Using machine learning to identify factors associated with practice location of the healthcare workforce.

INTRODUCTION: Past studies examined factors associated with rural practice, but none employed newer machine learning (ML) methods to explore potential predictors. The primary aim of this study was to identify factors related to practice in a rural area. Secondary aims were to capture a more precise understanding of the demographic characteristics of the healthcare professions workforce in Utah (USA) and to assess the viability of ML as a predictive tool. METHODS: This study incorporated four datasets - the 2017 dental workforce, the 2016 physician workforce, the 2014 nursing workforce and the 2017 pharmacy workforce - collected by the Utah Medical Education Council. Supervised ML techniques were used to identify factors associated with practice location, the outcome variable of interest. RESULTS: The study sample consisted of 11 259 healthcare professionals with an average age of 46.6 years, of which 36.6% were males and 94.5% Caucasian. Four ML methods were applied to assess model performance by comparing accuracy, sensitivity, specificity and area under the receiver operating characteristic (ROC) curve. Of the methods used, support vector machine performed the best (accuracy 99.7%, precision 100%, sensitivity 100%, specificity 99.4% and ROC 0.997). The models identified income and rural upbringing as the top factors associated with rural practice. CONCLUSION: By far, income emerged as the most important factor associated with rural practice, suggesting that attractive income offers might help rural communities address health professional shortages. Rural upbringing was the next most important predictive factor, validating and updating earlier research. The performance of the ML algorithms suggests their usefulness as a tool to model other databases for individualized prediction.

Examination of orthodontic expenditures and trends in the United States from 1996 to 2016: disparities across demographics and insurance payers.

BACKGROUND: Orthodontics prevent and treat facial, dental, and occlusal anomalies. Untreated orthodontic problems can lead to significant dental public health issues, making it important to understand expenditures for orthodontic treatment. This study examined orthodontic expenditures and trends in the United States over 2 decades. METHODS: This study used data collected by the Medical Expenditure Panel Survey to examine orthodontic expenditures in the United States from 1996 to 2016. Descriptive statistics for orthodontic expenditures were computed and graphed across various groups. Trends in orthodontic expenditures were adjusted to the 2016 United States dollar to account for inflation and deflation over time. Sampling weights were applied in estimating per capita and total expenditures to account for non-responses in population groups. RESULTS: Total orthodontic expenditures in the United States almost doubled from $11.5 billion in 1996 to $19.9 billion in 2016 with the average orthodontic expenditure per person increasing from $42.69 in 1996 to $61.52 in 2016. Black individuals had the lowest per capita orthodontic visit expenditure at $30.35. Out-of-pocket expenses represented the highest total expenditure and although the amount of out-of-pocket expenses increased over the years, they decreased as a percentage of total expenditures. Public insurance increased the most over the study period but still accounted for the smallest percentage of expenditures. Over the course of the study, several annual decreases were interspersed with years of increased spending CONCLUSION: While government insurance expenditure increased over the study period, out of pocket expenditures remained the largest contributor. Annual decreases in expenditure associated with economic downturns and result from the reliance on out-of-pocket payments for orthodontic care. Differences in spending among groups suggest disparities in orthodontic care among the US population.

A National Study Exploring the Association Between Fluoride Levels and Dental Fluorosis.

Importance: While the effects of fluoride on health have been widely researched, fewer high-quality studies examine the association of fluoride levels in water and dental fluorosis. Objective: To investigate the association between fluoride exposure from drinking water and dental fluorosis. Design, Setting, and Participants: This cross-sectional study used the 2013-2014 and 2015-2016 National Health and Nutrition Examination Survey (NHANES) data (January 1, 2013, through December 31, 2016). NHANES uses a complex sampling technique to develop nationally representative sample estimates of the US population that consists of interviews and physical assessments. Children and adolescents aged 6 to 15 years were included because NHANES contains their data for all 3 forms of fluoride measures: plasma fluoride levels, water levels of fluoride, and dietary fluoride supplementation. Data were analyzed from January 1 to April 30, 2023. Exposures: Water and plasma fluoride levels were measured by laboratory personnel. Dietary fluoride supplement data were self-reported. Main Outcomes and Measures: The Dean's Fluorosis Index was used to evaluate fluorosis status for each tooth. The dental fluorosis severity value was based on the second most affected tooth. Independent variables included plasma and water fluoride concentrations and dietary fluoride supplementation. An independent samples t test was used to compare fluoride exposures between groups, and Pearson correlation assessed the association between plasma and water fluoride levels. To assess whether fluoride exposures were associated with dental fluorosis, logistic regression analyses were conducted. Results: There were 1543 participants in the 2013-2014 NHANES cycle (weighted proportion male, 51.9%; mean [SD] age, 11.0 [2.7] years) and 1452 in the 2015-2016 cycle (weighted proportion male, 52.6%; mean [SD] age, 11.1 [2.8] years). A weighted 87.3% exhibited some degree of fluorosis in the 2013-2014 cycle and 68.2% in the 2015-2016 cycle. Higher fluoride levels in water and plasma were significantly associated with higher odds of dental fluorosis (adjusted odds ratios, 2.378 [95% CI, 2.372-2.383] in the 2013-2014 cycle and 1.568 [95% CI, 1.564-1.571] in the 2015-2016 cycle). Conclusions and Relevance: The findings of this cross-sectional study suggest that exposure to higher concentrations of fluoride in water and having higher plasma levels of fluoride were associated with a greater risk of dental fluorosis. Further research can help policy makers develop policies that balance substantial caries prevention with the risk of dental fluorosis.

Predicting all-cause 90-day hospital readmission for dental patients using machine learning methods.

INTRODUCTION: Hospital readmission rates are an indicator of the health care quality provided by hospitals. Applying machine learning (ML) to a hospital readmission database offers the potential to identify patients at the highest risk for readmission. However, few studies applied ML methods to predict hospital readmission. This study sought to assess ML as a tool to develop prediction models for all-cause 90-day hospital readmission for dental patients. METHODS: Using the 2013 Nationwide Readmissions Database (NRD), the study identified 9260 cases for all-cause 90-day index admission for dental patients. Five ML classification algorithms including decision tree, logistic regression, support vector machine, k-nearest neighbors, and artificial neural network (ANN) were implemented to build predictive models. The model performance was estimated and compared by using area under the receiver operating characteristic curve (AUC), and accuracy, sensitivity, specificity, and precision. RESULTS: Hospital readmission within 90 days occurred in 1746 cases (18.9%). Total charges, number of diagnosis, age, number of chronic conditions, length of hospital stays, number of procedures, primary expected payer, and severity of illness emerged as the top eight important features in all-cause 90-day hospital readmission. All models had similar performance with ANN (AUC = 0.743) slightly outperforming the rest. CONCLUSION: This study demonstrates a potential annual saving of over $500 million if all of the 90-day readmission cases could be prevented for 21 states represented in the NRD. Among the methods used, the prediction model built by ANN exhibited the best performance. Further testing using ANN and other methods can help to assess important readmission risk factors and to target interventions to those at the greatest risk.

In an era of uncertainty: Impact of COVID-19 on dental education.

PURPOSE/OBJECTIVES: The coronavirus disease 2019 (COVID-19) pandemic arguably represents the worst public health crisis of the 21st century. However, no empirical study currently exists in the literature that examines the impact of the COVID-19 pandemic on dental education. This study evaluated the impact of COVID-19 on dental education and dental students' experience. METHODS: An anonymous online survey was administrated to professional dental students that focused on their experiences related to COVID-19. The survey included questions about student demographics, protocols for school reopening and student perceptions of institutional responses, student concerns, and psychological impacts. RESULTS: Among the 145 respondents, 92.4% were pre-doctoral dental students and 7.6% were orthodontic residents; 48.2% were female and 12.6% students lived alone during the school closure due to the pandemic. Students' age ranged from 23 to 39 years. Younger students expressed more concerns about their emotional health (P = 0.01). In terms of the school's overall response to COVID-19, 73.1% students thought it was effective. The majority (83%) of students believed that social distancing in school can minimize the development of COVID-19. In general, students felt that clinical education suffered after transitioning to online but responded more positively about adjustments to other online curricular components. CONCLUSIONS: The COVID-19 pandemic significantly impacted dental education. Our findings indicate that students are experiencing increased levels of stress and feel their clinical education has suffered. Most students appear comfortable with technology adaptations for didactic curriculum and favor masks, social distancing, and liberal use of sanitizers.

Prediction of 30-Day Hospital Readmissions for All-Cause Dental Conditions using Machine Learning.

INTRODUCTION: It is unknown whether patients admitted for all-cause dental conditions (ACDC) are at high risk for hospital readmission, or what are the risk factors for dental hospital readmission. OBJECTIVE: We examined the prevalence of, and risk factors associated with, 30-day hospital readmission for patients with an all-cause dental admission. We applied artificial intelligence to develop machine learning (ML) algorithms to predict patients at risk of 30-day hospital readmission. METHODS: This study used data extracted from the 2013 Nationwide Readmissions Database (NRD). There were a total of 11,341 cases for all-cause index admission for dental patients admitted to the hospitals. Descriptive statistics were used to analyze patient characteristics. This study applied five techniques to build risk prediction models and to identify risk factors. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), and accuracy, sensitivity, specificity and precision. RESULTS: There were 11% of patients admitted for ACDC readmitted within 30 days of hospital discharge. On average, the total charge per patient was $131,004 for those with 30-day readmission (n=1254) and $69,750 for those without readmission (n=10,087). Factors significantly associated with 30-day hospital readmission included total charges, number of diagnoses, age, number of chronic conditions, length of hospital stays, number of procedures, Medicare insurance and Medicaid insurance, and severity of illness. Model performance from all methods was similar with the artificial neural network showing the highest AUC of 0.739. CONCLUSION: Our results demonstrate that readmission after hospitalization with ACDC is fairly common. If one-third of the 30-day readmission cases can be avoided, there is a potential annual saving of over $25 million among the twenty-one states represented in the NRD. The ML algorithms can predict hospital readmission in dental patients and should be further tested to aid the reduction of hospital readmission and enhancement of patient-centered care.

Artificial intelligence in dentistry: Harnessing big data to predict oral cancer survival.

BACKGROUND: Oral cancer is the sixth most prevalent cancer worldwide. Public knowledge in oral cancer risk factors and survival is limited. AIM: To come up with machine learning (ML) algorithms to predict the length of survival for individuals diagnosed with oral cancer, and to explore the most important factors that were responsible for shortening or lengthening oral cancer survival. METHODS: We used the Surveillance, Epidemiology, and End Results database from the years 1975 to 2016 that consisted of a total of 257880 cases and 94 variables. Four ML techniques in the area of artificial intelligence were applied for model training and validation. Model accuracy was evaluated using mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), R 2 and adjusted R 2. RESULTS: The most important factors predictive of oral cancer survival time were age at diagnosis, primary cancer site, tumor size and year of diagnosis. Year of diagnosis referred to the year when the tumor was first diagnosed, implying that individuals with tumors that were diagnosed in the modern era tend to have longer survival than those diagnosed in the past. The extreme gradient boosting ML algorithms showed the best performance, with the MAE equaled to 13.55, MSE 486.55 and RMSE 22.06. CONCLUSION: Using artificial intelligence, we developed a tool that can be used for oral cancer survival prediction and for medical-decision making. The finding relating to the year of diagnosis represented an important new discovery in the literature. The results of this study have implications for cancer prevention and education for the public.

Health and dental care expenditures in the United States from 1996 to 2016.

INTRODUCTION: As total health and dental care expenditures in the United States continue to rise, healthcare disparities for low to middle-income Americans creates an imperative to analyze existing expenditures. This study examined health and dental care expenditures in the United States from 1996 to 2016 and explored trends in spending across various population subgroups. METHODS: Using data collected by the Medical Expenditure Panel Survey, this study examined health and dental care expenditures in the United States from 1996 to 2016. Trends in spending were displayed graphically and spending across subgroups examined. All expenditures were adjusted for inflation or deflation to the 2016 dollar. RESULTS: Both total health and dental expenditures increased between 1996 and 2016 with total healthcare expenditures increasing from $838.33 billion in 1996 to $1.62 trillion in 2016, a 1.9-fold increase. Despite an overall increase, total expenditures slowed between 2004 and 2012 with the exception of the older adult population. Over the study period, expenditures increased across all groups with the greatest increases seen in older adult health and dental care. The per capita geriatric dental care expenditure increased 59% while the per capita geriatric healthcare expenditure increased 50% across the two decades. For the overall US population, the per capita dental care expenditure increased 27% while the per capita healthcare expenditure increased 60% over the two decades. All groups except the uninsured experienced increased dental care expenditure over the study period. CONCLUSIONS: Healthcare spending is not inherently bad since it brings benefits while exacting costs. Our findings indicate that while there were increases in both health and dental care expenditures from 1996 to 2016, these increases were non-uniform both across population subgroups and time. Further research to understand these trends in detail will be helpful to develop strategies to address health and dental care disparities and to maximize resource utilization.

Exploring the Intersection between Social Determinants of Health and Unmet Dental Care Needs Using Deep Learning.

The goals of this study were to develop a risk prediction model in unmet dental care needs and to explore the intersection between social determinants of health and unmet dental care needs in the United States. Data from the 2016 Medical Expenditure Panel Survey were used for this study. A chi-squared test was used to examine the difference in social determinants of health between those with and without unmet dental needs. Machine learning was used to determine top predictors of unmet dental care needs and to build a risk prediction model to identify those with unmet dental care needs. Age was the most important predictor of unmet dental care needs. Other important predictors included income, family size, educational level, unmet medical needs, and emergency room visit charges. The risk prediction model of unmet dental care needs attained an accuracy of 82.6%, sensitivity of 77.8%, specificity of 87.4%, precision of 82.9%, and area under the curve of 0.918. Social determinants of health have a strong relationship with unmet dental care needs. The application of deep learning in artificial intelligence represents a significant innovation in dentistry and enables a major advancement in our understanding of unmet dental care needs on an individual level that has never been done before. This study presents promising findings and the results are expected to be useful in risk assessment of unmet dental care needs and can guide targeted intervention in the general population of the United States.