Secondhand smoke (SHS) exposure is a significant health problem that is rarely addressed in primary care. According to the American Cancer Society, SHS—also known as environmental tobacco smoke—comes from two sources: smokers when they exhale tobacco smoke (mainstream smoke) and the lighted ends of cigarettes, cigars, pipes, and hookahs (sidestream smoke). When nonsmokers are exposed to SHS, it's known as involuntary or passive smoking.1 It's well known that involuntary smoking is associated with negative health and economic outcomes; however, there is limited research outside of pediatric settings on how to effect behavior change in health care providers to improve their SHS screening and education of all patients. This gap in research and a lack of validated screening tools designed to evaluate SHS exposure led to the development of a quality improvement (QI) initiative in a primary care clinic in the southeastern United States. Before initiating the QI project, one of us (CM) surveyed the clinic's health care providers and confirmed that, despite their knowledge of the negative health outcomes and costs associated with SHS exposure, none regularly screened all their patients for such exposure. Providers reported screening some pediatric patients but only as part of well-child visits. The lack of comprehensive SHS screening and patient education warranted practice change.
Purpose of the initiative. The purpose of this QI initiative was to use selected interventions to facilitate behavior modification among health care providers to improve the SHS screening and education of all patients in our primary care setting. These interventions included clinician education, visual reminders, flag reminders, and a monetary incentive. The proposed outcomes included an increase in the number of screenings providers performed and an increase in the number of patients screened and counseled. We also proposed adding to the current electronic health record (EHR) an improved SHS screening question with pop-up flag reminders for both screening and education.
The first Surgeon General's Report on Smoking and Health was issued in 1964, and in 2014, to mark the 50th anniversary of that groundbreaking publication, the acting surgeon general released The Health Consequences of Smoking: 50 Years of Progress, along with a user-friendly consumer guide.2, 3 And while there has been progress in reducing tobacco use over the past half century, the toll smoking took in that period is staggering. More than 20 million people have died as a result of smoking since the first surgeon general's report, and of these, an estimated 2.5 million nonsmokers died from SHS exposure. About half of all children between the ages of three and 18 years are regularly exposed to SHS. And unless more is done to further reduce current smoking rates, an estimated 5.6 million children alive today—one out of every 13—will die early deaths as a result of smoking.
According to the American Cancer Society, tobacco smoke contains approximately 7,000 chemicals, of which 70 are carcinogenic.1 Any degree of SHS exposure is a health threat; there is no safe level of exposure.1 However, large proportions of the national and global populations remain exposed, continually and involuntarily, to SHS.4 More than a decade ago, the World Health Organization (WHO) issued the WHO Report on the Global Tobacco Epidemic, 2008: The MPOWER Package, the first in a series of reports organized around six policy recommendations for health policymakers5:
- Monitor tobacco use and prevention policies.
- Protect people from tobacco smoke.
- Offer help to quit tobacco use.
- Warn about the dangers of tobacco.
- Enforce bans on tobacco advertising, promotion, and sponsorship.
- Raise taxes on tobacco to limit exposure.
The WHO estimates that approximately 40% of the world's population is protected by at least one of these measures, and 20% by two or more.6
Provider behavior. Despite health care providers’ recognition of the need for widespread counseling to limit SHS exposure, the majority don't routinely screen and educate their patients.7, 8 Also, research that evaluates practical efforts to improve screening and counseling to reduce SHS exposure has been inadequate.9 The lack of both research and provider compliance constitutes a gap in patient care that may lead to negative health outcomes.
Comparable studies. Recent studies that focused on clinical behavior change to improve screening for and education on SHS exposure have shown varied results, and most examined pediatric populations. Bunik and colleagues assessed the ONE Step initiative, a smoking cessation program that required medical assistants to ask caregivers whether they smoked, pediatric providers to advise smokers to smoke outside the home and to quit if they were ready, providers to refer smokers to a telephone quitline, and the addition of prompts and documentation regarding smoking to the EHR. The results included a significant improvement in documentation of these steps in screening, and a survey found that 97% of caregivers considered the discussion about SHS acceptable.9
Kum-Nji and colleagues found that 61% of 232 inner-city pediatric patients from newborns to 12-year-olds were exposed to SHS, and 81% of mothers reported that their child's pediatrician had counseled them on the dangers of SHS.10 However, the researchers found documentation of the children's SHS exposure status in only 45% of their charts.
Provider behavior in the hospital setting. Kruse and Rigotti found that the addition of a screening question about SHS exposure (“Is smoking allowed in your home or car?”) to an EHR in a hospital setting resulted in nurses screening and educating 91% of 34,295 adult admissions and 86% of 1,406 pediatric admissions over one year.11
However, in a study of 140 pediatric inpatients; their families; and the ED providers, residents, and nurses who cared for them, Wilson and colleagues found that “the majority of smoke-exposed children were not identified as exposed based on documentation of admission screenings.”12 The researchers discovered this by first tracking providers’ frequency of SHS screening documentation and the number of children they identified as smoke exposed on admission and in interviews with parents, then comparing these data to the number of smoke-exposed children identified via analysis of cotinine in their saliva. Cotinine is the main metabolite of nicotine and a reliable biomarker for tobacco smoke exposure.13 A salivary cotinine level greater than or equal to 1 ng/mL was found in 46% of the pediatric patients, indicating SHS exposure, whereas in admission assessments and structured interviews, ED providers, pediatric residents, and nurses identified 18%, 38%, and 12% of children, respectively, as having been exposed. The researchers also found that ED providers, pediatric residents, and nurses documented their SHS-related assessment and interview findings 46%, 42%, and 79% of the time, respectively.12
Such inconsistencies in the application of screening and education recommendations indicate that primary care and other health care institutions should prioritize the implementation of QI initiatives focused on the modification of provider behavior and include an EHR component to ensure that all patient populations receive screening for and education on the dangers of SHS.
Health effects. Negative health outcomes associated with SHS exposure necessitate improvements in SHS screening, education, and documentation in the EHR. Approximately 41,000 nonsmoking adults die from SHS exposure every year.14 Research has shown a significant association between SHS exposure and ischemic heart disease, chronic obstructive pulmonary disease, and stroke.4 A meta-analysis of 18 studies that examined the relationship between SHS exposure and more than 6,000 cases of lung cancer in nonsmokers in China showed a similar increased cancer risk in both men and women.15
Pediatric health effects. Oberg and colleagues collected data from 192 countries to conduct a retrospective analysis of the worldwide disease burden associated with SHS and found that in 2004, 40% of children, 33% of adult male nonsmokers, and 35% of adult female nonsmokers were exposed to SHS.16 More than 600,000 deaths attributable to SHS exposure occurred the same year, with more than half (379,000) from ischemic heart disease. Twenty-eight percent of all those who died because of SHS exposure were children. When the disease burden attributable to SHS exposure was measured in disability-adjusted life years (DALYs), 61% of the 10.9 million DALYs lost were in children, the majority of these (more than 5.9 million) because of lower respiratory infections in children younger than five years.
Intrauterine SHS exposure, as measured by cotinine levels in umbilical cord and maternal and newborn peripheral blood, was also found to be associated with lower than average birth weight and shorter length at birth.17 Active and passive smokers typically show distinctively different concentrations of this biomarker, which can be measured in blood, saliva, and urine.18 Researchers have also found that children exposed to SHS are at higher than average risk for asthma, viral upper respiratory infections, acute otitis media, allergic rhinitis, gastroenteritis, and conjunctivitis.10
Economic impact. Another reason for clinicians to be more committed to improved SHS screening and patient education is the significant economic costs associated with SHS exposure. Tobacco smoking has an estimated economic impact of $300 billion per year, of which approximately $130 billion is for medical care and $156 billion is related to lost productivity, including $5.6 billion in lost productivity due to SHS exposure.2, 3 Such statistical data and research support the need for practice change.
We collaborated to develop this QI initiative and its specific interventions; a clinical preceptor, clinical staff members, and the health care organization's administrative staff provided further assistance with implementation. Organizational stakeholders were involved in the approval process of the QI initiative and components of the project.
The design of this project included a retrospective chart review, preintervention and postintervention questionnaires, and interventions designed to change the behavior of our primary care clinic's providers in regard to SHS screening and patient education. The existing SHS screening question in the EHR asked only whether the patient was exposed to SHS at home (“Does anyone smoke at home or in your household?”), which was inadequate for determining whether there had been SHS exposure outside the home, such as in workplaces, cars, public transportation, restaurants, entertainment venues, and other public spaces. (South Carolina is one of 22 states that currently have no laws to protect people from SHS in public places and workplaces, so it's likely that most patients had some degree of SHS exposure; see www.lung.org/our-initiatives/tobacco/smokefree-environments/smokefree-air-laws.html.) Nevertheless, the chart review was conducted to determine whether clinicians were asking the screening question and documenting patients’ responses in the EHR. Prior to implementing the interventions, clinicians responded to a questionnaire to determine their baseline knowledge and screening practices.
The implementation of interventions to promote clinician behavior changes that would improve SHS screening and patient education then commenced. One of us (CM) conducted an on-site in-service at one of our regular lunchtime staff meetings to educate the clinic's entire staff, including all participating clinicians and clerical staff members, on the importance of screening all patients for SHS exposure and teaching them the risks associated with it. (The staff comprises three LPNs, two NPs, one physician, one clinical nurse educator, one facility manager, and five administrative assistants who staff the front desk.) The education component included a PowerPoint presentation, followed by a question-and-answer period. Clinical providers were told that one monetary incentive in the form of a $50 gift card would be awarded to the provider who recorded the most SHS screenings in the six-week data collection period.
Visual reminders to prompt clinicians to screen and educate patients included buttons worn by clinicians and posters displayed on bulletin boards and walls in the lobby and in patients’ rooms. In addition, we developed a way to emulate the pop-up windows in the EHR system that alert users to uncompleted but required items (this was necessary because the local university's institutional review board [IRB] didn't permit the use of the EHR in the QI initiative). When patients entered the clinic, clerical staff at the front desk informed them of the initiative and, in accordance with IRB requirements, provided an informed consent statement for them to sign. The clerical staff gave those who consented to participate two small black and white checkered flags and instructed them to hand one to the LPN who conducted the initial assessment and the other to the provider (NP, physician, or nurse educator) who treated them. The flags served as reminders to the provider to conduct the SHS screening and educate patients on the risks associated with SHS, much like an EHR alert flag or pop-up window. The clinical providers also gave patients bookmarks that listed the dangers of SHS and an index card showing some of the harmful chemicals SHS contains. Because the QI team wasn't using the EHR for this screening, the clinicians had been supplied with printed paper forms with the screening questions and spaces to document the patients’ answers and to note that the patient education had been performed.
Sample patient population. The sample consisted of primary care patients and their dependents who were treated at a primary care clinic in the southeastern United States. The clinic provides primary care services for corporate employees and their dependents. Its mission centers on the delivery of cost-effective, high-quality, patient-centered care that improves the well-being of the overall corporate community.
Before implementing any interventions, a random sample of 617 patient charts were selected and analyzed. The chart analysis provided each participating provider's baseline screening percentage. During the six-week data collection period, the clinic's clerical staff solicited patients in the regular clinical population, as well as patients managed by clinical nurse educators, to participate in the QI initiative at check-in. The sample totaled 401 patients; 240 patients agreed to participate. Patients ranged from two years of age to those at the end of the life cycle. The clinic's total patient population is approximately 5,000; 95% of patients are from 18 to 87 years of age, and 5% are from two to 17 years of age.
Patient education. The educational information provided to patients included the following1, 19-21:
- SHS can travel up to 20 feet and stay in the air for several hours.
- SHS can increase the risk of heart disease by 25% to 30% and increase the risk of lung cancer and stroke by 20% to 30%.
- There are some 7,000 chemicals in SHS, 70 of which are known carcinogens.
- There is no safe level of SHS exposure.
- Air purification and ventilation of indoor spaces does not eliminate SHS; it travels through doorways, cracks in walls, ventilation systems, plumbing, and electrical systems.
- If an individual can smell SHS, then that individual is inhaling SHS.
- One in four (58 million) nonsmokers continue to be exposed to SHS.
Cost analysis. We used inexpensive materials in the QI initiative, but the costs were only partially covered by the health care organization, which supplied paper, pencils, pens, hard copies of PowerPoint slides, and a laminated sheet on the chemical components of SHS that clinical providers showed patients. One of us (CM) covered approximately $300 in costs for items such as the buttons worn by clinical staff members, the printing of bookmarks and posters, and the $50 monetary incentive for the clinician who completed the most screenings by the end of the six-week data collection period.
Measures used in the analysis of processes and outcomes of the interventions included four questionnaires, two each for patients and providers. The preimplementation provider questionnaire allowed us to evaluate clinicians’ baseline knowledge and screening practices (see Table 1). The postimplementation provider questionnaire gave us clinical feedback on the QI initiative, barriers to screening and education, and the proportion of providers who intended to continue screening patients for SHS exposure. It contained three questions: “Will you continue to screen since the initiative is complete?”; “What interventions helped you the most to remember to screen?”; and “Were there any barriers that prevented you from screening during the initiative?”
The preimplementation patient questionnaire was developed to screen patients for SHS exposure. It was filled out by the clinical provider and consisted of four yes–no questions, the last for the clinician to answer (see Preimplementation Patient Questionnaire). We developed the postimplementation patient questionnaire to provide additional information on the patient's interaction with the provider. This questionnaire was given to patients at the front desk as they were checking out and consisted of three yes–no questions (see Postimplementation Patient Questionnaire).
Data analysis was conducted using IBM SPSS statistical software Version 24 and Pearson's χ2 test.
Ethical considerations. Ethical aspects of the QI initiative centered on confidentiality, conflict of interest, and clinical safety. There was little-to-no risk foreseen as a result of questionnaire use: there were no patient identifiers on any of the questionnaires and collected data were stored in a locked office on password-protected computers.
Patient participation in the QI initiative was voluntary and we obtained all participants’ informed consent. Patients could refuse to participate, and those who had given consent could withdraw at any time without penalty or loss of any benefit. Patients who agreed to participate received no financial gain. Patients were not required to answer all questions.
Our retrospective preintervention chart review (N = 617) revealed that in 88 charts (14.3%), the provider had recorded screening the patient for SHS exposure. Out of 401 prospective participants, 240 (59.9%) agreed to participate in the QI initiative. After implementation of the interventions to promote clinician behavior changes, a second chart review showed that clinicians provided SHS screening and education to 237 (98.8%) of the 240 participants. The seven clinical providers who screened patients included three LPNs, two NPs, one clinical nurse educator, and one physician.
Demographics. Simple demographic data were collected on 240 patients. Among the 194 patients who reported their age, mean age was 45.5 years (SD, 14.014; age range, 18–87 years). A total of 139 participants identified their race as white (57.9%), 31 (12.9%) identified as black, three (1.2%) identified as “other,” and 67 (27.9%) did not report their race. We did not collect additional demographic data that might have influenced SHS exposure, such as education level, smoking status, income level, and employment status.
Data analysis. We believed that the use of flag reminders, visual reminders, education, and a monetary incentive would provide impetus to providers in the primary care setting to screen for and provide health promotion education on SHS exposure, and contribute to the inclusion of a permanent screening question with a flag reminder in our organization's EHR. The percentage of charts in which SHS screening was recorded significantly increased (χ2 = 523.92, P < 0.0001) from 14.3% (88/617) before implementation of the interventions, as noted in the retrospective chart review, to 98.8% (237/240) after implementation, as a result of the combined interventions in the QI initiative.
Patient questionnaire responses. Three-quarters (n = 178) of the patients reported current or previous SHS exposure. However, 83% (n = 197) also reported never having been screened or counseled by a clinical provider. Because of an error in data collection, we cannot report with accuracy the results of the postintervention patient questionnaire; however, approximately 19 out of 20 patients found the education and counseling informative, and approximately nine out of 10 considered adopting changes in behavior that would minimize their exposure.
Provider questionnaire responses. Of the seven participating providers, five (71%) reported they would continue to screen patients after the conclusion of the QI initiative, whereas two (29%) reported they would not. Four of the seven providers (57%) rated flag reminders as the most important influence on behavior change, suggesting the importance of the EHR pop-up alerts to changing clinical practice (see Figure 1). Five providers (71%) reported barriers to screening and education, including a lack of time during office visits, a perceived inapplicability of screening to a provider's clinical role, and clinical inertia (the failure to initiate or intensify treatment to achieve treatment goals). Those who reported that clinical inertia was a barrier relied heavily on the flag reminders to initiate screening.
Contextual barriers. Contextual elements that influenced the interventions included inefficient scheduling of patients’ clinic visits (including double booking or booking patients with different providers), an inadequate SHS screening question in the EHR, and a high volume of interruptions during office visits. In addition, patients’ use of the clinic as an urgent care center rather than a primary care facility negatively affected the efficacy of the interventions. Such factors reduced the amount of time clinicians could spend with patients and interrupted the continuity of care.
Key findings of the QI initiative included an association between the interventions implemented and a significant increase in the rates of SHS screening and disease prevention education. Five out of seven (71%) clinical providers planned to continue screening after the initiative was completed. While we expected that the monetary incentive might be the most effective intervention in changing providers’ behavior, in fact, flag reminders proved to be the most effective. Five out of seven clinicians reported continuing barriers to SHS screening, including lack of time, clinical inertia, and perceived inapplicability to the provider's clinical role.
Three-quarters (75%) of our patients reported current and previous SHS exposure, while 83% reported never having been screened or counseled by a clinical provider. Such findings are important and confirm the need for future research and initiatives involving implementation of interventions that promote behavior changes in both clinicians and patients.
Our hope that the positive results of the QI initiative would lead to the incorporation of improved SHS screening questions with flag reminders in the organization's EHR was not immediately fulfilled, because the organization decided to contract with a new EHR provider. However, the new EHR system does contain a screening question related to smoking history and SHS exposure.
Strengths of the project included staff and organizational engagement and cost-effective interventions. Screening questions with flag reminders incorporated into EHRs may facilitate clinical behavior change that improves health outcomes. The identification and reduction of barriers that stymie behavior change in clinical practice is essential to health promotion and disease prevention. Despite the minimization of barriers to screening and education, some clinicians said they would not continue to attempt behavior change. While the majority of screened patients found education useful, some did not and would not consider changing their behavior to minimize SHS exposure.
This initiative provided patients with health promotion and disease prevention education to prevent SHS exposure and subsequent health effects. Patients’ improved knowledge and awareness may lead to improved health outcomes and lower risk of disease. The initiative also highlighted the importance of eliminating barriers to behavior change among both clinicians and patients, improving EHR efficacy, and adopting flag reminders to improve screening and health promotion education. The need for additional research on clinical behavior change within primary care organizations and health care in general is warranted. There was no associated opportunity cost or trade-offs in implementing the initiative, because the time needed for screening and education was part of the regular office visit for those patients who had given their consent. The associated expense of the initiative was partially offset by the organization.
Limitations of the study included the small sample size. Because of an organizational concern that the time required to perform initiative-related activities might interfere with scheduled patient visits, no data were collected on why some patients declined to participate. The preimplementation chart analysis did not include a question on whether education was provided, because the existing screening question in the EHR did not address clinician-provided patient education. The clinicians who participated in the QI initiative could not record patients’ answers to our screening questions in the EHR, because the existing screening question was inadequate to our purposes, and organizational constraints made it impossible to change it. Despite some missing data on patients’ ages and ethnicities, the project design and outcomes were not negatively affected. This project did not determine whether behavior change was sustained over longer time periods.
Implications. The future of SHS screening and education depends on technological innovation, research, legislation, national guidelines, and clinical practice. Future research should center on initiatives for new and sustained behavior change by providers and patients regarding incentives and reminders. Research should also discern the impact of behavior change on health and economic outcomes within organizations and health care overall. Clinicians who work with high-risk populations, including minorities, the economically disadvantaged, pediatric patients, and those who live in multiunit housing, should be especially vigilant about screening for SHS exposure. The U.S. Preventive Services Task Force (USPSTF) recommends annual lung cancer screenings with low-dose computed tomography scans for adults from 55 to 80 years who have a 30-pack-per-year smoking history and who currently smoke or have quit within the last 15 years.22 The USPSTF should explore the development of similar guidelines for nonsmokers exposed to SHS. Researchers also need to develop reliable tools for screening SHS exposure.
Organizations and their information technology departments must collaborate to incorporate efficient screening questions with flag reminders in EHRs in primary care and other health care settings. The inclusion of screenings as a national quality measure for all patients, with government incentives to improve provider compliance, should also be considered by the Agency for Healthcare Research and Quality, the Centers for Medicare and Medicaid Services, and the USPSTF. Lastly, the development of new and comprehensive local, state, and federal legislation to limit public exposure is warranted.