Abstract
Despite significant declines in the use of cigarettes, a significant proportion of adults smoke. This study explores the association between smoking and health-related quality of life (HRQoL) by age. The 2016 Behavioral Risk Factor Surveillance System survey was administered to adults in 50 states and District of Columbia (n = 437,195). Physically unhealthy days (PUDs) and mentally unhealthy days (MUDs)) were regressed on age strata (18–24, 25–34, 35–44, 45–54, 55–64, ≥ 65 years) and smoking status (never, former, someday, and everyday) using negative binomial regression models with adjustment for sociodemographic covariates. For each age group, everyday smoking highly predicted PUDs and MUDs. Predicted PUDs increased with age; predicted MUDs decreased with age. Among adults aged 45–54 and 55–64 years, 3-day difference in PUDs was observed between never smokers and everyday smokers. Among young adults (18–24 years), a 4.3-day difference in MUDs was observed between everyday and never smokers. The discrepancies were nonlinear with age. The observed relationship between smoking and HRQoL provides novel information about the need to consider age when designing community-based interventions. Additional research can provide needed depth to understanding the relationship between smoking and HRQoL in specific age groups.
1 |. INTRODUCTION
Smoking is an established risk factor for a plethora of diseases and other adverse health effects, which include cardiovascular disease, cancer, pulmonary disease, adverse reproductive outcomes, and exacerbation of chronic health conditions, thereby smoking continues to be a leading preventable cause of disability, disease, and death in the United States (U.S. Department of Health and Human Services [HHS], 2014). The biological mechanisms for the causal relationship between tobacco smoke and disease are now well documented (Centers for Disease Control and Prevention [CDC], 2010). Despite significant declines in the use of cigarettes (CDC, 2016), most recent estimates indicate that smoking is responsible for 480,000 deaths annually and annual smoking-attributable economic costs of $289–332.5 billion, including $132.5–175.9 billion for direct medical care of adults (HHS, 2014).
In addition to pathophysiological disease caused by smoking (CDC, 2010), there now exists a large body of research documenting the relationship between smoking and mental illness. For example, in 2013 the CDC reported that among a nationally representative sample of adults aged ≥18 years with any mental illness, 36.1% were current smokers compared to 21.4 % with no mental illness (CDC, 2013). Other studies also have documented the strong association between depression, anxiety, schizophrenia, and serious psychological distress and cigarette smoking (Dube et al., 2009; Hall & Prochaska, 2009; McClave, Davis, McKnight, & Dube, 2010; Trosclair & Dube, 2010).
In addition to the associations observed between smoking and mental illness, studies have also documented that current cigarette smoking is associated with poor health-related quality of life (HRQoL) across the lifespan (Dube, Thompson, Homa, & Zack, 2013; Heikkinen, Jallinoja, Saarni, & Patja, 2008; McClave, Dube, Strine, & Mokdad, 2009). HRQoL has become a health measure of particular interest and attention. HRQoL encompasses aspects of self-reported health appraisals that include subjective experiences and internal processing as it relates to health and well-being across multiple dimensions, which go beyond objectifying health status.
While studies have demonstrated that current smoking is associated with poor HRQoL, few studies have examined the association between smoking and current physical and mental health across age strata among adults. The epidemiologic transition theory is an important framework and is used to understand patterns of disease and behavioral risks in populations with changing demographics, such as age (Omran, 1971). For example, from 2010 to 2016, the number of adults aged 50 years and older in the United States has steadily increased (U.S. Census, 2017), largely due to the “baby boom” population. In the United States, medical advances in infectious disease control and reproductive health have resulted in reduction in birth and death rates and increased longevity. With reductions in infectious diseases and an increase in life expectancy, chronic diseases such as cardiovascular disease, malignancies, and chronic obstructive pulmonary disease are now the leading causes of death (McKeown, 2009; Omran, 1971), for which smoking is a leading risk factor. The epidemiologic transition theory can be utilized to inform how changing demographic patterns in communities, such as age structure, may contribute to the relationship between smoking and HRQoL.
We therefore sought to assess the association between smoking status, age, and HRQoL using the 2016 Behavioral Risk Factor Surveillance System (BRFSS). These results could be valuable to public health practitioners, could inform the development of novel, age-specific smoking interventions, and could be important information for health promotion campaigns.
2 |. METHOD
The BRFSS is a standardized, random-digit dialed, population-based, state-representative telephone survey that assesses key behavioral risk factors and chronic conditions among noninstitutionalized adults aged 18 years and older in the United States and participating territories annually. BRFSS data have been found to provide valid and reliable prevalence estimates of chronic conditions and health behaviors when compared with national household surveys (CDC, 2003; Nelson, Holtzman, Bolen, Stanwyck, & Mack, 2001; Nelson, Powell-Griner, Town, & Kovar 2003). Since 2011, the BRFSS had been conducting both landline telephone- and cellular telephone-based surveys (Hu, Balluz, Battaglia, & Frankel, 2011). Also, starting in 2011, a new weighting methodology called iterative proportional fitting (or “raking”) replaced the poststratification method to weight BRFSS data. The 2016 BRFSS raking method includes categories of age by gender, detailed race and ethnicity groups, education levels, marital status, regions within states, gender by race and ethnicity, telephone source, renter or owner status, and age groups by race and ethnicity.
The BRFSS survey consists of three components: a core questionnaire, optional modules, and state-specific questions. The BRFSS is exempt from institutional review board review by the Human Research Protection Office, Centers for Disease Control and Prevention and by Georgia State University. Cooperation rate is the proportion of all respondents interviewed among all eligible units in which a respondent was selected and actually contacted. The BRFSS has calculated 2016 response rates using AAPOR Response Rate #4 (https://www.cdc.gov/brfss/annual_data/2016/pdf/2016-sdqr.pdf). The response rate is the number of respondents who completed the survey as a proportion of all eligible and likely eligible people. The median survey response rate for all states, territories, and Washington, DC, in 2016 was 47.0%, and ranged from 30.7% to 65.0%. The technical information, questionnaire, and survey data are available online at https://www.cdc.gov/brfss/data_documentation/index.htm. The current analysis was based on 2016 BRFSS data with complete information on our outcomes and covariates obtained from 50 states and the District of Columbia (n = 437,195).
2.1 |. Measures
2.1.1 |. Mentally and physically unhealthy days
Two domains ascertaining mental and physical health are from the HRQoL measures within the BRFSS core questionnaire. The first measure, physically unhealthy days (PUDS), records the responses to the following question: “Now thinking about your physical health, which includes physical illness and injuries, for how many days during the past 30 days was your physical health not good?” The second measure, mentally unhealthy days (MUDS), records the responses to the following question: “Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good?” The values ranged from 0 to 30 for both measures.
2.1.2 |. Smoking status
Questions used to ascertain smoking status were “Have you smoked at least 100 cigarettes in your entire lifetime?” and “Do you now smoke cigarettes every day, some days, or not at all?” Respondents who reported smoking less than 100 cigarettes in his/her lifetime were categorized as never smokers. Respondents who reported having smoked 100 cigarettes in their lifetime and who were currently smoking were categorized as current smokers. Respondents who reported having smoked 100 cigarettes in their lifetime and who were not currently smoking were categorized as former smokers. We also examine current smokers as to two separate groups–everyday smokers and someday smokers.
2.1.3 |. Age group
Age was defined using six groupings: 18–24, 25–34, 35–44, 45–54, 55–64, ≥ 65 years.
2.1.4 |. Demographic and socioeconomic covariates
We included the following demographic covariates in the regression analysis: sex (male, female); race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, and other); and marital status (married or partner; previously married [divorced, separated or widowed]; never married). Socioeconomic covariates were as follows: educational attainment (< high school; high school or equivalent; some college; bachelor degree and higher); annual household income (< 15K, 15– < 25K, 25– < 35K, 35– < 50K, 50K or more; don’t know/not sure/missing); and health insurance coverage yes/no (“Do you have any kind of health care coverage, including health insurance, prepaid plans such as HMOs [health maintenance organizations], or government plans such as Medicare?”). State identifier was also entered into the regression model as a covariate.
2.2 |. Statistical Analysis
For all statistical analyses, we used SAS–callable SUDAAN (release 9.4) to account for the complex sampling design. We first used the CROSSTAB and DESCRIPT procedures to produce descriptive statistics on outcomes and covariates by smoking status. Because the PUDS and MUDS data were non-normally distributed, we used the NBREG procedure with survey command in STATA (version 14) to determine the actual probability distribution of the data. We found that negative binomial distribution fit the data better than a Poisson distribution. We then ran two negative binomial regression models using MUDS or PUDS as the dependent variables. We used, as independent variables, the previously described smoking status, six age strata, interaction of smoking status by age strata, demographic and socioeconomic characteristic categories, and healthcare coverage.
Incident rate ratios (IRRs) and 95% confidence intervals were obtained from the NBREG procedure. An IRR = 1.5 means the incident rate for group B (e.g., unhealthy days) is 1.5 times the incidence rate for the reference group (Group A), holding the other covariates constant. Adjusted (or predicted) estimates of unhealthy days for each combination of smoking status and age strata were the average probability if everyone in the data were treated as if they belonged to that specific category. The standard errors of the adjusted means were obtained by Delta method because it is appropriate to estimate standard errors from a nonlinear transformation data in large samples. A P-value less than 0.05 is considered statistically significant. Bonferroni correction was applied for multiple pairwise comparisons of predicted means of PUDs and MUDs between smoking status groups with each age stratum. For six pairwise comparisons within each age stratum, the new critical P value would be 0.05/6 ≈0.008.
3 |. RESULTS
Aggregate BRFSS 2016 indicated that 15% of the U.S. adult population were smokers, 26% were former smokers. Former smokers were, on average, 10 or more years older than never smokers, someday smokers, and everyday smokers. Someday and every day smokers were less educated, received less household income, less likely to be married, and disproportionally less likely to have insurance coverage during the last 12 months (Table 1).
TABLE 1.
Descriptive characteristics for health-related quality of life and sociodemographic variables by smoking status (BRFSS 2016)
| Never smoker (n = 256,659) |
Former smoker (n = 133,265) |
Someday smoker (n = 19,458) |
Everyday smokers (n = 48,622) |
|
|---|---|---|---|---|
| Continuous variables | Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) |
| Physically Unhealthy days | 3.1 (0.0) | 4.8 (0.0) | 5.6 (0.1) | 6.5 (0.1) |
| Mentally Unhealthy days | 2.9 (0.0) | 3.5 (0.0) | 5.8 (0.1) | 7.0 (0.1) |
| Categorical variables | % (SE) | % (SE) | % (SE) | % (SE) |
| Sex | ||||
| Women | 57.7 (0.2) | 47.0 (0.2) | 45.8 (0.6) | 48.5 (0.4) |
| Men | 42.3 (0.2) | 53.0 (0.2) | 54.2 (0.6) | 51.5 (0.4) |
| Age group | ||||
| 18 to 24 yr | 11.3 (0.1) | 2.1 (0.1) | 10.1 (0.4) | 5.5 (0.2) |
| 25 to 34 yr | 16.2 (0.1) | 8.9 (0.2) | 21.8 (0.5) | 15.8 (0.3) |
| 35 to 44 yr | 14.7 (0.1) | 11.1 (0.2) | 16.1 (0.5) | 17.0 (0.3) |
| 45 to 54 yr | 17.6 (0.1) | 14.9 (0.2) | 18.7 (0.5) | 22.8 (0.3) |
| 55 to 64 yr | 18.0 (0.1) | 22.7 (0.2) | 20.4 (0.5) | 23.8 (0.3) |
| ≥ 65 yr | 22.2 (0.1) | 40.3 (0.2) | 12.9 (0.4) | 15.2 (0.3) |
| Race/ethnicity | ||||
| White non-Hispanic | 64.6 (0.2) | 79.1 (0.2) | 60.7 (0.6) | 73.8 (0.4) |
| Black non-Hispanic | 11.2 (0.1) | 6.6 (0.1) | 14.6 (0.4) | 10.7 (0.2) |
| Hispanic | 16.1 (0.1) | 9.1 (0.2) | 16.2 (0.5) | 8.5 (0.3) |
| Other | 8.1 (0.1) | 5.2 (0.1) | 8.5 (0.4) | 7.0 (0.2) |
| Education attainment | ||||
| < High school | 7.6 (0.1) | 8.1 (0.1) | 14.0 (0.5) | 15.2 (0.3) |
| High school | 21.8 (0.1) | 26.3 (0.2) | 32.0 (0.6) | 38.6 (0.4) |
| > High school | 70.6 (0.2) | 65.6 (0.2) | 54.0 (0.6) | 46.2 (0.4) |
| Annual household Income | ||||
| < $25,000 | 19.6 (0.1) | 19.9 (0.2) | 35.5 (0.6) | 36.7 (0.4) |
| $25,000-$74,999 | 31.6 (0.2) | 35.6 (0.2) | 32.9 (0.6) | 35.5 (0.4) |
| > $75,000 | 33.3 (0.2) | 31.0 (0.2) | 18.7 (0.5) | 15.2 (0.3) |
| Income missing | 15.5 (0.1) | 13.5 (0.2) | 12.9 (0.4) | 12.6 (0.2) |
| Marital status | ||||
| Married/couple | 59.2 (0.2) | 62.3 (0.2) | 41.5 (0.6) | 44.1 (0.4) |
| Separated/divorced /widowed | 18.3 (0.1) | 27.0 (0.2) | 29.8 (0.6) | 33.5 (0.4) |
| Never married | 22.5 (0.2) | 10.6 (0.2) | 28.7 (0.6) | 22.4 (0.3) |
| Healthcare coverage | ||||
| No | 9.3 (0.1) | 6.1 (0.1) | 16.1 (0.5) | 16.7 (0.3) |
| Yes | 90.7 (0.1) | 93.9 (0.1) | 83.9 (0.5) | 83.3 (0.3) |
Note. SE = standard error. Chi-square tests between smoking status and other categorical variables were all significant (p < 0.001).
Results for main effects of the covariates on the physically and mentally unhealthy days are presented in Table 2. The following factors were significantly associated with high levels of physically and mentally unhealthy days during the past 30 days (IRR > 1 and 95% CI did not include 1): women, not currently married. For example, PUDs for women was 1.17 times the PUDs for men; MUDs for women was 1.43 times the MUDs for men. The following factors were significantly associated with less physically and mentally unhealthy days during the past 30 days (IRR < 1 and 95% CI did not include (a): non-Hispanic Black and Hispanics (vs. non-Hispanic White), higher household income, and having healthcare coverage. There was also significant State level variation in the PUDs and MUDs (p < 0.001).
TABLE 2.
Incident rate ratio (IRR) and 95% confidence interval for physically and mentally unhealthy days in association with covariates (BRFSS 2016)
| Variables | Physically unhealthy days IRR (95% CI) |
Mentally unhealthy days IRR (95% CI) |
|---|---|---|
| Sex | ||
| Male | Ref | ref |
| Female | 1.17 [1.13, 1.20] | 1.43 [1.40-1.48] |
| Race/ethnicity | ||
| White non-Hispanic | Ref | ref |
| Black non-Hispanic | 0.93 [0.88, 0.97] | 0.92 [0.88, 0.97] |
| Hispanic | 0.92 [0.88, 0.97] | 0.80 [0.76, 0.84] |
| Other | 1.03 [0.97, 1.09] | 0.97 [0.92, 1.02] |
| Education | ||
| Less than high school | Ref | ref |
| High school or equivalent | 0.81 [0.78, 0.85] | 0.85 [0.81, 0.89] |
| Some college or higher | 0.74 [0.71, 0.77] | 0.82 [0.78, 0.86] |
| Household income | ||
| < $25,000 | Ref | ref |
| $25,000-$74,999 | 0.60 [0.58, 0.62] | 0.67 [0.65, 0.69] |
| > $75,000 | 0.39 [0.37, 0.40] | 0.48 [0.46, 0.50] |
| Missing income | 0.69 [0.66, 0.71] | 0.68 [0.65, 0.70] |
| Marital status | ||
| Currently married or unmarried couple | Ref | ref |
| Separated, divorced, or widowed | 1.19 [1.15, 1.22] | 1.34 [1.30, 1.39] |
| Never married | 1.09 [1.04, 1.13] | 1.25 [1.20, 1.30] |
| Healthcare coverage | ||
| No 0 | Ref | ref |
| Yes 1 | 1.18 [1.13, 1.24] | 1.04 [1.00, 1.09] |
Note. SE = standard error. Never smoker was defined as never smoking at least 100 cigarettes during lifetime; former smoker was defined as having smoked at least 100 cigarettes and currently does not smoke; someday and everyday smoker were defined as smoking at least 100 cigarettes and currently smokes someday or everyday, respectively. The IRR was obtained from zero-inflated negative binomial regression model adjusted for sex, race/ethnicity, education attainment, household income, marital status, healthcare coverage and state.
Overall for each age group, everyday smokers ranked highest for predicted PUDs and MUDs, followed by someday smokers and former smokers. Never smokers always had lowest number of predicted PUDs and MUDs (Table 3, Figure 1, and Figure 2). Predicted PUDs increased with age; predicted MUDs decreased with age.
TABLE 3.
Predicted number of days (SE) for smoking by age interactions for physically and mentally unhealthy days
| Physically unhealthy days | Mentally unhealthy days | ||
|---|---|---|---|
| Age group | Smoking group | Predicted means (SE) | Predicted means (SE) |
| 18–24 years | Everyday smoker | 3.00 (0.22)c | 8.28 (0.41)ac |
| Someday smoker | 2.33 (0.19)d | 5.83 (0.36)ad | |
| Former smoker | 2.50 (0.21)e | 6.15 (0.37)e | |
| Never smoker | 1.83 (0.06)cde | 4.02 (0.10)acde | |
| 25–34 years | Everyday smoker | 3.58 (0.17)ac | 7.13 (0.22)ac |
| Someday smoker | 2.91 (0.20)d | 5.53 (0.29)d | |
| Former smoker | 2.97 (0.15)ae | 5.45 (0.19)e | |
| Never smoker | 2.17 (0.06)cde | 3.57 (0.08)cde | |
| 35–44 years | Everyday smoker | 4.65 (0.17)ac | 6.98 (0.22)ac |
| Someday smoker | 4.13 (0.28)d | 5.93 (0.32)abd | |
| Former smoker | 3.68 (0.14)ae | 4.82 (0.17)be | |
| Never smoker | 2.75 (0.07)cde | 3.30 (0.08)cde | |
| 45–54 years | Everyday smoker | 6.44 (0.17)ac | 6.64 (0.19)ac |
| Someday smoker | 5.88 (0.26)bd | 5.62 (0.24)abd | |
| Former smoker | 4.74 (0.13)abe | 4.45 (0.13)be | |
| Never smoker | 3.56 (0.08)cde | 3.27 (0.07)cde | |
| 55–64 years | Everyday smoker | 6.86 (0.17)ac | 5.51 (0.16)c |
| Someday smoker | 6.87 (0.28)bd | 5.09 (0.23)bd | |
| Former smoker | 5.67 (0.11)abe | 3.94 (0.98)be | |
| Never smoker | 4.12 (0.08)cde | 2.86 (0.06)acde | |
| ≥ 65 years | Everyday smoker | 5.74 (0.20)c | 3.61 (0.18)ac |
| Someday smoker | 5.21 (0.29)d | 2.80 (0.17)bd | |
| Former smoker | 5.24 (0.08)e | 2.29 (0.06)be | |
| Never smoker | 4.15 (0.06)cde | 1.84 (0.04)cde |
Note. SE = standard error. Never smoker was defined as never smoking at least 100 cigarettes during lifetime; former smoker was defined as having smoked at least 100 cigarettes and currently does not smoke; someday and everyday smoker were defined as smoking at least 100 cigarettes and currently smokes someday or everyday, respectively. The predicted days were obtained from zero-inflated negative binomial regression model adjusted for sex, race/ethnicity, education attainment, household income, marital status, healthcare coverage and state. The groups with same letters were significantly different from each other.
FIGURE 1.
Physically unhealthy days (SE) by smoking status and age
FIGURE 2.
Mentally unhealthy days (SE) by smoking status and age
The largest differences in PUDs between everyday smokers and never smokers occurred at 45–54 and 55–64 years of age, when PUDs reached a peak for everyday smokers. For the 45–54 age group, the never smokers reported an average of 3.6 PUDs, while the everyday smokers reported 6.4 PUDs, representing almost a three-day difference. For the 55–64 age group, the never smokers reported an average of 4.1 PUDs, while the everyday smokers reported 6.9 PUDs, also representing an almost a three-day difference. These differences were significantly greater than other age groups: 18–24 (1.2 days), 25–34 (1.4 days), 35–44 (2 days), and ≥ 65 years (1.6 days). There was not a significant difference in PUDs between someday smokers and everyday smokers. A significant difference in predicted PUDs between everyday smokers and former smokers occurred at 25–64 years of age. PUDs for someday smokers aged 45–64 years occurred about one-day higher on average than former smokers. For all age groups, PUDs for former smokers occurred about one-day higher on average than never smokers.
For the youngest age group (18–24 years), the largest differences in MUDs (diff = 4.3 days) occurred between never (MUD = 4.0 days) and everyday smokers (MUD = 8.3 days). The discrepancy steadily shrank with age: 25–34 (diff = 3.6 days), 35–44 (diff = 3.7 days), 45–54 (diff = 3.3 days), 55–64 (diff = 2.7 days), and ≥ 65 years (diff = 1.8 days). Although there were significant differences in MUDs between everyday smokers and someday smokers aged 18–54 years, the differences shrank with age, ranging from 2.5 days for 18–24-year-olds to 1.0 day for 45–54-year-olds. For the 35–64 age group, MUDs for someday smokers occurred about one-day higher on average than former smokers. Although there were differences in MUDs between former smokers and never smokers of all age groups, the differences shrank with age, ranging from 2.1 days for 18–24-year-olds to 0.5 day for those ≥ 65 years.
4 |. DISCUSSION
Smoking has declined significantly among the U.S. population and the CDC’s most recent estimate suggests that 15.9% of adults are current smokers (CDC, 2016; Clarke, Norris, & Schiller, 2017). Estimates from the present study using the BRFSS align closely with the national estimates and indicate that about 36 million of adults (14.9%) across the 50 states are current smokers. Our findings also showed significant age differences in the burden of self-reported mentally and physically unhealthy days by smoking status. In particular, we found that for adults who smoke, there were large differences among young adults in the number of self-reported MUDs. However, for PUDs, the largest differences were observed for middle-aged and older adults who were current smokers. These findings suggest that perceptions of global health for these two key dimensions (mental and physical) vary by both age and smoking.
For middle-aged and older adults, physical health is exacerbated by risk behaviors, such as smoking. Close to 1 in 5 adults aged 45 to 64 years reported current smoking (CDC, 2016) and this estimate has remained stable since 2002 (Dube & Wu, 2015). Indeed, the long-term effects of smoking on physical health outcomes are well established and the plausibility for greater PUDs among middle-aged and older adults provides further evidence of the negative health consequences associated with cigarettes. Interestingly, we surmise that the diminished group difference in the oldest age group is most likely due to differential mortality, where the most unhealthy current smokers tend to die at a young age.
The group of someday smokers is a heterogeneous group, which may comprise smokers who smoke less than everyday smokers and who have reduced smoking over time; therefore, it is not surprising to find that their HRQOL measures were in between everyday smokers and former smokers. Similarly, for most age strata, the former smoker group demonstrated a HRQOL profile in between someday smokers and never smokers. This indicated that although quitting smoking at any age is beneficial, some of the adverse effects on physical health may be irreversible but the mental health impacts related to the addiction to tobacco may still remain, though to a lesser degree, after reducing or quitting smoking.
In the present study, MUDs was greater among young adults compared to older adults, regardless of smoking status. However, MUDs was greatest among everyday smokers among all age groups. While we cannot provide any aspects of causality with these findings, the recognition of mentally unhealthy days may be more evident for young adults who smoke. Young adults may experience challenges during this transitory stage. In particular, they may be in the midst of completing their education, making decisions about career and future, establishing complete independence from family and parents, and beginning their own families. Thus, it is not surprising that transition into adulthood may be a stressful period, and may partially explain the observed findings of greater mentally unhealthy days among young adults who smoke compared to the other age groups. Because of the natural progression of physical disease with age, most young adults who smoke may not manifest or perceive poor physical health. Furthermore, the differences we observed in mental and physical health perceptions also underscore the importance of examining other domains of well-being (e.g., social, emotional, intellectual) across age.
Support for the findings of young adults reporting high level of MUDs were reported in other studies, which documented the association between mental illness and smoking. A 2013 report released by the CDC noted that young adults with mental illness had a higher prevalence of smoking than the other age groups (CDC, 2013). The relationship between smoking and mental illness is not surprising. Nicotine, tobacco products contain, can have mood-altering effects by stimulating the central nervous system. This action of nicotine may temporarily ameliorate symptoms often experienced with stress and poor mental well-being (Hall & Prochaska, 2009). Moreover, smoking and poor mental well-being are documented to co-occur, and as empirical evidence indicates, this relationship has been observed among young adults as well as youth (CDC, 2013; DiFranza et al., 2004; Dube et al., 2013).
The success of tobacco control efforts through implementation of coordinated effective strategies cannot be emphasized enough (CDC, 2014). However, if progress is to continue, understanding the phenomena of smoking and subjective well-being across populations will help to deepen our knowledge of the subtleties of the behavior that are not readily observable by simply monitoring trends. Our findings provide an opportunity to highlight that for young adults who smoke, tobacco prevention and control efforts may need to include components that focus on the stress response and other mental health issues. Clinicians who see young patients have an opportunity to not only screen for tobacco use but also consider screening for mental well-being. Smokers in settings such as universities and colleges are also target populations in which the co-occurrence of smoking and mental well-being can be addressed.
More importantly, HRQoL continues to gain national attention as an important global health outcome to track in the U.S. population. Healthy People 2020 has approved two HRQOL objectives–one for physical health and one for mental health–using the NIH Patient Reported Outcomes Measurement Information Systems (PROMIS) global health measure (Hays, Bjorner, Revicki, Spritzer, & Cella, 2009). The CDC PUDs and MUDs have been shown to measure the same constructs as the PROMIS items when they have been examined together to assess their psychometric properties (Barile et al., 2013).
4.1 |. Limitations
There are several limitations to our study. First, causality cannot be established due to the cross-sectional design of this study; our data can simply provide the association of MUDs and PUDs with smoking status across all age groups. Second, the BRFSS is based upon self-reports, and reporting and recall bias may have occurred for both smoking and reporting the number of mentally and physically unhealthy days.
4.2 |. Conclusion
Despite these limitations, our study’s findings are important to consider, especially in light of how age interacts with smoking as a key indicator of health, HRQoL. Wider implementation of population-based effective strategies known to prevent and reduce tobacco use is needed. These strategies include mass media campaigns, implementing complete smoking bans in public places, voluntary smoke-free rules in private places such as homes and cars, increasing the price of tobacco products, and providing coordination of cessation services for smokers. Given the changing age structure of the U.S. population, additional interventions that focus on middle-aged and older adults are needed. For example, cessation interventions may focus on workplace and healthcare settings, which can target middle-aged adults who are established smokers (DiFranza et al., 2004; Fiore, 2008). Finally, our findings also underscore that across age, the relationship between smoking and HRQoL requires a closer examination of both mental and physical health so that appropriate and coordinated cessation interventions may be developed to address overall health and well-being.
Footnotes
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. We have no financial conflicts of interest to report.
REFERENCES
- Barile JP, Reeve BB, Smith AW, Zack MM, Mitchell SA, Kobau R,… Thompson WW. (2013). Monitoring population health for Healthy People 2020: Evaluation of the NIH PROMIS® Global Health, CDC Healthy Days, and satisfaction with life instruments. Quality of Life Research, 22,1201–1211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2014). Best practices for comprehensive tobacco control programs—2014. Retrieved from https://www.cdc.gov/tobacco/stateandcommunity/best_practices/index.htm
- Centers for Disease Control and Prevention. (2003). Public health surveillance for behavioral risk factors in a changing environment. Recommendations from the Behavioral Risk Factor Surveillance Team. Morbidity Mortality Weekly Report, 52(RR-9), 1–12. [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2010). How tobacco smoke causes disease:The biology and behavioral basis for smoking-attributable disease: A report of the Surgeon General. Atlanta: Author; Retrieved from https://www.ncbi.nlm.nih.gov/books/NBK53017/ [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2013). Vital signs: Current cigarette smoking among adults aged ≥18 years with mental illness—United States, 2009–2011. Morbidity & Mortality Weekly Report, 62,81–87. [PMC free article] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2016). Current cigarette smoking among adults—United States, 2005–2015. Morbidity and Mortality Weekly Report, 65(44), 1205–1211. [DOI] [PubMed] [Google Scholar]
- Clarke TC, Norris T, & Schiller JS (2017). Early release of selected estimates based on data from 2016 National Health Interview Survey. National Center for Health Statistics; Retrieved from https://www.cdc.gov/nchs/nhis.htm [Google Scholar]
- DiFranza JR, Savageau JA, Rigotti NA, Ockene JK, McNeill AD, Coleman M, & Wood C (2004). Trait anxiety and nicotine dependence in adolescents—A report from the DANDY study. Addictive Behaviors, 29, 911–919. [DOI] [PubMed] [Google Scholar]
- Dube SR, Caraballo RS, Dhingra SS, Pearson WS, McClave AK, Strine TW, & Berry JT (2009). The relationship between smoking status and serious psychological distress: Findings from the 2007 Behavioral Risk Factor Surveillance System. International Journal of Public Health, 54(Suppl 1), 68–74. [DOI] [PubMed] [Google Scholar]
- Dube SR, Thompson WW, Homa DM, & Zack MM (2013). Health-related quality of life and smoking among adolescents: The National Health and Nutrition Examination Survey, 2001–2008. Nicotine and Tobacco Research, 15, 492–500. 10.1093/ntr/nts163 [DOI] [PubMed] [Google Scholar]
- Dube SR, & Wu LT (2015). Cigarette smoking among adults aged 45 and older in the 16 United States, 2002–2011 In Crome IB, Wu L, Rao R, &Crome P (Eds.), Substance use and older people. Chichester: Wiley-Blackwell. [Google Scholar]
- Fiore MC et al. (2008). Treating tobacco use and dependence: 2008 update Clinical practice guideline. Rockville, MD: US Department of Health and Human Services, Public Health Service; Retrieved from https://www.surgeongeneral.gov/tobacco/index.html [Google Scholar]
- Hall SM, & Prochaska JJ (2009). Treatment of smokers with co-occurring disorders: Emphasis on integration in mental health and addiction treatment settings. Annual Review of Clinical Psychology, 5,409–431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hays RD, Bjorner JB, Revicki DA, Spritzer KL, & Cella D (2009). Development of physical and mental health summary scores from the Patient-Reported Outcomes Measurement Information System (PROMIS) global items. Quality of Life Research, 18,873–880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Heikkinen H, Jallinoja P, Saarni SI, & Patja K (2008). The impact of smoking on health-related and overall quality of life: A general population survey in Finland. Nicotine & Tobacco Research, 10, 1199–1207. 10.1080/14622200802163142 [DOI] [PubMed] [Google Scholar]
- Hu SS, Balluz L, Battaglia MP, & Frankel MR (2011). Improving public health surveillance using a dual-frame survey of landline and cell phone numbers. American Journal of Epidemiology, 173,703–711. 10.1093/aje/kwq442 [DOI] [PubMed] [Google Scholar]
- McClave A, Davis S, McKnight L, & Dube SR (2010). Smoking characteristics of adults with selected lifetime mental illnesses: Results from the 2007 National Health Interview Survey. American Journal of Public Health, 100,2464–2472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- McClave A, Dube SR, Strine TW, & Mokdad AH (2009). The health-related quality of life of smokers, smokers attempting to quit, and non-smokers. Preventive Medicine, 48,173–179. [DOI] [PubMed] [Google Scholar]
- McKeown RE (2009). The epidemiologic transition: Changing patterns of mortality and population dynamics. American Journal of Lifestyle Medicine, 3(1 Suppl), 19S–26S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nelson DE, Holtzman D, Bolen J, Stanwyck CA, & Mack KA (2001). Reliability and validity of measures from the Behavioral Risk Factor Surveillance System (BRFSS). International Journal of Public Health, 46,1–42. [PubMed] [Google Scholar]
- Nelson DE, Powell-Griner E, Town M, & Kovar MG (2003). A comparison of national estimates from the National Health Interview Survey and the Behavioral Risk Factor Surveillance System. American Journal of Public Health, 93,1335–1341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Omran AR (1971). The epidemiologic transition. A theory of the epidemiology of population change. Milbank Memorial Fund, 49(4), 509–538. [PubMed] [Google Scholar]
- Trosclair A, & Dube SR (2010). Smoking among adults reporting lifetime depression, anxiety, anxiety with depression, and major depressive episode, United States, 2005–2006. Addictive Behaviors, 35,438–443. [DOI] [PubMed] [Google Scholar]
- U.S. Census Bureau, Population Division. (2017). Annual estimates of the resident population for selected age groups by sex for the United States, states, counties and Puerto Rico Commonwealth and Municipios: April 1, 2010 to July 1, 2016; June 2017 Retrieved from https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk
- U.S. Department of Health and Human Services. (2014). The health consequences of smoking-50 years of progress: A report of the Surgeon General. Atlanta: Author. [Google Scholar]