Smoking and finances: baseline characteristics of low income daily smokers in the FISCALS cohort

Background Financial stress is a barrier to successful smoking cessation and a key predictor of relapse. Little is known about the financial situation of low-income Australian daily smokers. This study aims to describe and investigate associations between the financial functioning, tobacco use and quitting behaviours of low income daily smokers. Methods Low-income Australian adult smokers in the ‘Financial Intervention for Smoking Cessation Among Low-income Smokers (FISCALS) randomised clinical trial completed a structured telephone questionnaire. Results The median number of cigarettes typically smoked by the 1047 participants was 23 per day. The median spent on tobacco per week was AU$80. Three quarters (73.0%) reported some financial stress and 43.2% reported smoking-induced deprivation. Financial stress was significantly associated with deprivation (IRR: 1.23, 95% CI 1.21, 1.26, p < 0.001). There were no significant associations either between adjusted financial stress or deprivation and motivation to quit or certainty of quit success. Conclusions Financial stress and smoking induced deprivation were prevalent among low-income daily smokers, but they were not associated with motivation to quit. Smoking cessation interventions need to be responsive to the role financial stress plays in reducing quit attempts and increasing relapse. Trial registration Australian and New Zealand Clinical trials Registry ACTRN12612000725864 6/07/2012 Electronic supplementary material The online version of this article (10.1186/s12939-017-0643-6) contains supplementary material, which is available to authorized users.


Background
Australians from disadvantaged areas are three times more likely to smoke daily than those from the most advantaged areas (20% c.f. 7%) [1]. This uneven distribution of tobacco use [2,3] and cessation [4] by socioeconomic status (SES; or its proxies) is also observed in other high income countries. As a consequence of this, proportionally larger reductions in smoking rates are needed among disadvantaged groups if national targets for smoking prevalence are to be achieved [5][6][7].
Tobacco taxation is an effective strategy to reduce smoking rates in low SES population groups [3,8,9].
However, financial stress has been identified as a barrier to successful cessation and a key predictor of smoking relapse [10,11]. There is a pernicious cycle where tobacco use increases financial pressures (e.g., difficulty paying for bills, food, rent etc.) [12], which in turn maintains smoking [13], by the use of price-minimization strategies (e.g., by switching to roll-your-own), rather than cessation [14]. This may mean that price-based interventions will deliver diminishing returns over time with socially disadvantaged smokers because rises in taxation increase financial stress and smoking, and decrease quit success [15].
Cross-sectional and longitudinal evaluation studies have indicated that Australian smokers vary in their SES and motivations, intentions, and timeframes for quitting (e.g., The International Tobacco Control (ITC) Four-Country Study, The National Drug Strategy Household Survey; and the Household, Income & Labour Dynamics in Australia (HILDA) survey). However, these data sources do not bring together detailed information about financial functioning, tobacco use and quitting. This was one of the aims of the Financial Intervention for Smoking Cessation Among Low-income Smokers (FISCALS) Randomised Clinical Trial (RCT). The 1047 FISCALS smokers represent a large sample of Australian daily smokers selected on the basis of their low-income (defined as being in receipt of a means-tested government pension or allowance) and their willingness to quit within one-month of trial enrolment. Their data provides novel insights into the interplay between financial stress, smoking and cessation that are necessary to guide the development and delivery of effective tobacco control interventions for disadvantaged smokers. This is essential if cigarette tax increases provide either diminishing returns in the future [15], or adversely affect low SES smokers who are unable to quit.
The aim of this paper was to describe and investigate associations between the financial functioning, tobacco use and quitting behaviours of low SES smokers on entry to the FISCALS trial.

Methods
The primary aim of the FISCALS RCT was to evaluate the efficacy of a financial education and support program (FESP) in improving smoking cessation rates among low SES smokers. The detailed protocol of the FISCALS RCT has been reported elsewhere [16]. This study reports only data collected at baseline. Key eligibility criteria for recruitment into FISCALS were: 1) being aged 18 years or older; 2) being in receipt of a government pension or allowance; 3) smoking more than 10 cigarettes per day; 4) being interested in quitting, and; 5) being willing to make a quit attempt within 1 month of recruitment. Participants were recruited to the trial between April, 2013 and September, 2014 via Quitlines, community newspaper advertisements, and Australian Government Department of Human Services Centrelink Customer Service Centres.

Survey instruments
Survey items were used to collect demographic information and to examine type and price of tobacco purchased, motivation to quit, methods used for past cessation attempts, financial functioning, financial stress, deprivation and smoking induced deprivation. Where possible we used validated measures and those previously appearing in the literature. (See Tables 2 and 3 for further information  regarding response options).

Tobacco use and quitting behaviours
Tobacco use We assessed typical number of cigarettes smoked per day, type of tobacco, weekly tobacco expenditure, and length of last quit attempt (in days).
Nicotine dependence We calculated a Heaviness of Smoking Index (HSI) [17] score for each participant, based on the number of cigarettes smoked and the time until first cigarette each day. Scores range from 0 to 6, with scores of 4 or more indicating 'heavy smokers' [18]. We also assessed frequency of strong urges to smoke with the question: "How often do you get strong urges to smoke?" Quitting behaviours We assessed past quit attempts with the question: "Have you ever tried to quit smoking?" We determined participants' confidence in the success of their next quit attempt by asking: "If you decided to give up smoking completely, how sure are you that you would succeed?" We assessed motivation to quit smoking with the question "How would you rate your current motivation to give up smoking".

Methods used for smoking cessation
NRT and other pharmacotherapy use We asked participants: "Prior to entering the study, had you ever used Nicotine Replacement Therapy (NRT)?" and "Have you ever used prescription medications to help you quit smoking". If yes, we then asked participants to specify the types of medications used. We also assessed perceptions of past and future NRT use with the questions: "On your last quit attempt using NRT, how effective did you find NRT to help you quit smoking?" and "How much do you think NRT will increase your chances of successfully quitting in your current quit attempt?" Quitline use We determined past use of Quitline services by asking: "Prior to entering this study, had you ever spoken to the Quitline?"

Financial functioning
We collected information on weekly household income. Participants were asked: "Have you ever sought professional help with managing your finances?" If yes, they were asked to specify if this was within the past 12 months.
Financial stress We assessed financial stress using seven of the eight items from the Financial Stress Inventory (FSI) [19]. The FSI requires participants to indicate: "In the last month did any of the following happen to you because of a lack of money" and then to select all of the following that apply: a) could not pay electricity, gas or telephone bills on time; b) could not pay the mortgage or rent on time; c) pawned or sold something; d) went without meals; e) was unable to heat the home; f ) asked for financial help from friends or family; g) asked for help from a welfare or community organisation. Scores ranged from 0 to 7 with anything greater than 0 indicating the presence of financial stress.
Deprivation We measured deprivation using six questions developed by the Australian Bureau of Statistics for the Household Expenditure Survey [20]. Participants were asked to identify if any of the following list of activities applied to them: a) Have a holiday away from home for at least 1 week a year; b) Have a night out once a fortnight; c) Have friends or family over for a meal once a month; d) Have a special meal once a week; e) Buy new and not second hand clothes, most of the time; f ) Spend time on leisure or hobby activities. If no, participants were then asked to indicate for each item: "Is the reason that you [don't have a holiday away from home for at least one week a year] because you: a) don't want to, b) can't afford to, or c) some other reason?" with the text in brackets reflecting the relevant item. The scale was constructed by adding the number of endorsed items, leading to a score in the range of 0-6.
Smoking induced deprivation (SID) We measured SID in two ways. Presence or absence was assessed with the question: "In the past month, has there been a time when the money you spent on cigarettes resulted in not having enough money for household essential such as food?" (yes/no). The International Tobacco Control survey uses the same question using a six-month window [21]. The second (categorical) measure of SID asked participants to indicate their agreement with the statement: "Spending money on cigarettes means missing out on things that are important to me" using a five-point Likert scale from 'strongly disagree' to 'strongly agree'.

Relative socio-economic advantage and disadvantage
We measured the relative socio-economic advantage or disadvantage of participants using the Socio-Economic Indexes for Areas (SEIFA) [22]. Areas are divided into five quintiles; the top quintile represents the most advantaged and the bottom quintile represents the most disadvantaged.

Data analysis
We conducted all analyses using Stata 14.1 [23]. Demographic information, tobacco use, quitting behaviours and financial functioning were analysed descriptively. Categorical variables were presented as number (%). None of the ordinal or continuous variables were normally distributed based on Shapiro-Wilk normality tests, so are presented as median and interquartile range (IQR).
Financial stress (FSI) and financial deprivation scores were analyzed for associations with motivation to quit and certainty of quit success. Because FSI and financial deprivation scores are count variables with no overdispersion, analysis was conducted using Poisson regression, with results presented as incidence-rate ratios (IRR). Analyses were conducted both unadjusted, and adjusted for potential confounding by age, sex, education, SEIFA and HSI. Type of tobacco used (factory made cigarettes or roll-your-own) was examined for associations with a range of financial functioning indicators: a) ability to pay bills on time (single-item FSI measure) using χ 2 ; b) weekly tobacco expenditure (Mann-Whitney U); c) proportion of household income spent on tobacco products (Kruskal-Wallis H); d) FSI score (Mann-Whitney U), and e) deprivation (χ 2 ). To control the Type I error rate due to the number of analyses conducted, we used a Bonferonni-adjusted critical p-value of p = 0.004 to assess statistical significance.

Demographic information
The RCT participation rate was 40.9% giving a final sample size of 1047 participants (SI Trial Enrolment (Additional file 1)). Almost half of the sample was recruited from New South Wales (49.7%), one quarter from Queensland (24.4%), and approximately one fifth from Victoria (14%). See Table 1 for further information about the demographic characteristics of the study participants.
Three fifths of participants (60.2%) said that they experienced strong cravings hourly or more and nearly all had made a previous quit attempt (92.6%), with 378 (36.1%) doing so in the past year, for a median of 7 days (IQR 3-21).

Methods used to quit smoking
Almost three quarters of participants had used NRT in a past quit attempt (70.5%), almost half had used prescription medications (49.3%). More than a third had called a Quitline (37.6%) (See Table 2).
Among those who had used NRT, 76.5% used patches. Of the 684 participants who reported NRT use, 67.3% considered it had been at least 'moderately' effective. Almost three quarters of the total sample (73.4%) thought NRT would increase their chances of quit success in their next attempt 'very much' or 'extremely'. For most, confidence in future quit success was at least moderate (80.5%), and motivation to quit smoking was high (median 8/10, IQR 7-10).

Financial functioning
Overall, 27% of the sample lived within the most disadvantaged SEIFA quintile [22], 27.7% were from the second most disadvantaged quintile. The majority (62.9%) reported a weekly household income of less than $579 per week ($30,108 p.a.) (See Table 3).
Almost three quarters of participants (73.0%) reported some financial stress as measured by the FSI and 10.6% of all participants had sought professional help to manage their finances in the past 12 months. Those who were more motivated to quit reported higher unadjusted financial stress (IRR: 1.04; 95% CI 1.01, 1.06, p = 0.003) but not when adjusting for other covariates (IRR: 1.03; 95% CI 1.01, 1.06, p = 0.005), as seen in Table 4. There was no significant association between certainty of quit success and financial stress score, either unadjusted (IRR: 1.02, 95% CI 0.99, 1.06, p = 0.220) or adjusted (IRR: 1.04, 95% CI 1.00, 1.08, p = 0.061).
The first question of the FSI has been used as a measure of financial stress in past research [17]. This item was endorsed by 40.5% of the sample and was significantly associated with the type of cigarette smoked: those who smoked FMC only were more likely to report not paying bills on time (42.1%) than those who smoked RYO only (35.3%; χ 2 (2) = 12.889, p = 0.002).
Of relevance to the calculation of household tobacco expenditure, almost half of participants (42.2%) resided in single smoker households, living either alone (29.0%) or only with children (13.2%). Roughly the same proportion of households included two adults (22.7% two adults alone; 15.4% two adults and children). The remainder of households accommodated three or more adults (12.1% ≥ three adults alone; 7.3% ≥ three adults and children). The median levels of expenditure represented 13.8% (IQR 8.3-21.1%) of total reported household income across all types of household. The available data do not allow the accurate calculation of equivalised household income (by person), but similar proportions of household income were spent on tobacco in single person households (i.e., 1 adult no children: Median 17.2%; IQR 10.6-26.4%) and single smoker households (i.e., 1 adult with children: Median 15.5%; IQR 9.0-23.7%). The association between type of cigarette smoked (only FMC; only RYO; or both) and proportion Data only for those who reported that they could not afford the activity of household income spent on tobacco was significant for each of the three household types described above (all households; single person; single smoker) ( Table 5).
Nearly half of the sample reported smoking induced deprivation (SID) in the past month (43.2%). Most (75.7%) either 'agreed' or 'strongly agreed' that spending money on cigarettes meant they were missing out on other important things. Binary SID was significantly associated with FSI (z = 17.076, p < 0.001), but not with cigarette type (χ 2 (2) = 3.134, p = 0.209).

Discussion
When compared to the Australian population, a majority of FISCALS participants were at or below the 10th percentile of incomes [24]. They were also heavier smokers than the Australian smokers in the ITC Four Country Survey (HSI 3.0 c.f. 3.9 respectively) [25]. FISCALS participants endorsed each financial stress item at twice the rate over the past month as those from the lowest income quintile in HILDA in the past 12 months [24]. They endorsed the first item of the scale (a single item measure of financial stress) three times more often than low income smokers from Australia, the UK, USA and Canada in the ITC sample (40.5% c.f., 13.6%) [21]. The rate of smoking induced deprivation reported by the FISCALS sample for the past month was four times that reported in the ITC survey for a six-month period. The data also reveal that over all household types tobacco expenditure accounted for 13.8% of total reported household income, and financial stress and deprivation were significantly associated.
Together these results suggest that heavy tobacco use, financial stress and deprivation are commonplace and interrelated. The data also clearly demonstrate that in our sample of low-income smokers tobacco was used despite an inability to meet financial obligations (e.g., paying bills) and smoking was given priority over other discretionary activities (e.g., leisure activities or holidays). However, high levels of motivation to quit indicated that our participants were seeking to change this situation.
Interestingly, neither (adjusted) financial stress nor deprivation were associated with increased motivation to quit or confidence in future quit attempt success. Thus, the experienced financial pressures and deprivation did not appear to increase cessation motivation or confidence in cessation success in the cohort. However, this may be due to the relatively high levels of financial stress and deprivation in this sample. That is, the lack of association may reflect difficulty in detecting effects due to a relative lack of variation in the sample, rather than to the absence of associations. It is also worth noting that the underlying causes of cessation motivation and confidence are complex, with a range of mechanisms related to both psychological and social factors. A 2010 metaanalysis of theory of planned behaviour [26] and smoking found that quit intention was most strongly related to perceived behavioural control, which may be associated with financial stress and deprivation. Further work is required to tease apart the complex patterns of associations and, potentially, causal relationships between the various psychological, social and environmental factors at play.
Increasing tobacco taxation is generally considered to be very successful way of reducing smoking in the population in general and in disadvantaged smoking populations in particular [3]. However, our data reveal that financial stress, a potential barrier to successful cessation [10,27,28], is highly prevalent in the FISCALS cohort of low income daily smokers. This creates a dilemma for policy makers: raising taxes may increase quit attempts,  but the success of these attempts may be undermined if financial stress increases as the price rises. It will be important to put redistributive strategies in place to minimise the impact of cigarette price rises on poor smokers, perhaps by increasing subsidies for smoking cessation.

Limitations
The main limitation of the study is that it relies upon selfreport data collected using a CATI. Self-report data may be affected by exaggeration and or socially desirable responding. However this risk was minimal because the accuracy of responses was potentially verifiable and there was no incentive to respond in a socially desirable way [29].
A second limitation of this study may be that the inclusion and exclusion criteria for the FISCALS trial means that the smokers are unlikely to be representative of all low-income smokers in Australia. Their financial situation may be worse and their motivation to quit smoking higher. Even so we know that the cessation success rate for low SES smokers is generally low [3]. Thus, many will continue to live in the challenging financial situation described here.

Conclusion
The data from FISCALS study provides a unique insight into the lives of a group of low income smokers wholike almost half of all low-income smokers in Australiasmoke more than 10 cigarettes per day. They were a group with high financial difficulty and declared motivation to quit, who had previously tried unsuccessfully to give up smoking. These results are consistent with those observed in other high cigarette tax environments [30] and suggest that low-income smokers will face increasingly difficult financial circumstances as tobacco products becomes less affordable. Furthermore, the effectiveness of tobacco price interventions may be undermined by the high prevalence of financial stress and smoking induced deprivation.
Our data reinforce the importance of researching and developing smoking cessation interventions that are responsive to the role that financial stress plays in reducing quit attempts and increasing relapse [11]. Actively managing the vicious cycle of financial stress and smoking may be required to reduce the increasing socioeconomic gradient in smoking prevalence in Australia and other high income countries.