scholarly journals Using multiple imputation to classify potential outcomes subgroups

2021 ◽  
pp. 096228022110028
Author(s):  
Yun Li ◽  
Irina Bondarenko ◽  
Michael R Elliott ◽  
Timothy P Hofer ◽  
Jeremy MG Taylor
Keyword(s):  
Multiple Imputation ◽  
Statistical Methods ◽  
The United States ◽  
Treatment Selection ◽  
Potential Outcomes ◽  
Model Framework ◽  
Medical Tests ◽  
Gene Assay ◽  
The Impact ◽  
Test Result

With medical tests becoming increasingly available, concerns about over-testing, over-treatment and health care cost dramatically increase. Hence, it is important to understand the influence of testing on treatment selection in general practice. Most statistical methods focus on average effects of testing on treatment decisions. However, this may be ill-advised, particularly for patient subgroups that tend not to benefit from such tests. Furthermore, missing data are common, representing large and often unaddressed threats to the validity of most statistical methods. Finally, it is often desirable to conduct analyses that can be interpreted causally. Using the Rubin Causal Model framework, we propose to classify patients into four potential outcomes subgroups, defined by whether or not a patient’s treatment selection is changed by the test result and by the direction of how the test result changes treatment selection. This subgroup classification naturally captures the differential influence of medical testing on treatment selections for different patients, which can suggest targets to improve the utilization of medical tests. We can then examine patient characteristics associated with patient potential outcomes subgroup memberships. We used multiple imputation methods to simultaneously impute the missing potential outcomes as well as regular missing values. This approach can also provide estimates of many traditional causal quantities of interest. We find that explicitly incorporating causal inference assumptions into the multiple imputation process can improve the precision for some causal estimates of interest. We also find that bias can occur when the potential outcomes conditional independence assumption is violated; sensitivity analyses are proposed to assess the impact of this violation. We applied the proposed methods to examine the influence of 21-gene assay, the most commonly used genomic test in the United States, on chemotherapy selection among breast cancer patients.

Counting Guns

2002 ◽  
Vol 26 (4) ◽  
pp. 699-708
Author(s):  
Gordon Wood ◽  
Robert Churchill ◽  
Edward Cook ◽  
James Lindgren ◽  
Wilbur Miller ◽  
...  
Keyword(s):  
Statistical Analysis ◽  
Social Science ◽  
Statistical Methods ◽  
Violent Crime ◽  
The United States ◽  
Real Income ◽  
Crime Rates ◽  
Gun Control ◽  
Gun Laws ◽  
The Impact

At the fall 2001 Social Science History Association convention in Chicago, the Crime and Justice network sponsored a forum on the history of gun ownership, gun use, and gun violence in the United States. Our purpose was to consider how social science historians might contribute nowand in the future to the public debate over gun control and gun rights. To date, we have had little impact on that debate. It has been dominated by mainstream social scientists and historians, especially scholars such as Gary Kleck, John Lott, and Michael Bellesiles, whose work, despite profound flaws, is politically congenial to either opponents or proponents of gun control. Kleck and Mark Gertz (1995), for instance, argue on the basis of their widely cited survey that gun owners prevent numerous crimes each year in theUnited States by using firearms to defend themselves and their property. If their survey respondents are to be believed, American gun owners shot 100,000 criminals in 1994 in selfdefense–a preposterous number (Cook and Ludwig 1996: 57–58; Cook and Moore 1999: 280–81). Lott (2000) claims on the basis of his statistical analysis of recent crime rates that laws allowing private individuals to carry concealed firearms deter murders, rapes, and robberies, because criminals are afraid to attack potentially armed victims. However, he biases his results by confining his analysis to the years between 1977 and 1992, when violent crime rates had peaked and varied little from year to year (ibid.: 44–45). He reports only regression models that support his thesis and neglects to mention that each of those models finds a positive relationship between violent crime and real income, and an inverse relationship between violent crime and unemployment (ibid.: 52–53)–implausible relationships that suggest the presence of multicollinearity, measurement error, or misspecification. Lott then misrepresents his results by claiming falsely that statistical methods can distinguish in a quasi-experimental way the impact of gun laws from the impact of other social, economic, and cultural forces (ibid.: 26, 34–35; Guterl 1996). Had Lott extended his study to the 1930s, the correlation between gun laws and declining homicide rates that dominates his statistical analysis would have disappeared. An unbiased study would include some consideration of alternative explanations and an acknowledgment of the explanatory limits of statistical methods.

Who Uses Fact-Checking Sites? The Impact of Demographics, Political Antecedents, and Media Use on Fact-Checking Site Awareness, Attitudes, and Behavior

2020 ◽  
Vol 25 (2) ◽  
pp. 217-237 ◽  
Author(s):  
Craig T. Robertson ◽  
Rachel R. Mourão ◽  
Esther Thorson
Keyword(s):  
The United States ◽  
Stage Model ◽  
Literature Analysis ◽  
Model Framework ◽  
Persuasive Message ◽  
Fact Checking ◽  
Audience Awareness ◽  
And Behavior ◽  
Attitudes And Behavior ◽  
The Impact

This study examines audience relationships to fact-checking sites in the United States. Focus is placed on predictors of audience awareness of, attitudes toward, and visits to such sites within a stage model framework drawn from the persuasive message literature. Analysis of survey data from a U.S. sample shows that liberals and liberal/mainstream news consumers are more aware of, positive toward, and likely to report using fact-checking sites. Conservatives are less positive and conservative news consumers see such sites as less useful to them. Findings indicate that while specific combinations of predictors of awareness, attitudes, and behavior vary, fact-checking sites have a particular appeal to liberals and liberal/mainstream news consumers. Results point to U.S. fact-checking sites being absorbed into wider ideological discourses and patterns of ideological news consumption.

scholarly journals Estimating the number of SARS-CoV-2 infections in the United States

2020 ◽  
Author(s):  
Dayton G. Thorpe ◽  
Kelsey Lyberger
Keyword(s):  
United States ◽  
New York ◽  
Infection Rate ◽  
New York State ◽  
Herd Immunity ◽  
The United States ◽  
Public Events ◽  
Response Team ◽  
The Impact ◽  
Test Result

AbstractWe apply a model developed by The COVID-19 Response Team [S. Flaxman, S. Mishra, A. Gandy, et al., “Estimating the number of infections and the impact of non-pharmaceutical interventions on COVID-19 in 11 European countries,” tech. rep., Imperial College London, 2020.] to estimate the total number of SARS-CoV-2 infections in the United States. Across the United States we estimate as of April 18, 2020 the fraction of the population infected was 4.6% [3.6%, 5.8%], 21 times the portion of the population with a positive test result. Excluding New York state, which we estimate accounts for over half of infections in the United States, we estimate an infection rate of 2.3% [2.1%, 2.8%].We include the timing of each state’s implementation of interventions including encouraging social distancing, closing schools, banning public events, and a lockdown / stay-at-home order. We assume fatalities are reported correctly and infer the number and timing of infections based on the infection fatality rate measured in populations that were tested universally for SARS-CoV-2. Underreporting of deaths would drive our estimates to be too low. Reporting of deaths on the wrong day could drive errors in either direction. This model does not include effects of herd immunity; in states where the estimated infection rate is very high - namely, New York - our estimates may be too high.

Demonstrating Nutrient Cost Gradients: A Brooklyn Case Study

2014 ◽  
Vol 84 (5-6) ◽  
pp. 244-251 ◽  
Author(s):  
Robert J. Karp ◽  
Gary Wong ◽  
Marguerite Orsi
Keyword(s):  
Low Income ◽  
Energy Content ◽  
Food Selection ◽  
The United States ◽  
United States Department ◽  
Micronutrient Deficiencies ◽  
Caloric Density ◽  
Low Income Families ◽  
The Cost ◽  
The Impact

Abstract. Introduction: Foods dense in micronutrients are generally more expensive than those with higher energy content. These cost-differentials may put low-income families at risk of diminished micronutrient intake. Objectives: We sought to determine differences in the cost for iron, folate, and choline in foods available for purchase in a low-income community when assessed for energy content and serving size. Methods: Sixty-nine foods listed in the menu plans provided by the United States Department of Agriculture (USDA) for low-income families were considered, in 10 domains. The cost and micronutrient content for-energy and per-serving of these foods were determined for the three micronutrients. Exact Kruskal-Wallis tests were used for comparisons of energy costs; Spearman rho tests for comparisons of micronutrient content. Ninety families were interviewed in a pediatric clinic to assess the impact of food cost on food selection. Results: Significant differences between domains were shown for energy density with both cost-for-energy (p < 0.001) and cost-per-serving (p < 0.05) comparisons. All three micronutrient contents were significantly correlated with cost-for-energy (p < 0.01). Both iron and choline contents were significantly correlated with cost-per-serving (p < 0.05). Of the 90 families, 38 (42 %) worried about food costs; 40 (44 %) had chosen foods of high caloric density in response to that fear, and 29 of 40 families experiencing both worry and making such food selection. Conclusion: Adjustments to USDA meal plans using cost-for-energy analysis showed differentials for both energy and micronutrients. These differentials were reduced using cost-per-serving analysis, but were not eliminated. A substantial proportion of low-income families are vulnerable to micronutrient deficiencies.

Wrestling with 9/11: Immigrant Perceptions and Perceptions of Immigrants

2006 ◽  
Vol 3 (2) ◽  
pp. 107-124 ◽  
Author(s):  
Caroline Brettell
Keyword(s):  
United States ◽  
Immigration Policy ◽  
Asian Indians ◽  
World Trade Center ◽  
The United States ◽  
Fort Worth ◽  
Foreign Born ◽  
Immigrant Populations ◽  
Trade Center ◽  
The Impact

Soon after 9/11 a research project to study new immigration into the Dallas Fort Worth metropolitan area got under way. In the questionnaire that was administered to 600 immigrants across five different immigrant populations (Asian Indians, Vietnamese, Mexicans, Salvadorans, and Nigerians) between 2003 and 2005 we decided to include a question about the impact of 9/11 on their lives. We asked: “How has the attack on the World Trade Center on September 11, 2001 affected your position as an immigrant in the United States?” This article analyzes the responses to this question, looking at similarities and differences across different immigrant populations. It also addresses the broader issue of how 9/11 has affected both immigration policy and attitudes toward the foreign-born in the United States. 

Sign in / Sign up

Export Citation Format

Share Document