Understanding Factors Influencing Breastfeeding Outcomes in a Sample of African American Women.

OBJECTIVES: Persistent disparities in breastfeeding rates among African American (AA) women compared to other population groups have motivated researchers to understand factors influencing breastfeeding choices using a variety of methods. Quantitative surveys are more commonly reported, however, qualitative work that amplifies voices of AA women is limited. METHODS: Participants were recruited from a randomized controlled feasibility trial focused on breastfeeding support for AA women in Detroit, MI. Thirteen women were enrolled in the qualitative portion of the study described here. Using the Socioecological model (SEM) as the theoretical foundation, semi-structured qualitative interviews were conducted to explore perceived facilitators and barriers to breastfeeding. Interviews were digitally recorded, transcribed, and analyzed using Theoretical thematic analysis. RESULTS: Women reported factors ranging from micro to macro SEM levels that discouraged or reinforced breastfeeding. Key challenges included breastfeeding-related discouragement issues, including factors that decreased confidence and led women to terminate breastfeeding (e.g., problems with latching, pumping, lack of comfort with breastfeeding in public, and work constraints). Facilitators included perceived mother and infant benefits, perseverance/commitment/self-motivation, pumping ability, and social support. Participant suggestions for expanding breastfeeding promotion and support included: (1) tangible, immediate, and proactive support; (2) positive non-judgmental support; (3) "milk supply" and "use of pump" education; and (4) self-motivation/willpower/perseverance. CONCLUSIONS FOR PRACTICE: Despite the identification of common facilitators, findings reveal AA women face many obstacles to meeting breastfeeding recommendations. Collaborative discussions between women and healthcare providers focused on suggestions provided by AA women should be encouraged.

Sleep Duration and Quality in Relation to Fruit and Vegetable Intake of US Young Adults: a Secondary Analysis.

BACKGROUND: Sleep is gaining recognition as a determinant of diet, yet this relationship remains understudied among young adults. We sought to examine how sleep duration and quality were related to fruit and vegetable (FV) intake within a diverse sample of young adults. METHODS: Participants (n = 1444) ages 21-30 (69% women, 15% African American, 35% full or part time in college) consuming < 5 servings/day of FV (eligibility criteria) completed a baseline survey to enroll in a randomized online FV intervention. Sleep questions included duration, perceived sleep quality, time to fall asleep, and insomnia symptoms. Overall and gender-stratified linear regression models compared average daily FV intake and sleep characteristics, adjusting for confounders. RESULTS: One-third (32%) of the participants reported < 7 h of sleep per night, and 36% noted insomnia symptoms ≥ 3 times per week. Women, a BMI > 30, African American race/ethnicity, less education, unemployment, higher depression, and stress were related to suboptimal sleep. Bivariate analyses showed that better sleep was associated with higher FV intake. After accounting for confounders, men with better sleep quality and shorter time to fall asleep had higher intakes of FV (1.12 serving/day difference in highest versus lowest quality [95% CI 0.48, 1.75] and a 0.52 serving/day higher intake difference for shortest versus longest fall asleep time [95% CI 0.90, 0.15], respectively). CONCLUSION: Sleep was highly prevalent in a diverse sample of community-based young adults and may contribute to lower FV intake among men. These associations highlight young adulthood as an important period for promoting healthy sleep habits.

Developing Machine Learning Models for Behavioral Coding.

OBJECTIVE: The goal of this research is to develop a machine learning supervised classification model to automatically code clinical encounter transcripts using a behavioral code scheme. METHODS: We first evaluated the efficacy of eight state-of-the-art machine learning classification models to recognize patient-provider communication behaviors operationalized by the motivational interviewing framework. Data were collected during the course of a single weight loss intervention session with 37 African American adolescents and their caregivers. We then tested the transferability of the model to a novel treatment context, 80 patient-provider interactions during routine human immunodeficiency virus (HIV) clinic visits. RESULTS: Of the eight models tested, the support vector machine model demonstrated the best performance, achieving a .680 F1-score (a function of model precision and recall) in adolescent and .639 in caregiver sessions. Adding semantic and contextual features improved accuracy with 75.1% of utterances in adolescent and 73.8% in caregiver sessions correctly coded. With no modification, the model correctly classified 72.0% of patient-provider utterances in HIV clinical encounters with reliability comparable to human coders (k = .639). CONCLUSIONS: The development of a validated approach for automatic behavioral coding offers an efficient alternative to traditional, resource-intensive methods with the potential to dramatically accelerate the pace of outcomes-oriented behavioral research. The knowledge gained from computer-driven behavioral research can inform clinical practice by providing clinicians with empirically supported communication strategies to tailor their conversations with patients. Lastly, automatic behavioral coding is a critical first step toward fully automated eHealth/mHealth (electronic/mobile Health) behavioral interventions.

Overweight adolescents and asthma: Revealing motivations and challenges with adolescent-provider communication.

OBJECTIVE: Among teens with asthma, challenges of disease management may be greater in those with a body mass index (BMI) >85th percentile compared to youth within the parameters for normal weight-for-age. This mixed-methods study assessed teens' awareness of the link between weight and asthma management, and perspectives on how medical providers might open a discussion about managing weight. METHOD: Teens aged 13-18, having BMI >85 percentile and chronic asthma, identified using health system databases and a staff email message board, were invited to complete a semi-structured, in-depth phone interview. Interviews were audio taped, transcribed, and qualitatively analyzed, using the Framework Method. Responses were summarized and themes identified. Descriptive summaries were generated for a 16-item survey of weight conversation starters. RESULTS: Of 35 teens interviewed, 24 (69%) were girls, 11 (31%) boys, 20 (63%) African-American. All teens reported having "the weight conversation" with their doctors, and preferred that parents be present. Half knew from their doctor about the link between being overweight and asthma, others knew from personal experience. Nearly all expressed the importance of providers initiating a weight management conversation. Most preferred conversation starters that recognized challenges and included parents' participation in weight management; least liked referred to "carrying around too much weight." CONCLUSIONS: Most teens responded favorably to initiating weight loss if it impacted asthma management, valued their provider addressing weight and family participation in weight management efforts. Adolescents' views enhance program development fostering more effective communication targeting weight improvement within the overall asthma management plan.

Providers' Experiences with a Melanoma Web-Based Course: a Discussion on Barriers and Intentions.

Primary care visits provide an opportunity for skin examinations with the potential to reduce melanoma mortality. The INFORMED (INternet curriculum FOR Melanoma Early Detection) Group developed a Web-based curriculum to improve primary care providers' (PCPs') skin cancer detection skills. This study details feedback obtained from participant focus groups, including the feasibility of implementing in other PCP practices. Practicing PCPs at Henry Ford Health System and Kaiser Permanente Northern California completed the curriculum. Feedback sessions were conducted with standardized questions focusing on four domains: (1) overall impressions of the curriculum, (2) recommendations for improvement, (3) current skin examination practices, and (4) suggestions for increasing skin screening by PCPs. Discussions at each site were audio recorded, transcribed verbatim, and de-identified. Providers (N = 54) had a positive impression of the Web-based curriculum, with suggestions to provide offline teaching aids and request assistance. Despite having improved confidence in diagnosing malignant lesions, many providers felt a lack of confidence in performing the screening and time constraints affected their current practices, as did institutional constraints. Providers intended to increase discussion with patients about skin cancer. The accessibility, effectiveness, and popularity of the curriculum indicate potential for implementation in the primary care setting. Participating providers noted that institutional barriers remain which must be addressed for successful dissemination and implementation.

Comparison of Interviewer-Administered and Automated Self-Administered 24-Hour Dietary Recalls in 3 Diverse Integrated Health Systems.

Twenty-four-hour dietary recalls provide high-quality intake data but have been prohibitively expensive for large epidemiologic studies. This study's goal was to assess whether the web-based Automated Self-Administered 24-Hour Recall (ASA24) performs similarly enough to the standard interviewer-administered, Automated Multiple-Pass Method (AMPM) 24-hour dietary recall to be considered a viable alternative. In 2010-2011, 1,081 adults from 3 integrated health systems in Detroit, Michigan; Marshfield, Wisconsin; and Kaiser-Permanente Northern California participated in a field trial. A quota design ensured a diverse sample by sex, age, and race/ethnicity. Each participant was asked to complete 2 recalls and was randomly assigned to 1 of 4 protocols differing by type of recall and administration order. For energy, the mean intakes were 2,425 versus 2,374 kcal for men and 1,876 versus 1,906 kcal for women by AMPM and ASA24, respectively. Of 20 nutrients/food groups analyzed and controlling for false discovery rate, 87% were judged equivalent at the 20% bound. ASA24 was preferred over AMPM by 70% of the respondents. Attrition was lower in the ASA24/AMPM study group than in the AMPM/ASA24 group, and it was lower in the ASA24/ASA24 group than in the AMPM/AMPM group. ASA24 offers the potential to collect high-quality dietary intake information at low cost with less attrition.

Patients' and family members' views on patient-centered communication during cancer care.

OBJECTIVES: To explore patients' and family members' views on communication during cancer care and to identify those aspects of clinician-patient communication which were most important to patients and family members. METHODS: We conducted a secondary data analysis of qualitative data from 137 patients with cancer and family members of patients with cancer. We used a modified version of the constant comparative method and coding paradigm of grounded theory. RESULTS: Patients want sensitive, caring clinicians who provide information that they need, when they need it, in a way that they can understand; who listen and respond to questions and concerns, and who attempt to understand the patient's experience. Effective information exchange and a positive interpersonal relationship with the clinician were of fundamental importance to patients and family members. These were interrelated; for instance, failure to provide information a patient needed could damage the relationship, whereas excellent listening could foster the relationship. Information exchange and relationship were also integral to decision-making, managing uncertainty, responding to emotions, and self-management. Clinicians who were responsive to patients' needs beyond the immediate medical encounter were valued. CONCLUSIONS: The complexity of cancer care today suggests that efforts to improve communication must be multilevel, acknowledging and addressing patient, clinician, organizational and policy barriers, and facilitators. Measurement tools are needed to assess cancer patients' and family members' experiences with communication over the course of cancer care to provide meaningful, actionable feedback to those seeking to optimize their effectiveness in communicating with patients with cancer.

Understanding the Impact of Perceived Social Support for Breastfeeding Among African American Women: Results From the Mama Bear Feasibility Trial.

PURPOSE: Perceived Social Support (PSS) can impact breastfeeding behaviors, and a lack of PSS potentially contributes to disparities in breastfeeding rates for African American women (AA). Objectives were to describe PSS at two timepoints and test associations between PSS and breastfeeding intensity for AA. METHODS: Data are from a feasibility trial of breastfeeding support among AA. The Hughes Breastfeeding Support Scale was used to measure PSS (Emotional, Informational, Tangible; total range = 30-120) in pregnancy (T1, n = 32) and early postpartum (T2, n = 31). Scale means were compared with t-tests. Associations between PSS at T1 and breastfeeding intensity (ie, quantitative measure of breastfeeding) were assessed with linear regression. RESULTS: Total PSS (mean ± SE) was high at both time points (T1 = 90.5 ± 4.8; T2 = 92.8 ± 3.1). At T2, older participants or those living with a partner had higher total PSS scores compared to those younger or living alone. Emotional PSS was significantly higher at T2 than T1 with no differences in tangible or informational PSS over time. Mixed-feeding, exclusive breastfeeding, and exclusive formula feeding was distributed at 39%, 32%, and 29%, respectively. Total PSS was not associated with breastfeeding intensity. CONCLUSION: Women reported high levels of social support, and emotional PSS increased over time in this small sample of AA. PSS and sources of PSS are understudied, especially among AA, and future studies should explore quantitative methods to assess PSS. The results of such assessments can then be used to design breastfeeding support interventions.

Breastfeeding Self-Efficacy as a Predictor of Breastfeeding Intensity Among African American Women in the Mama Bear Feasibility Trial.

Background: Improving breastfeeding rates among African American (AA) families is an important public health goal. Breastfeeding self-efficacy, a known predictor of breastfeeding behavior, has seldom been assessed among AAs, in relation to breastfeeding intensity (% breastfeeding relative to total feeding) or as a protective factor in combating the historical breastfeeding challenges faced by people of color. We aimed to test the association between breastfeeding self-efficacy assessed during pregnancy and breastfeeding intensity assessed in the early postpartum period. Methods: This was a secondary data analysis of a randomized controlled feasibility trial of breastfeeding support and postpartum weight management. AA women were recruited during pregnancy from a prenatal clinic in Detroit, MI. Data presented, in this study, were collected at enrollment (n = 50) and ∼6 weeks postpartum (n = 31). Linear regression models were used, adjusting for potential confounders. Results: There were no differences in breastfeeding intensity by study arm; data are from all women with complete data on targeted variables. Age ranged from 18 to 43 years, 52% were Women, Infant's, and Children program enrollees, and 62% had ≥ some college. Breastfeeding self-efficacy during pregnancy was a significant predictor of breastfeeding intensity in the early postpartum period (ß = 0.125, p < 0.05) with only slight attenuation in the fully adjusted model (ß = 0.123, p < 0.05). Implications for Practice: Our results confirm that self-efficacy is an important predictor of breastfeeding practice. Furthermore, the simple act of assessing breastfeeding self-efficacy permits an opportunity for women to reflect on breastfeeding possibilities, and can inform individualized confidence-building interventions to improve the disproportionately low breastfeeding rates among AAs. Clinical Trial Registration number NCT03480048.

Changes in fruit and vegetable consumption in relation to changes in sleep characteristics over a 3-month period among young adults.

OBJECTIVE: To evaluate whether increases in fruit and vegetable (FV) consumption were associated with concomitant changes in insomnia symptoms, sleep duration, and quality. DESIGN: Secondary longitudinal analysis of a randomized trial, baseline to 3 months. SETTING: Integrated health care systems in Detroit, Michigan and Danville, Pennsylvania. PARTICIPANTS: About 1165 young adults who were low consumers of FV (<3 servings/day) at baseline. INTERVENTION: Online 3-arm program designed to increase FV consumption. MEASUREMENTS: We categorized FV changes into 4 categories: no change or decrease, 1 serving increase, 2 serving increase, and 3 or more serving increase. We then compared the changes in chronic insomnia classification (yes or no), sleep duration, quality, and time to fall asleep (all self-reported) across the FV change categories. Analyses were both overall and stratified by gender, adjusting for potential confounders (depression, physical activity, education, children, and study site). RESULTS: Average age ± SD was 26 ± 2.8 years (71% women). At 3-month follow-up, participants on average increased FV intake by 1.2 ± 1.4 servings. Women who increased FV intake by 3+ servings showed improvements in insomnia symptoms (2-fold higher odds of improvement; 95% CI 1.1 to 3.6), sleep quality (0.2-point higher sleep quality score; 95% CI -0.01, 0.3), and time to fall asleep (4.2 minutes; 95% CI -8, 0) compared to women who did not change or decreased their FV intake. Associations were not as apparent among men. CONCLUSION: Young women with low consumption of FV may experience improvements in insomnia-related sleep difficulties by increasing their consumption of FV.

Influence of a One-Time Web-Based Provider Intervention on Patient-Reported Outcomes After the Well-Child Visit: A Feasibility Study.

PURPOSE: Patient-centered care promotes positive health outcomes in pediatrics. We created a provider-focused intervention and implemented it in a pragmatic clustered randomized controlled trial to improve health-related quality of life (HRQOL) among pediatric patients. METHODS: A one-time (1-1.5-hour) webinar focusing on patient-centered care and motivational interviewing, using obesity screening as an example, was developed. Pediatric providers were recruited and randomized to either intervention (webinar) or control (usual care) arms. All well-child visits to these providers for a period of up to 5 months following webinar completion (or study enrollment for controls) were identified, and these family/patients were invited to complete a survey to assess HRQOL postvisit. Reported outcomes were compared between intervention and control participants using clustered t-tests, chi-squared tests and multiple linear regression models. RESULTS: We recruited 20 providers (10 intervention, 10 control) to the study; 469 parents/guardians and 235 eligible children seeing these providers completed the postvisit survey. Parents/guardians of 8-12-year-old children in the intervention group reported higher school functioning compared to controls (83.5 vs 75.8; P=0.023). There were no other differences in children's HRQOL between intervention and control groups. CONCLUSIONS: A one-time, web-based provider intervention is feasible to implement in pediatrics. Modest evidence, requiring further study, indicates that instructing providers on patient-centered care in the well-child visit may improve aspects of pediatric HRQOL (ie, school functioning) compared to usual care. However, this was a brief intervention, with multiple outcomes tested and no evaluation of pre- and postintervention provider knowledge, thus additional study is needed.

Pilot study of a brief provider and EMR-based intervention for overweight teens with asthma.

INTRODUCTION: Asthma-related morbidity is increased in overweight patients, yet providers are given little guidance on how to discuss weight and asthma management with overweight teens. OBJECTIVE: We piloted an electronic medical record (EMR)-based tailored discussion guide (TDG) and a brief provider training, to address weight management in overweight teens with asthma. The primary outcome was intervention impact on patient-reported asthma outcomes (e.g., asthma control and morbidity). Secondary outcomes included change in BMI, patient-centeredness, and change in healthy behaviors. METHODS: Teens aged 13-18 years with persistent asthma and a body mass index ≥ 85th percentile for their age and sex were eligible. Parents of eligible teens were contacted before an upcoming appointment to allow teen enrollment during the clinic visit. Providers reviewed Motivational Interviewing (MI) concepts and were trained in the TDG for support of conversations around weight and asthma management. Measures included asthma outcomes retrieved from the EMR at 6- and 12-month post-baseline, teen impressions of patient-provider communication at 6-week post-enrollment, and teen report of healthy behaviors at 6- and 12-month post-baseline. RESULTS: Of 44 teens enrolled (77% African-American, 63% female), mean BMI for intervention (n=25) and control groups (n=19) at baseline were similar. Thirty participants (68%) completed a 6-week questionnaire. Compared to controls, at 6 months, intervention teens reported fewer days of limited activity and "uncontrolled asthma," but at 12 months, only restricted activity remained lower, and BMI was not reduced. Intervention teens reported clinic visits that were more patient-centered than controls, including discussion of asthma treatment options with provider, feeling ready to follow an asthma treatment routine, and receiving helpful tips about reaching a healthy weight. The healthy behavior "dinner with family" showed improvement for intervention teens at 6 and 12 months. The feasibility study also revealed a need to improve recruitment strategies and to streamline intervention delivery. CONCLUSION: Modest improvements in patient-reported asthma outcomes and health behaviors were observed. There was strong evidence that the TDG supports provider discussion of weight and asthma to create a more patient-centered conversation from the perspective of participating teens. Challenges to recruitment and clinic adaptation must be addressed before advancing to a full-scale trial. TRIAL REGISTRATION: NCT02575326 Teen Asthma Control Encouraging a Healthier Lifestyle, www.cllinicaltrials.gov.

A randomized clinical trial evaluating online interventions to improve fruit and vegetable consumption.

OBJECTIVES: We assessed change in fruit and vegetable intake in a population-based sample, comparing an online untailored program (arm 1) with a tailored behavioral intervention (arm 2) and with a tailored behavioral intervention plus motivational interviewing-based counseling via e-mail (arm 3). METHODS: We conducted a randomized controlled intervention trial, enrolling members aged 21 to 65 years from 5 health plans in Seattle, Washington; Denver, Colorado; Minneapolis, Minnesota; Detroit, Michigan; and Atlanta, Georgia. Participants reported fruit and vegetable intake at baseline and at 3, 6, and 12 months. We assessed mean change in fruit and vegetable servings per day at 12 months after baseline, using a validated self-report fruit and vegetable food frequency questionnaire. RESULTS: Of 2540 trial participants, 80% were followed up at 12 months. Overall baseline mean fruit and vegetable intake was 4.4 servings per day. Average servings increased by more than 2 servings across all study arms (P<.001), with the greatest increase (+2.8 servings) among participants of arm 3 (P=.05, compared with control). Overall program satisfaction was high. CONCLUSIONS: This online nutritional intervention was well received, convenient, easy to disseminate, and associated with sustained dietary change. Such programs have promise as population-based dietary interventions.

Body mass index measurement and obesity prevalence in ten U.S. health plans.

OBJECTIVE: The objective of this study was to examine the frequency of body mass index (BMI) measurement before the implementation of two new Healthcare Effectiveness Data and Information Set (HEDIS) performance measures for obesity that require U.S. health plans to annually report the frequency of BMI and BMI percentile measurement among all adults and children who had at least one outpatient visit during the past two years. DESIGN: Cross-sectional study. SETTING: A consortium of ten U.S. health plans and care delivery systems from the Health Maintenance Organization Research Network, which together provide care to more than 6.5 million adults and children. PARTICIPANTS: Children and adults, age 2 years and older, who were continuously enrolled in one of ten U.S. health plans for at least one full year from 2005 to 2006. METHODS: We extracted available anthropometric data for 3.7 million adults and 1.2 million children with at least one visit captured from ten electronic medical record databases from 2005 to 2006. RESULTS: We found that the availability of BMI measurements for adults ranged widely across health plans from 28% to 88%, and availability of BMI percentiles for children ranged from 21% to 81%. Among adults and children with BMI measures in these ten health plans, the overall prevalence of overweight and obesity were very similar to those reported in the 2005 to 2006 U.S. national surveys that used measured heights and weights. CONCLUSION: The newly approved HEDIS performance measures likely represent an important step in addressing the quality of obesity care in the United States. The current study demonstrates that these HEDIS measures are achievable, especially among health plans that have implemented electronic medical records. Future research should assess the relationship between BMI assessment, provider counseling and treatment practices, and long-term changes in obesity rates among different population groups.

Engagement and retention: measuring breadth and depth of participant use of an online intervention.

BACKGROUND: The Internet provides us with tools (user metrics or paradata) to evaluate how users interact with online interventions. Analysis of these paradata can lead to design improvements. OBJECTIVE: The objective was to explore the qualities of online participant engagement in an online intervention. We analyzed the paradata in a randomized controlled trial of alternative versions of an online intervention designed to promote consumption of fruit and vegetables. METHODS: Volunteers were randomized to 1 of 3 study arms involving several online sessions. We created 2 indirect measures of breadth and depth to measure different dimensions and dynamics of program engagement based on factor analysis of paradata measures of Web pages visited and time spent online with the intervention materials. Multiple regression was used to assess influence of engagement on retention and change in dietary intake. RESULTS: Baseline surveys were completed by 2513 enrolled participants. Of these, 86.3% (n = 2168) completed the follow-up surveys at 3 months, 79.6% (n = 2027) at 6 months, and 79.4% (n = 1995) at 12 months. The 2 tailored intervention arms exhibited significantly more engagement than the untailored arm (P < .01). Breadth and depth measures of engagement were significantly associated with completion of follow-up surveys (odds ratios [OR] = 4.11 and 2.12, respectively, both P values < .001). The breadth measure of engagement was also significantly positively associated with a key study outcome, the mean increase in fruit and vegetable consumption (P < .001). CONCLUSIONS: By exploring participants' exposures to online interventions, paradata are valuable in explaining the effects of tailoring in increasing participant engagement in the intervention. Controlling for intervention arm, greater engagement is also associated with retention of participants and positive change in a key outcome of the intervention, dietary change. This paper demonstrates the utility of paradata capture and analysis for evaluating online health interventions. TRIAL REGISTRATION: NCT00169312; http://clinicaltrials.gov/ct2/show/NCT00169312 (Archived by WebCite at http://www.webcitation.org/5u8sSr0Ty).

Recruitment to a randomized web-based nutritional intervention trial: characteristics of participants compared to non-participants.

BACKGROUND: Web-based behavioral programs efficiently disseminate health information to a broad population, and online tailoring may increase their effectiveness. While the number of Internet-based behavioral interventions has grown in the last several years, additional information is needed to understand the characteristics of subjects who enroll in these interventions, relative to those subjects who are invited to enroll. OBJECTIVE: The aim of the study was to compare the characteristics of participants who enrolled in an online dietary intervention trial (MENU) with those who were invited but chose not to participate, in order to better understand how these groups differ. METHODS: The MENU trial was conducted among five health plans participating in the HMO Cancer Research Network in collaboration with the University of Michigan Center for Health Communication Research. Approximately 6000 health plan members per site, between the ages of 21 and 65, and stratified by gender with oversampling of minority populations, were randomly selected for recruitment and were mailed an invitation letter containing website information and a US$2 bill with the promise of US$20 for completing follow-up surveys. Administrative and area-based data using geocoding along with baseline survey data were used to compare invitees (HMO members sent the introductory letter), responders (those who entered a study ID on the website), and enrollees (those who completed the enrollment process). Generalized estimating equation multivariate and logistic regression models were used to assess predictors of response and enrollment. RESULTS: Of 28,460 members invited to participate, 4270 (15.0%) accessed the website. Of the eligible responders, 2540 (8.9%) completed the consent form and baseline survey and were enrolled and randomized. The odds of responding were 10% lower for every decade of increased age (P < .001), while the likelihood of enrolling was 10% higher for every decade increase in age (P < .001). Women were more likely to respond and to enroll (P < .001). Those living in a census tract associated with higher education levels were more likely to respond and enroll, as well as those residing in tracts with higher income (P < .001). With a 22% (n = 566) enrollment rate for African Americans and 8% (n = 192) for Hispanics, the enrolled sample was more racially and ethnically diverse than the background sampling frame. CONCLUSIONS: Relative to members invited to participate in the Internet-based intervention, those who enrolled were more likely to be older and live in census tracts associated with higher socioeconomic status. While oversampling of minority health plan members generated an enrolled sample that was more racially and ethnically diverse than the overall health plan population, additional research is needed to better understand methods that will expand the penetration of Internet interventions into more socioeconomically diverse populations. TRIAL REGISTRATION: Clinicaltrials.gov NCT00169312; http://clinicaltrials.gov/ct2/show/NCT00169312 (Archived by WebCite at http://www.webcitation.org/5jB50xSfU).

Deep Neural Architectures for Discourse Segmentation in E-Mail Based Behavioral Interventions.

Communication science approaches to develop effective behavior interventions, such as motivational interviewing (MI), are limited by traditional qualitative coding of communication exchanges, a very resource-intensive and time-consuming process. This study focuses on the analysis of e-Coaching sessions, behavior interventions delivered via email and grounded in the principles of MI. A critical step towards automated qualitative coding of e-Coaching sessions is segmentation of emails into fragments that correspond to MI behaviors. This study frames email segmentation task as a classification problem and utilizes word and punctuation mark embeddings in conjunction with part-of-speech features to address it. We evaluated the performance of conditional random fields (CRF) as well as multi-layer perceptron (MLP), bi-directional recurrent neural network (BRNN) and convolutional recurrent neural network (CRNN) for the task of email segmentation. Our results indicate that CRNN outperforms CRF, MLP and BRNN achieving 0.989 weighted macro-averaged F1-measure and 0.825 F1-measure for new segment detection.

5-2-1-0 Lifestyle risk factors predict obesity in Millennials.

The Making Effective Nutritional Choices Generation Y (MENU GenY) study is a web-based intervention trial aimed at improving food choices in those aged 21-30 years. We report baseline levels of the 5-2-1-0 healthy lifestyle patterns to predict a body mass index (BMI) ≥30 vs <30 kg m-2 . Overall, 1674 young adults (69% female) from two large health systems enroled and completed an online survey asking questions about lifestyle habits. A multivariable binary logistic regression model was utilized to predict a BMI ≥30 while controlling for known predictors of obesity. Consuming >3 daily servings of fruits/vegetables (odds ratio, OR = 0.90, 95% confidence interval, CI = 0.81, 0.99), and reporting >2.5 hours/week of vigorous physical activity (OR = 0.93, 95% CI = 0.89-0.96, P < 0.001) was associated with a BMI <30. Conversely, time sitting (OR = 1.07, 95% CI = 1.04, 1.11) and consuming sugar-sweetened beverages (OR = 1.08, 95% CI = 1.00, 1.15) were related to a BMI ≥30. In this cohort of 20-30-year-olds, we observed a consistent relationship between obesity and the 5-2-1-0 healthy lifestyle patterns previously reported among children and adolescents.

Identifying Effective Motivational Interviewing Communication Sequences Using Automated Pattern Analysis.

Motivational Interviewing (MI) is an evidence-based strategy for communicating with patients about behavior change. Although there is strong empirical evidence linking "MI-consistent" counselor behaviors and patient motivational statements (i.e., "change talk"), the specific counselor communication behaviors effective for eliciting patient change talk vary by treatment context and, thus, are a subject of ongoing research. An integral part of this research is the sequential analysis of pre-coded MI transcripts. In this paper, we evaluate the empirical effectiveness of the Hidden Markov Model, a probabilistic generative model for sequence data, for modeling sequences of behavior codes and closed frequent pattern mining, a method to identify frequently occurring sequential patterns of behavior codes in MI communication sequences to inform MI practice. We conducted experiments with 1,360 communication sequences from 37 transcribed audio recordings of weight loss counseling sessions with African-American adolescents with obesity and their caregivers. Transcripts had been previously annotated with patient-counselor behavior codes using a specialized codebook. Empirical results indicate that Hidden Markov Model and closed frequent pattern mining techniques can identify counselor communication strategies that are effective at eliciting patients' motivational statements to guide clinical practice.

Effect of incentives and mailing features on online health program enrollment.

BACKGROUND: With the growing use of Internet-based interventions, strategies are needed to encourage broader participation. This study examined the effects of combinations of monetary incentives and mailing characteristics on enrollment, retention, and cost effectiveness for an online health program. METHODS: In 2004, a recruitment letter was mailed to randomly selected Midwestern integrated health system members aged 21-65 and stratified by gender and race/ethnicity; recipients were randomly pre-assigned to one of 24 combinations of incentives and various mailing characteristics. Enrollment and 3-month retention rates were measured by completion of online surveys. Analysis, completed in 2005, compared enrollment and retention factors using t tests and chi-square tests. Multivariate logistic regression modeling assessed the probability of enrollment and retention. RESULTS: Of 12,289 subjects, 531 (4.3%) enrolled online, ranging from 1% to 11% by incentive combination. Highest enrollment occurred with unconditional incentives, and responses varied by gender. Retention rates ranged from 0% to 100%, with highest retention linked to higher-value incentives. The combination of a $2 bill prepaid incentive and the promise of $20 for retention (10% enrollment and 71% retention) was optimal, considering per-subject recruitment costs ($32 enrollment, $70 retention) and equivalent enrollment by gender and race/ethnicity. CONCLUSIONS: Cash incentives improved enrollment in an online health program. Men and women responded differently to mailing characteristics and incentives. Including a small prepaid monetary incentive ($2 or $5) and revealing the higher promised-retention incentive was cost effective and boosted enrollment.