Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (4 results)

  • Article
  • Peer Reviewed
UCLA Previously Published Works (2017)
There is a widely held belief that children’s general and psychological health benefits from owning and/or interacting with pets. In our study, we aimed to determine whether children who live with a dog or cat in their home have better mental and physical health outcomes compared with children without such a pet. Our study design consisted of a secondary analysis of household survey data from the 2003 California Health Interview Survey. Children in pet-owning households (n = 2,236 households with a dog or cat) were compared with children in non-pet owning households (n = 2,955 house-holds) using a weighted propensity score regression approach. Double robust regression analyses were used to examine the association between living with a dog or cat and health outcomes, while accounting for confounding factors. Our results demonstrated strong confounding effects. Unadjusted analyses found that children in pet-owning households were significantly healthier than children in non-owning households in terms of, for example, better general health, higher activity level, and less concern from parents regarding mood, behavior, and learning ability. However, when estimates were adjusted using the double robust approach, the effects were smaller and no longer statistically significant. The results indicate that the benefits of owning pets observed in this study were largely explained by confounding factors.
    Creative Commons 'BY' version 4.0 license
    • Article
    • Peer Reviewed
    UCLA Previously Published Works (2016)
    PURPOSE: Much variation in individual-level cognitive function in late life remains unexplained, with little exploration of area-level/contextual factors to date. Income inequality is a contextual factor that may plausibly influence cognitive function. METHODS: In a nationally-representative cohort of older Americans from the Health and Retirement Study, we examined state- and metropolitan statistical area (MSA)-level income inequality as predictors of individual-level cognitive function measured by the 27-point Telephone Interview for Cognitive Status (TICS-m) scale. We modeled latency periods of 8-20 years, and controlled for state-/metropolitan statistical area (MSA)-level and individual-level factors. RESULTS: Higher MSA-level income inequality predicted lower cognitive function 16-18 years later. Using a 16-year lag, living in a MSA in the highest income inequality quartile predicted a 0.9-point lower TICS-m score (β = -0.86; 95% CI = -1.41, -0.31), roughly equivalent to the magnitude associated with five years of aging. We observed no associations for state-level income inequality. The findings were robust to sensitivity analyses using propensity score methods. CONCLUSIONS: Among older Americans, MSA-level income inequality appears to influence cognitive function nearly two decades later. Policies reducing income inequality levels within cities may help address the growing burden of declining cognitive function among older populations within the United States.
      Cover page: Lagged Associations of Metropolitan Statistical Area- and State-Level Income Inequality with Cognitive Function: The Health and Retirement Study.
      • Article
      • Peer Reviewed
      Western Journal of Emergency Medicine: Integrating Emergency Care with Population Health, Volume 11, Issue 1 (2010)

      Background: The purpose of syndromic surveillance is early detection of a disease outbreak. Such systems rely on the earliest data, usually chief complaint. The growing use of electronic medical records (EMR) raises the possibility that other data, such as emergency department (ED) diagnosis, may provide more specific information without significant delay, and might be more effective in detecting outbreaks if mechanisms are in place to monitor and report these data.

      Objective: The purpose of this study is to characterize the added value of the primary ICD-9 diagnosis assigned at the time of ED disposition compared to the chief complaint for patients with influenza-like illness (ILI).

      Methods: The study was a retrospective analysis of the EMR of a single urban, academic ED with an annual census of over 60, 000 patients per year from June 2005 through May 2006. We evaluate the objective in two ways. First, we characterize the proportion of patients whose ED diagnosis is inconsistent with their chief complaint and the variation by complaint. Second, by comparing time series and applying syndromic detection algorithms, we determine which complaints and diagnoses are the best indicators for the start of the influenza season when compared to the Centers for Disease Control regional data for Influenza-Like Illness for the 2005 to 2006 influenza season using three syndromic surveillance algorithms: univariate cumulative sum (CUSUM), exponentially weighted CUSUM, and multivariate CUSUM.

      Results: In the first analysis, 29% of patients had a different diagnosis at the time of disposition than suggested by their chief complaint. In the second analysis, complaints and diagnoses consistent with pneumonia, viral illness and upper respiratory infection were together found to be good indicators of the start of the influenza season based on temporal comparison with regional data. In all examples, the diagnosis data outperformed the chief-complaint data.

      Conclusion: Both analyses suggest the ED diagnosis contains useful information for detection of ILI. Where an EMR is available, the short time lag between complaint and diagnosis may be a price worth paying for additional information despite the brief potential delay in detection, especially considering that detection usually occurs over days rather than hours. [West J Emerg Med. 2010; 11(1):1-9].

        Cover page: Emergency Department Chief Complaint and Diagnosis Data to Detect Influenza-Like Illness with an Electronic Medical RecordCreative Commons 'BY-NC' version 4.0 license
        • Article
        • Peer Reviewed
        UC Irvine Previously Published Works (2019)

        Purpose

        To implement a multilevel, church-based intervention with diverse disparity populations using community-based participatory research and evaluate feasibility, acceptability, and preliminary effectiveness in improving obesity-related outcomes.

        Design

        Cluster randomized controlled trial (pilot).

        Setting

        Two midsized (∼200 adults) African American baptist and 2 very large (∼2000) Latino Catholic churches in South Los Angeles, California.

        Participants

        Adult (18+ years) congregants (n = 268 enrolled at baseline, ranging from 45 to 99 per church).

        Intervention

        Various components were implemented over 5 months and included 2 sermons by pastor, educational handouts, church vegetable and fruit gardens, cooking and nutrition classes, daily mobile messaging, community mapping of food and physical activity environments, and identification of congregational policy changes to increase healthy meals.

        Measures

        Outcomes included objectively measured body weight, body mass index (BMI), and systolic and diastolic blood pressure (BP), plus self-reported overall healthiness of diet and usual minutes spent in physical activity each week; control variables include sex, age, race-ethnicity, English proficiency, education, household income, and (for physical activity outcome) self-reported health status.

        Analysis

        Multivariate linear regression models estimated the average effect size of the intervention, controlling for pair fixed effects, a main effect of the intervention, and baseline values of the outcomes.

        Results

        Among those completing follow-up (68%), the intervention resulted in statistically significantly less weight gain and greater weight loss (-0.05 effect sizes; 95% confidence interval [CI] = -0.06 to -0.04), lower BMI (-0.08; 95% CI = -0.11 to -0.05), and healthier diet (-0.09; 95% CI = -0.17 to -0.00). There was no evidence of an intervention impact on BP or physical activity minutes per week.

        Conclusion

        Implementing a multilevel intervention across diverse congregations resulted in small improvements in obesity outcomes. A longer time line is needed to fully implement and assess effects of community and congregation environmental strategies and to allow for potential larger impacts of the intervention.
          Cover page: Eat, Pray, Move: A Pilot Cluster Randomized Controlled Trial of a Multilevel Church-Based Intervention to Address Obesity Among African Americans and Latinos
          Top