Person:

Emmons, Karen

Background: Economic theory provides the philosophical foundation for valuing costs in judging medical and public health interventions. When evaluating smoking cessation interventions, accurate data on costs are essential for understanding resource consumption. Smoking cessation interventions, for which prior data on resource costs are typically not available, present special challenges. We develop a micro-costing methodology for estimating the real resource costs of outreach motivational interviewing (MI) for smoking cessation and relapse prevention among low-income pregnant women and report results from a randomized controlled trial (RCT) employing the methodology. Methodological standards in cost analysis are necessary for comparison and uniformity in analysis across interventions. Estimating the costs of outreach programs is critical for understanding the economics of reaching underserved and hard-to-reach populations. Methods: Randomized controlled trial (1997-2000) collecting primary cost data for intervention. A sample of 302 low-income pregnant women was recruited from multiple obstetrical sites in the Boston metropolitan area. MI delivered by outreach health nurses vs. usual care (UC), with economic costs as the main outcome measures. Results: The total cost of the MI intervention for 156 participants was $48,672 or $312 per participant. The total cost of $311.8 per participant for the MI intervention compared with a cost of $4.82 per participant for usual care, a difference of $307 ([CI], $289.2 to $322.8). The total fixed costs of the MI were $3,930 and the total variable costs of the MI were $44,710. The total expected program costs for delivering MI to 500 participants would be 147,430, assuming no economies of scale in program delivery. The main cost components of outreach MI were intervention delivery, travel time, scheduling, and training. Conclusion: Grounded in economic theory, this methodology systematically identifies and measures resource utilization, using a process tracking system and calculates both component-specific and total costs of outreach MI. The methodology could help improve collection of accurate data on costs and estimates of the real resource costs of interventions alongside clinical trials and improve the validity and reliability of estimates of resource costs for interventions targeted at underserved and hard-to-reach populations.

Introduction: Intervention studies have been designed to change dietary and lifestyle factors associated with chronic diseases, but self-reported behavior change may incorporate intervention-related bias. This study examines plasma nutrient concentration and correlations with self-reports in the Healthy Directions intervention study. The Healthy Directions intervention studies were designed to increase multivitamin use, fruit and vegetable consumption, and physical activity in working-class, multiethnic populations. Methods: Participants in both studies completed interviewer-administered questionnaires that collected information on sociodemographic and health behavior characteristics. Postintervention blood samples were collected from 209 participants and pooled in pairs within study and within intervention group. Results: We found significantly higher plasma concentrations of retinol (P = .01) and α-carotene (P = .03) in the intervention than in the usual care group. Self-reported multivitamin users had significantly higher concentrations of retinol (P < .001), β-carotene (P = .02), and α-tocopherol (P < .001). Those who reported four or more fruit and vegetable servings per day had higher lutein and zeaxanthin (P = .05) and β-cryptoxanthin (P = .05) concentrations than those consuming fewer. Plasma nutrient concentrations were associated with reported multivitamin use and fruit and vegetable intake, but the correlations were generally higher in the usual care group. Conclusion: We found significant postintervention differences in plasma carotenoid and tocopherol concentrations by treatment group, multivitamin use, and fruit and vegetable intake. However, because we only obtained postintervention blood samples, we were unable to assess preintervention-to-postintervention changes in plasma nutrients. Self-reported intakes were significantly correlated with plasma nutrient concentrations, but the strength of the correlations differed by group, suggesting some intervention-related bias in the questionnaire responses.

Background: Colorectal cancer screening rates are low among poor and disadvantaged patients. Patient navigation has been shown to increase breast and cervical cancer screening rates, but few studies have looked at the potential of patient navigation to increase colorectal cancer screening rates. Methods: The objective was to determine the feasibility and effectiveness of a patient navigator-based intervention to increase colorectal cancer screening rates in community health centers. Patients at the intervention health center who had not been screened for colorectal cancer and were designated as "appropriate for outreach" by their primary care providers received a letter from their provider about the need to be screened and a brochure about colorectal cancer screening. Patient navigators then called patients to discuss screening and to assist patients in obtaining screening. Patients at a demographically similar control health center received usual care. Results: Thirty-one percent of intervention patients were screened at six months, versus nine percent of control patients (p < .001). Conclusion: A patient navigator-based intervention, in combination with a letter from the patient's primary care provider, was associated with an increased rate of colorectal cancer screening at one health center as compared to a demographically similar control health center. Our study adds to an emerging literature supporting the use of patient navigators to increase colorectal cancer screening in diverse populations served by urban health centers.

Background: Obesity and hypertension and their associated health complications disproportionately affect communities of color and people of lower socioeconomic status. Recruitment and retention of these populations in research trials, and retention in weight loss trials has been an ongoing challenge. Methods: Be Fit, Be Well was a pragmatic randomized weight loss and hypertension management trial of patients attending one of three community health centers in Boston, Massachusetts. Participants were asked to complete follow-up assessments every 6-months for two years. We describe challenges encountered and strategies implemented to recruit and retain trial participants over the 24-month intervention. We also identify baseline participant characteristics associated with retention status. Retention strategies included financial incentives, contact between assessment visits, building relationships with health center primary care providers (PCPs) and staff, and putting participant convenience first. Results: Active refusal rates were low with 130 of 2,631 patients refusing participation (4.9%). Of 474 eligible persons completing telephone screening, 365 (77.0%) completed their baseline visit and were randomized into the study. The study population was predominantly non-Hispanic Black (71.2%), female (68.5%) and reported annual household income of less than $35,000 (70.1%). Recruitment strategies included use of passive approval of potential participants by PCPs, use of part-time staff, and outsourcing calls to a call center. A total of 314 (86.0%) people completed the 24-month visit. Retention levels varied across study visits and intervention condition. Most participants completed three or more visits (69.6%), with 205 (56.2%) completing all four. At 24-months, lower retention was observed for males and the intervention condition. Retention strategies included building strong relationships with clinic staff, flexibility in overcoming participant barriers through use of taxi vouchers, night and weekend appointments, and keeping participants engaged via newsletters and social gatherings. Conclusion: We were able to retain 86.0% of participants at 24-months. Recruitment and retention of high percentages of racial/ethnic minorities and lower income samples is possible with planning, coordination with a trusted community setting and staff (e.g. community health centers and RAs), adaptability and building strong relationships. Trial registration Clinicaltrials.gov Identifier: NCT00661817

Background: Partnership for Health-2 (PFH-2) is a web-based version of Partnership for Health, an evidence-based smoking cessation intervention for childhood cancer survivors. This paper describes the PFH-2 intervention and baseline data collection. Methods: 374 childhood and young adult cancer survivors were recruited from five cancer centers and participated in the baseline assessment. At baseline, participants completed measures of their smoking behavior, self-efficacy and stage of change for quitting smoking as well as psychological and environmental factors that could impact their smoking behavior. Results: At baseline, 93% of survivors smoked in the past seven days; however, 89% smoked a pack or less during this period. Forty-seven percent were nicotine dependent, and 55% had made at least one quit attempt in the previous year. Twenty-two percent of survivors were in contemplation for quitting smoking; of those 45% were somewhat or very confident that they could quit within six months. Sixty-three percent were in preparation for quitting smoking; however, they had relatively low levels of confidence that they could quit smoking in the next month. In multivariate analyses, stage of change, self-efficacy, social support for smoking cessation, smoking policy at work and home, fear of cancer recurrence, perceived vulnerability, depression, BMI, and contact with the healthcare system were associated with survivors' smoking behavior. Discussions/Conclusions: A large proportion of the sample was nicotine dependent, yet motivated to quit. Individual- interpersonal- and environmental-level factors were associated with survivors' smoking behavior. Smoking is particularly dangerous for childhood and young adult cancer survivors. This population may benefit from a smoking cessation intervention designed to build self-efficacy and address other known predictors of smoking behavior.

Background. We examined cross-sectional relationships of social resources with health behaviors in adults ages 18–93 years. Methods. Baseline data from a 2009 risk behavior intervention trial were used to measure social resources, physical activity, and fruit and vegetable intake in 2,440 adults. To evaluate associations overall and within 4 age groups (18–34, 35–49, 50–64, and 65–93 y), we used multivariable regression. Results. Mean (SD) age was 49.4 (15) years, physical activity was 346 (304) minutes/week, and fruit and vegetable intake was 3.4 (2.4) servings/day. Mean social resource score was 1.2 (0–4 scale) in 18–34 year olds, 1.1 in all other age groups (P = 0.04). In multivariable models, for each one-point increment in social resource score, the odds ratio for getting 150–959 minutes of physical activity/wk (compared to <150 min/wk) was 3.7 (95% CI 3.0–4.6). Each one-point increment in score was also associated with 29% (95% CI: 23–35%) more servings of fruit and vegetables. We did not observe effect modification by age group. Conclusions:. Although younger adults reported slightly higher resources than older adults, the magnitude of association between social resources and healthful behaviors did not differ between them.

Many skin cancer screenings occur in non-traditional community settings, with the beach being an important setting due to beachgoers being at high risk for skin cancer. This study is a secondary analysis of data from a randomized trial of a skin cancer intervention in which participants (n = 312) had a full-body skin examination by a clinician and received a presumptive diagnosis (abnormal finding, no abnormal finding). Participants’ pursuit of follow-up was assessed post-intervention (n = 283). Analyses examined: (1) participant’s recall of screening results; and (2) whether cognitive and behavioral variables were associated with follow-up being as advised. Just 12% of participants (36/312) did not correctly recall the results of their skin examination. One-third (33%, 93/283) of participants’ follow-up was classified as being not as advised (recommend follow-up not pursued, unadvised follow-up pursued). Among participants whose follow-up was not as advised, 71% (66/93) did not seek recommended care. None of the measured behavioral and cognitive variables were significantly associated with recall of screening examination results or whether follow-up was as advised. Research is needed to determine what factors are associated with follow-up being as advised and to develop messages that increase receipt of advised follow-up care.

Background: To determine whether exposure to a peer-led intervention focused on colorectal cancer (CRC) screening, physical activity, and multi-vitamin intake can lead to increased intentions to be screened for CRC once age eligible among adults under the age of 50. Methods: Participants were residents of low-income housing sites, and CRC screening intentions were assessed at baseline and at follow-up (approximately 2 years later) to determine changes in screening intentions and factors associated with changes in intentions. Results: Participants (n = 692) were 78.4% female, 42.6% Hispanic and 50.8% black. At follow-up, 51% maintained their intention to be screened and 14.6% newly intended to get screened. Individuals newly intending to get screened were more likely to have participated in the intervention, be older, male, and born in Puerto Rico or the United States compared to those who maintained their intention not to get screened (p < 0.05). Conclusion: Exposure to CRC prevention messages before the age of 50 can increase screening intentions among individuals who did not initially intend to get screened. Peer-led interventions to promote CRC screening should include individual less than 50 years of age, as this may contribute to increased screening at the recommended age threshold.

Introduction: Cancer screening rates among Latinas are suboptimal. The objective of this study was to explore how Latinas perceive cancer screening and the use and design of interactive voice response (IVR) messages to prompt scheduling of 1 or more needed screenings. Methods: Seven focus groups were conducted with Latina community health center patients (n = 40) in need of 1 or more cancer screenings: 5 groups were of women in need of 1 cancer screening (breast, cervical, or colorectal), and 2 groups were of women in need of multiple screenings. A bilingual researcher conducted all focus groups in Spanish using a semistructured guide. Focus groups were recorded, transcribed, and translated into English for analysis. Emergent themes were identified by using thematic content analysis. Results: Participants were familiar with cancer screening and viewed it positively, although barriers to screening were identified (unaware overdue for screening, lack of physician referral, lack of insurance or insufficient insurance coverage, embarrassment or fear of screening procedures, fear of screening outcomes). Women needing multiple screenings voiced more concern about screening procedures, whereas women in need of a single screening expressed greater worry about the screening outcome. Participants were receptive to receiving IVR messages and believed that culturally appropriate messages that specified needed screenings while emphasizing the benefit of preventive screening would motivate them to schedule needed screenings. Conclusion: Participants’ receptiveness to IVR messages suggests that these messages may be an acceptable strategy to promote cancer screening among underserved Latina patients. Additional research is needed to determine the effectiveness of IVR messages in promoting completion of cancer screening.

Background: This paper presents the study design, intervention components, and baseline data from Open Doors to Health, a study designed to address social contextual factors in colorectal cancer (CRC) prevention for low-income, racial/ethnic minority populations. Methods: A cluster randomized design with 12 housing sites as the primary sampling units was used: 6 sites were assigned to a "Peer-led plus Screening Access" (PL) condition, and 6 were assigned to "Screening Access only" (SCR) condition. Study-related outcomes were CRC screening, physical activity (measured as mean steps/day), and multivitamin use. Results: At baseline (unweighted sample size = 1554), two-thirds self-reported that they were current with screening recommendations for CRC (corrected for medical records validation, prevalence was 52%), with half having received a colonoscopy (54%); 96% had health insurance. Mean steps per day was 5648 (se mean = 224), and on average 28% of the sample reported regular multivitamin use. Residents reported high levels of social support [mean = 4.40 (se = .03)] and moderately extensive social networks [mean = 2.66 (se = .02)]. Conclusion: Few studies have conducted community-based studies in public housing communities; these data suggest areas for improvement and future opportunities for intervention development and dissemination. Findings from the randomized trial will determine the effectiveness of the intervention on our health-related outcomes as well as inform future avenues of research.