Impact of smoking cessation counseling among acute myocardial infarction patients on post-hospitalization mortality rates: a systematic review

The researchers have adopted the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline¹⁴ to report this study. A systematic review of Medline (PubMed), ScienceDirect, CINAHL, and Google Scholar was conducted for studies on the impact of SCC for AMI patients on mortality rate. Search terms included SCC, smoking quit counseling, tobacco cessation counseling, mortality rate, death rate, AMI, and acute coronary syndrome. No limit on either English language or year of publication was applied. All possible combinations of the keywords were searched, and a manual literature search was further performed by reviewing studies referenced in retrieved papers. As an example of literature search, we can mention the PubMed search forming part of the present study, which was carried out using the following combinations of search terms: ((smoking cessation counseling) AND (acute myocardial infarction)) AND (mortality), ((smoking cessation counseling) AND (acute coronary syndrome)) AND (mortality rate), ((((smoking cessation counseling) OR (tobacco cessation counseling)) OR (smoking quit counseling)) AND (mortality rate)) AND (acute myocardial infarction), ((((smoking cessation counseling) OR (tobacco cessation counseling)) OR (smoking quit counseling)) AND (mortality rate)) AND (acute coronary syndrome), etc. All possible combinations of word search were done. The literature search yielded 100 abstracts that were further screened for eligibility based on this study's inclusion criteria.

Institutional review board approval was not required for this literature review. The investigation conforms to the principles outlined in the Declaration of Helsinki.¹⁴

The Population, Issue of interest, Comparison, Outcome, and Timeframe (PICOT) framework guided the question and inclusion criteria. This study review protocol was registered in Prospero during the period of the COVID crisis in 2020. The review protocol was registered automatically (registration No. CRD42020143071). Studies were included in the review if they met the following criteria. The population consists of adult patients with a confirmed diagnosis of AMI; the intervention is smoking or tobacco cessation or quit counseling during hospitalization; the comparison is patients who did not receive the SCC; the outcome is mortality (all-cause, specific cause, immediate and or delayed mortality); and the time is the immediate and delayed mortality of AMI patients. All study designs were included. Two investigators independently screened titles and abstracts identified by the searches to check their eligibility. If there were any disagreements between researchers and titles and abstracts didn’t provide sufficient information, then the full text was reviewed to check if the paper met inclusion criteria. Full texts of all publications that were decided to meet standards potentially were then examined to determine final inclusion. Any disagreement between reviewers was resolved by consensus or a third reviewer.

Each study was reviewed by 2 reviewers. Disagreement regarding data abstraction was resolved by consensus or a third reviewer. Data were summarized using narrative form based on elements of data abstraction. The abstracted data included the study purpose, designs, participants and population, primary and secondary outcomes, definitions of measures, results, and limitations.

Smoker patients were not routinely counseled to quit smoking during their post-AMI hospital stay.^15,16,18,19 The smoking cessation rate ranged between 33.9% and 52.1% in the reviewed studies. Please refer to Table 2 for more details of individual studies.

Studies had reported variation in characteristics of patients who received SCC. Patients who were less likely to receive SSC were non-white,^15,16,19 had diabetes mellitus, heart failure, were older patients,^15,16,18,19 had hypertension, had stroke,^15,19 were discharged to a skilled nursing facility,¹⁵ had a higher APACHE II score,¹⁶ and were women.¹⁵ On the other hand, patients who were most likely to receive SCC had a history of chronic obstructive pulmonary disease,^15,16,19 peripheral vascular disease,¹⁹ Killip class >2,¹⁹ treatment by revascularization [Percutaneous Coronary Intervention (PCI)/Coronary Artery Bypass Graft (CABG)] within 30 d, aspirin on admission for eligible patients, and beta-blockers on admission for qualified patients and fibrinolytic therapy.¹⁹

All 5 retrospective studies reviewed showed a reduction in AMI patients’ mortality after discharge from hospital among patients who received SCC compared with patients who did not receive SCC.^{14, 15, 16, 17, 18} Brown et al.¹⁶ had sampled 788 AMI smoker patients from 117 North Carolina care facilities in 2005 and reported a reduction of 5 years’ crude mortality among patients who were counseled (relative hazard 0.78, 95% confidence interval [CI] 0.63–0.97).¹⁶ A sample of 16,743 AMI smoker patients from 2971 acute care facilities in the US was studied by Houston et al.¹⁷ in 2005, and they found an improved 30 d mortality (1%), 60 d mortality (1.9%), and 2 years mortality (5%) among patients who received SCC.¹⁷

Moreover, Van Spall et al.¹⁹ investigated the 1-year mortality of AMI patients from 83 teaching and community hospitals in Ontario and a sample size of 9041 AMI patients. The authors documented a significant reduction of the 1-year mortality (hazard ratio 0.63, CI 0.44–0.9) among patients counseled to stop smoking.¹⁹ Bucholz et al.⁸ conducted a study in 2017 and investigated 13,815 AMI smoker patients from acute care non-governmental hospitals in the US. The investigators reported a lower mortality associated with 30 d (22.6%) and 17 years (7.5%) among patients who received SCC.⁸

Mohiuddin et al.¹⁸ conducted a randomized controlled study at a university-affiliated teaching hospital to compare the effect of intensive smoking cessation programs versus the standard smoking cessation counseling before patient discharge on mortality. The total number of participants was 209, among which 109 were categorized into intervention groups and 100 into usual care groups. The interventional group received SCC for 60 min/week for a minimum of 3 months, delivered by a trained tobacco cessation counselor. This study reported that smokers who received intensive counseling programs had a lower mortality rate (2.8%) than AMI smokers who received only verbal counseling upon discharge (12.0%).¹⁸ Data about the mortality rate risk reduction for each study are presented in Table 3.

The 2-year mortality rate risk reduction was reported only from 1 study and was 8%, with a hazard ratio of (0.92), and a 95% CI of (0.86–0.98).¹⁶ The 5 years mortality risk reduction was almost the same^{15, 19} and was reported as 22%, with a hazard risk of (0.782, 0.78), and 95% confidence intervals of (0.744–0.823) and (0.63–0.97) consecutively. At the same time, the long-term mortality of 17 years was only reported as 7.5%, with a hazard ratio of 0.925, and a 95% confidence interval of (0.893–0.959).¹⁹ Control for possible confounding variables was reported in all studies included in this systematic review.^15,16,18,19

Only 1 study included in this review was a randomized controlled trial. The other studies were cross-sectional retrospective reviews, and analyzed data from a national database. The authors have assessed the retrieved studies’ methodological features that might affect the estimates of the mortality rates among AMI patients. The assessment was done in both the study and the outcome. No formal risk of bias assessment was used because only 1 randomized controlled trial was used, and the reviewers’ decision at that time was to proceed with the consensus agreement of the risk of bias, considering the report²⁰ that the current biases reporting tools are lacking clear criteria for risk of bias assessment. Overall, all studies have described the sample population, determined the relevant inclusion and exclusion criteria, and defined the criteria of AMI and mortality rates. Only 1 study has detailed the data abstraction techniques adopted and its reliability assessment. Moreover, the randomized controlled trial was a single-center study that lacked a blindness protocol.

Our present review has identified several limitations of the included studies. The majority were non-randomized, observational studies with no control groups, had relied on health provider documentation of patient smoking habits and SCC, and contained no clear definition of smoking. All analyses were in the US and Canada. Moreover, we have searched only Medline (PubMed), ScienceDirect, CINAHL, and Google Scholar in this review. Other possible related publications may be present in other databases or even be not indexed.

Further, this review has focused on a single outcome measure, which is the mortality rate of AMI. One primary outcome of SCC is to quit smoking. However, this was not measured in all reviewed studies, and it was assumed that patients who received SCC were more likely to quit than a patient who didn’t have SCC.