Natural language processing for mental health interventions: a systematic review and research framework.

Neuropsychiatric disorders pose a high societal cost, but their treatment is hindered by lack of objective outcomes and fidelity metrics. AI technologies and specifically Natural Language Processing (NLP) have emerged as tools to study mental health interventions (MHI) at the level of their constituent conversations. However, NLP's potential to address clinical and research challenges remains unclear. We therefore conducted a pre-registered systematic review of NLP-MHI studies using PRISMA guidelines (osf.io/s52jh) to evaluate their models, clinical applications, and to identify biases and gaps. Candidate studies (n = 19,756), including peer-reviewed AI conference manuscripts, were collected up to January 2023 through PubMed, PsycINFO, Scopus, Google Scholar, and ArXiv. A total of 102 articles were included to investigate their computational characteristics (NLP algorithms, audio features, machine learning pipelines, outcome metrics), clinical characteristics (clinical ground truths, study samples, clinical focus), and limitations. Results indicate a rapid growth of NLP MHI studies since 2019, characterized by increased sample sizes and use of large language models. Digital health platforms were the largest providers of MHI data. Ground truth for supervised learning models was based on clinician ratings (n = 31), patient self-report (n = 29) and annotations by raters (n = 26). Text-based features contributed more to model accuracy than audio markers. Patients' clinical presentation (n = 34), response to intervention (n = 11), intervention monitoring (n = 20), providers' characteristics (n = 12), relational dynamics (n = 14), and data preparation (n = 4) were commonly investigated clinical categories. Limitations of reviewed studies included lack of linguistic diversity, limited reproducibility, and population bias. A research framework is developed and validated (NLPxMHI) to assist computational and clinical researchers in addressing the remaining gaps in applying NLP to MHI, with the goal of improving clinical utility, data access, and fairness.

An Integrative Engagement Model of Digital Psychotherapy: Exploratory Focus Group Findings.

BACKGROUND: Digital mental health interventions, such as 2-way and asynchronous messaging therapy, are a growing part of the mental health care treatment ecosystem, yet little is known about how users engage with these interventions over the course of their treatment journeys. User engagement, or client behaviors and therapeutic relationships that facilitate positive treatment outcomes, is a necessary condition for the effectiveness of any digital treatment. Developing a better understanding of the factors that impact user engagement can impact the overall effectiveness of digital psychotherapy. Mapping the user experience in digital therapy may be facilitated by integrating theories from several fields. Specifically, health science's Health Action Process Approach and human-computer interaction's Lived Informatics Model may be usefully synthesized with relational constructs from psychotherapy process-outcome research to identify the determinants of engagement in digital messaging therapy. OBJECTIVE: This study aims to capture insights into digital therapy users' engagement patterns through a qualitative analysis of focus group sessions. We aimed to synthesize emergent intrapersonal and relational determinants of engagement into an integrative framework of engagement in digital therapy. METHODS: A total of 24 focus group participants were recruited to participate in 1 of 5 synchronous focus group sessions held between October and November 2021. Participant responses were coded by 2 researchers using thematic analysis. RESULTS: Coders identified 10 relevant constructs and 24 subconstructs that can collectively account for users' engagement and experience trajectories in the context of digital therapy. Although users' engagement trajectories in digital therapy varied widely, they were principally informed by intrapsychic factors (eg, self-efficacy and outcome expectancy), interpersonal factors (eg, the therapeutic alliance and its rupture), and external factors (eg, treatment costs and social support). These constructs were organized into a proposed Integrative Engagement Model of Digital Psychotherapy. Notably, every participant in the focus groups indicated that their ability to connect with their therapist was among the most important factors that were considered in continuing or terminating treatment. CONCLUSIONS: Engagement in messaging therapy may be usefully approached through an interdisciplinary lens, linking constructs from health science, human-computer interaction studies, and clinical science in an integrative engagement framework. Taken together, our results suggest that users may not view the digital psychotherapy platform itself as a treatment so much as a means of gaining access to a helping provider, that is, users did not see themselves as engaging with a platform but instead viewed their experience as a healing relationship. The findings of this study suggest that a better understanding of user engagement is crucial for enhancing the effectiveness of digital mental health interventions, and future research should continue to explore the underlying factors that contribute to engagement in digital mental health interventions. TRIAL REGISTRATION: ClinicalTrials.gov NCT04507360; https://clinicaltrials.gov/ct2/show/NCT04507360.

Leveraging Mobile Technology for Public Health Promotion: A Multidisciplinary Perspective.

Health behaviors are inextricably linked to health and well-being, yet issues such as physical inactivity and insufficient sleep remain significant global public health problems. Mobile technology-and the unprecedented scope and quantity of data it generates-has a promising but largely untapped potential to promote health behaviors at the individual and population levels. This perspective article provides multidisciplinary recommendations on the design and use of mobile technology, and the concomitant wealth of data, to promote behaviors that support overall health. Using physical activity as anexemplar health behavior, we review emerging strategies for health behavior change interventions. We describe progress on personalizing interventions to an individual and their social, cultural, and built environments, as well as on evaluating relationships between mobile technology data and health to establish evidence-based guidelines. In reviewing these strategies and highlighting directions for future research, we advance the use of theory-based, personalized, and human-centered approaches in promoting health behaviors.

Disparate impacts on online information access during the Covid-19 pandemic.

The COVID-19 pandemic has stimulated important changes in online information access as digital engagement became necessary to meet the demand for health, economic, and educational resources. Our analysis of 55 billion everyday web search interactions during the pandemic across 25,150 US ZIP codes reveals that the extent to which different communities of internet users enlist digital resources varies based on socioeconomic and environmental factors. For example, we find that ZIP codes with lower income intensified their access to health information to a smaller extent than ZIP codes with higher income. We show that ZIP codes with higher proportions of Black or Hispanic residents intensified their access to unemployment resources to a greater extent, while revealing patterns of unemployment site visits unseen by the claims data. Such differences frame important questions on the relationship between differential information search behaviors and the downstream real-world implications on more and less advantaged populations.

Estimating the Burden of Influenza-like Illness on Daily Activity at the Population Scale Using Commercial Wearable Sensors.

Importance: The severity of viral infections can vary widely, from asymptomatic cases to complications leading to hospitalizations and death. Milder cases, despite being more prevalent, often go undocumented, and their public health burden is not accurately estimated. Objective: To estimate the true burden of influenza-like illness (ILI) in the US population using a surrogate measure of daily steps lost as measured by commercial wearable sensors. Design, Setting, and Participants: This cohort study modeled data from 15 122 US adults who reported ILI symptoms during the 2018-2019 influenza season (before the COVID-19 pandemic) and who had a sufficient density of wearable sensor data at symptom onset. Participants' minute-level step data as measured by commercial wearable sensors were collected from October 1, 2018, through June 30, 2019. Minute-level activity time series were transformed into day-level time series per user, indicating the total number of steps daily. Main Outcomes and Measures: The primary end point was the number of steps lost during the period of 4 days before symptom onset (the latent phase) through 11 days after symptom onset (the symptomatic phase). The association between covariates and steps lost during this interval was also examined. Results: Of the 15 122 participants in this study, 13 108 (86.7%) were women, and the median age was 32 years (IQR, 27-38 years). For their ILI event, 2836 of 15 080 participants (18.8%) sought medical attention, and only 61 (0.4%) were hospitalized. Over the course of an ILI lasting 10 days, the mean cumulative loss was 4437 steps (95% CI, 4143-4731 steps). After weighting, there was an estimated overall nationwide reduction in mobility equivalent to 255.2 billion steps (95% CI, 232.9-277.6 billion steps) lost because of ILI symptoms during the study period. This finding reflects significant changes in routines, mobility, and employment and is equivalent to 15% of the active US population becoming completely immobilized for 1 day. Moreover, 60.6% of this reduction in steps (154.6 billion steps [95% CI, 138.1-171.2 billion steps]) occurred among persons who sought no medical care. Age and educational level were positively associated with steps lost. Conclusions and Relevance: These findings suggest that most of the burden of ILI in this study would have been invisible to health care and public health reporting systems. This approach has applications for public health, health care, and clinical research, from estimating costs of lost productivity at population scale, to measuring effectiveness of anti-ILI treatments, to monitoring recovery after acute viral syndromes such as during long COVID-19.

Large-scale diet tracking data reveal disparate associations between food environment and diet.

An unhealthy diet is a major risk factor for chronic diseases including cardiovascular disease, type 2 diabetes, and cancer1-4. Limited access to healthy food options may contribute to unhealthy diets5,6. Studying diets is challenging, typically restricted to small sample sizes, single locations, and non-uniform design across studies, and has led to mixed results on the impact of the food environment7-23. Here we leverage smartphones to track diet health, operationalized through the self-reported consumption of fresh fruits and vegetables, fast food and soda, as well as body-mass index status in a country-wide observational study of 1,164,926 U.S. participants (MyFitnessPal app users) and 2.3 billion food entries to study the independent contributions of fast food and grocery store access, income and education to diet health outcomes. This study constitutes the largest nationwide study examining the relationship between the food environment and diet to date. We find that higher access to grocery stores, lower access to fast food, higher income and college education are independently associated with higher consumption of fresh fruits and vegetables, lower consumption of fast food and soda, and lower likelihood of being affected by overweight and obesity. However, these associations vary significantly across zip codes with predominantly Black, Hispanic or white populations. For instance, high grocery store access has a significantly larger association with higher fruit and vegetable consumption in zip codes with predominantly Hispanic populations (7.4% difference) and Black populations (10.2% difference) in contrast to zip codes with predominantly white populations (1.7% difference). Policy targeted at improving food access, income and education may increase healthy eating, but intervention allocation may need to be optimized for specific subpopulations and locations.

A computational approach to measure the linguistic characteristics of psychotherapy timing, responsiveness, and consistency.

Although individual psychotherapy is generally effective for a range of mental health conditions, little is known about the moment-to-moment language use of effective therapists. Increased access to computational power, coupled with a rise in computer-mediated communication (telehealth), makes feasible the large-scale analyses of language use during psychotherapy. Transparent methodological approaches are lacking, however. Here we present novel methods to increase the efficiency of efforts to examine language use in psychotherapy. We evaluate three important aspects of therapist language use - timing, responsiveness, and consistency - across five clinically relevant language domains: pronouns, time orientation, emotional polarity, therapist tactics, and paralinguistic style. We find therapist language is dynamic within sessions, responds to patient language, and relates to patient symptom diagnosis but not symptom severity. Our results demonstrate that analyzing therapist language at scale is feasible and may help answer longstanding questions about specific behaviors of effective therapists.

Efficient and Explainable Risk Assessments for Imminent Dementia in an Aging Cohort Study.

As the aging US population grows, scalable approaches are needed to identify individuals at risk for dementia. Common prediction tools have limited predictive value, involve expensive neuroimaging, or require extensive and repeated cognitive testing. None of these approaches scale to the sizable aging population who do not receive routine clinical assessments. Our study seeks a tractable and widely administrable set of metrics that can accurately predict imminent (i.e., within three years) dementia onset. To this end, we develop and apply a machine learning (ML) model to an aging cohort study with an extensive set of longitudinal clinical variables to highlight at-risk individuals with better accuracy than standard rudimentary approaches. Next, we reduce the burden needed to achieve accurate risk assessments for those deemed at risk by (1) predicting when consecutive clinical visits may be unnecessary, and (2) selecting a subset of highly predictive cognitive tests. Finally, we demonstrate that our method successfully provides individualized prediction explanations that retain non-linear feature effects present in the data. Our final model, which uses only four cognitive tests (less than 20 minutes to administer) collected in a single visit, affords predictive performance comparable to a standard 100-minute neuropsychological battery and personalized risk explanations. Our approach shows the potential for an efficient tool for screening and explaining dementia risk in the general aging population.

Daily, weekly, seasonal and menstrual cycles in women's mood, behaviour and vital signs.

Dimensions of human mood, behaviour and vital signs cycle over multiple timescales. However, it remains unclear which dimensions are most cyclical, and how daily, weekly, seasonal and menstrual cycles compare in magnitude. The menstrual cycle remains particularly understudied because, not being synchronized across the population, it will be averaged out unless menstrual cycles can be aligned before analysis. Here, we analyse 241 million observations from 3.3 million women across 109 countries, tracking 15 dimensions of mood, behaviour and vital signs using a women's health mobile app. Out of the daily, weekly, seasonal and menstrual cycles, the menstrual cycle had the greatest magnitude for most of the measured dimensions of mood, behaviour and vital signs. Mood, vital signs and sexual behaviour vary most substantially over the course of the menstrual cycle, while sleep and exercise behaviour remain more constant. Menstrual cycle effects are directionally consistent across countries.

Characterizing COVID-19 and Influenza Illnesses in the Real World via Person-Generated Health Data.

The fight against COVID-19 is hindered by similarly presenting viral infections that may confound detection and monitoring. We examined person-generated health data (PGHD), consisting of survey and commercial wearable data from individuals' everyday lives, for 230 people who reported a COVID-19 diagnosis between March 30, 2020, and April 27, 2020 (n = 41 with wearable data). Compared with self-reported diagnosed flu cases from the same time frame (n = 426, 85 with wearable data) or pre-pandemic (n = 6,270, 1,265 with wearable data), COVID-19 patients reported a distinct symptom constellation that lasted longer (median of 12 versus 9 and 7 days, respectively) and peaked later after illness onset. Wearable data showed significant changes in daily steps and prevalence of anomalous resting heart rate measurements, of similar magnitudes for both the flu and COVID-19 cohorts. Our findings highlight the need to include flu comparator arms when evaluating PGHD applications aimed to be highly specific for COVID-19.

Boba: Authoring and Visualizing Multiverse Analyses.

Multiverse analysis is an approach to data analysis in which all "reasonable" analytic decisions are evaluated in parallel and interpreted collectively, in order to foster robustness and transparency. However, specifying a multiverse is demanding because analysts must manage myriad variants from a cross-product of analytic decisions, and the results require nuanced interpretation. We contribute Baba: an integrated domain-specific language (DSL) and visual analysis system for authoring and reviewing multiverse analyses. With the Boba DSL, analysts write the shared portion of analysis code only once, alongside local variations defining alternative decisions, from which the compiler generates a multiplex of scripts representing all possible analysis paths. The Boba Visualizer provides linked views of model results and the multiverse decision space to enable rapid, systematic assessment of consequential decisions and robustness, including sampling uncertainty and model fit. We demonstrate Boba's utility through two data analysis case studies, and reflect on challenges and design opportunities for multiverse analysis software.

Assessing the relationship between routine and schizophrenia symptoms with passively sensed measures of behavioral stability.

Increased stability in one's daily routine is associated with well-being in the general population and often a goal of behavioral interventions for people with serious mental illnesses like schizophrenia. Assessing behavioral stability has been limited in clinical research by the use of retrospective scales, which are susceptible to reporting biases and memory inaccuracies. Mobile passive sensors, which are less susceptible to these sources of error, have emerged as tools to assess behavioral patterns in a range of populations. The present study developed and examined a metric of behavioral stability from data generated by a passive sensing system carried by 61 individuals with schizophrenia for one year. This metric-the Stability Index-appeared orthogonal from existing measures drawn from passive sensors and matched the predictive performance of state-of-the-art features. Specifically, greater stability in social activity (e.g., calls and messages) were associated with lower symptoms, and greater stability in physical activity (e.g., being still) appeared associated with elevated symptoms. This study provides additional support for the predictive value of individualized over population-level data in psychiatric populations. The Stability Index offers also a promising tool for generating insights about the impact of behavioral stability in schizophrenia-spectrum disorders.

Data-Driven Implications for Translating Evidence-Based Psychotherapies into Technology-Delivered Interventions.

Mobile mental health interventions have the potential to reduce barriers and increase engagement in psychotherapy. However, most current tools fail to meet evidence-based principles. In this paper, we describe data-driven design implications for translating evidence-based interventions into mobile apps. To develop these design implications, we analyzed data from a month-long field study of an app designed to support dialectical behavioral therapy, a psychotherapy that aims to teach concrete coping skills to help people better manage their mental health. We investigated whether particular skills are more or less effective in reducing distress or emotional intensity. We also characterized how an individual's disorders, characteristics, and preferences may correlate with skill effectiveness, as well as how skill-level improvements correlate with study-wide changes in depressive symptoms. We then developed a model to predict skill effectiveness. Based on our findings, we present design implications that emphasize the importance of considering different environmental, emotional, and personal contexts. Finally, we discuss promising future opportunities for mobile apps to better support evidence-based psychotherapies, including using machine learning algorithms to develop personalized and context-aware skill recommendations.

Best practices for analyzing large-scale health data from wearables and smartphone apps.

Smartphone apps and wearable devices for tracking physical activity and other health behaviors have become popular in recent years and provide a largely untapped source of data about health behaviors in the free-living environment. The data are large in scale, collected at low cost in the "wild", and often recorded in an automatic fashion, providing a powerful complement to traditional surveillance studies and controlled trials. These data are helping to reveal, for example, new insights about environmental and social influences on physical activity. The observational nature of the datasets and collection via commercial devices and apps pose challenges, however, including the potential for measurement, population, and/or selection bias, as well as missing data. In this article, we review insights gleaned from these datasets and propose best practices for addressing the limitations of large-scale data from apps and wearables. Our goal is to enable researchers to effectively harness the data from smartphone apps and wearable devices to better understand what drives physical activity and other health behaviors.

Passively-sensed Behavioral Correlates of Discrimination Events in College Students.

A deep understanding of how discrimination impacts psychological health and well-being of students could allow us to better protect individuals at risk and support those who encounter discrimination. While the link between discrimination and diminished psychological and physical well-being is well established, existing research largely focuses on chronic discrimination and long-term outcomes. A better understanding of the short-term behavioral correlates of discrimination events could help us to concretely quantify such experiences, which in turn could support policy and intervention design. In this paper we specifically examine, for the first time, what behaviors change and in what ways in relation to discrimination. We use actively-reported and passively-measured markers of health and well-being in a sample of 209 first-year college students over the course of two academic quarters. We examine changes in indicators of psychological state in relation to reports of unfair treatment in terms of five categories of behaviors: physical activity, phone usage, social interaction, mobility, and sleep. We find that students who encounter unfair treatment become more physically active, interact more with their phone in the morning, make more calls in the evening, and spend more time in bed on the day of the event. Some of these patterns continue the next day. Our results further our understanding of the impact of discrimination and can inform intervention work.

Goal-setting And Achievement In Activity Tracking Apps: A Case Study Of MyFitnessPal.

Activity tracking apps often make use of goals as one of their core motivational tools. There are two critical components to this tool: setting a goal, and subsequently achieving that goal. Despite its crucial role in how a number of prominent self-tracking apps function, there has been relatively little investigation of the goal-setting and achievement aspects of self-tracking apps. Here we explore this issue, investigating a particular goal setting and achievement process that is extensive, recorded, and crucial for both the app and its users' success: weight loss goals in MyFitnessPal. We present a large-scale study of 1.4 million users and weight loss goals, allowing for an unprecedented detailed view of how people set and achieve their goals. We find that, even for difficult long-term goals, behavior within the first 7 days predicts those who ultimately achieve their goals, that is, those who lose at least as much weight as they set out to, and those who do not. For instance, high amounts of early weight loss, which some researchers have classified as unsustainable, leads to higher goal achievement rates. We also show that early food intake, self-monitoring motivation, and attitude towards the goal are important factors. We then show that we can use our findings to predict goal achievement with an accuracy of 79% ROC AUC just 7 days after a goal is set. Finally, we discuss how our findings could inform steps to improve goal achievement in self-tracking apps.

Modeling Individual Cyclic Variation in Human Behavior.

Cycles are fundamental to human health and behavior. Examples include mood cycles, circadian rhythms, and the menstrual cycle. However, modeling cycles in time series data is challenging because in most cases the cycles are not labeled or directly observed and need to be inferred from multidimensional measurements taken over time. Here, we present Cyclic Hidden Markov Models (CyH-MMs) for detecting and modeling cycles in a collection of multidimensional heterogeneous time series data. In contrast to previous cycle modeling methods, CyHMMs deal with a number of challenges encountered in modeling real-world cycles: they can model multivariate data with both discrete and continuous dimensions; they explicitly model and are robust to missing data; and they can share information across individuals to accommodate variation both within and between individual time series. Experiments on synthetic and real-world health-tracking data demonstrate that CyHMMs infer cycle lengths more accurately than existing methods, with 58% lower error on simulated data and 63% lower error on real-world data compared to the best-performing baseline. CyHMMs can also perform functions which baselines cannot: they can model the progression of individual features/symptoms over the course of the cycle, identify the most variable features, and cluster individual time series into groups with distinct characteristics. Applying CyHMMs to two real-world health-tracking datasets-of human menstrual cycle symptoms and physical activity tracking data-yields important insights including which symptoms to expect at each point during the cycle. We also find that people fall into several groups with distinct cycle patterns, and that these groups differ along dimensions not provided to the model. For example, by modeling missing data in the menstrual cycles dataset, we are able to discover a medically relevant group of birth control users even though information on birth control is not given to the model.

Modeling Interdependent and Periodic Real-World Action Sequences.

Mobile health applications, including those that track activities such as exercise, sleep, and diet, are becoming widely used. Accurately predicting human actions in the real world is essential for targeted recommendations that could improve our health and for personalization of these applications. However, making such predictions is extremely difficult due to the complexities of human behavior, which consists of a large number of potential actions that vary over time, depend on each other, and are periodic. Previous work has not jointly modeled these dynamics and has largely focused on item consumption patterns instead of broader types of behaviors such as eating, commuting or exercising. In this work, we develop a novel statistical model, called TIPAS, for Time-varying, Interdependent, and Periodic Action Sequences. Our approach is based on personalized, multivariate temporal point processes that model time-varying action propensities through a mixture of Gaussian intensities. Our model captures short-term and long-term periodic interdependencies between actions through Hawkes process-based self-excitations. We evaluate our approach on two activity logging datasets comprising 12 million real-world actions (e.g., eating, sleep, and exercise) taken by 20 thousand users over 17 months. We demonstrate that our approach allows us to make successful predictions of future user actions and their timing. Specifically, TIPAS improves predictions of actions, and their timing, over existing methods across multiple datasets by up to 156%, and up to 37%, respectively. Performance improvements are particularly large for relatively rare and periodic actions such as walking and biking, improving over baselines by up to 256%. This demonstrates that explicit modeling of dependencies and periodicities in real-world behavior enables successful predictions of future actions, with implications for modeling human behavior, app personalization, and targeting of health interventions.

I'll Be Back: On the Multiple Lives of Users of a Mobile Activity Tracking Application.

Mobile health applications that track activities, such as exercise, sleep, and diet, are becoming widely used. While these activity tracking applications have the potential to improve our health, user engagement and retention are critical factors for their success. However, long-term user engagement patterns in real-world activity tracking applications are not yet well understood. Here we study user engagement patterns within a mobile physical activity tracking application consisting of 115 million logged activities taken by over a million users over 31 months. Specifically, we show that over 75% of users return and re-engage with the application after prolonged periods of inactivity, no matter the duration of the inactivity. We find a surprising result that the re-engagement usage patterns resemble those of the start of the initial engagement period, rather than being a simple continuation of the end of the initial engagement period. This evidence points to a conceptual model of multiple lives of user engagement, extending the prevalent single life view of user activity. We demonstrate that these multiple lives occur because the users have a variety of different primary intents or goals for using the app. These primary intents are associated with how long each life lasts and how likely the user is to re-engage for a new life. We find evidence for users being more likely to stop using the app once they achieved their primary intent or goal (e.g., weight loss). However, these users might return once their original intent resurfaces (e.g., wanting to lose newly gained weight). We discuss implications of the multiple life paradigm and propose a novel prediction task of predicting the number of lives of a user. Based on insights developed in this work, including a marker of improved primary intent performance, our prediction models achieve 71% ROC AUC. Overall, our research has implications for modeling user re-engagement in health activity tracking applications and has consequences for how notifications, recommendations as well as gamification can be used to increase engagement.

Psychomotor function measured via online activity predicts motor vehicle fatality risk.

Impaired psychomotor performance severely increases the risk of fatal and non-fatal car accidents. However, we currently lack methods to continuously and non-intrusively monitor psychomotor performance. We show we can estimate psychomotor function at population scale from 16 billion observations of typing speeds during the input of web search queries. We show that these estimates exhibit diurnal variation with a substantial increase during typical sleep times, matching published accident risk rates. Further, we show that psychomotor impairment, as measured by keystroke timing, predicts motor vehicle fatality risk on a population level (Spearman ρ = 0.61; p « 10-10). The methods and results highlight a promising direction of harnessing ambient streams of data, such as patterns of interactions with devices, as large-scale sensors to continuously and non-intrusively monitor human psychomotor performance at population scale.