Robert Yaffee | New York University (original) (raw)

Papers by Robert Yaffee

Journal of Statistical Software, 2007

Stata 10 (Time Series and Forecasting) StataCorp LP, College Station, TX. USD 1,795 (corporate), ... more Stata 10 (Time Series and Forecasting) StataCorp LP, College Station, TX. USD 1,795 (corporate), USD 985 (educational) for single user Stata/SE 10 (exact price varies by version and purchaser status). http://www.stata.com/ A review of Stata versions 9 and 10 time series and forecasting capability Stata 10 is a powerful, versatile, and flexible statistical package with a wide range of userfriendly and accurate time series analytical and forecasting commands. It was released in June 2007, about two years after Stata 9 appeared. Prospective users want to know what features the package offers, while veteran users want to know what is novel about the latest version(s). This review addresses time series and forecasting capability in Stata 10 and notes prominent new features in versions 9 and 10, what is still needed and what might be refined. Because all users need to have confidence in the accuracy of the package, this review examines accuracy of the principal time series algorithms. It also briefly describes associated graphics, diagnostics, documentation, help facilities, output, customization, and support. Altogether, these features characterize a very well-designed and well-crafted and powerful time series analysis and forecasting package (Baum 2004).

Journal of the American Geriatrics Society, 1994

Traditional mental status and psychometric assessments bottom out in the late stages of Alzheimer... more Traditional mental status and psychometric assessments bottom out in the late stages of Alzheimer disease (AD). A method adapted from cognitive testing in infants, the Ordinal Scales of Psychological Development was modified (M-OSPD) and applied to a severely demented population. The concurrent validity of this method was tested in comparison with Functional Assessment Staging (FAST). Internal consistency as a measure for reliability was also determined. Cross sectional study. Subjects were generally evaluated in their residence, usually a nursing home or a private home. Severely cognitively impaired subjects who fulfilled criteria for probable AD were studied. Evaluation consisted of clinical global, mental status, functional, and cognitive assessments including the Global Deterioration Scale (GDS) and the Mini-Mental State Examination (MMSE). Seventy patients were evaluated. Traditional mental status assessments (eg, the MMSE) manifested virtually uniform bottom scores in all GDS stage 7 subjects (n = 46), and GDS stage 6 subjects had MMSE scores within one standard deviation unit of zero. In contrast, the M-OSPD scale continued to show results in the last stages of the disease. The Spearman correlation coefficient between the M-OSPD total score and the 11 FAST substages represented in this sample was -0.77 (P < 0.001). The results indicate that patients who are functionally more impaired also show continuing increments in cognitive loss. These cognitive and functional assessments for measuring the magnitude of deterioration in AD can be applied to the estimated half-million nursing home residents presently labeled "untestable" with the goal of optimization of care and residual capacities.

Panel data analysis is an increasingly popular form of longitudinal data analysis among social an... more Panel data analysis is an increasingly popular form of longitudinal data analysis among social and behavioral science researchers. A panel is a cross-section or group of people who are surveyed periodically over a given time span. In this article, we will consider a small sample of panel data analytic applications in the social sciences. Then we will address the data structure for panel analysis. Principal models of panel analysis will be summarized, along with some of their relative advantages and disadvantages. We will discuss a test to determine whether to use fixed or random effects models. After a synopsis of methods of estimations tailored to different situations, we will conclude with a brief discussion of popular software capable of performing panel analysis. Some Applications of Panel Analysis Panel data analysis is a method of studying a particular subject within multiple sites, periodically observed over a defined time frame. Within the social sciences, panel analysis has...

ABSTRACT An OxMetrics Show Case

ABSTRACT Acknowledgments Acknowledgments

Our objectives were to examine predictive parameters of psychological impacts, resulting from the... more Our objectives were to examine predictive parameters of psychological impacts, resulting from the Chornobyl accident, on residents living in the oblasts of Kiev and Zhitomyr. We tested drivers for psycho-social depression based on estimates radiological dose received from radioactivity release during the accident and the perception of increased health effects associated with this radiation. To obtain a representative sam-ple of individuals, we attached computer generated random numbers to area codes provided by the telephone company. In January 2009, Russia created an intervening crisis by interrupting supplies of natural gas to the Ukraine. We employed modified scenario forecasting to circumvent crisis effects that could otherwise undermine the internal validity of our study. State space methods were used to model and graph trajectories of psycho-social depression reported by male and female respondents. Results of the dose reconstruction process revealed that the dose received by ...

oforeval.ado version 4.3.2 is an update of an earlier Stata program I wrote to perform ex post fo... more oforeval.ado version 4.3.2 is an update of an earlier Stata program I wrote to perform ex post forecast evaluations of ARIMA or ARFIMA programs. I list it here to help any interested persons who wish to download it at no cost. I hope it helps.

Version 4.3.2 of foreval.ado performs ex ante forecast evaluation following the generation of an ... more Version 4.3.2 of foreval.ado performs ex ante forecast evaluation following the generation of an ex ante forecast from an ARIMA or an ARFIMA model using Stata.

The American Journal of Geriatric Psychiatry, 2009

There are a paucity of long-term studies from the United States concerning predictors of outcome ... more There are a paucity of long-term studies from the United States concerning predictors of outcome among depressed older community adults. This article examines predictors of depression in a biracial sample of older persons in Brooklyn, NY. The authors conducted a naturalistic study of 110 persons aged 55 years and older living in randomly selected block groups who had a Center for Epidemiologic Studies-Depression (CES-D) score of > or = 8 at baseline. Persons were reassessed on an average of 3 years later. Their mean age was 69 years, 52% were women, and 35% were whites, and 65% were blacks, among whom 71% were African Caribbeans. Using George's Social Antecedent Model of Depression, the authors examined the impact of 13 predictor variables on two outcome measures: presence of either subsyndromal or syndromal depression (CES-D score > or = 8) and presence of syndromal depression (CES-D score > or = 16). To control for design effects, the authors used SUDAAN for the data analysis. On follow-up, 82% and 88% of subsyndromally and syndromally depressed persons at baseline, respectively, were depressed (CES-D > or = 8). In logistic regression, baseline depressive symptoms, baseline anxiety symptoms, greater increase in anxiety symptoms during the follow-up period, and higher locus of control were predictors of any level of depression. These four variables along with greater paranoid ideation and/or psychoses and more reliable social contacts were significant predictors of syndromal depression on follow-up. There were no inter- or intraracial differences in outcome. Depressed community elders in Brooklyn have highly unfavorable outcomes. Preventive strategies that target at-risk persons-i.e., especially those with baseline subsyndromal depression, greater anxiety symptoms, and more paranoid ideation and/or psychoses-may reduce the development of severe or persistent depression.

The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry, 2006

There have been few multiracial epidemiologic community-based studies of phobia in older adults. ... more There have been few multiracial epidemiologic community-based studies of phobia in older adults. The aim of this study was to determine the prevalence of phobia and associated factors among older persons living in a northeastern urban area. Using 1990 census data for Brooklyn, NY, the authors attempted to interview all persons age 55+ in randomly selected block groups. The final sample consisted of 214 whites and 860 blacks. The authors used an adaptation of George's Social Antecedent Model for examining the association of 18 individual variables and one interactive variable with the presence of a phobia. The dependent variable was derived from the Guy's/Age Concern community survey. The sample was weighted by race and gender. To control for design effects, the authors used SUDAAN for the data analysis. A total of 8.9% of the sample met criteria for a current phobia and 10.2% met phobia criteria at some time during their life. Using logistic regression analysis, the authors ...

The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry, 2005

The authors sought to determine differences between depressed older black residents in an urban c... more The authors sought to determine differences between depressed older black residents in an urban community who do and do not use formal mental health services. The Treatment Group (TG) consisted of 106 black patients age >or=55 with a diagnosis of depression who were recruited from outpatient psychiatric programs in Brooklyn, NY. The Untreated Community Group (UCG) consisted of 101 cognitively intact black subjects age >or=55 from randomly selected block groups in Brooklyn who met symptom criteria for major or minor depression and had no previous history of psychiatric treatment. In logistic regression analysis, TG persons were significantly more likely than the UCG persons to be female, younger, born in the United States, to have impaired daily functioning, have a family history of mental illness, and believe that environmental factors and religious activities could influence mental illness. TG persons were significantly less likely to have social network members who provided ...

When I read this article, I wondered why there was such rush to publish in gambling study researc... more When I read this article, I wondered why there was such rush to publish in gambling study research and publishing. The study by Kearney, Roblek, Thurman, and Turnbough (1996) appears to be afflicted with serious flaws in research design that call for discussion. The paper begins by associating gambling with problems of adolescents. However, the focus of the paper shifts quickly from problem or pathological gambling to a report of unusual recreational activities of different types of students within the gambling capital of the country. The methodology seems impaired. The internal validity of the study is plagued with several problems. Differential selection of the sample precludes representativeness. Low sample size deprives the findings of statistical power and conclusion validity. Lack of control for key confounding variables deprives the paper of implied construct validity and explanatory power, rendering apparent differences probably

The purpose of this article is to respond to and expand on the ideas presented by Father McGowan ... more The purpose of this article is to respond to and expand on the ideas presented by Father McGowan in The Ethics of Gambling Research: An Agenda for Mature Analysis. We provide specific recommendations for future research and public policy in the field of gambling studies. We suggest that key conceptual definitions-such as, problem, compulsive, and pathological gambling-should be clarified, established, and distinguished from one another before gambling research is conducted. Proper methodological procedures are recommended, where power analyses, pilot studies, and representative samples are appropriately conducted and analyzed. Retrospective and Prospective studies are considered and differentiated while Discrete Time Event History Analysis-namely, Life Tables Analysis and Discrete Time Logistic Regression-are proposed.

Forecasters are expected to provide evaluations of their forecasts along with their forecasts. Th... more Forecasters are expected to provide evaluations of their forecasts along with their forecasts. The forecast assessments demonstrate comparative, adequate, or optimal accuracy by common forecasting criteria to provide acceptable credence in the forecasts. To assist the Stata user in this process, Robert Yaffee has written Stata programs to evaluate ARIMA and GARCH models. He explains how these assessment programs are applied to one-step-ahead and dynamic forecasts, ex post and ex ante forecasts, conditional and unconditional forecasts, as well as combinations of forecasts. In his presentation, he will also demonstrate how assessment can be applied to rolling origin forecasts of time-series models.

Psychiatric Services, 2005

The authors explored racial differences in the prevalence of depression and its associated factor... more The authors explored racial differences in the prevalence of depression and its associated factors among older persons. Using 1990 census data for Brooklyn, New York, the authors attempted to interview all cognitively intact persons age 55 years and older in randomly selected block groups. The sample was weighted by ethnicity and gender. The authors adapted George's Social Antecedent Model of Depression to allow examination of 20 independent variables and the nominal dependent variable consisting of three levels of depression. The data were analyzed with SUDAAN. Syndromal depression was found among 8 percent of blacks and 10 percent of whites. Subsyndromal depression was found among 13 percent of blacks and 28 percent of whites. No racial differences were found in rates of syndromal depression, but significant racial differences were found in rates of subsyndromal depression and of any type of depression. Nonlinear effects on both types of depression were found, and higher levels of stress had a greater impact on whites than on blacks. The racial difference in subsyndromal depression was explained by its lower prevalence among French-speaking African Caribbeans. Many racial differences were found in the variables associated with syndromal and subsyndromal depression. Race had an independent effect on the rate of subsyndromal depression and an interactive effect with stress on the rate of both syndromal and subsyndromal depression. For each racial group, different elements may play a role in the etiology, maintenance, and relief of depression. The findings underscore the importance of recognizing within-group and between-group racial differences in depression.

Psychiatric Services, 2006

The objectives of this study were to estimate the prevalence of, and factors associated with, anx... more The objectives of this study were to estimate the prevalence of, and factors associated with, anxiety in epilepsy. We conducted a cross-sectional analysis using data from the Neurological Disease and Depression Study. The prevalence of anxiety and associated factors were assessed using descriptive statistics and logistic regression. Of the total sample (n = 250 patients), nearly 40.0% of participants had anxiety according to the Hospital Anxiety and Depression Scale. The most prevalent symptom of anxiety was "worrying thoughts" (35.6%). After adjustment for age and sex, depression (odds ratio [OR] = 8.97, 95% confidence interval [CI] = 4.38-18.40), medication side effects (OR = 1.79, 95% CI = 1.04-3.05), smoking (OR = 4.35, 95% CI = 2.27-8.31), and illicit substance use (OR = 2.42, 95% CI = 1.18-4.96) were significantly associated with higher odds of anxiety, whereas higher education (OR = 0.47, 95% CI = 0.28-0.80) was associated with lower odds of anxiety. Furthermore, participants with anxiety reported more severe epilepsy, debilitating seizures, and overall lower quality of life. Evidence from our study reveals a high prevalence of anxiety in persons with epilepsy and that anxiety is associated with a variety of negative outcomes. These findings further emphasize the need for more studies to understand the impact of anxiety and its relationship with various sociodemographic and clinical factors.

Journal of Statistical Software, 2007

Stata 10 (Time Series and Forecasting) StataCorp LP, College Station, TX. USD 1,795 (corporate), ... more Stata 10 (Time Series and Forecasting) StataCorp LP, College Station, TX. USD 1,795 (corporate), USD 985 (educational) for single user Stata/SE 10 (exact price varies by version and purchaser status). http://www.stata.com/ A review of Stata versions 9 and 10 time series and forecasting capability Stata 10 is a powerful, versatile, and flexible statistical package with a wide range of userfriendly and accurate time series analytical and forecasting commands. It was released in June 2007, about two years after Stata 9 appeared. Prospective users want to know what features the package offers, while veteran users want to know what is novel about the latest version(s). This review addresses time series and forecasting capability in Stata 10 and notes prominent new features in versions 9 and 10, what is still needed and what might be refined. Because all users need to have confidence in the accuracy of the package, this review examines accuracy of the principal time series algorithms. It also briefly describes associated graphics, diagnostics, documentation, help facilities, output, customization, and support. Altogether, these features characterize a very well-designed and well-crafted and powerful time series analysis and forecasting package (Baum 2004).

Journal of the American Geriatrics Society, 1994

Traditional mental status and psychometric assessments bottom out in the late stages of Alzheimer... more Traditional mental status and psychometric assessments bottom out in the late stages of Alzheimer disease (AD). A method adapted from cognitive testing in infants, the Ordinal Scales of Psychological Development was modified (M-OSPD) and applied to a severely demented population. The concurrent validity of this method was tested in comparison with Functional Assessment Staging (FAST). Internal consistency as a measure for reliability was also determined. Cross sectional study. Subjects were generally evaluated in their residence, usually a nursing home or a private home. Severely cognitively impaired subjects who fulfilled criteria for probable AD were studied. Evaluation consisted of clinical global, mental status, functional, and cognitive assessments including the Global Deterioration Scale (GDS) and the Mini-Mental State Examination (MMSE). Seventy patients were evaluated. Traditional mental status assessments (eg, the MMSE) manifested virtually uniform bottom scores in all GDS stage 7 subjects (n = 46), and GDS stage 6 subjects had MMSE scores within one standard deviation unit of zero. In contrast, the M-OSPD scale continued to show results in the last stages of the disease. The Spearman correlation coefficient between the M-OSPD total score and the 11 FAST substages represented in this sample was -0.77 (P < 0.001). The results indicate that patients who are functionally more impaired also show continuing increments in cognitive loss. These cognitive and functional assessments for measuring the magnitude of deterioration in AD can be applied to the estimated half-million nursing home residents presently labeled "untestable" with the goal of optimization of care and residual capacities.

Panel data analysis is an increasingly popular form of longitudinal data analysis among social an... more Panel data analysis is an increasingly popular form of longitudinal data analysis among social and behavioral science researchers. A panel is a cross-section or group of people who are surveyed periodically over a given time span. In this article, we will consider a small sample of panel data analytic applications in the social sciences. Then we will address the data structure for panel analysis. Principal models of panel analysis will be summarized, along with some of their relative advantages and disadvantages. We will discuss a test to determine whether to use fixed or random effects models. After a synopsis of methods of estimations tailored to different situations, we will conclude with a brief discussion of popular software capable of performing panel analysis. Some Applications of Panel Analysis Panel data analysis is a method of studying a particular subject within multiple sites, periodically observed over a defined time frame. Within the social sciences, panel analysis has...

ABSTRACT An OxMetrics Show Case

ABSTRACT Acknowledgments Acknowledgments

Our objectives were to examine predictive parameters of psychological impacts, resulting from the... more Our objectives were to examine predictive parameters of psychological impacts, resulting from the Chornobyl accident, on residents living in the oblasts of Kiev and Zhitomyr. We tested drivers for psycho-social depression based on estimates radiological dose received from radioactivity release during the accident and the perception of increased health effects associated with this radiation. To obtain a representative sam-ple of individuals, we attached computer generated random numbers to area codes provided by the telephone company. In January 2009, Russia created an intervening crisis by interrupting supplies of natural gas to the Ukraine. We employed modified scenario forecasting to circumvent crisis effects that could otherwise undermine the internal validity of our study. State space methods were used to model and graph trajectories of psycho-social depression reported by male and female respondents. Results of the dose reconstruction process revealed that the dose received by ...

oforeval.ado version 4.3.2 is an update of an earlier Stata program I wrote to perform ex post fo... more oforeval.ado version 4.3.2 is an update of an earlier Stata program I wrote to perform ex post forecast evaluations of ARIMA or ARFIMA programs. I list it here to help any interested persons who wish to download it at no cost. I hope it helps.

Version 4.3.2 of foreval.ado performs ex ante forecast evaluation following the generation of an ... more Version 4.3.2 of foreval.ado performs ex ante forecast evaluation following the generation of an ex ante forecast from an ARIMA or an ARFIMA model using Stata.

The American Journal of Geriatric Psychiatry, 2009

There are a paucity of long-term studies from the United States concerning predictors of outcome ... more There are a paucity of long-term studies from the United States concerning predictors of outcome among depressed older community adults. This article examines predictors of depression in a biracial sample of older persons in Brooklyn, NY. The authors conducted a naturalistic study of 110 persons aged 55 years and older living in randomly selected block groups who had a Center for Epidemiologic Studies-Depression (CES-D) score of > or = 8 at baseline. Persons were reassessed on an average of 3 years later. Their mean age was 69 years, 52% were women, and 35% were whites, and 65% were blacks, among whom 71% were African Caribbeans. Using George's Social Antecedent Model of Depression, the authors examined the impact of 13 predictor variables on two outcome measures: presence of either subsyndromal or syndromal depression (CES-D score > or = 8) and presence of syndromal depression (CES-D score > or = 16). To control for design effects, the authors used SUDAAN for the data analysis. On follow-up, 82% and 88% of subsyndromally and syndromally depressed persons at baseline, respectively, were depressed (CES-D > or = 8). In logistic regression, baseline depressive symptoms, baseline anxiety symptoms, greater increase in anxiety symptoms during the follow-up period, and higher locus of control were predictors of any level of depression. These four variables along with greater paranoid ideation and/or psychoses and more reliable social contacts were significant predictors of syndromal depression on follow-up. There were no inter- or intraracial differences in outcome. Depressed community elders in Brooklyn have highly unfavorable outcomes. Preventive strategies that target at-risk persons-i.e., especially those with baseline subsyndromal depression, greater anxiety symptoms, and more paranoid ideation and/or psychoses-may reduce the development of severe or persistent depression.

The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry, 2006

There have been few multiracial epidemiologic community-based studies of phobia in older adults. ... more There have been few multiracial epidemiologic community-based studies of phobia in older adults. The aim of this study was to determine the prevalence of phobia and associated factors among older persons living in a northeastern urban area. Using 1990 census data for Brooklyn, NY, the authors attempted to interview all persons age 55+ in randomly selected block groups. The final sample consisted of 214 whites and 860 blacks. The authors used an adaptation of George's Social Antecedent Model for examining the association of 18 individual variables and one interactive variable with the presence of a phobia. The dependent variable was derived from the Guy's/Age Concern community survey. The sample was weighted by race and gender. To control for design effects, the authors used SUDAAN for the data analysis. A total of 8.9% of the sample met criteria for a current phobia and 10.2% met phobia criteria at some time during their life. Using logistic regression analysis, the authors ...

The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry, 2005

The authors sought to determine differences between depressed older black residents in an urban c... more The authors sought to determine differences between depressed older black residents in an urban community who do and do not use formal mental health services. The Treatment Group (TG) consisted of 106 black patients age >or=55 with a diagnosis of depression who were recruited from outpatient psychiatric programs in Brooklyn, NY. The Untreated Community Group (UCG) consisted of 101 cognitively intact black subjects age >or=55 from randomly selected block groups in Brooklyn who met symptom criteria for major or minor depression and had no previous history of psychiatric treatment. In logistic regression analysis, TG persons were significantly more likely than the UCG persons to be female, younger, born in the United States, to have impaired daily functioning, have a family history of mental illness, and believe that environmental factors and religious activities could influence mental illness. TG persons were significantly less likely to have social network members who provided ...

When I read this article, I wondered why there was such rush to publish in gambling study researc... more When I read this article, I wondered why there was such rush to publish in gambling study research and publishing. The study by Kearney, Roblek, Thurman, and Turnbough (1996) appears to be afflicted with serious flaws in research design that call for discussion. The paper begins by associating gambling with problems of adolescents. However, the focus of the paper shifts quickly from problem or pathological gambling to a report of unusual recreational activities of different types of students within the gambling capital of the country. The methodology seems impaired. The internal validity of the study is plagued with several problems. Differential selection of the sample precludes representativeness. Low sample size deprives the findings of statistical power and conclusion validity. Lack of control for key confounding variables deprives the paper of implied construct validity and explanatory power, rendering apparent differences probably

The purpose of this article is to respond to and expand on the ideas presented by Father McGowan ... more The purpose of this article is to respond to and expand on the ideas presented by Father McGowan in The Ethics of Gambling Research: An Agenda for Mature Analysis. We provide specific recommendations for future research and public policy in the field of gambling studies. We suggest that key conceptual definitions-such as, problem, compulsive, and pathological gambling-should be clarified, established, and distinguished from one another before gambling research is conducted. Proper methodological procedures are recommended, where power analyses, pilot studies, and representative samples are appropriately conducted and analyzed. Retrospective and Prospective studies are considered and differentiated while Discrete Time Event History Analysis-namely, Life Tables Analysis and Discrete Time Logistic Regression-are proposed.

Forecasters are expected to provide evaluations of their forecasts along with their forecasts. Th... more Forecasters are expected to provide evaluations of their forecasts along with their forecasts. The forecast assessments demonstrate comparative, adequate, or optimal accuracy by common forecasting criteria to provide acceptable credence in the forecasts. To assist the Stata user in this process, Robert Yaffee has written Stata programs to evaluate ARIMA and GARCH models. He explains how these assessment programs are applied to one-step-ahead and dynamic forecasts, ex post and ex ante forecasts, conditional and unconditional forecasts, as well as combinations of forecasts. In his presentation, he will also demonstrate how assessment can be applied to rolling origin forecasts of time-series models.

Psychiatric Services, 2005

The authors explored racial differences in the prevalence of depression and its associated factor... more The authors explored racial differences in the prevalence of depression and its associated factors among older persons. Using 1990 census data for Brooklyn, New York, the authors attempted to interview all cognitively intact persons age 55 years and older in randomly selected block groups. The sample was weighted by ethnicity and gender. The authors adapted George's Social Antecedent Model of Depression to allow examination of 20 independent variables and the nominal dependent variable consisting of three levels of depression. The data were analyzed with SUDAAN. Syndromal depression was found among 8 percent of blacks and 10 percent of whites. Subsyndromal depression was found among 13 percent of blacks and 28 percent of whites. No racial differences were found in rates of syndromal depression, but significant racial differences were found in rates of subsyndromal depression and of any type of depression. Nonlinear effects on both types of depression were found, and higher levels of stress had a greater impact on whites than on blacks. The racial difference in subsyndromal depression was explained by its lower prevalence among French-speaking African Caribbeans. Many racial differences were found in the variables associated with syndromal and subsyndromal depression. Race had an independent effect on the rate of subsyndromal depression and an interactive effect with stress on the rate of both syndromal and subsyndromal depression. For each racial group, different elements may play a role in the etiology, maintenance, and relief of depression. The findings underscore the importance of recognizing within-group and between-group racial differences in depression.

Psychiatric Services, 2006

The objectives of this study were to estimate the prevalence of, and factors associated with, anx... more The objectives of this study were to estimate the prevalence of, and factors associated with, anxiety in epilepsy. We conducted a cross-sectional analysis using data from the Neurological Disease and Depression Study. The prevalence of anxiety and associated factors were assessed using descriptive statistics and logistic regression. Of the total sample (n = 250 patients), nearly 40.0% of participants had anxiety according to the Hospital Anxiety and Depression Scale. The most prevalent symptom of anxiety was "worrying thoughts" (35.6%). After adjustment for age and sex, depression (odds ratio [OR] = 8.97, 95% confidence interval [CI] = 4.38-18.40), medication side effects (OR = 1.79, 95% CI = 1.04-3.05), smoking (OR = 4.35, 95% CI = 2.27-8.31), and illicit substance use (OR = 2.42, 95% CI = 1.18-4.96) were significantly associated with higher odds of anxiety, whereas higher education (OR = 0.47, 95% CI = 0.28-0.80) was associated with lower odds of anxiety. Furthermore, participants with anxiety reported more severe epilepsy, debilitating seizures, and overall lower quality of life. Evidence from our study reveals a high prevalence of anxiety in persons with epilepsy and that anxiety is associated with a variety of negative outcomes. These findings further emphasize the need for more studies to understand the impact of anxiety and its relationship with various sociodemographic and clinical factors.

Hantavirus pulmonary syndrome (HPS) is a zoonotic disease associated with the deer mouse (Peromy... more Hantavirus pulmonary syndrome (HPS) is a zoonotic disease associated with the deer mouse (Peromyscus maniculatus) in the United States. Prediction of increases in deer mouse population density could result in advance warning of high-risk conditions and help direct public health interventions. We used temperature, precipitation, and monthly counts of rodents captured at a Montana trapping site to develop two causal and 16 univariate forecasting models as well as to compare their efficacy in predicting deer mouse population densities. A naïve model, which carried forward the last observed population abundance value, was used as a basis of comparison. We generated 12 forecasts over a rolling origin fixed three-month window over three years. Only the univariate state-space local level models outperformed the naïve model. Failure of the causal models may be related to their dependence on accurate forecasts of temperature and precipitation variables well into the 3-month forecast horizon. Future models will incorporate remote measures of habitat quality including satellite-derived indices of greenness and primary productivity.

Hantavirus pulmonary syndrome (HPS) is a zoonotic disease associated with the deer mouse (Peromys... more Hantavirus pulmonary syndrome (HPS) is a zoonotic disease associated with the deer mouse (Peromyscus maniculatus) in the United States. Prediction of increases in deer mouse population density could result in advance warning of high-risk conditions and help direct public health interventions. We used temperature, precipitation, and monthly
counts of rodents captured at a Montana trapping site to develop two causal and 16 univariate forecasting models as well as to compare their efficacy in predicting deer mouse population densities. A naïve model, which carried forward the last observed population abundance value, was used as a basis of comparison. We generated 12 forecasts over a rolling origin fixed three-month window over three years. Only the
univariate state-space local level models outperformed the naïve model. Failure of the causal models may be related to their dependence on accurate forecasts of temperature and precipitation variables well into the 3-month forecast horizon. Future models will incorporate remote measures of habitat quality including satellite-derived indices of
greenness and primary productivity.

Our objectives were to examine predictive parameters of psychological impacts, resulting from the... more Our objectives were to examine predictive parameters of psychological impacts, resulting from the Chornobyl accident, on residents living in the oblasts of Kiev and Zhitomyr. We tested drivers for psycho-
social depression based on estimates radiological dose received from radioactivity released during the accident and the perception of increased health e ffects associated with this radiation. To obtain a representative sample of individuals, we attached computer generated random numbers to area codes provided by the telephone
company. In January 2009, Russia created an intervening crisis by interrupting supplies of natural gas to the Ukraine. We employed scenario forecasting to circumvent crisis eff ects that could otherwise would have undermined the internal validity of our study. State space methods were used to model and graph trajectories of psycho-social depression reported by male and female respondents. The dose reconstruction process revealed that the dose received by this population was too low to identify pathological disease or injury. Our
empirical analysis revealed that psychological impacts of the nuclear incident stemmed from perceived risks, rather than actual exposure to ionizing radiation directly associated with the Chornobyl nuclear accident. This research was funded by National Science Foundation HSD Grant 082 6983.