Predicting COVID-19 new cases in California with Google Trends data and a machine learning approach.

BACKGROUND: Google Trends data can be a valuable source of information for health-related issues such as predicting infectious disease trends. OBJECTIVES: To evaluate the accuracy of predicting COVID-19 new cases in California using Google Trends data, we develop and use a GMDH-type neural network model and compare its performance with a LTSM model. METHODS: We predicted COVID-19 new cases using Google query data over three periods. Our first period covered March 1, 2020, to July 31, 2020, including the first peak of infection. We also estimated a model from October 1, 2020, to January 7, 2021, including the second wave of COVID-19 and avoiding possible biases from public interest in searching about the new pandemic. In addition, we extended our forecasting period from May 20, 2020, to January 31, 2021, to cover an extended period of time. RESULTS: Our findings show that Google relative search volume (RSV) can be used to accurately predict new COVID-19 cases.  We find that among our Google relative search volume terms, "Fever," "COVID Testing," "Signs of COVID," "COVID Treatment," and "Shortness of Breath" increase model predictive accuracy. CONCLUSIONS: Our findings highlight the value of using data sources providing near real-time data, e.g., Google Trends, to detect trends in COVID-19 cases, in order to supplement and extend existing epidemiological models.

Association of social vulnerability and influenza vaccination rates for Annual Medicare Enrollees at the county-level in the United States.

INTRODUCTION: Influenza is a preventable acute respiratory illness with a high potential to cause serious complications and is associated with high mortality and morbidity in the US. We aimed to determine the specific community-level vulnerabilities for different race/ethnic communities that are most predictive of influenza vaccination rates. METHODS: We conducted a machine learning analysis (XGBoost) to identify community-level social vulnerability features that are predictive of influenza vaccination rates among Medicare enrollees across counties in the US and by race/ethnicity. RESULTS: Population density per square mile in a county is the most important feature in predicting influenza vaccination in a county, followed by unemployment rates and the percentage of mobile homes. The gain relative importance of these features are 11.6%, 9.2%, and 9%, respectively. Among whites, population density (17% gain relative importance) was followed by the percentage of mobile homes (9%) and per capita income (8.7%). For Black/African Americans, the most important features were population density (12.8%), percentage of minorities in the county (8.0%), per capita income (6.9%), and percent of over-occupied housing units (6.8%). Finally, for Hispanics, the top features were per capita income (8.4%), percentage of mobile homes (8.0%), percentage of non-institutionalized persons with a disability (7.9%), and population density (7.6%). CONCLUSIONS: Our study may have implications for the success of large vaccination programs in counties with high social vulnerabilities. Further, our findings suggest that policies and interventions seeking to increase rates of vaccination in race/ethnic minority communities may need to be tailored to address their specific socioeconomic vulnerabilities.

Social vulnerability and initial COVID-19 community spread in the US South: a machine learning approach.

BACKGROUND AND OBJECTIVES: More than 93 million COVID-19 cases and more than 1 million COVID-19 deaths have been reported in the USA by August 2022. The disproportionate effect of the pandemic and its severe impact on vulnerable communities raised concerns. This research aimed to identify and rank Social Vulnerability Index (SVI) factors highly predictive of the spread of COVID-19 in the US South at the beginning of the pandemic. METHODS: We used Extreme Gradient Boosting (XGBoost) machine learning methodology and SVI data, and the number of COVID-19 cases across all counties in the US South to predict the number of positive cases within 30 days of a county's first case. RESULTS: Our results showed that the percentage of mobile homes is the most important feature in predicting the increase in COVID-19. Also, population density per square mile, per capita income, percentage of housing in structures with 10+ units, percentage of people below poverty and percentage of people with no high school diploma are important predictors of COVID-19 community spread, respectively. CONCLUSIONS: SVI can help assess the vulnerability or resilience of communities to the spread of COVID-19 and can help identify communities at high risk of COVID-19 spread.

Social vulnerability predictors of drug poisoning mortality: A machine learning analysis in the United States.

BACKGROUND AND OBJECTIVES: Drug poisoning is a leading cause of unintentional deaths in the United States. Despite the growing literature, there are a few recent analyses of a wide range of community-level social vulnerability features contributing to drug poisoning mortality. Current studies on this topic face three limitations: often studying a limited subset of vulnerability features, focusing on small sample sizes, or solely including local data. To address this gap, we conducted a national-level analysis to study the impacts of several social vulnerability features in predicting drug mortality rates in the United States. METHODS: We used machine learning to investigate the role of 16 social vulnerability features in predicting drug mortality rates for US counties in 2014, 2016, and 2018-the most recent available data. We estimated each vulnerability feature's gain relative contribution in predicting drug poisoning mortality. RESULTS: Among all social vulnerability features, the percentage of noninstitutionalized persons with a disability is the most influential predictor, with a gain relative contribution of 18.6%, followed by population density and the percentage of minority residents (13.3% and 13%, respectively). Percentages of households with no available vehicles, mobile homes, and persons without a high school diploma are the following features with gain relative contributions of 6.3%, 5.8%, and 5.1%, respectively. CONCLUSION AND SCIENTIFIC SIGNIFICANCE: We identified social vulnerability features that are most predictive of drug poisoning mortality. Public health interventions and policies targeting vulnerable communities may increase the resilience of these communities and mitigate the overdose death and drug misuse crisis.

Estimating Excess Deaths by Race/Ethnicity in the State of California During the COVID-19 Pandemic.

INTRODUCTION: To examine excess mortality among minorities in California during the COVID-19 pandemic. METHODS: Using seasonal autoregressive integrated moving average time series, we estimated counterfactual total deaths using historical data (2014-2019) of all-cause mortality by race/ethnicity. Estimates were compared to pandemic mortality trends (January 2020 to January 2021) to predict excess deaths during the pandemic for each race/ethnic group. RESULTS: Our findings show a significant disparity among minority excess deaths, including 7892 (24.6% increase), 4903 (20.4%), 30,186 (47.7%), and 22,027 (12.6%) excess deaths, including deaths identified as COVID-19-related, for Asian, Black, Hispanic, and White non-Hispanic individuals, respectively. Estimated increases in all-cause deaths excluding COVID-19 deaths were 1331, 1436, 3009, and 5194 for Asian, Black, Hispanic, and White non-Hispanic individuals, respectively. However, the rate of excess deaths excluding COVID-19 recorded deaths per 100 k was disproportionately high for Black (66 per 100 k) compared to White non-Hispanic (36 per 100 k). The rates for Asians and Hispanics were 23 and 19 per 100 k. CONCLUSIONS: Our findings emphasize the importance of targeted policies for minority populations to lessen the disproportionate impact of COVID-19 on their communities.

Impact of Florida's prescription drug monitoring program on drug- related fatal vehicle crashes: a difference-in-differences approach.

BACKGROUND: Prescription drug use has soared in the USA within the last two decades. Prescription drugs can impair motor skills essential for the safe operation of a motor vehicle, and therefore can affect traffic safety. As one of the epicentres of the opioid epidemic, Florida has been struck by high opioid misuse and overdose rates, and has concurrently suffered major threats to traffic disruptions safety caused by driving under the influence of drugs. To prevent prescription opioid misuse in Florida, Prescription Drug Monitoring Programs (PDMPs) were implemented in September 2011. OBJECTIVE: To examine the impact of Florida's implementation of a mandatory PDMP on drug-related MVCs occurring on public roads. METHODS: We employed a difference-in-differences approach to estimate the difference in prescription drug-related fatal crashes in Florida associated with its 2011 PDMP implementation relative to those in Georgia, which did not use PDMPs during the same period (2009-2013). The analyses were conducted in 2020. RESULTS: In Florida, there was a significant decline in drug-related vehicle crashes during the 22 months post-PDMP. PDMP implementation was associated with approximately two (-2.21; 95% CI -4.04 to -0.37; p<0.05) fewer prescribed opioid-related fatal crashes every month, indicating 25% reduction in the number of monthly crashes. We conducted sensitivity analyses to investigate the impact of PDMP implementation on central nervous system depressants and stimulants as well as cocaine and marijuana-related fatal crashes but found no robust significant reductions. CONCLUSIONS: The implementation of PDMPs in Florida provided important benefits for traffic safety, reducing the rates of prescription opioid-related vehicle crashes.

The role of good governance in the race for global vaccination during the COVID-19 pandemic.

Governments have developed and implemented various policies and interventions to fight the COVID-19 pandemic. COVID-19 vaccines are now being produced and distributed globally. This study investigated the role of good governance and government effectiveness indicators in the acquisition and administration of COVID-19 vaccines at the population level. Data on six World Bank good governance indicators for 172 countries for 2019 and machine-learning methods (K-Means Method and Principal Component Analysis) were used to cluster countries based on these indicators and COVID-19 vaccination rates. XGBoost was used to classify countries based on their vaccination status and identify the relative contribution of each governance indicator to the vaccination rollout in each country. Countries with the highest COVID-19 vaccination rates (e.g., Israel, United Arab Emirates, United States) also have higher effective governance indicators. Regulatory Quality is the most important indicator in predicting COVID-19 vaccination status in a country, followed by Voice and Accountability, and Government Effectiveness. Our findings suggest that coordinated global efforts led by the World Health Organization and wealthier nations may be necessary to assist in the supply and distribution of vaccines to those countries that have less effective governance.

Spatial Modeling of COVID-19 Vaccine Hesitancy in the United States.

Vaccine hesitancy refers to delay in acceptance or refusal of vaccines despite the availability of vaccine services. Despite the efforts of United States healthcare providers to vaccinate the bulk of its population, vaccine hesitancy is still a severe challenge that has led to the resurgence of COVID-19 cases to over 100,000 people during early August 2021. To our knowledge, there are limited nationwide studies that examined the spatial distribution of vaccination rates, mainly based on the social vulnerability index (SVI). In this study, we compiled a database of the percentage of fully vaccinated people at the county scale across the continental United States as of 29 July 2021, along with SVI data as potential significant covariates. We further employed multiscale geographically weighted regression to model spatial nonstationarity of vaccination rates. Our findings indicated that the model could explain over 79% of the variance of vaccination rate based on Per capita income and Minority (%) (with positive impacts), and Age 17 and younger (%), Mobile homes (%), and Uninsured people (%) (with negative effects). However, the impact of each covariate varied for different counties due to using separate optimal bandwidths. This timely study can serve as a geospatial reference to support public health decision-makers in forming region-specific policies in monitoring vaccination programs from a geographic perspective.

Analysis of Excess Deaths During the COVID-19 Pandemic in the State of Florida.

Objectives. To determine the number of excess deaths (i.e., those exceeding historical trends after accounting for COVID-19 deaths) occurring in Florida during the COVID-19 pandemic.Methods. Using seasonal autoregressive integrated moving average time-series modeling and historical mortality trends in Florida, we forecasted monthly deaths from January to September of 2020 in the absence of the pandemic. We compared estimated deaths with monthly recorded total deaths (i.e., all deaths regardless of cause) during the COVID-19 pandemic and deaths only from COVID-19 to measure excess deaths in Florida.Results. Our results suggest that Florida experienced 19 241 (15.5%) excess deaths above historical trends from March to September 2020, including 14 317 COVID-19 deaths and an additional 4924 all-cause, excluding COVID-19, deaths in that period.Conclusions. Total deaths are significantly higher than historical trends in Florida even when accounting for COVID-19-related deaths. The impact of COVID-19 on mortality is significantly greater than the official COVID-19 data suggest.

Comparison of Use of Health Care Services and Spending for Unauthorized Immigrants vs Authorized Immigrants or US Citizens Using a Machine Learning Model.

Importance: Knowledge about use of health care services (health care utilization) and expenditures among unauthorized immigrant populations is uncertain because of limitations in ascertaining legal status in population data. Objective: To examine health care utilization and expenditures that are attributable to unauthorized and authorized immigrants vs US-born individuals. Design, Setting, and Participants: This cross-sectional study used the data on documentation status from the Los Angeles Family and Neighborhood Survey (LAFANS) to develop a random forest classifier machine learning model. K-fold cross-validation was used to test model performance. The LAFANS is a randomized, multilevel, in-person survey of households residing in Los Angeles County, California, consisting of 2 waves. Wave 1 began in April 2000 and ended in January 2002, and wave 2 began in August 2006 and ended in December 2008. The machine learning model was then applied to a nationally representative database, the 2016-2017 Medical Expenditure Panel Survey (MEPS), to predict health care expenditures and utilization among unauthorized and authorized immigrants and US-born individuals. A generalized linear model analyzed health care expenditures. Logistic regression modeling estimated dichotomous use of emergency department (ED), inpatient, outpatient, and office-based physician visits by immigrant groups with adjusting for confounding factors. Data were analyzed from May 1, 2019, to October 14, 2020. Exposures: Self-reported immigration status (US-born, authorized, and unauthorized status). Main Outcomes and Measures: Annual health care expenditures per capita and use of ED, outpatient, inpatient, and office-based physician care. Results: Of 47 199 MEPS respondents with nonmissing data, 35 079 (74.3%) were US born, 10 816 (22.9%) were authorized immigrants, and 1304 (2.8%) were unauthorized immigrants (51.7% female; mean age, 47.6 [95% CI, 47.4-47.8] years). Compared with authorized immigrants and US-born individuals, unauthorized immigrants were more likely to be aged 18 to 44 years (80.8%), Latino (96.3%), and Spanish speaking (95.2%) and to have less than 12 years of education (53.7%). Half of unauthorized immigrants (47.1%) were uninsured compared with 15.9% of authorized immigrants and 6.0% of US-born individuals. Mean annual health care expenditures per person were $1629 (95% CI, $1330-$1928) for unauthorized immigrants, $3795 (95% CI, $3555-$4035) for authorized immigrants, and $6088 (95% CI, $5935-$6242) for US-born individuals. Conclusions and Relevance: Contrary to much political discourse in the US, this cross-sectional study found no evidence that unauthorized immigrants are a substantial economic burden on safety net facilities such as EDs. This study illustrates the value of machine learning in the study of unauthorized immigrants using large-scale, secondary databases.

Cost-effectiveness of Universal and Targeted Hepatitis C Virus Screening in the United States.

Importance: Between 2 and 3.5 million people live with chronic hepatitis C virus (HCV) infection in the US, most of whom (approximately 75%) are not aware of their disease. Despite the availability of effective HCV treatment in the early stages of infection, HCV will result in thousands of deaths in the next decade in the US. Objective: To investigate the cost-effectiveness of universal screening for all US adults aged 18 years or older for HCV in the US and of targeted screening of people who inject drugs. Design, Setting, and Participants: This simulated economic evaluation used cohort analyses in a Markov model to perform a 10 000-participant Monte Carlo microsimulation trail to evaluate the cost-effectiveness of HCV screening programs, and compared screening programs targeting people who inject drugs with universal screening of US adults age 18 years or older. Data were analyzed in December 2019. Exposures: Cost per quality-adjusted life-year (QALY). Main Outcomes and Measures: Cost per QALY gained. Results: In a 10 000 Monte Carlo microsimulation trail that compared a baseline of individuals aged 40 years (men and women) and people who inject drugs in the US, screening and treatment for HCV were estimated to increase total costs by $10 457 per person and increase QALYs by 0.23 (approximately 3 months), providing an incremental cost-effectiveness ratio of $45 465 per QALY. Also, universal screening and treatment for HCV are estimated to increase total costs by $2845 per person and increase QALYs by 0.01, providing an incremental cost-effectiveness ratio of $291 277 per QALY. Conclusions and Relevance: The findings of this study suggest that HCV screening for people who inject drugs may be a cost-effective intervention to combat HCV infection in the US, which could potentially decrease the risk of untreated HCV infection and liver-related mortality.