Predicting Self-Reported Social Risk in Medically Complex Adults Using Electronic Health Data.

BACKGROUND: Social barriers to health care, such as food insecurity, financial distress, and housing instability, may impede effective clinical management for individuals with chronic illness. Systematic strategies are needed to more efficiently identify at-risk individuals who may benefit from proactive outreach by health care systems for screening and referral to available social resources. OBJECTIVE: To create a predictive model to identify a higher likelihood of food insecurity, financial distress, and/or housing instability among adults with multiple chronic medical conditions. RESEARCH DESIGN AND SUBJECTS: We developed and validated a predictive model in adults with 2 or more chronic conditions who were receiving care within Kaiser Permanente Northern California (KPNC) between January 2017 and February 2020. The model was developed to predict the likelihood of a "yes" response to any of 3 validated self-reported survey questions related to current concerns about food insecurity, financial distress, and/or housing instability. External model validation was conducted in a separate cohort of adult non-Medicaid KPNC members aged 35-85 who completed a survey administered to a random sample of health plan members between April and June 2021 (n = 2820). MEASURES: We examined the performance of multiple model iterations by comparing areas under the receiver operating characteristic curves (AUCs). We also assessed algorithmic bias related to race/ethnicity and calculated model performance at defined risk thresholds for screening implementation. RESULTS: Patients in the primary modeling cohort (n = 11,999) had a mean age of 53.8 (±19.3) years, 64.7% were women, and 63.9% were of non-White race/ethnicity. The final, simplified model with 30 predictors (including utilization, diagnosis, behavior, insurance, neighborhood, and pharmacy-based variables) had an AUC of 0.68. The model remained robust within different race/ethnic strata. CONCLUSIONS: Our results demonstrated that a predictive model developed using information gleaned from the medical record and from public census tract data can be used to identify patients who may benefit from proactive social needs assessment. Depending on the prevalence of social needs in the target population, different risk output thresholds could be set to optimize positive predictive value for successful outreach. This predictive model-based strategy provides a pathway for prioritizing more intensive social risk outreach and screening efforts to the patients who may be in greatest need.

Telehealth During COVID-19 for Adults with Multiple Chronic Conditions: Associations with Self-Reported Food Insecurity and with Physical Limitations.

Background: Adults with chronic medical conditions complicated by food insecurity or physical limitations may have higher barriers to accessing telehealth implemented during the COVID-19 pandemic. Objective: To examine the relationships of self-reported food insecurity and physical limitations with changes in health care utilization and medication adherence comparing the year before (March 2019-February 2020) and the first year of the COVID-19 pandemic (April 2020-March 2021) among patients with chronic conditions insured by Medicaid or Medicare Advantage. Methods: A prospective cohort study of 10,452 Kaiser Permanente Northern California members insured by Medicaid and 52,890 Kaiser Permanente Colorado members insured by Medicare Advantage was conducted. Difference-in-differences (DID) between the pre-COVID and COVID years in telehealth versus in-person health care utilization and adherence to chronic disease medicines by food insecurity and by physical limitation status were measured. Results: Food insecurity and physical limitations were each associated with small but significantly greater shifts from in-person to telehealth. Medicare Advantage members with physical limitations also had significantly greater decline in adherence to chronic medications from year to year compared with those without physical limitations (DID from pre-COVID year to COVID year ranged from 0.7% to 3.6% greater decline by medication class, p < 0.01). Conclusions: Food insecurity and physical limitations did not present significant barriers to the transition to telehealth during the COVID pandemic. The greater decrease in medication adherence among older patients with physical limitations suggests that care systems must further address the needs of this high-risk population.

Associations Between Different Self-reported Social Risks and Neighborhood-level Resources in Medicaid Patients.

BACKGROUND: Adverse social conditions are a key contributor to health disparities. Improved understanding of how social risk factors interact with each other and with neighborhood characteristics may inform efforts to reduce health disparities. DATA: A questionnaire of 29,281 patients was collected through the enrollment of Medicaid beneficiaries in a large Northern California integrated health care delivery system between May 2016 and February 2020. EXPOSURES: Living in the least resourced quartile of neighborhoods as measured by a census-tract level Neighborhood Deprivation Index score. MAIN OUTCOMES: Five self-reported social risk factors: financial need, food insecurity, housing barriers, transportation barriers, and functional limitations. RESULTS: Nearly half (42.0%) of patients reported at least 1 social risk factor; 22.4% reported 2 or more. Mean correlation coefficient between social risk factors was ρ=0.30. Multivariable logistic models controlling for age, race/ethnicity, sex, count of chronic conditions, and insurance source estimated that living in the least resourced neighborhoods was associated with greater odds of food insecurity (adjusted odds ratio=1.07, 95% confidence interval: 1.00-1.13) and transportation barriers (adjusted odds ratio=1.20, 95% confidence interval: 1.11-1.30), but not financial stress, housing barriers, or functional limitations. CONCLUSIONS AND RELEVANCE: We found that among 5 commonly associated social risk factors, Medicaid patients in a large Northern California health system typically reported only a single factor and that these factors did not correlate strongly with each other. We found only modestly greater social risk reported by patients in the least resourced neighborhoods. These results suggest that individual-level interventions should be targeted to specific needs whereas community-level interventions may be similarly important across diverse neighborhoods.

Accounting for Social Risk Does not Eliminate Race/Ethnic Disparities in COVID-19 Infection Among Insured Adults: a Cohort Study.

BACKGROUND: Communities of color have been disproportionately impacted by the COVID-19 epidemic in the USA. OBJECTIVES: To examine the relationship of self-reported social health needs with SARS-COV-2 infection by race/ethnicity among insured adults with access to high-quality health care. DESIGN AND PARTICIPANTS: A prospective cohort study of 26,741 adult Kaiser Permanente Northern California members insured by Medicaid and 58,802 Kaiser Permanente Colorado members insured by Medicare Advantage who completed social risk assessments prior to the onset of the COVID-19 pandemic. MAIN MEASURES: We examined the independent relationships of demographic, medical, and social factors on SARS-COV-2 testing and positivity between March 1, 2020, and November 30, 2020, by race/ethnicity. KEY RESULTS: Findings were similar in the two cohorts, with Latino (16-18%), Asian (11-14%), and Black (11-12%) members having the highest prevalence of SARS-COV-2 infection (ORs adjusted for age, gender, and use of interpreter ranging from 1.68 to 2.23 compared to White member [7-8%], p < 0.001). Further adjustment for medical comorbidity (e.g., obesity, diabetes, chronic lung disease); neighborhood measures; and self-reported social risk factors (e.g., trouble paying for basics, food insecurity, housing concerns, transportation barriers) did not appreciably change these results. CONCLUSIONS: Compared to non-Latino White members, members of other race/ethnic groups had higher positivity rates that were only minimally reduced after controlling for medical and neighborhood conditions and self-reported social risk factors. These findings suggest that traditional infection transmission factors such as essential work roles and household size that have disproportionate representation among communities of color may be important contributors to SARS-COV-2 infection among insured adults.

Telephone-based social health screening by pharmacists in the nonadherent Medicare population.

BACKGROUND: Unmet social health needs are associated with medication nonadherence. Although pharmacists are well positioned to address medication nonadherence, there is limited experience with screening for and addressing social health needs. OBJECTIVES: To compare the prevalence of social health needs among Medicare patients with higher vs lower social health risk using a predictive model. To also evaluate pre-post changes in medication adherence and health care use following a pharmacist-initiated social health screening. METHODS: A social health screening workflow was implemented into a routine pharmacist adherence program at an integrated health care delivery system. The social health screening was conducted during medication adherence outreach phone calls with Medicare members who were overdue for statin, blood pressure, or diabetes medications. We developed a social health need predictive algorithm to flag higher-risk patients and tested this algorithm against a random subset of lower-risk patients. Screening conversations were guided by a focus group that developed open-ended questions to identify social health needs. Comparisons in social health needs were made between higher- and lower-risk patients. Use and adherence outcomes were compared pre and post for patients who accepted a referral to social health resources and patients who declined a referral. RESULTS: 1,217 patients were contacted and screened for social health needs by pharmacists. Patients flagged by the social risk algorithm were more likely to report social health needs (28.7% vs 12.7% in the unflagged group; P < 0.01). Commonly reported needs included transportation (43%), finances (34%), caregiving (22%), mental health (11%), and food access (10%). 221 patients accepted a referral to a central resource website and call center that connected patients to local services. One year after screening dates, patients who did not accept a referral spent more time in the hospital (mean change +0.7 days, SD = 7.3, P < 0.01), had fewer primary care visits (mean change -0.5 visits, SD = 6.5, P < 0.01), and had a shorter length of membership (mean change -0.4 months, SD = 1.9, P < 0.01). Patients who accepted a referral had increased statin adherence (62.3% adherent pre vs 74.7% post, P = 0.02). CONCLUSIONS: We implemented a workflow for pharmacists to screen for social health needs. The social health need prediction model doubled the identification rate of patients who have needs. Intervening on social health needs during these calls may improve statin adherence and may have no adverse effect on health care utilization or health plan membership. DISCLOSURES: Social health risk predictive model development and validation was funded by the Agency for Healthcare Research and Quality (AHRQ R18HS027343).

Prevalence and determinants of oral and cervicogenital HPV infection: Baseline analysis of the Michigan HPV and Oropharyngeal Cancer (MHOC) cohort study.

We determined baseline oral and cervicogenital human papillomavirus (HPV) prevalence and determinants of infection in the Michigan HPV and Oropharyngeal Cancer (MHOC) study. We enrolled 394 college-age and older participants of both sexes in Ann Arbor, Michigan and the surrounding area. All participants provided an oral sample at baseline, and 130 females provided a cervicogenital sample. Samples were tested for 18 HPV genotypes using polymerase chain reaction (PCR) MassArray. Participants filled out sociodemographic and behavioral questionnaires. Prevalence ratios for HPV oral or cervicogenital prevalence by predictor variables were estimated in univariable log-binomial models. Analysis was conducted 2018-20. In the full cohort, baseline oral HPV prevalence was 10.0% for any detected genotype (among the 338 valid oral tests at baseline) and 6.5% for high-risk types, and cervicogenital prevalence was 20.0% and 10.8%, respectively (among the 130 first valid cervicogenital tests). Oral HPV prevalence did not vary by sex, with 10.5% of women and 9.0% of men having an infection. We found a high prevalence of oral and cervicogenital HPV infection in college-age participants reporting no lifetime sexual partners. Reporting a single recent partner was associated with a lower oral HPV prevalence (PR 0.39, 95% CI: 0.16, 0.96) than reporting no recent (but at least one ever) partner. No similar protective effect was seen for cervicogenital HPV. Both oral and cervicogenital prevalence increased with the number of recent partners for most sexual behaviors. We observed an ecological fallacy masking the direction of impact of vaccination on HPV prevalence in the full cohort compared to the college-aged and the age 23+ populations considered separately. Substance use was not significantly associated with oral or cervicogenital HPV infection. Many studies report substantially higher oral HPV infection prevalence in men than in women. That difference may not be uniform across populations in the US.

Oral human papillomavirus prevalence, persistence, and risk-factors in HIV-positive and HIV-negative adults.

BACKGROUND: HIV has been shown to increase the likelihood of oral HPV infection. In this study, we evaluated the risk of oral HPV in HIV infected patients compared with HIV-negative controls. METHODS: 101 healthy adult volunteers (HIV-) and 245 adults living with HIV infection (HIV+) were recruited from 5 academic medical centers. Questionnaires and saliva samples were obtained every 3-8 months over a period of 2 years (2015-2017). DNA was isolated from the saliva samples and tested for 18 high- and low-risk genotypes. RESULTS: Oral HPV was detected in 23% of HIV + vs. 10% of HIV- participants (p < 0.0001). Men had a higher oral HPV prevalence than women (27% vs. 15% HIV+, p = 0.03, 16% vs. 5% HIV-, p = 0.01). Risk factors among HIV + participants included more lifetime deep kissing and oral sex partners, and history of AIDS. Persistent oral HPV was detected in 23% of HIV + vs. 5% of HIV- participants (p < 0.001). Among 8 HIV + participants with CD4 counts <200 cell/µL none had cleared their HPV infection during the study. CONCLUSIONS: Risk of oral HPV infection and persistence was significantly higher in HIV + adults with a history of poorly controlled HIV, which may put them at increased risk of HPV-associated cancer.

Use of Latent Class Analysis and k-Means Clustering to Identify Complex Patient Profiles.

Importance: Medically complex patients are a heterogeneous group that contribute to a substantial proportion of health care costs. Coordinated efforts to improve care and reduce costs for this patient population have had limited success to date. Objective: To define distinct patient clinical profiles among the most medically complex patients through clinical interpretation of analytically derived patient clusters. Design, Setting, and Participants: This cohort study analyzed the most medically complex patients within Kaiser Permanente Northern California, a large integrated health care delivery system, based on comorbidity score, prior emergency department admissions, and predicted likelihood of hospitalization, from July 18, 2018, to July 15, 2019. From a starting point of over 5000 clinical variables, we used both clinical judgment and analytic methods to reduce to the 97 most informative covariates. Patients were then grouped using 2 methods (latent class analysis, generalized low-rank models, with k-means clustering). Results were interpreted by a panel of clinical stakeholders to define clinically meaningful patient profiles. Main Outcomes and Measures: Complex patient profiles, 1-year health care utilization, and mortality outcomes by profile. Results: The analysis included 104 869 individuals representing 3.3% of the adult population (mean [SD] age, 70.7 [14.5] years; 52.4% women; 39% non-White race/ethnicity). Latent class analysis resulted in a 7-class solution. Stakeholders defined the following complex patient profiles (prevalence): high acuity (9.4%), older patients with cardiovascular complications (15.9%), frail elderly (12.5%), pain management (12.3%), psychiatric illness (12.0%), cancer treatment (7.6%), and less engaged (27%). Patients in these groups had significantly different 1-year mortality rates (ranging from 3.0% for psychiatric illness profile to 23.4% for frail elderly profile; risk ratio, 7.9 [95% CI, 7.1-8.8], P < .001). Repeating the analysis using k-means clustering resulted in qualitatively similar groupings. Each clinical profile suggested a distinct collaborative care strategy to optimize management. Conclusions and Relevance: The findings suggest that highly medically complex patient populations may be categorized into distinct patient profiles that are amenable to varying strategies for resource allocation and coordinated care interventions.

Pea aphid as both host and vector for the phytopathogenic bacterium Pseudomonas syringae.

Aphids are widespread agricultural pests that are capable of disseminating plant viral diseases; however, despite coming into frequent contact with epiphytic bacteria, aphids are considered to have no role in bacterial transmission. Here, we demonstrate the ability of pea aphids to vector the phytopathogen Pseudomonas syringae pv. syringae B728a (PsyB728a). While feeding on plants colonized by epiphytic bacteria, aphids acquire the bacteria, which colonize the digestive tract, multiply, and are excreted in the aphid honeydew, resulting in inoculation of the phyllosphere with up to 10(7) phytopathogenic bacteria per cm(2). Within days of ingesting bacteria, aphids succumb to bacterial sepsis, indicating that aphids serve as an alternative, nonplant host for PsyB728a. The related strain Pseudomonas syringae pv. tomato DC3000 is >1,000-fold less virulent than PsyB728a in the pea aphid, suggesting that PsyB728a possesses strain-specific pathogenicity factors that allow it to exploit aphids as hosts. To identify these factors, we performed a mutagenesis screen and recovered PsyB728a mutants that were hypovirulent, including one defective in a gene required for flagellum formation and motility. These interactions illustrate that aphids can also vector bacterial pathogens and that even seemingly host-restricted pathogens can have alternative host specificities and lifestyles.