BACKGROUND: Suicide is a significant public health issue. Many risk prediction tools have been developed to estimate an individual\'s risk of suicide. Risk prediction models can go beyond individual risk assessment; one important application of risk prediction models is population health planning. Suicide is a result of the interaction among the risk and protective factors at the individual, health care system, and community levels. Thus, policy and decision makers can play an important role in suicide prevention. However, few prediction models for the population risk of suicide have been developed.
OBJECTIVE: This study aims to develop and validate prediction models for the population risk of suicide using health administrative data, considering individual-, health system-, and community-level predictors.
METHODS: We used a case-control study design to develop sex-specific risk prediction models for suicide, using the health administrative data in Quebec, Canada. The training data included all suicide cases (n=8899) that occurred from January 1, 2002, to December 31, 2010. The control group was a 1% random sample of living individuals in each year between January 1, 2002, and December 31, 2010 (n=645,590). Logistic regression was used to develop the prediction models based on individual-, health care system-, and community-level predictors. The developed model was converted into synthetic estimation models, which concerted the individual-level predictors into community-level predictors. The synthetic estimation models were directly applied to the validation data from January 1, 2011, to December 31, 2019. We assessed the performance of the synthetic estimation models with four indicators: the agreement between predicted and observed proportions of suicide, mean average error, root mean square error, and the proportion of correctly identified high-risk regions.
RESULTS: The sex-specific models based on individual data had good discrimination (male model: C=0.79; female model: C=0.85) and calibration (Brier score for male model 0.01; Brier score for female model 0.005). With the regression-based synthetic models applied in the validation data, the absolute differences between the synthetic risk estimates and observed suicide risk ranged from 0% to 0.001%. The root mean square errors were under 0.2. The synthetic estimation model for males correctly predicted 4 of 5 high-risk regions in 8 years, and the model for females correctly predicted 4 of 5 high-risk regions in 5 years.
CONCLUSIONS: Using linked health administrative databases, this study demonstrated the feasibility and the validity of developing prediction models for the population risk of suicide, incorporating individual-, health system-, and community-level variables. Synthetic estimation models built on routinely collected health administrative data can accurately predict the population risk of suicide. This effort can be enhanced by timely access to other critical information at the population level.

UNASSIGNED: Regular physical activity and exercise are fundamental components of a healthy lifestyle for youth living with type 1 diabetes (T1D). Yet, few youth living with T1D achieve the daily minimum recommended levels of physical activity. For all youth, regardless of their disease status, minutes of physical activity compete with other daily activities, including digital gaming. There is an emerging area of research exploring whether digital games could be displacing other physical activities and exercise among youth, though, to date, no studies have examined this question in the context of youth living with T1D.
UNASSIGNED: We examined characteristics of digital gaming versus nondigital gaming (other exercise) sessions and whether youth with T1D who play digital games (gamers) engaged in less other exercise than youth who do not (nongamers), using data from the Type 1 Diabetes Exercise Initiative Pediatric study.
UNASSIGNED: During a 10-day observation period, youth self-reported exercise sessions, digital gaming sessions, and insulin use. We also collected data from activity wearables, continuous glucose monitors, and insulin pumps (if available).
UNASSIGNED: The sample included 251 youths with T1D (age: mean 14, SD 2 y; self-reported glycated hemoglobin A1c level: mean 7.1%, SD 1.3%), of whom 105 (41.8%) were female. Youth logged 123 digital gaming sessions and 3658 other exercise (nondigital gaming) sessions during the 10-day observation period. Digital gaming sessions lasted longer, and youth had less changes in glucose and lower mean heart rates during these sessions than during other exercise sessions. Youth described a greater percentage of digital gaming sessions as low intensity (82/123, 66.7%) when compared to other exercise sessions (1104/3658, 30.2%). We had 31 youths with T1D who reported at least 1 digital gaming session (gamers) and 220 youths who reported no digital gaming (nongamers). Notably, gamers engaged in a mean of 86 (SD 43) minutes of other exercise per day, which was similar to the minutes of other exercise per day reported by nongamers (mean 80, SD 47 min).
UNASSIGNED: Digital gaming sessions were longer in duration, and youth had less changes in glucose and lower mean heart rates during these sessions when compared to other exercise sessions. Nevertheless, gamers reported similar levels of other exercise per day as nongamers, suggesting that digital gaming may not fully displace other exercise among youth with T1D.

UNASSIGNED: This Research Letter describes the increasing trend of almost-constant social media use among California adolescents and the association with serious psychological distress, focusing on the influence of familial and experiential factors.

OBJECTIVE: To examine the impact of About Us, an innovative healthy relationships intervention that promotes positive adolescent romantic relationships and the use of effective contraceptives, on improving behavior, attitudes, and intentions related to sexual intercourse, relationship communication, and conflict resolution at 3- and 9-month follow-up, compared to services as usual.
METHODS: This was a multi-site, two-group, parallel, randomized-controlled trial with an intervention/comparison allocation ratio of 3:2 conducted at seven high schools in California between February 2018 and May 2021.
RESULTS: Overall, our study did not find statistically significant evidence of improved behavior, attitudes, and intentions related to sexual intercourse, relationship communication, and conflict resolution among participants (14-18 years old) randomized to the intervention group (n = 316) compared to services as usual (n = 217) during follow-up (group x time; p > .05). Exploratory within group analyses showed that, compared to baseline, at the 3-month follow-up, the prevalence of reporting having had sex increased in the control group relative to intervention group (+19% vs. +9%, p < .01). Our sub-group analyses showed that changes in condom use intentions scores differed across school sites (group x time x school; p < .01); mixed (positive and negative) trends were observed for intervention effect, and schools with positive intervention effect trends tended to have greater program participation.
CONCLUSIONS: About Us did not show statistically significant positive impacts on primary or secondary outcomes as anticipated. Our exploratory findings show evidence of some promising trends of intervention effects at the school-level, suggesting a need for better tailored intervention components and/or delivery to address the unique environmental contexts of participants. Overall, the context of study implementation was negatively affected by the COVID-19 pandemic and challenges related to using a non-classroom delivery intervention approach. Combined, these factors may have contributed to the study null findings. Moreover, it is difficult to know (or determine) the intervention\'s impact under more ideal conditions (i.e., no COVID pandemic).

BACKGROUND: Valid assessment tools are needed when investigating adherence to national dietary and lifestyle guidelines.
OBJECTIVE: The relative validity of the new digital food frequency questionnaire, the DIGIKOST-FFQ, against 7-day weighed food records and activity sensors was investigated.
METHODS: In total, 77 participants were included in the validation study and completed the DIGIKOST-FFQ and the weighed food record, and of these, 56 (73%) also used the activity sensors. The DIGIKOST-FFQ estimates the intake of foods according to the Norwegian food-based dietary guidelines (FBDGs) in addition to lifestyle factors.
RESULTS: At the group level, the DIGIKOST-FFQ showed good validity in estimating intakes according to the Norwegian FBDG. The median differences were small and well below portion sizes for all foods except \"water\" (median difference 230 g/day). The DIGIKOST-FFQ was able to rank individual intakes for all foods (r=0.2-0.7). However, ranking estimates of vegetable intakes should be interpreted with caution. Between 69% and 88% of the participants were classified into the same or adjacent quartile for foods and between 71% and 82% for different activity intensities. The Bland-Altman plots showed acceptable agreements between DIGIKOST-FFQ and the reference methods. The absolute amount of time in \"moderate to vigorous intensity\" was underestimated with the DIGIKOST-FFQ. However, estimated time in \"moderate to vigorous intensity,\" \"vigorous intensity,\" and \"sedentary time\" showed acceptable correlations and good agreement between the methods. The DIGIKOST-FFQ was able to identify adherence to the Norwegian FBDG and physical activity recommendations.
CONCLUSIONS: The DIGIKOST-FFQ gave valid estimates of dietary intakes and was able to identify individuals with different degrees of adherence to the Norwegian FBDG and physical activity recommendations. Moderate physical activity was underreported, water was overreported, and vegetables showed poor correlation, which are important to consider when interpreting the data. Good agreement was observed between the methods in estimating dietary intakes and time in \"moderate to vigorous physical activity,\" \"sedentary time,\" and \"sleep.\"

BACKGROUND: Tinnitus is a complex and heterogeneous disease that has been identified as a common manifestation of COVID-19. To gain a comprehensive understanding of tinnitus symptoms in individuals following COVID-19 infection, we conducted an online survey called the China Ear Nose and Throat Symptom Survey in the COVID-19 Pandemic (CENTSS) among the Chinese population.
OBJECTIVE: Our objective was to investigate tinnitus and ear-related symptoms after COVID-19 infection in the Chinese population, with the aim of providing a solid empirical foundation for improved health care. The findings from CENTSS can contribute to the development of enhanced management strategies for tinnitus in the context of long COVID. By gaining a better understanding of the factors contributing to tinnitus in individuals with COVID-19, health care providers can tailor interventions to address the specific needs of affected patients. Furthermore, this study serves as a basis for research on the long-term consequences of COVID-19 infection and its associated tinnitus symptoms.
METHODS: A quantitative, online, cross-sectional survey study design was used to explore the impact of the COVID-19 pandemic on experiences with tinnitus in China. Data were collected through an online questionnaire designed to identify the presence of tinnitus and its impacts. Descriptive statistics were used to analyze individuals\' demographic characteristics, COVID-19 infection-related ear symptoms, and the cognitive and emotional implications of tinnitus. Univariable and multivariable logistic regression analyses were used to model the cross-sectional baseline associations between demographic characteristics, noise exposure, educational level, health and lifestyle factors, and the occurrence of tinnitus.
RESULTS: Between December 19, 2022, and February 1, 2023, we obtained responses from 1262 Chinese participants representing 24 regions, with an average age of 37 years. Among them, 540 patients (42.8%) reported experiencing ear-related symptoms after COVID-19 infection. Only 114 (9%) of these patients sought medical attention specifically for their ear symptoms, while 426 (33.8%) did not seek hospital care. Tinnitus emerged as the most prevalent and impactful symptom among all ear-related symptoms experienced after COVID-19 infection. Of the respondents, female participants (688/888, 77.78%), younger individuals (<30 years), individuals with lower education levels, participants residing in western China, and those with a history of otolaryngology diseases were more likely to develop tinnitus following COVID-19 infection.
CONCLUSIONS: In summary, tinnitus was identified as the most common ear-related symptom during COVID-19 infection. Individuals experiencing tinnitus after COVID-19 infection were found to have poorer cognitive and emotional well-being. Different ear-related symptoms in patients post-COVID-19 infection may suggest viral invasion of various parts of the ear. It is therefore crucial to monitor and manage hearing-related changes resulting from COVID-19 as clinical services resume.

Adolescents face issues regarding physical health, mental health, sexual health, drug and alcohol problems, stress, and peer pressure. Little is known about adolescents\' help-seeking behaviours in relation to health concerns. The general practitioner (GP) is usually the first point of contact for adolescents. The aim of this systematic review was to identify, describe, and summarize evidence on barriers and enablers experienced by adolescents when accessing GP-led primary care services. Systematic searches using four electronic databases (PsycINFO, MEDLINE, CINAHL, and SocINDEX) were conducted and the quality of the included studies was appraised. Six studies were included in this review. Findings indicate that barriers to GP access relate to trust, confidentiality, privacy, and communication. Adolescents also reported barriers such as transport, cost, and lack of information. Adolescents reported enablers being services that are sensitive to their needs, healthcare professionals who understand them, and services that are flexible regarding out of hours access. Listening to and acting on the voice of adolescents is important to developing youth-friendly services.

UNASSIGNED: Medical history contributes approximately 80% to a diagnosis, although physical examinations and laboratory investigations increase a physician\'s confidence in the medical diagnosis. The concept of artificial intelligence (AI) was first proposed more than 70 years ago. Recently, its role in various fields of medicine has grown remarkably. However, no studies have evaluated the importance of patient history in AI-assisted medical diagnosis.
UNASSIGNED: This study explored the contribution of patient history to AI-assisted medical diagnoses and assessed the accuracy of ChatGPT in reaching a clinical diagnosis based on the medical history provided.
UNASSIGNED: Using clinical vignettes of 30 cases identified in The BMJ, we evaluated the accuracy of diagnoses generated by ChatGPT. We compared the diagnoses made by ChatGPT based solely on medical history with the correct diagnoses. We also compared the diagnoses made by ChatGPT after incorporating additional physical examination findings and laboratory data alongside history with the correct diagnoses.
UNASSIGNED: ChatGPT accurately diagnosed 76.6% (23/30) of the cases with only the medical history, consistent with previous research targeting physicians. We also found that this rate was 93.3% (28/30) when additional information was included.
UNASSIGNED: Although adding additional information improves diagnostic accuracy, patient history remains a significant factor in AI-assisted medical diagnosis. Thus, when using AI in medical diagnosis, it is crucial to include pertinent and correct patient histories for an accurate diagnosis. Our findings emphasize the continued significance of patient history in clinical diagnoses in this age and highlight the need for its integration into AI-assisted medical diagnosis systems.

BACKGROUND: Digital interventions are increasingly popular for the provision of nonpharmacological pain interventions, but few exist for adolescents with menstrual pain. User-centered design involves incorporating users across phases of digital health intervention design, development, and implementation and leads to improved user engagement and outcomes. A needs assessment is the first step of this approach.
OBJECTIVE: The goal of this study was to conduct a needs assessment to understand menstrual pain management needs and preferences and mindfulness experiences, preferences, and knowledge of adolescents with menstrual pain to inform the future development of an app for managing menstrual pain.
METHODS: We used an explanatory sequential mixed method design that included a survey followed by focus groups. Adolescents aged 13-17 years completed a survey (n=111) and participated in focus groups (n=16). Data were analyzed using descriptive statistics and thematic content analysis and synthesized to provide specific recommendations based on adolescent responses.
RESULTS: Adolescents (n=111) who completed the survey reported a moderate understanding of mindfulness and menstrual pain. Over three-quarters (n=87, 78%) of participants practiced some form of mindfulness and 87% (n=97) of survey participants used nonpharmacological pain management strategies. Teens had a moderate perception that mindfulness could help their menstrual pain (mean 4.51/10, SD 2.45, with higher scores suggesting more interest). Themes were generated related to mindfulness experiences, menstrual pain knowledge and experiences, and app functionality. These themes underscored adolescents\' need for continued support and flexible access to mindfulness activities; their awareness of multiple influences to pain, with potential for further education in this area; and the need for menstrual pain-specific content, along with content relevant to typical day-to-day experiences of adolescents.
CONCLUSIONS: Adolescents with menstrual pain have an interest in using a mindfulness app for pain but have unique needs that need to be addressed to ensure app engagement and relevance for this population. Concrete recommendations for future app development are provided.

BACKGROUND: Real-time surveillance of emerging infectious diseases necessitates a dynamically evolving, computable case definition, which frequently incorporates symptom-related criteria. For symptom detection, both population health monitoring platforms and research initiatives primarily depend on structured data extracted from electronic health records.
OBJECTIVE: This study sought to validate and test an artificial intelligence (AI)-based natural language processing (NLP) pipeline for detecting COVID-19 symptoms from physician notes in pediatric patients. We specifically study patients presenting to the emergency department (ED) who can be sentinel cases in an outbreak.
METHODS: Subjects in this retrospective cohort study are patients who are 21 years of age and younger, who presented to a pediatric ED at a large academic children\'s hospital between March 1, 2020, and May 31, 2022. The ED notes for all patients were processed with an NLP pipeline tuned to detect the mention of 11 COVID-19 symptoms based on Centers for Disease Control and Prevention (CDC) criteria. For a gold standard, 3 subject matter experts labeled 226 ED notes and had strong agreement (F1-score=0.986; positive predictive value [PPV]=0.972; and sensitivity=1.0). F1-score, PPV, and sensitivity were used to compare the performance of both NLP and the International Classification of Diseases, 10th Revision (ICD-10) coding to the gold standard chart review. As a formative use case, variations in symptom patterns were measured across SARS-CoV-2 variant eras.
RESULTS: There were 85,678 ED encounters during the study period, including 4% (n=3420) with patients with COVID-19. NLP was more accurate at identifying encounters with patients that had any of the COVID-19 symptoms (F1-score=0.796) than ICD-10 codes (F1-score =0.451). NLP accuracy was higher for positive symptoms (sensitivity=0.930) than ICD-10 (sensitivity=0.300). However, ICD-10 accuracy was higher for negative symptoms (specificity=0.994) than NLP (specificity=0.917). Congestion or runny nose showed the highest accuracy difference (NLP: F1-score=0.828 and ICD-10: F1-score=0.042). For encounters with patients with COVID-19, prevalence estimates of each NLP symptom differed across variant eras. Patients with COVID-19 were more likely to have each NLP symptom detected than patients without this disease. Effect sizes (odds ratios) varied across pandemic eras.
CONCLUSIONS: This study establishes the value of AI-based NLP as a highly effective tool for real-time COVID-19 symptom detection in pediatric patients, outperforming traditional ICD-10 methods. It also reveals the evolving nature of symptom prevalence across different virus variants, underscoring the need for dynamic, technology-driven approaches in infectious disease surveillance.