Chondra M. Lockwood’s research while affiliated with Oregon Health & Science University and other places

Objective To evaluate the effects of a research-tested, team-based health promotion and wellness program combined with digital technologies and implemented in a diverse worksite setting among hospital, clinic, and university employees. Methods A prospective cohort study of employees completing biometrics and questionnaires before and after the initial 12-session wellness program and its 12-session booster, 1 year later. Results After both the initial intervention and booster, blood pressure and weight were reduced, with greater reductions among employees with pre-hypertension and hypertension and those with a body mass index ≥25. After both the initial intervention and booster, there was a significant increase in (1) daily intake of fruit and vegetable servings, (2) days/week of ≥30 min of exercise, (3) days/week of strength training, and (4) levels of moderately vigorous and vigorous daily physical activity. Self-reported indices of both depression and work-related stress were reduced, while participants reported increased happiness. Post booster, average sleep quality, and sleep duration increased among higher risk employees reporting ≤6 h of daily sleep. Employees reported receiving encouragement from co-workers to engage in healthful diet and physical activities, and exercised together more, and indicated that they would recommend the program to other employees. Longitudinal analysis revealed the durability of the initial intervention outcomes with further beneficial effects after the booster. Conclusion A research tested, comprehensive team-based health promotion and wellness program, combined with digital technologies, improved employee health behaviors, mood, sleep, elements of co-worker cohesion, and biometrics among a diverse multi-site workforce. Positive program effects were durable, with enhanced results after the booster.

Mediation analysis is a newer statistical tool that is becoming more prominent in nutrition research. Its use provides insight into the relationship among variables in a potential causal chain. For intervention studies, it can define the influence of different programmatic components and, in doing so, allows investigators to identify and refine a program's critical aspects. We present an overview of mediation analysis, compare mediators with other variables (confounders, moderators, and covariates), and illustrate how mediation analysis permits interpretation of the change process. A framework is outlined for the critical appraisal of articles purporting to use mediation analysis. The framework's utility is demonstrated by searching the nutrition literature and identifying articles citing mediation cross referenced with the terms "nutrition," "diet," "food," and "obesity." Seventy-two articles were identified that involved human subjects and behavior outcomes, and almost half mentioned mediation without tests to define its presence. Tabulation of the 40 articles appropriately assessing mediation demonstrates an increase in these techniques' appearance and the breadth of nutrition topics addressed. Mediation analysis is an important new statistical tool. Familiarity with its methodology and a framework for assessing articles will allow readers to critically appraise the literature and make informed independent evaluations of works using these techniques.

Adolescence and emerging adulthood are critical windows for establishing life-long behaviors. We assessed long-term outcomes of a prospective randomized harm reduction/health promotion program for female high school athletes. The intervention's immediate beneficial effects on diet pill use and unhealthy eating behaviors have been reported; however, tobacco, alcohol and marijuana use were not immediately altered (Elliot et al, 2004). One to three years following graduation, positive benefits in those domains became evident, and intervention students reported significantly less lifetime use of cigarettes, marijuana, and alcohol. Sport teams may be effective vehicles for gender-specific interventions to promote competency skills and deter harmful actions, and those benefits may manifest when acquired abilities are applied in new environments following high school graduation.

This article describes a program, PRODCLIN (distribution of the PRODuct Confidence Limits for INdirect effects), written for SAS, SPSS, and R, that computes confidence limits for the product of two normal random variables. The program is important because it can be used to obtain more accurate confidence limits for the indirect effect, as demonstrated in several recent articles (MacKinnon, Lockwood, & Williams, 2004; Pituch, Whittaker, & Stapleton, 2005). Tests of the significance of and confidence limits for indirect effects based on the distribution of the product method have more accurate Type I error rates and more power than other, more commonly used tests. Values for the two paths involved in the indirect effect and their standard errors are entered in the PRODCLIN program, and distribution of the product confidence limits are computed. Several examples are used to illustrate the PRODCLIN program. The PRODCLIN programs in rich text format may be downloaded from www.psychonomic.org/archive.

An intermediate endpoint is hypothesized to be in the middle of the causal sequence relating an independent variable to a dependent variable. The intermediate variable is also called a surrogate or mediating variable and the corresponding effect is called the mediated, surrogate endpoint, or intermediate endpoint effect. Clinical studies are often designed to change an intermediate or surrogate endpoint and through this intermediate change influence the ultimate endpoint. In many intermediate endpoint clinical studies the dependent variable is binary, and logistic or probit regression is used. The purpose of this study is to describe a limitation of a widely used approach to assessing intermediate endpoint effects and to propose an alternative method, based on products of coefficients, that yields more accurate results. The intermediate endpoint model for a binary outcome is described for a true binary outcome and for a dichotomization of a latent continuous outcome. Plots of true values and a simulation study are used to evaluate the different methods. Distorted estimates of the intermediate endpoint effect and incorrect conclusions can result from the application of widely used methods to assess the intermediate endpoint effect. The same problem occurs for the proportion of an effect explained by an intermediate endpoint, which has been suggested as a useful measure for identifying intermediate endpoints. A solution to this problem is given based on the relationship between latent variable modeling and logistic or probit regression. More complicated intermediate variable models are not addressed in the study, although the methods described in the article can be extended to these more complicated models. Researchers are encouraged to use an intermediate endpoint method based on the product of regression coefficients. A common method based on difference in coefficient methods can lead to distorted conclusions regarding the intermediate effect.

The most commonly used method to test an indirect effect is to divide the estimate of the indirect effect by its standard error and compare the resulting z statistic with a critical value from the standard normal distribution. Confidence limits for the indirect effect are also typically based on critical values from the standard normal distribution. This article uses a simulation study to demonstrate that confidence limits are imbalanced because the distribution of the indirect effect is normal only in special cases. Two alternatives for improving the performance of confidence limits for the indirect effect are evaluated: (a) a method based on the distribution of the product of two normal random variables, and (b) resampling methods. In Study 1, confidence limits based on the distribution of the product are more accurate than methods based on an assumed normal distribution but confidence limits are still imbalanced. Study 2 demonstrates that more accurate confidence limits are obtained using resampling methods, with the bias-corrected bootstrap the best method overall.

The paper describes advances in statistical methods for prevention research with a particular focus on substance abuse prevention. Standard analysis methods are extended to the typical research designs and characteristics of the data collected in prevention research. Prevention research often includes longitudinal measurement, clustering of data in units such as schools or clinics, missing data, and categorical as well as continuous outcome variables. Statistical methods to handle these features of prevention data are outlined. Developments in mediation, moderation, and implementation analysis allow for the extraction of more detailed information from a prevention study. Advancements in the interpretation of prevention research results include more widespread calculation of effect size and statistical power, the use of confidence intervals as well as hypothesis testing, detailed causal analysis of research findings, and meta-analysis. The increased availability of statistical software has contributed greatly to the use of new methods in prevention research. It is likely that the Internet will continue to stimulate the development and application of new methods.

We employed meta-analytic techniques to examine the validity of the claim that social support and social negativity are unrelated. It was hypothesized that the inverse relation between social support and social negativity increases as the specificity of the level of the assessment increases. A total of 280 effect sizes (e.g., concurrent correlations between self-report measures of social support and negativity) were extracted from 87 journal articles and book chapters. In fixed effects models, the mean weighted effect size was -.08 for general assessments, -.18 for category-specific assessments, and -.34 for individual-specific assessments. Among effect sizes derived from individual-specific assessments, the relation between social support and social negativity was particularly strong when the provider was a spouse/significant other (weighted M = -.43). In weighted multiple regression analyses employing both fixed and random effect sizes models, level of assessment/type of provider remained significant (p < .001) in the presence of several covariates. Among the covariates, the strongest predictor was social negativity scale. Social support and social negativity appear to be moderately, inversely related when (a) the assessment focuses on the spouse/significant other relationship and (b) the Social Conflict Scale is used to measure social negativity.

To assess the deterrent effect of mandatory, random drug testing among high school (HS) athletes in a controlled setting. Two high schools, one with mandatory drug testing (DT) consent before sports participation, and a control school (C), without DT, were assessed during the 1999-2000 school year. Athletes (A) and nonathletes (NA) in each school completed confidential (A) or anonymous (NA) questionnaires developed for this study, respectively, at the beginning and end of the school year. Positive alcohol or drug tests required parent notification and mandatory counseling without team or school suspension. Thirty percent of the DT athletes were tested. Data were analyzed using the end of the school year measure, adjusted for the initial questionnaire results. Demographics of the athlete sample revealed that mean age was 15.5 years with 81.5% white, 9.6% Hispanic, 4.5% Asian, 2.6% American Indian/Native Alaskan, 1.3% African-American, and 1.3% Native Hawaiian/Pacific Islander. A (n = 276) and NA (n = 507) were assessed at the beginning (baseline) and at the end of the school year (A, n = 159; NA, n = 338). The past 30-day index of illicit drugs (4-fold difference) and athletic enhancing substances (3-fold difference) were lower (p < .05) among DT athletes at follow-up without difference in alcohol use. However, most drug use risk factors, including norms of use, belief in lower risk of drugs, and poorer attitudes toward the school, increased among DT athletes (p < .05). Although a reduction in the illicit drug use index was present among nonathletes at the DT school, at the end of the school year, it did not achieve statistical significance (p < .10). Random DT may have reduced substance use among athletes. However, worsening of risk factors and small sample size suggests caution to this drug prevention approach. A larger long-term study to confirm these findings is necessary.