Abstract
Introduction.
U.S. reductions in smoking have not been experienced equally. Smoking prevalence is greater among persons of lower education, lower income, and unemployed. We evaluated whether a cessation intervention for job-seekers would result in significantly fewer cigarettes smoked per day and a greater likelihood of tobacco abstinence and re-employment, compared to the control condition at 6-months follow-up.
Methods.
Unemployed, job-seekers who smoked daily were recruited from five employment development departments in the San Francisco Bay Area, October 2015 to February 2018. Intention to quit smoking was not required. Participants were randomized to a brief motivationally-tailored, computer-assisted counseling intervention or referred to a toll-free quitline. Midstudy, 8-weeks of combination nicotine replacement was added to the intervention. Expired carbon monoxide and cotinine testing verified abstinence. Data were analyzed fall 2019.
Results.
Participants (N=360; 70% men; 43% African American, 27% non-Hispanic Caucasian; 19% unhoused) averaged 12 cigarettes/day (SD=6), 67% smoked within 30 minutes of wakening; 27% were in preparation stage to quit. During the 6-month study period, intervention participants were more likely to make a quit attempt (71% vs. 58%, p=.021) and reported significantly greater reduction in cigarettes/day than control participants (median reduction: 6.9 vs. 5.0, p=.038); however, bioconfirmed abstinence (3%) and re-employment (36%) did not differ by treatment group.
Conclusions.
In a diverse sample with economic hardships, quit attempts and smoking reduction were greater in the intervention group; however, few achieved abstinence, and neither abstinence nor re-employment differed by condition. A priority group, further research is needed on smoking and re-employment.
Clinical Trials Registration:
NCT02478697, https://clinicaltrials.gov/ct2/show/NCT02478697
Keywords: Smoking, Tobacco, Randomized Controlled Trial, Unemployment, Job-seeking
Introduction1
The US smoking prevalence has declined significantly over the past 50 years; however, inequities persist1,2 and in some areas have widened.3 Disproportionately affected by tobacco’s harms are persons of lower education, lower income, the unemployed, and those with mental illness and substance use disorders.4,5 In California, with the second lowest smoking prevalence in the nation, pooling 2012–2018 data, 16.0% of unemployed job-seekers reported smoking compared to 12.9% of employed adults.6
Tobacco use is costly. As of January 2020, the average retail price for a pack of cigarettes, including taxes, was $6.64 nationally and $8.31 in California, where the current study was conducted.7 For employers, smoking also incurs an estimated $5,816 in excess costs per employed smoker per year,8–10 including $3.6 billion annually in smoking-related abstenteeism.11
Smokers are not a protected class; employers may evaluate candidates based on smoking status, and smoking is prohibited in many workplaces.12,13 Smoking is consistently associated with unemployment.14–18 The reliance on cross-sectional designs, however, has left the causal direction undetermined.
In the only prospective study among unemployed job-seekers, smoking was associated with a lower likelihood of re-employment.19 At one-year follow-up, 27% of those who smoked compared to 56% of nonsmokers were re-employed, and among those re-employed, the average hourly wage was $5 lower among smokers compared to nonsmokers. The findings were suggestive evidence of a causal association between smoking and continued unemployment. Worth testing is treatment of tobacco use among unemployed job-seekers with the goal of securing re-employment and enhancing health and financial wellbeing.
With a randomized controlled design, and partnering with employment development departments (EDDs), the current study evaluated a brief, motivationally-tailored tobacco cessation intervention for job-seekers who smoke. Hypothesized was that at 6-months follow-up, the intervention would result in significantly fewer cigarettes smoked per day, higher tobacco quit rates, and in turn, a greater likelihood of re-employment, compared to the control group.
Materials and Methods
Sample.
Participants were recruited at five EDDs in the San Francisco Bay Area by direct approach onsite by study staff, flyers posted in the EDDs, and word-of-mouth. Research staff screened interested clients to determine eligibility. Inclusion criteria were adult age (18+); English literate; residency in the San Francisco Bay Area; unemployed or underemployed (working <40 hours in the past month or <10 hours in the past week); established current daily smoking (100+ cigarettes smoked in one’s lifetime, smoking daily, with a carbon monoxide [CO] breath sample ≥7 ppm); and active job-seeking evidenced by an updated resume, recent job application, or attendance at an onsite job seminar. Chronic unemployment (2+ years) was an exclusion criterion due to the relatively short follow-up period (6-months) and the study Community Advisory Board’s (CAB) advisement that chronic unemployment often has unique challenges (e.g., incarceration, substance use problems). A collateral contact source was required. Intention to quit smoking was not required. Leading exclusion criteria were not actively job-seeking (n=350, 31%) and nondaily smoking (n=227, 20%) (Figure 1).
Figure 1:
Total IMPACT Study CONSORT flow diagram
Procedures.
Stanford University’s Institutional Review Board approved the study procedures, designed in consultation with the study’s CAB, with representatives from the partner EDDs. Participants provided signed informed consent. Confidentiality was assured with emphasis that individual data would not be shared with EDD staff. Incentives were $25 for assessments at baseline and each follow-up (~40 minutes each), with a $25 bonus for completing all three, for a total possible incentive of $100. All participants received study-branded items (pen, post-its, folder) with the team’s contact details. Upon completing the baseline assessment, participants were randomized through an online survey system programmed to block on recruitment site, stage of change, and heaviness of smoking (>1 pack/week).20,21 Randomization was sequential, and staff and participants were blinded to the next group assignment. Study staff communicated randomized assignments to participants and then implemented the procedures for the intervention or control condition appropriately. Follow-ups could be completed remotely. Recruitment was conducted October 2015 to February 2018, with the follow-up period January 2016 to September 2018.
Assessments.
Measures used in prior research with the population of interest19 assessed participants’ gender; age; race/ethnicity; education; income; marital status; housing situation; access to reliable transportation; last employment; reason(s) the employment ended; and experiences of being criticized at work for smoking; or job loss for smoking. Participants indicated any criminal record reportable on a job application. Health-related measures included a global health rating (poor to excellent); health insurance status; past-year medical and dental visits; physical limitations to work (yes/no); physical pain interference with work (not at all, a little bit, moderately, quite a bit, extremely); frequency of alcohol use; past 30-day problems due to alcohol (never, rarely, sometimes, often, always); and past-week depressive symptoms assessed with the 10-item Center for Epidemiologic Studies Depression scale (CESD; total scores: 0–30; 10+ indicates depression).22
At baseline, participants reported their usual number of cigarettes smoked per day; time to first cigarette upon wakening;23 daily tobacco costs; purchasing of individual cigarettes (“loosies”); menthol smoking; and other tobacco use, including current cigar uses, past-year blunt use, and lifetime, past-year, and current e-cigarette use. Stage of change for quitting smoking was assessed as precontemplation (no intention to quit), contemplation (intention to quit in next 6 months); or preparation (30-day intention to quit, and past year 24-hour quit attempt).24. Past-week exposure to secondhand smoke and living with others who smoke were assessed.
The primary outcome was quitting smoking. Tobacco abstinence was self-reported and biochemically verified at follow-ups as no tobacco use (including e-cigarettes) in the past 7 days. Consensus guidelines recommend 7-day point prevalence abstinence for cessation induction studies with individuals unprepared to quit smoking.25,26 Nicotine replacement therapy (NRT) use was reported. At follow-ups, participants reporting no tobacco use in the past 7 days provided an expired CO sample, tested using a Bedfont Smokerlyzer. Participants with CO<7 ppm, who reported no e-cigarette or NRT use, provided a saliva sample for testing cotinine with an Accutest® NicAlertTM test strip. Salivary cotinine levels <15 ng/ml confirmed nonsmoking. Change in cigarettes smoked per day was calculated from baseline to 3- and 6-months follow-up as a secondary outcome. Additionally, making a 24-hour quit attempt in the past 3-months was assessed at each follow-up, and those reporting a quit attempt were asked about specific quit strategies, including: taking cessation medication; calling the state quitline; referring to study-provided print materials; gradual reduction to quit; quitting “cold turkey”; and using e-cigarettes, given their growth in popularity and marketing as switch devices.
The employment outcome of interest was securing a position for at least 80 cumulative hours, to exclude one-time gigs and temporary work, assessed at 3- and 6-month follow-ups.
Intervention.
For privacy and delivery within EDDs, and with consideration of portability for future dissemination, the study CAB encouraged use of technology. All partner sites had computer stations. Printed materials for participants to keep also were encouraged, as was a motivationally-tailored approach, because a minority of job-seekers was anticipated to be prepared to quit smoking in the near future. For dissemination considerations, the intervention initially provided referrals for low/no-cost cessation medications in the community. When access proved challenging, the intervention was altered midstudy to provide combination NRT at no cost to intervention participants.
The intervention package consisted of a transtheoretical model (TTM)-tailored computer-delivered program completed at baseline and 3-months follow-up with printed individualized reports that reflected changes in participants’ responses over time; a stage-tailored printed manual emphasizing motivational, cognitive-behavioral, and relapse prevention strategies; and a brief check-in session at baseline with a research team member trained in tobacco cessation and motivational interviewing. The printed manual had quotes and photos from job-seekers,27 information on smoking and re-employment, and best practices for concealing evidence of tobacco use.19 Total intervention contact time was ~45 minutes. The TTM computer-tailored intervention approach was chosen because it has been shown efficacious in worksite settings for targeting smoking and other risk behaviors.20,28 Combined counseling and cessation medications is a best practice for treating tobacco use;29 and we expected that offering treatment within the EDD setting would optimize engagement, relative to referral to the state quitline.
Given low access to low/no-cost cessation medication, midway through study recruitment, 8 weeks of study-provided combination NRT (patch + gum/lozenge) was offered to intervention participants smoking 5+ cigarettes daily, without medical contraindications. Proper usage guidelines, including dosing and tapering instructions, were provided.
Control.
Although treating tobacco use is atypical in EDDs, a no-treatment control condition was considered unethical. The control group received the American Heart Association’s “Quit Smoking for Good” pamphlet and referral to the California quit-line.
Power Calculation.
With 7-day abstinence estimates of 5% and 15% at 6-months in the control and intervention groups, respectively, a required sample size of 140 to 148 per group would result in a minimum statistical power of .80 with type I error of 0.05, for a two-tailed test of the hypothesis of a difference between groups in abstinence. Anticipating 20% attrition, a sample size of N=360 was sought. Further, two groups of 112 each had 80% power to detect a moderately sized effect of changing rates of employment from 30% to 50%. Change in cigarettes per day as a continuous measure was anticipated sufficiently powered with sample size based upon the dichotomous outcomes of abstinence and re-employment.
Analyses.
Data were analyzed fall 2019. Chi-squares tested for group differences; Fisher’s Exact tests were examined when expected values in any cell were <5. A Mann-Whitney U test examined group differences with continuous, non-normally distributed outcomes. We tested outcomes both for complete cases (i.e., missing=missing) and by assuming those lost to follow-up continued to smoke and remained unemployed (i.e., missing=smoking and unemployed).
Results
Descriptive Characteristics.
The study groups were balanced on all measured baseline variables (Table 1). The sample (N=360) was 70% men, averaging 43 years of age (SD=11), identifying as 43% African American, 27% non-Hispanic Caucasian, 8% Hispanic, and 17% multiracial; 56% were single/never married; 78% did not have a college degree; 22% were unhoused; 76% lived in an urban area. Among those reporting income, 73% earned <$25,000 a year; 78% had access to reliable transportation. Most (72%) reported regular employment within the past 6 months. Top reason(s) employment had ended was lay-off or finished contract work (32%), being fired (18%), quitting (10%), relocating (8%), or a physical/mental health condition (8%). A sizeable minority (23%) reported being criticized at work for smoking; 3% were fired for smoking cigarettes; 21% indicated criminal history reportable on a job application.
Table 1:
Baseline Characteristics of the Sample by Condition and Overall
| Intervention n=179 | Control n=181 | Full Sample N=360 | |
|---|---|---|---|
| Age: M (SD) | 42.94 (11.23) | 42.95 (11.52) | 42.95 (11.36) |
| Gender | |||
| Male | 70% (125) | 70% (127) | 70% (252) |
| Female | 30% (54) | 29% (52) | 29% (106) |
| Transgender | 0% (0) | 1% (2) | 1% (2) |
| Race / Ethnicity | |||
| African-American | 41% (73) | 45% (81) | 43% (154) |
| Non-Hispanic Caucasian | 25% (53) | 30% (45) | 27% (98) |
| Hispanic | 8% (15) | 7% (12) | 8% (27) |
| Multiracial | 15% (27) | 19% (35) | 17% (62) |
| Other/Unknown | 6% (11) | 4% (8) | 5% (19) |
| Education Level | |||
| < HS degree | 10% (18) | 18% (33) | 14% (51) |
| HS degree/GED | 36% (65) | 29% (53) | 33% (118) |
| Some College | 29% (52) | 33% (59) | 31% (111) |
| College Degree | 20% (36) | 16% (29) | 18% (65) |
| Graduate Degree | 4% (8) | 4% (7) | 4% (15) |
| Marital Status | |||
| Single/Never Married | 56% (100) | 55% (100) | 56% (200) |
| Married/Cohabitating | 18% (33) | 19% (35) | 19% (68) |
| Divorced/Separated/Widowed | 26% (46) | 25% (46) | 25% (92) |
| Income Level | |||
| 10,000 or less | 33% (59) | 30% (54) | 31% (113) |
| 11,000 to 25,000 | 18% (32) | 20% (37) | 19% (69) |
| 26,000 to 50,000 | 11% (20) | 12% (21) | 11% (41) |
| Greater than 50,000 | 7% (13) | 8% (15) | 8% (28) |
| Refused/Don’t know | 31% (55) | 30% (54) | 30% (109) |
| County | |||
| Suburban | 21% (37) | 27% (49) | 24% (86) |
| Urban | 79% (142) | 73% (132) | 76% (274) |
| Living Situation | |||
| Rent/Own Home | 38% (68) | 37% (67) | 38% (135) |
| Friend/Relative’s Home | 25% (44) | 25% (45) | 25% (89) |
| Treatment Center | 5% (9) | 6% (11) | 6% (20) |
| SRO/Hotel/Motel | 8% (14) | 11% (20) | 9% (34) |
| Unhoused | 24% (42) | 20% (36) | 22% (78) |
| Other/Refused | 1% (2) | 1% (2) | 1% (4) |
| Access to Reliable Transportation | 78% (139) | 78% (142) | 78% (281) |
| Own Automobile | 25% (45) | 23% (42) | 24% (87) |
| Public Transportation | 56% (100) | 59% (107) | 58% (207) |
| Other (e.g., friends, walking, biking) | 19% (34) | 17% (32) | 18% (66) |
| Health Insured | 77% (137) | 77% (139) | 77% (276) |
| Saw a Healthcare Provider in Past Year | 68% (122) | 73% (132) | 71% (254) |
| Saw a Dentist in Past Year | 44% (62) | 34% (78) | 39% (140) |
| Physical Disability Affecting Work | 16% (29) | 19% (35) | 18% (64) |
| Physical Pain Interfering with Work | |||
| Moderately to Extremely | 26% (46) | 26% (47) | 26% (93) |
| Self-Rated Health | |||
| Excellent/Very Good | 44% (79) | 40% (73) | 42% (152) |
| Good | 40% (71) | 40% (72) | 40% (143) |
| Fair/Poor | 16% (28) | 20% (36) | 18% (63) |
| CESD Depression Score: M (SD) | 10.07 (6.19) | 10.04 (6.33) | 10.05 (6.26) |
| CESD Depression Score >=10 | 46% (83) | 50% (91) | 48% (174) |
| Time since last Regular Employment | |||
| < 3 months | 57% (102) | 52% (95) | 55% (197) |
| 3.1 – 6 months | 16% (28) | 18% (32) | 17% (60) |
| 6.1 – 12 months | 13% (24) | 23% (41) | 18% (65) |
| 12.1 – 24 months | 14% (25) | 7% (13) | 10% (38) |
| Reason(s) Last Employment Ended | |||
| Laid Off or Contract Work Ended | 34% (58) | 31% (57) | 32% (115) |
| Fired | 15% (27) | 20% (36) | 18% (63) |
| Quit | 10% (18) | 6% (18) | 10% (36) |
| Relocated | 7% (13) | 9% (16) | 8% (29) |
| Physical / Mental Health | 9% (16) | 9% (16) | 8% (32) |
| Other (e.g., return to school, family) | 26% (47) | 23% (42) | 25% (89) |
| Criticized at Work for Smoking | 21% (37) | 26% (47) | 23% (84) |
| Fired for Smoking | 3% (5) | 4% (7) | 3% (12) |
| Reportable Criminal History | 18% (33) | 24% (44) | 21% (77) |
| Cigarettes Smoked per Day: M(SD) | 12.27 (6.20) | 11.46 (6.28) | 11.86 (6.24) |
| Smokes Menthol Cigarettes | 64% (115) | 63% (114) | 64% (229) |
| Daily Costs of Tobacco: Median (IQR) | $6.00 (4 – 9) | $6.00 (4 – 9) | $6.00 (4 – 9) |
| Purchases Single Cigarettes (“loosies”) | 17% (30) | 20% (37) | 19% (67) |
| Time to First Cigarette upon Waking | |||
| Within 5 minutes | 31% (56) | 35% (62) | 33% (118) |
| 6 – 30 minutes | 35% (63) | 32% (58) | 34% (121) |
| 31 – 60 minutes | 16% (28) | 15% (27) | 15% (55) |
| > 60 minutes | 17% (31) | 18% (33) | 18% (64) |
| Stage of Change for Quitting Smoking | |||
| Precontemplation | 33% (59) | 33% (59) | 33% (118) |
| Contemplation | 41% (74) | 39% (71) | 40% (145) |
| Preparation | 26% (46) | 28% (51) | 27% (97) |
| Other Tobacco Product Use | |||
| Current Cigar / Cigarillo Use | 9% (17) | 9% (17) | 9% (34) |
| E-cigarette Use | |||
| Lifetime | 49% (88) | 44% (80) | 47% (168) |
| Past year | 32% (58) | 35% (63) | 34% (121) |
| Current | 2% (4) | 4% (7) | 3% (11) |
| Past Year Blunt Use | 38% (68) | 35% (64) | 37% (132) |
| Past Year Alcohol Use* | |||
| None | 27% (49) | 25% (46) | 26% (95) |
| Monthly or Less | 16% (29) | 21% (39) | 19% (68) |
| 2 to 4 times a Month | 23% (41) | 22% (39) | 22% (80) |
| 2 or more times a Week | 32% (59) | 30% (55) | 32% (114) |
| Problems with Alcohol Past 30 Days* | |||
| Never | 63% (113) | 61% (109) | 62% (222) |
| Rarely / Sometimes | 30% (53) | 31% (56) | 30% (109) |
| Often / Always | 7% (12) | 8% (15) | 8% (27) |
| Past Week Secondhand Smoke Exposure | 74% (132) | 72% (128) | 73% (260) |
| Lives with Someone who Smokes | 61% (108) | 51% (91) | 56% (199) |
| Lives with 2+ People who Smoke | 34% (61) | 31% (56) | 33% (117) |
Three people had missing data on past year alcohol use and two did not respond to the item on problems with alcohol in the past 30 days.
Most (82%) rated their health as good to excellent; 77% had health insurance; 71% had seen a healthcare provider in the past year; 39% had seen a dentist; 26% reported moderate to extreme pain interfered with their work; and 18% reported a physical disability affected their work. Nearly half (48%) scored as depressed; 38% reported past-month alcohol problems.
Tobacco Use.
At baseline, the sample averaged 12 cigarettes per day (SD=6); 64% smoked menthol cigarettes; 67% smoked within 30 minutes of wakening, an indicator of addiction. The sample spent a median of $6.00 a day on tobacco (interquartile range [IQR] $4.00–$9.00), going from $6.00 pre-April 1, 2017 (IQR $4.00–$8.00) to $10.00 post-April 1, 2017 (IQR $5.00–$12.00). Purchasing of “loosies” was reported by 19% of participants.
Baseline stage distribution for quitting smoking was 33% precontemplation, 40% contemplation, and 27% preparation. In addition to smoking cigarettes, 9% currently smoked cigars/cigarillos; 37% reported past-year blunt use; and e-cigarette use was 47% lifetime, 34% past-year, and 3% current. Most (73%) reported past-week secondhand smoke exposure; 56% lived with someone who smokes cigarettes; and 33% lived with 2+ people who smoke.
Treatment Fidelity.
Of the 179 participants randomized to the intervention, 173 (97%) completed the computer-based program and check-in session at baseline and 119 (66%) at the 3-month follow-up (Figure 1). Of the 89 intervention participants offered NRT, 51 (57%) accepted. All 181 control participants received the quit smoking brochure and quitline referral.
Retention.
At 3 months, retention was 76% overall: 79% control vs. 72% intervention, X2df=1=2.76, p=0.097 (Figure 1). Before the 3-month follow-up, one intervention participant died from cardiac arrest. The individual had not received study-provided NRT, and cause of death was considered unrelated to study participation. At 6 months, retention was 88% overall; 89% control vs. 86% intervention, X2df=1=0.70, p=.403. Between the 3- and 6-month follow-up, one control participant died by drug overdose, considered unrelated to study participation.
24-hour Quit Attempts and Quit Strategies.
During the 6-month study period, 109 (71%) intervention and 97 (58%) control participants reported at least one 24-hour quit attempt (X2df=1=5.31, p=.021). Among participants making a 24-hour quit attempt: 75 (69%) intervention and 69 (71%) control participants quit “cold turkey” (X2df=1=0.13, p=.716); 59 (54%) intervention and 43 (44%) control participants reduced to quit (X2df=1=1.97, p=.160); 47 (43%) intervention and 33 (34%) control participants used cessation medication (X2df=1=1.79, p=.181); 34 (31%) intervention and 14 (14%) control participants referred to study-provided written materials (X2df=1=8.07, p=.005); 14 (13%) intervention and 16 (17%) control participants used an e-cigarette (X2df=1=0.55, p=.458); and 3 (3%) intervention and 1 (1%) control participant called the state quitline, (X2 df=1=0.80, p=.371).
Tobacco Abstinence.
Among the complete cases, bioconfirmed quit rates at the 3-month follow-up were 2 of 129 (1%) in the intervention and 0 of 144 (0%) in the control group (Fisher’s Exact p=.222). One additional control participant self-reported abstinence but failed to submit for bioconfirmation. Excluding a deceased individual and inferring participants lost to follow-up to be smoking, bioconfirmed quit rates at the 3-month follow-up were 2 of 178 (1%) in the intervention and 0 of 181 (0%) in the control group (Fisher’s Exact p=0.245).
At the 6-month follow-up, 26 participants (8%) self-reported no tobacco use, of whom 10 (38%) were bioconfirmed as tobacco-free. Of the 16 (62%) not bioconfirmed, 10 did not submit for testing (3 intervention, 7 control), and 4 failed either the CO or cotinine test (2 intervention, 2 control). Two additional participants (1 intervention, 1 control) reported using e-cigarettes at the 6-month assessment with no other tobacco products; however, neither submitted for CO-biconfirmation of being smoke-free. Among the complete cases, bioconfirmed quit rates at the 6-month follow-up were 5 of 154 (3%) in the intervention and 5 of 160 (3%) in the control group (Fisher’s Exact p=1.000). Excluding deceased individuals (n=2) and inferring participants lost to follow-up to be smoking, bioconfirmed quit rates at the 6-month follow-up were 5 of 178 (3%) in the intervention and 5 of 180 (3%) in the control group (Fisher’s Exact p=1.000).
Smoking Reduction.
At the 3-month follow-up, intervention participants reported reducing their smoking by a median of 6.0 cigarettes per day (IQR: 2.1–9.8) compared to 3.6 cigarettes per day (IQR: 0.2–8.4) in the control group; Mann Whitney U test p=.002. By the 6-month follow-up, intervention participants had reduced their smoking by a median of 6.9 cigarettes per day (IQR: 2.7–11.0) compared to 5.0 cigarettes per day (IQR: 1.4–9.4) in the control group; Mann Whitney U test p=.038.
Re-Employment Status.
At the 3-month follow-up, among the complete cases, re-employment rates were 56 of 144 (39%) in the intervention and 44 of 129 (34%) in the control group, X2df=1=0.67, p=0.413. Excluding a deceased individual and inferring participants lost to follow-up to be unemployed, re-employment rates were 56 of 178 (31%) in the intervention and 44 of 181 (24%) in the control group, X2df=1=1.73, p=0.189.
At the 6-month follow-up, among the complete cases, re-employment rates were 62 of 154 (40%) in the intervention and 69 of 161 (43%) in the control group, X2df=1=0.22, p=0.640. Excluding deceased individuals (n=2) and inferring participants lost to follow-up to be unemployed, re-employment rates were 62 of 178 (35%) in the intervention and 69 of 180 (38%) in the control group, X2df=1=0.47, p=0.492.
Conclusions
Despite elevated smoking prevalence among the unemployed, this is the first randomized controlled trial to evaluate a cessation intervention with unemployed job-seekers. The intervention provided counseling and cessation medication, consistent with clinical practice guidelines;29 was informed by formative work and CAB input;19,27 and was stage-tailored, assuming a minority was prepared to quit in the next month. Computer-delivery maximized integration into EDD settings, and we found that nearly all intervention participants completed the first counseling session, and two-thirds completed the second. In contrast, only 2% of the sample called the state quitline, which is comparable to national estimates.30
During the 6-month study period, intervention participants were more likely to make a 24-hour quit attempt and to report referring to their study materials during a quit attempt. From baseline to 3- and 6-months, we observed significant reductions in cigarettes per day among participants in the intervention group. However, few (3%) in the intervention group or overall, achieved abstinence; hence, we were unable to test the effect of quitting smoking on success with re-employment. That nearly two-thirds of the sample made a 24-hour quit attempt, yet only 3% of the sample was bioconfirmed abstinent at 6-months, suggests motivation to be tobacco-free but clear challenges to sustaining abstinence. Notable challenges included high levels of nicotine addiction, depression, alcohol problems, physical pain, poverty, housing insecurity, and secondhand smoke exposure. As one participant summarized at study end, “I want to quit but don’t look forward to withdrawal.”
By 6-months follow-up, over a third of the sample was re-employed. The higher re-employment rate observed in the current study (36% at 6-months) compared to the prior study in the same urban area (27% at 1-year), is likely reflective of the strengthened economy; unemployment hit record lows (below 3%) during the trial.31 When unemployment rises, the association between smoking and difficulty finding work may be stronger. There are few prospective studies of smoking and employment. A study of change from insecure (i.e., fixed term) to secure (i.e., permanent) employment found an increase in job security and job satisfaction, with no associated change in smoking behavior.32
All participants were recruited from the San Francisco Bay Area; many relied on others for housing or were unhoused. Though facing economic hardships, most had access to reliable transportation. San Francisco provides universal healthcare, and most participants had health insurance, had seen a healthcare provider in the past year, and rated their physical health as good to excellent. Prior research has documented diversion of funds from basic needs such as food and even housing toward cigarettes among low-income individuals who smoke.33–35 In a qualitative study of unemployed job-seekers who smoked, cigarettes were characterized as a “protected purchase,” for which essential household items were sacrified to maintain one’s tobacco addiction.36 Participants in the current study spent a median of $6 per day on tobacco. Following California’s $2.00 per pack tax increase, the median cost rose to $10 per day, which over a year would be $3650. Notably, cigarettes are priced significantly lower in neighborhoods with a greater proportion of low-income residents.37 Nearly one in five participants purchased single “loosies,” which are prohibited for sale in the US. That some participants reported spending no money on tobacco suggests panhandling or “bumming” cigarettes also occurred.
Study strengths include a demographically diverse sample; the randomized design; attention to all forms of tobacco product use; the community partnership; and the 88% (n=315) 6-month retention. Study exclusion criteria were kept to a minimum, and intention to quit smoking was not required for participation. Top reasons for exclusion were not actively job-seeking and nondaily tobacco use or low CO reading, followed by working >40 hours in the past month. Together, this represents over two-thirds of those excluded. The incentive was appropriate given study time and travel demands.
Regarding study limitations, measures were self-reported, participants were recruited from a single geographic area, and follow-up was limited to 6-months. Findings may not generalize to rural areas. The follow-up duration was likely insufficient for moving a sample that at baseline was largely unprepared to quit smoking to quit in the next 30 days. Many participants were facing immediate stressors of economic hardship and were focused on finding re-employment. While the financial costs and negative stigma of smoking were salient motivaters to quit, the drive for nicotine and stressful discomfort of nicotine withdrawal were strong, and quitting smoking remained more of a future goal for many. As one participant expressed at the final survey, “I… thank staff for the support they gave me to help me quit smoking… goal of quitting by my birthday is with [sic] reach.” Quit rates likely would have been higher had the sample been restricted to those prepared to quit in the next 30 days, but then only a minority would have been eligible, and the aim was to have a program relevant for all job-seekers who smoke. Future efforts should attend to co-occurring depressive symptoms and alcohol problems, housing insecurity, and high levels of secondhand smoke exposure.
In a diverse sample of job-seekers, quit attempts and smoking reduction were greater in the intervention group; however, few successfully quit smoking, and neither abstinence nor re-employment significantly differed by treatment group in this sample. As this is a priority group with economic hardships to aid in smoking cessation, further research is needed.
Highlights.
Tobacco use is increasingly concentrated among those of lower income and unemployed.
In a randomized trial, we tested a quit smoking intervention with job-seekers.
Quit attempts and smoking reduction were greater in the intervention group.
Few achieved abstinence; neither abstinence nor re-employment differed by condition.
A priority group, further research is needed.
Acknowledgments:
We acknowledge Richard Johnson, Amy Rogers, Mia Grigg, and Tim McClain for their service and contributions to the study’s Community Advisory Board, providing input into study measures and intervention approaches, facilitating locations for recruitment of job-seekers, and participating in interpretation and dissemination of study findings. We acknowledge Beatriz Anguiano, Anne Michalek, and Peter Soyster who assisted with study recruitment, assessments, and intervention delivery. We acknowledge Dr. Priya Fielding-Singh for running initial descriptive analyses of the data. We also appreciate the many managers and staff at the participating sites for their support with the study. The computerized tailored tobacco cessation program and print workbook based on the Transtheoretical Model were developed by Pro-Change Behavior Systems, Inc. and modified for use with a community of unemployed job-seekers.
Funding: Conduct of the study was supported by the State of California Tobacco Related Disease Research Program (TRDRP) Research Award #24RT-0035. The National Heart, Lung and Blood Institute (NHLBI) postdoctoral training grant #T32 HL007034 supported CBJ on this study. The article’s contents are solely the responsibility of the authors and do not necessarily represent the official views of TRDRP or NHLBI. A grant from the Agency for Health Research & Quality provided support for some of Professor Baiocchi’s time (KHS022192A).
Abbrieviations:
- EDD
Employment Development Department
- CO
carbon monoxide
- CAB
Community Advisory Board
- CESD
Center for Epidemiologic Studies Depression
- NRT
Nicotine Replacement Therapy
Footnotes
Financial Disclosure: Judith Prochaska has consulted to technology and pharmaceutical companies focused on smoking cessation and has served as an expert in litigation against the tobacco companies. All other authors have no financial disclosures.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.U.S. Department of Health and Human Services. The Health Consequences of Smoking: 50 Years of Progress A Report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2014. [Google Scholar]
- 2.Leas EC, Schleicher NC, Prochaska JJ, Henriksen L . Place-Based Inequity in Smoking Prevalence in the Largest Cities in the United States . JAMA Internal Medicine. 2019;179(3):442–444. 10.1001/jamainternmed.2018.5990 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mills SD, Golden SD, Queen TL, Kong AY, Ribisl KM . Are state-level income-based disparities in adult smoking declining? Preventive Medicine. 2020;133:106019 10.1016/j.ypmed.2020.106019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lasser K, Boyd J, Woolhandler S. Smoking and Mental Illness: A Population-based Prevalence Study Year Book of Psychiatry and Applied Mental Health. 2002;2002(1):179–180. [Google Scholar]
- 5.Centers for Disease Control and Prevention. Vital signs: current cigarette smoking among adults aged≥ 18 years--United States, 2005–2010. MMWR Morbidity and mortality weekly report. 2011;60(35):1207. [PubMed] [Google Scholar]
- 6.Current smoking status – adults age 18–65 by current employment status. http://ask.chis.ucla.edu. Accessed October 26, 2019.
- 7.Boonn A State excise and sales taxes per pack of cigarettes total amounts & state rankings. Campaign for Tobacco-Free Kids. https://www.tobaccofreekids.org/assets/factsheets/0202.pdf. Published 2020. Accessed March 16, 2020.
- 8.Bunn WB III, Stave GM, Downs KE, Alvir JMJ, Dirani R. Effect of smoking status on productivity loss. J Occup Environ Med. 2006;48(10):1099–1108. [DOI] [PubMed] [Google Scholar]
- 9.Goetzel RZ, Carls GS, Wang S, et al. The relationship between modifiable health risk factors and medical expenditures, absenteeism, short-term disability, and presenteeism among employees at Novartis. Journal of Occupational and Environmental Medicine. 2009;51(4):487–499. [DOI] [PubMed] [Google Scholar]
- 10.Berman M, Crane R, Seiber E, Munur M. Estimating the cost of a smoking employee. Tobacco Control. 2014;23(5):428–433. 10.1136/tobaccocontrol-2012-050888 [DOI] [PubMed] [Google Scholar]
- 11.Asay G, Roy K, Lang J, Payne R, Howard D. Absenteeism and Employer Costs Associated With Chronic Diseases and Health Risk Factors in the US Workforce. Preventing chronic disease. 2016;13:E141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Asch DA, Muller RW, Volpp KG. Conflicts and compromises in not hiring smokers. New England Journal of Medicine. 2013;368(15):1371–1373. [DOI] [PubMed] [Google Scholar]
- 13.Palmer v. Liggett Group, Inc, 825 620(Court of Appeals, 1st Circuit 1987).
- 14.Okechukwu C, Bacic J, Cheng K-W, Catalano R. Smoking among construction workers: The nonlinear influence of the economy, cigarette prices, and antismoking sentiment. Social Science & Medicine. 2012;8(75):1379–1386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Prochaska JJ, Shi Y, Rogers A. Tobacco use among the job-seeking unemployed in California . Preventive Medicine. 2013;5(56):329–332. [DOI] [PubMed] [Google Scholar]
- 16.Khlat M, Sermet C, Le Pape A. Increased Prevalence of Depression, Smoking, Heavy Drinking and Use of Psycho-Active Drugs among Unemployed Men in France. European Journal of Epidemiology. 2004;19(5):445. [DOI] [PubMed] [Google Scholar]
- 17.De Vogli R, Santinello M. Unemployment and smoking: does psychosocial stress matter? Tobacco control 2005;14(6):389–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Freyer-Adam J, Gaertner B, Tobschall S, John U. Health risk factors and self-rated health among job-seekers. BMC Public Health. 2011;11(1):659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Prochaska JJ, Michalek AK, Brown-Johnson C, et al. Likelihood of unemployed smokers vs nonsmokers attaining reemployment in a one-year observational study. JAMA internal medicine. 2016;176(5):662–670. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Prochaska JJ, Velicer WF, Prochaska JO, Delucchi K, Hall SM. Comparing intervention outcomes in smokers treated for single versus multiple behavioral risks. Health Psychology. 2006;25(3):380–388. 10.1037/0278-6133.25.3.380 [DOI] [PubMed] [Google Scholar]
- 21.Gritz ER, Carr CR, Rapkin D, et al. Predictors of long-term smoking cessation in head and neck cancer patients. Cancer Epidemiology and Prevention Biomarkers. 1993;2(3):261–270. [PubMed] [Google Scholar]
- 22.Andresen EM, Malmgren JA, Carter WB, Patrick DL. Screening for depression in well older adults: Evaluation of a short form of the CES-D . American journal of preventive medicine. 1994;10(2):77–84. [PubMed] [Google Scholar]
- 23.Heatherton TF, Kozlowski LT, Frecker RC, Rickert W, Robinson J. Measuring the heaviness of smoking: Using self-reported time to the first cigarette of the day and number of cigarettes smoked per day. British Journal of Addiction. 1989;84(7):791–800. 10.1111/j.1360-0443.1989.tb03059.x [DOI] [PubMed] [Google Scholar]
- 24.Prochaska JO, DiClemente CC, Norcross JC. In search of how people change: Applications to addictive behaviors. Am Psychol. 1992;47(9):1102. [DOI] [PubMed] [Google Scholar]
- 25.Hughes JR, Keely JP, Niaura RS, et al. Measures of abstinence in clinical trials: Issues and recommendations. Nicotine Tob Res. 2003;5(1):13–25. [PubMed] [Google Scholar]
- 26.Velicer WF, Prochaska JO. A comparison of four self-report smoking cessation outcome measures. Addictive Behaviors. 2004;29(1):51–60. [DOI] [PubMed] [Google Scholar]
- 27.Michalek AK, Wong SL, Brown-Johnson CG, Prochaska JJ. Smoking and Unemployment: A Photo Elicitation Project. Tobacco Use Insights 2020;13:1179173X20921446 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Prochaska JO, Butterworth S, Redding CA, et al. Initial efficacy of MI, TTM tailoring and HRI’s with multiple behaviors for employee health promotion. Preventive Medicine. 2008;46(3):226–231. 10.1016/j.ypmed.2007.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tobacco Use and Dependence Guideline Panel. Treating Tobacco Use and Dependence: 2008 Update. Agency for Healthcare Research and Quality. https://www.ahrq.gov/professionals/clinicians-providers/guidelines-recommendations/tobacco/index.html. Accessed June 2019, 2019.
- 30.Anderson CM. Quitline Services: Current Practice and Evidence Base. Phoenix, AZ: North American Quitline Consortium,; 2016. [Google Scholar]
- 31.State of California EDD, Labor Market Information Division. Unemployment Rate Historical Trend for San Francisco-Redwood City-South San Francisco Metropolitan Division. https://www.labormarketinfo.edd.ca.gov/file/lfmonth/sanf$pds.pdf Published 2019. Accessed October 26, 2019.
- 32.Virtanen M, Kivimäki M, Elovainio M, Vahtera J, Ferrie JE. From insecure to secure employment: changes in work, health, health related behaviours, and sickness absence. Occupational and environmental medicine. 2003;60(12):948–953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Armour BS, Pitts MM, Lee C-w. Cigarette smoking and food insecurity among low-income famlies in the United States, 2001, Working Paper, No. 2007–19. In. Atlanta, GA: Federal Reserve Bank of Atlanta; 2007. [Google Scholar]
- 34.Busch SH, Jofre-Bonet M, Falba TA, Sindelar JL. Burning a hole in the budget: tobacco spending and its crowd-out of other goods. Applied Health Economics and Health Policy. 2004;34(4):263–272. [DOI] [PubMed] [Google Scholar]
- 35.Rogers E, Palacios J, Vargas E, et al. Financial Hardship, Motivation to Quit and Post-Quit Spending Plans among Low-Income Smokers Enrolled in a Smoking Cessation Trial. Substance Abuse: Research and Treatment. 2019;13:1178221819878765 10.1177/1178221819878765 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Guillaumier A, Bonevski B, Paul C. „Cigarettes are priority’: a qualitative study of how Australian socioeconomically disadvantaged smokers respond to rising cigarette prices. Health Education Research. 2015;30(4):599–608. 10.1093/her/cyv026 [DOI] [PubMed] [Google Scholar]
- 37.Mills SD, Golden SD, Henriksen L, et al. Neighbourhood disparities in the price of the cheapest cigarettes in the USA. J Epidemiol Community Health. 2019:jech-2018–210998. [DOI] [PMC free article] [PubMed] [Google Scholar]