Abstract
Objective
To determine the prevalence and predictors of receiving a smoking cessation medication prescription at discharge.
Methods
Retrospective analysis of ongoing Human Studies Committee-approved clinical trial data at large tertiary care center, The University of Kansas Medical Center. Patients included were smokers over 18, either Spanish or English speaking, those admitted between October 1, 2016 through May 31, 2018. Other eligibility criteria include access to a telephone or mobile phone, not currently be pregnant or breastfeeding, have no significant co-morbidity that precludes participation (acute, life-threatening illness, and communication barriers such as tracheal tube or altered mental status). Those included in this analysis were those randomized into the trial who expressed interest in receiving a smoking cessation medication prescription at discharge.
Results
Two hundred fourteen patients were recommended a prescription by their smoking cessation counselor, 88 patients (41.12%) were approved a prescription at discharge. Out of those approved, 50.70 (14.05 SD) was the average age, 12.84 (8.47 SD) was the average number of cigarettes used per day, 47 patients (53.41%) were White, 49 patients (55.68%) were admitted through the emergency department, 55 patients (62.50%) had used smoking cessation medication in the past, 49 patients (55.68%) had used inpatient smoking cessation, 36 patients (40.91%) had Medicaid. A binary logistic regression determined to show insurance status (P = 0.042) and use of inpatient smoking cessation medication use (P < 0.001) as statistically significant predictors of receiving a prescription at discharge.
Conclusion
It was determined that among the population recommended for medication, 41.12% actually received a prescription at discharge. The variables of “health insurance status” and “use of inpatient smoking cessation medication” demonstrated to be predictors of receiving a prescription. It is important to further study this as many patients rely on a prescription to afford these medications that are useful in a quit attempt.
Background
Most (70%) tobacco users want to quit but fewer than 10% succeed on any given quit attempt.1 An inpatient hospital stay provides a unique opportunity for patients who use tobacco to consider quitting while also providing patients with access to a range of health care staff able to provide various cessation services.2 According to the 2012 Cochrane review, evidence-based and effective cessation interventions that start in the hospital and continue for at least 1 month postdischarge promote smoking cessation.3 In addition, this meta-analysis found that adding nicotine replacement therapy (NRT) to intensive inpatient smoking cessation counseling can raise the odds of quitting by 50%.3 In the United States, the Joint Commission (JC) recommended performance measures for initiating treatment for tobacco users in the hospital and linking patients to cessation counseling and medications postdischarge. Unfortunately, because the JC performance measures are only recommendations and not required, many hospitalized tobacco users never receive evidence-based care.4 In the United Kingdom, a multidisciplinary project led by the Royal Pharmaceutical Society found that national-level quality initiatives exist to optimize discharge prescribing in order to improve transitions in care, but few focus on prescribing specifically for tobacco users who are trying to remain tobacco free after discharge.5 In Canada, the Ottawa Model for smoking cessation has been on the forefront of improving this transition of care, and outcome analyses have shown that effectively incorporating inpatient tobacco users algorithmically into counseling and postdischarge treatment for smoking cessation is cost accessible, results in lower rates of readmission, and improves long-term cessation rates.6
Some studies have examined prescribing for tobacco cessation during the inpatient stay. In Australia, a secondary analysis of a pharmacist-led smoking cessation intervention clinical trial in tertiary care hospitals found that patients admitted for respiratory and cardiac conditions, as well as heavy tobacco users, were more likely to receive inpatient medication.7 It also found that patients on numerous medications were just as likely to receive NRT, showing no evidence that regimen complexity or “pill burden” affected use of cessation medications. To date, no studies in the United States have described how many patients receive prescriptions for medication on discharge, to extend support into the critical month after hospitalization and enable patients to remain tobacco free.
Objective
The purpose of the current study is to describe the prevalence and determinants of cessation medication prescribing at discharge among hospitalized tobacco users. Understanding rates of cessation treatment prescribing at discharge is essential to determining how well hospitals are seizing the opportunity presented by a patient’s inpatient stay to provide a life-saving preventive intervention.
Methods
The present analysis includes all parent trial study participants admitted between October 1, 2016, and April 30, 2018. The parent study was designed to compare the effects of 2 different ways of offering tobacco treatment to patients: Opt Out versus Opt In. Parent trial study inclusion criteria included being a current smoker, older than 18 years of age, and a speaker of Spanish or English or both. To ensure participants could receive at least 1 form of cessation medication, eligibility for NRT was also included as an inclusion criterion. Counselors followed a physician-generated checklist with cautions or contraindications for cessation medications. For example, all participants had to be medically eligible to use NRT (i.e., no diagnosis of acute myocardial infarction, ST-elevation myocardial infarction, or other acute heart conditions). Patients with a seizure disorder or other contraindications for bupropion or varenicline were permitted to participate but were excluded from receiving a prescription for the contraindicated medication. Other eligibility criteria included having access to a telephone or mobile phone and not currently being pregnant or breastfeeding. Smokeless tobacco users were eligible to participate if they also smoked cigarettes. Eligible patients were randomized to parent study arms through a prompt on a handheld tablet after baseline patient information was collected. In the Opt Out arm, all patients received counseling and medications unless they indicated that they did not want tobacco treatment. In the Opt In arm, patients were first asked if they were planning to stay quit when they left the hospital, and only those patients who were planning on quitting and accepted outpatient medications as part of their treatment plan were included in the analysis. In the Opt Out arm, patients who accepted outpatient medications as part of their treatment plan regardless of their plans to stay quit when they left the hospital were also included.
The counseling content was the same across both study arms (for all patients in the Opt Out arm and for patients planning to stay quit in the Opt In arm). Counselors were certified Tobacco Treatment Specialists with bachelor’s or master’s degrees in counseling, social work, or other social services. Counseling included developing a manual-guided tobacco treatment plan for staying tobacco free after discharge by identifying motivators to quit, avoiding “triggers” to smoke, and obtaining or adhering to pharmacotherapy. Counselors provided information about all forms of Food and Drug Administration–approved first-line cessation medications (NRT, bupropion, and varenicline) and offered to arrange prescriptions for inpatient and outpatient cessation medications. Following the session, in accordance with a protocol created with inpatient prescribers, counselors sent prescription recommendations to the patient’s physician and inpatient pharmacy team for approval. This recommendation included a request for a “back up” request for another form of cessation medication, usually an NRT, in case the prescriber deemed the preferred medication to be medically not acceptable. The present analysis includes all parent trial study participants, regardless of arm, admitted between October 1, 2016, and April 30, 2018, who accepted the counselors’ offer to arrange a smoking cessation medication prescription at discharge.8
Measures
Data were collected at baseline as part of an inpatient interview performed by smoking cessation counselors and from chart review. The clinical trial included a standardized survey that recorded patient responses directly into a secure REDCap database.9
Measures included demographic information, tobacco use characteristics, and hospital stay data markers. Demographic information included age, sex, race, ethnicity (Hispanic or not Hispanic), education level, and health insurance type. Tobacco use characteristics included cigarettes per day, heaviness of smoking index (HSI), past use of smoking cessation medication (yes or no), and level of confidence in quitting or staying tobacco free (1–10–point scale). Hospital stay data markers included admission through the emergency department (ED) (yes or no), use of inpatient smoking cessation medication (yes or no), discharge day (weekend or weekday), length of stay, and time between end of counselor visit and time of discharge.
Analysis
The present project exported de-identified data sets from the REDCap database for analysis in IBM SPSS Statistics version 25 (Armonk, NY). Bivariate statistics compared demographic and other characteristics of patients who did and did not receive a prescription at discharge. Chi-square tests and independent sample t tests were performed for categorical and numerical outcomes, respectively. Missing data for analysis was excluded pairwise, as each bivariate test separately excluded data points that were missing from individual variables. Variables that were significantly associated with receipt of a prescription at the P < 0.1 level were a priori designated to be included in a logistic regression to determine predictors of receiving a prescription at discharge. Missing data was excluded listwise, and any case with missing data was excluded from the analysis. Additional variables that could potentially be associated with receipt of a prescription were also included in the logistic regression. These variables were dependence (HSI), previous medication use, ED admission, and confidence in ability to quit.
Results
A total of 214 inpatient tobacco users accepted counselor recommendation for discharge medication, including either NRT, varenicline, or bupropion. The population studied was mostly white (57%), had a mean age of 51, and used an average of 13.5 cigarettes per day. Most patients were admitted through the ED (54%) and the average inpatient stay was 105.7 hours.
Among participants, 35% were issued an inpatient prescription for smoking cessation medication. Less than half (41%) were issued a prescription at discharge. Descriptive statistics (Table 1) revealed in univariate testing that the variables of “Insurance type (P = 0.04)” and “Use of inpatient smoking cessation medication (P < 0.01)” were associated with receiving a prescription at discharge. Logistic regression (Table 2) revealed the variables of “Insurance type” (P = 0.03) and “Use of inpatient medication” (P = 0.001) remained predictors of receiving a prescription at discharge, after adjusting for other variables in the model. Patients with Medicaid and Medicare were more likely to receive a prescription than were patients with private or Veterans Affairs (VA) insurance, the reference variable.
Table 1.
Patient characteristics by prescription approval status
| Variable (N = 214 unless otherwise noted) | Total (N = 214) | Approved Rx at discharge (n = 88) | Not-approved Rx at discharge (n = 126) | P value |
|---|---|---|---|---|
| Demographic information | - | - | - | - |
| Age, y, mean (SD) | 51.03 (14.34) | 50.70 (14.05) | 51.25 (14.60) | 0.784 |
| Male, n (% of column) | 115 (53.74) | 43 (48.86) | 72 (57.14) | 0.232 |
| White, n (% of column) N = 210 | 120 (57.14) | 47 (53.41) | 73 (57.94) | 0.353 |
| Hispanic or Latino, n (% of column) N = 212 | 14 (6.60) | 5 (5.68) | 9 (7.14) | 0.675 |
| Education, some college or +, n (% of column) N = 197 | 92 (46.70) | 36 (40.91) | 56 (44.44) | 0.424 |
| Health insurance status | - | - | - | 0.042a |
| Medicaid, n (% of column) | 46 (21.50) | 25 (28.41) | 21 (16.67) | - |
| Medicare, n (% of column) | 81 (37.90) | 36 (40.91) | 45 (35.71) | - |
| Private or VA, n (% of column) | 61 (28.50) | 17 (19.32) | 44 (34.92) | - |
| No primary insurance, n (% of column) | 26 (12.15) | 10 (11.36) | 16 (12.70) | - |
| Tobacco usage characteristics | - | - | - | - |
| Cigarettes per day, mean (SD) | 13.48 (8.89) | 12.84 (8.47) | 13.93 (9.18) | 0.380 |
| Heaviness of smoking index, mean (SD) N = 212 | 2.39 (1.46) | 2.46 (1.32) | 2.34 (1.55) | 0.570 |
| Past use of smoking cessation medication, n (%) | 128 (59.81) | 55 (62.50) | 73 (57.94) | 0.503 |
| Confidence in quitting or staying quit N = 209 | 7.24 (2.68) | 7.28 (2.63) | 7.21 (2.73) | 0.868 |
| Hospital stay data markers | - | - | - | - |
| Admission through ED, n (% of column) | 116 (54.21) | 49 (55.68) | 67 (53.17) | 0.717 |
| Use of inpatient smoking cessation medication, n (% of column) | 75 (35.05) | 49 (55.68) | 26 (20.63) | < 0.001b |
| Weekend discharge, n (% of column) | 30 (14.02) | 12 (5.61) | 18 (8.41) | 0.893 |
| Length of stay, h, mean (SD) | 131.44 (131.36) | 101.73 (63.31) | 118.76 (129.98) | 0.744 |
| Time between end of counselor visit and time of discharge, h, mean (SD) | 86.78 (129.54) | 99.48 (63.33) | 115.68 (124.56) | 0.827 |
Abbreviations used: VA, Veterans Affairs; ED, emergency department; Rx, prescription.
Statistical significance P < 0.05.
Statistical significance P < 0.01.
Table 2.
Predictors of receiving a prescription at discharge
| Variable | Expβ | 95% CI for expβ | P value |
|---|---|---|---|
| Health insurance status | - | - | - |
| Reference: Private or VA | - | - | - |
| Medicaid | 3.49 | 1.42–8.60 | 0.007a |
| Medicare | 2.99 | 1.34–6.70 | 0.008a |
| No primary insurance | 2.33 | .77–7.09 | 0.136 |
| Inpatient medication used | 5.07 | 2.65–9.68 | < 0.001a |
Abbreviations used: VA, Veterans Affairs; Expβ, odds ratio.
Statistical significance P < 0.01.
Discussion
Less than half of inpatient tobacco users for whom counselors requested a prescription for smoking cessation medication at discharge actually received one. Despite the comprehensive coverage of cessation medications by VA benefits10 and Affordable Care Act (ACA) requirements for private plans to cover cessation medications,11 having VA or private insurance coverage reduced patients’ odds of receiving a prescription at discharge.
The majority (56%) of all patients who received a prescription at discharge had also been prescribed cessation medication during their inpatient stay. Medically managing patients with smoking cessation therapy during their inpatient stay can potentially facilitate providing a prescription for cessation medication at discharge. This finding is consistent with that of Regan et al.,12 that NRT use during the hospital stay was associated with self-reported NRT use after hospitalization. The relationship between inpatient and outpatient medication prescribing may be due to medication reconciliation, which is performed at many hospitals, including the study hospital, during discharge medication review. Although the process varies among institutions, in general terms the reconciliation process involves combining a patient’s preadmission and admission medication lists into a streamlined list of potential discharge medications for review and approval by providers prior to discharge.13
The model for medication reconciliation is constantly being improved to better identify gaps in care and improve efficiencies.14 Ferguson et al.13 evaluated an e-referral approach involving both inpatient and outpatient pharmacies to review and optimize medication changes between care settings and to assess and improve patient medication adherence. In this model, pharmacy staff contacts patients following discharge to discuss medication concerns and follow-up on medications that were added or discontinued in the patient’s regimen at discharge. If providers could ensure that smoking cessation therapy is prescribed while tobacco users are inpatients, this e-referral approach would carry over information on inpatient tobacco cessation treatment onto the discharge medication list and the agenda for the postdischarge discussion, which could have the added benefit of promoting adherence to cessation therapy during the postdischarge quit attempt.
Limitations
There were limitations in this secondary analysis. The parent trial is a population-based study based on a random sample of all hospitalized tobacco users, who are then randomized to receive Opt In versus Opt Out care for smoking cessation. Participants included all patients in both study arms who agreed to accept a prescription after discharge, so they may not be a representative sample of all inpatient tobacco users. In addition, not all hospitals have dedicated tobacco treatment services that work with inpatient prescribers to order scripts on discharge. This feature of our study may affect generalizability to community and rural inpatient facilities. In general terms the medication reconciliation process involves combining a patient’s preadmission and admission medication lists into a streamlined list of potential discharge medications for review and approval by providers prior to discharge13; exactly how this process occurs, however, varies widely and could have a different impact on prescription rates at different hospitals. Data were limited to those collected in the clinical trial and may not have included all variables that influence the likelihood of receiving a prescription. Follow-up information on whether patients actually filled their prescriptions is not available, nor are data on quit rates, as this is an ongoing and blinded clinical trial.
Conclusion
Even though tobacco use is a major preventable cause of death in the United States, most tobacco users leave the hospital without evidence-based care to help them quit. Based on these study findings, efforts to improve cessation medication delivery should focus on getting patients started on cessation medications during their hospital stay. Although somewhat limited in its potential impact on provider behavior, an additional potential strategy could be to inform providers of the expanded private insurance coverage of cessation medications prompted by the ACA and medication coverage provided by the VA. Many pharmaceutical companies offer prescription assistance programs to help smokers with no drug coverage to obtain medications at little or no cost; these programs, however, have complex enrollment procedures and are poor substitutes for insurance coverage.15 Pharmacist-led interventions have been shown to improve cardiovascular medication adherence through follow-up in-hospital counseling on medications, patient education prior to discharge, and telephone follow-up after discharge.16 These principles may be adapted for tobacco treatment, emphasizing the importance of a careful transition of care from hospital to community. Further studies are warranted to capitalize on hospitalization as an opportunity to help tobacco users quit by improving prescribing at discharge.
Acknowledgments
The authors thank the smoking cessation counselors for performing a role in resources by counseling and enrolling patients in the trial.
Funding: Research reported in this publication was supported by the National Institutes of Health from both R01 HL131512 and P30 CA168524.
Footnotes
Disclosure: The authors declare no relevant conflicts of interest or financial relationships.
Contributor Information
Vivek N. Patel, Department of Population Health, The University of Kansas Medical Center, Kansas City, KS.
Kimber P. Richter, Department of Population Health, The University of Kansas Medical Center, Kansas City, KS.
Laura M. Mussulman, Department of Population Health, The University of Kansas Medical Center, Kansas City, KS.
Niaman Nazir, Department of Population Health, The University of Kansas Medical Center, Kansas City, KS.
Byron Gajewski, Department of Biostatistics, The University of Kansas Medical Center, Kansas City, KS.
References
- 1.Babb S, Malarcher A, Schauer G, Asman K, Jamal A. Quitting smoking among adults –United States, 2000–2015. MMWR Morb Mortal Wkly Rep. 2017;65(52):1457–1464. [DOI] [PubMed] [Google Scholar]
- 2.Cassar MP. Smoking Cessation interventions in the cardiology ward. BMJ Qual Improv Rep. 2015;4(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Rigotti NA, Clair C, Munafo MR, Stead LF. Interventions for smoking cessation in hospitalised patients. Cochrane Database Syst Rev. 2012: CD001837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Faseru B, Turner M, Casey G, et al. Evaluation of a hospital-based tobacco treatment service: outcomes and lessons learned. J Hosp Med. 2011;6(4): 211–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Royal Pharmaceutical Society. Keeping patients safe when they transfer between care providers - getting the medicines right. Available at: https://www.rpharms.com/Portals/0/RPS%20document%20library/Open%20access/Publications/Keeping%20patients%20safe%20transfer%20of%20care%20report.pdf. Accessed September 20, 2018.
- 6.Mullen KA, Manuel DG, Hawken SJ, et al. Effectiveness of a hospital-initiated smoking cessation programme: 2-year health and healthcare outcomes. Tob Control. 2017;26(3):293–299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chui CY, Thomas D, Taylor S, et al. Factors associated with nicotine replacement therapy use among hospitalised smokers. Drug Alcohol Rev. 2018;37(4):514–519. [DOI] [PubMed] [Google Scholar]
- 8.Faseru B, Ellerbeck EF, Catley D, et al. Changing the default for tobacco-cessation treatment in an inpatient setting: study protocol of a randomized controlled trial. Trials. 2017;18(1):379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research Electronic Data Capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.U. S. Department. of Veterans Affairs. Mental health: tobacco and health.Available at: https://www.mentalhealth.va.gov/quit-tobacco/index.asp. Accessed November 11, 2018.
- 11.American Lung Association. Tobacco cessation treatment: what is covered?. Available at: https://www.lung.org/our-initiatives/tobacco/cessation-and-prevention/tobacco-cessation-treatment-what-is-covered.html. Accessed November 16, 2018.
- 12.Regan S, Viana JC, Reyen M, Rigotti NA. Prevalence and predictors of smoking by inpatients during a hospital stay. Arch Intern Med. 2012;172(21):1670–1674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ferguson J, Seston L, Ashcroft DM. Refer-to-pharmacy: a qualitative study exploring the implementation of an electronic transfer of care initiative to improve medicines optimisation following hospital discharge. BMC Health Serv Res. 2018;18(1):424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Holbrook A, Bowen JM, Patel H, et al. Process mapping evaluation of medication reconciliation in academic teaching hospitals: a critical step in quality improvement. BMJ, (Open). 2016;6(12):e013663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Leon-Salas A, Hunt JJ, Richter KP, Nazir N, Ellerbeck EF, Shireman TI. Pharmaceutical assistance programs to support smoking cessation medication access. J Am Pharm Assoc (2003). 2017;57(1):67–71.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cutrona SL, Choudhry NK, Fischer MA, et al. Modes of delivery for interventions to improve cardiovascular medication adherence. Am J Manag Care. 2010;16(12):929–942. [PMC free article] [PubMed] [Google Scholar]