Statistical dialog management for health interventions

Research endeavors on spoken dialogue systems in the 1990s and 2000s have led to the deployment of commercial spoken dialogue systems (SDS) in microdomains such as customer service automation, reservation/booking and question answering systems. Recent research in SDS has been focused on the development of applications in different domains (e.g. virtual counseling, personal coaches, social companions) which requires more sophistication than the previous generation of commercial SDS. The focus of this research project is the delivery of behavior change interventions based on the brief intervention counseling style via spoken dialogue systems. ^ Brief interventions (BI) are evidence-based, short, well structured, one-on-one counseling sessions. Many challenges are involved in delivering BIs to people in need, such as finding the time to administer them in busy doctors' offices, obtaining the extra training that helps staff become comfortable providing these interventions, and managing the cost of delivering the interventions. Fortunately, recent developments in spoken dialogue systems make the development of systems that can deliver brief interventions possible. ^ The overall objective of this research is to develop a data-driven, adaptable dialogue system for brief interventions for problematic drinking behavior, based on reinforcement learning methods. The implications of this research project includes, but are not limited to, assessing the feasibility of delivering structured brief health interventions with a data-driven spoken dialogue system. Furthermore, while the experimental system focuses on harmful alcohol drinking as a target behavior in this project, the produced knowledge and experience may also lead to implementation of similarly structured health interventions and assessments other than the alcohol domain (e.g. obesity, drug use, lack of exercise), using statistical machine learning approaches. ^ In addition to designing a dialog system, the semantic and emotional meanings of user utterances have high impact on interaction. To perform domain specific reasoning and recognize concepts in user utterances, a named-entity recognizer and an ontology are designed and evaluated. To understand affective information conveyed through text, lexicons and sentiment analysis module are developed and tested.^

Statistical Dialog Management for Health Interventions

Research endeavors on spoken dialogue systems in the 1990s and 2000s have led to the deployment of commercial spoken dialogue systems (SDS) in microdomains such as customer service automation, reservation/booking and question answering systems. Recent research in SDS has been focused on the development of applications in different domains (e.g. virtual counseling, personal coaches, social companions) which requires more sophistication than the previous generation of commercial SDS. The focus of this research project is the delivery of behavior change interventions based on the brief intervention counseling style via spoken dialogue systems. Brief interventions (BI) are evidence-based, short, well structured, one-on-one counseling sessions. Many challenges are involved in delivering BIs to people in need, such as finding the time to administer them in busy doctors' offices, obtaining the extra training that helps staff become comfortable providing these interventions, and managing the cost of delivering the interventions. Fortunately, recent developments in spoken dialogue systems make the development of systems that can deliver brief interventions possible. The overall objective of this research is to develop a data-driven, adaptable dialogue system for brief interventions for problematic drinking behavior, based on reinforcement learning methods. The implications of this research project includes, but are not limited to, assessing the feasibility of delivering structured brief health interventions with a data-driven spoken dialogue system. Furthermore, while the experimental system focuses on harmful alcohol drinking as a target behavior in this project, the produced knowledge and experience may also lead to implementation of similarly structured health interventions and assessments other than the alcohol domain (e.g. obesity, drug use, lack of exercise), using statistical machine learning approaches. In addition to designing a dialog system, the semantic and emotional meanings of user utterances have high impact on interaction. To perform domain specific reasoning and recognize concepts in user utterances, a named-entity recognizer and an ontology are designed and evaluated. To understand affective information conveyed through text, lexicons and sentiment analysis module are developed and tested.

A computer based approach to improve the dietary and physical activity patterns of a diverse group of adolescents

Diet and physical activity patterns have been implicated as major factors in the increasing prevalence of childhood and adolescent obesity. It is estimated that between 16 and 33 percent of children and adolescents in the United States are overweight (CDC, 2000). Moreover, the CDC estimates that less than 50% of adolescents are physically active on a regular basis (CDC, 2003). Interventions must be focused to modify these behaviors. Facilitating the understanding of proper nutrition and need for physical activity among adolescents is the first step in preventing overweight and obesity and delaying the development of chronic diseases later in life (Dwyer, 2000). The purpose of this study was to compare the outcomes of students receiving one of two forms of education (both emphasizing diet and physical activity), to determine whether a computer based intervention (CBI) program using an interactive, animated CD-ROM would elicit a greater behavior change in comparison to a traditional didactic intervention (TDI) program. A convenience sample of 254 high school students aged 14-19 participated in the 6-month program. A pre-test post-test design was used, with follow-up measures taken at three months post-intervention. ^ No change was noted in total fat, saturated fat, fruit/vegetables, or fiber intake for any of the groups. There was also no change in perceived self-efficacy or perceived social support. Results did, however, indicate an increase in nutrition knowledge for both intervention groups (p<0.001). In addition, the CBI group demonstrated more positive and sustained behavior changes throughout the course of the study. These changes included a decrease in BMI (ppre/post<0.001, ppost/follow-up<0.001), number of meals skipped (ppre/post<0.001), and soda consumption (ppre/post=0.003, ppost/follow-up=0.03) and an increase in nutrition knowledge (ppre/post<0.001, ppre/follow-up <0.001), physical activity (ppre/post<0.05, p pre/follow-up<0.01), frequency of label reading (ppre/follow-up <0.0l) and in dairy consumption (ppre/post=0.03). The TDI group did show positive gains in some areas post intervention, however a return to baseline behavior was shown at follow-up. Findings of this study suggest that compared to traditional didactic teaching, computer-based nutrition and health education has greater potential to elicit change in knowledge and behavior as well as promote maintenance of the behavior change over time. ^

Adapting Mpowerment to a Rural Area

As HIV/AIDS continues to disproportionately impact men who have sex with men (MSM) (CDC, 2010a), effective and timely prevention strategies for this population must be developed. Specifically, evidence-based interventions that can be easily adapted and have proven effectiveness are needed. Hence, the purpose of the current study was to assess the impact of the Mpowerment Project (Hayes, Rebchook, & Kegeles, 2003), a community level HIV prevention program originally designed for young urban gay men, when adapted for rural gay men. The Mpowerment Project is recognized as evidence- based intervention by the CDC (CDC, 2009b). The current study is an extension of this research, assessing Mpowerment model fidelity and the behavioral and attitudinal changes that occurred among participants. Data were collected from participants in a rural area of southeast Idaho from 2002-2004. Data were collected prior to M-Group participation and at a three months follow-up. The 66 individuals completing the M-Group pre and posttest assessment also attended a minimum of three study events and a maximum of 226 events. Results revealed no significant changes in attitudinal variables and all but one behavioral variable among Rural Mpowerment (R-MP) participants. The one significant behavior change was an increase in reported safer sex discussion among friends, indicating a possible change in the social norm regarding safer sex. Results also indicate that program fidelity was maintained and the Mpowerment Project is adaptable to rural areas. However, there was no indication of attitudinal changes in participants of this study. There were no changes in behavioral variables aside from discussion about safer sex with friends increasing. The lack of evidence-based interventions for rural gay men highlights the need for further research on the community impact of the Mpowerment Project on rural participants.

Adapting Mpowerment to a rural area

As HIV/AIDS continues to disproportionately impact men who have sex with men (MSM) (CDC, 2010a), effective and timely prevention strategies for this population must be developed. Specifically, evidence-based interventions that can be easily adapted and have proven effectiveness are needed. Hence, the purpose of the current study was to assess the impact of the Mpowerment Project (Hayes, Rebchook, & Kegeles, 2003), a community level HIV prevention program originally designed for young urban gay men, when adapted for rural gay men. The Mpowerment Project is recognized as evidence- based intervention by the CDC (CDC, 2009b). The current study is an extension of this research, assessing Mpowerment model fidelity and the behavioral and attitudinal changes that occurred among participants. Data were collected from participants in a rural area of southeast Idaho from 2002-2004. Data were collected prior to M-Group participation and at a three months follow-up. The 66 individuals completing the M-Group pre and posttest assessment also attended a minimum of three study events and a maximum of 226 events. Results revealed no significant changes in attitudinal variables and all but one behavioral variable among Rural Mpowerment (R-MP) participants. The one significant behavior change was an increase in reported safer sex discussion among friends, indicating a possible change in the social norm regarding safer sex. Results also indicate that program fidelity was maintained and the Mpowerment Project is adaptable to rural areas. However, there was no indication of attitudinal changes in participants of this study. There were no changes in behavioral variables aside from discussion about safer sex with friends increasing. The lack of evidence-based interventions for rural gay men highlights the need for further research on the community impact of the Mpowerment Project on rural participants.

Fourth generation evaluation revisited: Testing a revised pause model

This research provides data which investigates the feasibility of using fourth generation evaluation during the process of instruction. A semester length course entitled "Multicultural Communications", (PUR 5406/4934) was designed and used in this study, in response to the need for the communications profession to produce well-trained culturally sensitive practitioners for the work force and the market place. A revised pause model consisting of three one-on-one indepth interviews conducted outside of the class, three reflections periods during the class and a self-reflective essay prepared one week before the end of the course was analyzed. Narrative and graphic summaries of participant responses produced significant results. The revised pause model was found to be an effective evaluation method for use in multicultural education under certain conditions as perceived by the participants in the study. participant self-perceived behavior change and knowledge acquisition was identified through use of the revised pause model. Study results suggest that by using the revised pause model of evaluation, instructors teaching multicultural education in schools of journalism and mass communication is yet another way of enhancing their ability to become both the researcher and the research subject. In addition, the introduction of a qualitative model has been found to be a more useful way of generating participant involvement and introspection. Finally, the instructional design of the course used in the study provides communication educators with a practical way of preparing their students be effective communicators in a multicultural world.

Effect of the Supplemental Nutrition Assistance Program (SNAP) and nutrition education on nutrition and health outcomes of HIV+ individuals

Factors associated with and barriers to participation in Supplemental Nutrition Assistance Program (SNAP) and the effect participation has on food security, nutrition status, disease status and quality of life was investigated in a cross-sectional study including 175 HIV infected individuals. In addition, the effect of a targeted nutrition education on nutrition knowledge, readiness to dietary behavior change, nutrition status, disease status and quality of life was also investigated among a subset of the population (N = 45) in a randomized clinical control trial. ^ SNAP participation rate was 70.3%, similar to the State of Florida and national participation rates. SNAP participation was positively and independently associated with being born in the US (P < 0.001), having monthly income less than $1000 (P = 0.006), and receiving antiretroviral treatment (P < 0.001). Participation barriers include denial of participation by program, recent incarceration, living in a shelter where participation is not allowed and unawareness of eligibility status. In regression analyses, SNAP participation was not significantly associated with improved food security, nutrition status, disease status and health related quality of life (HRQOL). Over half (56%) of the population experienced food insecurity and had inadequate intakes of half of the nutrients assessed. Illicit drug, alcohol and cigarette use were high in this population (31%, 55% and 63% respectively), and affected food security, nutrients intake, disease status and HRQOL. The nutrition education intervention resulted in a trend towards improvements nutrition knowledge, self-efficacy, and readiness to change without impacting nutrition status, disease state and quality of life. ^ Food insecurity and other nutrition related issues, with implications for treatment, management and cost of HIV disease, continue to plague infected individuals living in poverty. More resources, including food and nutrition programs, specifically targeted towards this population are needed to address these issues.^

Effect of the Supplemental Nutrition Assistance Program (SNAP) and Nutrition Education on Nutrition and Health Outcomes of HIV+ Individuals

Factors associated with and barriers to participation in Supplemental Nutrition Assistance Program (SNAP) and the effect participation has on food security, nutrition status, disease status and quality of life was investigated in a cross-sectional study including 175 HIV infected individuals. In addition, the effect of a targeted nutrition education on nutrition knowledge, readiness to dietary behavior change, nutrition status, disease status and quality of life was also investigated among a subset of the population (N = 45) in a randomized clinical control trial. SNAP participation rate was 70.3%, similar to the State of Florida and national participation rates. SNAP participation was positively and independently associated with being born in the US (P < 0.001), having monthly income less than $1000 (P = 0.006), and receiving antiretroviral treatment (P < 0.001). Participation barriers include denial of participation by program, recent incarceration, living in a shelter where participation is not allowed and unawareness of eligibility status. In regression analyses, SNAP participation was not significantly associated with improved food security, nutrition status, disease status and health related quality of life (HRQOL). Over half (56%) of the population experienced food insecurity and had inadequate intakes of half of the nutrients assessed. Illicit drug, alcohol and cigarette use were high in this population (31%, 55% and 63% respectively), and affected food security, nutrients intake, disease status and HRQOL. The nutrition education intervention resulted in a trend towards improvements nutrition knowledge, self-efficacy, and readiness to change without impacting nutrition status, disease state and quality of life. Food insecurity and other nutrition related issues, with implications for treatment, management and cost of HIV disease, continue to plague infected individuals living in poverty. More resources, including food and nutrition programs, specifically targeted towards this population are needed to address these issues.

The utility of an integrated qualitative/quantitative data analytic strategy (IQ-DAS) to evaluate the impact of youth development interventions on positive qualitative change in the life course

This study reports one of the first controlled studies to examine the impact of a school based positive youth development program (Lerner, Fisher, & Weinberg, 2000) on promoting qualitative change in life course experiences as a positive intervention outcome. The study built on a recently proposed relational developmental methodological metanarrative (Overton, 1998) and advances in use of qualitative research methods (Denzin & Lincoln, 2000). The study investigated the use the Life Course Interview (Clausen, 1998) and an integrated qualitative and quantitative data analytic strategy (IQDAS) to provide empirical documentation of the impact the Changing Lives Program on qualitative change in positive identity in a multicultural population of troubled youth in an alternative public high school. The psychosocial life course intervention approach used in this study draws its developmental framework from both psychosocial developmental theory (Erikson, 1968) and life course theory (Elder, 1998) and its intervention strategies from the transformative pedagogy of Freire's (1983/1970). Using the 22 participants in the Intervention Condition and the 10 participants in the Control Condition, RMANOVAs found significantly more positive qualitative change in personal identity for program participants relative to the non-intervention control condition. In addition, the 2X2X2X3 mixed design RMANOVA in which Time (pre, post) was the repeated factor and Condition (Intervention versus Control), Gender, and Ethnicity the between group factors, also found significant interactions for the Time by Gender and Time by Ethnicity. Moreover, the directionality of the basic pattern of change was positive for participants of both genders and all three ethnic groups. The pattern of the moderation effects also indicated a marked tendency for participants in the intervention group to characterize their sense of self as more secure and less negative at the end of the their first semester in the intervention, that was stable across both genders and all three ethnicities. The basic differential pattern of an increase in the intervention condition of a positive characterization of sense of self relative to both pre test and relative to the directionality of the movement of the non-intervention controls, was stable across both genders and all three ethnic groups.

The utility of an Integrated Qualitative /Quantitative Data Analytic Strategy (IQ -DAS) to evaluate the impact of youth development interventions to positive qualitative change in the life course

This study reports one of the first controlled studies to examine the impact of a school based positive youth development program (Lerner, Fisher, & Weinberg, 2000) on promoting qualitative change in life course experiences as a positive intervention outcome. The study built on a recently proposed relational developmental methodological metanarrative (Overton, 1998) and advances in use of qualitative research methods (Denzin & Lincoln, 2000). The study investigated the use the Life Course Interview (Clausen, 1998) and an integrated qualitative and quantitative data analytic strategy (IQ-DAS) to provide empirical documentation of the impact the Changing Lives Program on qualitative change in positive identity in a multicultural population of troubled youth in an alternative public high school. The psychosocial life course intervention approach used in this study draws its developmental framework from both psychosocial developmental theory (Erikson, 1968) and life course theory (Elder, 1998) and its intervention strategies from the transformative pedagogy of Freire's (1983/1970). ^ Using the 22 participants in the Intervention Condition and the 10 participants in the Control Condition, RMANOVAs found significantly more positive qualitative change in personal identity for program participants relative to the non-intervention control condition. In addition, the 2X2X2X3 mixed design RMANOVA in which Time (pre, post) was the repeated factor and Condition (Intervention versus Control), Gender, and Ethnicity the between group factors, also found significant interactions for the Time by Gender and Time by Ethnicity. ^ Moreover, the directionality of the basic pattern of change was positive for participants of both genders and all three ethnic groups. The pattern of the moderation effects also indicated a marked tendency for participants in the intervention group to characterize their sense of self as more secure and less negative at the end of the their first semester in the intervention, that was stable across both genders and all three ethnicities. The basic differential pattern of an increase in the intervention condition of a positive characterization of sense of self relative to both pre test and relative to the directionality of the movement of the non-intervention controls, was stable across both genders and all three ethnic groups. ^

Effects of two intervention strategies on the behavior of nurses and nurse students related to advance directives

Advance directives are one mechanism for preserving the rights of individuals to exercise some control over their health care when serious illness may prevent them from direct participation. Nurses, as the health care providers with the closest and most sustained contact with critically ill and dying patients, are positioned to assist patients to plan for future health care needs. Although a majority of nurses favor the concept of advance directives for their patients and for themselves, they have not played a significant role in facilitating advance health care planning with their patients nor implemented advance health care planning for themselves.^ Research has also shown that differing forms of education and counseling increase the completion rates for advance directives in selected populations, mostly the elderly and seriously ill. Not yet developed are effective educational strategies to assist nurses and nurse students to make optimal contributions in assisting their clients' plans for future health care decision-making. This study sought to determine whether specific learning strategies (a) increased the involvement of nurses and nurse students in facilitating advance care planning with patients and (b) increased the percentage of the nurses' and nurse students' own personal advance care planning activities.^ The study compared two learning interventions and two populations, nurses and nurse students. The participants were randomly assigned to one of the two learning interventions, L1 or L2. Participants in L1 received a lecture, discussion and exploration of the forces impacting on advance directive behavior. Participants in L2 received the same intervention components with the additional component of group practice completing advance directives.^ Analysis of the data by chi-square and logistic regression did not support the hypotheses that the practice component would make a difference in the participants' facilitation of advance care planning with patients or in their own personal advance care planning activities. There were significant differences in post-intervention behavior between the nurse and nurse student groups. The nurses in the study did significantly more facilitation of advance care planning with patients and completed significantly more advance care documents than the nurse students post-intervention. However, the nurse students held more post-intervention family discussions than did the nurses. ^

Family and group cognitive behavior therapy: Evaluation of treatment outcome and treatment specificity

This dissertation examined the efficacy of family cognitive behavior treatment (FCBT) and group cognitive behavior treatment (GBCT) for reducing anxiety disorders in children and adolescents using several approaches: clinical significant change, equivalence testing, and analyses of variance. It also examined treatment specificity in terms of targeting family/parents (in FCBT) and peers/group (in GCBT) contextual variables using two main approaches: analyses of variance and structural equation modeling (SEM). The sample consisted of 143 children and their parents who presented to the Child Anxiety and Phobia Program housed within the Child and Family Psychosocial Research Center at Florida International University. Diagnostic interviews and questionnaires were administered to assess youth anxiety. Questionnaires were administered to assess child and parent views of family/parents and peers/group contextual variables. In terms of clinical significant change, results indicated that 84.6% of youth in FCBT and 71.2% of youth in GBCT no longer met diagnostic criteria for their primary/targeted anxiety disorder. In addition, results from analyses of variance indicated that FCBT and GCBT were both efficacious in reducing anxiety disorders in youth across both child and parent ratings. Results using both analyses of variance and structural equation modeling also indicated that there was no meaningful treatment specificity between FCBT and GCBT in terms of either family/parents or peers/group contextual variables. That is, child social skills improved in GCBT in which these skills were targeted and in FCBT in which these skills were not targeted; parenting skills improved in FCBT in which these skills were targeted and in GCBT in which these skills were not targeted. Clinical implications and future research recommendations are discussed.

Impact of a cognitive-behavioral weight control program on body weight, diet quality, and smoking cessation in weight-concerned female smokers

Many people use smoking as a weight control mechanism and do not want to quit because they fear weight gain. These weight-concerned smokers tend to be female, are significantly less likely to stop smoking, are less likely to join smoking cessation programs, and will relapse more often than smokers who are not weight-concerned. Research suggests that a woman’s confidence in her ability to control her weight after quitting relates positively with her intention to quit smoking. Likewise, success in smoking cessation has been associated with increased self-efficacy for weight control. It has been shown that success in changing one negative health behavior may trigger success in changing another, causing a synergistic effect. Recently research has focused on interventions for weight-concerned smokers who are ready to quit smoking. The present study investigated the effect of a cognitive based weight control program on self-efficacy for weight control and the effect on smoking behavior for a group of female weight concerned smokers. Two hundred and sixteen subjects who wanted to lose weight but who were not ready to quit smoking were recruited to participate in a 12-week, cognitive-behavioral weight control program consisting of twelve one-hour sessions. Subjects were randomly assigned to either (1) the weight-control program (intervention group), or (2) the control group. Results of this study demonstrated that subjects in the intervention group increased self-efficacy for weight control, which was associated with improved healthy eating index scores, weight loss, increased self-efficacy for quitting smoking, a decrease in number of cigarettes smoked and triggered positive movement in stage of change towards smoking cessation compared to the control subjects. For these subjects, positive changes in self-efficacy for one behavior (weight control) appeared to have a positive effect on their readiness to change another health behavior (smoking cessation). Further study of the psychological variables that influence weight-concerned female smokers’ decisions to initiate changes in these behaviors and their ability to maintain those changes are warranted.

An investigation of multiple pathways of developmental intervention change

Convergence among treatment, prevention, and developmental intervention approaches has led to the recognition of the need for evaluation models and research designs that employ a full range of evaluation information to provide an empirical basis for enhancing the efficiency, efficacy, and effectiveness of prevention and positive development interventions. This study reports an investigation of a positive youth development program using an Outcome Mediation Cascade (OMC) evaluation model, an integrated model for evaluating the empirical intersection between intervention and developmental processes. The Changing Lives Program (CLP) is a community supported positive youth development intervention implemented in a practice setting as a selective/indicated program for multi-ethnic, multi-problem at risk youth in urban alternative high schools. This study used a Relational Data Analysis integration of quantitative and qualitative data analysis strategies, including the use of both fixed and free response measures and a structural equation modeling approach, to construct and evaluate the hypothesized OMC model. Findings indicated that the hypothesized model fit the data (χ2 (7) = 6.991, p = .43; RMSEA = .00; CFI = 1.00; WRMR = .459). Findings also provided preliminary evidence consistent with the hypothesis that in addition to having effects on targeted positive outcomes, PYD interventions are likely to have progressive cascading effects on untargeted problem outcomes that operate through effects on positive outcomes. Furthermore, the general pattern of findings suggested the need to use methods capable of capturing both quantitative and qualitative change in order to increase the likelihood of identifying more complete theory informed empirically supported models of developmental intervention change processes.

Children as Change Agents: The Influence of Integrating Environmental Education into Home Learning Projects on Families and Community Members

Environmental education focusing on sustainability should be integrated into the elementary curriculum. Learning experiences should include home learning projects that are relevant to the students, involve active participation, and stimulate discussions with their families. Students can use their knowledge and skills to influence the attitudes and behavior of their parents and other community members.