The Great Recession and Employee Alcohol Use: A U.S. Population Study
Abstract
This is the first study to examine broadly the overall net change in U.S. population estimates of alcohol use related to a recession—The Great Recession—among individuals who remain employed. The alcohol variables included drinker status, usual frequency and quantity of alcohol use, frequency of heavy drinking and intoxication, as well as contextual assessments of the frequency and quantity of alcohol use during the workday and after work. The moderating influence of gender, race, and age also was explored. Data for this repeated cross-sectional study were obtained from two national telephone surveys of U.S. workers. The first survey occurred prior to the Great Recession (2002–2003; N = 2,501), whereas the second survey occurred during and after the official end of the Great Recession (2008–2011; N = 2,581). The results revealed that the recession was related to a higher proportion of drinkers among middle-aged employees, but not among young employees. Gender and race did not moderate the relation of the recession to drinker status. Among drinkers, the recession was not related to usual alcohol use (frequency and quantity), but was positively related to the frequency of heavy drinking and intoxication. Further, the recession had a differential relation to the contextual alcohol measures. It was negatively related to the frequency and quantity of workday alcohol use, but was positively related to the frequency and quantity of after-work alcohol use. Among drinkers, gender, race, and age did not moderate the relation of the recession to alcohol use.
Understanding the factors influencing excessive and ill-timed alcohol use among employed adults is an important social and organizational issue because it is prevalent; can lead to compromised health, role performance and safety on and off the job; and has associated costs for employees, employers, and society (Bouchery, Harwood, Sacks, Simon, & Brewer, 2011; Cherpitel, 2007; Frone, 2013; Normand, Lempert, & O’Brien, 1994; Rehm, Taylor, & Room, 2006). Past research exploring environmental causes of employee alcohol use has generally focused on work-related factors, such as work stress, physical availability of alcohol at work, workplace social control, and workplace alcohol use norms (Frone, 2013). However, employee alcohol use also may be affected by factors outside the workplace, including macro-environmental factors that might have an impact on conditions at work and outside work. One such macro-environmental factor is economic recession. To date, no research has explored the net impact of economic recession on employee alcohol use. The general goal of the present study, therefore, is to explore the relation of the most recent recession, referred to as the Great Recession, to U.S. population estimates of alcohol use among individuals who remain employed.
The Great RecessionIn December 2007, the U.S. officially entered a recession. Starting in the fall of 2008 and continuing into early 2009, the recession intensified with a dramatic collapse of financial markets and an equally dramatic increase in mass layoffs (Council of Economic Advisers, 2010; Grusky, Western, & Wimer, 2011). Although the recession was officially over in June 2009, its effects were long-lasting. For example, the annual unemployment rate increased from 5% in December 2007 to 10% in October 2010, and only decreased to 8% by January 2013. The Dow Jones Industrial Average low of 6,547 in March 2009 did not recover to the prerecession high of 14,165 in October 2007 until March 2013. Polls by Hart Research Associates (2014) found that 64% of Americans believed the US was in a recession in December 2012 and 57% still believed this in March 2014. Finally, the most recent recession was the longest economic downfall since the Great Depression of the 1930s and spread worldwide.
Recessions and Alcohol Use
The most obvious and studied conditions resulting from economic recessions are job loss and economic stress (e.g., Dee, 2001; de Goeij et al., 2015; Mulia, Zenmore, Murphy, Liu, & Catalano, 2014; Popovici & French, 2013; Richman et al., 2012). Studies that explore the relation of such variables to alcohol consumption during a period of recession or economic decline provide information on gross changes in alcohol use due to these separate processes. However, these processes may have opposing effects (de Goeij et al., 2015). For instance, after reviewing the literature to assess the effect of economic crises on alcohol consumption, de Goeij et al. (2015) found that exposure to stressors resulting from or amplified by recessions and the consequent distress can increase alcohol use, and that reduced financial resources may lead to reduced alcohol intake due to a loss in discretionary income. Although studies exploring the processes linking a recession to alcohol use provide useful information, they do not directly assess the overall net impact of a recession per se. To directly evaluate the overall net impact of a recession on population changes in alcohol use requires data from repeated cross-sectional studies conducted before and during, or shortly after, the recession (e.g., Firebaugh, 1997).
Only one study was located that explored the net relation of the Great Recession to alcohol use in the U.S. Bor, Basu, Coutts, McKee, and Stuckler (2013) compared alcohol use in the overall U.S. adult population that was assessed prior to (2006/2007) and during (2008/2009) the Great Recession. That study found a significant but small decrease in the proportion of drinkers from pre-recession (52%) to the Great Recession (51.6%). However, among drinkers, the recession was positively associated with the number of drinking days during the prior month and the number of monthly binge drinking episodes. The recession was not related to the average number of drinks on drinking days.
No research has explored the net change in alcohol use related to the Great Recession, or provided a broad assessment of net change in alcohol use related to any recession, among the employed population. Even though job loss is not experienced, a recession can impact the lives of employed individuals in a number of ways. Green, Felstead, Gallie, and Inanc (in press) found that, among those remaining employed, the Great Recession was associated with increased workloads and job insecurity, and lower quality of work life. Fischer and Hout (2006) suggested that each layoff during a recession makes 2.1 additional workers insecure about their own jobs. In addition to changes in work conditions and job insecurity, those who remain employed may experience a host of other negative financial outcomes, such as depreciation of property value and losses in retirement accounts. Collectively, all of these adverse experiences may influence the alcohol use of employed adults. Therefore, the present study compares U.S. population estimates of several alcohol use outcomes among employed adults that were assessed before (2002–2003) and during/after (2008–2011) the Great Recession.
The first outcome is drinker status. Bor et al. (2013) found a small decrease in the proportion of drinkers in the overall U.S. adult population associated with the Great Recession, which was attributed to an increase in a subgroup of individuals who lost the financial resources, perhaps due to job loss, to purchase alcohol. However, in a sample of employed adults, there may be less direct loss of discretionary income. In contrast, the increased exposure to various work-related stressors mentioned earlier, such as work overload, job insecurity, and lower quality of work life, as well as financial stressors, such as unrealized and/or realized losses in retirement accounts and property values, may increase the likelihood of alcohol use during the recession. Using the lens of self-medication (also known as tension-reduction and affect-regulation) models of stress-induced alcohol use (Conger, 1956; Cooper, Frone, Russell, & Mudar, 1995; McCarthy, Curtin, Piper, & Baker, 2010), which propose that individuals use alcohol to reduce the tension and distress caused by exposure to stressors, it is hypothesized that the Great Recession will be associated with a higher proportion of employed drinkers.
Among alcohol users, the second set of alcohol outcomes includes usual frequency of consumption, usual number of drinks per drinking occasion, frequency of heavy drinking, and frequency of intoxication. As noted earlier, although their study did not focus specifically on the subpopulation of employed adults, Bor et al. (2013) reported that the Great Recession was associated with a net increase in the number of monthly drinking days and binge episodes in the general U.S. adult population. A study by Dee (2001) that assessed fluctuations in state unemployment rates from 1984 to 1995 found that the prevalence of binge drinking during the past month increased during economic downturns among those remaining employed. However, Dee’s (2001) analysis of employed adults only explored changes in a single alcohol outcome in relation to gross change in one of several processes that might link recessions to alcohol use among those remaining employed. Based on self-medication models of stress-induced alcohol use (Conger, 1956; Cooper et al., 1995; McCarthy et al., 2010) and the results reported by Bor et al. (2013) and Dee (2001), it is hypothesized that the Great Recession will be associated with a net increase in levels of usual frequency and quantity of alcohol use, heavy drinking, and intoxication in the population of employed adults.
The third set of outcomes are contextual measures of alcohol use that are only relevant among employed alcohol users and have never been explored in studies of economic recession and alcohol use. These measures capture the frequency and quantity of alcohol use during the workday and immediately after work. The dynamics occurring during a recession may create differential effects on workday and after-work alcohol use. The Great Recession resulted in substantial organizational downsizing (Grusky et al., 2011) and was associated with increases in workload and job insecurity among those remaining employed (Fisher & Hout, 2006; Green et al., in press). To the extent that work demands increased among those who retained employment and the available labor force increased with downsizing, employers may have been more likely to monitor employee work behavior and employees may have been more sensitive to this possibility leading to elevated levels of job insecurity. Also, because alcohol use is legal, employer substance use policies generally proscribe alcohol use during the workday, but do not address alcohol use outside work, such as after-work use (Frone & Trinidad, 2012). Therefore, based on these arguments and self-medication models of stress-induced alcohol use, it is hypothesized that the Great Recession will be associated with a decrease in alcohol use during the workday to avoid job loss and an increase in alcohol use after work to reduce stress.
Finally, it is possible that the relation of the Great Recession to alcohol use differed across subgroups of the employed population based on gender, race, and age. However, little research has explored demographic differences in either gross or net change in alcohol use due to an economic recession (Zemore, Mulia, Jones-Webb, Liu, & Schmidt, 2013). Compared with women, men have been socialized to externalize distress and perceive fewer social sanctions associate with drinking (Cooper, Russell, Skinner, Frone, & Mudar, 1992; Nolen-Hoeksema, 2004). Consistent with this expectation, Dee (2001) reported that economic downturns, represented by increases in state unemployment rates, were more strongly related to a higher prevalence of binge drinking during the past month among employed men compared with employed women. Therefore, the Great Recession may be associated with a larger net increase in alcohol use among employed men compared with employed women.
Relative to Whites, members of racial minorities have an economic disadvantage and are more likely to experience racial discrimination. This may make racial minorities more vulnerable to stress-induced alcohol use resulting from recessions (Lo & Cheng, 2013; Zemore et al., 2013). However, research in the general U.S. population has provided mixed support for differential effects of race on gross change in alcohol use associated with unemployment and financial stress resulting from the Great Recession (Lo & Cheng, 2013; Zemore et al., 2013). Also, comparing employed Whites with employed Blacks, Dee (2001) failed to find a difference in the relation between state unemployment rates and past-month binge drinking. Nonetheless, the present study provides an exploratory examination of the potential moderating effect of race on the net relation between the Great Recession and alcohol use among employed adults.
The net relation of the Great Recession to employee alcohol use also may differ across age groups. Compared with younger workers, older workers have more family-related demands and financial responsibilities (Pew Research Center, 2013). These demands and financial responsibilities may combine with losses in retirement savings and increases in job demands and job insecurity to make the recession more stressful among older employees compared with younger employees. Therefore, the present study explores age differences in the net relation between the recession and alcohol use among employed adults.
Method
Sample and study design
Data obtained prior to the Great Recession came from a random U.S. telephone survey called the National Survey of Workplace Health and Safety conducted from January 2002 to June 2003 (Study 1). Data obtained during and after the Great Recession came from a random U.S. telephone survey called the National Survey of Work Stress and Health conducted from December 2008 to April 2011 (Study 2). The population from which the participants for both studies were sampled was all non-institutionalized adults, ages 18–65 years, who were employed in the civilian labor force, and who resided in households in the 48 U.S. contiguous states and the District of Columbia. The present analyses were restricted to wage and salary workers (owner/operators were not included) who had data on all required variables—2,501 participants from Study 1 and 2,581 participants from Study 2. Additional information on the design of Study 1 and Study 2 can be found in Frone (2006) and Frone (2015), respectively.
Sampling Weights
Equivalent procedures were used in both studies to compute sampling weights. The weights adjust for the probability of selection and nonresponse, and post-stratify the sample to average population totals for gender, race, age, and U.S. Census region. For more detail on the sampling weights, see Frone (2006).
Measures
Recession
A recession indicator variable was created where individuals from Study 1 were assigned a score of 0 and individuals from Study 2 were assigned a score of 1.
Alcohol use
The same measures were used in both studies. Individuals were asked whether or not they consumed alcohol during the prior 12 months. Drinker status was coded 0 for those answering no and 1 for those answering yes.
The second set of alcohol outcomes assessed overall alcohol use during the past 12 months among drinkers. Usual alcohol use was assessed by asking about the typical frequency of consumption and the typical number of drinks per drinking day. Heavy drinking was assessed by asking about the frequency of consuming 5+ (men) or 4+ (women) drinks per day. Frequency of intoxication was assessed by asking how often the participant drank enough to become intoxicated or drunk. The one quantity measure used an open-ended response format. Response anchors for the three items assessing frequency of use ranged from 0 (never) to 4 (everyday).
The third set of alcohol outcomes assessed the context of alcohol use during the past 12 months among drinkers. Workday alcohol use was assessed with two measures—frequency of drinking during the workday and the typical number of drinks consumed during the workday. After-work alcohol use was assessed with two measures—frequency of initiating drinking within two hours of leaving work and the typical number of drinks consumed after work. The two quantity measures used an open-ended response format. Response anchors for the two measures assessing frequency of use ranged from 0 (never) to 4 (everyday).
Covariates
Several covariates were included in the analyses to adjust for potential differences in the composition of the samples across the two surveys due to sampling or changes in the employed population: gender (0 = women, 1 = men), race (1 = White, 2 = Black, 3 = Hispanic, 4 = other), age (in years), years of formal education (10 ordinal categories; 1 = less than high school, 2 = some high school, 3 = high school graduate/GED, 4 = trade/technical/vocational school 5 = some college, 6 = associate’s degree, 7 = bachelor’s degree, 8 = some graduate school, 9 = master’s degree, 10 = PhD/professional degree), total annual family income from all sources (U.S. dollars), marital status (1 = never married, 2 = married/living as married, 3= divorced, 4 = widowed), U.S. Census geographic divisions (9 nominal categories, see Table 1), occupation (9 aggregated nominal categories based on the Standard Occupational Classification system, see Table 1), job tenure (in years), number of weekly work hours, seasonal job (0 = no, 1 = yes), and union membership (0 = no, 1 = yes).
Table 1
Sample characteristics
| Variable | Study 1 (N =2,501) | Study 2 (N = 2,581) | Combined Sample (N = 5,082) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Unweighted N | Unweighted Percentage or Mean | Weighted Percentage or Mean | Unweighted N | Unweighted Percentage or Mean | Weighted Percentage | Unweighted N | Unweighted Percentage or Mean | Weighted Percentage or Mean | |
|
| |||||||||
| Gendera | |||||||||
| Male | 1,109 | 44.3% | 52.2% | 1,030 | 39.9% | 51.5% | 2,139 | 42.1% | 51.9% |
| Female | 1,392 | 55.7% | 47.8% | 1,551 | 60.1% | 48.5% | 2,943 | 57.9% | 48.1% |
|
| |||||||||
| Racea | |||||||||
| White | 1,946 | 77.8% | 72.3% | 2,075 | 80.4% | 68.5% | 4,021 | 79.1% | 70.4% |
| Black | 296 | 11.8% | 12.5% | 236 | 9.1% | 13.0% | 532 | 10.5% | 12.7% |
| Hispanic | 141 | 5.6% | 8.1% | 129 | 5.0% | 9.5% | 270 | 5.3% | 8.8% |
| Other | 118 | 4.7% | 7.1% | 141 | 5.5% | 9.0% | 259 | 5.1% | 8.0% |
|
| |||||||||
| Agea,b | 2,501 | 39.6 | 38.8 | 2,581 | 45.7 | 40.3 | 5,082 | 42.7 | 39.6 |
|
| |||||||||
| Educationa,b | 2,501 | 5.5 | 5.5 | 2,581 | 6.1 | 5.9 | 5,082 | 5.8 | 5.7 |
|
| |||||||||
| Family Income (median)a,b | 2,501 | 50,000 | 50,000 | 2,581 | 68,000 | 65,000 | 5,082 | 60,000 | $60,000 |
|
| |||||||||
| Marital Statusa | |||||||||
| Not married | 661 | 26.4% | 26.5% | 518 | 20.1% | 27.7% | 1,179 | 23.2% | 27.1% |
| Married/living as married | 1,331 | 53.2% | 57.6% | 1,553 | 60.2% | 59.3% | 2,884 | 56.8% | 58.5% |
| Divorced | 466 | 18.6% | 14.7% | 434 | 16.8% | 11.6% | 900 | 17.7% | 13.1% |
| Widowed | 43 | 1.7% | 1.2% | 76 | 2.9% | 1.4% | 119 | 2.3% | 1.3% |
|
| |||||||||
| U.S. Census Divisiona | |||||||||
| New England | 159 | 6.4% | 5.9% | 168 | 6.5% | 4.9% | 327 | 6.4% | 5.4% |
| Middle Atlantic | 350 | 14.0% | 13.8% | 426 | 16.5% | 14.0% | 776 | 15.3% | 13.9% |
| East North Central | 461 | 18.4% | 16.8% | 498 | 19.3% | 15.2% | 959 | 18.9% | 16.0% |
| West North Central | 193 | 7.7% | 6.7% | 250 | 9.7% | 7.6% | 443 | 8.7% | 7.2% |
| South Atlantic | 443 | 17.7% | 18.1% | 428 | 16.6% | 20.8% | 871 | 17.1% | 19.5% |
| East South Central | 145 | 5.8% | 5.7% | 123 | 4.8% | 7.0% | 268 | 5.3% | 6.4% |
| West South Central | 260 | 10.4% | 10.0% | 192 | 7.4% | 8.7% | 452 | 8.9% | 9.4% |
| Mountain | 162 | 6.5% | 7.1% | 160 | 6.2% | 6.2% | 322 | 6.3% | 6.6% |
| Pacific | 328 | 13.1% | 15.9% | 336 | 13.0% | 15.6% | 664 | 13.1% | 15.8% |
|
| |||||||||
| Occupationsa,b | |||||||||
| Management/business/financial | 299 | 12.0% | 11.3% | 375 | 14.5% | 13.2% | 674 | 13.3% | 12.3% |
| Professional | 659 | 26.3% | 25.3% | 911 | 35.3% | 31.9% | 1,570 | 30.9% | 28.7% |
| Service | 392 | 15.7% | 15.9% | 335 | 13.0% | 15.3% | 727 | 14.3% | 15.6% |
| Sales | 217 | 8.7% | 8.5% | 189 | 7.3% | 7.7% | 406 | 8.0% | 8.1% |
| Office/administrative support | 452 | 18.1% | 17.2% | 415 | 16.1% | 15.1% | 867 | 17.1% | 16.2% |
| Construction/extraction/farming/fishing/forestry | 102 | 4.1% | 4.7% | 52 | 2.0% | 2.5% | 154 | 3.0% | 3.8% |
| Installation/maintenance/repair | 105 | 4.2% | 4.9% | 74 96 | 2.9% | 3.5% | 179 | 3.5% | 4.2% |
| Production | 140 | 5.6% | 5.8% | 3.7% | 3.7% | 236 | 4.6% | 4.7% | |
| Transportation/material moving | 135 | 5.4% | 6.4% | 134 | 5.2% | 6.6% | 269 | 5.3% | 6.5% |
|
| |||||||||
| Job Tenure (years)a,b | 2,501 | 4.9 | 4.7 | 2,581 | 6.5 | 5.3 | 5,082 | 5.7 | 5.0 |
|
| |||||||||
| Number of Weekly Work Hoursa,b | 2,501 | 42.4 | 42.4 | 2,581 | 41.5 | 41.2 | 5,082 | 41.9 | 41.8 |
|
| |||||||||
| Seasonal Job | |||||||||
| No | 2,382 | 95.2% | 94.9% | 2,455 | 95.1% | 93.4% | 4,837 | 95.2% | 94.1% |
| Yes | 119 | 4.8% | 5.1% | 126 | 4.9% | 6.6% | 245 | 4.8% | 5.9% |
|
| |||||||||
| Union Member | |||||||||
| No | 2,072 | 82.9% | 82.5% | 2,108 | 81.7% | 83.3% | 4,180 | 82.3% | 82.9% |
| Yes | 429 | 17.1% | 17.4% | 473 | 18.3% | 16.7% | 902 | 17.8% | 17.1% |
|
| |||||||||
| Recession | |||||||||
| No | 2,501 | 49.2% | 49.3% | ||||||
| Yes | 2,581 | 50.8% | 50.7% | ||||||
Study Design and Data Analysis
To explore net population change in employee alcohol use associated with the Great Recession, this study used a repeated cross-sectional study design where the same measures are assessed using independent probability samples (i.e., different respondents) that were drawn from the U.S. population of employed adults both (a) before and (b) during and after the formal end of the recession. This design differs from a panel study where the same measures are assessed at multiple time points using the same sample of individuals. Although both types of studies are useful for assessing change, repeated cross-sectional surveys are generally better suited to explore overall net change at the level of populations (e.g., Firebaugh, 1997).
All analyses were conducted using Stata (Version 14, Stata Corporation, 2015) because it allowed the use of sampling weights in all analyses and estimation of correct standard errors based on Taylor linearization (e.g., Lehtonen & Pahkinen, 2004). To explore the relation of the Great Recession to the alcohol use outcomes, logistic regression was used for drinker status, ordinal logistic regression was used for outcomes representing frequency of use, and negative binomial regression was used for the outcomes representing number of drinks consumed (Long, 1997; Wooldridge, 2002). For each outcome, the covariates and the recession indicator variable entered the regression equation on Step 1, followed by the interactions between the recession indicator variable and gender, race, and age on Step 2.
Results
Sample Characteristics
Table 1 presents unweighted and weighted descriptive information for the overall samples from Study 1 and Study 2, as well as the overall pooled sample of employed adults. The unweighted and weighted results in Table 1 show that the two samples did not differ in the proportion of workers reporting seasonal jobs or belonging to a union. Based on unweighted results, statistically significant differences existed between the two samples in terms of gender, marital status, and U.S. Census Division (i.e., geographic location of residence). However, after applying the sampling weights, there were no longer statistically significant differences between the samples on these variables.
Several variables showed statistically significant differences across the two samples even after applying sampling weights—age, education, median family income, occupation, job tenure, and number of weekly work hour—though some weighted differences were fairly small in absolute magnitude1 As noted earlier, to adjust for the relation of any compositional differences in the samples across Study 1 and Study 2 to the alcohol outcomes, each of the demographic characteristics shown in Table 1 were used as covariates in the weighted regression analyses described below.
Drinker Status
The results of the logistic regression analysis testing the relation of the recession to drinker status is presented in Table 2. The results from Step 1 show that, while adjusting for the demographic covariates, there was no marginal relation between the recession and drinker status (odds ratio = 1.19, ns). However, the results from Step 2 show that there was a significant interaction between the recession and age (odds ratio = 1.02, p < .05). To explore the nature of this interaction, conditional relations were computed for several ages covering the range from 18 to 65 years old. The conditional relations, presented in Table 3, show that the recession was not related to a change in the proportion of drinkers among young employees, but it was related to an increase in the proportion of drinkers among middle-aged employees. These results provide conditional support for the hypothesized positive relation between the recession and drinker status.
Table 2
Logistic regression results for drinker status
| Predictors | Odds Ratio | [95% CI] |
|---|---|---|
|
| ||
| Step 1 | ||
|
| ||
| Gender (men) | 1.31** | [1.083, 1.593] |
|
| ||
| Race | ||
| White | RG | RG |
| Black | .50*** | [.384, .653] |
| Hispanic | .81 | [.564, 1.150] |
| Other | .49*** | [.349, .695] |
|
| ||
| Age | .98*** | [.972, .989] |
|
| ||
| Education | 1.09*** | [1.035, 1.151] |
|
| ||
| Family Income | 1.05* | [1.009, 1.089] |
|
| ||
| Marital Status | ||
| Not married | RG | RG |
| Married/living as married | .68** | [.519, .894] |
| Divorced | 1.15 | [.860, 1.547] |
| Widowed | .84 | [.494, 1.429] |
|
| ||
| U.S. Census Division | ||
| New England | RG | RG |
| Middle Atlantic | 1.21 | [.823, 1.791] |
| East North Central | 1.00 | [.695, 1.437] |
| West North Central | 1.03 | [.681, 1.568] |
| South Atlantic | .70 | [.484, 1.012] |
| East South Central | .43*** | [.271, .672] |
| West South Central | .64* | [.421, .969] |
| Mountain | .51** | [.329, .798] |
| Pacific | .70 | [.472, 1.045] |
|
| ||
| Occupations | ||
| Management/business/financial | RG | RG |
| Professional | .71* | [.523, .970] |
| Service | .76 | [.527, 1.086] |
| Sales | .75 | [.506, 1.112] |
| Office/administrative support | .93 | [.662, 1.316] |
| Construction/extraction/farming/ | ||
| fishing/forestry/ | .77 | [.453, 1.325] |
| Installation/maintenance/repair | .95 | [.564, 1.616] |
| Production | .57* | [.349, .920] |
| Transportation/material moving | .76 | [.463, 1.238] |
|
| ||
| Job Tenure (years) | .98** | [.967, .993] |
|
| ||
| Number of Weekly Work Hours | 1.01 | [.996, 1.015] |
|
| ||
| Seasonal Job (yes) | .96 | [.642, 1.432] |
|
| ||
| Union Member (yes) | .91 | [.728, 1.138] |
|
| ||
| Recession | 1.19 | [.998, 1.410] |
|
| ||
| Step 2 | ||
|
| ||
| Recession × Gender | .95 | [.682, 1.329] |
|
| ||
| Recession × Black | 1.20 | [.724, 1.994] |
|
| ||
| Recession × Hispanic | 1.30 | [.641, 2.647] |
|
| ||
| Recession × Other | 1.90 | [.962, 3.751] |
|
| ||
| Recession × Age | 1.02* | [1.002, 1.032] |
Note: N = 5,082. RG = reference group. The coefficient for family income has been rescaled to represent increments of $10,000.
Table 3
Conditional relation of the recession to drinker status by age
| Age | Odds Ratio | [95% CI] |
|---|---|---|
| 18 years old | .85 | [.563, 1.283] |
| 25 years old | .94 | [.685, 1.300] |
| 35 years old | 1.10 | [.890, 1.349] |
| 45 years old | 1.27** | [1.081, 1.496] |
| 55 years old | 1.48*** | [1.175, 1.853] |
| 65 years old | 1.71** | [1.212, 2.423] |
N = 5,082.
Overall Alcohol Use
The regression analyses for overall alcohol use are presented in Table 4. The Step 1 results indicate that, while adjusting for the demographic covariates, the recession was unrelated to the usual frequency (b = .040, ns) and quantity (b = .063, ns) of alcohol use. However, the recession was related to increases in both heavy drinking (b = .442, p < .001) and intoxication (b = .496, p < .001). The Step 2 results show that the relation of the recession to the four assessments of overall alcohol use did not differ across gender, race, and age. These results fail to support the hypothesized positive relation of the recession to usual alcohol consumption, but they do support the hypothesized positive relation to heavy alcohol use.
Table 4
Regression results for overall alcohol use among drinkers
| Predictors | Usual Frequency of Alcohol Use | Drinks per Drinking Day | Frequency of Heavy Drinking | Frequency of Intoxication | ||||
|---|---|---|---|---|---|---|---|---|
| b | (SE) | b | (SE) | b | (SE) | b | (SE) | |
|
| ||||||||
| Step 1 | ||||||||
|
| ||||||||
| Gender (men) | .589*** | (.082) | .340*** | (.036) | 1.145*** | (.097) | .486*** | (.093) |
|
| ||||||||
| Race | ||||||||
| White | RG | RG | RG | RG | ||||
| Black | −.492*** | (.128) | −.190** | (.073) | −1.01*** | (.200) | −.745*** | (.167) |
| Hispanic | −.280 | (.149) | .040 | (.072) | −.373 | (.196) | −.705*** | (.207) |
| Other | −.692*** | (.170) | −.158* | (.069) | −.738*** | (.214) | −.393 | (.218) |
|
| ||||||||
| Age | .015*** | (.004) | −.012*** | (.001) | −.037*** | (.005) | −.053*** | (.005) |
|
| ||||||||
| Education | .063** | (.023) | −.057*** | (.009) | −.092*** | (.025) | −.042 | (.024) |
|
| ||||||||
| Family Income | .011 | (.007) | −.001 | (.001) | .005* | (.002) | −.002 | (.002) |
|
| ||||||||
| Marital Status | ||||||||
| Not married | RG | RG | RG | RG | ||||
| Married/living as married | −.339*** | (.106) | −.229*** | (.041) | −.519*** | (.118) | −.647*** | (.107) |
| Divorced | −.349* | (.144) | −.036 | (.054) | −.253 | (.170) | −.315* | (.146) |
| Widowed | −.481 | (.267) | −.177* | (.080) | −.383 | (.311) | −.576 | (.312) |
|
| ||||||||
| U.S. Census Division | ||||||||
| New England | RG | RG | RG | RG | ||||
| Middle Atlantic | −.143 | (.166) | −.067 | (.054) | .002 | (.185) | −.032 | (.171) |
| East North Central | −.279 | (.163) | −.012 | (.057) | .063 | (.178) | .094 | (.164) |
| West North Central | −.048 | (.176) | .029 | (.071) | .192 | (.199) | .038 | (.186) |
| South Atlantic | −.179 | (.168) | −.065 | (.059) | −.090 | (.190) | −.057 | (.170) |
| East South Central | −.206 | (.237) | .014 | (.096) | −.246 | (.290) | −.386 | (.225) |
| West South Central | −.326 | (.185) | −.041 | (.093) | −.260 | (.224) | −.234 | (.222) |
| Mountain | −.031 | (.197) | −.041 | (.075) | −.049 | (.246) | .143 | (.227) |
| Pacific | .061 | (.178) | −.087 | (.062) | −.040 | (.192) | .162 | (.178) |
|
| ||||||||
| Occupations | ||||||||
| Management/business/financial | RG | RG | RG | RG | ||||
| Professional | −.351** | (.113) | −.047 | (.043) | −.137 | (.129) | −.239 | (.132) |
| Service | −.239 | (.149) | .037 | (.065) | −.013 | (.171) | −.280 | (.177) |
| Sales | −.291 | (.163) | .024 | (.076) | −.232 | (.201) | −.178 | (.184) |
| Office/administrative support | −.283* | (.144) | −.029 | (.052) | .059 | (.168) | −.181 | (.165) |
| Construction/extraction/farming/fishing/forestry/ | −.181 | (.237) | .208* | (.105) | .339 | (.270) | .537 | (.293) |
| Installation/maintenance/repair | .120 | (.222) | .043 | (.082) | −.051 | (.239) | .273 | (.264) |
| Production | −.132 | (.204) | −.001 | (.084) | .119 | (.210) | .047 | (.234) |
| Transportation/material moving | −.411* | (.197) | −.030 | (.073) | −.175 | (.224) | −.172 | (.227) |
|
| ||||||||
| Job Tenure (years) | −.003 | (.007) | −004* | (.002) | −.010 | (.007) | −.019* | (.009) |
|
| ||||||||
| Number of Weekly Work Hours | .001 | (.004) | .002 | (.002) | .001 | (.004) | −.002 | (.004) |
|
| ||||||||
| Seasonal Job (yes) | −.064 | (.181) | .132 | (.088) | .219 | (.228) | .042 | (.229) |
|
| ||||||||
| Union Member (yes) | .065 | (.101) | .081 | (.050) | .182 | (.114) | .156 | (.119) |
|
| ||||||||
| Recession | .040 | (.077) | .063 | (.033) | .442*** | (.088) | .496*** | (.087) |
|
| ||||||||
| Step 2 | ||||||||
|
| ||||||||
| Recession × Gender | .056 | (.145) | −.017 | (.059) | −.003 | (.167) | .191 | (.163) |
|
| ||||||||
| Recession × Black | −.136 | (.249) | .000 | (.148) | .180 | (.390) | −.056 | (.321) |
|
| ||||||||
| Recession × Hispanic | −.170 | (.287) | .012 | (.143) | .453 | (.374) | −.027 | (.414) |
|
| ||||||||
| Recession × Other | .168 | (.341) | .098 | (.134) | .339 | (.441) | .599 | (.445) |
|
| ||||||||
| Recession × Age | .006 | (.006) | .003 | (.003) | .011 | (.007) | .011 | (.007) |
Note: N = 3,818. Unstandardized regression coefficients are reported. RG = reference group. The coefficients for family income have been rescaled to represent increments of $10,000. See Methods section for a discussion of the regression analyses for the various alcohol outcomes.
Contextual Alcohol Use
The regression analyses for contextual alcohol use are shown in Table 5. The Step 1 results indicate that, while adjusting for the demographic covariates, the recession was associated with a decrease in the frequency (b = −354, p < .05) and quantity (b = −500, p < .001) of alcohol use during the workday, and an increase in the frequency (b = .321, p < .001) and quantity (b = .269, p < .001) of after-work alcohol use. The Step 2 results show that the relation of the recession to the contextual alcohol use outcomes did not differ across gender, race, and age. These results support the hypothesized differential relation of the recession to workday and after-work alcohol use.
Table 5
Regression results for contextual alcohol use among drinkers
| Predictors | Frequency of Workday Alcohol Use | Number of Drinks During the Workday | Frequency of After-Work Alcohol Use | Number of Drinks After Work | ||||
|---|---|---|---|---|---|---|---|---|
| b | (SE) | b | (SE) | b | (SE) | b | (SE) | |
|
| ||||||||
| Step 1 | ||||||||
|
| ||||||||
| Gender (men) | .856*** | (.156) | .854*** | (.153) | .400*** | (.084) | .372*** | (.057) |
|
| ||||||||
| Race | ||||||||
| White | RG | RG | RG | RG | ||||
| Black | −.261*** | (.294) | −.072 | (.314) | −787*** | (.146) | −.660*** | (.108) |
| Hispanic | −.165 | (.332) | −.256 | (.300) | −.593** | (.195) | −.275* | (.127) |
| Other | −.811* | (.335) | −.804* | (.330) | −.512** | (.179) | −.182 | (.134) |
|
| ||||||||
| Age | −.017* | (.008) | −.021** | (.008) | .007 | (.004) | −.053*** | (.005) |
|
| ||||||||
| Education | .154** | (.051) | .152*** | (.045) | .092*** | (.023) | −.009 | (.015) |
|
| ||||||||
| Family Income | .006 | (.008) | .013 | (.009) | .007 | (.007) | .003 | (.002) |
|
| ||||||||
| Marital Status | ||||||||
| Not married | RG | RG | RG | RG | ||||
| Married/living as married | −.292 | (.211) | −.146 | (.206) | −.294*** | (.106) | −.256*** | (.069) |
| Divorced | −.308 | (.265) | .258 | (.242) | −.366** | (.147) | −.154 | (.099) |
| Widowed | −.502 | (.517) | .345 | (.496) | −.394 | (.301) | −.254 | (.215) |
|
| ||||||||
| U.S. Census Division | ||||||||
| New England | RG | RG | RG | RG | ||||
| Middle Atlantic | .169 | (.326) | .219 | (.310) | −.091 | (.152) | −.055 | (.102) |
| East North Central | .010 | (.326) | .471 | (.321) | −.015 | (.148) | .044 | (.100) |
| West North Central | −.204 | (.396) | −.202 | (.376) | .077 | (.170) | .103 | (.124) |
| South Atlantic | .106 | (.326) | .358 | (.323) | −.017 | (.156) | −.025 | (.106) |
| East South Central | .440 | (.433) | .582 | (.403) | .164 | (.249) | −.002 | (.145) |
| West South Central | −.314 | (.397) | −.242 | (.375) | .066 | (.182) | .115 | (.124) |
| Mountain | .415 | (.411) | .619 | (.378) | .233 | (.206) | .003 | (.119) |
| Pacific | .129 | (.347) | .138 | (.328) | .333* | (.166) | .084 | (.107) |
|
| ||||||||
| Occupations | ||||||||
| Management/business/financial | RG | RG | RG | RG | ||||
| Professional | −1.01*** | (.197) | −.907*** | (.193) | −.480*** | (.129) | −.216*** | (.060) |
| Service | −.987*** | (.279) | −.663* | (.286) | −.505** | (.165) | −.292** | (.103) |
| Sales | −.773** | (.308) | −.631* | (.268) | −.169 | (.164) | .063 | (.110) |
| Office/administrative support | −.937*** | (.287) | −.779** | (.271) | −.392** | (.139) | −.220* | (.094) |
| Construction/extraction/farming/fishing/forestry/ | −.779 | (.630) | −.530 | (.470) | −.144 | (.290) | −.192 | (.145) |
| Installation/maintenance/repair | −1.48** | (.479) | −1.079* | (.458) | .124 | (.257) | −.130 | (.126) |
| Production | −2.14*** | (.525) | −1.637** | (.620) | −.627** | (.214) | −.278 | (.149) |
| Transportation/material moving | −1.82** | (.694) | −1.271 | (.668) | −.833*** | (.220) | −.477** | (.167) |
|
| ||||||||
| Job Tenure (years) | .026 | (.014) | .016 | (.012) | .000 | (.007) | −.001 | (.004) |
|
| ||||||||
| Number of Weekly Work Hours | .016* | (.008) | .010 | (.009) | .011** | (.004) | .007** | (.003) |
|
| ||||||||
| Seasonal Job (yes) | −.087 | (.434) | −.097 | (.362) | −.106 | (.235) | −.038 | (.154) |
|
| ||||||||
| Union Member (yes) | −.874*** | (.257) | −.725** | (.294) | −.103 | (.106) | .028 | (.063) |
|
| ||||||||
| Recession | −.354* | (.164) | −.500*** | (.157) | .321*** | (.080) | .269*** | (.051) |
|
| ||||||||
| Step 2 | ||||||||
|
| ||||||||
| Recession × Gender | .243 | (.306) | .018 | (.304) | .221 | (.150) | .100 | (.095) |
|
| ||||||||
| Recession × Black | −.882 | (606) | −.620 | (.622) | .124 | (.279) | −.044 | (.208) |
|
| ||||||||
| Recession × Hispanic | −.828 | (.709) | −.749 | (.606) | −.097 | (.382) | −.069 | (.252) |
|
| ||||||||
| Recession × Other | −1.356 | (.714) | −1.313 | (.696) | .235 | (.366) | .241 | (.445) |
|
| ||||||||
| Recession × Age | −.009 | (.013) | .005 | (.013) | −.006 | (.007) | −.001 | (.004) |
Note: N = 3,818. Unstandardized regression coefficients are reported. RG = reference group. The coefficients for family income have been rescaled to represent increments of $10,000. See Methods section for a discussion of the regression analyses for the various alcohol outcomes.
Discussion
This study is the first to explore the net change in alcohol use among employed adults in the U.S. related to the Great Recession, and provides the first broad assessment of net change in alcohol use related to any economic recession among the employed population. It is also the first study to explore the net relation of a recession to alcohol use in specific contexts. The results revealed that the relation of the Great Recession to drinker status differed by age. Consistent with the notion that middle-aged employees have more family and financial responsibilities than young employees, which may exacerbate the stressors imposed by the recession, the proportion of drinkers increased among middle-aged employees but not among younger employees. However, the relation of the recession to drinker status did not differ across gender and race subgroups.
The results also suggest that, among drinkers, the Great Recession was unrelated to usual frequency and quantity of alcohol use among employees. However, the recession was related to more frequent heavy drinking and intoxication. This differential pattern of results is consistent with self-medication models of stress-induced alcohol use (Conger, 1956; Cooper et al., 1995; McCarthy et al., 2010). The Great Recession also had differential relations with the contextual alcohol use outcomes among drinkers. Perhaps due to increases in workloads, concerns over performance monitoring, and job insecurity during the recession, the frequency and quantity of alcohol use during the workday decreased. However, the increased exposure to work-related and financial stressors and resulting distress that came with the recession may explain the increased frequency and quantity of alcohol use initiated within two hours of leaving work. Taken together, these findings suggest that during the recession, employees reduced alcohol use that could further increase the chance of job loss, but increased alcohol use in close proximity to the workday (shortly after work) that would not likely increase job loss. Using contextual measures of alcohol use provided unique information that could not be directly discerned from assessments of overall usual and excessive alcohol use. Finally, among drinkers, there was no evidence that the net relation of the Great Recession to any of the alcohol use outcomes differed across gender, race, or age.
Limitations
The present study has two main limitations that should be taken into account when interpreting the reported findings. First, this study relied on self-reports of alcohol use. Although it is naive to assume all self-reports are veridical, Turkkan (2000) and Baldwin (2000) point out that there may be no better measurement method with behaviors that may be hidden. Moreover, collateral informants are subject to the same types of biases as the target respondent (e.g., Connors & Maisto, 2003). The differential pattern of results comparing (a) usual alcohol use with excessive alcohol use and (b) workday alcohol use with after-work alcohol use provides some evidence that the various self-reported alcohol measures were assessing different dimensions of drinking behavior. Second, because the timing of a recession cannot be manipulated or predicted ahead of time, it is necessary to rely on observational data that may not provide a strong basis for causal conclusions. Despite controlling for differences in the composition of the workforce before and during/after the Great Recession, it is always possible that some other societal change co-occurred with the Great Recession and may be responsible for the observed changes in population-level alcohol use. However, this alternate societal change would not only have to be a potential cause of increased heavy drinking and intoxication, it would have to be a reasonable cause of the differential change in workday and after-work alcohol use.
Implications for Research and Practice
Because the timing of major economic downturns (and other major calamites) cannot be predicted ahead of time, research on net population change in alcohol use relies on ongoing national surveys that ideally can provide data using the same measures before, during, and after such events. In the U.S., there exist ongoing surveys that assess overall alcohol use to varying degrees and contain minimal information on basic employment status, such as the National Survey on Drug Use and Health and the Behavioral Risk Factor Surveillance System (BRFSS) surveys. The BRFSS surveys served as the data source for the Bor et al. (2013) and Dee (2001) studies described earlier. Although such ongoing studies can help assess the relation of economic recessions to various dimension of overall alcohol use, no ongoing national surveys assess alcohol use in specific contexts relevant to major subpopulations, such as employed adults (e.g., alcohol use during the workday and immediately after work). Therefore, exploring the association of recessions or other major events to contextual alcohol use currently relies on the fortuitous availability of data. To better understand the epidemiology of alcohol use among employed adults, the incorporation of a common set of contextual alcohol measures in ongoing national surveys would be beneficial. Alternately, an ongoing national survey focusing on the health of the employed population, taking into account contextually relevant outcomes and predictors (i.e., dimensions of the work environment) should be considered.
The present study provides additional and unique population-level support for the collective set of self-medication models of alcohol use (Conger, 1956; Cooper et al., 1995; McCarthy et al., 2010). Future research should attempt to identify the specific stressors that increase in the workforce during economic downturns (e.g., Green et al, in press). Such knowledge, along with the use of broader self-medication models incorporating mediating and moderating processes (e.g., Frone, 1999; Wolff, Rospenda, Richman, Liu, & Milner, 2013), can lead to more intensive longitudinal (panel and daily-process) studies to explore how and for whom these stressors, and by extension economic downturns, are related to changes in alcohol use among employed individuals.
Although the present results revealed that the Great Recession was associated with lower levels of workday alcohol use, the overall net effect associated with the recession was an increase in alcohol use among those who remain employed. This increase in alcohol use should be of concern to employers and policymakers for two reasons. The first reason for concern is that the increased alcohol use suggests that the stress and negative personal outcomes associated with recessions extend beyond those who lose employment. Even individuals who maintain employment during an economic downturn may experience a variety of work-related (Green et al., in press) and financial stressors that lead to negative health-related outcomes. The second reason for concern is that the increased alcohol use was exclusively due to excessive (heavy consumption and intoxication) and ill-timed (right after work) alcohol use that may interfere with responsibilities at work and at home, and may be more strongly associated with acute and longer-term negative consequences.
The present results, therefore, suggest that those managing work organizations need to be cognizant of the negative impact of economic downturns on their workforce and need to consider the implementation of interventions to prevent and reduce associated alcohol problems. This is particularly true because economic downturns are not rare, though some are more severe than others. For instance, since 1945 there have been 12 economic contractions in the U.S. (National Bureau of Economic Research, 2012). However, there is no research directly exploring interventions that may help reduce excessive and ill-timed alcohol use resulting from exposure to an economic downturn among those who maintain employment.
Nonetheless, from a broader research literature on employee assistance programs and workplace wellness programs, several types of interventions have focused on preventing or reducing employee alcohol misuse (see Ames & Bennett, 2011; Frone, 2013, for reviews). These interventions involve, either singly or in combination, several strategies: (a) providing educational information regarding heathy life styles and the negative effects of heavy drinking; (b) skills training in stress and emotion management, encouraging coworkers to get help for alcohol and drug use problems, and resilience (e.g., personal competence, social competence, personal responsibility, family coherence, and obtaining and providing social support); and (c) screening and brief intervention. In general, past evaluation research shows that educational interventions used on their own have no effect on reducing alcohol use or related problems. There is evidence that some multifaceted interventions involving skills training or screening and brief intervention can reduce alcohol use, though the effects are generally small and the evaluation studies have methodological limitations that need to be addressed.
More attention, therefore, needs to be devoted to the development and evaluation of workplace interventions to reduce heavy or ill-timed alcohol use among employees (Ames & Bennett, 2011; Frone, 2013). The development of workplace interventions need to incorporate program elements addressing the impact of external macro-economic shocks on employees and their alcohol use. Economic recessions may create sources of stress that are not as salient in nonrecessionary periods. For instance, the financial stress resulting from unrealized or realized losses in 401(k) accounts or concerns about pension plans may require that financial management be addressed during economic declines. Recent research also suggests that cognitively-based stress management training (i.e., reducing negative and increasing positive preservative cognition) may be useful in addressing excessive and ill-timed alcohol among employees (Frone, 2015).
In summary, to address effectively alcohol misuse among employees, especially during economic recessions, more workplace intervention research needs to be conducted (see Ames & Bennett, 2011). And, as noted by Baumeister and Alghamdi (2015, p. 624), “Meanwhile, though, practitioners cannot generally wait until research has identified what works best. Practitioners need to do the best they can for people who have issues right now. So it is urgent that research at least try to keep up.”
Conclusion
This study shows that when exploring the potential effect of economic recessions on alcohol use in the general workforce, research needs to be more inclusive. In addition to workers who lose jobs, research needs to include workers who maintain employment. Even among the employed, economic downturns can create sources of work-related and financial stress that may lead to lower levels of alcohol use during the workday, but higher levels of excessive and ill-timed alcohol use away from work. Therefore, among the employed subpopulation, researchers needs to better delineate the stressors that are caused by or exacerbated during economic recessions, the processes linking these stressors to excessive and ill-timed alcohol use, and the types of interventions required to mitigate the experienced stressors and dysfunctional alcohol use.
Acknowledgments
Data collection was supported by grants (R01-AA12412, R01-AA016592) from the National Institute on Alcohol Abuse and Alcoholism to Michael R. Frone. The content of this project is solely the responsibility of the author and does not necessarily represent the official views of the National Institute on Alcohol Abuse and Alcoholism or the National Institutes of Health. These agencies had no role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.
Footnotes
1It should be noted that several of the observed differences between Study 1 and Study 2 in the weighted demographic variables, such as education, family income, occupation, and weekly work hours, are consistent with the impact of the recession on the workforce (e.g., Goodman & Mance, 2011; Hout, Levanon, & Cumberworth, 2011). The hardest hit occupations in terms of job loss were in construction, manufacturing, and transportation occupations, and the least affected occupations were professional and managerial occupations. Also, during the recession, there was an inverse relation between job loss and education, and average weekly work hours declined. Collectively, these changes may explain (a) the higher proportion of participants in managerial and professional occupations, and the lower proportion of participants in construction, extraction, installation, maintenance, production, and transportation occupations; (b) the increase in total family income; (c) the slight increase in average education; and (d) the 0slight decrease in average weekly work hours in Study 2 compared with Study 1.
References
- Ames GM, Bennett JB. Prevention interventions of alcohol problems in the workplace: A review and guiding framework. Alcohol Research & Health. 2011;34:175–187. [PMC free article] [PubMed] [Google Scholar]
- Baldwin W. Information no one else knows: The value of self-report. In: Stone AA, Turkkan JS, Bachrach CA, Jobe JB, Kurtzman HS, Cain VS, editors. The science of self-report: Implications for research and practice. Mahwah, NJ: Lawrence Erlbaum; 2000. pp. 3–7. [Google Scholar]
- Baumeister RF, Alghamdi N. Resource-based interventions in the workplace: Integration, commentary, and recommendations. Journal of Occupational and Organizational Psychology. 2015;88:623–629. [Google Scholar]
- Bor J, Basu S, Coutts McKee M, Stuckler D. Alcohol use during the Great Recession of 2008–2009. Alcohol and Alcoholism. 2013;48:343–348. [PubMed] [Google Scholar]
- Bouchery EE, Harwood HJ, Sacks JJ, Simon CJ, Brewer RD. Economic costs of excessive alcohol consumption in the U.S., 2006. American Journal of Preventive Medicine. 2011;41:516–524. [PubMed] [Google Scholar]
- Cherpitel CJ. Alcohol and injuries: A review of international emergency room studies since 1995. Drug and Alcohol Review. 2007;26:201–214. [PubMed] [Google Scholar]
- Conger J. Reinforcement theory and the dynamics of alcoholism. Quarterly Journal of Studies on Alcohol. 1956;17:296–305. [PubMed] [Google Scholar]
- Connors GJ, Maistro SA. Drinking reports from collateral individuals. Addiction. 2003;98(Suppl. No. 2):21–29. [PubMed] [Google Scholar]
- Cooper ML, Frone MR, Russell M, Mudar P. Drinking to regulate positive and negative emotions: A motivational model of alcohol use. Journal of Personality and Social Psychology. 1995;69:990–1005. [PubMed] [Google Scholar]
- Cooper ML, Russell M, Skinner JB, Frone MR, Mudar P. Stress and alcohol use: The moderating effect of gender, coping and alcohol expectancies. Journal of Abnormal Psychology. 1992;101:139–150. [PubMed] [Google Scholar]
- Council of Economic Advisers. Economic report of the president. Washington, DC: U.S. Government Printing Office; 2010. [Google Scholar]
- Dee TS. Alcohol abuse and economic conditions: Evidence from repeated cross-sections of individual-level data. Health Economics. 2001;10:257–270. [PubMed] [Google Scholar]
- de Goeij MCM, Suhrcke M, Toffolutti V, van de Mheen D, Schoenmakers TM, Hunst AE. How economic crises affect alcohol consumption and alcohol-related health problems: A realist systematic review. Social Science and Medicine. 2015;131:131–146. [PubMed] [Google Scholar]
- Firebaugh G. Analyzing repeated surveys. Thousand Oaks, CA: Sage; 1997. [Google Scholar]
- Fischer CS, Hout M. Century of difference. New York: Russell Sage Foundation; 2006. [Google Scholar]
- Frone MR. Work stress and alcohol use. Alcohol Research and Health. 1999;23:284–291. [PMC free article] [PubMed] [Google Scholar]
- Frone MR. Prevalence and distribution of illicit drug use in the workforce and in the workplace: Findings and implications from a U.S. national survey. Journal of Applied Psychology. 2006;91:856–869. [PubMed] [Google Scholar]
- Frone MR. Alcohol and illicit drug use in the workforce and workplace. Washington, DC: American Psychological Association; 2013. [Google Scholar]
- Frone MR. Relations of Negative and Positive Work Experiences to Employee Alcohol Use: Testing the Intervening Role of Negative and Positive Work Rumination. Journal of Occupational Health Psychology. 2015;20:148–160. [PMC free article] [PubMed] [Google Scholar]
- Frone MR, Trinidad JR. Relation of supervisor social control to employee substance use: Considering the dimensionality of social control, temporal context of substance use, and substance legality. Journal of Studies on Alcohol and Drugs. 2012;73:303–310. [PMC free article] [PubMed] [Google Scholar]
- Goodman CJ, Mance SM. Employment loss and the 2007–09 recession: An overview. Monthly Labor Review. 2011;134:3–12. [Google Scholar]
- Green F, Felstead A, Gallie D, Inanc H. Job-related well-being through the great Recession. Journal of Happiness Studies (in press) [Google Scholar]
- Grusky DB, Western B, Wimer C. The consequences of the great recession. In: Grusky DB, Western B, Wimer C, editors. The great recession. New York, NY: Russell Sage Foundation; 2011. pp. 3–20. [Google Scholar]
- Hart Research Associates. NBC News/Wall Street Journal Survey. Washington, DC: Author; 2014. (Study 14133) http://online.wsj.com/public/resources/documents/WSJNBCpoll03052014.pdf. [Google Scholar]
- Hout M, Levanon A, Cumberworth E. Job loss and unemployment. In: Grusky DB, Western B, Wimer C, editors. The great recession. New York, NY: Russell Sage Foundation; 2011. pp. 59–20. [Google Scholar]
- Lehtonen R, Pahkinen E. Practical methods for design and analysis of complex surveys. 2nd. New York: Wiley; 2004. [Google Scholar]
- Lo CC, Cheng TC. Heavy drinking during periods of high unemployment: 15- year trend study of the role of race/ethnicity. Drug and Alcohol Dependence. 2013;133:383–390. [PubMed] [Google Scholar]
- Long JS. Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage; 1997. [Google Scholar]
- McCarthy DE, Curtin JJ, Piper ME, Baker TB. Negative reinforcement: Possible clinical implications of an integrative model. In: Kassel JD, editor. Substance abuse and emotion. Washington, DC: American Psychological Association; 2010. pp. 15–42. [Google Scholar]
- Mulia N, Zemore SE, Murphy R, Liu H, Catalano R. Economic loss and alcohol consumption and problems during the 2008 to 2009 U.S. recession. Alcoholism: Clinical and Experimental Research. 2014;38:1026–1034. [PMC free article] [PubMed] [Google Scholar]
- National Bureau of Economic Research. US business cycle expansions and contractions. Cambridge, MA: Author; 2012. http://www.nber.org/cycles.html. [Google Scholar]
- Nolen-Hoeksema S. Gender differences in risk factors and consequences for alcohol use and problems. Clinical Psychology Review. 2004;24:981–1010. [PubMed] [Google Scholar]
- Normand J, Lempert RO, O’Brien CP. Under the influence? Drugs and the American work force. Washington, DC: National Academy Press; 1994. [Google Scholar]
- Pew Research Center. The sandwich generation: Rising financial burdens for middle-aged Americans. Washington, DC: Author; 2013. [Google Scholar]
- Popovici I, French MT. Does unemployment lead to greater alcohol consumption? Industrial Relations. 2013;52:444–466. [PMC free article] [PubMed] [Google Scholar]
- Rehm J, Taylor B, Room R. Global burden of disease from alcohol, illicit drugs, and tobacco. Drug and Alcohol Review. 2006;25:503–513. [PubMed] [Google Scholar]
- Richman JA, Rospenda KM, Johnson TP, Cho YI, Vijayasira G, Cloninger L, Wolff JM. Drinking in the age of the great recession. Journal of Addictive Behaviors. 2012;31:158–172. [PMC free article] [PubMed] [Google Scholar]
- Stata Corporation. Stata statistical software: Release 14. College Station, TX: StataCorp; 2015. [Google Scholar]
- Turkkan JS. General issues in self-report. In: Stone AA, Turkkan JS, Bachrach CA, Jobe JB, Kurtzman HS, Cain VS, editors. The science of self-report: Implications for research and practice. Mahwah, NJ: Lawrence Erlbaum; 2000. pp. 1–2. [Google Scholar]
- Wolff JM, Rospenda KM, Richman JA, Liu L, Milner LA. Work-family conflict and alcohol use: Examination of a moderated mediation model. Journal of Addictive Diseases. 2013;32:85–98. [PMC free article] [PubMed] [Google Scholar]
- Wooldridge JM. Econometric analysis of cross section and panel data. Cambridge, MA: MIT Press; 2002. [Google Scholar]
- Zemore SE, Mulia N, Jones-Webb RJ, Liu H, Schmidt L. The 2008–2009 recession and alcohol outcomes: Differential exposure and vulnerability for black and latino populations. Journal of Studies on Alcohol and Drugs. 2013;74:9–20. [PMC free article] [PubMed] [Google Scholar]