 adolescents in a psychiatric hospital. Consistent with previous research, results indicated that females were more likely to engage in NSSI than males. Additionally, increased sensation seeking was associated with greater likelihood of ever engaging in NSSI and a greater number of different NSSI methods tried. Though we expected sensation seeking would be significantly related to lifetime NSSI frequency and earlier onset of NSSI, it was not. Findings suggest that individual differences may alter relations between the benefits and barriers of NSSI and that measuring sensation seeking in adolescents, especially females, and especially those experiencing psychological distress, may identify those at highest risk for engaging in NSSI and may allow for targeted intervention with these individuals.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.