PURPOSE: This study examined the use of behavioral design strategies to improve healthier food sales. DESIGN: A quasi-experimental, one-group, repeated measures design examined changes in food sales following behavioral design adjustments. SETTING: United States military base hospital dining facility. SUBJECTS: U.S. military service members, retirees, and civilian employees. INTERVENTION: Behavioral design changes included placement, layout, messaging, default healthy bundling, a stoplight rating system, strategic positioning of healthy items on menu boards, and an increase in healthier snacks. MEASURES: Food sales were assessed by point-of-sales data. ANALYSIS: T-tests examined total sales of each food adjusted weekly between baseline and intervention and intervention and post-intervention. 16 food items targeted by the intervention were examined. Weekly food sales were calculated for the 18-week baseline, 18-week intervention, and 9-week post-intervention. Further, analysis estimated negative binomial models for food item sales. RESULTS: The hospital dining facility served 600 to 900 meals per day. Weekly foods sales decreased during the intervention for desserts, cooked starches, hummus, and yogurt (P < 0.01). Sales increased during the intervention for fruit cups, cooked vegetables, vegetable and turkey burgers, grilled chicken, packaged salads, French fries, hamburgers, and hot dogs (P < 0.02). CONCLUSION: This study demonstrates that a mixture of behavioral design strategies can be operationalized with reasonable fidelity and can lead to increases in the sales of some healthy foods in military worksites.

Immigrants and refugees seeking admission to the United States must first undergo an overseas medical exam, overseen by the US Centers for Disease Control and Prevention (CDC), during which all persons ≥15 years old receive a chest x-ray to look for signs of tuberculosis. Although individual screening sites often implement quality control (QC) programs to ensure radiographs are interpreted correctly, the CDC does not currently have a method for conducting similar QC reviews at scale. We obtained digitized chest radiographs collected as part of the overseas immigration medical exam. Using radiographs from applicants 15 years old and older, we trained deep learning models to perform three tasks: identifying abnormal radiographs; identifying abnormal radiographs suggestive of tuberculosis; and identifying the specific findings (e.g., cavities or infiltrates) in abnormal radiographs. We then evaluated the models on both internal and external testing datasets, focusing on two classes of performance metrics: individual-level metrics, like sensitivity and specificity, and sample-level metrics, like accuracy in predicting the prevalence of abnormal radiographs. A total of 152,012 images (one image per applicant; mean applicant age 39 years) were used for model training. On our internal test dataset, our models performed well both in identifying abnormalities suggestive of TB (area under the curve [AUC] of 0.97; 95% confidence interval [CI]: 0.95, 0.98) and in estimating sample-level counts of the same (-2% absolute percentage error; 95% CIC: -8%, 6%). On the external test datasets, our models performed similarly well in identifying both generic abnormalities (AUCs ranging from 0.89 to 0.92) and those suggestive of TB (AUCs from 0.94 to 0.99). This performance was consistent across metrics, including those based on thresholded class predictions, like sensitivity, specificity, and F1 score. Strong performance relative to high-quality radiological reference standards across a variety of datasets suggests our models may make reliable tools for supporting chest radiography QC activities at CDC.

To describe child, caregiver, and household characteristics associated with fruit and vegetable intakes among US children aged 1-5 years, we examined fruit and vegetable intakes (less than daily vs. daily) using data from the 2021 National Survey of Children's Health among children aged 1-5 years. Multiple logistic regression provided adjusted odds ratios for factors associated with (1) daily fruit and (2) daily vegetable intakes. Among children aged 1-5 years, 68% (n = 11,124) consumed fruit daily, and 51% (n = 8292) consumed vegetables daily. Both daily fruit and daily vegetable intake were associated with child age, child race and ethnicity, and frequency of family meals. For example, children who ate a family meal 4-6 days/week (aOR 0.69; 95% CI 0.57, 0.83) or 0-3 days/week (aOR 0.57; 95% CI 0.46, 0.72) were less likely to consume fruit daily compared to children who had a family meal every day. Participation in food assistance programs, food insufficiency, and household income were not significantly associated with odds of daily fruit or daily vegetable intake in the adjusted models. Several factors were associated with daily fruit and vegetable intake among children aged 1-5. Strategies aimed at increasing fruit and vegetable consumption in early childhood may consider these child, caregiver, and household characteristics. Pediatric healthcare providers, early childhood education centers, and families of young children may be important partners in this work.

Establishing healthy dietary and physical activity patterns among youths is an important public health strategy for improving health and preventing chronic diseases; however, few adolescents meet U.S. government recommendations for dietary or physical activity behaviors, and disparities by sex and race/ethnicity exist. CDC analyzed data from the 2019 Youth Risk Behavior Survey to update estimates of dietary and physical activity behaviors among U.S. high school students overall and by sex and race/ethnicity. In addition, 2-year comparisons (2017 and 2019) and trends in prevalence of these behaviors during 2009-2019 were examined. In 2019, overall, during the 7 days before the survey, 41.8% of students had eaten fruit or drunk 100% fruit juices <1 time/day; 40.7% had eaten vegetables <1 time/day; and 16.7% had not eaten breakfast on all 7 days. Moreover, although 57.4% of students had played on ≥1 sports team during the 12 months before the survey, less than half of students had been physically active for ≥60 minutes/day on all 7 days (23.2%), had exercised to strengthen or tone their muscles on ≥3 days/week (49.5%), had met both aerobic and muscle-strengthening physical activity guidelines (16.5%), or had attended physical education classes on all 5 days in an average school week (25.9%). Trend data indicate limited progress in shifting dietary and physical activity behaviors. That is, with the exception of decreases in the percentage of students who had consumed soda ≥1 time/day (2009: 29.2%; 2019: 15.1%), sports drinks ≥1 time/day (2015: 13.8%; 2019: 10.6%), and <3 glasses/day of plain water (2015: 50.5%; 2019: 44.6%), high school students' dietary and physical activity behaviors have not improved and, in certain cases, have worsened. These findings support the need for multicomponent approaches, including policy and environmental changes, and opportunities for adolescents to learn about and practice making healthy choices.

Objective: To examine associations between comfort eating in response to loneliness or stress and weight change during the COVID-19 pandemic among U.S. adults. Design(s): Quantitative, cross-sectional study. Setting(s): The 2021 SummerStyles survey data. Subjects: U.S. adults (>=18 years; N = 4068). Measures: The outcome was reported weight changes since the start of the COVID-19 pandemic with four responses: lost weight, weight remained the same, gained weight, and don't know. The exposure variable was frequency of comfort eating in response to loneliness or stress during the past year with three responses: never/rarely, sometimes, or often/always. Analysis: We used chi-square analysis to examine the independence of survey variables related to weight changes, and comfort eating in response to loneliness or stress during the COVID-19 pandemic. Next, we used a multinomial logistic regression to estimate adjusted odds ratios for weight changes by comfort eating in response to loneliness or stress frequency. Result(s): Overall, 20.1% of adults reported losing weight, 39.9% remained about the same weight, 30.4% gained weight, and 9.4% did not know about their weight change during the COVID-19 pandemic. Taking comfort by eating in response to loneliness or stress was reported by over 33% of participants (often/always = 8.3%; sometimes = 25.3%). Weight change and comfort eating during the COVID-19 pandemic significantly varied by sociodemographic factors. Respondents that sometimes or often/always reported taking comfort by eating in response to loneliness or stress were more likely to report losing weight (Adjusted Odds Ratio ranges: 1.62-2.99) or gaining weight (Adjusted Odds Ratio ranges: 3.10-4.61) than those who never/rarely took comfort by eating in response to loneliness or stress. Conclusion(s): Taking comfort by eating when stressed/lonely was significantly associated with reported weight changes during the COVID-19 pandemic. Weight changes may lead to additional health complications. Implementing evidence-based strategies to reduce loneliness or stress and support healthy eating during the COVID-19 pandemic may benefit weight management and future well-being. Copyright © 2023

Objective: When novel diseases such as COVID-19 emerge, predictors of clinical outcomes might be unknown. Using data from electronic medical records (EMR) allows evaluation of potential predictors without selecting specific features a priori for a model. We evaluated different machine learning models for predicting outcomes among COVID-19 inpatients using raw EMR data. Material(s) and Method(s): In Premier Healthcare Data Special Release: COVID-19 Edition (PHD-SR COVID-19, release date March, 24 2021), we included patients admitted with COVID-19 during February 2020 through April 2021 and built time-ordered medical histories. Setting the prediction horizon at 24 hours into the first COVID-19 inpatient visit, we aimed to predict intensive care unit (ICU) admission, hyperinflammatory syndrome (HS), and death. We evaluated the following models: L2-penalized logistic regression, random forest, gradient boosting classifier, deep averaging network, and recurrent neural network with a long short-term memory cell. Result(s): There were 57,355 COVID-19 patients identified in PHD-SR COVID-19. ICU admission was the easiest outcome to predict (best AUC=79%), and HS was the hardest to predict (best AUC=70%). Models performed similarly within each outcome. Discussion(s): Although the models learned to attend to meaningful clinical information, they performed similarly, suggesting performance limitations are inherent to the data. Conclusion(s): Predictive models using raw EMR data are promising because they can use many observations and encompass a large feature space; however, traditional and deep learning models may perform similarly when few features are available at the individual patient level. Copyright The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.

BACKGROUND: Sugar-sweetened beverage (SSB) intake is associated with adverse health outcomes. OBJECTIVE: We examined associations between the knowledge of health risks related to SSB and SSB intake among adolescents. DESIGN: A cross-sectional study using 2021 YouthStyles survey data. PARTICIPANTS/SETTINGS: 831 US adolescents (12-17 years old). MAIN OUTCOME MEASURES: The outcome variable was SSB intake (none, 1-6 times/week, and ≥1 time/day). Exposure variables were knowledge of seven SSB-related health risks. STATISTICAL ANALYSES PERFORMED: Seven multinomial regressions were used to estimate adjusted odds ratios (AOR) for drinking SSB, according to knowledge of SSB-related health risks and after controlling for sociodemographics. RESULTS: Overall, 29% of adolescents consumed SSB ≥1 time/day. Although most adolescents identified cavities (75.4%), weight gain (74.6%), and diabetes (69.7%) as being related to drinking SSB, fewer adolescents identified related conditions such as high blood pressure (31.7%), high cholesterol (25.8%), heart disease (24.6%), and some cancers (18.0%). Compared to non-SSB consumers, drinking SSB ≥1 time/day was significantly higher among adolescents who lacked knowledge of associations between SSB intake and weight gain (AOR = 2.0), heart disease (AOR = 1.9), or some cancers (AOR = 2.3) after controlling for covariates. CONCLUSIONS: Among US adolescents, knowledge of SSB-related health risks varied by condition, ranging from 18% (some cancers) to 75% (cavities and weight gain). There were increased odds of drinking SSB among those unaware that weight gain, heart disease, and some cancers are associated with SSB intake. Intervention could evaluate whether increasing certain types of knowledge may influence youth SSB intake.

BACKGROUND: The COVID-19 pandemic impacted some dietary habits of Americans. OBJECTIVE: We examined characteristics associated with a high intake of sweet foods and sugar-sweetened beverages (SSB) during the COVID-19 pandemic among US adults. DESIGN: This was a cross-sectional study. PARTICIPANTS/SETTINGS: The SummerStyles survey data were collected in 2021 among 4034 US adults (≥18 years). MAIN OUTCOME MEASURES: The frequencies were measured of consuming various sweet foods (chocolate/candy, doughnuts/sweet rolls/Danish/muffins/Pop-Tarts, cookies/cake/pie/brownies, and ice cream/frozen desserts) and SSB (regular sodas, sweetened coffee/tea drinks fruit drinks, sports drinks, and energy drinks) during the COVID-19 pandemic. The responses were categorized into 0, >0 to <1, 1 to <2, and ≥2 times/day. The descriptive variables were sociodemographics, food insecurity, weight status, metropolitan status, census regions, and eating habit changes during the COVID-19 pandemic. STATISTICAL ANALYSES PERFORMED: Multinomial regressions were used to estimate adjusted odds ratios (AOR) for being a high consumer of sweet foods and SSB after controlling for characteristics. RESULTS: During 2021, 15% of adults reported consuming sweet foods ≥2 times/day, and 30% reported drinking SSB ≥2 times/day. The factors that were significantly associated with greater odds of high sweet food intake (≥2 times/day) were lower household income (AOR = 1.53 for <$35,000 vs. ≥$100,000), often/sometimes experiencing food insecurity (AOR = 1.41 vs. never), and eating more sweet foods than usual since start of the pandemic (AOR = 2.47 vs. same as usual). The factors that were significantly associated with greater odds of high SSB intake (≥2 times/day) were males (AOR = 1.51), lower education (AOR = 1.98 for ≤high school; AOR = 1.33 for some college vs. college graduate), currently having children (AOR = 1.65), living in nonmetropolitan areas (AOR = 1.34), and drinking more SSB than usual since the pandemic began (AOR = 2.23 vs. same as usual). Younger age, Black race, and reductions in consumption during COVID-19 were related to lower sweet food and SSB intakes. CONCLUSIONS: Our findings, which identified high consumers of sweet foods or SSB, can be used to inform efforts to reduce consumers' added sugars intake during pandemic recovery and support their health.

Frequent intake of sugar-sweetened beverages (SSBs) is associated with adverse health outcomes such as obesity, type 2 diabetes, and cardiovascular disease. Little is known about when, where, and why U.S. adults consume SSBs. This study, using data from an online survey distributed in 2021, examined the occasions, locations, and reasons for consuming SSBs and the characteristics of the adults who consume them. Nearly 7 of 10 adults reported consuming a SSB (1-6 times) in the past 7 days, and more than a third (38%) reported doing so once or more per day (on average). For comparative purposes, the sample was limited to adults who reported consuming SSBs within the last 7 days. Mealtimes were reported as the most frequent occasion for the intake of SSBs (43%) and SSBs were most often consumed at home (70%). Over half of respondents (56%) reported they consume SSBs because they enjoy the taste. Younger adults (18-34 years old) were more likely to consume SSBs in social settings than older adults (≥50 years old). Hispanic adults were less likely to consume SSBs at the beginning of the day compared to non-Hispanic White adults. Younger (18-34 years old) and middle-aged (35-49 years old) adults were more likely to consume SSBs in restaurants, at work, and in cars than older adults (≥50 years old). Women were less likely to consume SSBs at work than men. Hispanic adults were less likely to consume SSBs in cars than non-Hispanic White adults, while those earning USD 50,000-<USD 100,000 were more likely to consume SSBs in cars than those earning ≥USD 100,000. Younger and middle-aged adults were more likely to consume SSBs due to cravings and enjoyment of the carbonation compared to older adults. These findings provide insights on specific populations for whom to tailor messaging and adapt interventions to help reduce SSB intake.

PURPOSE: The high intake of added sugars from foods or beverages increases the risk of obesity, hypertension, dyslipidemia, and cardiovascular disease. Because state-level data are lacking, we estimated dietary intake of added sugars by state and factors associated with intake among US adults. DESIGN: Nationally representative, cross-sectional, in-person, household survey. SETTING: 50 states and DC. SAMPLE: 52,279 US adults from pooled data from 2010 and 2015 National Health Interview Surveys. MEASURES: Estimated total added sugars intake (tsp/day) using the National Cancer Institute's scoring algorithm that converts responses from the Dietary Survey Questionnaire screener to estimated total added sugars intake (tsp/day). ANALYSIS: Mean dietary-added sugars intake estimates and standard error were calculated for adults' characteristics and by state for all 50 states and the District of Columbia. Differences by adult's characteristics were assessed by pairwise t-tests (p &lt; 0.05). All analyses accounted for complex survey design and sampling weights. RESULTS: Overall, US adults consumed 17.0 tsp of added sugars/day (range: 14.8 tsp/day in Alaska to 1.2 tsp/day in Kentucky). Added sugars intake varied by states and sociodemographic characteristics. CONCLUSION: Findings may inform efforts to reduce added sugars intake to lower the high burden of chronic disease.

The 2020-2025 Dietary Guidelines for Americans (DGA) recommends less than 10% of total daily calories come from added sugars. However, many adults overconsume added sugars putting them at risk for poor health outcomes. We examined characteristics of high added sugars consumers among US adults (≥20 years) and described their top 10 sources of added sugars intake using National Health and Nutrition Examination Survey 2015-2018 data (n = 9647). We defined high consumers as consuming &gt;15% of daily calories from added sugars (1.5 times higher than the DGA). We used the National Cancer Institute method to estimate usual intake of energy and percent of calories from added sugars. Top 10 sources were identified based on their percentage contribution to total added sugars intake on a given day. T-tests were used to examine differences by age, sex, race/ethnicity, education, income, marital status, and weight status. Overall, mean usual total energy intake and added sugars intake was 2068 kcal/day and 264 kcal/day, respectively, and 30% of adults were classified as high consumers. The prevalence of high added sugars consumers was significantly higher among 20-30-year-olds (29%), 31-50-year-olds (33%), and 51-70-year-olds (29%) than those aged ≥70 years (22%); non-Hispanic Black (39%) and non-Hispanic White (31%) adults than Hispanics (26%); adults with &lt;high school (37%), high school/GED (38%), or some college (34%) than adults with college or higher (15%); adults living in lower-income households (39% for federal poverty income ratio &lt; 130% and 35% for 130%-&lt;350%) than high-income households (21%). The prevalence of high consumers did not differ by sex or weight status. Top sources of added sugars were sweetened beverages (42%), tea (12%), sweet bakery products (11%), and jams/syrups/sugars (7%). Our findings can inform intervention efforts to decrease added sugars intake to support health.

BACKGROUND: High consumption of added sugars is related to adverse health consequences. OBJECTIVE: The objective of this study was to examine characteristics of US youth who report high intakes of added sugars, as well as the eating occasions and top sources of added sugars that contributed to intakes among consumers with high added sugars intake. Design and participants/setting: We conducted a cross-sectional study using 2015-2018 NHANES data among 5280 US youths (2-19 years). MAIN OUTCOME MEASURES: Outcome measure was usual percent of calories from added sugars using 2 days of dietary recall based on the National Cancer Institute method. High consumers were defined as consuming greater than 15% of total daily calorie intake from added sugars (1.5 times higher than the 2020-2025 Dietary Guidelines for Americans recommendation of &lt;10% of total daily calorie intake). Explanatory measures were selected sociodemographics (e.g., age, sex, race/ethnicity). Eating occasions were breakfast, lunch, dinner, and snack. STATISTICAL ANALYSES PERFORMED: We used t-tests to compare mean differences between sociodemographic groups. RESULTS: Overall, 34% of US youths were classified as high consumers of added sugars. The prevalence of high consumers of added sugars significantly varied by some sociodemographics (i.e., age, race/ethnicity, and head of household's education level). The prevalence of high added sugars consumers was significantly greater among 12-19-year-olds (41%) and 6-11-year-olds (37%) compared to 2-5-year-olds (19%), non-Hispanic Black (42%) and non-Hispanic White (42%) persons compared to Hispanic persons (19%), and those with a head of household's education level of high school/some college (40%) compared to households with college degree or higher (29%). The prevalence of high consumers did not differ by sex, income, or weight status. Of eating occasions, the amount of added sugars youths consumed was highest during snack occasions among high consumers. Top five sources of added sugars among high consumers on a given day were sweetened beverages, sweet bakery products, candy, other desserts, and ready-to-eat cereals. CONCLUSION: One in three US youths consumed more than 15% of total calories from added sugars. High added sugars intake was more prevalent among certain subgroups such as 12-19-year-olds and non-Hispanic Black or non-Hispanic White youth. Our findings can provide information for intervention efforts to decrease added sugars intake to promote child health.

This case study describes how we paired free SARS-CoV-2 rapid antigen testing with on-site case investigation and contact tracing at a drive-through site in an underresourced area of Salt Lake City. Residents of this area had lower rates of employment and health insurance and higher rates of poverty than in the Utah general population. People were given an option to remain on-site and wait until their test results were ready. If a vehicle occupant received a positive test result, the case investigation occurred on-site; contact tracing with the other vehicle occupants was also initiated. People were provided resources to support isolation and quarantine. Bilingual staff who spoke Spanish were incorporated into the workflow. From December 2020 through April 2021, public health staff administered 39 587 rapid tests; 4094 people received a positive test result and 1133 stayed for on-site case investigation. More than half (60.5%) of people with a positive test result who agreed to stay for on-site case investigation were Hispanic or self-reported belonging to a non-Hispanic racial minority group (American Indian/Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, or other racial identities). Pairing rapid antigen testing with on-site case investigation and contact tracing is feasible and improved the timeliness of case investigation by ≥1 day. On-site vaccination services were later integrated. Future emergency responses might consider assisting underresourced communities with on-site services that provide convenient and accessible public health interventions. By providing dependable and reliable services, we were able to achieve buy-in and become a consistent resource for those in the community.

Background: The COVID-19 pandemic has triggered stress, anxiety, and disruption to many individuals’ daily lives, which might impact eating habits. Objective: To examine changes in eating habits related to less healthy foods and beverages during the early phase of the COVID-19 pandemic among US adults. Design: Cross-sectional study. Participants/setting: Authors used SummerStyles data gathered in June 2020 among 3916 US adults (≥18 years). Main outcome measures: The outcome of interest was the reported frequency of consuming more (1) unhealthy snacks and desserts including chips, cookies, and ice cream and (2) sugar-sweetened beverages (SSBs) like regular soda, fruit drinks, sports/energy drinks, sweetened coffee/teas during the COVID-19 pandemic. Responses were categorized as Never/Rarely, Sometimes, or Often/Always. Explanatory variables were sociodemographics, weight status, and census regions. Statistical analyses performed: We used multinomial regressions to calculate adjusted odds ratios (AOR) for Sometimes or Often/Always consuming more unhealthy snacks/desserts (vs. Never/Rarely); and Sometimes or Often/Always more SSBs (vs. Never/Rarely). Results: Overall, 36% of adults reported sometimes consuming more unhealthy snacks/desserts; 16% did so often/always. Twenty-two percent of adults reported sometimes drinking more SSBs; 10% did so often/always. Factors significantly associated with higher odds of reporting often/always consuming more unhealthy snacks/desserts were younger adults (AOR range = 1.51–2.86 vs. adults ≥65 years), females (AOR = 1.58 vs. males), non-Hispanic Black (AOR = 1.89 vs. non-Hispanic White), lower household income (AOR = 2.01 for <USD 35,000 vs. ≥USD 100,000), and obesity (AOR = 1.56 vs. underweight/healthy weight). Factors significantly associated with odds of Often/Always drinking more SSBs were being younger (AOR range = 2.26–4.39 vs. adults ≥65 years), non-Hispanic Black (AOR = 3.25 vs. non-Hispanic White), Hispanic (AOR = 1.75 vs. non-Hispanic White), non-Hispanic Other race/ethnicity (AOR = 2.41 vs. non-Hispanic White), lower education (AOR = 2.03 for ≤high school; AOR = 1.80 for some college vs. college graduate), lower household income (AOR range = 1.64–3.15 vs. ≥USD 100,000), and obesity (AOR = 1.61 vs. underweight/healthy weight). Conclusions: Consuming more sugary foods and SSBs during the first phase of the pandemic was higher in younger adults, lower-income adults, people of racial/ethnic minority groups, and adults with obesity. Dietary shifts to less healthy foods and drinks may influence metabolic health if sustained long-term. Implementing strategies to support individual’s healthy eating habits during the ongoing pandemic and the pandemic recovery may benefit health and wellness. © 2022 by the authors. Licensee MDPI, Basel, Switzerland.

The 2020-2025 Dietary Guidelines for Americans* advise incorporating more fruits and vegetables into U.S. residents' diets as part of healthy dietary patterns. Adults should consume 1.5-2 cup-equivalents of fruits and 2-3 cup-equivalents of vegetables daily.() A healthy diet supports healthy immune function (1) and helps to prevent obesity, type 2 diabetes, cardiovascular diseases, and some cancers (2); having some of these conditions can predispose persons to more severe illness and death from COVID-19 (3). CDC used the most recent 2019 Behavioral Risk Factor Surveillance system (BRFSS) data to estimate the percentage of states' adult population who met intake recommendations overall and by sociodemographic characteristics for 49 states and the District of Columbia (DC). Overall, 12.3% of adults met fruit recommendations, ranging from 8.4% in West Virginia to 16.1% in Connecticut, and 10.0% met vegetable recommendations, ranging from 5.6% in Kentucky to 16.0% in Vermont. The prevalence of meeting fruit intake recommendations was highest among Hispanic adults (16.4%) and lowest among males (10.1%); meeting vegetable intake recommendations was highest among adults aged 51 years (12.5%) and lowest among those living below or close to the poverty level (income to poverty ratio [IPR] <1.25) (6.8%). Additional policies() and programs that will increase access to fruits and vegetables in places where U.S. residents live, learn, work, and play, might increase consumption and improve health.

PURPOSE: We examined associations between academic grades and positive health behaviors, individually and collectively, among U.S. high school students. DESIGN: Cross-sectional study design. SETTING: Data were from the 2017 national Youth Risk Behavior Survey. Response rates were 75% for schools, 81% for students, and 60% overall (n = 14,765 students). SUBJECTS: Youth in grades 9th-12th. MEASURES: We focused on youth behaviors that can prevent or delay the onset of chronic health conditions. Seven dietary, 3 physical activity, 2 sedentary screen time, and 4 tobacco product use behaviors were assessed. Variables were dichotomized (0/1) to indicate that a score was given to the positive health behavior response (e.g.,, did not smoke cigarettes = 1). A composite score was created by summing each positive health behavior response among 16 total health behaviors. ANALYSIS: Multivariable logistic regression analyses for each individual health behavior, and a multivariable negative binomial regression for the composite score, were conducted with self-reported academic grades, controlling for sex, grade in school, race/ethnicity, and body mass index (BMI) categories. RESULTS: Controlling for covariates, students who reported mostly A's had 2.0 (P < .001) more positive health behaviors; students who reported mostly B's had 1.3 (P < .001) more positive health behaviors; and students who reported mostly C's had .78 (P < .001) more positive health behaviors, compared to students who reported mostly D's/F's. CONCLUSIONS: Higher academic grades are associated with more positive individual and cumulative health behaviors among high school students. Understanding these relationships can help inform efforts to create a healthy and supportive school environment and strive for health equity.

BACKGROUND: Optimized symptom-based COVID-19 case definitions that guide public health surveillance and individual patient management in the community may assist pandemic control. METHODS: We assessed diagnostic performance of existing cases definitions (e.g. influenza-like illness, COVID-like illness) using symptoms reported from 185 household contacts to a PCR-confirmed case of COVID-19 in Wisconsin and Utah, United States. We stratified analyses between adults and children. We also constructed novel case definitions for comparison. RESULTS: Existing COVID-19 case definitions generally showed high sensitivity (86-96%) but low positive predictive value (PPV) (36-49%; F-1 score 52-63) in this community cohort. Top performing novel symptom combinations included taste or smell dysfunction and improved the balance of sensitivity and PPV (F-1 score 78-80). Performance indicators were generally lower for children (< 18 years of age). CONCLUSIONS: Existing COVID-19 case definitions appropriately screened in household contacts with COVID-19. Novel symptom combinations incorporating taste or smell dysfunction as a primary component improved accuracy. Case definitions tailored for children versus adults should be further explored.

PURPOSE: During a pandemic, persons might experience worry because of threats (real or perceived), or as part of stress-related reactions. We aimed to provide insight into Americans' worry about food during COVID-19. Design, Subjects, Measures: Online survey data from June 2020 (n = 4,053 U.S. adults; 62.7% response rate) was used to assess 2 outcomes: worry about food availability (FA); food safety (FS). Adults with missing information about FA and FS were excluded from analysis (final n = 3,652). ANALYSIS: We used descriptive statistics and multivariable logistic regression to examine characteristics associated with the outcomes and estimate adjusted odds ratios (aOR) and 95% confidence intervals (95%CI) for associations between sociodemographic variables and outcomes. RESULTS: 58.3% of respondents reported worry about FA; 57.5% about FS, with higher odds of worry for FA and FS (versus referents) in lower income households (FA: aOR = 1.76 95%CI [1.30, 2.39]; FS: 1.84[1.35, 2.51]); unemployed (1.54[1.05, 2.28]; 1.90[1.26,2.81]); non-Hispanic Black (1.55[1.14,2.12]); 2.25[1.65,3.07]); Hispanic (1.39[1.06,1.82]; 1.94[1.46,2.56]). CONCLUSION: Findings highlight the importance of strategies to reduce consumer worry about FA and FS and negative food behaviors, and the need for continued access to hunger safety net programs, which could have positive effects on nutrition security.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in children is often asymptomatic or results in only mild disease.1 Data on the extent of transmission of SARS-CoV-2 from children and adolescents in the household setting, including transmission to older persons who are at increased risk for severe disease, are limited.2 After an outbreak of coronavirus disease 2019 (Covid-19) at an overnight camp,3 we conducted a retrospective cohort study involving camp attendees and their household contacts to assess secondary transmission and factors associated with household transmission (additional details are provided in the Methods section in the Supplementary Appendix, available with the full text of this letter at NEJM.org).

Frequent intake of sugar-sweetened beverages (SSBs) is associated with adverse health outcomes, including obesity, type 2 diabetes, and cardiovascular disease. We used combined data from the 2010 and 2015 National Health Interview Survey to examine the prevalence of SSB intake among US adults in all 50 states and the District of Columbia. Approximately two-thirds of adults reported consuming SSBs at least daily, including more than 7 in 10 adults in Hawaii, Arkansas, Wyoming, South Dakota, Connecticut, and South Carolina, with significant differences in sociodemographic characteristics. Efforts to decrease SSB consumption could consider the sociodemographic and geographic differences in SSB intake when designing equitable interventions.

Background: Obesity remains a significant public health issue in the United States. Each week, millions of infants and children are cared for in early care and education (ECE) programs, making it an important setting for building healthy habits. Since 2010, 39 states promulgated licensing regulations impacting infant feeding, nutrition, physical activity, or screen time practices. We assessed trends in ECE regulations across all 50 states and the District of Columbia (D.C.) and hypothesized that states included more obesity prevention standards over time. Methods: We analyzed published ratings of state licensing regulations (2010-2018) and describe trends in uptake of 47 high-impact standards derived from Caring for Our Children's, Preventing Childhood Obesity special collection. National trends are described by (1) care type (Centers, Large Care Homes, and Small Care Homes); (2) state and U.S. region; and (3) most and least supported standards. Results: Center regulations included the most obesity prevention standards (∼13% in 2010 vs. ∼29% in 2018) compared with other care types, and infant feeding and nutrition standards were most often included, while physical activity and screen time were least supported. Some states saw significant improvements in uptake, with six states and D.C. having a 30%-point increase 2010-2018. Conclusions: Nationally, there were consistent increases in the percentage of obesity prevention standards included in ECE licensing regulations. Future studies may examine facilitators and barriers to the uptake of obesity prevention standards and identify pathways by which public health and health care professionals can act as a resource and promote obesity prevention in ECE.

IMPORTANCE: Criterion-standard specimens for tuberculosis diagnosis in young children, gastric aspirate (GA) and induced sputum, are invasive and rarely collected in resource-limited settings. A far less invasive approach to tuberculosis diagnostic testing in children younger than 5 years as sensitive as current reference standards is important to identify. OBJECTIVE: To characterize the sensitivity of preferably minimally invasive specimen and assay combinations relative to maximum observed yield from all specimens and assays combined. DESIGN, SETTING, AND PARTICIPANTS: In this prospective cross-sectional diagnostic study, the reference standard was a panel of up to 2 samples of each of 6 specimen types tested for Mycobacterium tuberculosis complex by Xpert MTB/RIF assay and mycobacteria growth indicator tube culture. Multiple different combinations of specimens and tests were evaluated as index tests. A consecutive series of children was recruited from inpatient and outpatient settings in Kisumu County, Kenya, between October 2013 and August 2015. Participants were children younger than 5 years who had symptoms of tuberculosis (unexplained cough, fever, malnutrition) and parenchymal abnormality on chest radiography or who had cervical lymphadenopathy. Children with 1 or more evaluable specimen for 4 or more primary study specimen types were included in the analysis. Data were analyzed from February 2015 to October 2020. MAIN OUTCOMES AND MEASURES: Cumulative and incremental diagnostic yield of combinations of specimen types and tests relative to the maximum observed yield. RESULTS: Of the 300 enrolled children, the median (interquartile range) age was 2.0 (1.0-3.6) years, and 151 (50.3%) were female. A total of 294 met criteria for analysis. Of 31 participants with confirmed tuberculosis (maximum observed yield), 24 (sensitivity, 77%; interdecile range, 68%-87%) had positive results on up to 2 GA samples and 20 (sensitivity, 64%; interdecile range, 53%-76%) had positive test results on up to 2 induced sputum samples. The yields of 2 nasopharyngeal aspirate (NPA) samples (23 of 31 [sensitivity, 74%; interdecile range, 64%-84%]), of 1 NPA sample and 1 stool sample (22 of 31 [sensitivity, 71%; interdecile range, 60%-81%]), or of 1 NPA sample and 1 urine sample (21.5 of 31 [sensitivity, 69%; interdecile range, 58%-80%]) were similar to reference-standard specimens. Combining up to 2 each of GA and NPA samples had an average yield of 90% (28 of 31). CONCLUSIONS AND RELEVANCE: NPA, in duplicate or in combination with stool or urine specimens, was readily obtainable and had diagnostic yield comparable with reference-standard specimens. This combination could improve tuberculosis diagnosis among children in resource-limited settings. Combining GA and NPA had greater yield than that of the current reference standards and may be useful in certain clinical and research settings.

According to the 2020-2025 Dietary Guidelines for Americans, persons should consume fruits and vegetables as part of a healthy eating pattern to reduce their risk for diet-related chronic diseases, such as cardiovascular disease, type 2 diabetes, some cancers, and obesity.* A healthy diet is important for healthy growth in adolescence, especially because adolescent health behaviors might continue into adulthood (1). The U.S. Department of Agriculture (USDA) recommends minimum daily intake of 1.5 cups of fruit and 2.5 cups of vegetables for females aged 14-18 years and 2 cups of fruit and 3 cups of vegetables for males aged 14-18 years.(†) Despite the benefits of fruit and vegetable consumption, few adolescents consume these recommended amounts (2-4). In 2013, only 8.5% of high school students met the recommendation for fruit consumption, and only 2.1% met the recommendation for vegetable consumption (2). To update the 2013 data, CDC analyzed data from the 2017 national and state Youth Risk Behavior Surveys (YRBSs) to describe the percentage of students who met intake recommendations, overall and by sex, school grade, and race/ethnicity. The median frequencies of fruit and vegetable consumption nationally were 0.9 and 1.1 times per day, respectively. Nationally, 7.1% of students met USDA intake recommendations for fruits (95% confidence interval [CI] = 4.0-10.3) and 2.0% for vegetables (upper 95% confidence limit = 7.9) using previously established scoring algorithms. State-specific estimates of the percentage of students meeting fruit intake recommendations ranged from 4.0% (Connecticut) to 9.3% (Louisiana), and the percentage meeting vegetable intake recommendations ranged from 0.6% (Kansas) to 3.7% (New Mexico). Additional efforts to expand the reach of existing school and community programs or to identify new effective strategies, such as social media approaches, might help address barriers and improve adolescent fruit and vegetable consumption.

PURPOSE: Examine association of health literacy (HL) and menu-labeling (ML) usage with sugar-sweetened beverage (SSB) intake among adults in Mississippi. DESIGN: Quantitative, cross-sectional study. SETTING: 2016 Mississippi Behavioral Risk Factor Surveillance System data. PARTICIPANTS: Adults living in Mississippi (n = 4549). MEASURES: Outcome variable was SSB intake (regular soda, fruit drinks, sweet tea, and sports/energy drinks). Exposure variables were 3 HL questions (find information, understand oral information, and understand written information) and ML usage among adults who eat at fast-food/chain restaurants (user, nonuser, and do not notice ML). ANALYSIS: Multinomial logistic regressions were used to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CI) for SSB intake >/=1 time/d (reference: 0 times/d) associated with HL and ML. RESULTS: In Mississippi, 46.8% of adults consumed SSB >/=1 time/d, and 26.9% consumed >/=2 times/d. The odds of consuming SSBs >/=1 time/d were higher among adults with lower HL (aOR = 1.7; 95% CI = 1.3-2.2) than those with higher HL. Among adults who ate at fast-food/chain restaurants, the odds of consuming SSBs >/=1 time/d were higher among nonusers of ML (aOR = 2.3; 95% CI = 1.7-3.1) and adults who did not notice ML (aOR = 1.8; 95% CI = 1.3-2.6) than ML users. CONCLUSION: Adults with lower HL and adults who do not use or notice ML consumed more SSBs in Mississippi. Understanding why lower HL and no ML usage are linked to SSB intake could guide the design of interventions to reduce SSB intake in this population.

OBJECTIVE: The Centers for Disease Control and Prevention (CDC) coordinates a labor-intensive process to measure the prevalence of autism spectrum disorder (ASD) among children in the United States. Random forests methods have shown promise in speeding up this process, but they lag behind human classification accuracy by about 5%. We explore whether more recently available document classification algorithms can close this gap. MATERIALS AND METHODS: Using data gathered from a single surveillance site, we applied 8 supervised learning algorithms to predict whether children meet the case definition for ASD based solely on the words in their evaluations. We compared the algorithms' performance across 10 random train-test splits of the data, using classification accuracy, F1 score, and number of positive calls to evaluate their potential use for surveillance. RESULTS: Across the 10 train-test cycles, the random forest and support vector machine with Naive Bayes features (NB-SVM) each achieved slightly more than 87% mean accuracy. The NB-SVM produced significantly more false negatives than false positives (P = 0.027), but the random forest did not, making its prevalence estimates very close to the true prevalence in the data. The best-performing neural network performed similarly to the random forest on both measures. DISCUSSION: The random forest performed as well as more recently available models like the NB-SVM and the neural network, and it also produced good prevalence estimates. NB-SVM may not be a good candidate for use in a fully-automated surveillance workflow due to increased false negatives. More sophisticated algorithms, like hierarchical convolutional neural networks, may not be feasible to train due to characteristics of the data. Current algorithms might perform better if the data are abstracted and processed differently and if they take into account information about the children in addition to their evaluations. CONCLUSION: Deep learning models performed similarly to traditional machine learning methods at predicting the clinician-assigned case status for CDC's autism surveillance system. While deep learning methods had limited benefit in this task, they may have applications in other surveillance systems.

Syndromic surveillance detects and monitors individual and population health indicators through sources such as emergency department records. Automated classification of these records can improve outbreak detection speed and diagnosis accuracy. Current syndromic systems rely on hand-coded keyword-based methods to parse written fields and may benefit from the use of modern supervised-learning classifier models. In this paper, we implement two recurrent neural network models based on long short-term memory (LSTM) and gated recurrent unit (GRU) cells and compare them to two traditional bag-of-words classifiers: multinomial naive Bayes (MNB) and a support vector machine (SVM). The MNB classifier is one of only two machine learning algorithms currently being used for syndromic surveillance. All four models are trained to predict diagnostic code groups as defined by Clinical Classification Software, first to predict from discharge diagnosis, and then from chief complaint fields. The classifiers are trained on 3.6 million de-identified emergency department records from a single United States jurisdiction. We compare performance of these models primarily using the F1 score, and we measure absolute model performance to determine which conditions are the most amenable to surveillance based on chief complaint alone. Using discharge diagnoses, the LSTM classifier performs best, though all models exhibit an F1 score above 96.00. Using chief complaints, the GRU performs best (F1=47.38), and MNB with bigrams performs worst (F1=39.40). We also note that certain syndrome types are easier to detect than others. For example, chief complaints using the GRU model predicts alcohol-related disorders well (F1=78.91) but predicts influenza poorly (F1=14.80). In all instances, the RNN models outperformed the bag-of-words classifiers suggesting deep learning models could substantially improve the automatic classification of unstructured text for syndromic surveillance.

Drug overdose is the leading cause of unintentional injury-associated death in the United States. Among 70,237 fatal drug overdoses in 2017, prescription opioids were involved in 17,029 (24.2%) (1). Higher rates of opioid-related deaths have been recorded in nonmetropolitan (rural) areas (2). In 2017, 14 rural counties were among the 15 counties with the highest opioid prescribing rates.* Higher opioid prescribing rates put patients at risk for addiction and overdose (3). Using deidentified data from the Athenahealth electronic health record (EHR) system, opioid prescribing rates among 31,422 primary care providers(dagger) in the United States were analyzed to evaluate trends from January 2014 to March 2017. This analysis assessed how prescribing practices varied among six urban-rural classification categories of counties, before and after the March 2016 release of CDC's Guideline for Prescribing Opioids for Chronic Pain (Guideline) (4). Patients in noncore (the most rural) counties had an 87% higher chance of receiving an opioid prescription compared with persons in large central metropolitan counties during the study period. Across all six county groups, the odds of receiving an opioid prescription decreased significantly after March 2016. This decrease followed a flat trend during the preceding period in micropolitan and large central metropolitan county groups; in contrast, the decrease continued previous downward trends in the other four county groups. Data from EHRs can effectively supplement traditional surveillance methods for monitoring trends in opioid prescribing and other areas of public health importance, with minimal lag time under ideal conditions. As less densely populated areas appear to indicate both substantial progress in decreasing opioid prescribing and ongoing need for reduction, community health care practices and intervention programs must continue to be tailored to community characteristics.

One broad goal of biomedical informatics is to generate fully-synthetic, faithfully representative electronic health records (EHRs) to facilitate data sharing between healthcare providers and researchers and promote methodological research. A variety of methods existing for generating synthetic EHRs, but they are not capable of generating unstructured text, like emergency department (ED) chief complaints, history of present illness, or progress notes. Here, we use the encoder-decoder model, a deep learning algorithm that features in many contemporary machine translation systems, to generate synthetic chief complaints from discrete variables in EHRs, like age group, gender, and discharge diagnosis. After being trained end-to-end on authentic records, the model can generate realistic chief complaint text that appears to preserve the epidemiological information encoded in the original record-sentence pairs. As a side effect of the model's optimization goal, these synthetic chief complaints are also free of relatively uncommon abbreviation and misspellings, and they include none of the personally identifiable information (PII) that was in the training data, suggesting that this model may be used to support the de-identification of text in EHRs. When combined with algorithms like generative adversarial networks (GANs), our model could be used to generate fully-synthetic EHRs, allowing healthcare providers to share faithful representations of multimodal medical data without compromising patient privacy. This is an important advance that we hope will facilitate the development of machine-learning methods for clinical decision support, disease surveillance, and other data-hungry applications in biomedical informatics.

RATIONALE: The interferon gamma release assays and tuberculin skin tests are used to support the diagnosis of both latent and active tuberculosis. However, we previously demonstrated that a negative tuberculin test in active tuberculosis is associated with disseminated disease and death. It is unknown whether the same associations exist for interferon gamma release assays. OBJECTIVES: To determine the association between these tests and site of tuberculosis and death among persons with active tuberculosis. METHODS: We analyzed interferon gamma release assay and tuberculin test results for all persons with culture-confirmed tuberculosis reported to the US National Tuberculosis Surveillance System from 2010-2014. We used logistic regression to calculate the association between these tests and site of disease and death. MEASUREMENTS AND MAIN RESULTS: 24,803 persons with culture-confirmed tuberculosis had either of these test results available for analysis. Persons with a positive tuberculin test had lower odds of disseminated disease (i.e., miliary or combined pulmonary and extrapulmonary disease), but there was no difference in the odds of disseminated disease with a positive interferon gamma release assay. However, persons who were positive to either of these tests had lower odds of death. An indeterminate interferon gamma release assay result was associated with greater odds of both disseminated disease and death. CONCLUSIONS: Despite perceived equivalence in clinical practice, interferon gamma release assay and tuberculin test results have different associations with tuberculosis site yet similar associations with the risk of death. Furthermore, an indeterminate interferon gamma release assay result in a person with active tuberculosis is not unimportant, and rather carries greater odds of disseminated disease and death. Prospective study may improve our understanding of the underlying mechanisms by which these tests are associated with disease localization and death.

Little is known about the relationship between body composition and metabolic risk factors in young adults. We hypothesized that low muscle mass (LMM) is associated with metabolic syndrome (MetS) and its components in young adults and that the associations vary by obesity. A cross-sectional analysis was conducted using the Korea National Health and Nutrition Examination Survey data. In total, 5300 young adults aged 19 to 39 years were evaluated. Low muscle mass was defined as an appendicular skeletal muscle mass/weight less than 1 SD below the mean for each participant's corresponding sex and age group. Obesity was defined as a body mass index greater than or equal to 25 kg/m2. The prevalence of LMM was higher in obese than nonobese participants (37.6% vs 9.6%). In the nonobese participants, the prevalence of MetS, high waist circumference, high triglycerides, and high blood pressure was significantly greater in the LMM group than in the high muscle mass group. In the nonobese group, compared with high muscle mass participants, those with LMM had odds ratios for MetS of 3.6 (95% confidence interval, 1.48-8.76; P < .001) and 3.6 (95% confidence interval, 1.48-8.71; P < .001) in men and women, respectively, after adjusting for confounding factors. However, no significant association of LMM with MetS or its components was found in obese participants. In conclusion, our results suggest that young adults with LMM may have a high risk of MetS, especially when they are nonobese. Interventions aimed at increasing muscle mass at younger ages may have the potential to reduce MetS.