Abstract
Previous work suggests that, compared to white adults, black adults have lower perceived risk for smoking-related diseases (SRDs), which may influence cessation behavior and health outcomes; however, racial differences in SRD risk perceptions among high-risk patients (i.e., i.e., a group that exhibits elevated risk for SRDs) following lung screening remain unknown. This paper thus examined differences in risk perceptions for lung cancer and other SRDs among black and white National Lung Screening Trial (NLST) participants. We administered a 10-item measure of perceived lifetime risk of lung cancer and other SRD (Smoking Risk Perceptions Scale; SRPS) to NLST participants at one year following lung screening to 1) establish the internal consistency of the SRPS for both black and white participants, 2) compare smoking-related disease risk perceptions between black and white participants, and 3) identify predictors of risk perceptions for black and white participants using multivariable linear regression models. We determined the SRPS items loaded onto two factors (personal and comparative risk; Cronbach’s alpha=0.93 and 0.95 for 1743 white and 194 black participants, respectively), thus demonstrating high internal consistency for both black and white adults. Compared to white participants, black adults demonstrated lower SRD risk perceptions (SRPS range=10–50, mean difference=2.55, SE=0.50, p<0.001), even after adjusting for smoking status and sociodemographics. Younger age, female gender, higher education, white race and current smoking status were independently associated with high risk perceptions. Sociodemographic factors associated with lower risk perceptions resemble factors related to continued smoking. Findings suggest current and former black smokers are at risk of having lower risk perceptions for lung cancer and SRDs than white adults following lung cancer screening; these differences may explain observed racial differences in cessation outcomes. Although similar factors influence black and white adults’ beliefs, risk perceptions may differentially impact smoking behavior among these groups. Behavior change models that guide tobacco treatment approaches, particularly for high-risk black smokers, should consider the influence of cultural factors on risk perceptions and cessation efforts.
Keywords: smoking cessation, racial disparities, lung cancer, screening
Lung cancer has been well studied in the context of exploring racial disparities in cancer care [1–6]. Although we have seen overall improvement in survival rates, lung cancer remains the leading cause of cancer-related deaths across all races. Blacks, however, carry a disproportionate burden of disease, demonstrating a higher incidence and mortality rate compared to their white counterparts [7]. While several factors may contribute to these inequities, including poor access to health care services, advanced staging at diagnosis, and patient-provider miscommunication [8], continued smoking has been posited to drive blacks’ risk for cancer and other smoking-related diseases.
The majority (87%) of lung cancer deaths can be attributed to smoking [9]. While smoking rates among black adults are comparable to those reported for whites, there are important racial differences in smoking cessation outcomes and risk for smoking-related disease that render blacks a high-risk population (i.e., a group that exhibits elevated risk for smoking-related disease). Specifically, compared with white smokers, black smokers are less likely to receive quit advice, initiate and comply with tobacco treatment, utilize pharmacological agents to aid cessation efforts, and maintain abstinence after quitting [10–16]. Racial differences in risk beliefs, particularly perceived risk for smoking-related disease, may partially explain these findings. However, few studies have examined these issues among high-risk current and former smokers. Previous work suggests that black adults consider their risks of smoking-related illness to be low. One small study with 144 smokers found that black smokers were more likely than whites to perceive their risk for lung cancer to be low relative to their non-smoking peers; additionally, they were less likely to believe that quitting would reduce their risk [17]. Relatedly, data from the 2005 Health Information National Trends (HINTS) survey of 1872 adults revealed that, in a community sample of smokers and nonsmokers, both black and white participants did not perceive their risk for lung cancer to be high; however, black participants were less likely to attribute lifestyle behavior to cancer risk [18]. These findings highlight the potential link between black adults’ risk beliefs and their continued smoking behavior.
The advent of lung screening by low-dose computed tomography (LDCT) offers current and former smokers at high risk for SRD the opportunity to reduce their risk of lung cancer mortality by detecting disease at an early stage [19]; as such, the United States Preventive Services Task Force issued a grade B recommendation for annual lung cancer screening for adults aged 55 to 80 who have a 30 pack-year smoking history and are currently smoking or have quit sometime in the past 15 years [20]. Lung screening offers the opportunity to introduce smoking cessation treatment to high-risk patients during a “teachable moment,” [21] since patients may be more motivated and receptive to cessation support; yet, it is possible that racial differences in perceived risk for lung cancer and other SRD may influence lung cancer screening decision-making and cessation behavior. To date, little is known about how high-risk black and white patients differ in their perceived risk for cancer and SRD following lung screening. Work led by Park and colleagues conducted with 572 National Lung Screening Trial (NLST) participants at study enrollment suggests that black patients have lower perceived risk for lung cancer and other SRD than their white counterparts [22–24]. The current study proposes to extend these findings by establishing the internal consistency, and thus reliability, of the Smoking Risk Perceptions Scale for both black and white NLST participants and using it to examine potential racial differences in perceived risk for lung cancer and other SRD among a larger cohort of high-risk black and white NLST participants at one-year following lung screening. Establishing the reliability of this scale is critical given evidence suggesting racial differences may exist in the consistency of smoking-related information self-reported in national surveys; specifically, factors such as response bias, which can vary by racial/ethnic background, may impact data quality [25–27]. Further, this study examines factors associated with risk perceptions among black and white NLST participants.
Methods
Study design
The current report is a cross-sectional analysis of a subset of black and white, former and current smokers enrolled in the NLST who were one-year post-lung cancer screening. The NLST is a collaborative effort between the American College of Radiology Imaging Network (ACRIN) and the National Cancer Institute Lung Screening Study. Patients were eligible for recruitment into the NLST if they were 55–74 years old, were either current or former (quit within 15 years) smokers with a minimum history of 30-pack years, and had no prior history of lung cancer. These criteria are in line with the US Preventive Service Task Force’s recommendations regarding screening for high-risk patients. Complete details on NLST design, recruitment and data collection procedures are described elsewhere [28].
Cohort and data collection
In brief, permission was obtained from the ACRIN/NLST executive committee to administer the smoking risk perceptions scale as a sub-study within the ACRIN arm of the trial. A total of eight ACRIN/NLST sites participated in the smoking risk perceptions sub-study. From 12/03–2/04, each of the participants at the 8 sites were offered the opportunity to complete the risk perception sub-study questionnaire at one year post-lung screening. We restricted our analyses to the sites whose surveyed participants were at least 4% black in efforts to make our sample of white and black participants more comparable. Four sites met this criterion: The Cancer Institute of New Jersey, New Brunswick, New Jersey; St. Elizabeth Health Center, Youngstown, Ohio; Jewish Hospital Heart and Lung Institute, Louisville, Kentucky; and Johns Hopkins University, Baltimore, Maryland.
Sociodemographic, medical and smoking history variables
Participants completed a questionnaire upon NLST enrollment that assessed a number of sociodemographic (e.g., age, gender, education, marital status), medical history (familial history of lung cancer, personal history of cancer or SRDs), and smoking history factors (e.g., smoking status, number of years smoked, number of years quit, and nicotine dependence utilizing the Fagerstrom Test for Nicotine Dependence; FTND) [23,28]. Lung screening results were categorized hierarchically as 1) positive, defined as showing a nodule or other abnormality suspicious for lung cancer, 2) positive for significant abnormalities not suspicious for lung cancer, or 3) negative, defined as having no major abnormalities.
Smoking status at one year post-lung screening
Information on smoking status was collected at one year post-lung screening by self-report with one question: “Do you now smoke cigarettes [one or more cigarettes per week]?”
Smoking-Related Disease Risk Perceptions and Behavior Change Constructs
Details of the 25-item questionnaire have been previously published [23]. This questionnaire is comprised of ten risk perception items and fifteen items that tapped into theoretically-based cognitive and emotional behavior change constructs.
Smoking Risk Perceptions Scale (SRPS).
This is a 10-item self-report scale that measures an individual’s perceived personal (individual risk) and comparative lifetime risk (risk related to others) for lung cancer and other SRDs. For perceived personal risk, four questions assessed the likelihood (‘very unlikely’ to ‘very likely’ on a 5-point Likert scale) and danger (‘strongly disagree’ to ‘strongly agree’ on a 5-point Likert scale) of developing lung cancer or SRD. Given the influence of different comparison groups on perceived risk [23], perceived comparative risk was assessed with six questions rated on a 5-point Likert scale using the referent groups average person, others of the same age and sex, and other former/current smokers. Specifically, participants were asked if they were in danger of developing lung cancer and other SRDs, compared to the average person (strongly agree to strongly disagree). Participants were also asked about their chances of developing lung cancer and a SRD, compared to others of the same age and sex and compared to other former/current smokers (much lower to much higher). The 10 items of the smoking risk perception scale were summed to create a composite score, with higher scores indicating higher perceived risk for lung cancer and other SRDs [range=10 (low) to 50 (high)].
Cognitive and emotional constructs of smoking risk.
Fifteen items capture theoretically-based cognitive and emotional behavior change constructs given their potential influence on perceived risk. Self-efficacy was measured with one question about confidence to quit smoking/remain quit (5-point Likert scale, ‘not at all’ to ‘extremely’). Perceived benefits of quitting smoking/staying quit was measured with 3 questions about the benefits of quitting in terms of decreasing risk for lung cancer, other SRDs, and increasing life expectancy (4-point Likert scale, ‘not at all’ to ‘very much’); these items were combined to create a composite score (range = 3–12; alpha=0.89) [23]. Perceived benefits of lung screening were assessed by five questions about the curability and benefits of lung cancer detected by screening (4-point Likert scale ‘very few’ to ‘most,’ range 1–4; and 5-point Likert scale ‘not at all’ to ‘extremely’). Perceived severity was assessed with four questions about the health consequences and severity of lung cancer and other SRDs (5-point Likert scale, ‘not at all’ to ‘extremely’); these items were combined to create a composite score (range 4–20; alpha=0.85) [23]. Worry about lung cancer and other SRDs was assessed with 4 questions about intensity (4-point Likert scale, ‘not at all’ to ‘extremely’) and frequency of worry (4-point Likert scale, ‘not at all’ to ‘all of the time’); items were combined to create a composite score (range = 4–16; alpha=0.89) [23]. Knowledge of smoking risks was assessed with three questions which asked participants to estimate the percentage of smokers who would get lung cancer (“Among 100 smokers, how many will get lung cancer because they smoke?”) and the average number of years of life lost due to smoking (“On average, smokers die nearly ___ years earlier than nonsmokers?”). In addition, participants were asked to quantify a smoker’s risk of developing lung cancer (“A smoker who smokes one pack of cigarettes a day is at how many times risk of developing lung cancer, compared to a non-smoker?”).
Statistical Analyses
All statistical analyses were conducted using SAS software, Version 9.4 of the SAS System for Linux. Copyright © [2002–2012] SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc., Cary, NC, USA. Baseline differences in sociodemographic, clinical, and smoking-related variables by race were assessed using Chi Square test/Fisher’s Exact tests for categorical variables, and two sample t-tests/Wilcoxon rank-sum tests for continuous variables. Cronbach’s alpha was used to assess the internal consistency of the total smoking risk perception scale and the subscales, separately for black and white participants. Linear regression models were used to examine the effect of race on risk perceptions, for lung cancer and SRDs at 12 months; unadjusted and adjusted analyses (adjusting for common confounders such as age, gender, income, education, and controlling for smoking status) as well as cognitive constructs (worry, anxiety, perceived benefits of screening/quitting) were conducted. Interactions between race and potential confounders were examined in the combined model. Interactions with p-values (two-sided) less than 0.05 were considered statistically significant. Models were also fit separately for white and black participants to calculate the significance level of the confounders in the subsets.
Results
Participant Characteristics
There were 1743 white participants and 194 black participants who completed the risk perception questionnaire at one of the four study sites one-year following lung screening (Table 1). White participants were more likely to be older (mean age=61.4 vs. 59.1; p<.0001), male (59.2% vs. 53.1%; p<.001) and married (65.4% vs. 29.4%; p>.0001). They were more likely to have completed more years of education (62.9% vs. 44.9%; p<.0001) and to have a higher household income. White participants were also more likely to have a history of emphysema (9.6% vs. 3.1%; p=.003), while black participants were more likely to have a history of asthma (12.4% vs. 7.4%, p=.01). There were no differences in lung screening results for white and black participants; the majority of the sample received a negative screening result (82.3% vs. 81.4%; p=.56).
Table 1.
Sociodemographic, smoking, and medical characteristics of participants
| White (N=1743) | Black (N=194) | p-value | |
|---|---|---|---|
| Sociodemographic characteristics | |||
| Age, mean (SD) | 61.43 (4.93) | 59.14 (4.05) | <.0001 |
| Gender, N (%) | <.0001 | ||
| Male | 1032 (59.21) | 91 (46.91) | |
| Female | 711 (40.79) | 103 (53.09) | |
| Education, N (%) | <.0001 | ||
| High School or less | 613 (35.17) | 102 (52.58) | |
| Post high school training, some college | 595 (34.14) | 55 (28.35) | |
| > College degree or higher | 500 (28.69) | 32 (16.49) | |
| Other/Unknown | 35 (2.01) | 5 (2.58) | |
| Household Income, N (%) | <.0001 | ||
| < $15,000 | 169 (9.70) | 63 (32.47) | |
| $15,000–34,999 | 406 (23.29) | 52 (26.80) | |
| 35,000–64,999 | 493 (28.28) | 28 (14.43) | |
| $65,000 and above | 434 (24.90) | 20 (10.31) | |
| Unknown | 241 (13.83) | 31 (15.98) | |
| Marital Status, N (%) | <.0001 | ||
| Married/Living as Married | 1140 (65.40) | 57 (29.38) | |
| Never Married/Widowed/Divorced/Separated | 599 (34.37) | 134 (69.07) | |
| Unknown | 4 (0.23) | 3 (1.55) | |
| Center, N (%) | <.0001 | ||
| Jewish Hospital Louisville, KY | 985 (56.51) | 33 (17.01) | |
| Johns Hopkins, Baltimore, MD | 649 (37.23) | 156 (80.41) | |
| Cancer Inst of NJ, New Brunswick, NJ | 64 (3.67) | 3 (1.55) | |
| St. Elizabeth, Youngstown, OH | 45 (2.58) | 2 (1.03) | |
| Insurance Status, N (%) | <.0001 | ||
| Private Insurance | 1116 (64.03) | 96 (49.48) | |
| Medicare | 164 (9.41) | 31 (15.98) | |
| Medicare and Private Insurance | 287 (16.47) | 19 (9.79) | |
| Medicaid | 6 (0.34) | 1 (0.52) | |
| Medicare and Medicaid | 11 (0.63) | 4 (2.06) | |
| Military or Veteran Administration | 52 (2.98) | 9 (4.64) | |
| Self Pay | 37 (2.12) | 3 (1.55) | |
| No Means of payment | 53 (3.04) | 29 (14.95) | |
| Other | 5 (0.29) | 0 (0.00) | |
| Unknown/Decline to answer | 12 (0.69) | 2 (1.03) | |
| Smoking Characteristics | |||
| Smoking Status at One Year, N (%) | <.001 | ||
| Former smoker | 901 (51.69) | 60 (30.93) | |
| Current smoker | 821 (47.10) | 132 (68.04) | |
| Not reported | 21 (1.20) | 2 (1.03) | |
| Smoking Characteristics, Mean (SD) | |||
| Pack Years | 59.10 (25.84) | 47.82 (19.94) | <.0001 |
| # Years Quit (Count current smokers as 0) | 3.45 (4.92) | 1.50 (3.50) | <.0001 |
| Fagerstrom test for nicotine dependence | 5.64 (2.32) | 5.22 (2.30) | 0.02 |
| Medical Characteristics | |||
| Personal History of, N (%) | |||
| Asthma | 129 (7.40) | 24 (12.37) | 0.01 |
| Chronic Bronchitis | 230 (13.20) | 22 (11.34) | 0.47 |
| Emphysema | 168 (9.64) | 6 (3.09) | 0.003 |
| Heart Disease/Heart Attack | 263 (15.09) | 24 (12.37) | 0.31 |
| Stroke | 52 (2.98) | 7 (3.61) | 0.63 |
| Cancer | 177 (10.15) | 13 (6.70) | 0.13 |
| SRD | 493 (28.28) | 39 (20.10) | 0.02 |
| Family history of lung cancer, N (%) | 425 (24.38) | 45 (23.20) | 0.71 |
| Baseline screening result, N (%) | 0.56 | ||
| Negative | 1435 (82.32) | 158 (81.44) | |
| Significant for non-lung cancer abnormalities | 113 (6.48) | 10 (5.15) | |
| Positive | 191 (10.96) | 26 (13.40) | |
| Unknown | 5 (0.29) | 0 (0.00) | |
Smoking Risk Perceptions Scale (SRPS)
The SRPS demonstrated high internal consistency. Cronbach’s alpha for the total 10-item questionnaire was 0.93 and 0.95 for white and black participants, respectively. For each, the ten risk items loaded onto two factors. Among black participants, Cronbach’s alpha was .87 for personal and .95 for comparative risk; among white participants, Cronbach’s alpha was .84 and .92 for personal and comparative risk, respectively. Compared to white patients (mean=35.52, SD=7.50), black patients had lower risk perceptions (mean=32.36, SD=9.32; p<.0001; Table 2). This difference remained significant when controlling for important demographic, smoking behavior, and cognitive and emotional determinants of smoking (see Table 3).
Table 2.
Smoking Risk Perceptions and Cognitive-Emotional Constructs
| White (N=1743) |
Black (N=194) |
P Value | |
|---|---|---|---|
| Risk Perceptions, mean (SD) | |||
| Risk perceptions for lung cancer and SRDs | 35.52 (7.50) | 32.36 (9.32) | <.0001 |
| Cognitive and emotional constructs, Mean (SD) | |||
| Perceived benefits of screening (White=1700, Black=187) | 4.86 (1.38) | 4.84 (1.47) | 0.94 |
| Perceived severity of lung cancer and SRDs (White=1648, Black=177) | 18.24 (1.90) | 18.12 (2.60) | 0.32 |
| Worry about lung cancer and SRDs (White=1667, Black=186) | 9.55 (2.80) | 9.76 (3.35) | 0.30 |
| Perceived benefits of quitting (White=1702, Black=188) | 9.47 (2.10) | 9.83 (2.07) | 0.02 |
| Confidence/Self-Efficacy to quit (White=1658, Black=181) | 3.69 (1.35) | 3.54 (1.4) | 0.005 |
| Knowledge of smoking risks | |||
| % smokers who will get lung cancer, mean (SD) (White=1680, Black =182) | 40.25 (22.66) | 46.54 (22.85) | 0.0005 |
| Average years decreased life for smokers, N (%) | <.0001 | ||
| 0–5 | 426 (24.44) | 49 (25.26) | |
| 6–10 | 813 (46.64) | 62 (31.96) | |
| 11+ | 150 (8.61) | 34 (17.53) | |
| Not Answered | 354 (20.31) | 49 (25.26) | |
| One pack/day smoker’s risk of developing lung cancer, N (%) | 0.01 | ||
| 0–2× risk | 483 (27.71) | 55 (28.35) | |
| 5× risk | 641 (36.78) | 52 (26.80) | |
| 10–20× risk | 588 (33.73) | 80 (41.24) | |
| Not Answered | 31 (1.78) | 7(3.61) | |
Table 3.
Multivariate Linear Regression model: factors associated with risk perceptions of lung cancer and smoking-related diseases (SRDs) at one-year among NLST participants (N=1651)
| B | SE | P-value | |
|---|---|---|---|
| Age (years) | −0.08 | 0.03 | 0.01 |
| Female | 0.39 | 0.30 | 0.22 |
| Spiral CT (ref. x-ray) | 0.09 | 0.28 | 0.75 |
| Education (ref. <High School) | |||
| Some college | 0.84 | 0.33 | 0.01 |
| College degree or more | 2.43 | 0.36 | <.0001 |
| Married | −0.05 | 0.31 | 0.88 |
| Black (ref. white) | −2.55 | 0.50 | <.0001 |
| Current Smoker at 1 year (ref. Former) | 1.40 | 0.31 | <.0001 |
| Total Pack Years | 0.01 | 0.01 | 0.24 |
| FTND score | 0.30 | 0.07 | <.0001 |
| Cognitive and Emotional Constructs | |||
| Perceived benefit of screening | 0.03 | 0.10 | 0.78 |
| Perceived benefit of quitting | −0.49 | 0.07 | <.0001 |
| Perceived severity of lung cancer and SRDs | 0.19 | 0.08 | 0.01 |
| Worry About Lung Cancer and SRDs | 1.53 | 0.05 | <.0001 |
Cognitive and Emotional Constructs of Perceived Risk
There were no significant differences between white and black participants in terms of lung cancer and SRD worry, perceived benefits of screening and of quitting, or perceived severity of having a diagnosis of lung cancer or an SRD. Knowledge of smoking risks was also similar across white and black participants. Both white and black participants overestimated the risk of lung cancer among smokers; on average, participants estimated that a little over 40% of smokers would get lung cancer (Table 2). However, white participants were more likely to underestimate the risks of smoking on mortality, with 9% of white and 18% of black participants accurately reporting that smokers die more than 10 years earlier than nonsmokers (p<.0001). A little over one third of black and white participants also accurately identified the tenfold risk for lung cancer that exists for smokers (p=.01).
Multivariate predictors of risk perception at one year
We built a linear regression model to determine if smoking status, race, age, gender, education level, marital status (binary), and screening arm could be used to predict risk perception among NLST participants (Table 3). Younger age (p=.01), white race (p<.0001), and fewer perceived benefits of quitting (p<.0001) were all significant predictors of higher smoking-related disease risk perception at one year. Additionally, higher education (some college compared to < high school, p=.01 and college degree or more compared to <high school, p<.0001), current smoking status (p<.0001), smoking addiction (p<.0001), greater perceived severity of diagnosis (p=.01), and more cancer worry (p<.0001) were also significant predictors of greater risk perceptions.
We built separate linear regression models for black and white participants to examine potential differences in predictors of risk perceptions. Factors independently associated with higher risk perceptions among white participants included younger age (β=−.08, SE=.03, p=.01), higher education (some college vs. < high school; β=.96, SE=.35, p=.005; college degree or more vs. <high school; β=2.40, SE=.37, p<.0001), current smoking status (β=1.32, SE=.32, p<.0001), more nicotine dependence ( β=.28, SE=.07, p=.0002), few perceived benefits of quitting (β=−.45, SE=.08, p<.0001), and greater worry about lung cancer and smoking-related disease (β=1.51, SE=.06, p<.0001). Factors independently associated with higher risk perceptions among black participants included higher education (college degree or more versus <high school; β=3.05, SE=1.40, p=.03), current smoking status (β=2.42, SE=1.17, p=.04), nicotine dependence (β=.61, SE=.26, p=.02), fewer perceived benefits of quitting (β=−.88, SE=.27, p=.0009) and greater worry (β=1.56, SE=.17, p<.0001) were significant predictors of higher risk perceptions.
Discussion
The purpose of this study was to establish the internal consistency of a smoking risk perception scale (SRPS) for high-risk black and white patients undergoing lung screening to help examine questions related to potential differences in and predictors of risk perceptions.
We found that the SRPS is a helpful tool that can be used to reliably assess perceived smoking-related risk for lung cancer and other SRD among high-risk black and white adults. The internal consistency for the total scale and for the two scale factors was excellent and comparable with that reported in the original validation of this scale [23], suggesting the scale effectively assesses black and white smoker’s perceptions of their personal and comparative risk for lung and SRDs. Using this risk perception scale, we confirmed existing reports documenting significant racial differences in beliefs about the harms of smoking [23;29–31]. Specifically, black adults held lower risk perceptions for smoking-related disease relative to white adults, even after adjusting for sociodemographic factors, smoking status and other important cognitive and emotional determinants of smoking. Importantly, our study highlights the strong and independent influence that race has on smoking risk beliefs, further underscoring the need to support cessation efforts using more targeted approaches.
Much of the existing research attributes low smoking risk beliefs among smokers to the presence of an “optimistic bias,” which is the tendency for individuals to underestimate their own personal health risks of smoking. However, several patterns emerged in our data that run counter to these theories. First, compared to whites, our sample of black adults was more likely to demonstrate accurate knowledge regarding the risks of being diagnosed with and dying from a smoking-related disease. Although knowledge items were not examined in our regression model due to data missingness across both black and white participants, lacking knowledge about the hazards of smoking has been associated with unrealistic optimism [30]. Moreover, black participants also endorsed lower confidence in their ability to quit smoking, which conflicts with work suggesting a potential link between lower quit self-efficacy and higher perceived risk [24,31]. Additionally, although age emerged as a factor associated with risk perceptions in our white sample, we did not see the same relationship with age amongst black participants. This is important, as older age has historically been linked to increased likelihood of having an “optimistic” outlook [31]. Together, our results suggest that factors beyond misplaced optimism may underlie black participants’ smoking risk beliefs. It could be that among current and former black smokers, emotional factors may have a stronger impact on perceptions of smoking risk.
Of note, our analyses found little to no differences in our models for black and white patients with respect to predictors of smoking risk perceptions. These findings question the applicability of traditional risk belief models for current and former black smokers and support the need to account for other variables when considering drivers of risk beliefs for this vulnerable group. For instance, culturally-salient factors such as wishful thinking, fatalistic beliefs, and medical mistrust have been posited to contribute to lower cancer risk perceptions in black communities [18, 30, 32]. Cultural models, such as the PEN-3, offer a useful framework for understanding how these and other variables may come together to influence risk beliefs [33–36]. Likewise, misconceptions about the hazards of light or nondaily smoking may also influence risk attributions [37]. Specifically, our previous NLST findings show that black participants averaged fewer daily cigarettes relative to white patients [38]. These smoking patterns may lead current and former black smokers to consider their risk for smoking-related disease to be lower. Future studies may benefit from identifying these smoking misconceptions and examining the role they may have on shaping risk perceptions in this population.
Risk beliefs are a fundamental part of many health behavior models. Some of the most widely-used theoretical frameworks, such as the Health Behavior Model [39] and the Precaution Adoption Model [40], maintain that individuals must judge their personal risk for illness to be high for behavior change to occur. However, given the pattern in our findings, we believe that important racial differences exist in the risk-behavior connection. Specifically, despite having comparably lower risk beliefs, evidence from our earlier work in this sample indicates black smokers have high intentions to quit and are more likely than white smokers to make 24-hour and 7-day quit attempts; however, they appear to have difficulty maintaining abstinence [38]. These patterns suggest that smoking risk beliefs may differentially affect smoking behavior for black smokers, and these subtle differences may not be adequately captured by existing health behavior models [32]. Moreover, we found additional differences in several determinants of behavior change that further question the validity of these models for black patients. White and black patients alike believed the consequences of smoking are severe, and they shared modest levels of worry about lung cancer and smoking-related diseases. Yet, black patients reported lower confidence to quit. These beliefs did not appear to interfere with their efforts to quit, but instead they may have interfered with black patients’ ability to remain abstinent. In truth, Orom et al noted that in situations wherein risk for a health threat is seen as high and control as low, vulnerable groups may engage a set of attributions to manage their fear [32]. One implication of this is that efforts to improve cessation rates at the time of lung screening may necessitate interventions that de-emphasize risk (and subsequently, potentially illness worry) and instead focus on improving smokers’ confidence and skills to convert quit attempts to sustained abstinence.
The information obtained in this study is helpful in light of the ongoing discussion on lung cancer screening in the US; however, there are important limitations worth considering. As we alluded to earlier, this sample consists of a select group of current and former smokers who have sufficient motivation and means with which to pursue lung screening. Risk perceptions and associated smoking cessation factors may be variable among a more general, under-resourced population of current and former black smokers. Our findings may thus not generalize to other black adults, particularly those who may avoid screening, were not eligible for screening or who may have a more prominent personal or familial history of smoking-related disease. Additionally, our measure for current smoking status (i.e., do you now smoke cigarettes [one or more cigarettes per week]?”) differs from measures used in some population-based observational studies (i.e., based on two items: ever smoke 100 cigarettes [former smoker] and current smoking “do you now smoke cigarettes”). This is because the NLST participants were either heavy former or current smokers, so a former smoker was categorized by a minimum of 15 years of smoking, not greater than 100 cigarettes. Although this may contribute to slight differences in smoking rates when compared to other studies [41–43], categorizing adults as smokers based on any cigarette use in the past week is per the recommendation of the Society for Research on Nicotine and Tobacco (SRNT), is used nationally in randomized clinical trials, and is consistent with the national tobacco organization’s (SRNT) characterization of current smoker. Importantly, we were unable to capture important knowledge items, such as awareness of the harms of light smoking, that may clarify the potential misconceptions current and former black smokers may have. Further, our small sample size and the small rates of having a positive screening result in our sample precluded our ability to compare and contrast differences in predictors of risk perceptions among this higher-risk group. Relatedly, given the differences in sample size for our black and white patients, certain variables that are statistically significant in our linear model for white participants may not be significant in our linear model for black patients; however, the point estimates could be comparable. With that said, the numbers of black participants included in this subset of NLST are reflective proportions of black participants in the overall NLST and in the US in general. Lastly, data missingness for our knowledge scale, which was comparable across race, precluded our ability to examine the impact of knowledge on smoking-related disease risk perceptions. It would be important to understand if and how this information would impact perceived risk for smoking-related disease.
Smoking cessation is an important part of any discussion during lung cancer screening, and acknowledging that there are differences in risk perceptions, which could engender disparities in smoking outcomes, might allow for more focused cessation efforts. The results of our study support previous observations by Park and colleagues [10, 22–24] and Lathan et al. [3, 4, 18] which document black smokers as having less perceived risk for lung cancer and other smoking-related illnesses. Yet, black smokers remain highly motivated to quit and are more likely to make quit efforts, more often exhibiting struggles remaining abstinent [38]. Existing risk belief and behavior change models are missing key cultural variables that may explain these conflicting patterns and the modest connection between risk, intentions and behavior. Elevating risk perceptions has the potential to increase worry and learned helplessness among current and former black smokers, diminishing their capacity to quit or stay quit. Interventions delivered during the time of lung screening should optimize black adults’ interest and willingness to quit by framing abstinence information in a way that empowers black individuals to sustain quits. Moreover, to dispel lingering misconceptions, efforts may also be better placed on developing their understanding of how and why quitting can contribute to health gains regardless of the number of cigarettes consumed. Programs that focus on providing cessation medications to counter nicotine dependence in addition to culturally-tailored information and skills to maximize their confidence and ability to convert quit attempts to sustained quits may be useful in achieving equitable cessation outcomes among this vulnerable group.
Table 4.
Multivariate Linear Regression Model: factors associated with risk perceptions of lung cancer and smoking-related diseases (SRDs) at one-year among white participants (N=1495)
| B | SE | P-value | |
|---|---|---|---|
| Age (years) | −0.08 | 0.03 | 0.01 |
| Female | 0.49 | 0.31 | 0.12 |
| Spiral CT (ref. x-ray) | 0.09 | 0.29 | 0.76 |
| Education (ref. <High School) | |||
| Some college | 0.96 | 0.35 | 0.005 |
| College degree or more | 2.40 | 0.37 | <0.0001 |
| Married | −0.14 | 0.32 | 0.65 |
| Current Smoker at 1 year (ref. Former) | 1.32 | 0.32 | <0.0001 |
| Total Pack Years | 0.01 | 0.01 | 0.09 |
| FTND score | 0.28 | 0.07 | 0.0002 |
| Cognitive and Emotional Constructs | |||
| Perceived benefit of screening | 0.02 | 0.11 | 0.85 |
| Perceived benefit of quitting | −0.45 | 0.08 | <0.0001 |
| Perceived severity of lung cancer and SRDs | 0.15 | 0.08 | 0.08 |
| Worry About Lung Cancer and SRDs | 1.51 | 0.06 | <0.0001 |
Table 5.
Multivariate Linear Regression Model: factors associated with risk perceptions of lung cancer and smoking-related diseases (SRDs) at one-year among black participants (N=156)
| B | SE | P-value | |
|---|---|---|---|
| Age (years) | 0.003 | 0.13 | 0.98 |
| Female | −0.70 | 1.04 | 0.50 |
| Spiral CT (ref. x-ray) | −0.04 | 1.00 | 0.97 |
| Education (ref. <High School) | |||
| Some college | −0.13 | 1.15 | 0.91 |
| College degree or more | 3.05 | 1.40 | 0.03 |
| Married | 0.95 | 1.12 | 0.40 |
| Current Smoker at 1 year (ref. Former) | 2.42 | 1.17 | 0.04 |
| Total Pack Years | −0.03 | 0.03 | 0.25 |
| FTND score | 0.61 | 0.26 | 0.02 |
| Cognitive and Emotional Constructs | |||
| Perceived benefit of screening | 0.02 | 0.37 | 0.95 |
| Perceived benefit of quitting | −0.88 | 0.27 | 0.0009 |
| Perceived severity of lung cancer and SRDs | 0.42 | 0.22 | 0.06 |
| Worry About Lung Cancer and SRDs | 1.56 | 0.17 | <.0001 |
Acknowledgements:
This work was supported by the American Cancer Society (grant number MRSG-005-05-CPPB to E.R.P); The National Cancer Institute at the National Institutes of Health (grant numbers U01 CA079778, U01 CA080098, K07 CA211955); and the United States Department of Veterans Affairs (grant number 1IK2CX000918-01A1)
Footnotes
Ethical approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
For this type of study formal consent is not required.
Conflict of Interest: All authors declare that they have no conflict of interest.
References
- 1.Bach PB, Cramer LD, Warren JL, et al. Racial differences in the treatment of early-stage lung cancer. N Engl J Med 1999; 34(16): 1198–1205. doi: 10.1056/NEJM199910143411606 [DOI] [PubMed] [Google Scholar]
- 2.Bach PB, Cramer LD, Schrag D, et al. The influence of hospital volume on survival after resection for lung cancer. N Eng J Med 2001; 345(3): 181–188. doi: 10.1056/NEJM200107193450306 [DOI] [PubMed] [Google Scholar]
- 3.Lathan CS, Neville BA, Earle CC. The effect of race on invasive staging and surgery in non-small-cell lung cancer. J Clin Oncol 2006; 24(3): 413–418. doi: 10.1200/JCO.2005.02.1758 [DOI] [PubMed] [Google Scholar]
- 4.Lathan CS, Neville BA, Earle CC. Racial composition of hospitals: Effects on surgery for early-stage non-small-cell lung cancer. J Clin Oncol 2008; 26(26): 4347–4352. doi: 10.1200/JCO.2007.15.5291 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Earle C, Venditti LN, Nuemann PJ, et al. Who gets chemotherapy for metastatic lung cancer? Chest, 2000; 117(5): 1239–1246. doi: 10.1378/chest.117.5.1239 [DOI] [PubMed] [Google Scholar]
- 6.Earle CC, Venditti L, Neumann, et al. Impact of referral patterns on the use of chemotherapy for lung cancer. J Clin Oncol 2000; 117(5): 1239–1246. doi: 10.1200/JCO.2002.07.142 [DOI] [Google Scholar]
- 7.Stewart J, Lung carcinoma in African Americans: A review of the current literature. Cancer 2001; 91(12): 2476–2481. doi: [DOI] [PubMed] [Google Scholar]
- 8.Cooper LA, Roter DL, Johnson RL, et al. Patient-centered communication, ratings of care, and concordance of patient and physician race. Ann Intern Med 2003; 139(11): 907–915. doi: 10.7326/0003-4819-139-11-200312020-00009 [DOI] [PubMed] [Google Scholar]
- 9.American Cancer Society. Cancer Facts & Figures 2017 Atlanta: American Cancer Society [Google Scholar]
- 10.Park ER, Japuntich SJ, Traeger L, et al. Disparities between blacks and whites in tobacco and lung cancer treatment. Oncologist 2011; 16(10): 1428–1434. doi: 10.1634%2Ftheoncologist.2011-0114 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.de Dios MA, Anderson BJ, Stanton C, Audet DA, Stein M. (Project Impact: A pharmacotherapy pilot trial investigating the abstinence and treatment adherence of Latino light smokers. J Subst Abuse Treat 2012; 43(3):322–30. doi: 10.1016/j.stat.2012.01.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Landrine H, Corral I, Campbell KM. Racial disparities in healthcare provider advice to quit smoking. Prev Med Rep 2018; 10:172–175. doi: 10.1016/j.pmedr.2018.03.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jones MR, Joshu CE, Navas-Acien A, Platz EA. Racial/Ethnic differences in duration of smoking among former smokers in the National Health and Nutrition Examination Surveys. Nicotine Tob Res 2018; 20(3):303–311. doi: 10.1093/ntr/ntw326 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bacio GA, Guzman IY, Shapiro JR, Ray LA. Differences in quit attempts between non-Hispanic Black and White daily smokers: The role or smoking motives. Addict Behav 2015; 39(12):1769–1772. doi: 10.1016/j.addbeh.2014.07.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Johnson L, Hartz S, Bierut L. Different Trajectories of smoking behaviors across racial groups. Drug and Alcohol Dependence 2017; 171:e96–e97. [Google Scholar]
- 16.Max W, Sung HY, Tucker LY, Stark B. Disproportionate cost of smoking for African Americans in California. Am J Public Health 2010; 100(1):152–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lyna P, McBride C, Samsa G, et al. Exploring the association between perceived risks of smoking and benefits to quitting: Who does not see the link? Addict Behav 2002; 27(2) :293–307. doi: 10.1016/S0306-4603(01)00175-7 [DOI] [PubMed] [Google Scholar]
- 18.Lathan CS, Okechukwu C, Drake B, et al. Racial differences in the perception of lung cancer: The 2005 Health Information National Trends Survey. Cancer 2010; 116(8): 1981–1986. doi: 10.1002/cncr.24923 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Marshall HM, Bowman RV, Yang IA, Fong, et al. Screening for lung cancer with low-dose computed tomography: A review of current status. J Thorac Dis 2013; 5(Suppl 5): S524–S539. doi: 10.3978/j.issn.2072-1439.2013.09.06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Final Update Summary: Lung Cancer: Screening. U.S. Preventive Services Task Force. 2015 July. https://www.uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/lung-cancer-screening.
- 21.Taylor KL, Cox LS, Zincke N, et al. Lung cancer screening as a teachable moment for smoking cessation. Lung Cancer 2007; 56: 125–134. doi: 10.1001/jamainternmed.2015.3558 [DOI] [PubMed] [Google Scholar]
- 22.Park ER, Kleimann S, Youatt EJ, et al. Black and white adults’ perspectives on the genetics of nicotine addiction susceptibility. Addict Behav 2011; 36(7): 769–772. doi: 10.1016/j.addbeh.2011.02.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Park ER, Ostroff JS, Rakowski W, et al. Risk perceptions among participants undergoing lung cancer screening: Baseline results from the National Lung Screening Trial. Ann Behav Med 2009; 37(3): 268–279. doi: 10.1007%2Fs12160-009-9112-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Park ER, Streck JM, Gareen IF, et al. A qualitative study of lung cancer risk perceptions and smoking beliefs among national lung screening trial participants. Nicotine Tob Res 2014; 16(2): 166–173. doi: 10.1093/ntr/ntt133 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Soulakova JN, Hartman AM, Liu B, Willis GB, Augustine S, Reliability of adult self-reported smoking history: data from the tobacco use supplement to the current population survey 2002–2003 cohort. Nicotine Tob Res 2012; 14(18): 952–960. doi: 10.1093/ntr/ntr313 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Soulakova JN, Huang H, Crockett LJ. Racial/ethnic disparities in consistent reporting of smoking-related behaviors. J Addict Behav Ther Rehabil 2015; 4(4). Doi: 10.4172/2324-9005.1000147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ramo DE, Hall SM, Prochaska JJ. Reliability and validity of self-reported smoking in an anonymous online survey with young adults. Health Psychol 2011; 30(6): 693–701. doi: 10.1037/a0023443 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.American College of Radiology Imaging Network. American College of Radiology Imaging Network Protocol 6654 data forms. 2011.
- 29.Lyna P, McBride C, Samsa G, Pollak KI. Exploring the association between perceived risks of smoking and benefits to quitting: Who does not see the link? Addict Behav 2002;27(2):293–307. [DOI] [PubMed] [Google Scholar]
- 30.Weinstein ND, Marcus SE, Moser RP. Smokers’ unrealistic optimism about their risk. Tob Control 2005; 14(1): 55–59. doi: 10.1136/tc.2004.008375 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Borrelli B, Hayes RB, Dunsiger S, et al. Risk perception and smoking behavior in medically ill smokers: A prospective study. Addiction June 2010; 105(6): 1100–1108. doi: 10.1111/j.1360-0443.2010.02900.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Orom H, O’Quin KE, Reilly S, et al. Perceived cancer risk and risk attributions among African-American residents of a low-income, predominantly African-American neighborhood. Ethn Health December 2015; 20(6): 543–556. doi: 10.1080/13557858.2014.950197 [DOI] [PubMed] [Google Scholar]
- 33.Scarinci IC, Bandura L, Hidalgo B, Cherrington A. Development of a theory-based (PEN-3 and Health Belief Model), culturally relevant intervention on cervical cancer prevention among Latina immigrants using intervention mapping. Health Promot Pract 2012; 13(1):29–40. doi: 10.1177/1524839910366416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Iwelunmor J, Newsome V, Airhihenbuwa CO. Framing the impact of culture on health: A systematic review of the PEN-3 Cultural Model and its application in public health research and interventions. Ethn Health 2014; 19(1):20–46. doi: 10.1080/13557858.2013.857768 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Blackstone S, Iwelunmor J, Plange-Rhule J, Gyamfi J, Quakyi NK, Ntim M, Addison A, Ogedegbe G. ‘I believe high blood pressure can kill me:’ Using the PEN-3 Cultural Model to understand patients’ perceptions of an intervention to control hypertension in Ghana. Ethn Health 2017; 4:1–14. doi: 10.80/13557858.2017.1346178 [DOI] [PubMed] [Google Scholar]
- 36.Poss JE. Developing a new model for cross-cultural research: Synthesizing the Health Belief Model and the Theory of Reasoned Action. ANS Adv Nurs Sci 2001; 23(4): 1–15. doi: 10.1097.00012272-200106000-00002 [DOI] [PubMed] [Google Scholar]
- 37.Tong EK, Ong MK, Vittinghoff E et al. Nondaily smokers should be asked and advised to quit. American Journal of Preventive Medicine 2006; 30(1): 23–30. doi: 10.1016/j.amepre.2005.08.048 [DOI] [PubMed] [Google Scholar]
- 38.Kumar P, Gareen IF, Lathan CS, et al. Racial differences in tobacco cessation and treatment usage after lung screening: An examination of the National Lung Screening Trial. Oncologist 2016; 21(1): 40–49. doi: 10.1634/theoncologist.2015-0325 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Rosenstock IM. Historical origins of the health belief model. Health Educ Monogr 1974; 2(4): 328–334. doi: 10.1177/109019817400200403 [DOI] [PubMed] [Google Scholar]
- 40.Weinstein ND. The precaution adoption process. Health Psychol 1988; 7(4): 355–386. doi: 10.1037/0278-6133.7.4.355 [DOI] [PubMed] [Google Scholar]
- 41.Soulakova J, Davis WW, Hartman A, Gibson J The impact of survey and response modes on current smoking prevalence estimates using TUS-CPS: 1992–2003. Surv Res Methods 2009; 3(3): 123–137 [PMC free article] [PubMed] [Google Scholar]
- 42.Lasser K, Boyd JW, Woolhandler S, Himmelstein DU, McCormick D, Bor DH. Smoking and mental illness: a population-based prevalence study. JAMA 2000; 284(20): 2606–2610. doi: 10.1001/jama.284.20.2606 [DOI] [PubMed] [Google Scholar]
- 43.Ryan H, Trosclair A, Gfroerer J. Adult current smoking: differences in definitions and prevalence estimates—NHIS and NSDUH, 2008. J Environ Public Health 2012; 2012. doi: 10.1155/2012/918368 [DOI] [PMC free article] [PubMed] [Google Scholar]