Tobacco use continues to be the leading cause of preventable death in the United States and is a major public health issue (1–3). An estimated 45 million people living in the United States smoke cigarettes (4).
This problem is heightened among individuals with psychiatric disorders, who are about twice as likely to smoke tobacco and who are heavier smokers compared with the general population (5). These individuals also have greater adverse health outcomes (6) and are two to six times more likely to die from smoking-related diseases (7). For example, patients with schizophrenia who are smokers have 20% shorter life spans and experience higher standardized mortality rates for cardiovascular disease, respiratory disease, lung cancer, and infections than the general population (8). Among persons who have been treated for narcotics addiction, the death rate of smokers is four times that of nonsmokers (9). Efforts to provide smoking cessation interventions for these individuals are in immediate demand (10).
To effectively address smoking behaviors, states and other regulatory bodies, such as The Joint Commission (TJC), have implemented specific smoking regulations in hospitals. Since 1991, TJC has been setting smoking cessation standards for accredited hospitals (11), including psychiatric inpatient units (12,13), to support implementation of smoking bans. Federal agencies and independent membership associations have developed guidelines, tool kits, and resource materials (14–17). These supportive resources suggest that there are smoking cessation best practices that a hospital should implement. Given that hospitalized individuals with mental illness are engaged in treatment in a controlled environment, hospitalization can provide a “teachable moment” to address an individual’s smoking behaviors by using a variety of resources and treatment options (18).
Between 50% and 90% of individuals with mental illness and addiction disorders are tobacco dependent, and up to 200,000 of them die each year from smoking-related illnesses (19). Given these startling numbers, it is imperative that outpatient and inpatient treatment providers treat these individuals through the continuum of care, from intake through treatment planning and discharge. There is a strong link between quit rates and inpatient counseling, education, behavioral techniques, nicotine replacement therapies (NRTs), and discharge follow-up (18). There is also evidence that smoking cessation interventions “lower the cost per quality-adjusted life year,” demonstrating a cost-benefit to addressing smoking behaviors (20). However, there is still a significant gap between science and service in providing smoking cessation services to patients (21).
By 1994, more than 96% of U.S. general hospitals complied with TJC tobacco control standards (22). Even though a group of state-operated or state-supported psychiatric hospitals still allows smoking (23), many more are adopting smoke-free policies (24). Reports from 2006, 2008, and 2011 showed a continuing trend toward increased numbers of hospitals adopting smoke-free policies (41%, 49%, and 79%, respectively) (23,25–27). The substantial increase in hospitals adopting smoking bans presents an opportunity to evaluate current best practices for smoking cessation care associated with implementation of smoke-free policies. Because data were collected in 2008 and in 2011, there is also an opportunity to conduct a longitudinal analysis of each component of smoking cessation care and further explore changes in the level of smoking cessation care provided.
A recent study that identified the smoking policies and procedures in public psychiatric inpatient units in Australia found that a majority of patients who smoked received inadequate and inconsistent smoking care (28). The objectives of this research project were to assess the smoking policy and its implementation and describe the level of smoking cessation care in a national sample of state-operated or state-supported psychiatric inpatient hospitals.
Design, sample, and procedures
This longitudinal study analyzed paired data about smoking policies and practices submitted by a national sample of state-operated or state-supported psychiatric inpatient hospitals in 2008 and 2011. The data were collected by the Smoking Policies and Practices in State Psychiatric Hospitals Survey, developed in 2006 to identify the trend among hospitals in response to the general movement toward a nonsmoking environment (25). The survey collects information about the hospital, including the demographic characteristics of patients, current smoking policy, milieu issues, intake, education, treatment and aftercare planning, and outcomes and barriers related to enacting a smoke-free policy. It also provides a standard definition of smoking and of a hospital’s premises.
Hospitals were identified by using the National Association of State Mental Health Program Directors members’ directory. Psychiatric hospitals were excluded if they were serving only children younger than age 12, if the hospital was closed or had merged, or if the contact information for both the hospital’s director and quality assurance manager were not available.
An electronic version of the survey was distributed to the psychiatric hospitals’ directors and quality assurance managers in 2008 (N=219 hospitals) and 2011 (N=206 hospitals). In 2008, 164 (75%) hospitals completed the survey, and 49% (N=80) prohibited smoking. In 2011, 165 (80%) hospitals completed the survey, and 79% (N=134) prohibited smoking. Results from each survey have been independently published (23,27).
Characteristics of psychiatric hospitals.
Respondents provided the type of population served (youths 12–17 years; adults 18–64 years; geriatric patients ≥65 years; and forensic patients), the hospital’s size (number of beds), and type of care provided (acute or long term).
From a list of seven statements, hospitals identified the one that best described the current smoking policy. Based on the policy selected, hospitals were categorized as prohibiting or allowing smoking. Hospitals allowing smoking also provided information on how they control access to smoking, from designation of smoking areas to escorting of patients to smoking areas, use of privileges to access smoking areas, and establishment of smoking times.
One multilevel question addressed the types of specialty training provided to staff, including interactions of prescription medications and smoking, assessment of smoking use and dependence, medication treatments for smoking, awareness of quit lines, wellness counseling, counseling for smoking dependence, and coordination with community resources.
Management of access to smoking-related products.
Two questions addressed the use of smokeless tobacco products and retail sales of tobacco products on hospital grounds.
Level of smoking cessation care.
One question addressed the assessment of a patient’s smoking status at intake. Three questions addressed a range of treatments offered, resources available, and education provided on smoking-related topics. Two questions addressed follow-up with the next level of care provider.
Statistical analyses were conducted by using SPSS, version 17.0 (29). Approval was received from the National Association of State Mental Health Program Directors Research Institute, Inc., Institutional Review Board.
Hospitals that completed and returned the surveys in both 2008 and 2011 were classified as participating hospitals. Hospitals that completed and returned the surveys in only one of both years were classified as nonparticipating hospitals in the current study. Pearson’s chi square analysis and independent sample t tests tested for differences between participating and nonparticipating hospitals. Descriptive statistical analyses were calculated for the participating hospitals’ demographic characteristics.
McNemar’s chi square test examined the change from 2008 to 2011 in the proportion of hospitals providing specialty training, assessment of smoking status at intake, smoking cessation treatments offered, availability of educational resources, promotion of and education about the risks of smoking, and referral for antismoking care and inclusion of the patient’s smoking status in the aftercare plan.
A multilevel variable was created to analyze the level of smoking cessation care provided by hospitals in 2008 and 2011. Hospitals were categorized as providing best, good, average, or poor care. The best level of care included offering all types of treatments (smoking counseling, NRT, and pharmacotherapy); providing breadth of resources on smoking cessation, including at least two passive resources (educational pamphlets, referral to quit lines, peer support, and referral to quit-smoking Web sites) and two active resources (group sessions, individual sessions with clinical staff, and healthy lifestyle counseling); and including a referral for smoking cessation care in the aftercare plan (partial follow-up) plus indicating the patient’s smoking status in the aftercare plan (complete follow-up).
A good level of care offered any type of smoking treatment, provided any type of resources, and included complete or partial referral for follow-up. An average level of smoking cessation care offered any type of treatment, provided any type of resources, and included no referral for follow-up. A poor level of smoking cessation care offered no smoking cessation treatment, provided any type of resources, and included no follow-up.
More participating (N=108) hospitals than nonparticipating hospitals (N=57) served forensic (χ2=4.01, df=1, p<.05) and multiple types of populations (χ2=5.11, df=1, p<.05).
Characteristics of psychiatric hospitals
Data from 108 hospitals representing 35 states and the District of Columbia were used for paired analysis. More hospitals were located in the South (N=38, 35%) than in the Northeast (N=34, 32%), the Midwest (N=26, 24%), and the West (N=10, 9%). A majority of hospitals (N=73, 68%) were located in rural areas. There were no significant regional differences in terms of hospitals’ participation in the survey and in the overall smoking cessation care provided to patients. [A map of participating hospitals is available online as a data supplement to this article.]
More hospitals served adult populations in 2008 (81%) than in 2011 (73%), but the proportion of hospitals serving forensic patients rose from 47% in 2008 to 60% in 2011. Hospitals served multiple types of populations in both years. A majority of hospitals that served youths and adults in 2008 (N=97) and 2011 (N=87) provided acute care (36% and 38%, respectively) or a combination of acute and long-term care services (35% and 39%, respectively). Table 1 summarizes the demographic characteristic of hospitals by year surveyed.
Table 1Characteristics of 108 state-operated or state-supported psychiatric hospitalsa
| Add to My POL
|Number of populations served|
| Data missing||3||2||0||—|
|Type of carec|
| Long term||28||29||20||23|
| Acute and long term||34||35||34||39|
| Prohibit smoking||52||48||90||83|
| Allow smoking||52||48||18||17|
| Data missing||4||4||0||—|
Significantly fewer hospitals were categorized as prohibiting smoking in 2008 (48%) than in 2011 (83%) (χ2=15.82, df=1, p<.05). Among hospitals that allowed smoking, the most common methods to control access continued to be designated smoking areas, escort to smoking areas, and established smoking times.
Overall, 61% of hospitals in 2008 and 65% of hospitals in 2011 provided training for hospital staff (Table 2).
Table 2Specialty training in smoking cessation care provided to staff at 108 inpatient psychiatric hospitalsa
| Add to My POL
|Prescription medication interaction with smoking||47||44||47||44||—|
|Assessment of smoking use and dependence||42||39||50||47||.33|
|Medication treatment for smoking||44||41||45||42||—|
|Awareness of quit lines||21||20||27||25||.42|
|Counseling for smoking dependence||37||35||36||34||—|
|Coordination with community resources||20||19||26||24||.41|
Management of access to smoking-related products
More hospitals did not allow the use of smokeless tobacco products in 2011 (N=98, 91%) than in 2008 (N=83, 77%) (χ2=.98, df=1, p<.05). Fewer hospitals did not allow retail sales of smoking tobacco products on hospital premises in 2008 (N=86, 80%) than in 2011 (N=104, 96%) (χ2=.05, df=1, p<.05).
Level of smoking cessation care
Table 3 shows the level of smoking cessation care provided by hospitals. Nearly all hospitals assessed smoking status at admission. NRT continued to be the treatment provided by more hospitals, and in 2011 a slight majority of hospitals provided all types of treatment (smoking counseling, NRT, and pharmacotherapy). Although the percentage of hospitals providing resources on smoking cessation increased from 2008 to 2011, only 42% of hospitals in 2011 offered breadth of options. There was a significant increase in hospitals making referrals and including the patient’s smoking status in the aftercare plan, but only 20% of hospitals had adopted this best practice in 2011. The number of hospitals providing no follow-up on smoking cessation significantly improved, dropping from 64% to 41% (p=.05). From 2008 to 2011, fewer hospitals provided average smoking cessation care (56% and 35%, respectively, p=.05), and more hospitals provided good smoking cessation care (22% and 43%, respectively, p=.05), including follow-up and breadth of resources.
Table 3Smoking cessation care provided by 108 inpatient psychiatric hospitals
| Add to My POL
|Assessment of smoking status at intake||94||87||106||98||.01|
| Smoking counseling||65||60||84||78||.01|
|Number of treatments|
| Data missing||1||0||1||0||—|
|Resources on smoking cessation|
| Passive onlyc||9||8||6||5||.61|
| Active onlyd||21||19||17||16||.61|
| Passive and active||74||69||81||75||.31|
| At least two passive and two active||34||32||45||42||.19|
| Data missing||1||1||1||1||—|
|Promotion of and education about smoking risks|
| Treatment planning||65||60||76||70||.13|
| Only referral in aftercare plan||24||22||20||19||.74|
| Only smoking status in aftercare plan||5||5||17||16||.05|
| Referral and smoking status in aftercare plan||5||5||22||20||.05|
| Data missing||5||4||5||4||—|
|Level of care|
| Data missing||5||4||5||4||—|
The number of hospitals prohibiting smoking increased 73% (from 48% to 83%) in a three-year period, demonstrating that psychiatric inpatient hospitals are actively enacting smoke-free policies. This remarkable increase in smoking bans provides confidence that certain long-held beliefs—such as smoking is an acceptable cultural norm for individuals with mental illness (30), smoking provides a therapeutic effect, (31), and individuals with mental illness cannot or prefer not to quit smoking (32)—have less influence over current patient care. It also makes evident that hospitals are responding to the health threats of smoking because of the increase in the number of hospitals enacting smoke-free policies that have an impact on patient, staff, and visitors. Many more hospitals have restricted the use of smokeless tobacco products as an alternative to smoking, and more hospitals have prohibited the sale of smoking products on hospital grounds.
Hospitals that continue to allow smoking use staff resources to support this unhealthy behavior. Staff resources are required for escorting patients to smoking areas and supervising smoking areas. Providing organized smoking times may encourage smoking and reinforces nicotine addiction through continuous dosing. Secondhand smoke is also a health hazard for nonsmoking staff and patients.
Although there was a marked increase in nonsmoking policies, there was no significant increase in the proportion of hospitals providing overall training to staff related to smoking cessation care. Less than half of the hospitals provided specialty training to staff. Staff training has been correlated with successful smoking bans. A survey of mental health care staff found that they are significantly less positive toward smoking-related policies and treatments than their counterparts in other areas of health care (33). Without staff support, smoking bans are less likely to be enforced.
An increase was found in the number of hospitals that assess smoking status at admission, provide a variety of smoking cessation treatments (including smoking counseling, NRTs, and pharmacotherapies), and educate patients during treatment planning about the risk of smoking. However, although 98% of hospitals in 2011 assessed patients’ smoking status at admission, only 42% took time during the intake evaluation to discuss the risks of smoking. When tobacco use is assessed as part of the intake process, the hospital begins a dialogue with the patient that helps in the prompt diagnosis of nicotine addiction and in the delineation of the best smoking cessation treatment possible (34). Therefore, intake provides an opportunity for process improvement at the hospital level that could translate into better health outcomes for patients who smoke. High rates of general medical and mental comorbidity have been found among patients with mental illness. Managing risk factors for cardiovascular disease and other physical conditions, such as smoking, is a first step to better health outcomes.
There was a marked increase in the proportion of hospitals providing complete follow-up (making referrals and including the patient’s smoking status at discharge); however, they constituted only 20% of hospitals in 2011. Coincidentally, only 24% of hospitals in 2011 provided training in coordination with community resources for smoking cessation.
The current survey found a collective belief that staff members do not refer patients to smoking cessation treatments or do not include patients’ smoking status at discharge because the patients have been smoke-free while hospitalized. This finding underscores the urgent need for staff education on the important role of community and outpatient smoking cessation programs in sustaining smoking abstinence. In fact, after discharge it takes only five weeks for patients to return to smoking (35). Although hospitals may educate patients during their hospital stay on the benefits of not smoking, high relapse rates suggest that hospitals should be more active in promoting smoking cessation treatment throughout the continuum of care. Staff training and best practices should be integrated to support patients to achieve longer-term smoking cessation after discharge from inpatient care (36).
It has been suggested that inpatient programs should include outpatient follow-up by face-to-face contacts or phone calls to encourage smoking cessation for at least one month after discharge or referral to community-based smoking cessation programs that offer ongoing counseling, support, and NRT and that the aftercare plan should include these recommendations (37). The rates of smoking resumption after discharge from hospitals that do not transmit patients’ smoking status to the next level of care provider should be explored to determine the effect of information referral on relapse rates.
Finally, there has been an increase in the number of hospitals providing good smoking cessation care and a corresponding decrease in the number of hospitals providing average care. Average care was defined as the provision of a minimum of inpatient treatments to address smoking. Good care was defined as the provision of any type of smoking treatment and follow-up planning. However, in 2011, only 12% and 43% of hospitals provided best and good smoking cessation care, respectively; and 35% of hospitals provided average care. Although hospitals are actively implementing smoking cessation policies, there is still a gap in the translation of policy into practice. Providing inadequate training to staff about the delivery of smoking cessation care and appropriate follow-up at discharge lowers the quality of smoking cessation care provided in state-operated or supported psychiatric inpatient hospitals.
This study was limited, first, by the loss of participating hospitals from 2008 to 2011. Sixty-five percent (N=108) of 2011 participating hospitals also participated in the 2008 survey (N=165); however, the smoking status of participating and nonparticipating hospitals was not significantly different, suggesting an equal loss from both groups.
Second, the study focused on state-operated or state-supported psychiatric inpatient hospitals. Smoke-free policies of general hospitals with psychiatric units and private free-standing psychiatric hospitals were not assessed. However, the study group provided valuable information on the conversion to smoke-free environments and the needs for professional development to provide breadth of treatment and continuum of care.
Finally, survey questions about practice addressed availability and not volume. Given low levels of staff training, the actual penetration and utilization of available services and resources to patients should be further investigated.
Although smoking policy implementation has increased in a three-year period, there is opportunity to improve the quality of smoking cessation treatment. Hospitals may look at the continuum of care to determine how best to improve their services while transitioning patients to the next care provider. Beginning shortly after admission, hospitals may align services provided during the inpatient stay with those that will be available in outpatient care. These services and the treatment goals must be clearly stated in the aftercare plan. In addition, hospitals may wish to establish links with peer support and wellness clinics, resources that are readily available in the outpatient treatment community and that may be necessary for patients to maintain their nonsmoking status. Hospitals that seek to make these resources available to patients during the inpatient stay may require training to provide these services and to coordinate with the outpatient treatment community.
This research study was fully funded by grant 047139 from the University of San Francisco’s Smoking Cessation Leadership Center. The authors thank Vera Hollen, M.A., for her careful review of the drafts.
The authors report no competing interests.