Faculty Bibliography

OBJECTIVES:Patient experience and presence of evidence-based facility services are 2 dimensions of assessing quality of addiction treatment facilities. However, the relationship between these two is not well described. The objective of this study was to explore associations between patient experience measures and service offerings at addiction treatment facilities. METHODS:We used data from cross-sectional surveys of addiction treatment facilities and persons involved in treatment at corresponding facilities to identify facility services (eg, availability of medications for alcohol use disorder, assistance with obtaining social services, etc) and patient experience measures (overall facility rating, extent helped by treatment, ability to deal with daily problems after treatment), respectively. We used hierarchical multiple logistic regression to test for associations between top-box scores for each patient experience outcome and facility services. RESULTS:We analyzed 9191 patient experience surveys from 149 facilities. Assistance with obtaining social services (adjusted odds ratio [95% confidence interval], 0.43 [0.28-0.66]) was associated with lower overall treatment facility ratings. Childcare (2.00 [1.04-3.84]) was associated with top-box scores for extent helped. Availability of cognitive behavioral therapy (2.67 [1.25-5.73]) and childcare (1.77 [1.08-2.92]) were associated with top-box scores for ability to deal with daily problems after treatment. Assistance with obtaining social services (0.61 [0.41-0.90]) was associated with lower scores for ability to deal with problems after treatment. CONCLUSIONS:Few addiction treatment facility services were associated with patient experience measures. Future work should explore bridging the gap between evidence-based services and positive patient experiences.

OBJECTIVE:We examined drive times to outpatient substance use disorder treatment providers that provide contingency management (CM) and those that integrate CM with medication for opioid use disorder (MOUD) services in 6 US states. METHODS:We completed cross-sectional geospatial analysis among census tracts in Delaware, Louisiana, Massachusetts, North Carolina, New York, and West Virginia. We excluded census tracts with a population of zero. Using data from the 2020 Shatterproof substance use treatment facility survey, our outcome was the minimum drive time in minutes from the census tract mean center of population to the nearest outpatient CM provider, outpatient CM provider with MOUD services, and federally qualified health centers (FQHC). We stratified census tracts by 2010 Rural-Urban Commuting Area codes and by state. RESULTS:The population was greater than zero in 11,719 of 11,899 census tracts. The median drive time to the nearest CM provider was 12.2 [interquartile range (IQR), 7.0-23.5) minutes and the median drive time to the nearest CM provider increased from 9.7 (IQR, 6.0-15.0) minutes in urban census tracts to 38.8 (IQR, 25.4-53.0) minutes in rural ( H = 3683, P < 0.001). The median drive time increased to the nearest CM provider with MOUD services [14.2 (IQR, 7.9-29.5) minutes, W = 18,877, P < 0.001] and decreased to the nearest FQHC [7.9 (IQR, 4.3-13.6) minutes, W = 11,555,894, P < 0.001]. CONCLUSIONS:These results suggest limited availability of CM, particularly within rural communities and for patients needing concurrent CM and MOUD treatment. Our results suggest greater adoption of CM within FQHCs could reduce urban-rural disparities in CM availability.

IMPORTANCE:Staffing shortages in nursing homes (NHs) threaten the quality of resident care, and the COVID-19 pandemic magnified critical staffing shortages within NHs. During the pandemic, the US Congress enacted the Paycheck Protection Program (PPP), a forgivable loan program that required eligible recipients to appropriate 60% to 75% of the loan toward staffing to qualify for loan forgiveness. OBJECTIVE:To evaluate characteristics of PPP loan recipient NHs vs nonloan recipient NHs and whether there were changes in staffing hours at NHs that received a loan compared with those that did not. DESIGN, SETTING, AND PARTICIPANTS:This economic evaluation used national data on US nursing homes that were aggregated from the Small Business Administration, Nursing Home Compare, LTCFocus, the Centers for Medicare & Medicaid Services Payroll Based Journal, the Minimum Data Set, the Area Deprivation Index, the Healthcare Cost Report Information System, and the US Department of Agriculture Rural-Urban Continuum Codes from January 1 to December 23, 2020. EXPOSURE:Paycheck Protection Program loan receipt status. MAIN OUTCOME AND MEASURES:Staffing variables included registered nurse, licensed practical nurse (LPN), and certified nursing assistant (CNA) total hours per week. Staffing hours were examined on a weekly basis before and after loan receipt during the study period. An event-study approach was used to estimate the staffing total weekly hours at NHs that received PPP loans compared with NHs that did not receive a PPP loan. RESULTS:Among 6008 US NHs, 1807 (30.1%) received a PPP loan and 4201 (69.9%) did not. The median loan amount was $664 349 (IQR, $407 000-$1 058 300). Loan recipients were less likely to be part of a chain (733 [40.6%] vs 2592 [61.7%]) and more likely to be for profit (1342 [74.3%] vs 2877 [68.5%]), be located in nonurban settings (159 [8.8%] vs 183 [4.4%]), have a greater proportion of Medicaid-funded residents (mean [SD], 60.92% [21.58%] vs 56.78% [25.57%]), and have lower staffing quality ratings (mean [SD], 2.88 [1.20] vs 3.03 [1.22]) and overall quality star ratings (mean [SD], 3.08 [1.44] vs 3.22 [1.44]) (P < .001 for all). Twelve weeks after PPP loan receipt, NHs that received a PPP loan experienced a mean difference of 26.19 more CNA hours per week (95% CI, 14.50-37.87 hours per week) and a mean difference of 6.67 more LPN hours per week (95% CI, 1.21-12.12 hours per week) compared with nursing homes that did not receive a PPP loan. No associations were found between PPP loan receipt and weekly RN staffing hours (12 weeks: mean difference, 1.99 hours per week; 95% CI, -2.38 to 6.36 hours per week). CONCLUSIONS AND RELEVANCE:In this economic evaluation, a forgivable loan program that required funding to be appropriated toward staffing was associated with a significant increase in CNA and LPN staffing hours among NH PPP loan recipients. Because the PPP loans are temporary, federal and state entities may need to institute sufficient and sustainable support to mitigate NH staffing shortages.

Understanding patient characteristics associated with scheduling and completing telehealth visits can identify potential biases or latent preferences related to telehealth usage. We describe patient characteristics associated with being scheduled for and completing audio and video visits. We used data from patients at 17 adult primary care departments in a large, urban public healthcare system from August 1, 2020 to July 31, 2021. We used hierarchical multivariable logistic regression to generate adjusted odds ratios (aOR) for patient characteristics associated with having been scheduled for and completed telehealth (vs in-person) visits and for video (vs audio) scheduling and completion during two time periods: a telehealth transition period (N = 190,949) and a telehealth elective period (N = 181,808). Patient characteristics were significantly associated with scheduling and completion of telehealth visits. Many associations were similar across time periods, but others changed over time. Patients who were older (≥ 65 years old vs 18-44 years old: aOR for scheduling 0.53/completion 0.48), Black (0.86/0.71), Hispanic (0.76/0.62), or had Medicaid (0.93/0.84) were among those less likely to be scheduled for or complete video (vs audio) visits. Patients with activated patient portals (1.97/3.34) or more visits (≥ 3 scheduled visits vs 1 visit: 2.40/1.52) were more likely to be scheduled for or complete video visits. Variation in scheduling/completion explained by patient characteristics was 7.2%/7.5%, clustering by provider 37.2%/34.9%, and clustering by facility 43.1%/37.4%. Stable and dynamic associations suggest persistent gaps in access and evolving preferences/biases. Variation explained by patient characteristics was relatively low compared with that explained by provider and facility clustering.

OBJECTIVES:This study aimed to describe addiction treatment facilities by their offerings of medications for alcohol use disorder (MAUD) and/or for opioid use disorder (MOUD), and by their offering services to groups with barriers to care: uninsured and publicly insured, youth, seniors, individuals preferring to receive care in Spanish, and sexual minority individuals. METHODS:We examined addiction treatment facility survey data in 6 US states. We performed bivariate analyses comparing facilities that offered MAUD, MOUD, and both (main outcomes). We then constructed a multivariable model to identify predictors of offering MAUD, MOUD, or both, including exposures that demonstrate programming for special populations. RESULTS:Among 2474 facilities, 1228 (50%) responded between October 2019 and January 2020. Programs were offered for youth (30%), elderly (40%), Spanish-speaking (37%), and sexual minority populations (39%), with 58% providing MAUD, 67% providing MOUD, and 56% providing both. Among those providing MAUD, MOUD, or both, a majority (>60% for all exposures) offered programming to vulnerable populations. With Delaware as reference, Louisiana (adjusted odds ratio [aOR], 0.28; 95% confidence interval [CI], 0.12-0.67) and North Carolina (aOR, 0.33; 95% CI, 0.15-0.72) facilities had lesser odds of offering both MAUD and MOUD. All exposures identifying facilities offering treatment to vulnerable groups were associated with offerings of MAUD and/or MOUD except for offerings to youth; these facilities had less odds of offering MOUD (aOR, 0.31; 95% CI, 0.31-0.62). CONCLUSIONS:There are facility-level disparities in providing MAUD and MOUD by state, and facilities with youth programming have lesser odds of offering MOUD than other facilities.

OBJECTIVES/OBJECTIVE:Patient experience and presence of evidence-based facility services are 2 dimensions of assessing quality of addiction treatment facilities. However, the relationship between these two is not well described. The objective of this study was to explore associations between patient experience measures and service offerings at addiction treatment facilities. METHODS:We used data from cross-sectional surveys of addiction treatment facilities and persons involved in treatment at corresponding facilities to identify facility services (eg, availability of medications for alcohol use disorder, assistance with obtaining social services, etc) and patient experience measures (overall facility rating, extent helped by treatment, ability to deal with daily problems after treatment), respectively. We used hierarchical multiple logistic regression to test for associations between top-box scores for each patient experience outcome and facility services. RESULTS:We analyzed 9191 patient experience surveys from 149 facilities. Assistance with obtaining social services (adjusted odds ratio [95% confidence interval], 0.43 [0.28-0.66]) was associated with lower overall treatment facility ratings. Childcare (2.00 [1.04-3.84]) was associated with top-box scores for extent helped. Availability of cognitive behavioral therapy (2.67 [1.25-5.73]) and childcare (1.77 [1.08-2.92]) were associated with top-box scores for ability to deal with daily problems after treatment. Assistance with obtaining social services (0.61 [0.41-0.90]) was associated with lower scores for ability to deal with problems after treatment. CONCLUSIONS:Few addiction treatment facility services were associated with patient experience measures. Future work should explore bridging the gap between evidence-based services and positive patient experiences.

OBJECTIVE:We examined drive times to outpatient substance use disorder treatment providers that provide contingency management (CM) and those that integrate CM with medication for opioid use disorder (MOUD) services in 6 US states. METHODS:We completed cross-sectional geospatial analysis among census tracts in Delaware, Louisiana, Massachusetts, North Carolina, New York, and West Virginia. We excluded census tracts with a population of zero. Using data from the 2020 Shatterproof substance use treatment facility survey, our outcome was the minimum drive time in minutes from the census tract mean center of population to the nearest outpatient CM provider, outpatient CM provider with MOUD services, and federally qualified health centers (FQHC). We stratified census tracts by 2010 Rural-Urban Commuting Area codes and by state. RESULTS:The population was greater than zero in 11,719 of 11,899 census tracts. The median drive time to the nearest CM provider was 12.2 [interquartile range (IQR), 7.0-23.5) minutes and the median drive time to the nearest CM provider increased from 9.7 (IQR, 6.0-15.0) minutes in urban census tracts to 38.8 (IQR, 25.4-53.0) minutes in rural (H = 3683, P < 0.001). The median drive time increased to the nearest CM provider with MOUD services [14.2 (IQR, 7.9-29.5) minutes, W = 18,877, P < 0.001] and decreased to the nearest FQHC [7.9 (IQR, 4.3-13.6) minutes, W = 11,555,894, P < 0.001]. CONCLUSIONS:These results suggest limited availability of CM, particularly within rural communities and for patients needing concurrent CM and MOUD treatment. Our results suggest greater adoption of CM within FQHCs could reduce urban-rural disparities in CM availability.

INTRODUCTION/OBJECTIVE:Previous studies have evaluated the implementation of standardized social determinants of health (SDOH) screening within healthcare settings, however, less is known about where screening gaps may exist following initial implementation based on facility characteristics. The objective of this study is to assess differences in screening rates for SDOH at a large, urban healthcare system. METHODS:We used electronic health record data obtained from NYC Health + Hospitals primary care sites from 2019 to 2022. We calculated the mean number of visits that were SDOH screened by visit type, facility size, and the percentages of community characteristics. We conducted 4 logistic regression models predicting the odds of screening for any SDOH and for specific SDOH needs (housing, food, and medical cost assistance) based on facility type, facility size, and the socioeconomic characteristics of the surrounding community. RESULTS:Among the 3 212 650 visits included, 16.90% were SDOH screened. Across all 4 multivariate logistic regression models predicting SDOH screening, a visit had significantly lower odds of being screened if based at a midsize or small facility, if it was a telemedicine visit, or based at a facility located in a zip-code with a higher percentage of SDOH needs. CONCLUSIONS:Our study found important differences in SDOH screening rates at a large, NYC-based health system based on size, visit type, and community level characteristics. In particular, our findings point to barriers related to facility size and telemedicine workflow that should be addressed to increase uptake of SDOH screening within different visits and facility types.

RATIONALE/BACKGROUND:During the coronavirus disease pandemic, audio-only and video telehealth visits became more widely available, but the relative patient satisfaction between telehealth and in-person modalities is not well-described. AIMS AND OBJECTIVES/OBJECTIVE:Our objective was to compare patient satisfaction with audio-only, video, and in-person adult primary care visits at a large, urban public healthcare system. METHODS:In this cross-sectional study, we used aggregated data from Press Ganey patient satisfaction surveys at 17 primary care facilities at New York City Health + Hospitals for visits between 1 June 2021 to 30 November 2021. We compared mean scores for questions common to surveys for each modality in domains of Access, Care Provider, and Overall Assessment using pairwise comparisons with two-tailed t-tests. RESULTS:There were 7,183/79,562 (9.0%) respondents for in-person visits and 1,009/15,092 (6.7%) respondents for telehealth visits. Compared to respondents for in-person visits, respondents for telehealth visits were more likely to be aged 35-64 years, Asian, and speak English as their primary language, and less likely to be ≥65 years old, Black or other race, and speak Spanish or another language as their primary language (p < 0.001). Patients reported higher mean satisfaction for Access measures for telehealth visits than in-person visits (p < 0.001). For Care Provider satisfaction questions, video visits generally had higher mean scores than in-person and, in turn, audio-only visits. For Overall Assessment questions, video visits had higher mean scores than in-person and, subsequently, audio-only visits. CONCLUSION/CONCLUSIONS:Of the visit modalities, video visits had the highest mean satisfaction scores across all domains. Telehealth may improve experiences with access, but audio-only visits may provide poorer visit experiences.

BACKGROUND:Appointment non-attendance has clinical, operational, and financial implications for patients and health systems. How telehealth services are associated with non-attendance in primary care is not well-described, nor are patient characteristics associated with telehealth non-attendance. OBJECTIVE:We sought to compare primary care non-attendance for telehealth versus in-person visits and describe patient characteristics associated with telehealth non-attendance. DESIGN/METHODS:An observational study of electronic health record data. PARTICIPANTS/METHODS:Patients with primary care encounters at 23 adult primary care clinics at a large, urban public healthcare system from November 1, 2019, to August 31, 2021. MAIN MEASURES/METHODS:We analyzed non-attendance by modality (telephone, video, in-person) during three time periods representing different availability of telehealth using hierarchal multiple logistic regression to control for patient demographics and variation within patients and clinics. We stratified by modality and used hierarchal multiple logistic regression to assess for associations between patient characteristics and non-attendance in each modality. KEY RESULTS/RESULTS:There were 1,219,781 scheduled adult primary care visits by 329,461 unique patients: 754,149 (61.8%) in-person, 439,295 (36.0%) telephonic, and 26,337 (2.2%) video visits. Non-attendance for telephone visits was initially higher than that for in-person visits (adjusted odds ratio 1.04 [95% CI 1.02, 1.07]) during the early telehealth availability period, but decreased later (0.82 [0.81, 0.83]). Non-attendance for video visits was higher than for in-person visits during the early (4.37 [2.74, 6.97]) and later (2.02 [1.95, 2.08]) periods. Telephone visits had fewer differences in non-attendance by demographics; video visits were associated with increased non-attendance for patients who were older, male, had a primary language other than English or Spanish, and had public or no insurance. CONCLUSIONS:Telephonic visits may improve access to care and be more easily adoptable among diverse populations. Further attention to implementation may be needed to avoid impeding access to care for certain populations using video visits.