Use of machine learning approaches to predict transition of retention in care among people living with HIV in South Carolina: a real-world data study.

Maintaining retention in care (RIC) for people living with HIV (PLWH) helps achieve viral suppression and reduce onward transmission. This study aims to identify the best machine learning model that predicts the RIC transition over time. Extracting from the enhanced HIV/AIDS reporting system, this study included 9765 PLWH from 2005 to 2020 in South Carolina. Transition of RIC was defined as the change of RIC status in each two-year time window. We applied seven classifiers, such as Random Forest, Support Vector Machine, eXtreme Gradient Boosting and Long-short-term memory, for each lagged response to predict the subsequent year's RIC transition. Classification performance was assessed using balanced prediction accuracy, the area under the curve (AUC), recall, precision and F1 scores. The proportion of the four categories of RIC transition was 13.59%, 29.78%, 9.06% and 47.57%, respectively. Support Vector Machine was the best approach for every lag model based on both the F1 score (0.713, 0.717 and 0.719) and AUC (0.920, 0.925 and 0.928). The findings could facilitate the risk augment of PLWH who are prone to follow-up so that clinicians and policymakers could come up with more specific strategies and relocate resources for intervention to keep them sustained in HIV care.

Sexual orientation, gender identity and virologic failure among people with HIV: a cohort study in all of US research program.

BACKGROUND: Sexual and gender minorities (SGMs) are at higher risk of HIV incidence compared to their heterosexual cisgender counterparts. Despite the high HIV disease burden among SGMs, there was limited data on whether they are at higher risk of virologic failure, which may lead to potential disease progression and increased transmission risk. The All of Us (AoU) Research Program, a national community-engaged program aiming to improve health and facilitate health equity in the United States by partnering with one million participants, provides a promising resource for identifying a diverse and large volunteer TGD cohort. Leveraging various data sources available through AoU, the current study aims to explore the association between sexual orientation and gender identity (SOGI) and longitudinal virologic failure among adult people with HIV (PWH) in the US. METHODS: This retrospective cohort study used integrated electronic health records (EHR) and self-reported survey data from the All of Us (AoU) controlled tier data, version 7, which includes participants enrolled in the AoU research program from May 31, 2017, to July 1, 2022. Based on participants' sexual orientation, gender identity, and sex assigned at birth, their SOGI were categorized into six groups, including cisgender heterosexual women, cisgender heterosexual men, cisgender sexual minority women, cisgender sexual minority men, gender minority people assigned female at birth of any sexual orientation, and gender minority people assigned male at birth of any sexual orientation. Yearly virologic failure was defined yearly after one's first viral load testing, and individuals with at least one viral load test > 50 copies/mL during a year were defined as having virologic failure at that year. Generalized linear mixed-effects models were used to explore the association between SOGI and longitudinal virologic failure while adjusting for potential confounders, including age, race, ethnicity, education attainment, income, and insurance type. RESULTS: A total of 1,546 eligible PWH were extracted from the AoU database, among whom 1,196 (77.36%) had at least one viral failure and 773 (50.00%) belonged to SGMs. Compared to cisgender heterosexual women, cisgender sexual minority women (adjusted Odds Ratio [aOR] = 1.85, 95% CI: 1.05-3.27) were at higher risk of HIV virologic failure. Additionally, PWH who were Black vs. White (aOR = 2.15, 95% CI: 1.52-3.04) and whose insurance type was Medicaid vs. Private insurance (aOR = 2.07, 95% CI: 1.33-3.21) were more likely to experience virologic failure. CONCLUSIONS: Maintaining frequent viral load monitoring among sexual minority women with HIV is warranted because it allows early detection of virologic failure, which could provide opportunities for interventions to strengthen treatment adherence and prevent HIV transmission. To understand the specific needs of subgroups of SGMs, future research needs to examine the mechanisms for SOGI-based disparities in virologic failure and the combined effects of multi-level psychosocial and health behavior characteristics.

Accuracy of 3-dimensional stereophotogrammetry: Comparison of the 3dMD and Bellus3D facial scanning systems with one another and with direct anthropometry.

INTRODUCTION: The objective of this study was to compare the accuracy of 3-dimensional (3D) digital facial photographs taken by the Bellus3D Face Camera Pro (Bellus3D) (Bellus3D Inc, Los Gatos, Calif) and the 3dMDface system (3dMD) (3dMD Inc, Atlanta, Ga) with one another and with direct anthropometry (DA). METHODS: A mannequin head was selected as the research object. Twenty facial landmarks were labeled on the basis of the 8 interlandmark distances and 5 angles that were defined. A 150-mm digital Vernier caliper (Mitutoyo Inc, Tokyo, Japan) with an accuracy of 0.02 mm was applied to directly measure the interlandmark distances, and the angles were calculated according to the law of cosines. All the measurements were conducted 3 times by each operator under identical conditions. Then, each scanner was used to acquire 3D photographs 5 times, generating 10 digital stereophotographs. Linear distances and angles were measured on the 3D facial photographs reconstructed with open-source MeshLab software (ISTI [Italian National Research Council], Rome, Italy). Each linear distance and angle were measured 3 times by 1 operator, and 3 examiners conducted the measurements independently. To obtain the trueness, equivalence tests were applied to compare the measurements of the 2 scanners and DA. In addition, the average absolute deviations were calculated to directly compare the trueness of 3 methods (Bellus3D vs 3dMD vs DA). Finally, the intraclass correlation coefficient was used to assess the interobserver agreement and the precision of 2 scanners. RESULTS: As for the trueness, 7 out of 8 of the linear distance measurements (N-Pn, Sn-Pog, ORE-IRE, OLE-ILE, RLC-RMC, LLC-LMC, and CR-CL) and 3 out of 5 of the angular measurements (MLA, NFA, and INI) obtained by 3dMD were equivalent to those obtained by DA. Five out of 8 measurements (N-Pn, Sn-Pog, RLC-RMC, LLC-LMC, and CR-CL) and 1 out of 5 of the angular measurements (MLA) obtained by Bellus3D were equivalent to the measurements obtained with DA. All but 3 of the measurements (ORE-IRE, NFA, and INI) obtained with Bellus3D were equivalent to 3dMD. The mean absolute difference between 3dMD and DA was 0.36 ± 0.20 mm and 0.45° ± 0.56°; the deviation between Bellus3D and DA was 0.61 ± 0.47 mm and 0.99° ± 0.61°; and the deviation between Bellus3D and 3dMD was 0.38 ± 0.37 mm and 0.62° ± 0.39°. Regarding the precision of the 2 scanners, the intraclass correlation coefficient value of 3dMD was approximately 1.00, and that of Bellus3D was 0.99. The interobserver agreement for each linear and angular measurement was 0.99. CONCLUSIONS: The trueness of each scanner was clinically acceptable for diagnosis and treatment planning. The precision of 3D photographs obtained by 3dMD and Bellus3D showed good scanning repeatability. The interobserver agreement between the 3 operators was rated as excellent (0.99).

The correlation of altitude with gingival status among adolescents in western China: a cross-sectional study.

Periodontal disease is common in Chinese adolescents. There is little information about the effect of different altitudes on gingival health. This study aimed to investigate the gingival status at different altitudes and to identify relative factors that affect adolescents' gingival status. A total of 1033 adolescents aged 12-14 years were included in this cross-sectional study in Ganzi (plateau, 1400 m, 2560 m, 3300 m) and Suining (plain, 300 m). Gingival status was assessed by the presence of gingival bleeding on probing (BOP) and dental calculus (DC). Demographic variables, socioeconomic status, dairy habits and oral health-related knowledge, attitudes and behaviors were obtained via questionnaire. Univariate and multivariate binary logistic regression analyses were performed to identify potential relative factors. A total of 64.09% and 77.15% of adolescents had BOP and DC, respectively. The prevalence rates of BOP and DC were higher in the plateau than the plain (P < 0.05). After adjusting for all other factors and interaction terms, residence altitudes of 2560 m [300 m as reference: P < 0.001, odds ratio (OR) = 4.072] and 3300 m (300 m as reference: P = 0.002, OR = 4.053) were significant relative factors of BOP, and an altitude of 2560 m (300 m as reference: P = 0.001, OR = 3.866, 1400 m as reference: P = 0.001, OR = 3.944) was an important relative factor of DC. Gingival bleeding and calculus deposits were common at different altitudes. High altitude was a significant relative factor of gingival bleeding and calculus deposits.

Influence of orthodontic treatment with premolar extraction on the spatial position of maxillary third molars in adult patients: a retrospective cohort cone-bean computed tomography study.

BACKGROUND: Based on low-dose radiation Cone-bean computed tomography (CBCT) images, This study aims to establish a space coordinate system, which offers more precise and comparable evaluation on changes of maxillary third molars influenced by orthodontic treatment with premolar extraction in adults. The system suggests promising application prospect in future studies related to CBCT superimposition and evaluation for its feasibility and efficiency. METHODS: Forty-nine maxillary third molars from 27 patients (mean age, 20.78 years) were included. CBCT images were obtained before and after orthodontic treatment with premolars extracted (mean treatment duration, 31.47 months). The changes in the position, angulation, and rotation of the third molars were evaluated with a space coordinate system using four landmarks: anterior nasal spine (ANS), posterior nasal spine (PNS), left and right orbitales. RESULTS: After orthodontic treatment, the third molars moved forward (adjusted mean, 1.44 mm) (p < 0.001) and downward (adjusted mean, 2.87 mm) (p < 0.001) accompanied by outward rotation of the crowns (adjusted mean, 5.38°) (p = 0.001), while changes in angulation were insignificant. CONCLUSIONS: This was the first study to systematically investigate the spatial position change of maxillary third molars in adult patients who received orthodontic treatment with premolar extraction. During the process, maxillary third molars moved downward and forward accompanied by outward rotation of the crowns. Orthodontists should take tooth movement potential into consideration when making extraction plans.

Violence against health care workers in China, 2013-2016: evidence from the national judgment documents.

BACKGROUND: Incidents of patient-initiated workplace violence against health care workers have been a subject of substantial public attention in China. Patient-initiated violence not only represents a risk of harm to health care providers but is also indicative of general tensions between doctors and patients which pose a challenge to improving health system access and quality. This study aims to provide a systematic, national-level characterization of serious workplace violence against health care workers in China. METHODS: This study extracted data from the China Judgment Online System, a comprehensive database of judgment documents. Three key phrases, "criminal case," "health care institution," and "health care worker" were used to search the China Judgment Online System for relevant cases between January 1, 2013, and December 31, 2016. Data extracted from identified cases was used to document the occurrence, the degree of risk, and the factors associated with serious workplace violence. RESULTS: In total, 459 criminal cases involving patient-initiated workplace violence against health care workers in China were reported and processed. The analysis revealed geographic heterogeneity in the occurrence of serious workplace violence, with lower incidence in western provinces compared to central and eastern provinces. Primary hospitals experienced the highest rates of serious workplace violence and emergency departments and doctors were at higher risk compared with other departments and health workers. Perpetrators were primarily male farmers aged 18 to 44 with low levels of education. The most frequently reported reasons of serious patient-initiated workplace violence included perceived medical malpractice by the perpetrator after the death of a patient, death of a patient with no other reason given, failures of the compensation negotiations after the death of a patient, and dissatisfaction with the treatment outcomes. CONCLUSIONS: Serious workplace violence against providers varies across regions and types of health care institutions in China. Perception of low-quality care is the most reported reason for violence. Efforts should be made to improve quality of care in the low-level health institutions and strengthen the doctor-patient communication during the whole course of service.

Computational phenotyping with the All of Us Research Program: identifying underrepresented people with HIV or at risk of HIV.

Objective: This study aims to identify the people living with HIV (PWH) and pre-exposure prophylaxis (PrEP) users in the All of Us (AoU) database by integrating information from both electronic health record (EHR)- and self-reported survey data. Methods: We identified PWH and PrEP users if they met the inclusion criterion by conditions, lab measurements, or medications related to HIV in EHR data or confirmed questions in the Survey data. Results: We evaluated the latest data release through July 1, 2022 in AoU. Through computational phenotyping, we identified 4575 confirmed and 3092 probable adult PWH and 564 PrEP users. PWH was most identified by a combination of medications and conditions (3324, 43.4%) and drug exposure alone (2191, 28.6%), then less commonly by survey data alone (608, 7.9%) and lab alone (81, 1.1%). Discussion and conclusion: Our methods serve as an overall framework for other researchers using AoU data for conducting HIV-related research.

Accuracy of RGB-D camera-based and stereophotogrammetric facial scanners: a comparative study.

OBJECTIVES: This study aimed to evaluate and compare the accuracy and inter-operator reliability of a low-cost red-green-blue-depth (RGB-D) camera-based facial scanner (Bellus3D Arc7) with a stereophotogrammetry facial scanner (3dMD) and to explore the possibility of the former as a clinical substitute for the latter. METHODS: A mannequin head was selected as the research object. In the RGB-D camera-based facial scanner group, the head was continuously scanned five times using an RGB-D camera-based facial scanner (Bellus3D Arc7), and the outcome data of each scan was then imported into CAD software (MeshLab) to reconstruct three-dimensional (3D) facial photographs. In the stereophotogrammetry facial scanner group, the mannequin head was scanned with a stereophotogrammetry facial scanner (3dMD). Selected parameters were directly measured on the reconstructed 3D virtual faces using a CAD software. The same parameters were then measured directly on the mannequin head using the direct anthropometry (DA) method as the gold standard for later comparison. The accuracy of the facial scanners was evaluated in terms of trueness and precision. Trueness was evaluated by comparing the measurement results of the two groups with each other and with that of DA using equivalence tests and average absolute deviations, while precision and inter-operator reliability were assessed using the intraclass correlation coefficient (ICC). A 3D facial mesh deviation between the two groups was also calculated for further reference using a 3D metrology software (GOM inspect pro). RESULTS: In terms of trueness, the average absolute deviations between RGB-D camera-based and stereophotogrammetry facial scanners, between RGB-D camera-based facial scanner and DA, and between stereophotogrammetry facial scanner and DA were statistically equivalent at 0.50±0.27 mm, 0.61±0.42 mm, and 0.28±0.14 mm, respectively. Equivalence test results confirmed that their equivalence was within clinical requirements (<1 mm). The ICC for each parameter was approximately 0.999 in terms of precision and inter-operator reliability. A 3D facial mesh analysis suggested that the deviation between the two groups was 0.37±0.01 mm. CONCLUSIONS: For facial scanners, an accuracy of <1 mm is commonly considered clinically acceptable. Both the RGB-D camera-based and stereophotogrammetry facial scanners in this study showed acceptable trueness, high precision, and inter-operator reliability. A low-cost RGB-D camera-based facial scanner could be an eligible clinical substitute for traditional stereophotogrammetry. CLINICAL SIGNIFICANCE: The low-cost RGB-D camera-based facial scanner showed clinically acceptable trueness, high precision, and inter-operator reliability; thus, it could be an eligible clinical substitute for traditional stereophotogrammetry.

Using Twitter Data to Estimate the Prevalence of Symptoms of Mental Disorders in the United States During the COVID-19 Pandemic: Ecological Cohort Study.

BACKGROUND: Existing research and national surveillance data suggest an increase of the prevalence of mental disorders during the COVID-19 pandemic. Social media platforms, such as Twitter, could be a source of data for estimation owing to its real-time nature, high availability, and large geographical coverage. However, there is a dearth of studies validating the accuracy of the prevalence of mental disorders on Twitter compared to that reported by the Centers for Disease Control and Prevention (CDC). OBJECTIVE: This study aims to verify the feasibility of Twitter-based prevalence of mental disorders symptoms being an instrument for prevalence estimation, where feasibility is gauged via correlations between Twitter-based prevalence of mental disorder symptoms (ie, anxiety and depressive symptoms) and that based on national surveillance data. In addition, this study aims to identify how the correlations changed over time (ie, the temporal trend). METHODS: State-level prevalence of anxiety and depressive symptoms was retrieved from the national Household Pulse Survey (HPS) of the CDC from April 2020 to July 2021. Tweets were retrieved from the Twitter streaming application programming interface during the same period and were used to estimate the prevalence of symptoms of mental disorders for each state using keyword analysis. Stratified linear mixed models were used to evaluate the correlations between the Twitter-based prevalence of symptoms of mental disorders and those reported by the CDC. The magnitude and significance of model parameters were considered to evaluate the correlations. Temporal trends of correlations were tested after adding the time variable to the model. Geospatial differences were compared on the basis of random effects. RESULTS: Pearson correlation coefficients between the overall prevalence reported by the CDC and that on Twitter for anxiety and depressive symptoms were 0.587 (P<.001) and 0.368 (P<.001), respectively. Stratified by 4 phases (ie, April 2020, August 2020, October 2020, and April 2021) defined by the HPS, linear mixed models showed that Twitter-based prevalence for anxiety symptoms had a positive and significant correlation with CDC-reported prevalence in phases 2 and 3, while a significant correlation for depressive symptoms was identified in phases 1 and 3. CONCLUSIONS: Positive correlations were identified between Twitter-based and CDC-reported prevalence, and temporal trends of these correlations were found. Geospatial differences in the prevalence of symptoms of mental disorders were found between the northern and southern United States. Findings from this study could inform future investigation on leveraging social media platforms to estimate symptoms of mental disorders and the provision of immediate prevention measures to improve health outcomes.

The use of one-stage meta-analytic method based on individual participant data for binary adverse events under the rule of three: a simulation study.

OBJECTIVE: In evidence synthesis practice, dealing with binary rare adverse events (AEs) is a challenging problem. The pooled estimates for rare AEs through traditional inverse variance (IV), Mantel-Haenszel (MH), and Yusuf-Peto (Peto) methods are suboptimal, as the biases tend to be large. We proposed the "one-stage" approach based on multilevel variance component logistic regression (MVCL) to handle this problem. METHODS: We used simulations to generate trials of individual participant data (IPD) with a series of predefined parameters. We compared the performance of the MVCL "one-stage" approach and the five classical methods (fixed/random effect IV, fixed/random effect MH, and Peto) for rare binary AEs under different scenarios, which included different sample size setting rules, effect sizes, between-study heterogeneity, and numbers of studies in each meta-analysis. The percentage bias, mean square error (MSE), coverage probability, and average width of the 95% confidence intervals were used as performance indicators. RESULTS: We set 52 scenarios and each scenario was simulated 1,000 times. Under the rule of three (a sample size setting rule to ensure a 95% chance of detecting at least one AE case), the MVCL "one-stage" IPD method had the lowest percentage bias in most of the situations and the bias remained at a very low level (<10%), when compared to IV, MH, and Peto methods. In addition, the MVCL "one-stage" IPD method generally had the lowest MSE and the narrowest average width of 95% confidence intervals. However, it did not show better coverage probability over the other five methods. CONCLUSIONS: The MVCL "one-stage" IPD meta-analysis is a useful method to handle binary rare events and superior compared to traditional methods under the rule of three. Further meta-analyses may take account of the "one-stage" IPD method for pooling rare event data.