Associations of At-Risk Drinking,Current Smoking,and Their Co-Occurrence With Primary Care Service Utilization

Abstract

Purpose

Smoking and at-risk drinking are each associated with lower primary care utilization, but the influence of their co-occurrence is not known. The current study compared associations of endorsement of one behavior vs endorsement of both with primary care utilization.

Design

Cross-sectional telephone survey.

Setting

All United States and Territories.

Subjects

246 801 adults aged 18–64.

Measures

The outcome was endorsement of attending a past-year primary care visit. Predictor variables included drinking and smoking status examined individually and combined.

Analysis

Multivariable logistic regressions, adjusted for socio-demographics and number of chronic health conditions.

Results

The odds of attending a past-year primary care visit were 24% lower for persons who drank at risk levels compared to the odds of persons who did not drink and 36% lower for persons who smoked vs those who did not smoke. Among persons who endorsed at least one risk behavior, the odds of attending a past-year primary care visit were 25–35% lower for those who engaged in multiple behaviors compared to the odds of persons who engaged in one behavior.

Conclusion

Substance use screening and intervention services in primary care may not be reaching individuals with the greatest need for services. Proactive outreach and identification of primary care utilization barriers are needed, with special consideration of those with co-occurring substance use.

Keywords

primary care primary care utilization alcohol smoking health risk behaviors multiple health risk behaviors

In Brief

Smoking and at-risk drinking are each associated with lower primary care utilization, but the influence of their co-occurrence on primary care utilization is not known. As such, the present study examined associations of at-risk drinking, current smoking, and their co-occurrence with attending a past-year primary care checkup using a nationally representative sample of adults aged 18–64. Results showed that at-risk drinking and current smoking were each associated with lower odds of attending a past-year primary care checkup. The key study finding is that their co-occurrence is associated with lower odds of attending a past-year primary care checkup compared to engaging in only one of these behaviors. Thus, substance use screening and intervention services in primary care may not be reaching individuals with the greatest need for services. Future research is needed to identify and alleviate barriers of primary care utilization among this population.

Purpose

Approximately 12.6% of adults in the United States (U.S.) engage in at-risk drinking (ie, either binge or heavy drinking).¹ At-risk drinking is linked to numerous chronic health conditions such as heart disease,² liver disease,³ and cancer.⁴ Approximately 13.7% of U.S. adults currently smoke,⁵ and smoking is the leading cause of preventable morbidity and mortality in the U.S.⁶ Smoking is a causal factor for numerous chronic health conditions, including heart disease, cancer, and asthma.^7,8 The co-occurrence of smoking and at-risk drinking is significant,^9,10 as over 43% of persons who smoke also meet criteria for at-risk drinking.¹¹ Furthermore, the co-occurrence of at-risk drinking and smoking is associated with increased risk of chronic health conditions compared to engaging in one behavior.^12-14

Many adults who drink excessively and/or smoke do not seek substance use treatment services.^15-17 Primary care providers provide critical prevention, diagnosis, and treatment services for chronic health conditions that are disproportionately experienced by persons who drink excessively and/or smoke.^18-21 Thus, they are well-positioned to identify and refer individuals who may need services for alcohol and tobacco use, but who may not otherwise seek services on their own.^22,23 The World Health Organization (WHO) recommended the implementation of screening, brief intervention, and referral to treatment services (SBIRT) for alcohol, tobacco, and substance use in primary care settings to fill the gap between prevention and treatment services for substance use.^22,24,25 SBIRT is widely implemented in primary care settings in the U.S.,²⁵ and has demonstrated effectiveness in identifying and treating individuals who drink above moderate levels (ie, at-risk drinking) or smoke.^15,26,27 However, screening and intervention services for health risk behaviors in primary care settings are reliant on individuals presenting to their healthcare providers,²⁸ and the relationship between at-risk drinking, smoking, and their co-occurrence with primary care utilization is largely understudied.

The role of primary care settings for identifying and serving persons who drink at-risk levels and smoke may be limited by the extent to which these persons use primary care services. Prior studies indicate mixed findings regarding associations of at-risk drinking with attending a primary care checkup.^29-32 One nationally representative U.S. study found that persons who drink at-risk levels had lower rates of attending a checkup than those who do not drink.²⁹ However, two U.S. studies that also used nationally representative samples found that the likelihood of past-year primary care attendance did not significantly differ by at-risk drinking status.^30,31 Finally, one German study found that, compared to not drinking, at-risk drinking was associated with higher odds of attending a past-year primary care visit among women, but not among men.³²

Persons who smoke are generally less likely to use primary care services than persons who do not smoke.^33-38 However, few U.S. studies have examined the relationship between smoking status and primary care attendance. One study found daily smoking to be associated with lower odds of attending a primary care checkup than non-smoking.³³ A study of Colorado primary care patients found that persons who do not smoke have significantly higher odds of attending a past-year primary care visit compared to persons who smoke.³⁴ Kahende et al.’s³⁶ study of a nationally representative sample of U.S. adults found that current smoking was not associated with primary care attendance. Prior studies have found that having previously smoked is associated with a higher likelihood of primary care attendance compared to never having smoked,^33,36 but some studies found no such association.^37,38

Research on health risk behaviors and primary care utilization has focused on individual health risk behaviors. In addition to the increased risk of chronic health conditions associated with multiple health risk behaviors,^12-14,39 those who engage in co-occurring at-risk drinking and smoking report more frequent alcohol and tobacco use and have greater difficulty in cessation of these behaviors.^40-42 Given the greater need for preventive services by persons who engage in these behaviors, the absence of research examining the association of multiple risk behaviors with primary care attendance is a significant gap in the literature. Therefore, the purpose of the current study is to explore the associations between at-risk drinking, current smoking, and their co-occurrence with primary care attendance. The primary research question examined is: Among those who engage in at least one health risk behavior, is engagement in multiple health risk behaviors associated with a past-year checkup? Secondary research questions examined include whether each individual behavior (drinking status and smoking status) is associated with attending a past-year checkup. Due to the mixed findings regarding associations between at-risk drinking and smoking with primary care attendance and the dearth of evidence surrounding associations between engagement in multiple health risk behaviors with primary care attendance, we refrain from making formal hypotheses with regard to the associations examined in the current study.

Methods

Design

Study data comes from the 2018 Behavioral Risk Factor Surveillance System (BRFSS),⁴³ which is a nationally representative survey of health behaviors, chronic health conditions, and healthcare access among U.S. adults. Data is collected through cellular and landline phone calls in all 50 states, the District of Columbia, Guam, and Puerto Rico.⁴³ A raking weighting methodology is used to improve the sample’s generalizability to individual states and territories.⁴³ BRFSS contains optional modules and state-added questions, which are administered to subsets of respondents. BRFSS core items are those administered to every survey respondent.⁴³ This study exclusively utilized data from core items because of their administration to respondents in all fifty states and participating territories.⁴³

Sample

The present study’s sample was composed of respondents between 18–64 years old with complete data on key variables (N = 246 801). Analyses examining engagement in multiple health risk behaviors included the subsample of adults who endorsed engagement in either current drinking or current smoking and adults who endorsed engagement in both behaviors (N = 158 162).

Measures

Attending a checkup in the past year was the outcome variable of this study. Past-year checkup was a dichotomous variable computed from responses to the following BRFSS item: “About how long has it been since you last visited a doctor for a routine checkup? [A routine checkup is a general physical exam, not an exam for a specific injury, illness, or condition.].” Item responses included “within past year”; “within past 2 years”; “within past 5 years”; or “5 or more years ago.” Responses of “within past year” were recoded as “yes” and other responses were recoded as “no.”

Drinking was a computed three-category variable of drinking behavior in the past-month that denoted at-risk drinking, non-at-risk drinking (ie, past-month drinking that did not meet criteria for at-risk drinking), and no drinking. Engagement in at-risk drinking in the past month was operationalized consistent with National Institute on Alcohol Abuse and Alcoholism (NIAAA) criteria⁴⁴: consumption of 4 or more drinks on one occasion or 8 or more drinks per week for women, and; consumption of 5 or more drinks on one occasion or 15 or more drinks per week for men.⁴⁵

Smoking status was a computed three-category variable that indicated if a respondent endorsed current smoking, former smoking, or non-smoking. Current smoking was operationalized as self-report of having smoked at least 100 cigarettes in one’s lifetime and currently smoking “some days” or “every day.” Participants who reported smoking 100 or more cigarettes and answered “not at all” to the current smoking item were coded as being a former smoker. Those who did not report smoking 100 cigarettes or more in their lifetime were coded as never having smoked (ie, non-smoking).

Covariates included sociodemographic characteristics and chronic health conditions that could confound the relationship between health risk behaviors and attending primary care. The referent category for each of the categorical variables included in multivariable logistic regression analyses are presented in Table 3. Sex was defined as male or female. Race/ethnicity classifications were Black non-Hispanic, Hispanic, Other Race non-Hispanic, and White non-Hispanic. Age was a five-category variable as follows: 18–24, 25–34, 35–44, 45–54, and 55–64. Educational attainment was a three-category variable representing high school diploma/GED or less, some college or technical school, or completed college or technical school. Marital status indicated whether a respondent was married/partnered or not married/partnered. Employment status indicated whether a respondent was employed (full or part time) or not employed. Metropolitan area refers to the participant’s county of residence and was coded as metropolitan (counties in a metropolitan statistical area with a population ≥50 000) or non-metropolitan (counties in a micropolitan statistical area or counties that are not in proximity to an urban core).⁴⁶ Health insurance status denoted whether a respondent reported any form of health insurance or was uninsured. Individual items from the BRFSS capture lifetime diagnosis of following chronic health conditions: heart attack, angina or coronary heart disease, stroke, asthma, skin cancer, other types of cancer, chronic obstructive pulmonary disease (or emphysema or chronic bronchitis), arthritis, depressive disorder, kidney disease, and diabetes. Positive responses to each of these items were summed to obtain the total number of chronic health conditions. Less than 1% of respondents reported 6 or more chronic health conditions. Thus, total number of chronic health conditions was coded with a truncated range of 0 to 5 or more. Income was considered but excluded due to multicollinearity with marital status and education in the adjusted models (variance inflation factors >4), which resulted in overfit models.

Analysis

Analyses were performed using Mplus (v. 7.4) and Stata 16. All estimates were weighted by employing BRFSS cluster, stratification, and individual person weights. Descriptive statistics included univariate frequencies for participant descriptive characteristics and unadjusted rates of attending a past-year checkup by health behavior status. Multivariable logistic regression analyses were conducted to examine associations between drinking status with a past-year checkup (Model 1); associations between smoking status with a past-year checkup (Model 2); and associations of co-occurring drinking and current smoking (vs engagement in one behavior) with a past-year checkup among adults who engaged in at least one of these behaviors (Model 3). In Model 1, no past month drinking was coded as the referent category and non-at-risk drinking and at-risk drinking were coded as indicators (ie, 0 = no drinking; 1 = non-at-risk drinking; 2 = at-risk drinking. In Model 2, non-smoking was coded as the referent category and former smoking and current smoking were coded as indicators (ie, 0 = non-smoking; 1 = former smoking; 2 = current smoking. Because persons who formerly and never smoked did not differ significantly from each other, these two categories were combined in Model 3. Thus, engagement in multiple health risk behaviors was represented by two dichotomous variables wherein “current smoking + non-at-risk drinking” and “current smoking + at-risk drinking” were each compared to “one behavior.” In Model 3, engagement in only one behavior (current smoking or non-at-risk drinking or at-risk drinking only) was coded as the referent category and engagement in non-at-risk drinking and current smoking and engagement in at-risk drinking and current smoking were coded as indicators, respectively, (ie, 0 = one behavior; 1 = non-at-risk drinking and current smoking; 2 = at-risk drinking and current smoking).

Adjusted rates of attending a primary care checkup were computed from the estimated models. All multivariable analyses adjusted for sociodemographics and chronic health conditions.

Results

Sample Characteristics

Table 1 shows that 72.5% of U.S. adults aged 18–64 attended a past-year checkup in 2018. Additionally, 21.1% of respondents engaged in at-risk drinking, and 17.2% reported current smoking. One-third of respondents reported either at-risk drinking or current smoking, of which 17.6% reported co-occurring at-risk drinking and current smoking. Table 2 shows the unadjusted rates of attending a past-year checkup. A lower percentage of persons who engaged in at-risk drinking (65.7%) attended a past-year check-up than persons who did not drink (75.2%) and persons who engaged in non-at-risk drinking (73.2%). Additionally, 65.1% of persons who currently smoked attended a past-year checkup compared to 74.9% of persons who formerly smoked and 73.8% of those who never smoked. Among persons who reported current drinking or smoking, 58.0% of persons who engaged in at-risk drinking and current smoking attended a past-year checkup, vs 65.6% of persons who engaged in non-at-risk drinking and current smoking, and 72.2% of persons who engaged in only one behavior.

Table 1.

Participant Characteristics, 2018 Behavioral Risk Factor Surveillance System (N = 246 801).

Characteristics	Prevalence (95% CI)
Past-year checkup
Did not attend	27.5 (27.1–27.8)
Attended	72.5 (72.2–72.9)
Sex
Women	50.5 (50.1–50.9)
Men	49.5 (49.1–49.9)
Race/Ethnicity
Black non-Hispanic	12.0 (11.8–12.3)
Hispanic	18.7 (18.4–19.1)
Other race	8.7 (8.4–8.9)
White non-Hispanic	59.1 (58.8–59.5)
Age
18–24	15.8 (15.5–16.1)
25–34	22.2 (21.9–22.6)
35–44	20.3 (20.0–20.6)
45–54	20.2 (19.9–20.5)
55–64	21.6 (21.3–21.9)
Education
High school graduate or less	39.8 (39.4–40.2)
Attended college or technical school	31.3 (30.9–31.7)
Graduated from college or technical school	29.0 (28.6–29.3)
Marital status
Married/Partnered	55.2 (54.9–55.6)
Not married	44.8 (44.4–45.2)
Employment status
Employed	68.8 (68.4–69.1)
Unemployed	31.2 (30.9–31.6)
Metropolitan area
Metropolitan	85.6 (85.4–85.8)
Nonmetropolitan	14.4 (14.2–14.6)
Health insurance status
Insured	85.7 (85.3–86.0)
Uninsured	14.4 (14.0–14.7)
Mean number of chronic health conditions	.9 (.8–.9)
At-risk drinking status
Non-drinking	44.7 (44.3–45.1)
Non-at-risk drinking	34.3 (33.9–34.7)
At-risk drinking	21.1 (20.7–21.4)
Smoking status
Non-smoking	62.7 (62.3–63.1)
Former smoking	20.1 (19.8–20.4)
Current smoking	17.2 (16.9–17.5)
Drinking and current smoking behaviors
None	37.7 (37.3–38.1)
Non-at-risk drinking, at-risk drinking, or current smoking	62.3 (61.9–62.7)
Co-occurring drinking and current smoking^a
One behavior only	83.5 (83.1–83.9)
Current smoking + non-at-risk drinking	7.3 (7.1–7.6)
Current smoking + at-risk drinking	9.2 (8.9–9.5)
At-risk drinking and current smoking behaviors
None	67.5 (67.1–67.8)
At-risk drinking or current smoking	32.5 (32.2–32.9)
Co-occurring at-risk drinking and current smoking^b
One behavior only	82.4 (81.9–83.0)
Current smoking + at-risk drinking	17.6 (17.1–18.1)

^aAmong those who engaged in either non-at-risk drinking, at-risk drinking, or current smoking.

^bAmong those who engaged in either at-risk drinking, or current smoking.

Table 2.

Prevalence Rates of Attending A Past-Year Checkup by Health Risk Behaviors (N = 246 801), 2018 Behavioral Risk Factor Surveillance System.

	Rates of attending a past-year checkup (95% confidence interval)
	Unadjusted	Adjusted
Drinking status
Non-drinking	75.2 (74.7–75.8)	74.3 (73.8–74.9)
Non-at-risk drinking	73.2 (72.6–73.8)	72.3 (71.7–72.9)
At-risk drinking	65.7 (64.9–66.5)	69.3 (68.6–70.1)
Smoking status
Non-smoking	73.8 (73.4–74.3)	74.1 (73.6–74.5)
Former smoking	74.9 (74.2–75.7)	73.5 (72.8–74.3)
Current smoking	65.1 (64.2–65.9)	65.5 (64.6–66.4)
Multiple health risk behaviors^a
One behavior^b	72.2 (71.7–72.7)	71.6 (71.1–72.1)
Current smoking + non-at-risk drinking	65.6 (63.9–67.3)	66.0 (64.3–67.7)
Current smoking + at-risk drinking	58.0 (56.3–59.6)	63.1 (61.6–64.6)

Note: Adjusted rates are adjusted for for sociodemographic characteristics and number of chronic health conditions.

^aAmong those who engaged in non-at-risk drinking, at-risk drinking, or current smoking.

^bEngagement in one behavior refers to engaging in only one of the following behaviors: at-risk drinking, non-risk drinking, or current smoking.

Associations of Drinking, Smoking, and Their Co-Occurrence With Attending a Past-Year Checkup

Table 3 summarizes model results. Model 1 shows that persons who drank at non-at-risk levels (adjusted odds ratio [AOR] = .89, 95% confidence interval [CI] = .85–.93) and persons who drank at risk levels (AOR = .76, 95% CI = .72–.80) had lower odds of attending a past-year checkup relative to the odds of those who did not drink, adjusting for sociodemographics and chronic health conditions. As shown in Table 2, adjusted rates of attendance were significantly lower for persons who drank at risk levels compared to persons in either of the other two drinking categories. The rate of attendance for persons who drank at non-at-risk levels was significantly lower than that of persons who did not drink.

Table 3.

Multivariable logistic regression analyses of attending a past-year checkup, 2018 Behavioral Risk Factor Surveillance System.

	Adjusted odds ratios and 95% confidence intervals
Predictor	Model 1 (N = 246 801)		Model 2 (N = 246 801)		Model 3 (N = 158 162)
Sex (men)
Women	1.43	(1.37–1.48)	1.44	(1.38–1.49)	1.41	(1.34–1.48)
Race (white non-Hispanic)
Black non-Hispanic	2.07	(1.93–2.21)	2.07	(1.93–2.22)	2.08	(1.91–2.27)
Hispanic	1.28	(1.20–1.36)	1.23	(1.16–1.32)	1.23	(1.13–1.33)
Other race	1.09	(1.01–1.18)	1.12	(1.03–1.21)	1.06	(.96–1.17)
Age (55–64)
18–24	.62	(.57–.66)	.58	(.54–.62)	.55	(.50–.60)
25–34	.51	(.48–.54)	.51	(.48–.54)	.49	(.46–.53)
35–44	.59	(.56–.63)	.60	(.56–.64)	.60	(.56–.64)
45–54	.80	(.75–.85)	.80	(.76–.86)	.79	(.73–.85)
Education (graduated college/technical school)
High school graduate or less	.89	(.85–.93)	.99	(.94–1.04)	.88	(.83–.93)
Some college/technical school	.92	(.88–.96)	.97	(.93–1.02)	.94	(.89–.99)
Marital status (married/Partnered)
Not married/partnered	.91	(.87–.95)	.93	(.90–.97)	.90	(.85–.94)
Employment status (employed)
Unemployed	.83	(.79–.87)	.81	(.77–.84)	.82	(.77–.87)
Metropolitan area (metropolitan)
Nonmetropolitan	.96	(.92–1.00)	.99	(.94–1.03)	.96	(.91–1.01)
Health insurance status (insured)
Uninsured	.30	(.28–.31)	.31	(.29–.32)	.29	(.27–.31)
Number chronic health conditions	1.29	(1.26–1.31)	1.32	(1.29–1.35)	1.28	(1.25–1.32)
At-risk drinking (non-drinking)
Non-at-risk drinking	.89	(.85–.93)
At-risk drinking	.76	(.72–.80)
Smoking status (non-smoking)
Former smoking			.97	(.92–1.02)
Current smoking			.64	(.60–.67)
Multiple behaviors (one behavior)^a
Current smoking + non-at-risk drinking					.75	(.69–.82)
Current smoking + at-risk drinking					.65	(.60–.70)

Note: Boldface indicates statistical significance. Reference groups are denoted in parentheses.

^aEngagement in one behavior refers to engaging in only one of the following behaviors: at-risk drinking, non-at-risk drinking, or current smoking.

Model 2 shows those who currently smoked had lower odds of attending a past-year checkup relative to the odds of persons who had never smoked (AOR = .64, 95% CI = .60–.67). Odds of attending a past-year checkup did not differ between persons who had formerly smoked and never smoked (AOR = .97, 95% CI = .92–1.02; Table 3). Table 2 shows that adjusted rates of a past-year checkup were significantly lower for persons who currently smoked compared to persons who formerly and never smoked, but rates did not differ between persons who formerly smoked and never smoked.

Table 3 shows that in Model 3, adults who engaged in more than one health risk behavior had lower odds of attending a past-year checkup than the odds of persons who engaged in one behavior: current smoking + non-at-risk drinking, AOR = .75, 95% CI = .69–.82; current smoking + at-risk drinking, AOR = .65, 95% CI = .60–.70. Adjusted rates of a past-year checkup were significantly lower among persons who engaged in more than one health risk behavior compared to those who engaged in one (Table 2).

Discussion

The present study used data from a large nationally representative sample to examine associations of drinking, current smoking, and their co-occurrence with attending a past-year checkup. This study builds upon the extant research because, to our knowledge, this is the first study to examine associations between co-occurring health risk behaviors and primary care utilization. Odds of attending an annual checkup were lower for persons who engaged in a health risk behavior compared to persons who did not engage in that behavior. These differences persisted even when adjusting for chronic health conditions and sociodemographic characteristics. New to the literature is the finding that persons who concurrently engaged in both drinking and smoking had lower odds of attending a past-year checkup than persons who engaged in only one of these behaviors.

The finding that at-risk drinking was associated with a lower likelihood of attending a primary care check-up is consistent with Naimi et al’s findings.²⁹ However, Heise³⁰ and Cherpitel and Ye³¹ found no association, and Hoebel et al³² found that at-risk drinking was associated with higher odds of attendance among women but not men. Previous null findings may have been a function of different time frames for alcohol consumption used across studies. Specifically, Heise³⁰ and Cherpitel and Ye³¹ examined past-year alcohol consumption. Similarly, Hoebel et al³² measured alcohol consumption with the AUDIT-C, which has been validated to screen for past-year at-risk drinking and alcohol use disorder.⁴⁷ Past-year recall of one’s drinking patterns should be less reliable than past-month recall, which could in turn introduce error that would obscure any associations with alcohol consumption. Another possible source of error could lie in the operationalization of alcohol consumption in these studies. Although Hoebel et al³² used recommended cut-off scores for the AUDIT-C to classify at-risk drinking (≥4 for women, ≥5 for men),⁴⁷ scores as low as 1 on items 2 and 3 are also indicative of at-risk drinking.⁴⁸ Heise³⁰ designated the range of 1–11 days of binge-drinking in the past year as “low at-risk” (vs ≥ 12 days, “high at risk”). This operationalization does not reflect standard criteria for at-risk drinking and divides those who would otherwise be considered “at-risk” in the past year into two arbitrarily defined at-risk drinking categories. Thus, both Hoebel et al’s³² and Heise’s³⁰ operationalizations could have resulted in moderate-risk categories that were not meaningfully different from their high-risk categories.

That those who currently smoked had lower odds of attending a primary care checkup than those who had never smoked is consistent with extant research.^33,34,37,38 That the odds of attending a checkup did not differ between persons who formerly smoked and persons who never smoked is consistent with Asada and Kephart³⁸ and Wacker et al³⁷; however, Culica et al³³ and Kahende et al³⁶ found that persons who formerly smoked were more likely to attend a checkup than those who never smoked. Studies have consistently found a negative association between smoking and primary care attendance even after adjusting for sociodemographic characteristics, insurance status, and numerous health conditions and health behaviors. Although many of these factors were significantly associated with primary care use across studies, they did not fully account for the relationship between smoking and primary care use; thus, additional research on what accounts for this association is needed.

The key finding of this study is that the co-occurrence of drinking (at at-risk and non-at-risk levels) and smoking is a more substantial risk factor for not attending a primary care checkup than engaging in only one behavior. This is important given that co-occurring at-risk drinking and smoking is linked to poorer overall health and chronic health conditions than engaging in only one or the other.^12-14,39 Co-occurring at-risk drinking and smoking is also linked to higher levels of alcohol and tobacco use severity and greater difficulty in cessation.^40-42 Thus, these persons are likely in the greatest need of primary care screening, intervention, and referral services, but they are also at higher risk of underutilizing primary care services than persons who engage in one of these behaviors.

One potential driver of the inverse relationships of drinking and smoking with primary care use is an avoidance of care among those who are most in need.^49,50 Although this is speculative, it is consistent with the extant research on barriers to help-seeking. Persons who engage in health-risk behaviors perceive their behaviors as low-risk or do not view themselves as susceptible to the harmful effects of at-risk drinking and smoking.^17,35,51 These individuals may avoid primary care services because of feelings of shame and embarrassment surrounding their drinking and smoking.^49,50 Furthermore, evidence suggests a link between polysubstance use and increased shame and internal stigmatization (although these findings were not specific to co-occurring drinking and smoking).^52,53 Thus, adults who engage in both health-risk behaviors may be less likely to attend primary care than those who engage in only one behavior due to higher levels of internal shame and stigmatization.

Moreover, persons who formerly drank at risk levels or smoked attend primary care and acute care settings at higher frequencies than persons who have not engaged in these health risk behaviors,^36,50,54 and this may provide further context on the relationship between health-risk behaviors and underutilization of primary care. It has been argued that persons who engage in these behaviors may avoid routine and preventative health care until their health deteriorates to the point that they need to quit and seek medical care.^50,54 Thus, the adverse health effects of at-risk drinking and smoking and the underutilization of routine care may contribute to worsening health which leads to frequent health care service engagement in the future. Although the dynamics that impede primary care utilization among persons who engage in these behaviors are multi-faceted, particular attention may need to be paid to the interpersonal and psychological barriers among these persons to increase their use of primary care services.

The current study focuses on patient characteristics as determinants of primary care attendance and, by extension, potential receipt of SBIRT services. However, it is important to recognize that receipt of SBIRT services is also determined by the proper delivery of these services in healthcare settings. While most persons who attend health care settings receive screening for alcohol and tobacco use, critical service delivery gaps have been identified.^55-60 An estimated 23–29% of ambulatory care attendees may not be screened for alcohol use^55,57 and 15–37% may not be screened for smoking.^56,58 Moreover, rates of receiving advice to quit smoking,^56,59 advice to quit drinking, and/or information about alcohol treatment^55,57,60 are low in ambulatory care settings. Thus, reaching those most in need of SBIRT in primary care requires an understanding of barriers at multiple levels (eg, patient-, provider-, institution-level).

Limitations

Current study findings should be considered in light of the following limitations. Data obtained from BRFSS is cross-sectional, and this study design cannot speak to causal relationships. All responses were self-reported, and responses may have been influenced by social desirability and recall biases. Study findings are pertinent to primary care services utilization and may not be generalizable to other healthcare services, such as specialty clinics. Recommendations on the frequency of attending a checkup vary between demographic groups⁶¹; thus, the present study’s examination of attending an annual checkup may not be applicable to all respondents. This study was limited to examining drinking and smoking behaviors, and other health risk behaviors that are relevant to primary care services utilization were not captured in this study.

Conclusions

Engaging in both current smoking and drinking (at risk and non-at-risk) was associated with a significantly lower likelihood of attending a primary care visit in the past year relative to engaging in only one health risk behavior. As such, these findings indicate there is substantial room for improvement in the reach of SBIRT interventions in primary care. Psychological and interpersonal barriers may be major contributors to the lower attendance among this population, but increased research on the mechanisms of the relationship between health risk behaviors and primary care utilization is necessary to corroborate this speculation. Such research stands to identify targets of intervention for programs aimed to increase primary care utilization among this population. More generally, further study regarding ways to alleviate barriers to primary care utilization among this population is needed to increase utilization of SBIRT and other vital primary care services. Research that identifies alternative settings in which these persons may be as or more likely to frequent as their non-drinking and non-smoking counterparts may allow for opportunities for increasing the reach of SBIRT services. Finally, healthcare settings traditionally rely on individuals to initiate contact in order to provide services, but there is a need for proactive outreach interventions to increase contact with those in need of services.²⁸

So What?

What is Already Known on this Topic?

Persons who smoke or drink at risk levels are a priority population for substance use screening and brief intervention services in primary care settings because they have an increased risk of chronic health conditions. However, they are generally less likely to use primary care services than persons who do not drink at risk levels or smoke.

What Does this Article Add?

Not only are at-risk drinking and smoking independent risk factors of not attending a primary care checkup, but their co-occurrence is incrementally associated with lower odds of attending a primary care checkup compared to engaging in only one of these behaviors.

What are the Implications for Health Promotion Practice or Research?

Persons with co-occurring at-risk drinking and smoking are in need of proactive outreach from healthcare professionals to increase contact. Research is needed to identify and reduce barriers of primary care attendance among this population.

Footnotes

Author Contributions

All authors contributed to the study design, drafting/editing of the manuscript, and approved the version to be published. John Moore conceptualized the study and led data analyses and manuscript drafting/editing. Yessenia Castro contributed to study conceptualization, data analyses, and manuscript drafting/editing. Catherine Cubbin contributed to study conceptualization and manuscript editing. Kirk von Sternberg contributed to study conceptualization and manuscript editing.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

John R. Moore

References

Grant

Chou

Saha

, et al. Prevalence of 12-month alcohol use, high-risk drinking, and DSM-IV alcohol use disorder in the United States, 2001-2002 to 2012-2013: Results from the national epidemiologic survey on alcohol and related conditions. JAMA Psychiatry. 2017;74(9):911-923. doi:10.1001/jamapsychiatry.2017.2161.

Roerecke

Rehm

. Chronic heavy drinking and ischaemic heart disease: a systematic review and meta-analysis. Open Heart. 2014;1(1):e000135. doi:10.1136/openhrt-2014-000135.

Centers for Disease Control and Prevention . Excessive alcohol use. https://www.cdc.gov/chronicdisease/resources/publications/factsheets/alcohol.htm. Published June 26, 2019. Accessed August 21, 2020.

Gupta

Wang

Holly

Bracci

. Risk of pancreatic cancer by alcohol dose, duration, and pattern of consumption, including binge drinking: a population-based study. Cancer Causes Control. 2010;21(7):1047-1059. doi:10.1007/s10552-010-9533-6.

Creamer

Wang

Babb

, et al. Tobacco product use and cessation indicators among adults — United States, 2018. MMWR Morb Mortal Wkly Rep. 2019;68(45):1013-1019. Published November 2019. doi:10.15585/mmwr.mm6845a2.

Henley

Thomas

Sharapova

, et al. Vital signs: Disparities in tobacco-related cancer incidence and mortality — United States, 2004–2013. MMWR Morb Mortal Wkly Rep. 2016;65(44):1212-1218. Published November 2016. doi:10.2307/24859134.

National Center for Chronic Disease Prevention Health Promotion (US) Office on Smoking and Health. The Health Consequences of Smoking—50 Years of Progress: A Report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention (US); 2014. Accessed July 31, 2020. http://www.ncbi.nlm.nih.gov/books/NBK179276/.

Schane

Ling

Glantz

. Health effects of light and intermittent smoking. Circulation. 2010;121(13):1518-1522. doi:10.1161/CIRCULATIONAHA.109.904235.

Blazer

L-T

. The epidemiology of at-risk and binge drinking among middle-aged and elderly community adults: National survey on drug use and health. Am J Psychiatry. 2009;166(10):1162-1169. doi:10.1176/appi.ajp.2009.09010016.

10.

Kranzler

Amin

Cooney

, et al.

Screening for health behaviors in ambulatory clinical settings: does smoking status predict hazardous drinking?

Addict Behav. 2002;27(5):737-749. doi:10.1016/S0306-4603(01)00206-4.

11.

Substance Abuse and Mental Health Services Administration . Results from the 2016 national survey on drug use and health: Detailed tables | CBHSQ Data. https://www.samhsa.gov/data/report/results-2016-national-survey-drug-use-and-health-detailed-tables Accessed August 1, 2020.

12.

Adams

. Psychopharmacology of tobacco and alcohol comorbidity: A review of current evidence. Curr Addict Rep. 2017;4(1):25-34. doi:10.1007/s40429-017-0129-z.

13.

Pelucchi

Gallus

Garavello

Bosetti

La Vecchia

. Cancer risk associated with alcohol and tobacco use: focus on upper aero-digestive tract and liver. Alcohol Res Health. 2006;29(3):193-198.

14.

Marrero

Fontana

Conjeevaram

Lok

. Alcohol, tobacco and obesity are synergistic risk factors for hepatocellular carcinoma. J Hepatol. 2005;42(2):218-224. doi:10.1016/j.jhep.2004.10.005.

15.

Madras

Compton

Avula

Stegbauer

Stein

Clark

. Screening, brief interventions, referral to treatment (SBIRT) for illicit drug and alcohol use at multiple healthcare sites: Comparison at intake and 6 months later. Drug Alcohol Depend. 2009;99(1):280-295. doi:10.1016/j.drugalcdep.2008.08.003.

16.

McCance-Katz

Satterfield

. SBIRT: A key to integrate prevention and treatment of substance abuse in primary care. Am J Addict. 2012;21(2):176-177. doi:10.1111/j.1521-0391.2011.00213.x.

17.

Chapman

Wong

Smith

. Self-exempting beliefs about smoking and health: Differences between smokers and ex-smokers. Am J Public Health. 1993;83(2):215-219. doi:10.2105/ajph.83.2.215.

18.

Basu

Berkowitz

Phillips

Bitton

Landon

Phillips

. Association of primary care physician supply with population mortality in the United States, 2005-2015. JAMA Intern Med. 2019;179(4):506-514. doi:10.1001/jamainternmed.2018.7624.

19.

Chowdhury

Mawokomatanda

, et al. Surveillance for certain health behaviors, chronic diseases, and conditions, access to health care, and use of preventive health services among states and selected local areas — Behavioral Risk Factor Surveillance System, United States, 2012. MMWR Surveill Summ. 2016;65(4):1-139. doi:10.2307/24805794.

20.

Ganguli

Shi

Orav

Rao

Ray

Mehrotra

. Declining use of primary care among commercially insured adults in the United States, 2008–2016. Ann Intern Med. 2020;172(4):240-247. doi:10.7326/M19-1834.

21.

Starfield

Shi

Macinko

. Contribution of primary care to health systems and health. Milbank Q. 2005;83(3):457-502. doi:10.1111/j.1468-0009.2005.00409.x.

22.

Babor

McRee

Kassebaum

Grimaldi

Ahmed

Bray

. Screening, Brief Intervention, and Referral to Treatment (SBIRT): Toward a public health approach to the management of substance abuse. Subst Abus. 2007;28(3):7-30. doi:10.1300/J465v28n03_03.

23.

Humeniuk

Ali

Babor

, et al. A randomized controlled trial of a brief intervention for illicit drugs linked to the alcohol, smoking and substance involvement screening test (ASSIST) in clients recruited from primary health-care settings in four countries. Addiction. 2012;107(5):957-966. doi:10.1111/j.1360-0443.2011.03740.x.

24.

Agerwala

McCance-Katz

. Integrating screening, brief intervention, and referral to treatment (SBIRT) into clinical practice settings: A brief review. J Psychoactive Drugs. 2012;44(4):307-317.

25.

Saitz

. Screening and brief intervention enter their 5th decade. Subst Abus. 2007;28(3):3-6. doi:10.1300/J465v28n03_02.

26.

Moyer

Preventive Services Task Force . Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2013;159(3):210-218. doi:10.7326/0003-4819-159-3-201308060-00652.

27.

Land

Rigotti

Levy

Schilling

Warner

. The effect of systematic clinical interventions with cigarette smokers on quit status and the rates of smoking-related primary care office visits. PLoS One. 2012;7(7):e41649. doi:10.1371/journal.pone.0041649.

28.

Prochaska

. Multiple health behavior research represents the future of preventive medicine. Prev Med. 2008;46(3):281-285. doi:10.1016/j.ypmed.2008.01.015.

29.

Naimi

Town

Mokdad

Brewer

. Health care access among U.S. adults who drink alcohol excessively: missed opportunities for prevention. Prev Chronic Dis. 2006;3(2). Accessed July 31, 2020. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1563973/.

30.

Heise

. Healthcare system use by risky alcohol drinkers: a secondary data analysis. J Am Acad Nurse Pract. 2010;22(5):256-263. doi:10.1111/j.1745-7599.2010.00500.x.

31.

Cherpitel

. Risky drinking, alcohol use disorders, and health services utilization in the U.S. general population: Data from the 2005 and 2010 National Alcohol Surveys. Alcohol Clin Exp Res. 2015;39(9):1698-1704. doi:10.1111/acer.12801.

32.

Hoebel

Starker

Jordan

Richter

Lampert

. Determinants of health check attendance in adults: findings from the cross-sectional German Health Update (GEDA) study. BMC Publ Health. 2014;14(1):913. doi:10.1186/1471-2458-14-913.

33.

Culica

Rohrer

Ward

Hilsenrath

Pomrehn

. Medical checkups: Who does not get them? Am J Public Health. 2002;92(1):88-91. doi:10.2105/AJPH.92.1.88.

34.

Preisser

Cohen

Wofford

, et al. Physician and patient predictors of health maintenance visits. Arch Fam Med. 1998;7(4):346. doi:10.1001/archfami.7.4.346.

35.

Jorm

Shepherd

Rogers

Blyth

. Smoking and use of primary care services: Findings from a population-based cohort study linked with administrative claims data. BMC Health Serv Res. 2012;12(1):263. doi:10.1186/1472-6963-12-263.

36.

Kahende

Adhikari

Maurice

Rock

Malarcher

. Disparities in health care utilization by smoking status – NHANES 1999-2004. Int J Environ Res Public Health. 2009;6(3):1095-1106. doi:10.3390/ijerph6031095.

37.

Wacker

Holle

Heinrich

, et al. The association of smoking status with healthcare utilisation, productivity loss and resulting costs: results from the population-based KORA F4 study. BMC Health Serv Res. 2013;13(1):278. doi:10.1186/1472-6963-13-278.

38.

Asada

Kephart

. Equity in health services use and intensity of use in Canada. BMC Health Serv Res. 2007;7(1):41. doi:10.1186/1472-6963-7-41.

39.

Kvaavik

Batty

Ursin

Huxley

Gale

. Influence of individual and combined health behaviors on total and cause-specific mortality in men and women: The United Kingdom Health and Lifestyle Survey. Arch Intern Med. 2010;170(8):711-718. doi:10.1001/archinternmed.2010.76.

40.

Falk

Hiller-Sturmhöfel

. An epidemiologic analysis of co-occurring alcohol and tobacco use and disorders. Alcohol Res Health. 2006;29(3):162-171.

41.

Roche

DJO

Ray

Yardley

King

. Current insights into the mechanisms and development of treatments for heavy-drinking cigarette smokers. Curr Addict Rep. 2016;3(1):125-137. doi:10.1007/s40429-016-0081-3.

42.

Substance Abuse and Mental Health Services Administration . Results from the 2013 national survey on drug use and health: Summary of national findings | CBHSQ Data. https://www.samhsa.gov/data/report/results-2013-national-survey-drug-use-and-health-summary-national-findings Accessed August 21, 2020.

43.

Centers for Disease Control and Prevention . Behavioral risk factor surveillance system (BRFSS) Overview 2018. https://www.cdc.gov/brfss/annual_data/2018/pdf/overview-2018-508.pdf Accessed August 21, 2020.

44.

National Institute on Alcohol Abuse and Alcoholism . Alcohol facts and statistics. Published April 25, 2019. https://www.niaaa.nih.gov/publications/brochures-and-fact-sheets/alcohol-facts-and-statistics Accessed August 21, 2020.

45.

Centers for Disease Control and Prevention . Behavioral risk factor surveillance system (BRFSS) codebook. 2018. https://www.cdc.gov/brfss/annual_data/2018/pdf/overview-2018-508.pdf Accessed August 21, 2020.

46.

Ingram

Franco

. 2013 NCHS urban-rural classification scheme for counties. Vital Health Stat. 2014;2(166):1-73. https://www.cdc.gov/nchs/data/series/sr_02/sr02_166.pdf. Accessed April 30,2021.

47.

Bradley

DeBenedetti

Volk

Williams

Frank

Kivlahan

. AUDIT-C as a brief screen for alcohol misuse in primary care. Alcohol Clin Exp Res. 2007;31(7):1208-1217. doi:10.1111/j.1530-0277.2007.00403.x.

48.

Babor

Higgins-Biddle

Saunders

Monteiro

. AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Care. 2nd ed. Geneva, Switzerland: World Health Organization; 2001.

49.

Burrows

Carlisle

. They don’t want it ramming down their throats. Learning from the perspectives of current and ex-smokers with smoking-related illness to improve communication in primary care: a qualitative study. Prim Health Care Res Dev. 2010;11(3):206-214. doi:10.1017/S1463423609990351.

50.

Green

Polen

Leo

, et al. Drinking patterns, gender and health III: Avoiding vs. seeking healthcare. Addict Res Theory. 2010;18(2):160-180.

51.

Arnett

. Optimistic bias in adolescent and adult smokers and nonsmokers. Addict Behav. 2000;25(4):625-632. doi:10.1016/S0306-4603(99)00072-6.

52.

Dearing

Stuewig

Tangney

. On the importance of distinguishing shame from guilt: relations to problematic alcohol and drug use. Addict Behav. 2005;30(7):1392-1404. doi:10.1016/j.addbeh.2005.02.002.

53.

Kulesza

Watkins

Ober

Osilla

Ewing

. Internalized stigma as an independent risk factor for substance use problems among primary care patients: Rationale and preliminary support. Drug Alcohol Depend. 2017;180:52-55. doi:10.1016/j.drugalcdep.2017.08.002.

54.

Rice

Conell

Weisner

Hunkeler

Fireman

. Alcohol drinking patterns and medical care use in an HMO setting. J Behav Health Serv Res. 2000;27(1):3-16. doi:10.1007/BF02287800.

55.

Glass

Bohnert

Brown

. Alcohol screening and intervention among United States adults who attend ambulatory healthcare. J Gen Intern Med. 2016;31(7):739-745. doi:10.1007/s11606-016-3614-5.

56.

Keith

Stanton

Gaalema

, et al. Disparities in US healthcare provider screening and advice for cessation across chronic medical conditions and tobacco products. J Gen Intern Med. 2017;32(9):974-980. doi:10.1007/s11606-017-4062-6.

57.

Sahker

Arndt

. Alcohol use screening and intervention by American primary care providers. Int J Drug Policy. 2017;41:29-33. doi:10.1016/j.drugpo.2016.11.013.

58.

Jamal

Dube

Malarcher

Shaw

Engstrom

Centers for Disease Control and Prevention (CDC) . Tobacco use screening and counseling during physician office visits among adults--national ambulatory medical care survey and national health interview survey, United States, 2005-2009. MMWR Suppl. 2012;61(2):38-45.

59.

Danesh

Paskett

Ferketich

. Disparities in receipt of advice to quit smoking from health care providers: 2010 National Health Interview Survey. Prev Chronic Dis. 2014;11:E131. Published 2014 Jul 31. doi:10.5888/pcd11.140053.

60.

Bandara

Samples

Crum

Saloner

. Is screening and intervention associated with treatment receipt among individuals with alcohol use disorder? Evidence from a national survey. J Subst Abuse Treat. 2018;92:85-90. doi:10.1016/j.jsat.2018.06.009.

61.

National Library of Medicine . Physical exam frequency. MedlinePlus. https://medlineplus.gov/ency/article/002125.htm. Accessed July 6, 2021.