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

Family Medicine

UC San Diego

Department of Family Medicine - Open Access Policy Deposits

This series is automatically populated with publications deposited by UC San Diego School of Medicine Department of Family Medicine researchers in accordance with the University of California’s open access policies. For more information see Open Access Policy Deposits and the UC Publication Management System.

How neighborhood socioeconomic status, green space, and walkability are associated with risk for fracture among postmenopausal women

(2025)

Abstract: Although most fractures, and about half of hip fractures, occur outdoors among older women, limited research has uncovered neighborhood predictors for fractures among older women. This study assessed the independent associations of neighborhood socioeconomic status, walkability, and green space with incident any and hip fracture among postmenopausal women. The Women’s Health Initiative recruited a national sample of postmenopausal women (50-79 years) across 40 U.S. clinical centers and conducted yearly assessments from 1993 to 2012 (n = 161 808). Women reporting a history of hip fracture or walking limitations were excluded from the analytic sample, yielding a final sample of 157 583 participants. Fracture events were self-reported and adjudicated annually. Walkability was calculated annually using measures of population density, land use mix, and presence/quantity of nearby high-traffic roadways. Neighborhood green space was calculated annually using measures of exposure to trees/vegetation. Neighborhood SES, walkability, and green space were categorized intro tertiles: high, intermediate, and low. The time-varying relationship between neighborhood environmental factors and age at first fracture (any; hip) was examined using extended Cox proportional hazards modeling with adjustment. Neighborhood socioeconomic status (intermediate vs low: hazard ratio = 1.03, 95% Confidence Interval (CI): 1.01-1.05; high vs low, hazard ratio = 1.01, 95% CI: 0.99-1.03) and green space (intermediate vs low, hazard ratio = 1.15, 95% CI: 1.12-1.18; high vs low hazard ratio = 1.18, 95% CI: 1.15-1.21) were associated with increased any incident fractures, while walkability had a mixed association (intermediate vs low hazard ratio = 1.06, 95% CI: 1.04-1.07; high vs low, hazard ratio = 0.97, 95% CI: 0.95-0.98). Neighborhood socioeconomic status, walkability, and green space did not have a relationship with hip fracture after adjustment for important covariates. Results indicate that macroscale neighborhood features did not protect against fractures. Additional research is needed to investigate more granual neighborhood features that might influence injury risk and support physical activity among postmenopausal women.

Cover page of Racial/ethnic differences in the associations between social support and cardiovascular morbidity and mortality in the Multi-Ethnic Study of Atherosclerosis (MESA)

Racial/ethnic differences in the associations between social support and cardiovascular morbidity and mortality in the Multi-Ethnic Study of Atherosclerosis (MESA)

(2025)

Background

Despite the established link between social support and cardiovascular disease (CVD) outcomes, few studies have examined racial/ethnic variation in these associations. This study utilized data from the Multi-Ethnic Study of Atherosclerosis (MESA) to investigate racial/ethnic differences in perceived social support and in the link between support and incident hard CVD events and mortality.

Method

Participants (N = 6,814) were 45-84 years of age who identified as White, Black, Hispanic/Latino, or Chinese without known clinical CVD at baseline (2000-2002). Racial/ethnic differences in perceived support (overall, emotional, informational, and instrumental) were tested using multiple regression with adjustments for demographic, socioeconomic, lifestyle/psychosocial, and clinical risk factors, and immigration history. Racial/ethnic differences in the association between perceived support and incident CVD events or mortality were tested using Cox proportional hazards models with progressive adjustments for the same covariates.

Results

At baseline, the mean age was 62.15 years (SD = 10.23); 38.5% identified as White, 27.8% as Black, 22.0% as Hispanic/Latino, and 11.8% as Chinese. Black and Hispanic/Latino participants reported higher levels of overall support, emotional support, and informational support than White participants (p's < 0.05). Chinese participants reported less informational support (p = .010) than White participants. Higher informational support was associated with decreased risk for hard CVD events. This association did not differ by race/ethnic group.

Conclusion

Despite racial/ethnic differences in perceptions of support, perceived informational support was protective against CVD for participants of all racial/ethnic backgrounds.

Cover page of Learning health system research as a catalyst for promoting physician wellness: EHR InBasket Spring cleaning and team-based compassion practice.

Learning health system research as a catalyst for promoting physician wellness: EHR InBasket Spring cleaning and team-based compassion practice.

(2025)

INTRODUCTION: Addressing physician burnout is critical for healthcare systems. As electronic health record (EHR) workload and teamwork have been identified as major contributing factors to physician well-being, we aimed to mitigate burnout through EHR-based interventions and a compassion team practice (CTP), targeting EHR workload and team cohesion. METHODS: A modified stepped wedge-clustered randomized trial was conducted, involving specialties with heavy InBasket workloads. EHR interventions included quick-action shortcuts and recommended practice for secure chats. The CTP comprised 30-s practice between physicians and their dyad partners. Survey and EHR data were collected over four assessment periods. Linear and generalized mixed-effects models assessed intervention effects, accounting for covariates. RESULTS: Forty-four physicians participated (20% participation rate). While burnout prevalence decreased from 58.5% at baseline to 50.0% at the end of the study, burnout reduction was not statistically significant after EHR (OR 0.43, 95% CI 0.12 to 1.61, p = 0.21) or EHR + CTP (OR 0.60, 95% CI 0.17 to 2.10, p = 0.42) interventions. Statistically significant greater perceived ease of EHR work resulted from both the EHR intervention (coefficient 0.76, 95% CI 0.22 to 1.29, p = 0.01) and EHR + CTP intervention (coefficient 0.80, 95% CI 0.26 to 1.35, p < 0.01). EHR + CTP increased perceived workplace supportiveness (coefficient 0.61, 95% CI -0.04 to 1.26, p = 0.07). Total number of InBasket messages/week increased significantly after EHR interventions (coefficient = 27.4, 95% CI 6.69 to 48.1, p = 0.011) and increased after EHR + CTP (18.5, 95% CI -3.15 to 40.2, p = 0.097). CONCLUSION: While burnout reduction was not statistically significant, EHR interventions positively impacted workload perceptions. CTP showed potential for improving perceived workplace supportiveness. Further research is needed to explore the efficacy of CTP with more participants. The interventions gained interest beyond our institution and prompted consideration for broader implementation.

National-level and state-level prevalence of overweight and obesity among children, adolescents, and adults in the USA, 1990-2021, and forecasts up to 2050.

(2024)

BACKGROUND: Over the past several decades, the overweight and obesity epidemic in the USA has resulted in a significant health and economic burden. Understanding current trends and future trajectories at both national and state levels is crucial for assessing the success of existing interventions and informing future health policy changes. We estimated the prevalence of overweight and obesity from 1990 to 2021 with forecasts to 2050 for children and adolescents (aged 5-24 years) and adults (aged ≥25 years) at the national level. Additionally, we derived state-specific estimates and projections for older adolescents (aged 15-24 years) and adults for all 50 states and Washington, DC. METHODS: In this analysis, self-reported and measured anthropometric data were extracted from 134 unique sources, which included all major national surveillance survey data. Adjustments were made to correct for self-reporting bias. For individuals older than 18 years, overweight was defined as having a BMI of 25 kg/m2 to less than 30 kg/m2 and obesity was defined as a BMI of 30 kg/m2 or higher, and for individuals younger than 18 years definitions were based on International Obesity Task Force criteria. Historical trends of overweight and obesity prevalence from 1990 to 2021 were estimated using spatiotemporal Gaussian process regression models. A generalised ensemble modelling approach was then used to derive projected estimates up to 2050, assuming continuation of past trends and patterns. All estimates were calculated by age and sex at the national level, with estimates for older adolescents (aged 15-24 years) and adults aged (≥25 years) also calculated for 50 states and Washington, DC. 95% uncertainty intervals (UIs) were derived from the 2·5th and 97·5th percentiles of the posterior distributions of the respective estimates. FINDINGS: In 2021, an estimated 15·1 million (95% UI 13·5-16·8) children and young adolescents (aged 5-14 years), 21·4 million (20·2-22·6) older adolescents (aged 15-24 years), and 172 million (169-174) adults (aged ≥25 years) had overweight or obesity in the USA. Texas had the highest age-standardised prevalence of overweight or obesity for male adolescents (aged 15-24 years), at 52·4% (47·4-57·6), whereas Mississippi had the highest for female adolescents (aged 15-24 years), at 63·0% (57·0-68·5). Among adults, the prevalence of overweight or obesity was highest in North Dakota for males, estimated at 80·6% (78·5-82·6), and in Mississippi for females at 79·9% (77·8-81·8). The prevalence of obesity has outpaced the increase in overweight over time, especially among adolescents. Between 1990 and 2021, the percentage change in the age-standardised prevalence of obesity increased by 158·4% (123·9-197·4) among male adolescents and 185·9% (139·4-237·1) among female adolescents (15-24 years). For adults, the percentage change in prevalence of obesity was 123·6% (112·4-136·4) in males and 99·9% (88·8-111·1) in females. Forecast results suggest that if past trends and patterns continue, an additional 3·33 million children and young adolescents (aged 5-14 years), 3·41 million older adolescents (aged 15-24 years), and 41·4 million adults (aged ≥25 years) will have overweight or obesity by 2050. By 2050, the total number of children and adolescents with overweight and obesity will reach 43·1 million (37·2-47·4) and the total number of adults with overweight and obesity will reach 213 million (202-221). In 2050, in most states, a projected one in three adolescents (aged 15-24 years) and two in three adults (≥25 years) will have obesity. Although southern states, such as Oklahoma, Mississippi, Alabama, Arkansas, West Virginia, and Kentucky, are forecast to continue to have a high prevalence of obesity, the highest percentage changes from 2021 are projected in states such as Utah for adolescents and Colorado for adults. INTERPRETATION: Existing policies have failed to address overweight and obesity. Without major reform, the forecasted trends will be devastating at the individual and population level, and the associated disease burden and economic costs will continue to escalate. Stronger governance is needed to support and implement a multifaceted whole-system approach to disrupt the structural drivers of overweight and obesity at both national and local levels. Although clinical innovations should be leveraged to treat and manage existing obesity equitably, population-level prevention remains central to any intervention strategies, particularly for children and adolescents. FUNDING: Bill & Melinda Gates Foundation.

Cover page of Regular cannabis smoking and carotid artery calcification in the Multi-Ethnic Study of Atherosclerosis (MESA)

Regular cannabis smoking and carotid artery calcification in the Multi-Ethnic Study of Atherosclerosis (MESA)

(2024)

Background

Studies on cannabis use and adverse cardiovascular outcomes have reported conflicting results. Research on its relationship to calcified arterial plaque remains limited.

Methods

Cross-sectional data from 2152 participants at Exam 6 (2016-2018) in the Multi-Ethnic Study of Atherosclerosis (MESA) were analyzed, including self-reported cannabis smoking patterns and carotid artery calcification (CAC) as measured via computed tomography. Multivariable relative and absolute risk regression models were used to estimate adjusted prevalence ratios (PRs) and prevalence differences, respectively, for the presence of calcified plaque. Multivariable linear regression was then used to compare group differences in the extent of CAC in those with calcified plaque.

Results

A minority of participants (n = 159, 7.4%) reported a history of regular cannabis smoking. Among all participants, 36.1% (n = 777) had detectable CAC. In models adjusted for demographics, behavioral, and clinical cardiovascular disease factors, a history of regular cannabis smoking was not associated with the prevalence of CAC in either common carotid artery (PR: 1.14, 95% CI: 0.88 to 1.49). In the subset of participants with calcified plaque, and in separate fully adjusted multivariable linear regression models, a history of regular cannabis smoking was not associated with increased calcium volume (difference = 7.7%, 95% CI: -21.8 to 48.5), calcium density (difference = 0.4%, 95% CI: -6.6 to 7.9), or Agatston score (difference = 32.1%, 95% CI: -31.8 to 155.8) in either carotid artery. Models exploring potential effect modification by age, race/ethnicity, and tobacco smoking status showed no significant association, except for higher CAC prevalence in men with a history of regular cannabis smoking.

Conclusions

In a racially and ethnically diverse cohort of older adults with a moderately high prevalence of CAC, no associations were found between a history of regular cannabis smoking, duration, or recency of cannabis smoking, and the prevalence of carotid calcified plaque. These findings were consistent across age, race/ethnicity, and cigarette smoking, except for an increased prevalence in men with a history of regular cannabis smoking. Similarly, in a subgroup with CAC, no association was found between a history of regular cannabis smoking and extent of calcification as measured by volume, density, and Agatston score.

Cover page of The Somali Distress and Resilience Scale: Development of a novel measure for Somali adults.

The Somali Distress and Resilience Scale: Development of a novel measure for Somali adults.

(2024)

Although resilience has been identified as an important mediator of negative mental health outcomes among refugee populations, there are few culturally specific measures of resilience among such communities and no such measure among Somalis. In this study we aimed to develop a culturally appropriate measure of resilience specific to Somali adults in San Diego, as an example of a vulnerable refugee community. A community-based, exploratory sequential mixed method investigation was conducted via focus group discussions (n = 4), cognitive interviews (n = 4), and iterative survey adaptation. Somali refugee adults in San Diego (N = 183) were surveyed with this novel scale, a standardized measure of resilience, and assessments of depression, anxiety, and PTSD. Results were analyzed via correlation coefficients and multivariate linear regression modeling. Qualitative findings supported the inclusion of items addressing both barriers and facilitators of good mental health, which resulted in the development of the Somali Distress and Resilience Survey (SDRS). Linear regression analysis revealed that the SDRS demonstrated significant associations with symptoms of depression and PTSD, while the standardized measure of resilience did not demonstrate associations with any of the mental health outcomes assessed. The SDRS identified obstacles to resilience among Somali individuals, placing them at risk of developing negative mental health outcomes. Our novel measure also demonstrated more robust relationships with these outcomes than a standardized measure of resilience, suggesting greater utility of the adapted scale. However, the SDRSs development raises larger questions about the limitations of developing and comprehensively evaluating novel resilience measures in a community-based setting.

Cover page of Optimizing Compassion Training in Medical Trainees Using an Adjunct mHealth App: A Preliminary Single-Arm Feasibility and Acceptability Study.

Optimizing Compassion Training in Medical Trainees Using an Adjunct mHealth App: A Preliminary Single-Arm Feasibility and Acceptability Study.

(2024)

BACKGROUND: While structured compassion training programs have shown promise for increasing compassion among medical trainees, a major challenge is applying the concepts and practices taught during the program into the complex, dynamic, time-pressured, and often hectic hospital workplace. OBJECTIVE: The purpose of this pilot study was to examine the feasibility, acceptability, and preliminary effects of Compassion Coach, a mobile health (mHealth) smartphone app designed to bolster a 6-week mindfulness and self-compassion training program for medical trainees. METHODS: In Compassion Coach, notifications to remind, encourage, and measure the perceived impact of informal mindfulness and compassion practices taught during the program were delivered at 7 AM, 12 PM, and 7 PM, respectively, 3 times per week over the course of the training program. The app also contained a library of guided audio formal mindfulness and compassion practices to allow quick and easy access. In this pilot study, we collected data from 29 medical students and residents who downloaded Compassion Coach and completed surveys assessing perceived effectiveness and acceptability. Engagement with the Compassion Coach app was passively tracked through notification response rate and library resource access over time. RESULTS: The average response rate to notifications was 58% (SD 29%; range 12%-98%), with a significant decline over time (P=.009; odds ratio 0.98, 95% CI 0.96-0.99). Across all participants and occasions, the majority agreed the informal practices prompted by Compassion Coach helped them feel grounded and centered (110/150, 73%), improved compassion (29/41, 71%), reduced burnout (106/191, 56%), and improved their mood (133/191, 70%). In total, 16 (55%) of the 29 participants accessed guided audio recordings on average 3 (SD 3.4) times throughout the program. At the posttreatment time point, most participants (13/18, 72%) indicated that Compassion Coach helped them engage in compassion practices in daily life, and half (9/18, 50%) indicated that Compassion Coach helped improve interactions with patients. CONCLUSIONS: Overall, preliminary results of Compassion Coach are encouraging and suggest the integration of a smartphone app with an ongoing mindfulness and self-compassion training program may bolster the effects of the program on medical trainees. However, there was variability in engagement with Compassion Coach and perceived helpfulness. Additional research is indicated to optimize this novel mHealth approach and conduct a study powered to formally evaluate effects.

Cover page of Type 1 Human Immunodeficiency Virus (HIV-1) Incidence, Adherence, and Drug Resistance in Individuals Taking Daily Emtricitabine/Tenofovir Disoproxil Fumarate for HIV-1 Pre-exposure Prophylaxis: Pooled Analysis From 72 Global Studies

Type 1 Human Immunodeficiency Virus (HIV-1) Incidence, Adherence, and Drug Resistance in Individuals Taking Daily Emtricitabine/Tenofovir Disoproxil Fumarate for HIV-1 Pre-exposure Prophylaxis: Pooled Analysis From 72 Global Studies

(2024)

Background

Oral pre-exposure prophylaxis (PrEP) with emtricitabine/tenofovir disoproxil fumarate (F/TDF) has high efficacy against HIV-1 acquisition. Seventy-two prospective studies of daily oral F/TDF PrEP were conducted to evaluate HIV-1 incidence, drug resistance, adherence, and bone and renal safety in diverse settings.

Methods

HIV-1 incidence was calculated from incident HIV-1 diagnoses after PrEP initiation and within 60 days of discontinuation. Tenofovir concentrations in dried blood spots (DBS), drug resistance, and bone/renal safety indicators were evaluated in a subset of studies.

Results

Among 17 274 participants, there were 101 cases with new HIV-1 diagnosis (.77 per 100 person-years; 95% confidence interval [CI]: .63-.94). In 78 cases with resistance data, 18 (23%) had M184I or V, 1 (1.3%) had K65R, and 3 (3.8%) had both mutations. In 54 cases with tenofovir concentration data from DBS, 45 (83.3%), 2 (3.7%), 6 (11.1%), and 1 (1.9%) had average adherence of <2, 2-3, 4-6, and ≥7 doses/wk, respectively, and the corresponding incidence was 3.9 (95% CI: 2.9-5.3), .24 (.060-.95), .27 (.12-.60), and .054 (.008-.38) per 100 person-years. Adherence was low in younger participants, Hispanic/Latinx and Black participants, cisgender women, and transgender women. Bone and renal adverse event incidence rates were 0.69 and 11.8 per 100 person-years, respectively, consistent with previous reports.

Conclusions

Leveraging the largest pooled analysis of global PrEP studies to date, we demonstrate that F/TDF is safe and highly effective, even with less than daily dosing, in diverse clinical settings, geographies, populations, and routes of HIV-1 exposure.

Cover page of Orthopedic Injections: A Longitudinal Musculoskeletal Curriculum in a Family Medicine Residency.

Orthopedic Injections: A Longitudinal Musculoskeletal Curriculum in a Family Medicine Residency.

(2024)

BACKGROUND AND OBJECTIVES: Musculoskeletal (MSK) complaints comprise more than 20% of all visits to health care providers each year. Despite required experiences in MSK care, family physicians report low confidence in diagnosing and treating MSK conditions. The purpose of this study was to analyze the effects of early and longitudinal exposure to MSK education on residents confidence in and likelihood of performing MSK physical exams and injections in future practice. METHODS: From 2017 to 2019, residents completed an annual survey assessing confidence in, frequency of, and future intentions to perform exams and injections for MSK conditions. We compared responses between family medicine residents who completed a 176-hour longitudinal sports medicine (LSM) curriculum distributed over all 3 years of residency and a comparable cohort of family medicine residents who completed a 188-hour concentrated MSK curriculum primarily in the final year of residency. We made comparisons using the Fisher exact test for categorical variables and an independent samples t test for numeric variables. RESULTS: We analyzed the 98 total responses from 50 residents. The proportion of residents reporting high ratings of their residency MSK education (26% to 60%), performing >5 injections (38% to 73%), reporting confidence in performing injections (12% to 40%), and indicating likelihood to perform MSK injections in the future (52% to 65%) were all greater in the LSM versus concentrated MSK curriculum cohorts (P<.05 for all). CONCLUSIONS: Early and longitudinal exposure to MSK care and sports medicine in family medicine residency led to both an increase in MSK injections during residency training and a greater desire to perform these injections in postresidency practice.

Cover page of Community Health Navigators for Cancer Screening Among Deaf, Deafblind, and Hard of Hearing Adults Who Use American Sign Language.

Community Health Navigators for Cancer Screening Among Deaf, Deafblind, and Hard of Hearing Adults Who Use American Sign Language.

(2024)

Deaf, deafblind, and hard of hearing (DDBHH) individuals experience barriers to accessing cancer screening, including ineffective patient-physician communication when discussing screening recommendations. For other underserved communities, culturally and linguistically aligned community health navigators (CHNs) have been shown to improve cancer screening and care. A needs assessment study was conducted to identify barriers and gather recommendations for CHN training resources. A community-based participatory needs assessment was conducted from May 2022 to June 2022 using three focus groups. Eight were cancer survivors, six advocates/navigators, and three clinicians. All questions were semi-structured and covered screening barriers, observations or personal experiences, perceived usefulness of having a CHN to promote cancer screening adherence, and training resources that may be useful to American Sign Language (ASL)-proficient CHNs, who are also culturally and linguistically aligned. Out of 20 focus group participants, seven self-identified as persons of color. Data highlighted systemic, attitudinal, communication, and personal-level barriers as recurrent themes. The most frequently cited barrier was access to training that supports the role and competencies of CHNs, followed by cultural considerations, access to cancer guidelines in ASL, dialect diversity in sign language, and the health system itself. Unaddressed barriers can contribute to health disparities, such as lower preventive cancer screening rates amongst DDBHH individuals. The next step is to translate recommendations into actionable tasks for DDBHH CHN training programs. As a result, CHNs will be well-equipped to help DDBHH individuals navigate and overcome their unique barriers to cancer screening and healthcare access.