Lauren Restivo | University of Baltimore (original) (raw)

Papers by Lauren Restivo

Research paper thumbnail of Weight Adjustment Methods using Multilevel Propensity Models and Random Forests

Using two national surveys as an example, this paper focuses on the selection of predictors for n... more Using two national surveys as an example, this paper focuses on the selection of predictors for non-response analysis and weight adjustments. Weighting classes for non-response adjustments are formed using a set of core variables that are correlated with response behavior and with survey outcomes. The intent is to minimize the potential of non-response bias by balancing it against acceptable increases in weighting variances due to weighting variability. The choice of variables to use when defining weighting classes is determined by measures of variable importance for predicting response. This paper compares two methods for variable selection and weighting class adjustments. One method uses response propensity models estimated with mixed effects logistic regression. The other method uses recursive partitioning, and more specifically, Random Forest algorithms. Mixed effects models are run using SAS Proc Glimmix that allows the modeling of random effects. This paper assesses the method...

Research paper thumbnail of Guantanamo Bay

Research paper thumbnail of Sexual Victimization Among Intimates

Sexual Victimization: Then and Now

Research paper thumbnail of Health Affairs Substance User Groups Understanding Patterns Of High-Cost Health Care Use Across Different

Substance use contributes to significant societal burdens, including high-cost health care use. H... more Substance use contributes to significant societal burdens, including high-cost health care use. However, these burdens may vary by type of substance and level of involvement. Using the 2009–13 National Surveys on Drug Use and Health, we examined all-cause hospitalizations and estimated costs across substance use profiles for alcohol, marijuana, and other illicit drugs. For each substance, we characterized differences between abstainers, nondiagnostic users (people who used the substance but did not meet diagnostic criteria for substance use disorder), and people with substance use disorders. In a multivariate analysis, we found that the odds of hospitalization were 16 percent lower for nondiagnostic marijuana users and 11 percent lower for nondiagnostic alcohol users, compared to abstainers. Neither alcoholnor marijuana-specific substance use disorders were associated with hospitalization. In contrast, substance use disorders for other illicit drugs were strongly associated with hos...

Research paper thumbnail of Examining collective efficacy and perceptions of policing in East Baltimore

This study examines how residents from a high-crime, high-poverty neighborhood in East Baltimore ... more This study examines how residents from a high-crime, high-poverty neighborhood in East Baltimore interact with one another, participate in their community, and perceive police. Using community surveys collected from 191 respondents, the study empirically measures collective efficacy, community participation, and police services and encounters. We predict that high levels of collective efficacy lead to more positive perceptions of police and an increased willingness to work with law enforcement. The results indicate that neighborhood trust is an important factor in shaping a community’s overall perception of police. Furthermore, older residents who own their homes are more likely to report a more positive perception of police, specifically police response.

Research paper thumbnail of Understanding Drivers of Crime in East Baltimore: Resident Perceptions of Why Crime Persists

American Journal of Criminal Justice, 2015

Research paper thumbnail of Understanding Patterns Of High-Cost Health Care Use Across Different Substance User Groups

Health affairs (Project Hope), 2016

Substance use contributes to significant societal burdens, including high-cost health care use. H... more Substance use contributes to significant societal burdens, including high-cost health care use. However, these burdens may vary by type of substance and level of involvement. Using the 2009-13 National Surveys on Drug Use and Health, we examined all-cause hospitalizations and estimated costs across substance use profiles for alcohol, marijuana, and other illicit drugs. For each substance, we characterized differences between abstainers, nondiagnostic users (people who used the substance but did not meet diagnostic criteria for substance use disorder), and people with substance use disorders. In a multivariate analysis, we found that the odds of hospitalization were 16 percent lower for nondiagnostic marijuana users and 11 percent lower for nondiagnostic alcohol users, compared to abstainers. Neither alcohol- nor marijuana-specific substance use disorders were associated with hospitalization. In contrast, substance use disorders for other illicit drugs were strongly associated with h...

Research paper thumbnail of Child Abuse

The Encyclopedia of Crime & Punishment, 2015

Research paper thumbnail of Factors Associated With Receiving Rapid HIV Testing Among Individuals on Probation or Parole

Journal of correctional health care : the official journal of the National Commission on Correctional Health Care, Oct 1, 2016

Despite the strong correlation between HIV and corrections, testing and prevention efforts have l... more Despite the strong correlation between HIV and corrections, testing and prevention efforts have largely been ignored among community corrections populations. The current study is a secondary analysis to compare characteristics of individuals under community corrections supervision who completed rapid HIV testing with those who refused such testing (N = 2,382) in Baltimore, Maryland, and Providence, Rhode Island. Results indicate that the following variables were significantly associated with the receipt of rapid HIV testing: being female (p = .024), Black race (p = .004), homeless (p = .016), early age of crime onset (p = .001), more drug use during the past 90 days (p = .033), and previously tested for hepatitis C virus/hepatitis B virus (p = .024). Such findings make it especially important that individuals under community supervision be linked with services in the community to ensure that HIV testing and health care planning occur simultaneously.

Research paper thumbnail of Gender and Crime

The Encyclopedia of Crime & Punishment, 2015

Research paper thumbnail of Weight Adjustment Methods using Multilevel Propensity Models and Random Forests

Using two national surveys as an example, this paper focuses on the selection of predictors for n... more Using two national surveys as an example, this paper focuses on the selection of predictors for non-response analysis and weight adjustments. Weighting classes for non-response adjustments are formed using a set of core variables that are correlated with response behavior and with survey outcomes. The intent is to minimize the potential of non-response bias by balancing it against acceptable increases in weighting variances due to weighting variability. The choice of variables to use when defining weighting classes is determined by measures of variable importance for predicting response. This paper compares two methods for variable selection and weighting class adjustments. One method uses response propensity models estimated with mixed effects logistic regression. The other method uses recursive partitioning, and more specifically, Random Forest algorithms. Mixed effects models are run using SAS Proc Glimmix that allows the modeling of random effects. This paper assesses the method...

Research paper thumbnail of Guantanamo Bay

Research paper thumbnail of Sexual Victimization Among Intimates

Sexual Victimization: Then and Now

Research paper thumbnail of Health Affairs Substance User Groups Understanding Patterns Of High-Cost Health Care Use Across Different

Substance use contributes to significant societal burdens, including high-cost health care use. H... more Substance use contributes to significant societal burdens, including high-cost health care use. However, these burdens may vary by type of substance and level of involvement. Using the 2009–13 National Surveys on Drug Use and Health, we examined all-cause hospitalizations and estimated costs across substance use profiles for alcohol, marijuana, and other illicit drugs. For each substance, we characterized differences between abstainers, nondiagnostic users (people who used the substance but did not meet diagnostic criteria for substance use disorder), and people with substance use disorders. In a multivariate analysis, we found that the odds of hospitalization were 16 percent lower for nondiagnostic marijuana users and 11 percent lower for nondiagnostic alcohol users, compared to abstainers. Neither alcoholnor marijuana-specific substance use disorders were associated with hospitalization. In contrast, substance use disorders for other illicit drugs were strongly associated with hos...

Research paper thumbnail of Examining collective efficacy and perceptions of policing in East Baltimore

This study examines how residents from a high-crime, high-poverty neighborhood in East Baltimore ... more This study examines how residents from a high-crime, high-poverty neighborhood in East Baltimore interact with one another, participate in their community, and perceive police. Using community surveys collected from 191 respondents, the study empirically measures collective efficacy, community participation, and police services and encounters. We predict that high levels of collective efficacy lead to more positive perceptions of police and an increased willingness to work with law enforcement. The results indicate that neighborhood trust is an important factor in shaping a community’s overall perception of police. Furthermore, older residents who own their homes are more likely to report a more positive perception of police, specifically police response.

Research paper thumbnail of Understanding Drivers of Crime in East Baltimore: Resident Perceptions of Why Crime Persists

American Journal of Criminal Justice, 2015

Research paper thumbnail of Understanding Patterns Of High-Cost Health Care Use Across Different Substance User Groups

Health affairs (Project Hope), 2016

Substance use contributes to significant societal burdens, including high-cost health care use. H... more Substance use contributes to significant societal burdens, including high-cost health care use. However, these burdens may vary by type of substance and level of involvement. Using the 2009-13 National Surveys on Drug Use and Health, we examined all-cause hospitalizations and estimated costs across substance use profiles for alcohol, marijuana, and other illicit drugs. For each substance, we characterized differences between abstainers, nondiagnostic users (people who used the substance but did not meet diagnostic criteria for substance use disorder), and people with substance use disorders. In a multivariate analysis, we found that the odds of hospitalization were 16 percent lower for nondiagnostic marijuana users and 11 percent lower for nondiagnostic alcohol users, compared to abstainers. Neither alcohol- nor marijuana-specific substance use disorders were associated with hospitalization. In contrast, substance use disorders for other illicit drugs were strongly associated with h...

Research paper thumbnail of Child Abuse

The Encyclopedia of Crime & Punishment, 2015

Research paper thumbnail of Factors Associated With Receiving Rapid HIV Testing Among Individuals on Probation or Parole

Journal of correctional health care : the official journal of the National Commission on Correctional Health Care, Oct 1, 2016

Despite the strong correlation between HIV and corrections, testing and prevention efforts have l... more Despite the strong correlation between HIV and corrections, testing and prevention efforts have largely been ignored among community corrections populations. The current study is a secondary analysis to compare characteristics of individuals under community corrections supervision who completed rapid HIV testing with those who refused such testing (N = 2,382) in Baltimore, Maryland, and Providence, Rhode Island. Results indicate that the following variables were significantly associated with the receipt of rapid HIV testing: being female (p = .024), Black race (p = .004), homeless (p = .016), early age of crime onset (p = .001), more drug use during the past 90 days (p = .033), and previously tested for hepatitis C virus/hepatitis B virus (p = .024). Such findings make it especially important that individuals under community supervision be linked with services in the community to ensure that HIV testing and health care planning occur simultaneously.

Research paper thumbnail of Gender and Crime

The Encyclopedia of Crime & Punishment, 2015