Customize your JAMA Network experience by selecting one or more topics from the list below.
Prochaska JJ, Michalek AK, Brown-Johnson C, et al. Likelihood of Unemployed Smokers vs Nonsmokers Attaining Reemployment in a One-Year Observational Study. JAMA Intern Med. 2016;176(5):662–670. doi:10.1001/jamainternmed.2016.0772
Studies in the United States and Europe have found higher smoking prevalence among unemployed job seekers relative to employed workers. While consistent, the extant epidemiologic investigations of smoking and work status have been cross-sectional, leaving it underdetermined whether tobacco use is a cause or effect of unemployment.
To examine differences in reemployment by smoking status in a 12-month period.
Design, Setting, and Participants
An observational 2-group study was conducted from September 10, 2013, to August 15, 2015, in employment service settings in the San Francisco Bay Area (California). Participants were 131 daily smokers and 120 nonsmokers, all of whom were unemployed job seekers. Owing to the study’s observational design, a propensity score analysis was conducted using inverse probability weighting with trimmed observations. Including covariates of time out of work, age, education, race/ethnicity, and perceived health status as predictors of smoking status.
Main Outcomes and Measures
Reemployment at 12-month follow-up.
Of the 251 study participants, 165 (65.7) were men, with a mean (SD) age of 48 (11) years; 96 participants were white (38.2%), 90 were black (35.9%), 24 were Hispanic (9.6%), 18 were Asian (7.2%), and 23 were multiracial or other race (9.2%); 78 had a college degree (31.1%), 99 were unstably housed (39.4%), 70 lacked reliable transportation (27.9%), 52 had a criminal history (20.7%), and 72 had received prior treatment for alcohol or drug use (28.7%). Smokers consumed a mean (SD) of 13.5 (8.2) cigarettes per day at baseline. At 12-month follow-up (217 participants retained [86.5%]), 60 of 108 nonsmokers (55.6%) were reemployed compared with 29 of 109 smokers (26.6%) (unadjusted risk difference, 0.29; 95% CI, 0.15-0.42). With 6% of analysis sample observations trimmed, the estimated risk difference indicated that nonsmokers were 30% (95% CI, 12%-48%) more likely on average to be reemployed at 1 year relative to smokers. Results of a sensitivity analysis with additional covariates of sex, stable housing, reliable transportation, criminal history, and prior treatment for alcohol or drug use (25.3% of observations trimmed) reduced the difference in employment attributed to smoking status to 24% (95% CI, 7%-39%), which was still a significant difference. Among those reemployed at 1 year, the average hourly wage for smokers was significantly lower (mean [SD], $15.10 [$4.68]) than for nonsmokers (mean [SD], $20.27 [$10.54]; F(1,86) = 6.50, P = .01).
Conclusions and Relevance
To our knowledge, this is the first study to prospectively track reemployment success by smoking status. Smokers had a lower likelihood of reemployment at 1 year and were paid significantly less than nonsmokers when reemployed. Treatment of tobacco use in unemployment service settings is worth testing for increasing reemployment success and financial well-being.
Cross-sectional surveys have shown a consistent association between tobacco smoking and unemployment. A study of 52 418 construction workers in the 2006-2007 US Current Population Survey reported a greater likelihood of unemployment among smokers (229 [11.1%]) than nonsmokers (136 [6.4]%).1 In the study’s fully adjusted model with sex, age, education, ethnicity, and household income as covariates, unemployment remained a significant predictor of current smoking (odds ratio [OR], 1.51; 95% CI, 1.38-1.65). Among 68 501 adults surveyed in the California Health Interview Survey from 2007 to 2009, unemployed job seekers had the highest smoking prevalence (679 [20.9%]) relative to unemployed individuals who were not seeking a job (2652 [15.9%]) and employed individuals (7189 [14.8%]); the difference remained significant when adjusting for demographic factors and other risk behaviors (eg, obesity, binge drinking) (OR, 1.23; 95% CI, 1.01-1.49).2 Analysis of data from the French National Health Survey from the early 1990s reported a smoking prevalence of 45% among employed men vs 67% for unemployed men (OR, 2.3; 95% CI, 1.7-3.1).3 Adjusting for demographic and social-psychological variables, analysis of data from Italy’s 2003 Health Determinants Surveillance System with 4002 civilians found that smoking remained a significant correlate of unemployment status (OR, 2.23; 95% CI, 1.28-3.88).4 Among 7906 jobseekers presenting to employment agencies in Germany between 2008 and 2009, the smoking prevalence was 57.7% overall (N = 4328) and exceeded 80% (women, 80.2%; men, 84.8%) for young adults (aged 18-24 years) unemployed more than 24 months.5
While consistent, the extant epidemiologic investigations have been cross-sectional, leaving it underdetermined as to whether smoking is a cause or effect of unemployment. Tobacco use among employees is associated with greater health care costs, unproductive time, and absenteeism.6-8 An employee who smokes costs private employers in the United States an estimated excess cost (above that for a nonsmoking employee) of $5816 per year.9 Concerned about the health risks and related costs associated with tobacco use, employers are increasingly taking action to reduce smoking in the workforce.10 Smokers are not a protected class entitled to special legal protections, based on a 1987 Federal Appeals Court ruling,11 and hiring policies requiring that employees do not use nicotine are legal in more than 20 states. Hence, employers can make judgments about tobacco use among prospective employees. For example, health care and other industries have implemented testing of applicants’ urine for nicotine or cotinine (a nicotine metabolite) as a contingency for employment. Employers have prohibited tobacco use during working hours, offered financial incentives for employees to quit smoking, or charged higher medical insurance co-payments for those who continue to smoke. In many states, employers are able to fire or discipline employees who smell of tobacco smoke at work.12 Employees who have claimed nicotine addiction under the Americans with Disabilities Act have not been successful, as the courts have refused to find that addiction to cigarette smoking is a disability.13
Research has not quantified the economic burden of tobacco use for job seekers. To evaluate whether tobacco use is indeed a detriment to employability, prospective trials that observe unemployed smokers and nonsmokers through the job search process are needed. Our study, using a longitudinal observational design, sought to examine differences in reemployment success by smoking status during a 12-month period. We hypothesized that nonsmokers would be more successful than smokers in gaining reemployment. Among those reemployed at 1 year, we examined hours employed and hourly pay by smoking status. To inform tobacco treatments for job seekers, we also assessed strategies and motivations for quitting.
Question Does reemployment success differ by smoking status?
Findings In this 2-group, 12-month prospective study with 251 unemployed job seekers, nonsmokers were 30% more likely on average to be reemployed at 1 year relative to smokers. Among those reemployed at 1 year, the average hourly wage was $5 higher for nonsmokers than smokers.
Meaning Given the disparities in reemployment by smoking status, treatment of tobacco use in unemployment service settings is worth testing for increasing reemployment success and financial well-being.
Smokers and nonsmokers were recruited from September 10, 2013, to August 15, 2015, from 2 employment development departments in adjacent California counties (1 urban, San Francisco, and 1 suburban, Marin) serving a combined estimated 7000 clients annually in the San Francisco Bay Area. Both counties had 100% smoke-free nonhospitality workplace laws that banned smoking of tobacco products in an enclosed space at a place of employment, with several designated exceptions, and comparable unemployment rates at the start of recruitment (Marin, 5.0%; San Francisco, 5.2%) and study completion (Marin, 3.5%; San Francisco, 3.6%). To be eligible, smokers had to report daily smoking with a carbon monoxide breath sample more than 10 ppm; nonsmokers had to deny tobacco use in the past year with a carbon monoxide breath sample less than 10 ppm.14 Daily marijuana users were excluded, as smoked cannabis can elevate carbon monoxide levels. Participants had to be 18 years or older, English literate, unemployed, actively job seeking at the time of study enrollment, able to provide collateral sources of contact for follow-up, and not actively planning to relocate out of the area. Recruitment efforts were reactive (via flyers) and proactive (via onsite outreach).
The Stanford University Institutional Review Board approved all study procedures, participants provided written informed consent, and confidentiality was assured. The computer-delivered surveys, administered at the employment service settings, took 60 minutes at baseline and 30 minutes at follow-up. Participants received up to $100 cash for their time in the study.
Participants reported their age, sex, race/ethnicity, educational level, marital status, housing status, transportation, and height and weight to calculate body mass index (calculated as weight in kilograms divided by height in meters squared). We assessed criminal history reportable on a job application; prior treatment for alcohol or drug use, including 12-step programs; and treatment for psychological or emotional problems. A question about general health had participants rate their health as fair, poor, good, very good, or excellent. The Kessler 6 scale assessed psychological distress, scored as low (total score, <5), moderate (5-12), or high (>12).15
The primary outcome of interest was reemployment at 12-month follow-up. We defined reemployment as current hired work at least 10 hours per week or 40 hours per month. Among those reemployed, we assessed their hourly wage. At baseline, we assessed the reason for leaving their last position, duration of unemployment, past year gross income, and career cluster(s) of interest, categorized per O*Net classifications (part of the American JobCenter Network [http://www.onetonline.org]) (Table 1).
Measures of tobacco use were usual number of cigarettes per day, the Fagerström Test for Cigarette Dependence,16 stage of change for quitting smoking,17 past 30-day use of other tobacco and nicotine products, daily cost of smoking, and preference for menthol tobacco products. We assessed tobacco-related work experiences (eg, perceived discrimination owing to tobacco use) with a 4-point scale ranging from strongly disagree to strongly agree, and work-related expectations as a result of tobacco abstinence (eg, increased productivity) with a 7-point Likert scale ranging from not likely at all to extremely likely (Table 2).19
We created a measure to assess discretionary spending priorities. Smokers ordered items based on what they were most likely to purchase, assuming finite resources, using their discretionary funds, defined as money available after one’s bills are paid (Table 3). Possible rank values ranged from 1 to 13. The online survey system randomly ordered the items for presentation.
The study was powered to detect an absolute difference in reemployment of 20% between smokers and nonsmokers at 12-month follow-up (eg, 50% vs 70%). With 2-tailed testing, a 0.05 type I error rate, and 80% retention, a sample size of 120 participants per group provided greater than 80% power for detecting this group difference. Analysis of variance, χ2, and gamma tests for ordinal associations tested for group differences in baseline variables. Among those reemployed at 12 months, we performed analysis of variance tests for group differences in hours worked and hourly wage by smoking status.
As this was an observational study comparing self-selected groups (smokers vs nonsmokers), we used a propensity score (PS) design20-25 to mitigate confounding and investigate the main causal hypothesis of interest. The PS design helped to account for inherent differences between smokers and nonsmokers that could produce biased estimates of the effect of smoking on successful reemployment by balancing the distribution of covariates between smokers and nonsmokers. Under certain assumptions, a PS approach allows statistical inference to be interpreted as causal inference.22
We conducted an inverse probability weight (IPW) analysis with trimmed observations, where the weight was the inverse (ie, reciprocal) of the PS; the smoking and nonsmoking groups were weighted so as to be similar (on average) to each other in baseline characteristics. Propensity score–based IPWs are used to enhance the internal validity of an analysis, while survey sampling weights are used to support external validity or generalizability. Once each observation is weighted by its IPW, the weighted average of the 2 groups are differenced, which estimates the risk difference (RD). In our analysis, the IPW-adjusted estimand was the causal effect of smoking status on unemployment status.22
Before using IPWs to weight the observations, one must ensure that the 2 comparison groups do not have members that are completely dissimilar from the other group (ie, the groups share a common support). Design-based PS analysis starts with careful consideration of which observational units should be included in the study—using only preexposure covariates and specifically excluding any information about the outcome information. Thus, the design-based portion of a PS analysis is distinct from fitting the outcome model, which is not the immediate goal of the PS approach. One important feature of a PS design is that it identifies the set of observational units with overlapping PS values; positivity (ie, the probability of not smoking is strictly between 0 and 1) is a key assumption of the PS approach. Positivity enforces that (in one particular sense) the exposed group and the unexposed groups are not distinct in terms of their baseline covariates, thus avoiding complete confounding by baseline covariates.
To fulfill these criteria, we trimmed observations; that is, we identified and dropped observations with extreme PS values estimated using variables other than the outcome, following Crump et al.26 It would not be totally incorrect to compare trimming in observational studies with the inclusion and exclusion criteria in randomized clinical trials. One guideline for trimming is that observations with PS values outside of the interval [0.1, 0.9] should be dropped.26 Using both sets of PS values (ie, from nonsmokers and smokers), our trimming points were instead defined as the highest minimum and lowest maximum PS values (ie, 0.038 and 0.889 for the 5-covariate model; 0.124 and 0.903 for the 10-covariate model); hence, overlap was empirically determined using our real data, which closely aligns with the guideline for trimming. The full trimming-based procedure involves first estimating the PS model covariates on the full data set, trimming observations, and then reestimating these covariates using the trimmed data set; the IPWs for the main analysis are then constructed using these final PS estimates.
Analyses were done in R statistical software (R Foundation for Statistical Computing) to estimate the PS through logistic regression using the glm package, and subsequently the IPW-adjusted RD for unemployment at 12 months for nonsmokers with respect to smokers. The boot package was used to estimate the SE of the IPW-adjusted RD estimator. The primary PS was specified using a logit link function and a linear model that included a set of 5 covariates deemed most relevant to reemployment (ie, time out of work, age, education, race/ethnicity, perceived health status). We also conducted a basic sensitivity analysis comparing the primary PS results with those of a PS model that doubled the number of covariates, adding sex, housing stability, reliable transportation, criminal history, and prior treatment for alcohol or drug use. Including more covariates in a correct PS model with unknown predictors is generally expected to increase the accuracy of IPW estimators while decreasing precision. A related tradeoff is that including more covariates in the PS model tends to lead to less overlap between smokers and nonsmokers; the PS analyst must trim more, therefore reducing precision. The difference in the proportion of data trimmed between the primary and sensitivity analyses was 4-fold. Last, we used analysis of variance to test for group differences in hours worked and hourly wage by smoking status among those reemployed at 12 months.
The full sample, 131 smokers and 120 nonsmokers, was 65.7% male (N = 165), with 96 white participants (38.2%), 90 black (35.9%), 24 Hispanic (9.6%), 18 Asian (7.2%), and 23 multiracial or other race (9.2%), with a mean (SD) age of 48 (11) years and mean (SD) body mass index (calculated as weight in kilograms divided by height in meters squared) of 26.7 (5.9); 129 were never married (51.4%), 99 were unstably housed (39.4%), 72 had received prior treatment for alcohol or drugs (28.7%) and 106 for psychological or emotional problems (42.2%), 52 had a criminal history (20.7%), 70 lacked reliable transportation (27.9%), 92 had a high school degree or less (36.7%), 81 had completed some college (32.3%), and 78 had a college degree (31.1%). A majority (142 [56.6%]) of the participants were unemployed for more than 6 months; 62.0% (n = 157) reported a gross income in the past year of less than $20 000. The most common reason for leaving their last employer was that their contract ended or they were laid off (145 [57.8%]), followed by being fired (38 [15.1%]).
Table 1 compares baseline characteristics of smokers and nonsmokers. Compared with nonsmokers, smokers were significantly younger and less educated; more likely to be men, African American or multiracial, never married, unstably housed, and an urban resident; had a criminal history, unreliable transportation, and prior treatment for alcohol or drug problems; and reported poorer health. Mental health measures and body mass index did not differ by smoking status. Smokers were more chronically unemployed and reported lower past year income than did nonsmokers. Smokers were less likely than nonsmokers to be seeking employment in business; education and training; health science; marketing, sales, and service; and science, technology, engineering, or math. The reason for leaving their last employer did not differ by smoking status.
At baseline, smokers consumed a mean (SD) of 13.5 (8.2) cigarettes per day for a mean (SD) of 28 (12) years; 63 (48.1%) smoked only menthol cigarettes, 56 (42.7%) smoked only nonmenthol cigarettes, and 12 (9.2%) smoked both kinds. A majority (75 [57.3%]) of smokers reported relighting their extinguished cigarettes, with 33.6% (n = 44) doing so daily. Fagerström scores were a mean (SD) of 5.6 (1.5), indicating a moderate level of dependence; 99 of smoking participants (75.6%) smoked within 30 minutes of waking. Nearly half (62 [47.3%]) of those who smoked cigarettes used additional tobacco products in the past 30 days, including cigars (30 [22.9%]), e-cigarettes (16 [12.2%]), cigarillos (14 [10.7%]), blunts (10 [7.6%]), pipes (9 [6.9%]), chewing tobacco (5 [3.8%]), hookah (3 [2.3%]), and snus (3 [2.3%]), a form of smokeless tobacco. Few used nicotine replacement in the past 30 days, including patch (10 [7.6]%), gum (10 [7.6%]), lozenge (2 [1.5%]), and nasal spray (2 [1.5%]). Participants spent a mean (SD) of $6.49 ($4.35) per day (median, $5.00) on tobacco. Typical purchasing was by the pack (112 [85.5%]), with 21 [16.0%] purchasing cigarettes individually (ie, “loosies”), and 8 [6.1%] rolling their own cigarettes. At baseline, 61 smokers (46.6%) were in the precontemplation stage of change for quitting smoking, 44 (33.6%) in the contemplation stage, and 26 (19.8%) in the preparation stage.
Table 2 summarizes the prior attempts to quit smoking and abstinence expectancies of the sample.18 Nearly all (119 [90.8%]) had made a 24-hour attempt to quit smoking in their lifetime and 85 [64.9%] had done so in the past year. Most (93 [71.0%]) were advised to quit smoking by a health care professional, yet few used evidence-based cessation approaches. Nearly half (59 [45.0%]) reported that an employer offered them a cigarette or encouraged them to smoke, 46 (35.1%) were criticized by an employer for smoking, and 11 (8.4%) were fired owing to tobacco use. Few smokers reported support for quitting smoking from an employer (9 [6.9%]) or career counselor (10 [7.6%]).
Table 3 summarizes ratings for smokers’ discretionary spending priorities. Tobacco was ranked at the top (lowest score) ahead of basic needs and job-seeking necessities, such as transportation funds and cellular telephone costs. Nicotine replacement had the lowest mean rank. Heavier smoking was associated with a higher prioritization (lower score) of tobacco (Pearson r = –0.19; P = .04). Among the top-ranked items, 20 respondents (15.3%) selected tobacco as their first priority; 19 (14.5%) selected nutritious food, 14 (10.7%) transportation, and 13 (9.9%) cellular telephone costs.
A total of 217 participants (86.5%) completed 12-month surveys; 89 reported being reemployed (41.0%). An additional 6 participants (3 nonsmokers, 3 smokers) reported working less than 10 hours per week. Study retention was comparable for smokers (109 [83.2%]) and nonsmokers (108 [90.0%]) (χ2df = 1, 2.47; P = .12). Among those completing the 12-month survey, 60 nonsmokers (55.6%) and 29 smokers (26.6%) were reemployed at 1 year. The unadjusted RD in reemployment is 0.29 (95% CI, 0.15-0.42).
In our primary PS analysis, the PS was specified using a logit link function and a linear model that included the covariates of time out of work, age, education, race/ethnicity, and perceived health status. Observations with PS less than 0.047 or greater than 0.903 were trimmed from the original sample of 109 smokers and 108 nonsmokers. The trimmed sample contained 107 smokers and 102 nonsmokers (3.7% excluded). The RD of reemployment for nonsmokers vs smokers was 0.30 (SE = 0.09) (95% CI, 0.12-0.48). That is, after controlling for the 5 covariates of greatest concern and trimming to ensure common support, we estimate that if all participants in the study changed from being smokers to being nonsmokers then there would be a 30% increase in reemployment.
In the sensitivity analysis, the model with 10 covariates was fit on a trimmed sample with 82 smokers and 80 nonsmokers (25.3% excluded), yielding an RD estimate of 0.24 (SE = 0.08) (95% CI, 0.07-0.39). Qualitatively, this sensitivity analysis agreed with our primary analysis, and the 95% CIs also overlapped, indicating no significant difference in the estimates. A reduced effect size was found using the 10-covariate model, so additional confounders may have contributed to the observed difference in reemployment between smokers and nonsmokers.
Participants who were reemployed at 1 year worked a mean (SD) of 32 (22) hours per week, with no difference by smoking status (F(1,82) = 1.19; P = .28). Among those reemployed at 1 year, the hourly wage for smokers was significantly lower (mean [SD], $15.10 [$4.68]) than for nonsmokers (mean [SD], $20.27 [$10.54]; F(1,86) = 6.50; P = .01).
Smoking status was generally stable over time: 6 smokers (5.7%) at baseline had quit at 12 months, while 8 baseline nonsmokers (7.4%) were smoking (5 of 8 were former smokers at baseline). Among continued smokers, mean (SD) cigarettes per day declined significantly over time, from 12.6 (6.3) at baseline to 10.2 (7.4) at 12 months (paired samples tdf = 97 = 3.65; P < .001). The reduction in cigarettes per day did not differ by reemployment status (F1,98 = 0.05; P = .83).
Although tobacco use has been associated with unemployment in cross-sectional population-based studies in the United States and Europe, the mechanism and direction of that association has not been investigated prospectively. Our study examined the association of smoking with reemployment during a 12-month time frame. In our primary and sensitivity PS analyses adjusting for covariates of interest, nonsmokers had a significant advantage in reemployment at 12 months relative to smokers.
Had we randomized participants into groups of smokers or nonsmokers, then we would conclude that not smoking increased the probability of reemployment at 12 months by 12% to 48% on average. Given that nicotine is addictive and tobacco use is harmful, it would be unethical to randomize a participant to smoke vs not smoke. Instead we prospectively tracked the reemployment success of smokers and nonsmokers in the job-seeking market. That the groups were preexisting raises concern about residual confounding. We therefore conclude that there is suggestive evidence that a causal association may exist between smoking status and reemployment at 12 months.
Among smokers reemployed at 1 year, on average, their hourly income was $5 less relative to reemployed nonsmokers: $15.10 vs $20.27, a 25.5% difference. Averaging 32 hours per week, this is a deficit exceeding $8300 annually. Our findings, which were self-reported among job seekers, are comparable with wage estimates for nonsmokers ($20.71) and smokers ($17.48) reported by Berman et al9, based on the US Bureau of Labor Statistics, adjusted to 2010 levels, and discounted at 15.6% for smokers per a report of the Medical Expenditure Panel Survey.27 Combining our estimated wage gap with the sample’s report of spending about $6.50 per day on tobacco (more than $2300 per year), the findings suggest a cost, on average, of more than $10 600 annually associated with tobacco use. With nearly 3 decades of smoking and evidence of very low rates of quitting during this 12-month observational study, the financial losses to smokers are significant.
An economically disadvantaged group, with most earning less than $20 000 gross in the past year (and residing in the San Francisco Bay Area), participants reported relighting extinguished cigarettes; a preference for menthol tobacco, which is often priced more cheaply; purchasing single cigarettes, which are illegal; and smoking cigars and cigarillos, which are taxed at a lower rate and are available for individual sale. Regulatory efforts to ban menthol, increase taxes, and enforce bans on individual sales of all forms of tobacco may help promote cessation among job-seeking smokers. Notably, smokers in our sample prioritized tobacco as more important than items relevant to job seeking, such as transportation costs, cellular telephone service, and grooming needs.
Study limitations included the exclusion criteria and sample size, which, while powered for the main outcome, did not allow for tests of association within career clusters. Participants were English-literate, not intending to relocate in the next 12 months, and residing in the San Francisco Bay Area, a geographical location with a very low smoking prevalence and probably unusually high stigma about smoking. As such, study findings may not be generalizeable to all job-seeking smokers in all regions. Although limited by its observational design, our study yielded novel findings.
Employment development departments are well placed for reaching tobacco users and addressing tobacco-related health and economic disparities. Our research team is now testing the effect of a tobacco cessation intervention on time to reemployment in a randomized controlled trial with job-seeking smokers. As a “one-stop shop” for employment resources, employment service agencies could raise awareness of tobacco-related costs, wage losses, health harms, and associations with lower reemployment success and serve as a connector to low-cost cessation services such as state quit-lines.
Accepted for Publication: February 15, 2016.
Corresponding Author: Judith J. Prochaska, PhD, MPH, Stanford Prevention Research Center, Department of Medicine, Stanford University, Medical School Office Bldg, Room X316, 1265 Welch Rd, Stanford, CA 94305 (firstname.lastname@example.org).
Published Online: April 11, 2016. doi:10.1001/jamainternmed.2016.0772.
Author Contributions: Drs Prochaska and Daza had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Prochaska, Baiocchi, Rogers.
Acquisition, analysis, or interpretation of data: Prochaska, Michalek, Brown-Johnson, Daza, Baiocchi, Anzai, Grigg, Chieng.
Drafting of the manuscript: Prochaska, Baiocchi, Anzai.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Prochaska, Michalek, Daza, Baiocchi.
Obtained funding: Prochaska, Rogers.
Administrative, technical, or material support: Prochaska, Michalek, Brown-Johnson, Anzai, Rogers, Chieng.
Study supervision: Prochaska, Michalek, Rogers.
Conflict of Interest Disclosures: Dr Prochaska reported providing expert witness testimony in litigation against tobacco companies and consulting with Pfizer on smoking cessation medications. No other conflicts were reported.
Funding/Support: This study was supported by two grants from the State of California Tobacco-Related Disease Research Program: a Pilot Community-Academic Research Award 21BT-0018 and a Research Award 24RT-0035. Postdoctoral training grant T32 HL007034 from the National Heart, Lung, and Blood Institute supported Drs Brown-Johnson and Daza. Grant KHS022192A from the Agency for Healthcare Research and Quality supported Dr Baiocchi.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The article’s contents are solely the responsibility of the authors and do not necessarily represent the official views of the State of California Tobacco Related Disease Research Program; the National Heart, Lung, and Blood Institute; or the Agency for Healthcare Research and Quality.
Additional Contributions: Richard Johnson, Employment Development Department, Workforce Service Branch for San Francisco and San Mateo Counties, and Tim McClain, Marin Employment Connection, contributed to the study’s Community Advisory Board, provided input into study measures, facilitated a location for recruitment of job seekers, and participated in interpretation and dissemination of study findings. They were not compensated for their contributions. Jorge Tapia, Employment Development Department, Workforce Service Branch for San Francisco and San Mateo Counties, supported the study activities, and Racy Ming, MA, Marin Employment Connection, supported the grant proposal and provided a location for recruitment of job seekers. They were not compensated for their contributions. In addition, we appreciate the many managers and staff at both sites for their support with the study.