System-Level Quality Improvement Initiatives for Tobacco Use in a Safety-Net Health System During the COVID-19 Pandemic

Association of COVID-19 and the Prevalence of In-person Versus Telehealth Primary Care Visits and Subsequent Impacts on Tobacco Use Assessment and Referral for Cessation Assistance

INTRODUCTION

The COVID-19 pandemic dramatically altered patterns of healthcare delivery. Smoking remains an important risk factor for multiple chronic conditions and may exacerbate more severe symptoms of COVID-19. Thus, it is important to understand how pandemic-induced changes in primary care practice patterns affected smoking assessment and cessation assistance.

AIMS AND METHODS

Electronic health record data from eight community health centers were examined from March 1, 2019 to February 28, 2022. Data include both telehealth (phone and video) and in-person office visits and represent 310 388 visits by adult patients. Rates of smoking assessment, provision of referral to counseling, and orders for smoking cessation medications were calculated. Comparisons by visit mode and time period were examined using generalized estimating equations and logistic regression.

RESULTS

The proportion of telehealth visits was < 0.1% 1 year prior to COVID-19 onset and, 54.5% and 34.1% 1 and 2 years after. The odds of asking about smoking status and offering a referral to smoking cessation counseling were significantly higher during in-person versus telehealth visits; adjusted odds ratios (AOR) (95% CI) = 15.0 (14.7 to 15.4) and AOR (95% CI) = 6.5 (3.0 to 13.9), respectively. The interaction effect of visit type * time period was significant for ordering smoking cessation medications.

CONCLUSIONS

Telehealth visits were significantly less likely to include smoking status assessment and referral to smoking cessation counseling compared to in-person visits. Given that smoking assessment and cessation assistance do not require face-to-face interactions with health care providers, continued efforts are needed to ensure provision at all visits, regardless of modality.

IMPLICATIONS

The COVID-19 pandemic dramatically altered patterns of healthcare-seeking and delivery with a considerable rise in telehealth visits. This study examined 1 year prior to the onset of COVID-19 and 2 years after to evaluate the assessment of tobacco use and assistance with tobacco cessation and differences during in-person versus telehealth visits. Tobacco assessment was 15 times more likely during in-person versus telehealth visits in the 2 years post onset of COVID-19. Given that telehealth visits are likely to continue, ensuring that patients are regularly assessed for tobacco regardless of visit modality is an important concern for health systems.

Factors Associated with Smoking Cessation and the Receipt of Cessation Services in a Public, Safety-Net Primary Care System

BACKGROUND

Prevalence of smoking is high among patients receiving care in safety-net settings, and there is a need to better understand patient factors associated with smoking cessation and receipt of cessation services.

OBJECTIVE

To identify patient factors associated with smoking cessation attempts and receipt of cessation counseling and pharmacotherapy in a large safety-net health system.

DESIGN

We conducted a retrospective cohort analysis using EHR data in a safety-net system in San Francisco, CA.

PARTICIPANTS

We included 7384 adult current smokers who had at least three unique primary care encounters with documented smoking status between August 2019 and April 2022.

MAIN MEASURES

We assessed four outcomes using multivariate generalized estimating equation models: (1) any cessation attempt, indicating a transition in smoking status from "current smoker" to "former smoker"; (2) sustained cessation, defined as transition in smoking status from current smoker to former smokers for two or more consecutive visits; (3) receipt of smoking cessation counseling from healthcare providers; and (4) receipt of pharmacotherapy.

KEY RESULTS

Of 7384 current adult smokers, 17.6% had made any cessation attempt, and of those 66.5% had sustained cessation. Most patients (81.1%) received counseling and 41.8% received pharmacotherapy. Factors associated with lower odds of any cessation attempt included being aged 45-64, non-Hispanic black, and experiencing homelessness. The factor associated with lower odds of sustained cessation was being male. Factors associated with lower odds of receiving counseling were being insured by Medicaid or being uninsured. Factors associated with lower odds of receiving pharmacotherapy included speaking languages other than English, being male, and identifying as racial and ethnic minorities.

CONCLUSIONS

Health system interventions could close the gap in access to smoking cessation services for unhoused and racial/ethnic minority patients in safety-net settings, thereby increasing cessation among these populations.

Telemedicine Implementation for Safety Net Populations: A Systematic Review

Background: Telemedicine systems were rapidly implemented in response to COVID-19. However, little is known about their effectiveness, acceptability, and sustainability for safety net populations. This study systematically reviewed primary care telemedicine implementation and effectiveness in safety net settings. Methods: We searched PubMed for peer-reviewed articles on telemedicine implementation from 2013 to 2021. The search was done between June and December 2021. Included articles focused on health care organizations that primarily serve low-income and/or rural populations in the United States. We screened 244 articles from an initial search of 343 articles and extracted and analyzed data from N = 45 articles. Results: Nine (20%) of 45 articles were randomized controlled trials. N = 22 reported findings for at least one marginalized group (i.e., racial/ethnic minority, 65 years+, limited English proficiency). Only n = 19 (42%) included African American/Black patients in demographics descriptions, n = 14 (31%) LatinX/Hispanic patients, n = 4 (9%) Asian patients, n = 4 (9%) patients aged 65+ years, and n = 4 (9%) patients with limited English proficiency. Results show telemedicine can provide high-quality primary care that is more accessible and affordable. Fifteen studies assessed barriers and facilitators to telemedicine implementation. Common barriers were billing/administrative workflow disruption (n = 9, 20%), broadband access/quality (n = 5, 11%), and patient preference for in-person care (n = 4, 9%). Facilitators included efficiency gains (n = 6, 13%), patient acceptance (n = 3, 7%), and enhanced access (n = 3, 7%). Conclusions: Telemedicine is an acceptable care modality to deliver primary care in safety net settings. Future studies should compare telemedicine and in-person care quality and test strategies to improve telemedicine implementation in safety net settings.

Conjoint Analysis of Telemedicine Preferences for Hypertension Management Among Adult Patients

Background: Telemedicine has been differentially utilized by different demographic groups during COVID-19, exacerbating inequities in health care. We conducted conjoint and latent class analyses to understand factors that shape patient preferences for hypertension management telemedicine appointments. Methods: We surveyed 320 adults, oversampling participants from households that earned <$50K per year (77.2%) and speak a language other than English at home (68.8%). We asked them to choose among 2 hypothetical appointments through 12 conjoint tasks measuring 6 attributes. Individual utilities for attributes were constructed using logit estimation, and latent classes were identified and compared by demographic and clinical characteristics. Results: Respondents preferred in-person visits (0.353, standard error [SE] = 0.039) and video appointments conducted through a secure patient portal (0.002, SE = 0.040). Respondents also preferred seeing a clinician with whom they have an established relationship (0.168, SE = 0.021). We found four latent classes: "in-person" (26.5% of participants) who strongly weighted in-person appointments, "cost conscious" (8.1%) who prioritized the lowest copay ($0 to $10), "expedited" (19.7%) who prioritized getting the earliest appointment possible (same/next day or at least within the next week), and "comprehensive" (45.6%) who had preferences for in-person care and telemedicine appointments through a secure portal, low copayments, and the ability to see a familiar clinician. Conclusions: Appointment preferences for hypertension management can be segmented into four groups that prioritize (1) in-person care, (2) low copayments, (3) expedited care, and (4) balanced preferences for in-person and telemedicine appointments. Evidence is needed to clarify whether aligning appointment offerings with patients' preferences can improve care quality, equity, and efficiency.

The First Year of the COVID-19 Pandemic: Changes in Preventive Services in Community Health Centers

INTRODUCTION

Community Health Centers provide comprehensive primary healthcare services to many underserved populations. It is unknown how routine preventive and chronic care services in Community Health Centers may have changed nationwide during the COVID-19 pandemic.

METHODS

The 2014-2020 Health Resources and Services Administration Uniform Data System of Community Health Centers was used, and data analysis was conducted from November 2021 to May 2022. Data for clinical quality measures in 2020 were treated as during the pandemic, whereas receipt of care in 2019 and before were treated as before the pandemic. Outcomes included 6 clinical quality measures of being up to date for colorectal cancer screening, cervical cancer screening, tobacco screening and cessation counseling, BMI screening and follow-up, depression screening and follow-up, and aspirin use for ischemic vascular disease. A mixed effects regression model was used to estimate changes in measures by year.

RESULTS

Between 2019 and 2020, receipt of preventive services declined for each of the 6 clinical quality measures: from 40.8% to 37.7% for colorectal cancer screening, from 48.8% to 44.9% for cervical cancer screening, from 85.8% to 83.4% for tobacco screening and cessation counseling, from 70.7% to 65.4% for BMI screening and follow-up, from 71.1% to 64.9% for depression screening and follow-up, and from 81.5% to 79.4% for aspirin use for ischemic vascular disease.

CONCLUSIONS

Receipt of preventive services in Community Health Centers declined during the COVID-19 pandemic for each of the 6 clinical quality measures considered in the study. Immediate action is required to support ongoing high-quality, primary healthcare services in Community Health Centers across the nation.

Contingency management to promote smoking cessation in people experiencing homelessness: Leveraging the electronic health record in a pilot, pragmatic randomized controlled trial

BACKGROUND

Cigarette smoking is disproportionately high among people experiencing homelessness (PEH). Contingency management (CM) is a strategy that has shown considerable efficacy for smoking cessation and has been used in short-term studies of smoking abstinence in PEH. We describe a pilot, pragmatic randomized controlled trial protocol, which leverages an electronic health record (EHR) infrastructure to assess the feasibility and acceptability of an extended CM intervention to improve long-term abstinence in PEH.

METHODS

We will conduct the study at three safety-net clinics in San Francisco among 90 adults experiencing homelessness who smoke cigarettes currently and have a desire to quit. We will encourage all participants to receive smoking cessation services that include behavioral counseling and pharmacotherapy through their clinics. We will randomly assign participants to an extended CM intervention group with escalating incentives contingent on abstinence or to a control group with fixed incentives for attending study visits. We will use the EHR to recruit participants, track receipt of counseling and pharmacotherapy during clinical care, and communicate with providers on participants' progress. CM participants will get escalating incentives for demonstration of carbon monoxide-verified abstinence over 6 months, with a total possible earnings of $475. Control participants will receive a fixed incentive of $5 for attending study visits, totaling $125. We will conduct the carbon-monoxide verified abstinence assessments-which will determine CM incentive amounts-daily during week 1, bi-weekly through week 4, weekly through week 13, and monthly through week 24. Measures of feasibility and acceptability, both quantitative and qualitative, will include assessments of screening and recruitment, adherence to study visits, engagement in smoking cessation clinical care, retention, and participant satisfaction. One of the primary clinical outcomes will be biochemically verified 7-day point prevalence abstinence at 6 months. We will measure secondary outcomes, which will include 7-day point prevalence abstinence at 2 weeks, 3 and 12 months.

DISCUSSION

This trial will allow us to assess the feasibility and acceptability of a CM cessation intervention among PEH. The protocol's clinical setting and use of EHRs gives it significant potential for scalability. If found to be feasible, acceptable, and subsequently efficacious in a larger trial, the intervention could reduce tobacco-related health disparities by increasing long-term smoking abstinence among this vulnerable population.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04982952. Registered on July 29, 2021.

Cigarette Smoking in Response to COVID-19: Examining Co-Morbid Medical Conditions and Risk Perceptions

During the initial wave of the Coronavirus Disease 2019 (COVID-19) pandemic in the U.S., information was mixed about the relative COVID-19 risks and potential benefits associated with cigarette smoking. Therefore, we sought to understand individual differences in the impact of COVID-19 on cigarette smoking in a sample of adults who reported recent use, with a particular focus on chronic medical conditions likely associated with increased COVID-19 risk. Participants completed an online survey of smoking behavior, demographic variables, medical history, and COVID-19 risk perceptions between July and August 2020 (N = 286). We examined whether medical conditions, COVID-19 risk perceptions and/or demographic characteristics were related to smoking changes in response to the pandemic (i.e., no change, decrease, increase) using multinomial logistical regression. Younger age, higher COVID-19 risk perceptions and Black versus White race were associated with greater odds of decreased smoking compared to no smoking change. Moreover, having at least one chronic medical condition was associated with greater odds of increased smoking relative to no change. The results have important implications for tobacco cessation treatment and preventive healthcare during the ongoing COVID-19 pandemic and other public health threats.