The Prevalence of Tobacco Use at Federally Qualified Health Centers in the United States, 2013

We explored tobacco use across federally qualified health centers (FQHCs) and compared data on state-level tobacco use between FQHC patients and the general population. We used data from the Uniform Data System (UDS) and the Behavioral Risk Factor Surveillance System (BRFSS) to generate estimates of 2013 prevalence of tobacco use among adults aged 18 years or older. According to UDS data, the overall prevalence of tobacco use was 25.8% in FQHCs compared with 20.6% in the general population represented by BRFSS data, an average of 5.2 percentage points (range, −4.9 to 20.9) higher among FQHCs. Among FQHCs, the burden of tobacco use and the opportunity for offering cessation assistance is substantial.


Objective
Tobacco use contributes to substantial illness and death in the United States (1). Although prevalence of tobacco use has declined during the past decade among some demographic groups, rates have remained steady and even increased among some socially and economically disadvantaged populations (2).
Federally qualified health centers (FQHCs), which provide comprehensive health services to economically disadvantaged populations in rural and urban communities in the United States, are required to collect data on tobacco use screening and tobacco cessation counseling rates as Uniform Data System (UDS) measures.
Understanding rates of tobacco use among FQHC clients can guide efforts to provide resources for tobacco cessation assistance where they are most needed (3-6).
Our study explores differences in tobacco use among FQHCs and compares state-level tobacco use between FQHC patients and the general population.

Methods
We used 2013 UDS FHQC data, which include quality-of-care indicators and patient demographics, to estimate tobacco use. We included only those FQHCs (967 of 1,202) that obtain tobacco use data from an electronic health record (EHR). Our denominator was the number of adults (≥18 y) having 1 or more medical visits to a FQHC in 2013. The numerator was the number of adults using any form of tobacco including cigarettes, cigars, and smokeless tobacco, as documented during routine patient care.
We estimated the prevalence of adult tobacco users in each state's FQHC population by summing the total number of tobacco users across FQHCs and dividing by the total number of adult FQHC patients. We also estimated the prevalence of tobacco use for each FQHC and calculated the median and lowest and highest values for FQHCs in each state.
We then compared data on state-level estimates of FQHC tobacco use with data from the 2013 Behavioral Risk Factor Surveillance System (BRFSS), a random-digit-dial telephone survey that collects data on population-level prevalence of health risk behaviors among US adults aged 18 years or older. Three BRFSS items are used to indicate tobacco use: 1) "Have you smoked at least 100 cigarettes in your entire life?", 2) "Do you now smoke cigarettes every day, some days, or not at all?", and 3) "Do you currently use chewing tobacco, snuff, or snus every day, some days, or not at all?" Survey participants that responded yes to question 1 and every day or some days to either question 2 or question 3 were The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors' affiliated institutions.
identified as tobacco users (7). We applied BRFSS sampling weights to estimate state-specific prevalence of tobacco use (8).

Results
In total, 1,202 FQHCs reported 2013 UDS data; 967 (80.4%) collected EHR-based tobacco use data. In this subset, the nearly 9 million adult patients seen were similar to the US population in percentage female (58.7%) and Hispanic (16.4%). FQHC patients were less likely than the US population to be older than 65 years (7.3% vs 14.1%) and more likely to be black (20.5% vs 13.0%) or other race (23.6% vs 8.6%). As expected, FQHC patients were more likely to be below the federal poverty level (71.7% vs 14.8%) and uninsured (34.8% vs 13.4%) or using government health insurance (50.8% vs 34.3%).
The overall proportion of tobacco use in FHQCs was 25.8%, and median prevalence was 29.3%, ranging from 0.4% to 94.4% across states (Table). BRFSS data from 2013 estimated US tobacco use at 20.6%, ranging from 12.1% to 30.8% across states.
Except for 5 states, state-level prevalence of tobacco use in FQHCs was higher than the BRFSS national average (Table). FQHC tobacco use prevalence and differences between FQHC and state-level estimates are displayed in the Figure  Prevalence of tobacco use among individual FQHCs varied widely, even within states. Fifty-four percent of FQHCs had a tobacco prevalence greater than 30%; 63 FQHCs had tobacco use rates higher than 50%.

Discussion
Our study is the first national assessment of the prevalence of tobacco use across FQHCs; previous reports focused on patient samples (9) or delivery of services among a subgroup of FQHCs (10). We found that in 2013 tobacco use among FQHC populations was considerably higher than for the general US population. Although the finding was not surprising, this report quantifies this difference for the first time. A second notable finding was the PREVENTING CHRONIC DISEASE wide range of tobacco use and high prevalence of tobacco use in some FQHCs, particularly in sites where more than half of adult patients use tobacco. Caring for patients in an environment where the prevalence of tobacco use is high poses substantial challenges and may require additional investment of resources to successfully offer tobacco cessation.
Assessing tobacco use rates is an important first step to targeting opportunities for intervention and quality improvement (4). Implementing clinical interventions and decision support tools to effectively act on EHR-documented tobacco use to support delivery of tobacco treatment has emerged as a national priority, especially in low-income settings (3-6).
This study has 2 main limitations. First, UDS data are collected for administrative purposes rather than for research; we cannot verify outlier values or dictate how variables are documented. Conversely, BRFSS data collection procedures are standardized but rely on self-report. In an effort to report the most robust data possible, we limited analyses to FQHCs that generated UDS quality elements using an EHR so that estimates are based on the patient population rather than a random sample of manually abstracted records. To ensure data reliability, we examined 2 prior years of UDS reporting for the top and bottom 5% of FQHC tobacco use values; 2013 reporting prevalence was similar in all cases. Second, BRFSS is able to separate data on rates of combustible tobacco use and rates of smokeless tobacco use and in 2013 reported all tobacco use at 20.6%, combustible at 19.4%, and smokeless tobacco at 4% (7). However, UDS data combine all tobacco use (combustible and smokeless), limiting our ability to report tobacco use separately.
Recommendations by the US Preventive Services Task Force to offer annual lung cancer screening using low-dose computed tomography (LDCT) to long-term smokers older than 55 years will significantly affect FQHCs caring for older adults (11). Although the UDS cannot provide information on the number of individuals eligible for lung cancer screening (data on age and packyear history are lacking), given tobacco user prevalence, the effort to implement the LDCT scans in FQHCs is substantial and will require an evaluation of costs and approaches to integrating smoking cessation (12). Understanding more about how FQHC clinicians, staff, and patients are addressing tobacco use -and how they plan to address lung cancer screening -is essential for guiding efforts to implement systems-and evidence-based practices to promote tobacco cessation and offer lung cancer screening to eligible patients.