Nanhua Zhang

TAILORING A FRUIT AND VEGETABLE INTERVENTION ON ETHNIC IDENTITY: RESULTS OF A RANDOMIZED STUDY

OBJECTIVE Many targeted interventions have been developed and tested with African Americans (AA); however, AAs are a highly heterogeneous group. One characteristic that varies across AAs is Ethnic Identity (EI). Little research has been conducted on how to incorporate EI into the design of health messages and programs. DESIGN We tested whether tailoring a print-based fruit and vegetable (F & V) intervention on EI would enhance program impact. AA adults were recruited from two integrated healthcare delivery systems and then randomized to receive three newsletters focused on F & V behavior change over three months. One set of newsletters was tailored only on demographic and social cognitive variables (control condition), whereas the other (experimental condition) was additionally tailored on EI. MAIN OUTCOME MEASURES The primary outcome for the study was F & V intake, assessed at baseline and three months later using the composite of two brief self-report frequency measures. RESULTS A total of 560 eligible participants were enrolled, of which 468 provided complete 3-month follow-up data. The experimental group increased their daily mean F & V intake by 1.1 servings compared to .8 servings in the control group (p = .13). Afrocentric experimental group participants showed a 1.4 increase in F & V servings per day compared to a .43 servings per day increase among Afrocentric controls (p < .05). CONCLUSIONS Although the overall between-group effects were not significant, tailoring dietary messages on ethnic identity may improve intervention impact for some AA subgroups.

Increasing donor designation through black churches: results of a randomized trial

Context African Americans are disproportionately represented among those awaiting a transplant, but many are reluctant to donate their organs. Objective To test the effectiveness of using lay health advisors to increase organ donation among church members. Design Churches were pair-matched by average estimated income and size and then randomized to 1 of 2 interventions: one addressing organ donation and the other addressing increasing consumption of fruits and vegetables. Setting Twenty-two African American churches in Southeast Michigan. Patients or Other Participants Church members were trained to serve as lay health advisors (called peer leaders). Interventions Peer leaders conducted organ donation discussions with church groups and showed a DVD created for this program that was tailored to African American churches. Main Outcome Measures The primary outcome was verified registration in the states donor registry. Participants also completed pre/post questionnaires regarding their attitudes about organ donation. Results Once clustering, baseline value, and demographics were adjusted for, the intervention and comparison groups did not differ on any of the 3 attitude scales on the posttest. In logistic regression analysis, with baseline donation status, demographics, and church clustering controlled for, the odds of self-reported enrollment at 1-year posttest did not differ by condition (odds ratio, 1.23; 95% CI, 0.87–1.72). A total of 211 enrollments in the state registry from participating churches were verified. Of these, 163 were from intervention churches and 48 were from comparison churches. Conclusions Use of lay health advisors through black churches can increase minority enrollment in a donor registry even absent change in attitudes.

Development of a Scale to Measure African American Attitudes toward Organ Donation

African American attitudes toward organ donation differ from other racial and ethnic groups. However, existing measures of organ donation attitudes do not adequately address ethnic identity and cultural factors. We examined the psychometric properties of a new 18-item organ donation scale among 1225 members of 21 African American churches in Southeast Michigan. We identified three factors: (1) Barriers; (2) Family/Race Benefits; and (3) Altruism. More positive donation attitudes on each subscale were observed for individuals who reported being enrolled as a donor. Among individuals not enrolled, higher scores were observed on scales two and three for those with stronger intentions to enroll.

Ethnic Identity, Questionnaire Content, and the Dilemma of Race Matching in Surveys of African Americans by African American Interviewers

We used data from two telephone-administered health surveys to explore African Americans’ preferences for interviewer race. The first survey utilized African American interviewers to assess ethnic identity and aspects of healthy eating among 617 African American adults. In the second survey, interviewers of varying races queried 534 African American adults about their motivations to eat healthier. The motivation survey contained almost no racial content, whereas 40% of the ethnic identity survey assessed racial content. Using only ethnic identity survey data, we found that respondents with Afrocentric or Black American identity components were more likely to prefer African American interviewers than respondents with solely assimilated, bicultural, or multicultural identity components. Ethnic identity survey respondents were also more likely to prefer racially/ethnically matched interviewers than motivation survey respondents. Ethnic identity respondents with a college or graduate degree reported lower hypothetical comfort with a white interviewer than respondents with a high school education.

A Pseudo-Bayesian Shrinkage Approach to Regression with Missing Covariates

We consider the linear regression of outcome Y on regressors W and Z with some values of W missing, when our main interest is the effect of Z on Y, controlling for W. Three common approaches to regression with missing covariates are (i) complete-case analysis (CC), which discards the incomplete cases, and (ii) ignorable likelihood methods, which base inference on the likelihood based on the observed data, assuming the missing data are missing at random (Rubin, 1976b), and (iii) nonignorable modeling, which posits a joint distribution of the variables and missing data indicators. Another simple practical approach that has not received much theoretical attention is to drop the regressor variables containing missing values from the regression modeling (DV, for drop variables). DV does not lead to bias when either (i) the regression coefficient of W is zero or (ii) W and Z are uncorrelated. We propose a pseudo-Bayesian approach for regression with missing covariates that compromises between the CC and DV estimates, exploiting information in the incomplete cases when the data support DV assumptions. We illustrate favorable properties of the method by simulation, and apply the proposed method to a liver cancer study. Extension of the method to more than one missing covariate is also discussed.

Organ Donation Attitudes and Practices Among Older Adults Participating in Evidence-Based Health Programs

Objective: The study assessed attitudes, barriers, and practices related to deceased organ donation and intent to enroll on a donor registry among older adults in health programs. Methods: We surveyed 1,286 adults (71% over age 60) participating in evidence-based self-management/lifestyle change programs in Michigan about their attitudes toward organ donation. Results: Three subscales with good psychometric properties were identified: General Benefits (Alpha .83), General Barriers (Alpha .79), and Age/Health Barriers (Alpha .61). Female gender, white race, higher education attainment/income, and intent to sign up on the donor registry were correlated with positive attitudes toward donation. Participants cited age and chronic disease status as barriers to enrolling as organ donors. Conclusion: Education on donor eligibility and perceived barriers could increase organ donation rates among participants in evidence-based health programs.

Subsample ignorable likelihood for accelerated failure time models with missing predictors

Missing values in predictors are a common problem in survival analysis. In this paper, we review estimation methods for accelerated failure time models with missing predictors, and apply a new method called subsample ignorable likelihood (IL) Little and Zhang (J R Stat Soc 60:591–605, 2011) to this class of models. The approach applies a likelihood-based method to a subsample of observations that are complete on a subset of the covariates, chosen based on assumptions about the missing data mechanism. We give conditions on the missing data mechanism under which the subsample IL method is consistent, while both complete-case analysis and ignorable maximum likelihood are inconsistent. We illustrate the properties of the proposed method by simulation and apply the method to a real dataset.