BackgroundNonsuicidal self-injury (NSSI) is common among adolescents with depressive disorders and is associated with substantial clinical burden. Machine learning may assist in identifying multidimensional patterns associated with NSSI, although its incremental value beyond direct clinical assessment remains uncertain. ObjectiveThis study compares 7 supervised machine learning model variants for classifying past-year NSSI status among adolescents with depressive disorders and conducts detailed analyses of interpretability, sensitivity, subgroup performance, and clinical utility using a random forest model as the focal exploratory model. MethodsWe retrospectively identified 437 adolescents aged 10-19 years who received inpatient or outpatient psychiatric care at 4 hospitals in Zhejiang Province, China, between January 2021 and December 2023. A total of 410 eligible participants were included, of whom 274 reported NSSI during the preceding year. The primary analysis included 67 assessment-time sociodemographic, physiological, biochemical, and clinical-behavioral features. The sample was divided into a training set (287/410, 70%) and an untouched held-out test set (123/410, 30%) using stratified random sampling. Hyperparameters were optimized using 5-fold cross-validation within the training set. Performance was evaluated using discrimination, classification metrics, calibration, and decision-curve analysis. Random forest model behavior was examined using Gini importance, permutation importance, and Shapley Additive Explanations (SHAP). ResultsThe radial basis function support vector machine achieved the highest mean cross-validated area under the receiver operating characteristic curve (AUROC) in the training set (0.720), whereas the random forest showed the highest observed AUROC and area under the precision-recall curve (AUPRC) and the lowest Brier score in the held-out test set. The random forest achieved an AUROC of 0.681 (95% CI 0.583-0.778), an AUPRC of 0.808 (95% CI 0.741-0.874), an accuracy of 0.667 (95% CI 0.585-0.748), a sensitivity of 0.720 (95% CI 0.622-0.817), a specificity of 0.561 (95% CI 0.415-0.707), and a Brier score of 0.212 (95% CI 0.188-0.236). Its calibration slope was 1.013. Pairwise AUROC differences between the random forest and alternative models were not statistically significant after Holm correction (Holm-adjusted P values ranged from .17 to .81). The full random forest model did not outperform a suicidal-ideation-only model (AUROC 0.701), whereas excluding suicidal ideation and previous suicide attempt reduced the AUROC to 0.540. A biochemical-only model showed near-chance discrimination (AUROC 0.513). Decision-curve analysis indicated that the random forest provided greater net benefit than the assess-all and assess-none strategies across threshold probabilities of approximately 0.51-0.80. Suicidal ideation showed the most consistent importance across Gini, SHAP, and permutation analyses. ConclusionsThe random forest had the highest observed AUROC and AUPRC and the lowest Brier score in the held-out test set among the evaluated algorithms, although the pairwise AUROC differences were not statistically significant after correction for multiple comparisons. Suicidality-related information contributed substantially to model discrimination, whereas routinely collected biochemical variables did not demonstrate clear incremental value. Because this was a retrospective cross-sectional classification study, the findings do not establish prospective risk prediction. External and prospective temporal validation is required before clinical implementation.
BackgroundStudies have indicated that maternal anxiety and depression during pregnancy can contribute to neurodevelopmental disorders and gut flora dysbiosis in children. However, it is unclear whether these factors increase the risk of autism spectrum disorder (ASD) with gastrointestinal (GI) symptoms in children. ObjectiveThis study aimed to analyze the relationship between maternal anxiety and depression during pregnancy and ASD with GI symptoms. MethodsA total of 470 children with ASD and 486 children with typical development were enrolled in the study. The caregivers of all enrolled children completed questionnaires, including information on maternal mental health and GI symptoms. ResultsThe percentage of maternal prenatal anxiety or depression was significantly higher in the ASD group than in the typical development group (P<.001). Children whose mothers experienced anxiety or depression during pregnancy had substantially higher risk of developing ASD than those whose mothers did not experience these conditions (unadjusted odds ratio [OR] 4.274, 95% CI 2.799-6.526; adjusted OR 3.897, 95% CI 2.481-6.119). For children with ASD, those whose mothers had prenatal anxiety or depression had a significantly higher risk of GI symptoms than those whose mothers did not have these conditions (unadjusted OR 1.883, 95% CI 1.209-2.934; adjusted OR 1.943, 95% CI 1.236-3.054). ConclusionsMothers who experience anxiety or depression during pregnancy may have a higher likelihood of their children developing ASD with GI symptoms.
BackgroundThe Canadian Occupational Performance Measure (COPM) has achieved global implementation across 35 countries in more than 20 language versions. However, its psychometric properties for children with physical disabilities remain unverified. ObjectiveThis study aimed to evaluate the interrater and test-retest reliability of the Arabic version of the COPM (Arabic-COPM) in this pediatric population. MethodsA cross-sectional study was conducted using a convenience sample of parents of children with physical disabilities recruited from multi-rehabilitation centers. For the interrater reliability assessment, 100 parents completed the Arabic-COPM through independent interviews with 2 raters. A subsample of 30 (30%) parents repeated the assessment after a 14-day interval for the test-retest reliability analysis, in which they identified their children’s 5 most significant daily activity problems and rated the performance and satisfaction levels for each problem using standardized 10-point scales. ResultsReliability analyses demonstrated excellent psychometric properties. Intraclass correlation coefficients (ICCs) for interrater reliability were 0.84 (performance) and 0.80 (satisfaction), while test-retest reliability coefficients were 0.83 and 0.91, respectively. Interrater agreement analysis revealed 73% concordance in problem prioritization between independent assessors. Bland-Altman plots indicated narrow limits of agreement for both performance (−1.69 to 1.60) and satisfaction scores (−1.03 to 0.98). Cohen κ analysis for problem categorization demonstrated slight-to-fair interrater agreement and slight-to-moderate test-retest agreement across COPM domains. Content analysis identified a predominant concern with daily activities (82%-85% of problems), particularly mobility functions, followed by dressing and eating subcategories. ConclusionsThe Arabic-COPM demonstrated excellent reliability metrics for both interrater consistency and temporal stability when administered to parents of children with physical disabilities. These robust measurement properties support its application as a validated instrument for implementing family-centered rehabilitation approaches in Arabic-speaking clinical contexts.
Background Over the past decade, smartphones have become an integral part of family life, shaping both daily routines and parent-child interactions. Although parental smartphone use in the presence of children is frequently framed as harmful through concepts such as phubbing and technoference, these constructs emphasize disruptive smartphone use and may bias interpretations toward negative effects. Moreover, most previous research has relied on cross-sectional self-report data, limiting insight into how parental smartphone use relates to children’s well-being in everyday life. Objective This study examined the associations between parental smartphone use in the presence of the child and children’s affective well-being, as well as perceived threats to children’s fundamental needs, using both self-reported and device-logged measures of smartphone use. Methods Individual parents (N=131) participated in a 10-day daily diary study. Each evening, parents reported their smartphone use in the presence of a selected target child, their child’s positive and negative affect, and perceived threats to their child’s fundamental psychological needs. In addition, Android users who provided consent installed a mobile sensing app that continuously recorded objective smartphone use throughout the study period. Multilevel models were used to examine both within-person (day-to-day fluctuations) and between-person associations separately for mothers and fathers while controlling for day of study, weekday vs weekend, and daily working hours. Results Higher parental smartphone use was largely unrelated to children’s affective well-being. However, mothers who, on average, reported greater overall smartphone use perceived greater threat to their child’s fundamental needs. Similarly, on days when fathers reported using their smartphones more than usual, they also perceived greater threats to their child’s fundamental needs. In regard to positive affect, no significant associations were found for children’s positive affect perceived by their parents, indicating that parents who, on average, used their smartphones more frequently around their child did not report lower child positive affect. Regarding negative affect, fathers who reported higher overall smartphone use also reported higher child negative affect. Additionally, on days when mothers used their smartphone more than usual, they perceived their child as displaying more negative affect. None of these associations emerged when parental smartphone use was assessed using device-logged data. Conclusions Taken together, these findings suggest that perceived child need threat may be linked to parents’ subjective perceptions and interpretations of their smartphone use rather than to their digitally traced smartphone behavior. This discrepancy underscores the importance of distinguishing between subjective and objective assessments of parental smartphone use and highlights the need for future research that integrates children’s perspectives and interactional context to clarify when and how parental smartphone use may impact children’s well-being.
Abstract Background Adolescents’ high engagement with smartphones creates opportunities for mobile health (mHealth) apps targeting dietary behaviors. However, despite the availability of diverse dietary mHealth apps, user engagement remains low. Existing quantitative literature has identified barriers and facilitators to app effectiveness in youth (aged 10-19 years) but has not fully captured the nuanced perceptions of adolescents (aged 12-19 years), who represent a unique life stage. Synthesizing similarities and differences in adolescents’ perceptions of what they value in dietary mHealth apps from qualitative studies may provide a more comprehensive understanding of app engagement and future app design. Objective We aimed to synthesize adolescents’ perceptions of using diverse dietary mHealth apps. Methods A scoping review was conducted according to the 5-stage framework by Arksey and O’Malley. Eligible studies were published in English and reported on adolescents (aged 12-19 years) with perceptions of dietary mHealth apps using qualitative methods. Apps that were not specific to diet, relied only on SMS text messaging, were delivered as part of a school-based intervention or telemedicine service, or were housed on a website or social media platform were excluded. Five evidence sources (PsycINFO, Embase, Medline, Web of Science, and CINAHL) were searched. Extracted data were charted by 1 researcher following a pilot test. Similarities and differences in adolescents’ perceptions were synthesized across studies using thematic analysis. Results Of the 1195 search results, 1.2% (n=14) met the eligibility criteria. We identified six themes that represented adolescents’ perceptions of using dietary mHealth apps: (1) functionality and design, (2) content and depth, (3) social connection and support, (4) entertainment and gamification, (5) goal setting and motivation, and (6) perceived effectiveness. Overall, adolescents favorably perceived apps that were easy to navigate, fostered social connections, were personalizable, and were gamified. Apps that were felt to increase awareness of personal eating habits, often through tracking calories and foods consumed, and foster awareness of personal well-being were especially discussed in relation to perceived app effectiveness. Conclusions The findings of this review provide an innovative synthesis of adolescents’ perceptions of diverse dietary mHealth apps, using qualitative data. Notably, dietary mHealth apps that were perceived as easy to navigate, tailored (eg, in terms of content), and gamified and those that included social engagement were perceived favorably, regardless of their dietary target and features. Apps that were perceived to improve adolescents’ current dietary awareness were particularly valued. Inclusion of these elements in dietary mHealth apps may promote greater engagement. App developers should work with adolescents to establish how specific apps can incorporate these elements into their design, including how best to track dietary habits and integrate AI into app features.
