Emily A. Scherer

CrossCheck: toward passive sensing and detection of mental health changes in people with schizophrenia

Early detection of mental health changes in individuals with serious mental illness is critical for effective intervention. CrossCheck is the first step towards the passive monitoring of mental health indicators in patients with schizophrenia and paves the way towards relapse prediction and early intervention. In this paper, we present initial results from an ongoing randomized control trial, where passive smartphone sensor data is collected from 21 outpatients with schizophrenia recently discharged from hospital over a period ranging from 2-8.5 months. Our results indicate that there are statistically significant associations between automatically tracked behavioral features related to sleep, mobility, conversations, smart-phone usage and self-reported indicators of mental health in schizophrenia. Using these features we build inference models capable of accurately predicting aggregated scores of mental health indicators in schizophrenia with a mean error of 7.6% of the score range. Finally, we discuss results on the level of personalization that is needed to account for the known variations within people. We show that by leveraging knowledge from a population with schizophrenia, it is possible to train accurate personalized models that require fewer individual-specific data to quickly adapt to new users.

mHealth for Schizophrenia: Patient Engagement With a Mobile Phone Intervention Following Hospital Discharge

Background mHealth interventions that use mobile phones as instruments for illness management are gaining popularity. Research examining mobile phone‒based mHealth programs for people with psychosis has shown that these approaches are feasible, acceptable, and clinically promising. However, most mHealth initiatives involving people with schizophrenia have spanned periods ranging from a few days to several weeks and have typically involved participants who were clinically stable. Objective Our aim was to evaluate the viability of extended mHealth interventions for people with schizophrenia-spectrum disorders following hospital discharge. Specifically, we set out to examine the following: (1) Can individuals be engaged with a mobile phone intervention program during this high-risk period?, (2) Are age, gender, racial background, or hospitalization history associated with their engagement or persistence in using a mobile phone intervention over time?, and (3) Does engagement differ by characteristics of the mHealth intervention itself (ie, pre-programmed vs on-demand functions)? Methods We examined mHealth intervention use and demographic and clinical predictors of engagement in 342 individuals with schizophrenia-spectrum disorders who were given the FOCUS mobile phone intervention as part of a technology-assisted relapse prevention program during the 6-month high-risk period following hospitalization. Results On average, participants engaged with FOCUS for 82% of the weeks they had the mobile phone. People who used FOCUS more often continued using it over longer periods: 44% used the intervention over 5-6 months, on average 4.3 days a week. Gender, race, age, and number of past psychiatric hospitalizations were associated with engagement. Females used FOCUS on average 0.4 more days a week than males. White participants engaged on average 0.7 days more a week than African-Americans and responded to prompts on 0.7 days more a week than Hispanic participants. Younger participants (age 18-29) had 0.4 fewer days of on-demand use a week than individuals who were 30-45 years old and 0.5 fewer days a week than older participants (age 46-60). Participants with fewer past hospitalizations (1-6) engaged on average 0.2 more days a week than those with seven or more. mHealth program functions were associated with engagement. Participants responded to prompts more often than they self-initiated on-demand tools, but both FOCUS functions were used regularly. Both types of intervention use declined over time (on-demand use had a steeper decline). Although mHealth use declined, the majority of individuals used both on-demand and system-prompted functions regularly throughout their participation. Therefore, neither function is extraneous. Conclusions The findings demonstrated that individuals with schizophrenia-spectrum disorders can actively engage with a clinically supported mobile phone intervention for up to 6 months following hospital discharge. mHealth may be useful in reaching a clinical population that is typically difficult to engage during high-risk periods.

Clinic- and home-based contingency management plus parent training for adolescent cannabis use disorders.

OBJECTIVE The aim of this study was to conduct a randomized test comparing 2 multicomponent, contingency management interventions, 1 with and 1 without a full parent training curriculum, and an individual treatment for adolescent cannabis use disorders. METHOD A total of 153 adolescents who met DSM-IV criteria for cannabis abuse or dependence were randomized to motivational enhancement therapy/cognitive-behavioral therapy (MET/CBT), MET/CBT+abstinence-based contingency management (CM), or MET/CBT+CM+Parent Training (PT). RESULTS Overall, during treatment, abstinence was greater for youth receiving clinic- and home-based CM without PT compared to those who received individual MET/CBT. There was no additional benefit of the full PT curriculum on marijuana use, youth externalizing problems, or parenting. CONCLUSION These results suggest that clinic- plus home-based CM for cannabis use disorders can increase rates of abstinence during treatment over and above an evidence-based treatment (individual MET/CBT), but in this study the addition of a comprehensive parenting training curriculum did not further enhance efficacy. CLINICAL TRIAL REGISTRATION INFORMATION Treatment for Adolescent Marijuana Abuse; http://clinicaltrials.gov; NCT00580671.

Food Insecurity and the Burden of Health-Related Social Problems in an Urban Youth Population

PURPOSE Our study objectives were to (1) determine the prevalence of food insecurity; (2) examine the association between presence and level of food insecurity with other health-related social problems; and (3) assess the predictive values of a two-item food insecurity screen in an urban youth population. METHODS Patients aged 15-25 years completed a Web-based screening tool. Validated questions were used to identify problems in seven health-related social domains (food insecurity, health care access, education, housing, income insecurity, substance use, and intimate partner violence). Chi-square and Kruskal-Wallis tests and logistic regression models controlled for age, sex, and race/ethnicity, assessed the association between food insecurity and health-related social problems. Predictive values of a two-item food insecurity screen compared with the United States Department of Agriculture Food Security Survey were calculated. RESULTS Among 400 patients (mean age 18 years; 69.2% female; 54.6% black; 58.9% public insurance), 32.5% screened positive for food insecurity. Increasing food insecurity level was significantly associated with cumulative burden of social problems (p < .001). In adjusted analyses, food insecurity was associated with problems with health care access (aOR = 2.6, 95% confidence interval [CI] 1.7-4.1), education (aOR = 2.8, 95% CI 1.6-5.1), housing (aOR = 2.8, 95% CI 1.8-4.4), income insecurity (aOR = 2.3, 95% CI 1.2-4.5), and substance use (aOR = 2.5, 95% CI 1.5-4.3). The two-item screen demonstrated sensitivity of 88.5% and specificity of 84.1%. CONCLUSIONS One-third of youth in sample experienced food insecurity, which was strongly associated with presence of other health-related social problems. The two-item screen effectively detected food insecurity. Food insecurity screening may lead to identification of other health-related social problems that when addressed early may improve adolescent health.

Patterns of Body Image Concerns and Disordered Weight- and Shape-Related Behaviors in Heterosexual and Sexual Minority Adolescent Males

This study investigates body image concerns and disordered weight- and shape-related behaviors across adolescence and young adulthood in males and how patterns vary by sexual orientation. Participants were 5,388 males from the U.S. national Growing Up Today Study. In 2001, 2003, and 2005 (spanning ages 15-20 years), participants reported sexual orientation, past-year desire for toned/defined muscles and concerns with weight and shape, and past-year binge eating, restrictive dieting, purging (vomiting or laxative use), and use of products to increase muscularity (e.g., creatine, steroids). Latent class analyses identified 2 patterns at ages 15-16 years and 3 patterns at 17-18 and 19-20 years: healthy (all ages; low body image concerns and weight- and shape-related behaviors; 54-74% of observations), muscle-concerned (ages 17-18 and 19-20; relatively high muscularity concern and product use; 18-21% of observations), and lean-concerned (all ages; relatively high weight and shape concern, dieting, and binge eating; 19-28% of observations). Latent transition analyses revealed that sexual minority males (i.e., mostly heterosexual, gay, and bisexual) were more likely than completely heterosexual males to be lean-concerned at ages 17-18 and 19-20 years and to transition to the lean-concerned class from the healthy class. There were no sexual orientation differences in odds of being muscle-concerned. Both heterosexual and sexual minority males are at risk for presenting body image concerns and weight- and shape-related behaviors that may have deleterious health consequences. Results suggest the need for screening for concerns and behaviors related to leanness and muscularity in early adolescence among all males, regardless of sexual orientation.

College Health Service Capacity to Support Youth With Chronic Medical Conditions

BACKGROUND AND OBJECTIVE: Twenty percent of US youth have a chronic medical condition and many attend college. Guidelines for transition from pediatric to adult care do not address college health services, and little is known about their capacity to identify, support, and provide care for these youth. The objective of this study was to describe college health center policies, practices, and resources for youth with chronic medical conditions (YCMC). METHODS: Survey of medical directors from health centers of a representative sample of 200 4-year US colleges with ≥400 enrolled undergraduate students. Patterns of identification, management, and support for youth with a general chronic medical condition and with asthma, diabetes, and depression, were investigated; χ2 and Fisher exact tests were used to ascertain differences by institutional demographics. RESULTS: Directors at 153 institutions completed the survey (76.5% response rate). Overall, 42% of schools had no system to identify YCMC. However, almost a third (31%) did identify and add to a registry of incoming YCMC on review of medical history, more likely in private (P < .001) and small (<5000 students, P = .002) colleges; 24% of health centers contacted YCMC to check-in/make initial appointments. Most institutions could manage asthma and depression (83% and 69%, respectively); 51% could manage diabetes on campus. CONCLUSIONS: Relatively few US colleges have health systems to identify and contact YCMC, although many centers have capacity to provide primary care and management of some conditions. Guidelines for transition should address policy and practices for pediatricians and colleges to enhance comanagement of affected youth.

CrossCheck: Integrating self-report, behavioral sensing, and smartphone use to identify digital indicators of psychotic relapse.

Objective: This purpose of this study was to describe and demonstrate CrossCheck, a multimodal data collection system designed to aid in continuous remote monitoring and identification of subjective and objective indicators of psychotic relapse. Method: Individuals with schizophrenia-spectrum disorders received a smartphone with the monitoring system installed along with unlimited data plan for 12 months. Participants were instructed to carry the device with them and to complete brief self-reports multiple times a week. Multimodal behavioral sensing (i.e., physical activity, geospatials activity, speech frequency, and duration) and device use data (i.e., call and text activity, app use) were captured automatically. Five individuals who experienced psychiatric hospitalization were selected and described for instructive purposes. Results: Participants had unique digital indicators of their psychotic relapse. For some, self-reports provided clear and potentially actionable description of symptom exacerbation prior to hospitalization. Others had behavioral sensing data trends (e.g., shifts in geolocation patterns, declines in physical activity) or device use patterns (e.g., increased nighttime app use, discontinuation of all smartphone use) that reflected the changes they experienced more effectively. Conclusion: Advancements in mobile technology are enabling collection of an abundance of information that until recently was largely inaccessible to clinical research and practice. However, remote monitoring and relapse detection is in its nascence. Development and evaluation of innovative data management, modeling, and signal-detection techniques that can identify changes within an individual over time (i.e., unique relapse signatures) will be essential if we are to capitalize on these data to improve treatment and prevention.

Wearable devices and mobile technologies for supporting behavioral weight loss among people with serious mental illness

Promoting physical activity is essential for addressing elevated cardiovascular risk and high obesity rates affecting people with serious mental illness. Numerous challenges interfere with exercise participation in this high-risk group including mental health symptoms, low motivation, and limited access to safe and affordable options for physical activity. Wearable devices and mobile health technologies may afford new opportunities for promoting physical activity and supporting behavioral weight loss efforts. This exploratory study examined whether daily step count measured using Fitbit wearable devices was associated with weight loss and improved fitness among individuals with serious mental illness enrolled in a 6-month lifestyle program. Participants (n=34) had a schizophrenia spectrum disorder (23.5%), major depression (50.0%), or bipolar disorder (26.5%), and wore Fitbits most of the days (M=86.2%; SD=18.4%) they were enrolled in the study. At 6-months, higher average daily step count was associated with greater weight loss (F=5.07; df=1,32; p=0.0314), but not improved fitness (F=1.92; df=1,31; p=0.176). These findings demonstrate that encouraging participants with serious mental illness enrolled in lifestyle interventions to collect more steps may contribute to greater weight loss. This suggests that wearable devices may offer a feasible and potentially effective strategy for supporting behavioral weight loss in community mental health settings.

Cigarette smoking and interest in quitting among overweight and obese adults with serious mental illness enrolled in a fitness intervention.

Abstract This study explored cigarette smoking, health status, and interest in quitting among overweight and obese adults with serious mental illness enrolled in a fitness intervention. Baseline data from two studies of the In SHAPE fitness intervention were combined. A total of 341 overweight or obese adults with serious mental illness were assessed on smoking, interest in quitting, cardiovascular fitness, lipids, body mass index, readiness to change diet, and psychiatric symptoms. Thirty-six percent (n = 122) of participants were categorized as current smokers. The majority of smokers (84%) were interested in quitting. Smokers were more likely to be younger, male, and less educated than non-smokers. Smokers had lower high-density lipoprotein cholesterol and were less ready to reduce dietary fat, after adjusting for age, gender, and education. Findings highlight the potential to address both fitness and smoking to reduce cardiovascular risk in individuals with serious mental illness.

Normal Weight with Central Obesity, Physical Activity, and Functional Decline: Data from the Osteoarthritis Initiative

To identify the risks of the combination of normal body mass index (BMI) and central obesity (normal weight and central obesity (NWCO)) on physical activity and function.