Search

Purpose: Implant‐mediated guided growth (IMGG) is used to address coronal plane deformity in skeletally immature patients. Few studies have reported on IMGG and simultaneous medial patellofemoral ligament (MPFL) reconstruction for paediatric patients with concurrent genu valgum and patellofemoral instability (PFI). This study aimed to report on the outcomes of these simultaneous procedures. Materials and Methods: This was a retrospective review of paediatric patients undergoing simultaneous MPFL reconstruction and IMGG between 2016 and 2023. Mechanical lateral distal femoral angle (mLDFA), hip–knee–ankle angle (HKA) and mechanical axis deviation (MAD) were measured on full‐length hip‐to‐ankle plain radiographs. Measurements were taken preoperatively, prior to implant removal and/or at final follow‐up with minimum 1‐year clinical follow‐up. Results: A total of 25 extremities in 22 patients (10 female) underwent simultaneous IMGG and MPFL reconstruction. The mean age at surgery was 12.6 ± 1.7 years. The mean duration of implant retention was 18.6 ± 11.3 months. Nineteen extremities (76%) underwent implant removal by final follow‐up. Preoperative HKA corrected from a mean of 5.8 ± 2.3° to −0.8 ± 4.5° at implant removal or final follow‐up (p < 0.001), with mLDFA and MAD similarly improving (both p < 0.001). HKA corrected a mean of 0.7 ± 0.9° per month, while mLDFA and MAD corrected a mean of 0.5 ± 0.6°/month and 2 ± 3 mm/month, respectively. At the time of implant removal or final follow‐up, 88% of patients demonstrated alignment within 5° of neutral. Only one extremity experienced subsequent PFI (4%). For 14 patients who underwent implant removal with further imaging at a mean of 7.8 ± 4.9 months, only one patient (7%) had a valgus rebound to an HKA > 5°. Conclusion: Simultaneous MPFL reconstruction and IMGG provided successful correction of lower extremity malalignment with only one recurrence of PFI. This approach is a reliable surgical option for skeletally immature patients with genu valgum and PFI. Level of Evidence: Level 4 case series. [ABSTRACT FROM AUTHOR]

Background: Adolescent idiopathic scoliosis results in three dimensional changes to a patient's body, which may change a patient's range of motion. Surface topography is an emerging technology to evaluate three dimensional parameters in patients with scoliosis. The goal of this paper is to introduce novel and reliable surface topographic measurements for the assessment of global coronal and sagittal range of motion of the spine in adolescents, and to determine if these measurements can distinguish between adolescents with lumbar scoliosis and those without scoliosis. Methods: This study is a retrospective cohort study of a prospectively collected registry. Using a surface topographic scanner, a finger to floor and lateral bending scans were performed on each subject. Inter- and intra-rater reliabilities were assessed for each measurement. ANOVA analysis was used to test comparative hypotheses. Results: Inter-rater reliability for lateral bending fingertip asymmetry (LBFA) and lateral bending acromia asymmetry (LBAA) displayed poor reliability, while the coronal angle asymmetry (CAA), coronal angle range of motion (CAR), forward bending finger to floor (FBFF), forward bending acromia to floor (FBAF), sagittal angle (SA), and sagittal angle normalized (SAN) demonstrated good to excellent reliability. There was a significant difference between controls and lumbar scoliosis patients for LBFA, LBAA, CAA and FBAF (p-values < 0.01). Conclusion: Surface topography yields a reliable and rapid process for measuring global spine range of motion in the coronal and sagittal planes. Using these tools, there was a significant difference in measurements between patients with lumbar scoliosis and controls. In the future, we hope to be able to assess and predict perioperative spinal mobility changes. [ABSTRACT FROM AUTHOR]

Background: Tibiofemoral bone bruise patterns seen on magnetic resonance imaging (MRI) are associated with ligamentous injuries in the acutely injured knee. Bone bruise patterns in multiligament knee injuries (MLKIs) and particularly their association with common peroneal nerve (CPN) injuries are not well described. Purpose: To analyze the tibiofemoral bone bruise patterns in MLKIs with and without peroneal nerve injury. Study Design: Case series; Level of evidence, 4. Methods: We retrospectively identified 123 patients treated for an acute MLKI at a level 1 trauma center between January 2001 and March 2021. Patients were grouped into injury subtypes using the Schenck classification. Within this cohort, patients with clinically documented complete (motor and sensory loss) and/or partial CPN palsies on physical examination were identified. Imaging criteria required an MRI scan on a 1.5 or 3 Tesla scanner within 30 days of the initial MLKI. Images were retrospectively interpreted for bone bruising patterns by 2 board-certified musculoskeletal radiologists. The location of the bone bruises was mapped on fat-suppressed T2-weighted coronal and sagittal images. Bruise patterns were compared among patients with and without CPN injury. Results: Of the 108 patients with a MLKI who met the a priori inclusion criteria, 26 (24.1%) were found to have a CPN injury (N = 20 complete; N = 6 partial) on physical examination. For CPN-injured patients, the most common mechanism of injury was high-energy trauma (N = 19 [73%]). The presence of a grade 3 posterolateral corner (PLC) injury (N = 25; odds ratio [OR], 23.81 [95% CI, 3.08-184.1]; P =.0024), anteromedial femoral condyle bone bruising (N = 24; OR, 21.9 [95% CI, 3.40-202.9]; P <.001), or a documented knee dislocation (N = 16; OR, 3.45 [95% CI, 1.38-8.62]; P =.007) was significantly associated with the presence of a CPN injury. Of the 26 patients with CPN injury, 24 (92.3%) had at least 1 anteromedial femoral condyle bone bruise. All 20 (100%) patients with complete CPN injury also had at least 1 anteromedial femoral condyle bone bruise on MRI. In our MLKI cohort, the presence of anteromedial femoral condyle bone bruising had a sensitivity of 92.3% and a specificity of 64.6% for the presence of CPN injury on physical examination. Conclusion: In our MLKI cohort, the presence of a grade 3 PLC injury had the greatest association with CPN injury. Additionally, anteromedial femoral condyle bone bruising on MRI was a highly sensitive finding that was significantly correlated with CPN injury on physical examination. The high prevalence of grade 3 PLC injuries and anteromedial tibiofemoral bone bruising suggests that these MLKIs with CPN injuries most commonly occurred from a hyperextension-varus mechanism caused by a high-energy blow to the anteromedial knee. [ABSTRACT FROM AUTHOR]

Case. We report an 11-year-old male who sustained a lesser tuberosity avulsion fracture of the proximal humerus upon making contact with a baseball during an at-bat. This injury was neglected for 14 months and was eventually detected by an axillary radiograph and subsequent MRI. He successfully underwent an open surgical repair and regained full range of motion and level of activity at 1-year follow-up. Conclusion. In pediatric baseball players, lesser tuberosity avulsion fractures may occur upon striking a baseball with a bat. Even after being neglected for several months, these injuries can be treated successfully with an open surgical repair. [ABSTRACT FROM AUTHOR]

Background: The onset of peak height velocity (PHV) guides the timing of interventions in the growing child. The purpose of the present study was to validate the Diméglio olecranon grading system and to compare these scores with the Risser/triradiate closure (TRC), proximal humerus, and Sanders hand scores.Methods: Eighty children with annual serial radiographs were selected from the Bolton-Brush collection. The olecranon apophysis was graded with use of lateral radiographs of the elbow. The mean age to PHV was determined for each stage, and reliability was calculated with use of an intraclass correlation coefficient (ICC). Olecranon stage was combined with age, sex, and height in a generalized estimating equation (GEE) model to predict PHV. Predictive performance of this model was evaluated with use of tenfold cross-validation such that the model was trained on 90% of the radiographs and was asked to predict the PHV of the remaining 10%.Results: PHV is closely associated with olecranon stage, with stage 1 occurring 3.0 years before PHV and stage 7 occurring 3.4 years after PHV. Stage 5 was found to occur at PHV. Scoring system reliability was high across an array of observers (ICC = 0.85 ± 0.07). The GEE model showed that this olecranon system outperforms the Risser/TRC system in predicting PHV and is comparable with the humerus and Sanders hand systems. When combined with age and sex, the olecranon system successfully predicted PHV such that 62% of PHV predictions were accurate within 6 months and 90% of PHV predictions were accurate within a year.Conclusions: Our data show that stage 5 occurs at PHV, contrary to previously published data. When combined with age and sex, the olecranon system successfully predicts PHV within a year in 90% of cases, establishing a single lateral view of the olecranon as a simple alternative to more complex grading systems. Last, we describe novel 3 variations in olecranon morphology and provide a guide for accurate olecranon staging.Clinical Relevance: Understanding PHV is critical in the treatment of many pediatric orthopaedic disorders. The revised olecranon staging system will allow for more accurate determination of this variable. [ABSTRACT FROM AUTHOR]

Background: A description of the precise locations of ligamentous and myotendinous injury patterns of acute posterolateral corner (PLC) injuries and their associated osseous and neurovascular injuries is lacking in the literature. Purpose: To characterize the ligamentous and myotendinous injury patterns and zones of injury that occur in acute PLC injuries and determine associated rates of peroneal nerve palsies and vascular injuries, as well as fracture and dislocation. Study Design: Case series; Level of evidence, 4. Methods: We retrospectively identified all patients treated for an acute multiligament knee injury (MLKI) at our level 1 trauma center from 2001 to 2018. From this cohort, all patients with PLC injuries were identified. Demographics, involved ligaments and tendons, neurovascular injury, and presence of fracture and dislocation were compared with the larger multiligament knee cohort. Incidence and location of injury of PLC structures—from proximal to midsubstance and distal injury—were recorded. Results: A total of 100 knees in 100 patients were identified as having MLKIs. A total of 74 patients (74%) had lateral-sided ligament injuries. Of these, 23 (31%) had a peroneal nerve palsy associated with their injury; 10 (14%), a vascular injury; and 23 (31%), a fracture. Patients with PLC injuries had higher rates of peroneal nerve injury as compared with those having acute MLKIs without a PLC injury (31% vs 4%; P =.005). Patients with a complete peroneal nerve palsy (n = 17) were less likely to regain function than those with a partial peroneal nerve palsy (n = 6; 12% vs 100%; P <.0001). Complete injury to the lateral collateral ligament (LCL) occurred in 71 of 74 (96%) PLC injuries, with 3 distinct patterns of injury demonstrated. Fibular avulsion of the LCL was the most common zone of injury (65%), followed by femoral avulsion (20%) and midsubstance tear (15%). Location of injury to the LCL was associated with the rate of peroneal nerve injury, with midsubstance tears and fibular avulsions associated with higher rates of peroneal nerve injury. Conclusion: MLKIs with involvement of the PLC are more likely to suffer peroneal nerve injury. The LCL is nearly always involved, and its location of injury is predictive of peroneal nerve injury. Patients with a complete peroneal nerve palsy at presentation are much less likely to regain function. [ABSTRACT FROM AUTHOR]

Highlights • An implementation of Shimp's Associative Learner (AL) was fit to multiple dependent variables from a concurrent VI VI schedule in-transition dataset. • AL also had difficulty accounting for the acquisition of consecutive reinforcers from the same alternative. • AL had difficulty accounting for the inter-response times and responding to the not-just-reinforced alternative. • An extended implementation of AL was proposed, which rectified many of the weaknesses of the previous implementation. Abstract From a model that generates event records, we may generate predictions for any dependent variable computable from those event records. In the present paper, we examined an implementation of Shimp's (1978) Associative Learner for two-alternative concurrent variable-interval variable-interval schedules in transition. In general, our Associative Learner implementation had difficulty accounting for the inter-response times as a function of the reinforcer ratio, the general frequency of inter-response times, the acquisition of consecutive reinforcers from the same alternative, and the response rates to the not-just-reinforced alternative after a reinforcer. An extended Associative Learner implementation was designed with provisions for increasing responding to the just-reinforced alternative after a reinforcer and organising bout-like responses. The extended implementation rectified many of the weaknesses of the base implementation. [ABSTRACT FROM AUTHOR]

A multivariate analysis is concerned with more than one dependent variable simultaneously. Models that generate event records have a privileged status in a multivariate analysis. From a model that generates event records, we may compute predictions for any dependent variable associated with those event records. However, because of the generality that is afforded to us by these kinds of models, we must carefully consider the selection of dependent variables. Thus, we present a conditional compromise heuristic for the selection of dependent variables from a large group of variables. The heuristic is applied to McDowell's Evolutionary Theory of Behavior Dynamics (ETBD) for fitting to a concurrent variable‐interval schedule in‐transition dataset. From the parameters obtained from fitting ETBD, we generated predictions for a wide range of dependent variables. Overall, we found that our ETBD implementation accounted well for various flavors of the log response ratio, but had difficulty accounting for the overall response rates and cumulative reinforcer effects. Based on these results, we argue that the predictions of our ETBD implementation could be improved by decreasing the base response probabilities, either by increasing the response latencies or by decreasing the sizes of the operant classes. [ABSTRACT FROM AUTHOR]

Models that generate event records have very general scope regarding the dimensions of the target behavior that we measure. From a set of predicted event records, we can generate predictions for any dependent variable that we could compute from the event records of our subjects. In this sense, models that generate event records permit us a freely multivariate analysis. To explore this proposition, we conducted a multivariate examination of Catania's Operant Reserve on single VI schedules in transition using a Markov Chain Monte Carlo scheme for Approximate Bayesian Computation. Although we found systematic deviations between our implementation of Catania's Operant Reserve and our observed data (e.g., mismatches in the shape of the interresponse time distributions), the general approach that we have demonstrated represents an avenue for modelling behavior that transcends the typical constraints of algebraic models. [ABSTRACT FROM AUTHOR]

Catania’s Operant Reserve (COR; Catania, 2005) is a computational model of operant behaviour. In COR, responding depletes the reserve while reinforcement replenishes the reserve. The replenishment to the reserve depends on the location of responses within the most recent inter-reinforcement interval. The rule that maps replenishment to the responses within an inter-reinforcement interval is given by a delay-of-reinforcement gradient (DOR). Previous research (Berg and McDowell, 2011) found that non-linear DORs produce sigmoidal response rates on single variable-interval schedules while a linear DOR produces hyperbolic response rates. Berg and McDowell took these sigmoidal response rates as evidence against the viability of COR for modelling variable-interval performance. However, Berg and McDowell did not consider the effect of the slope of each DOR. The present conjecture is that the response rates from COR can be made hyperbolic by manipulating the area under the DOR. Our results show that the manipulation of the area under the DOR allows COR to produce hyperbolic response rates regardless of the parametric form of the DOR. Hence, if COR is to be used to model single variable-interval performance, the reinforcement effect over time should be taken into consideration through the manipulation of the area under the DOR. [ABSTRACT FROM AUTHOR]

BackgroundUnfractionated heparin is widely used in the intensive care unit as an anticoagulant. However, weight-based heparin dosing has been shown to be suboptimal and may place patients at unnecessary risk during their intensive care unit stay. ObjectiveIn this study, we intended to develop and validate a machine learning–based model to predict heparin treatment outcomes and to provide dosage recommendations to clinicians. MethodsA shallow neural network model was adopted in a retrospective cohort of patients from the Multiparameter Intelligent Monitoring in Intensive Care III (MIMIC III) database and patients admitted to the Peking Union Medical College Hospital (PUMCH). We modeled the subtherapeutic, normal, and supratherapeutic activated partial thromboplastin time (aPTT) as the outcomes of heparin treatment and used a group of clinical features for modeling. Our model classifies patients into 3 different therapeutic states. We tested the prediction ability of our model and evaluated its performance by using accuracy, the kappa coefficient, precision, recall, and the F1 score. Furthermore, a dosage recommendation module was designed and evaluated for clinical decision support. ResultsA total of 3607 patients selected from MIMIC III and 1549 patients admitted to the PUMCH who met our criteria were included in this study. The shallow neural network model showed results of F1 scores 0.887 (MIMIC III) and 0.925 (PUMCH). When compared with the actual dosage prescribed, our model recommended increasing the dosage for 72.2% (MIMIC III, 1240/1718) and 64.7% (PUMCH, 281/434) of the subtherapeutic patients and decreasing the dosage for 80.9% (MIMIC III, 504/623) and 76.7% (PUMCH, 277/361) of the supratherapeutic patients, suggesting that the recommendations can contribute to clinical improvements and that they may effectively reduce the time to optimal dosage in the clinical setting. ConclusionsThe evaluation of our model for predicting heparin treatment outcomes demonstrated that the developed model is potentially applicable for reducing the misdosage of heparin and for providing appropriate decision recommendations to clinicians.

BackgroundMonitoring critically ill patients in intensive care units (ICUs) in real time is vitally important. Although scoring systems are most often used in risk prediction of mortality, they are usually not highly precise, and the clinical data are often simply weighted. This method is inefficient and time-consuming in the clinical setting. ObjectiveThe objective of this study was to integrate all medical data and noninvasively predict the real-time mortality of ICU patients using a gradient boosting method. Specifically, our goal was to predict mortality using a noninvasive method to minimize the discomfort to patients. MethodsIn this study, we established five models to predict mortality in real time based on different features. According to the monitoring, laboratory, and scoring data, we constructed the feature engineering. The five real-time mortality prediction models were RMM (based on monitoring features), RMA (based on monitoring features and the Acute Physiology and Chronic Health Evaluation [APACHE]), RMS (based on monitoring features and Sequential Organ Failure Assessment [SOFA]), RMML (based on monitoring and laboratory features), and RM (based on all monitoring, laboratory, and scoring features). All models were built using LightGBM and tested with XGBoost. We then compared the performance of all models, with particular focus on the noninvasive method, the RMM model. ResultsAfter extensive experiments, the area under the curve of the RMM model was 0.8264, which was superior to that of the RMA and RMS models. Therefore, predicting mortality using the noninvasive method was both efficient and practical, as it eliminated the need for extra physical interventions on patients, such as the drawing of blood. In addition, we explored the top nine features relevant to real-time mortality prediction: invasive mean blood pressure, heart rate, invasive systolic blood pressure, oxygen concentration, oxygen saturation, balance of input and output, total input, invasive diastolic blood pressure, and noninvasive mean blood pressure. These nine features should be given more focus in routine clinical practice. ConclusionsThe results of this study may be helpful in real-time mortality prediction in patients in the ICU, especially the noninvasive method. It is efficient and favorable to patients, which offers a strong practical significance.

BackgroundPrevious studies have shown that electronic cigarette (e-cigarette) users might be more vulnerable to COVID-19 infection and could develop more severe symptoms if they contract the disease owing to their impaired immune responses to viral infections. Social media platforms such as Twitter have been widely used by individuals worldwide to express their responses to the current COVID-19 pandemic. ObjectiveIn this study, we aimed to examine the longitudinal changes in the attitudes of Twitter users who used e-cigarettes toward the COVID-19 pandemic, as well as compare differences in attitudes between e-cigarette users and nonusers based on Twitter data. MethodsThe study dataset containing COVID-19–related Twitter posts (tweets) posted between March 5 and April 3, 2020, was collected using a Twitter streaming application programming interface with COVID-19–related keywords. Twitter users were classified into two groups: Ecig group, including users who did not have commercial accounts but posted e-cigarette–related tweets between May 2019 and August 2019, and non-Ecig group, including users who did not post any e-cigarette–related tweets. Sentiment analysis was performed to compare sentiment scores towards the COVID-19 pandemic between both groups and determine whether the sentiment expressed was positive, negative, or neutral. Topic modeling was performed to compare the main topics discussed between the groups. ResultsThe US COVID-19 dataset consisted of 4,500,248 COVID-19–related tweets collected from 187,399 unique Twitter users in the Ecig group and 11,479,773 COVID-19–related tweets collected from 2,511,659 unique Twitter users in the non-Ecig group. Sentiment analysis showed that Ecig group users had more negative sentiment scores than non-Ecig group users. Results from topic modeling indicated that Ecig group users had more concerns about deaths due to COVID-19, whereas non-Ecig group users cared more about the government’s responses to the COVID-19 pandemic. ConclusionsOur findings show that Twitter users who tweeted about e-cigarettes had more concerns about the COVID-19 pandemic. These findings can inform public health practitioners to use social media platforms such as Twitter for timely monitoring of public responses to the COVID-19 pandemic and educating and encouraging current e-cigarette users to quit vaping to minimize the risks associated with COVID-19.

BackgroundInstagram is a popular social networking platform for users to upload pictures sharing their experiences. Instagram has been widely used by vaping companies and stores to promote electronic cigarettes (e-cigarettes), as well as by public health entities to communicate the risks of e-cigarette use (vaping) to the public. ObjectiveWe aimed to characterize current vaping-related content on Instagram through descriptive analyses. MethodsFrom Instagram, 42,951 posts were collected using vaping-related hashtags in November 2019. The posts were grouped as (1) pro-vaping, (2) vaping warning, (3) neutral to vaping, and (4) not related to vaping based on the attitudes to vaping expressed within the posts. From these Instagram posts and the corresponding 18,786 unique Instagram user accounts, 200 pro-vaping and 200 vaping-warning posts as well as 200 pro-vaping and 200 vaping-warning user accounts were randomly selected for hand coding. Furthermore, follower counts and media counts of the Instagram user accounts as well as the “like” counts and hashtags of the posts were compared between pro-vaping and vaping-warning groups. ResultsThere were more posts in the pro-vaping group (41,412 posts) than there were in the vaping-warning group (1539 posts). The majority of pro-vaping images were product display images (163/200, 81.5%), and the most popular image type in vaping-warning posts was educational (95/200, 47.5%). The highest proportion of pro-vaping user account type was vaping store (110/189, 58.1%), and the store account type had the highest mean number of posts (10.33 posts/account). The top 3 vaping-warning user account types were personal (79/155, 51%), vaping-warning community (37/155, 23.9%), and community (35/155, 22.6%), of which the vaping-warning community had the highest mean number of posts (3.68 posts/account). Pro-vaping user accounts had more followers (median 850) and media (median 232) than vaping-warning user accounts had (follower count: median 191; media count: 92). Pro-vaping posts had more “likes” (median 22) and hashtags (mean 20.39) than vaping-warning posts had (“like” count: median 12; hashtags: mean 7.16). ConclusionsInstagram is dominated by pro-vaping content, and pro-vaping posts and user accounts seem to have more user engagement than vaping-warning accounts have. These results highlight the importance of regulating e-cigarette posts on social media and the urgency of identifying effective communication content and message delivery methods with the public about the health effects of e-cigarettes to ameliorate the epidemic of vaping in youth.

BackgroundIn late December 2019, a pneumonia caused by SARS-CoV-2 was first reported in Wuhan and spread worldwide rapidly. Currently, no specific medicine is available to treat infection with COVID-19. ObjectiveThe aims of this study were to summarize the epidemiological and clinical characteristics of 175 patients with SARS-CoV-2 infection who were hospitalized in Renmin Hospital of Wuhan University from January 1 to January 31, 2020, and to establish a tool to identify potential critical patients with COVID-19 and help clinical physicians prevent progression of this disease. MethodsIn this retrospective study, clinical characteristics of 175 confirmed COVID-19 cases were collected and analyzed. Univariate analysis and least absolute shrinkage and selection operator (LASSO) regression were used to select variables. Multivariate analysis was applied to identify independent risk factors in COVID-19 progression. We established a nomogram to evaluate the probability of progression of the condition of a patient with COVID-19 to severe within three weeks of disease onset. The nomogram was verified using calibration curves and receiver operating characteristic curves. ResultsA total of 18 variables were considered to be risk factors after the univariate regression analysis of the laboratory parameters (P

BackgroundIn recent years, flavored electronic cigarettes (e-cigarettes) have become popular among teenagers and young adults. Discussions about e-cigarettes and e-cigarette use (vaping) experiences are prevalent online, making social media an ideal resource for understanding the health risks associated with e-cigarette flavors from the users’ perspective. ObjectiveThis study aimed to investigate the potential associations between electronic cigarette liquid (e-liquid) flavors and the reporting of health symptoms using social media data. MethodsA dataset consisting of 2.8 million e-cigarette–related posts was collected using keyword filtering from Reddit, a social media platform, from January 2013 to April 2019. Temporal analysis for nine major health symptom categories was used to understand the trend of public concerns related to e-cigarettes. Sentiment analysis was conducted to obtain the proportions of positive and negative sentiment scores for all reported health symptom categories. Topic modeling was applied to reveal the topics related to e-cigarettes and health symptoms. Furthermore, generalized estimating equation (GEE) models were used to quantitatively measure potential associations between e-liquid flavors and the reporting of health symptoms. ResultsTemporal analysis showed that the Respiratory category was consistently the most discussed health symptom category among all categories related to e-cigarettes on Reddit, followed by the Throat category. Sentiment analysis showed higher proportions of positive sentiment scores for all reported health symptom categories, except for the Cancer category. Topic modeling conducted on all health-related posts showed that 17 of the top 100 topics were flavor related. GEE models showed different associations between the reporting of health symptoms and e-liquid flavor categories, for example, lower association of the Beverage flavors with Respiratory compared with other flavors and higher association of the Fruit flavors with Cardiovascular than other flavors. ConclusionsThis study identified different potential associations between e-liquid flavors and the reporting of health symptoms using social media data. The results of this study provide valuable information for further investigation of the health effects associated with different e-liquid flavors.

BackgroundThe number of electronic cigarette (e-cigarette) users has been increasing rapidly in recent years, especially among youth and young adults. More e-cigarette products have become available, including e-liquids with various brands and flavors. Various e-liquid flavors have been frequently discussed by e-cigarette users on social media. ObjectiveThis study aimed to examine the longitudinal prevalence of mentions of electronic cigarette liquid (e-liquid) flavors and user perceptions on social media. MethodsWe applied a data-driven approach to analyze the trends and macro-level user sentiments of different e-cigarette flavors on social media. With data collected from web-based stores, e-liquid flavors were classified into categories in a flavor hierarchy based on their ingredients. The e-cigarette–related posts were collected from social media platforms, including Reddit and Twitter, using e-cigarette–related keywords. The temporal trend of mentions of e-liquid flavor categories was compiled using Reddit data from January 2013 to April 2019. Twitter data were analyzed using a sentiment analysis from May to August 2019 to explore the opinions of e-cigarette users toward each flavor category. ResultsMore than 1000 e-liquid flavors were classified into 7 major flavor categories. The fruit and sweets categories were the 2 most frequently discussed e-liquid flavors on Reddit, contributing to approximately 58% and 15%, respectively, of all flavor-related posts. We showed that mentions of the fruit flavor category had a steady overall upward trend compared with other flavor categories that did not show much change over time. Results from the sentiment analysis demonstrated that most e-liquid flavor categories had significant positive sentiments, except for the beverage and tobacco categories. ConclusionsThe most updated information about the popular e-liquid flavors mentioned on social media was investigated, which showed that the prevalence of mentions of e-liquid flavors and user perceptions on social media were different. Fruit was the most frequently discussed flavor category on social media. Our study provides valuable information for future regulation of flavored e-cigarettes.