Data Source Concordance for Infectious Disease Epidemiology (preprint)

Background: As highlighted by the COVID-19 pandemic, researchers are eager to make use of a wide variety of data sources, both government-sponsored and alternative, to characterize the epidemiology of infectious diseases. To date, few studies have investigated the strengths and limitations of sources currently being used for such research. These are critical for policy makers to understand when interpreting study findings. Methods: To fill this gap in the literature, we compared infectious disease reporting for three diseases (measles, mumps, and varicella) across four different data sources: Optum (health insurance billing claims data), HealthMap (online news surveillance data), Morbidity and Mortality Weekly Reports (official government reports), and National Notifiable Disease Surveillance System (government case surveillance data). We reported the yearly number of national- and state-level disease-specific case counts and disease clusters according to each of our sources during a five-year study period (2013-2017). Findings: Our study demonstrated drastic differences in reported infectious disease incidence across data sources. When compared against the other three sources of interest, Optum data showed substantially higher, implausible standardized case counts for all three diseases. Although there was some concordance in identified state-level case counts and disease clusters, all four sources identified variations in state-level reporting. Interpretation: Researchers should consider data source limitations when attempting to characterize the epidemiology of infectious diseases. Some data sources, such as billing claims data, may be unsuitable for epidemiological research within the infectious disease context.

The Impact of State Paid Sick Leave Policies on Longitudinal Weekday Workplace Mobility During the COVID-19 Pandemic (preprint)

Objectives. To evaluate whether the Families First Coronavirus Response Act (FFCRA) modified the association between pre-existing state paid sick leave (PSL) and weekday workplace mobility between February 15 and July 7, 2020. Methods. The 50 US states and Washington, D.C. were divided into exposure groups based on the presence or absence of pre-existing state PSL policies. Derived from Google COVID-19 Community Mobility Reports, the outcome was measured as the daily percent change in weekday workplace mobility. Mixed-effects, interrupted time series regression was performed to evaluate weekday workplace mobility after the implementation of the FFCRA on April 1st, 2020. Results. States with pre-existing PSL policies exhibited a greater drop in mobility following the passage of the FFCRA ({beta}=-8.86,95%CI:-11.6,-6.10,P< 001). This remained significant after adjusting for state-level health, economic, and sociodemographic indicators ({beta}=-3.13,95%CI:-5.92,-0.34,P=.039). Conclusions. Pre-existing PSL policies contributed to a significant decline in weekday workplace mobility after the FFCRA, which may have influenced local health outcomes. Policy implications The presence of pre-existing state policies may differentially influence the impact of federal legislation enacted during emergencies.

An Inductive Approach to Using Google Search Trends To Identify Attitudes and Areas of COVID-19 Vaccine Hesitancy (preprint)

In this study, we present and apply a novel inductive approach that uses Google Search Trends data for public health surveillance of COVID-19 vaccine hesitancy from January to April 2021. In contrast to previous studies that use researcher-curated keywords, our study combed through over 2 million social media posts on Twitter to identify over 3,000 keywords related to vaccination, enabling us to capture potential sources of vaccine hesitancy that would otherwise be missed. These keywords were used to retrieve Google Search Trends for 774 queries that yielded nonzero search volumes for at least 45 U.S. states. We then used the CovidCast survey of vaccine acceptance at the U.S. state level to calculate a Spearman’s rank correlation coefficient for each term. Negatively correlated queries, we argue, can help shed light on potential sources of vaccine hesitancy. As noted by Flahault et al. (2020), in order to achieve “precision global health” and “real-time action”, researchers must creatively combine multiple data sources in order to deliver targeted interventions. We are hopeful that our approach will be useful for practitioners in addressing the problem of vaccine hesitancy.

Evaluation of County-Level Heterogeneity in Excess Mortality in Colorado from March to September 2020 (preprint)

Importance: Tracking the direct and indirect impact of the coronavirus disease 2019 (COVID-19) pandemic on all-cause mortality in the United States has been hindered by the lack of testing and by reporting delays. Evaluating excess mortality, or the number of deaths above what is expected in a given time period, provides critical insights into the true burden of the COVID-19 pandemic caused by the novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). Stratifying mortality data by demographics such as age, sex, race, ethnicity, and geography helps quantify how subgroups of the population have been differentially affected. Similarly, stratifying mortality data by cause of death reveals the public health effects of the pandemic in terms of other acute and chronic diseases.Objective: To provide stratified estimates of excess mortality in Colorado from March to September 2020.Design, Setting, and Population: This study evaluated the number of excess deaths both directly due to SARS-CoV-2 infection and from all other causes between March and September 2020 at the county level in Colorado. Data were obtained from the Vital Statistics Program at the Colorado Department of Public Health and Environment. These estimates of excess mortality were derived by comparing population- adjusted mortality rates in 2020 with rates in the same months from 2015 to 2019.Results: We found evidence of excess mortality in Colorado between March and September 2020. Two peaks in excess deaths from all causes were recorded in the state, one mid-April and the other at the end of June. Since the first documented SARS-CoV-2 infection on March 5th, we estimated that the excess mortality rate in Colorado was two times higher than the officially reported COVID-19 mortality rate. State-level cumulative excess mortality from all causes reached 71 excess deaths per 100k residents (~4000 excess deaths in the state); in contrast, 35 deaths per 100k directly due to SARS-CoV-2 were recorded in the same period (~1980 deaths. Excess mortality occurred in 52 of 64 counties, accounting for 99% of the state’s population. Most excess deaths recorded from March to September 2020 were associated with acute events (estimated at 44 excess deaths per 100k residents and at 9 after excluding deaths directly due to SARS-CoV-2) rather than with chronic conditions (~21 excess deaths per 100k). Among Coloradans aged 14-44, 1.4 times more deaths occurred in those months than during the same period in the five previous years. Hispanic White males died of COVID-19 at the highest rate during this time (~90 deaths from COVID-19 per 100k residents); however, Non-Hispanic Black/African American males were the most affected in terms of overall excess mortality (~204 excess deaths per 100k). Beyond inequalities in COVID-19 mortality per se, these findings signal considerable regional and racial-ethnic disparities in excess all-cause mortality that need to be addressed for a just recovery and in future public health crises.

Evaluation of County-Level Heterogeneity in Excess Mortality in Colorado from March to September 2020 (preprint)

1. Importance Tracking the direct and indirect impact of the coronavirus disease 2019 (COVID-19) pandemic on all-cause mortality in the United States has been hindered by the lack of testing and by reporting delays. Evaluating excess mortality, or the number of deaths above what is expected in a given time period, provides critical insights into the true burden of the COVID-19 pandemic caused by the novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). Stratifying mortality data by demographics such as age, sex, race, ethnicity, and geography helps quantify how subgroups of the population have been differentially affected. Similarly, stratifying mortality data by cause of death reveals the public health effects of the pandemic in terms of other acute and chronic diseases. Objective To provide stratified estimates of excess mortality in Colorado from March to September 2020. Design, Setting, and Population This study evaluated the number of excess deaths both directly due to SARS-CoV-2 infection and from all other causes between March and September 2020 at the county level in Colorado. Data were obtained from the Vital Statistics Program at the Colorado Department of Public Health and Environment. These estimates of excess mortality were derived by comparing population-adjusted mortality rates in 2020 with rates in the same months from 2015 to 2019. Results We found evidence of excess mortality in Colorado between March and September 2020. Two peaks in excess deaths from all causes were recorded in the state, one mid-April and the other at the end of June. Since the first documented SARS-CoV-2 infection on March 5th, we estimated that the excess mortality rate in Colorado was two times higher than the officially reported COVID-19 mortality rate. State-level cumulative excess mortality from all causes reached 71 excess deaths per 100k residents (∼4000 excess deaths in the state); in contrast, 35 deaths per 100k directly due to SARS-CoV-2 were recorded in the same period (∼1980 deaths. Excess mortality occurred in 52 of 64 counties, accounting for 99% of the state’s population. Most excess deaths recorded from March to September 2020 were associated with acute events (estimated at 44 excess deaths per 100k residents and at 9 after excluding deaths directly due to SARS-CoV-2) rather than with chronic conditions (∼21 excess deaths per 100k). Among Coloradans aged 14-44, 1.4 times more deaths occurred in those months than during the same period in the five previous years. Hispanic White males died of COVID-19 at the highest rate during this time (∼90 deaths from COVID-19 per 100k residents); however, Non-Hispanic Black/African American males were the most affected in terms of overall excess mortality (∼204 excess deaths per 100k). Beyond inequalities in COVID-19 mortality per se, these findings signal considerable regional and racial-ethnic disparities in excess all-cause mortality that need to be addressed for a just recovery and in future public health crises.

Evaluating Criminal Justice Reform During COVID-19: The Need for a Novel Sentiment Analysis Package (preprint)

Existing natural language processing lexicons that underlie current sentiment analysis (SA) algorithms may not perform adequately in certain academic disciplines depending on contextual complexities. The health and safety of incarcerated persons and correctional personnel have been prominent in news media discourse during the COVID-19 pandemic, potentially highlighting the need for a novel SA lexicon and algorithm that is tailored for the examination of public health policy in the context of the criminal justice system. We utilized a text corpus consisting of news articles at the intersection of COVID-19 and criminal justice to analyze the performance of existing lexicons collected across state-level outlets between January and May 2020. Our results demonstrated that sentence sentiment scores provided by three popular SA packages differ considerably from manually-curated ratings. This dissimilarity was especially pronounced when the text was more polarized, whether negatively or positively. A randomly selected set of 1,000 manually scored sentences, and the corresponding binary document term matrices, were used to train two new sentiment prediction algorithms (i.e., linear regression and random forest regression) to verify the performance of the manually-curated ratings. By better accounting for the unique context in which incarceration-related terminologies are used in news media, both of our proposed models outperformed all existing SA packages considered for comparison. Our findings suggest that there is a need to develop a novel lexicon, and potentially an accompanying algorithm, for analysis of text related to public health within the criminal justice system, as well as criminal justice more broadly.

Digital phenotyping of complex psychological responses to the COVID-19 pandemic (preprint)

Background: The novel coronavirus disease 2019 (COVID-19) has negatively impacted mortality, economic conditions, and mental health. A large scale study on psychological reactions to the pandemic to inform ongoing population-level symptom tracking and response to treatment is currently lacking. Methods: Average intake scores for standard depression and anxiety symptom scales were tracked from January 1, 2017 to June 9, 2020 for patients seeking treatment from a digital mental health service to gauge the relationship between COVID-19 and self-reported symptoms. We applied natural language processing (NLP) to unstructured therapy transcript data from patients seeking treatment during the height of the pandemic in the United States between March 1, 2020 and June 9, 2020 to identify words associated with COVID-19 mentions. This analysis was used to identify symptoms that were present beyond those assessed by standard depression and anxiety measures. Results: Depression and anxiety symptoms reported by 169,889 patients between January 1, 2017 and June 9, 2020 were identified. There was no detectable change in intake depression symptom scores. Intake anxiety symptom scores increased 1.42 scale points [95% CI: 1.18, 1.65] between March 15, 2020 and April 1, 2020, when scores peaked. In the transcript data of these 169,889 patients, plus an expanded sample of 49,267 patients without symptom reports, term frequency-inverse document frequency (tf-idf) identified 2,377 positively correlated and 661negatively correlated terms that were significantly (FDR<.01) associated with mentions of the virus. These terms were classifiable into 24 symptoms beyond those included in the diagnostic criteria for anxiety or depression. Conclusions: The COVID-19 pandemic may have increased intake anxiety symptoms for individuals seeking digital mental health treatment. NLP analyses suggest that standard symptom scales for depression and anxiety alone are inadequate to fully assess and track psychological reactions to the pandemic. Symptoms of grief, trauma, obsession-compulsion, agoraphobia, hypochondriasis, panic, and non- suicidal self-injury should be monitored as part of a new COVID-19 Syndrome category.

Modeling Between-Population Variation in COVID-19 Dynamics in Hubei, Lombardy, and New York City (preprint)

As the COVID-19 pandemic continues, formulating targeted policy interventions that are informed by differential SARS-CoV2 transmission dynamics will be of vital importance to national and regional governments. We develop an individual-level model for SARS-CoV2 transmission that accounts for location-dependent distributions of age, household structure, and comorbidities. We use these distributions together with age-stratified contact matrices to instantiate specific models for Hubei, China; Lombardy, Italy; and New York City, United States. Using data on reported deaths to obtain a posterior distribution over unknown parameters, we infer differences in the progression of the epidemic in the three locations. We also examine the role of transmission due to particular age groups on total infections and deaths. The effect of limiting contacts by a particular age group varies by location, indicating that strategies to reduce transmission should be tailored based on population-specific demography and social structure. These findings highlight the role of between-population variation in formulating policy interventions. Across the three populations though, we find that targeted “salutary sheltering" by 50% of a single age group may substantially curtail transmission when combined with the adoption of physical distancing measures by the rest of the population. Code may be found at https://github.com/bwilder0/covid_abm_release.

The Role of Environmental Factors on Transmission Rates of the COVID-19 Outbreak: An Initial Assessment in Two Spatial Scales. (preprint)

A novel coronavirus (SARS-CoV-2) was identified in Wuhan, Hubei Province, China, in December 2019 and has caused over 240,000 cases of COVID-19 worldwide as of March 19, 2020. Previous studies have supported an epidemiological hypothesis that cold and dry environments facilitate the survival and spread of droplet-mediated viral diseases, and warm and humid environments see attenuated viral transmission (e.g., influenza). However, the role of temperature and humidity in transmission of COVID-19 has not yet been established. Here, we examine the spatial variability of the basic reproductive numbers of COVID-19 across provinces and cities in China and show that environmental variables alone cannot explain this variability. Our findings suggest that changes in weather alone (i.e., increase of temperature and humidity as spring and summer months arrive in the Northern Hemisphere) will not necessarily lead to declines in case count without the implementation of extensive public health interventions.