Digital Phenotyping for Mood Disorders: Methodology-Oriented Pilot Feasibility Study.

BACKGROUND: In the burgeoning area of clinical digital phenotyping research, there is a dearth of literature that details methodology, including the key challenges and dilemmas in developing and implementing a successful architecture for technological infrastructure, patient engagement, longitudinal study participation, and successful reporting and analysis of diverse passive and active digital data streams. OBJECTIVE: This article provides a narrative rationale for our study design in the context of the current evidence base and best practices, with an emphasis on our initial lessons learned from the implementation challenges and successes of this digital phenotyping study. METHODS: We describe the design and implementation approach for a digital phenotyping pilot feasibility study with attention to synthesizing key literature and the reasoning for pragmatic adaptations in implementing a multisite study encompassing distinct geographic and population settings. This methodology was used to recruit patients as study participants with a clinician-validated diagnostic history of unipolar depression, bipolar I disorder, or bipolar II disorder, or healthy controls in 2 geographically distinct health care systems for a longitudinal digital phenotyping study of mood disorders. RESULTS: We describe the feasibility of a multisite digital phenotyping pilot study for patients with mood disorders in terms of passively and actively collected phenotyping data quality and enrollment of patients. Overall data quality (assessed as the amount of sensor data obtained vs expected) was high compared to that in related studies. Results were reported on the relevant demographic features of study participants, revealing recruitment properties of age (mean subgroup age ranged from 31 years in the healthy control subgroup to 38 years in the bipolar I disorder subgroup), sex (predominance of female participants, with 7/11, 64% females in the bipolar II disorder subgroup), and smartphone operating system (iOS vs Android; iOS ranged from 7/11, 64% in the bipolar II disorder subgroup to 29/32, 91% in the healthy control subgroup). We also described implementation considerations around digital phenotyping research for mood disorders and other psychiatric conditions. CONCLUSIONS: Digital phenotyping in affective disorders is feasible on both Android and iOS smartphones, and the resulting data quality using an open-source platform is higher than that in comparable studies. While the digital phenotyping data quality was independent of gender and race, the reported demographic features of study participants revealed important information on possible selection biases that may result from naturalistic research in this domain. We believe that the methodology described will be readily reproducible and generalizable to other study settings and patient populations given our data on deployment at 2 unique sites.

Digital phenotyping of student mental health during COVID-19: an observational study of 100 college students.

Objective: This study assessed the feasibility of capturing smartphone based digital phenotyping data in college students during the COVID-19 pandemic with the goal of understanding how digital biomarkers of behavior correlate with mental health. Participants: Participants were 100 students enrolled in 4-year universities. Methods: Each participant attended a virtual visit to complete a series of gold-standard mental health assessments, and then used a mobile app for 28 days to complete mood assessments and allow for passive collection of GPS, accelerometer, phone call, and screen time data. Students completed another virtual visit at the end of the study to collect a second round of mental health assessments. Results: In-app daily mood assessments were strongly correlated with their corresponding gold standard clinical assessment. Sleep variance among students was correlated to depression scores (ρ = .28) and stress scores (ρ = .27). Conclusions: Digital Phenotyping among college students is feasible on both an individual and a sample level. Studies with larger sample sizes are necessary to understand population trends, but there are practical applications of the data today.

Estimated GFR, Albuminuria, and Physical Function: The Brain in Kidney Disease (BRINK) Cohort Study.

Rationale & Objective: Chronic kidney disease (CKD) is associated with impaired physical performance. However, the association between albuminuria, a marker of vascular endothelial dysfunction, and physical performance has not been fully characterized. We hypothesized that estimated glomerular filtration rate (eGFR) and albuminuria would be independently associated with physical performance. Study Design: Cross-sectional analysis. Setting & Participants: A total of 571 adults with and without CKD. Predictors: Creatinine-based eGFR (eGFRCr) and cystatin C-based eGFR (eGFRCysC) and urine albumin to creatinine ratio (UACR). Outcome: Short Physical Performance Battery (SPPB). Analytical Approach: Univariate and multivariable logistic regression models were used to examine associations of eGFR and UACR with impaired physical performance. Results: Of the 571 participants (mean age, 69.3 years), 157 (27.5%) had eGFRCr (mL/min/1.73m2) <30, 276 (48.3%) had eGFRCr 30-<60, and 138 (24.2%) had eGFRCr ≥60; 303 (55.3%) participants had eGFRcysC <30, 141 (25.7%) had eGFRcysC 30-<60, and 104 (19.0%) had eGFRcysC ≥60. Impaired physical performance was observed in 222 (38.9%) participants. Separate univariate analyses showed that lower eGFRCr, lower eGFRCysC, and higher UACR were associated with higher odds of impaired physical performance. In the adjusted model with eGFRCr or eGFRCysC, UACR, and covariates, UACR retained a statistically significant association with impaired physical performance (adjusted odds ratio [OR], 2.04; 95% confidence interval [CI], 1.21-3.47 for UACR from 30-300 mg/g vs <30 mg/g and adjusted OR, 1.93; 95% CI, 1.01-3.69 for UACR >300 mg/g vs <30 mg/g), but eGFRCr and eGFRCysC did not. Limitations: Cross-sectional analysis, estimated rather than measured GFR. Conclusions: Only UACR was associated with worse physical performance in the fully adjusted model, suggesting that vascular endothelial function and inflammation may be important mechanisms of decreased physical function. Similar results were found using eGFRCr or eGFRCysC, suggesting that confounding based on muscle mass does not explain the lack of an association between eGFRCr and physical performance.

Similarity matrix-based anomaly detection for clinical intervention.

The use of digital phenotyping methods in clinical care has allowed for improved investigation of spatiotemporal behaviors of patients. Moreover, detecting abnormalities in mobile sensor data patterns can be instrumental in identifying potential changes in symptomology. We propose a method that temporally aligns sensor data in order to achieve interpretable measures of similarity between time points. These computed measures can then be used for anomaly detection, baseline routine computation, and trajectory clustering. In addition, we apply this method on a study of 695 college participants, as well as on a patient with worsening anxiety and depression. With varying temporal constraints, we find mild correlations between changes in routine and clinical scores. Furthermore, in our experiment on an individual with elevated depression and anxiety, we are able to cluster GPS trajectories, allowing for improved understanding and visualization of routines with respect to symptomology. In the future, we aim to apply this method on individuals that undergo data collection for longer periods of time, thus allowing for a better understanding of long-term routines and signals for clinical intervention.

Kidney-Metabolic Factors Associated with Cognitive Impairment in Chronic Kidney Disease: A Pilot Study.

INTRODUCTION: The associations of kidney-metabolic biomarkers with cognitive impairment (CI) beyond the estimated glomerular filtration rate (eGFR, in mL/min/1.73 m2) and albuminuria levels are not well understood. In exploratory analysis, our objective was to determine the extent that three kidney-metabolic factors, previously proposed as mechanisms of CI and commonly abnormal in chronic kidney disease (CKD), were associated with prevalent CI in CKD participants, adjusted for kidney function measures. METHODS: The study cohort included community-dwelling individuals aged ≥45 years with CKD (eGFR <60), not requiring dialysis, recruited from four health systems. We examined the serum biomarkers bicarbonate (CO2), TNFαR1, and cholesterol as primary exposures. A structured neuropsychological battery conducted by trained staff measured global and domain-specific cognitive performance. Logistic regression analyses estimated the cross-sectional associations between kidney-metabolic measures and global and cognitive domain-specific moderate/severe (Mod/Sev) CI, adjusted for the eGFR, urinary albumin-creatinine ratio (UACR, mg/g), demographics, comorbid conditions, and other kidney-metabolic biomarkers commonly abnormal in CKD. RESULTS: Among 436 CKD participants with mean age 70 years, 16% were Black, the mean eGFR was 34, and the median [IQR] UACR was 49 [0.0, 378] mg/g. In adjusted models, increased TNFαR1 was associated with global Mod/Sev CI (odds ratio [95% confidence interval] = 1.40 [1.02, 1.93]; p = 0.04); low bicarbonate (CO2 <20 mEq/L) with Mod/Sev memory impairment (3.04 [1.09, 8.47]; p = 0.03), and each 10-mg/dL lower cholesterol was associated with Mod/Sev executive function/processing speed impairment (1.12 [1.02, 1.23]; p = 0.02). However, after adjustment for multiple comparisons, these associations were no longer significant nor were any other kidney-metabolic factors significant for any CI classification. CONCLUSION: In exploratory analyses in a CKD population, three kidney-metabolic factors were associated with CI, but after adjustment for multiple comparisons, were no longer significant. Future studies in larger CKD populations are needed to assess these potential risk factors for CI.

Assessing mental health apps marketplaces with objective metrics from 29,190 data points from 278 apps.

OBJECTIVE: Utilizing a standard framework that may help clinicians and patients to identify relevant mental health apps, we sought to gain a comprehensive picture of the space by searching for, downloading, and reviewing 278 mental health apps from both the iOS and Android stores. METHODS: 278 mental health apps from the Apple iOS store and Google Play store were downloaded and reviewed in a standardized manner by trained app raters using a validated framework. Apps were evaluated with this framework comprising 105 questions and covering app origin and accessibility, privacy and security, inputs and outputs, clinical foundation, features and engagement style, and interoperability. RESULTS: Our results confirm that app stars and downloads-even for the most popular apps by these metrics-did not correlate with more clinically relevant metrics related to privacy/security, effectiveness, and engagement. Most mental health apps offer similar functionality, with 16.5% offering both mood tracking and journaling and 7% offering psychoeducation, deep breathing, mindfulness, journaling, and mood tracking. Only 36.4% of apps were updated with a 100-day window, and 7.5% of apps had not been updated in four years. CONCLUSION: Current app marketplace metrics commonly used to evaluate apps do not offer an accurate representation of individual apps or a comprehensive overview of the entire space. The majority of apps overlap in terms of features offered, with many domains and other features not well represented. Selecting an appropriate app continues to require personal matching given no clear trends or guidance offered by quantitative metrics alone.

Anomaly detection to predict relapse risk in schizophrenia.

The integration of technology in clinical care is growing rapidly and has become especially relevant during the global COVID-19 pandemic. Smartphone-based digital phenotyping, or the use of integrated sensors to identify patterns in behavior and symptomatology, has shown potential in detecting subtle moment-to-moment changes. These changes, often referred to as anomalies, represent significant deviations from an individual's baseline, may be useful in informing the risk of relapse in serious mental illness. Our investigation of smartphone-based anomaly detection resulted in 89% sensitivity and 75% specificity for predicting relapse in schizophrenia. These results demonstrate the potential of longitudinal collection of real-time behavior and symptomatology via smartphones and the clinical utility of individualized analysis. Future studies are necessary to explore how specificity can be improved, just-in-time adaptive interventions utilized, and clinical integration achieved.

To test or not to test: A study examining the return rates of rosacea patients treated with a pulsed dye laser.

Pulsed dye laser (PDL) is an effective treatment option for erythematotelangiectatic rosacea. The use of a test spot allows patients to experience the procedure on a small area prior to further treatment. The purpose of this study was to elucidate whether the use of a no charge test spot influenced return rates for further PDL treatment. Data were obtained retrospectively using International Classification of Diseases (ICD)-10 codes for rosacea. Sixty charts were identified: 26 patients initially received a PDL test area free of charge, whereas 34 patients initially underwent full PDL treatment. Patients who experienced the test spot laser treatment had a lower return rate compared to those that directly underwent full PDL treatment. However, this difference was not statistically significant (Fisher's exact test p = 0.2883). Future studies evaluating and identifying factors that influence PDL return rates are needed. Abbreviations: ETR: Erythematotelangiectatic rosacea; PDL: pulsed dye laser; ICD: International classification of diseases.

A multidisciplinary approach to the successful management of Gorlin syndrome.

Gorlin-Goltz syndrome (GGS) is a rare genetic syndrome with variable expressivity and autosomal dominant inheritance. The major features of GGS include numerous basal cell carcinomas (BCCs), keratocysts of the jaw, palmar/plantar pits and calcification of the falx cerebri. Authors report the case of a 51-year-old male with a 19-year history of GGS and multiple BCCs of the head and neck. He presented with a large ulcerating lesion on the right side of his face involving cutaneous, subcutaneous and muscular tissues of the temporal and orbital region. Additionally, magnetic resonance imaging revealed involvement of the right zygomatic bone, infratemporal fossa and mandible. This case is notable in that BCC invasion of the facial bones is rare. Extensive resection and reconstruction with a latissimus dorsi microvascular free muscle flap was performed. The success of this challenging case exemplifies the need for a multidisciplinary team that included dermatology, plastic surgery, oculoplastics and otolaryngology.

Altering leukocyte recruitment following traumatic brain injury with ghrelin therapy.

BACKGROUND: Traumatic brain injury (TBI)-induced cerebral inflammation involves several mediators including activation of resident microglia, infiltration of leukocytes, and release of proinflammatory cytokines and chemokines at the site of injury. Invading leukocytes, mainly neutrophil and inflammatory monocytes, contribute to ongoing post-TBI cerebral edema and neuronal injury. Based on the beneficial effect of ghrelin hormone treatment following TBI, we hypothesized that ghrelin may alter the infiltrating inflammatory cell profile. METHODS: A weight drop model was used to create severe TBI. C57 mice were divided into three groups: sham, no TBI or ghrelin treatment; TBI, TBI only; TBI/ghrelin, animals were treated with ghrelin 20 µg (intraperitoneally) immediately following TBI and again 1 hour later. Seven days after injury, brain sections were immunostained with Iba-1 and CD11b to assess the recruitment and activation of resident microglia and infiltrated leukocytes. Alternatively, brain dissociates were isolated, and flow cytometry was used to gate for microglia (CD11b, CD45 cells), monocytes (CD11b, CD45, F4/80 cells), and neutrophils (CD11b, CD45, F4/80 cells) to measure their recruitment to injury site. RESULTS: TBI resulted in a rapid invasion (16-fold) of inflammatory leukocytes to the site of injury, which persisted for at least 1 week. Ghrelin treatment significantly reduced infiltration of peripheral leukocytes (2.8-fold). In particular, recruitment of CD11bCD45 inflammatory monocytes (2.4-fold) and CD11bCD45F4/80 neutrophils (1.7-fold) was reduced following ghrelin treatment. There were no observed ghrelin-mediated changes in either the number of CD11bCD45 resident microglia or its activation state. CONCLUSION: Together, our data demonstrate that ghrelin attenuated leukocyte recruitment, which correlates with improved histologic outcome following TBI.

Three distinct actin-attached structural states of myosin in muscle fibers.

We have used thiol cross-linking and electron paramagnetic resonance (EPR) to resolve structural transitions of myosin's light chain domain (LCD) and catalytic domain (CD) that are associated with force generation. Spin labels were incorporated into the LCD of muscle fibers by exchanging spin-labeled regulatory light chain for endogenous regulatory light chain, with full retention of function. To trap myosin in a structural state analogous to the elusive posthydrolysis ternary complex A.M'.D.P, we used pPDM to cross-link SH1 (Cys(707)) to SH2 (Cys(697)) on the CD. LCD orientation and dynamics were measured in three biochemical states: relaxation (A.M.T), SH1-SH2 cross-linked (A.M'.D.P analog), and rigor (A.M.D). EPR showed that the LCD of cross-linked fibers has an orientational distribution intermediate between relaxation and rigor, and saturation transfer EPR revealed slow rotational dynamics indistinguishable from that of rigor. Similar results were obtained for the CD using a bifunctional spin label to cross-link SH1-SH2, but the CD was more disordered than the LCD. We conclude that SH1-SH2 cross-linking traps a state in which both the CD and LCD are intermediate between relaxation (highly disordered and microsecond dynamics) and rigor (highly ordered and rigid), supporting the hypothesis that the cross-linked state is an A.M'D.P analog on the force generation pathway.