Large-scale modeling for housing condition prediction using machine learning algorithms.

PurposeThis paper seeks to provide researchers and librarians with an overview of the US Census Bureau's American Community Survey (ACS), with a specific focus on practical issues that users must face when choosing and using ACS datasets.Design/methodology/approachEach of the following issues are explored subsequent to a general overview of the ACS: choosing among census datasets from different census programs, interpreting and choosing between the different ACS period estimates, selecting census geography, understanding and recalculating margins of error, and accessing the data. Samples of ACS tables and formulas for creating derived estimates are used to illustrate how to interpret and work with the data.FindingsThe ACS datasets are fundamentally different from the decennial census as they are period estimates created from rolling sample surveys. The ACS has a steeper learning curve; this complexity is due in part to the number of choices users must make between datasets, but the primary challenge is learning how to understand and work with estimates as opposed to population counts.Originality/valueWhile other papers have discussed the benefits and challenges of the ACS, this paper is structured around the practical issues that researchers must face when using it. Special consideration is given to calculating derived estimates using spreadsheet formulas, as this is a key task that many users will need to perform and spreadsheets are the most likely tool users will employ to manipulate the data.

Purpose: To determine if socioeconomic, demographic, and clinical characteristics are associated with postpartum readmission. Methods: A retrospective cohort study evaluating all pregnant patients that delivered at seven hospitals within a large academic health system in New York between January 1, 2018 and March 1, 2020. Demographic information, medical comorbidities, and characteristics of antepartum, intrapartum, and postpartum care were compared between patients who were readmitted within 6 weeks postpartum and those who were not. Postpartum patients who presented to the emergency department but remained less than 23 hours were excluded. Patient ZIP codes were linked to data from the United States Census Bureau's American Community Survey and used as a proxy for neighborhood socioeconomic status. Mixed effects logistic regression was used to evaluate factors associated with an increased risk of postpartum readmission while adjusting for potential confounders. Results: A total of 57,507 delivery hospitalizations were evaluated, and 1,481 (2.5%) patients were readmitted. Black race (aOR: 1.56, 95% CI: 1.30-1.86, p < 0.001) and public health insurance (aOR: 1.19, 95% CI: 1.05-1.35, p = 0.007) were associated with an increased likelihood of postpartum readmission. Chronic hypertension (aOR: 2.83, 95% CI: 2.33-3.44, p < 0.001), body mass index >25 kg/m2 (aOR: 1.22, 95% CI: 1.05-1.42, p = 0.01), gestational weight gain >40 lb (aOR: 1.19, 95% CI: 1.01-1.40, p = 0.04), and administration of blood products (aOR: 2.18, 95% CI: 1.68-2.82, p < 0.001) were associated with an increased odd of readmission. Neighborhood characteristics were not associated with postpartum readmission. Conclusion: Efforts to reduce postpartum readmissions should focus on high-risk populations. Specific sociodemographic and clinical characteristics are associated with this complication.

Racial and ethnic minority groups experience a disproportionate burden of SARS-CoV-2 illness and studies suggest that cancer patients are at a particular risk for severe SARS-CoV-2 infection. The objective of this study was examine the association between neighborhood characteristics and SARS-CoV-2 infection among patients with cancer. We performed a cross-sectional study of New York City residents receiving treatment for cancer at a tertiary cancer center. Patients were linked by their address to data from the US Census Bureau's American Community Survey and to real estate tax data from New York's Department of City Planning. Models were used to both to estimate odds ratios (ORs) per unit increase and to predict probabilities (and 95% CI) of SARS-CoV2 infection. We identified 2350 New York City residents with cancer receiving treatment. Overall, 214 (9.1%) were infected with SARS-CoV-2. In adjusted models, the percentage of Hispanic/Latino population (aOR=1.01; 95% CI, 1.005-1.02), unemployment rate (aOR=1.10; 95% CI, 1.05-1.16), poverty rates (aOR=1.02; 95% CI, 1.0002-1.03), rate of >1 person per room (aOR=1.04; 95% CI, 1.01-1.07), average household size (aOR=1.79; 95% CI, 1.23-2.59) and population density (aOR=1.86; 95% CI, 1.27-2.72) were associated with SARS-CoV-2 infection. Among cancer patients in New York City receiving anti-cancer therapy, SARS-CoV-2 infection was associated with neighborhood- and building-level markers of larger household membership, household crowding, and low socioeconomic status. We performed a cross-sectional analysis of residents of New York City receiving treatment for cancer in which we linked subjects to census and real estate date. This linkage is a novel way to examine the neighborhood characteristics that influence SARS-COV-2 infection. We found that among patients receiving anti-cancer therapy, SARS-CoV-2 infection was associated with building and neighborhood-level markers of household crowding, larger household membership, and low socioeconomic status. With ongoing surges of SARS-CoV-2 infections, these data may help in the development of interventions to decrease the morbidity and mortality associated with SARS-CoV-2 among cancer patients.

To evaluate the likelihood of linking electronic health records (EHRs) to restricted individual-level American Community Survey (ACS) data based on patient health condition. Electronic health records (2019-2021) are derived from a primary care registry collected by the American Board of Family Medicine. These data were assigned anonymized person-level identifiers (Protected Identification Keys [PIKs]) at the U.S. Census Bureau. These records were then linked to restricted individual-level data from the ACS (2005-2022). We used logistic regressions to evaluate match rates for patients with health conditions across a range of severity: hypertension, diabetes, and chronic kidney disease. Among more than 2.8 million patients, 99.2% were assigned person-level identifiers (PIKs). There were some differences in the odds of receiving an identifier in adjusted models for patients with hypertension (OR = 1.70, 95% CI: 1.63, 1.77) and diabetes (OR = 1.17, 95% CI: 1.13, 1.22), relative to those without. There were only small differences in the odds of matching to ACS in adjusted models for patients with hypertension (OR = 1.03, 95% CI: 1.03, 1.04), diabetes (OR = 1.02, 95% CI: 1.01, 1.03), and chronic kidney disease (OR = 1.05, 95% CI: 1.03, 1.06), relative to those without. Our work supports evidence-building across government consistent with the Foundations for Evidence-Based Policymaking Act of 2018 and the goal of leveraging data as a strategic asset. Given the high PIK and ACS match rates, with small differences based on health condition, our findings suggest the feasibility of enhancing the utility of EHR data for research focused on health.

Some progress has been made in gender diversity in undergraduate medical education and the physician workforce, but much remains to be done to improve workforce disparities for women, particularly women from underrepresented populations, such as Latinas. This study examines the current level of representation and demographic characteristics of Latina physicians, including age, language use, nativity, and citizenship status. The authors used data from the 2014-2018 U.S. Census Bureau's American Community Survey (ACS) 5-year estimates for their analyses. During the time period covered by this analysis, ACS response rates ranged from 92.0% to 96.7%. The authors included in this study individuals who self-reported their occupation as physician and who self-identified their race/ethnicity as either non-Hispanic White (NHW) or Hispanic/Latino, regardless of race. The authors used person-level sampling weights provided by the ACS to convert the original 1% sample to a 100% enumeration of the population. According to the ACS 2014-2018 5-year estimates, NHW physicians make up 65.8% (660,031/1,002,527) of physicians in the United States. Women comprise 36.1% (361,442) of the total U.S. physician population; however, Hispanic/Latina women comprise only 2.4% (24,411). The female physician population is younger than the male physician population, and Hispanic female physicians are the youngest. Latina physicians are far more likely to speak Spanish at home than NHW physicians. Immigrants make up 40.1% (9,782/24,411) of the Hispanic female physician population, and 12.3% (3,012/24,411) of Hispanic female physicians are not U.S. citizens. This study suggests that Latina physicians in the United States are younger, more likely to be bilingual and speak Spanish at home, and very underrepresented, compared with NHW female and male physicians. Increasing their share of the U.S. physician workforce would benefit the pursuit of health equity for an ever more diverse population.

Executive Summary1 This paper provides a statistical portrait of the US undocumented population, with an emphasis on the social and economic condition of mixed-status households - that is, households that contain a US citizen and an undocumented resident. It is based primarily on data compiled by the Center for Migration Studies (CMS). Major findings include the following: • There were 3.3 million mixed-status households in the United States in 2014. • 6.6 million US-born citizens share 3 million households with undocumented residents (mostly their parents). Of these US-born citizens, 5.7 million are children (under age 18). • 2.9 million undocumented residents were 14 years old or younger when they were brought to the United States. • Three-quarters of a million undocumented residents are self-employed, having created their own jobs and in the process, creating jobs for many others. • A total of 1.3 million, or 13 percent of the undocumented over age 18, have college degrees. • Of those with college degrees, two-thirds, or 855,000, have degrees in four fields: engineering, business, communications, and social sciences. • Six million undocumented residents, or 55 percent of the total, speak English well, very well, or only English. • The unemployment rate for the undocumented was 6.6 percent, the same as the national rate in January 2014.2 • Seventy-three percent had incomes at or above the poverty level. • Sixty-two percent have lived in the United States for 10 years or more. • Their median household income was $41,000, about $12,700 lower than the national figure of $53,700 in 2014 (US Census Bureau 2015). Based on this profile, a massive deportation program can be expected to have the following major consequences: • Removing undocumented residents from mixed-status households would reduce median household income from $41,300 to $22,000, a drop of $19,300, or 47 percent, which would plunge millions of US families into poverty. • If just one-third of the US-born children of undocumented residents remained in the United States following a mass deportation program, which is a very low estimate, the cost of raising those children through their minority would total $118 billion. • The nation's housing market would be jeopardized because a high percentage of the 1.2 million mortgages held by households with undocumented immigrants would be in peril. • Gross domestic product (GDP) would be reduced by 1.4 percent in the first year, and cumulative GDP would be reduced by $4.7 trillion over 10 years. CMS derived its population estimates for 2014 using a series of statistical procedures that involved the analysis of data collected by the US Census Bureau's American Community Survey (ACS). The privacy of all respondents in the survey is legally mandated, and, for the reasons listed in the Appendix, the identity of undocumented residents cannot be derived from the data. A detailed description of the methodology used to develop the estimates is available at the CMS website.3

BackgroundYoung adults’ early adoption of new cell phone technologies have created challenges to survey recruitment but offer opportunities to combine random digit dialing (RDD) sampling with web mode data collection. The National Young Adult Health Survey was designed to test the feasibility of this methodology.ObjectiveIn this study, we compared response rates across the telephone mode and web mode, assessed sample representativeness, examined design effects (DEFFs), and compared cigarette smoking prevalence to a gold standard national survey.MethodsWe conducted a survey experiment where the sampling frame was randomized to single-mode telephone interviews, telephone-to-web sequential mixed mode, and single-mode web survey. A total of 831 respondents aged 18 to 34 years were recruited via RDD at baseline. A soft launch was conducted prior to main launch. We compared the web mode to the telephone modes (ie, single-mode and mixed mode) at wave 1 based on the American Association for Public Opinion Research response rate 3 for screening and extended surveys. Base-weighted demographic distributions were compared to the American Community Survey. The sample was calibrated to the US Census Bureau's American Community Survey to calculate DEFFs and to compare cigarette smoking prevalence to the National Health Interview Survey. Prevalence estimates are estimated with sampling weights and are presented with unweighted sample sizes. Consistency of estimates was judged by 95% CI.ResultsThe American Association for Public Opinion Research response rate 3 was higher in the telephone mode than in the web mode (24% and 30% vs 6.1% and 12.5%, for soft launch and main launch, respectively), which was reflected in response rate 3 for screening and extended surveys. During the soft launch, the extended survey and eligibility rate were low for respondents pushed to the web mode. To boost productivity and survey completes for the web condition, the main launch used cell phone numbers from the sampling frame where the sample vendor matched the number to auxiliary data, which suggested that the number likely belonged to an adult in the target age range. This increased the eligibility rate, but the screener response rate was lower. Compared to population distribution from the US Census Bureau, the telephone mode overrepresented men (57.1% [unweighted n=412] vs 50.9%) and those enrolled in college (40.3% [unweighted n=269] vs 23.8%); it also underrepresented those with a Bachelor of Arts or Science (34.4% [unweighted n=239] vs 55%). The web mode overrepresented White, non-Latinos (70.7% [unweighted n=90] vs 54.4%) and those with some college education (30.4% [unweighted n=40] vs 7.6%); it also underrepresented Latinos (13.6% [unweighted n=20] vs 20.7%) and those with a high school or General Education Development diploma (15.3% [unweighted n=20] vs 29.3%). The DEFF measure was 1.28 (subpopulation range 0.96-1.93). The National Young Adult Health Survey cigarette smoking prevalence was consistent with the National Health Interview Survey overall (15%, CI 12.4%-18% [unweighted 149/831] vs 13.5%, CI 12.3%-14.7% [unweighted 823/5552]), with notable deviation among 18- to 24-year-olds (15.6%, CI 11.3%-22.2% [unweighted 51/337] vs 8.7%, CI 7.1%-10.6% [unweighted 167/1647]), and those with education levels lower than Bachelor of Arts or Science (24%, CI 19.3%-29.4% [unweighted 123/524] vs 17.1%, CI 15.6%-18.7% [unweighted 690/3493]).ConclusionsRDD sampling for a web survey is not feasible for young adults due to its low response rate. However, combining this methodology with RDD telephone surveys may have a great potential for including media and collecting autophotographic data in population surveys.

Research ObjectiveGender diversification is rapidly occurring within the dental profession. Gender differences in practice choice (eg, employment vs practice ownership) and practice participation (eg, part‐time vs full‐time work hours) have been reported. It is important to understand how changes in the delivery system including workforce participation and preferences might impact the availability of services especially for underserved populations. The objectives of this study were to evaluate the variation in practice preferences among female and male dentists and to assess potential associations with socioeconomic and family factors.Study DesignThe current study used data from the Census Bureau's American Community Survey (ACS). The variables extracted describe dentists' sociodemographic characteristics including gender, age, race/ethnicity, presence of children or elders in the household, household income, and marital status, as well as dental practice characteristics such as employment status, practice setting, and work hours.Population StudiedThe 5‐year Public Use Microdata Sample data (2014–2018) from the ACS were utilized. Person‐level data collected through the questionnaire contain demographic, social, and economic information. The analytical sample consisted of 9993 dentists and 520,925 people living in the dentists' households. Survey data was weighted to generate unbiased estimates representative for the US population.Principal FindingsFemale dentists were younger (mean 43.3 years vs 53.6; P &lt; 0.001), more racially and ethnically diverse (58.1% vs 78.6% White, non‐Hispanic; P &lt; 0.001), and proportionally more foreign‐born (35.2% vs 17.7%; P &lt; 0.001) compared to male dentists. Female dentists were more likely to report being employed (55.1% vs 34.8%; P &lt; 0.001) or working part‐time (less than 30 hours per week) in dental practice (15.5% vs 11.0%; P = 0.001) than male dentists. Female dentists with children were more likely to report being an employee (64.3% vs 52.8%; P &lt; 0.001) and working part‐time (19.5% vs 14.3%; P &lt; 0.001) compared to female dentists without children in their care. Proportionally fewer female dentists left the workforce (ie, retired) in the past 12 months than male dentists (2.4% vs 3.6%; P = 0.002). Average income in the past 12 months was significantly lower (P &lt; 0.001) for female dentists (mean = $141,267; 95% Confidence Interval [CI] = $ 135,452 ‐$147,082) compared to male dentists (mean = $185,923; 95% CI = $135,452–$147,082).ConclusionsThe data indicate that the dental workforce is diversifying in gender and by race/ethnicity. The data about dental practice by gender is consistent with the current literature; female dentists were significantly more likely to be employed than to be self‐employed/a practice owner. This preference is consistent with evolving practice models including consolidations of smaller practices into group practices and changing business models for the profession. The wage gap by gender, which has been previously researched, is notable and difficult to explain.Implications for Policy or PracticeThe percentage of women entering and graduating from dental schools has increased over recent years achieving equity in numbers in dental schools in the US. Differences in employment status and workforce participation by gender are important preferences that should be monitored over time to ascertain if availability of services is affected by these trends.Primary Funding SourceHealth Resources and Services Administration.

In the United States (US), individuals of Puerto Rican heritage die of drug overdoses at higher rates than other Hispanic groups or non-Hispanic Whites; yet, little is known about the extent to which drug overdose mortality affects island-born, versus US-born, Puerto Ricans. The distinction between Puerto Rican-born and US-born provides a starting point for culturally tailored services and interventions, as place of birth often informs language preferences and cultural identifications. Therefore, this study analyzed 2013-2019 death certificate data from the National Center for Health Statistics for 415,111 US deaths attributed to drug overdose. Drug overdose deaths were compared for island-born Puerto Ricans (N=3516), US-born Puerto Ricans (N=4949), and individuals not of Puerto Rican heritage (N=406,646). Drug overdose mortality rates, including age-specific and directly age-standardized rates, were calculated for each subgroup using population estimates from the US Census Bureau's American Community Survey. Results indicated that age-adjusted drug overdose mortality rates over the period of 2013-2019 were significantly higher for island-born than US-born Puerto Rican men (46.8 versus 34.6, per 100,000), with rates in both groups significantly higher than for men not of Puerto Rican heritage (24.0 per 100,000). For women, in contrast, drug overdose mortality rates were lower for island-born than US-born Puerto Ricans (8.6 versus 11.1, per 100,000). Within stateside Puerto Rican communities, island-born men experience a disproportionate burden of drug overdose mortality, necessitating targeted, culturally appropriate interventions built around the specific norms, circumstances, and lived experiences shared by Puerto Rican migrants who use drugs.

Social determinants of health and coronavirus disease 2019 in pregnancy

Naturalization has long been recognized as a crucial step in the full integration of immigrants into US society. Yet until now, sufficient information on the naturalization-eligible has not been available that would allow the federal government, states, localities, and non-governmental service providers to develop targeted strategies on a local level to assist this population to naturalize and to overcome barriers to eligibility. This paper remedies that deficiency by providing detailed estimates on the naturalization-eligible from data collected in the US Census Bureau's American Community Survey (ACS). Naturalization rates have traditionally been calculated by dividing the naturalized or the “naturalization eligible” populations by all foreign-born persons; i.e., the naturalized, legal non-citizens, and undocumented residents. By including the unauthorized in this calculation, naturalization rates have appeared misleadingly low for populations that can naturalize. By contrast, the Center for Migration Studies of New York (CMS) provides “naturalization eligibility” rates, which it calculates by dividing the “naturalization eligible” by the foreign-born population, minus undocumented residents and legal residents who arrived after mid-2008. The paper reports that 8.6 million US residents were eligible to naturalize in 2013. This figure approximates the 8.8 million estimate of the US Department of Homeland Security (DHS). Mexican nationals constitute the largest naturalization-eligible population at 2.7 million, followed by Indian (337,000), Chinese (320,000), Cuban (316,000), and Canadian (313,000) nationals. Fifty countries have 25,000 or more naturalization-eligible persons. The large number of legally resident Mexican nationals and this population's high naturalization eligibility rate mean that US states with large Mexican populations have relatively high percentages of legal foreign-born residents who can naturalize. The overall “naturalization eligibility” rate was 31 percent in 2013, including 48 percent for Mexican nationals. Nine of the 25 largest US naturalization-eligible populations by source country have naturalization eligibility rates in excess of 40 percent, including Mexico (48 percent), Canada (45 percent), El Salvador (42 percent), the United Kingdom (41 percent), Guatemala (44 percent), Japan (56 percent), Honduras (48 percent), and Brazil (41 percent). On a state level, California, Texas, New York, and Florida contain roughly five million of the US naturalization-eligible or about 58 percent of the total population. The paper finds that a large number of naturalization-eligible immigrants may have difficulty meeting the naturalization requirements or may need intensive support to do so. This population likely includes substantial percentages of the 2.87 million naturalization-eligible who have lived in the United States for more than 25 years; 1.16 million who do not speak English; 3.0 million with less than a high school education; and the 1.8 million with incomes below the poverty level. On the other hand, high percentages of eligible immigrants would seem to be well-situated to naturalize, including those who have lived in the United States for more than 10 years (78 percent); are age 35 or older (74 percent); are married (64 percent); speak English well, very well, or only English (65 percent); have access to both a computer and the internet (74 percent); earn income above the poverty level (79 percent); and have health insurance (72 percent).

BackgroundDescribing the US emergency medical services workforce is important to understand gaps in recruitment and retention and inform efforts to improve diversity. Our objective was to describe the characteristics and temporal trends of emergency medical technicians (EMTs) and paramedics in the United States.MethodsWe performed a repeated cross‐sectional evaluation of US Census Bureau's American Community Survey 1‐year Public Use Microdata Sample data sets from 2011–2019. We included respondents working as an EMT or paramedic. Survey‐weighted descriptive statistics of demographic and employment characteristics were calculated. Trend analysis was conducted using joinpoint regression to estimate slope and annual percent change (APC).ResultsThe total estimated number of EMTs and paramedics in the United States increased from 216,310 (95%CI 204,957–227,663) in 2011 to 289,830 (95%CI 276,918–302,743) in 2019 (APC 3.0%; 95%CI 1.4%, 4.7%). There was a slight increase in the proportion of females (2011, 31%; 2019, 35%). There was a significant decrease in proportion of non‐Hispanic whites (2011, 80%; 2019, 72%; APC −1.5%; 95%CI −2.0%, −0.9%) with concurrent increases in other racial/ethnic groups (e.g., Hispanics, 2011, 10%; 2019, 13%). About half worked >40 hours per week, with little change over time. Between 15% and 18% lived and worked in different states, and about 40% traveled ≥30 minutes to their workplace.ConclusionsThe number of EMTs and paramedics actively working in EMS as their primary paid occupation has increased over time. However, there have been only modest changes in their demographic diversity.

The Impact of Socioeconomic Status on Time to Decannulation Among Children With Tracheostomies.

Disparities in cervical cancer incidence and neighborhood socioeconomic inequality in New York City

Doing well by doing good: linking access with quality