STRUCTURED ABSTRACT
Objective:
To assess associations between co-occurring preoperative smoking and risky alcohol use on the likelihood of adverse surgical outcomes.
Summary Background Data.
Risky alcohol use and smoking are known surgical risk factors with a high co-occurrence and additive adverse effects on multiple organ systems that impact surgical health, yet no research has evaluated the impact of co-occurrence on surgical outcomes.
Methods:
This investigation analyzed 200,816 patients from the Michigan Surgical Quality Collaborative database between 7/1/2012 to 12/31/2018. Patients were classified based on past year risky alcohol use (>2 drink/day) and cigarette smoking into four groups: 1) risky alcohol and smoking, 2) risky alcohol only, 3) smoking only, and 4) no risky alcohol/smoking. We fitted logistic regression models, applying propensity score weights incorporating demographic, clinical, and surgical factors to assess associations between alcohol and smoking and 30-day postoperative outcomes; surgical complications, readmission, reoperation, and emergency department (ED) visits.
Results:
Risky alcohol and smoking, risky alcohol only, and smoking only were reported by 2,852 (1.4%), 2,840 (1.4%), and 44,042 (22%) patients, respectively. Relative to all other groups, the alcohol and smoking group had greater odds of surgical complications, readmission, and reoperation. Relative to the no alcohol and smoking group, the alcohol only group higher odds of reoperation and smoking only group had higher odds of ED visits.
Conclusions:
The combination of smoking and risky drinking conferred the highest likelihood of complications, readmission, and reoperation before surgery. Co-occurring alcohol and smoking at the time of surgery warrants special attention as a patient risk factor and deserves additional research.
Mini Abstract
General surgery patients who smoke cigarettes and drink alcohol (>2 drinks/day) before surgery have higher odds of experiencing adverse surgical outcomes compared to patients who only smoke, only drink, or do neither.
Introduction
Each year 14% of Americans meet criteria for an alcohol use disorder and 13% smoke cigarettes.1,2 In surgical populations, the prevalence of these behaviors is higher, with as many as one in four surgical patients smoking and/or drinking alcohol at risky levels.3,4 In fact, smoking and risky alcohol use (defined as consuming >2 drinks per day) prior to surgery are two of the most common risk factors for poor outcomes after surgery.5 Individuals who smoke cigarettes have increased odds of general infections, unplanned intubation, cardiopulmonary complications, surgical site infections, and other adverse outcomes following surgery.6–12 Risky alcohol use is associated with a nearly two-fold increase in the likelihood of postoperative infections, pulmonary complications, prolonged hospital stay, and admission to the intensive care unit.13,14 Smoking- and alcohol-associated surgical complications are not specific to certain operations or subpopulations, but are evident across a range of operations even after controlling for relevant covariates.6,13–16 The causal relationship between smoking, alcohol consumption, and surgical complications is further evidenced by studies linking short-term preoperative abstinence to lower likelihood of surgical complications in randomized controlled trials.17,18
Although numerous studies have evaluated the relationship between smoking or risky alcohol use on surgical outcomes in isolation, no study has evaluated surgical outcomes among those who both drink alcohol at risky levels and smoke cigarettes. Such an analysis is especially important given the very high co-occurrence of smoking and risky alcohol use19–21 and well-documented additive or even multiplicative deleterious effects of co-use on multiple organ systems that impact surgical health.22–25 Knowledge of the combined impact of risky alcohol use and smoking before surgery has the potential to aid conversations between patients and surgical teams when considering the risks of undergoing operations as well as to potentially motivate interventions to encourage reduction and abstinence of both behaviors.
This study used quality improvement data from a state-wide consortium (Michigan Surgical Quality Collaborative; MSQC), to compare the incidence of adverse post-operative outcomes among general surgical patients who drink at risky levels (> 2 drinks/day), smoke tobacco cigarettes in the past year, do both, or do neither. Outcomes include presence of ≥ 1 surgical complication, post-operative readmission, reoperation, or emergency department (ED) visit in within 30 days of surgery. We hypothesized that individuals who both smoke and drink at risky levels would have a higher likelihood of adverse surgical outcomes and events, relative to the individuals who do not smoke tobacco or drink more than 2 drinks a day or do either in isolation, and this likelihood would be highest among individuals who both smoke and drink alcohol at risky levels.
Methods
Patients who underwent general surgeries included in the MSQC registry from 7/1/2012 to 12/31/2018 were included in this retrospective study. MSQC is a sanctioned Patient Safety Organization with the Agency for Healthcare Research and Quality. It includes 69 hospitals in the state of Michigan. At each site, trained nurse reviewers collect data related to patient characteristics, intraoperative processes of care, and 30-day postoperative patient outcomes utilizing a sampling algorithm that minimizes selection bias in accordance with established policies and procedures. Regular data audits ensure registry data quality and validity.26,27 Data collection for MSQC is Institutional Review Board (IRB) exempt at participating hospitals, and the current study was reviewed and considered exempt by the Michigan Medicine IRB.
Cohort selection
In total, there were 272,820 adult surgical patients identified in the MSQC database between 7/1/2012 to 12/31/2018 who had a complete 30-day follow up (Figure 1). After excluding those with underwent hysterectomy surgery, vascular surgery, or had missing data, 200,816 general surgery cases remained and comprised the study cohort. This included all general surgeries collected through MSQC representing eleven procedure groups: appendectomy, cholecystectomy, bowel, esophagectomy, gastrectomy, hepatectomy, hernia, pancreatectomy, thyroidectomy, adrenalectomy, and splenectomy. (See Supplemental Table 1 for all surgery CPT codes).Hysterectomy was excluded because it is performed on individuals of female sex only, and the primary analysis adjusted for sex as a covariate. Vascular surgery was excluded because of the differential effects of smoking and risky alcohol use on cardiovascular surgeries that may differ from general surgeries.
Figure 1.
Cohort Flowchart
Variables
The primary dependent variables included: 1) risky alcohol use, defined as >2 drinks/day in the two weeks prior to admission; 2) smoking; defined as any tobacco cigarette smoking in the past year; 3) both smoking and risky alcohol use, and 4) no risky alcohol/smoking. Presence/absence of smoking and risky alcohol use determinations were made by trained MSQC chart reviewers based on review of patient electronic health records. Tobacco use is defined by MSQC as tobacco cigarette smoking only and excludes cigars, electronic cigarettes, and marijuana.
Covariates included patient sex, race, age group, insurance status, American Society of Anesthesiologists (ASA) classification, comorbidities (obese, diabetes, chronic obstructive pulmonary disease (COPD), sleep apnea, hypertension, transfusion in 72 hours prior to surgery, renal failure requiring dialysis, steroid, sepsis, peripheral vascular disease, bleeding disorder, 10% loss of body weight 6 months prior to surgery, coronary artery disease, ascites, congestive heart failure, pneumonia), open wound, wound classification, surgical approach and procedure group.
Outcomes
We evaluated four surgical outcomes: surgical complications, reoperation, readmission, and ED visits.
Surgical Complications:
Surgical complications were defined as presence of one or more complications in the 30 days following surgery (excluding those present prior to surgery). The complications included cardiac arrest requiring cardiopulmonary resuscitation, cardiac dysrhythmias, C-difficile infection, central line-associated bloodstream infection, deep vein thrombosis requiring therapy, myocardial infarction, pulmonary embolism, pneumonia, sepsis, septic shock, severe sepsis, deep incisional surgical site infection (SSI), organ/space SSI, superficial incisional SSI, stroke/cerebrovascular accident, transfusions within first 72 hours postop, unplanned intubation, catheter associated urinary tract infection, and non-catheter associated urinary tract infection.
Reoperation:
Reoperation was defined as returning back to the operating room after the principal procedure, for any reason, related to the principal procedure or not.
Readmission:
Readmission was defined as an inpatient (surpasses 2 consecutive midnights) acute care hospitalization at any hospital within 30 days of surgery.
ED visit:
ED visits were defined as any visits where patient was treated and then discharged directly from the ED, urgent care clinic, or observation unit after being there less than 2 midnights within 30 days of operation date.
Statistical Analyses
First, descriptive statistics were calculated for demographic and clinical characteristics. Chi-square tests were used to compare the categorical variables among the four smoking/alcohol groups. Next, logistic regression models with the smoking and risky alcohol use groups as the only independent variable were fitted to estimate the odds of the different surgical outcomes of interest (surgical complication, readmission, reoperation, ED visit). Lastly, we fitted the logistic regression models applying propensity score weights.28,29 The goal of using propensity score weighting was to better balance the four groups on demographic variables, comorbidities, surgical approach, and procedure group.30 Propensity scores were derived from a multinomial logistic regression model, which regressed the four groups on covariates listed above. From this model, propensity score weights were then calculated as the inverse of the propensity scores, and also reflected the sample size for each of the four groups. In primary analyses, the no risky alcohol/smoking group was used as the reference. We also repeated the models with smoking only and risky alcohol only, respectively, as reference groups to test whether estimates for the risky alcohol and smoking group were significantly different from the estimates for these two groups. All statistical tests were 2-tailed and statistical significance was set at p<=0.05. Analyses were performed with SAS 9.4.
Results
Among the final cohort of general surgical patients, 2,840 (1.4%) were classified with risky alcohol use only, 44,042 (21.9%) with smoking only, 2,852 (1.4%) with both risky alcohol and smoking, and 151,082 (75.2%) with neither smoking or risky alcohol use.
The study cohort was majority female (54.4%), Non-Hispanic white (75.3%), and the median age was 56 (Table 1). Surgical priority groups included elective (63.8%), urgent (18.2%), and emergent (18.1%). The most common procedure types included cholecystectomy (28.9%) and hernia repair (26.8%) (Supplemental table 2). All participant and surgery characteristics and comorbidity prevalence differed significantly between risky alcohol and smoking groups. Incidence of 30-day surgical complications, hospital readmission, reoperation and ED visits are presented in Table 2.
Table 1.
Characteristics of Study Sample
| Overall Cohort n=200,816 |
No Risky Alcohol or Smoking n=151,082 (75.2%) | Risky Alcohol Only n=2,840 (1.4%) | Smoking Only n=44,042 (21.9%) | Risky Alcohol and Smoking n=2,852 (1.4%) | P-value | |
|---|---|---|---|---|---|---|
| Female | 109344 (54.4) | 84300 (55.8) | 525 (18.5) | 23789 (54.0) | 730 (25.6) | <0.0001 |
| Race/Ethnicity | <0.0001 | |||||
| Non-Hispanic White | 151187 (75.3) | 114643 (75.9) | 2293 (80.7) | 32240 (73.2) | 2011 (70.5) | |
| Other | 49629 (24.7) | 36439 (24.1) | 547 (19.3) | 11802 (26.8) | 841 (29.5) | |
| Age Group | <0.0001 | |||||
| <45 | 56104 (27.9) | 39030 (25.8) | 363 (12.8) | 16113 (36.6) | 598 (21.0) | |
| 45–64 | 81551 (40.6) | 57584 (38.1) | 1350 (47.5) | 20842 (47.3) | 1775 (62.2) | |
| 65+ | 63161 (31.5) | 54468 (36.1) | 1127 (39.7) | 7087 (16.1) | 479 (16.8) | |
| Insurance | <0.0001 | |||||
| Commercial Insurance (non-HMO) | 71009 (35.4) | 56104 (37.1) | 1043 (36.7) | 13076 (29.7) | 786 (27.6) | |
| Other | 129807 (64.6) | 94978 (62.9) | 1797 (63.3) | 30966 (70.3) | 2066 (72.4) | |
| Weight | <0.0001 | |||||
| Underweight | 3887 (1.9) | 2237 (1.5) | 36 (1.3) | 1448 (3.3) | 166 (5.8) | |
| Normal | 45877 (22.8) | 32093 (21.2) | 594 (20.9) | 12068 (27.4) | 1122 (39.3) | |
| Overweight | 61534 (30.6) | 46932 (31.1) | 1114 (39.2) | 12649 (28.7) | 839 (29.4) | |
| Obesity | 89518 (44.6) | 69820 (46.2) | 1096 (38.6) | 17877 (40.6) | 725 (25.4) | |
| ASA Group | <0.0001 | |||||
| ASA 3 or 4 or 5 | 88993 (44.3) | 66539 (44.0) | 1403 (49.4) | 19411 (44.1) | 1640 (57.5) | |
| ASA 1 or 2 | 111823 (55.7) | 84543 (56.0) | 1437 (50.6) | 24631 (55.9) | 1212 (42.5) | |
| Diabetes | 28122 (14.0) | 22704 (15.0) | 275 (9.7) | 4959 (11.3) | 184 (6.5) | <0.0001 |
| Chronic obstructive pulmonary disease | 15196 (7.6) | 9019 (6.0) | 225 (7.9) | 5517 (12.5) | 435 (15.3) | <0.0001 |
| Sleep Apnea | 34675 (17.3) | 27258 (18.0) | 613 (21.6) | 6362 (14.4) | 442 (15.5) | <0.0001 |
| Hypertension | 89441 (44.5) | 70185 (46.5) | 1634 (57.5) | 16315 (37.0) | 1307 (45.8) | <0.0001 |
| Preop transfusion any RBCs within 72 hours prior to surgery | 2405 (1.2) | 1791 (1.2) | 56 (2.0) | 465 (1.1) | 93 (3.3) | <0.0001 |
| Dialysis | 1775 (0.9) | 1374 (0.9) | 17 (0.6) | 348 (0.8) | 36 (1.3) | 0.0050 |
| Steroids | 7376 (3.7) | 5793 (3.8) | 78 (2.7) | 1429 (3.2) | 76 (2.7) | <0.0001 |
| Preoperative sepsis | 15036 (7.5) | 10782 (7.1) | 285 (10.0) | 3527 (8.0) | 442 (15.5) | <0.0001 |
| Peripheral Vascular Disease | 4283 (2.1) | 2883 (1.9) | 77 (2.7) | 1205 (2.7) | 118 (4.1) | <0.0001 |
| Bleeding disorder | 7606 (3.8) | 5825 (3.9) | 153 (5.4) | 1471 (3.3) | 157 (5.5) | <0.0001 |
| 10% loss of body weight 6 months prior to surgery | 3845 (1.9) | 2540 (1.7) | 48 (1.7) | 1161 (2.6) | 96 (3.4) | <0.0001 |
| Coronary Artery Disease | 23599 (11.8) | 18396 (12.2) | 426 (15.0) | 4442 (10.1) | 335 (11.7) | <0.0001 |
| Ascites | 1864 (0.9) | 1228 (0.8) | 57 (2.0) | 446 (1.0) | 133 (4.7) | <0.0001 |
| Congestive Heart Failure | 1589 (0.8) | 1239 (0.8) | 29 (1.0) | 281 (0.6) | 40 (1.4) | <0.0001 |
| Pneumonia | 1377 (0.7) | 972 (0.6) | 27 (1.0) | 310 (0.7) | 68 (2.4) | <0.0001 |
| Open wound with or without infection | 3294 (1.6) | 2388 (1.6) | 52 (1.8) | 772 (1.8) | 82 (2.9) | <0.0001 |
| Wound classification | ||||||
| Clean | 49155 (24.5) | 36094 (23.9) | 774 (27.3) | 11290 (25.6) | 997 (35.0) | <0.0001 |
| Contaminated/Infected | 151661 (75.5) | 114988 (76.1) | 2066 (72.7) | 32752 (74.4) | 1855 (65.0) | |
| Surgical Approach | <0.0001 | |||||
| Open | 78977 (39.3) | 64655 (42.8) | 1434 (50.5) | 19158 (43.5) | 1616 (56.7) | |
| Minimally invasive | 121839 (60.7) | 86427 (57.2) | 1406 (49.5) | 24884 (56.5) | 1236 (43.3) | |
| Surgical Priority | <0.0001 | |||||
| Emergent | 36261 (18.1) | 26260 (17.4) | 598 (21.1) | 8581 (19.5) | 822 (28.8) | |
| Urgent | 36498 (18.2) | 26986 (17.9) | 501 (17.6) | 8397 (19.1) | 614 (21.5) | |
| Elective | 128057 (63.8) | 97836 (64.8) | 1741 (61.3) | 27064 (61.5) | 1416 (49.6) |
Note: HMO= Health maintenance organization, RBC= Red Blood Cell
Table 2.
Incidence of 30-day postoperative outcomes by smoking and risky alcohol use status
| Overall Cohort n=200,816 | No Risky Alcohol or Smoking n=151,082 (75.2%) |
Risky Alcohol Only n=2,840 (1.4%) | Smoking Only n=44,042 (21.9%) |
Risky Alcohol and Smoking n=2,852 (1.4%) | |
|---|---|---|---|---|---|
| Reoperation | 8566 | 6060 (4.0) | 154 (5.4) | 2100 (4.8) | 261 (9.2) |
| Readmission | 13358 | 9822 (6.5) | 193 (6.8) | 3081 (7.0) | 262 (9.2) |
| Surgical Complications | 20568 | 15275 (10.1) | 364 (12.8) | 4386 (10.0) | 543 (19.0) |
| Emergency Department visit | 17279 | 11872 (7.9) | 217 (7.6) | 4907 (11.1) | 283 (9.9) |
Note: Readmission and Emergency Department visit wasn’t specified to be related to the principal surgery prior to 2020.
Reoperation
In the propensity score weighted model (Figure 2), the odds of reoperation were significantly higher in the risky alcohol and smoking group (OR 1.77, 95% CI 1.53–2.06) relative to the no risky alcohol or smoking group (Table 3). The estimate for the risky alcohol and smoking group was also significantly different from either of the estimates for the smoking only or risky alcohol only groups. The odds of reoperation were significantly higher in the risky alcohol only group (OR 1.29, 95% CI 1.09–1.53), and smoking only group (OR 1.15, 95% CI 1.10–1.22) in the propensity score weighted model.
Figure 2.
Odds of adverse surgical outcomes by alcohol and smoking status
Table 3.
Unadjusted and propensity score adjusted outcomes by risky alcohol and smoking status (Reference group: No risky alcohol/smoking)
| Unadjusted Models | Propensity Score Weighted Models | |||||
|---|---|---|---|---|---|---|
| Risky Alcohol and Smoking | Risky Alcohol Only | Smoking Only | Risky Alcohol and Smoking | Risky Alcohol Only | Smoking Only | |
| Reoperation | ||||||
| Odds Ratio | 2.412 | 1.371 | 1.198 | 1.773 *† | 1.291 | 1.154 |
| 95% CI | [2.118, 2.746] | [1.163, 1.616] | [1.138, 1.260] | [1.529, 2.057] | [1.089, 1.531] | [1.096, 1.216] |
| P-value | <.0001 | <.0001 | <.0001 | <.0001 | 0.003 | <.0001 |
| Readmission | ||||||
| Odds Ratio | 1.457 | 1.049 | 1.082 | 1.277 *† | 1.038 | 1.026 |
| 95% CI | [1.281, 1.657] | [0.905, 1.215] | [1.037, 1.128] | [1.115, 1.463] | [0.895, 1.204] | [0.982, 1.071] |
| P-value | <.0001 | 0.528 | <.0001 | <.0001 | 0.622 | 0.249 |
| Surgical complication | ||||||
| Odds Ratio | 2.091 | 1.307 | 0.983 | 1.365 *† | 1.051 | 0.985 |
| 95% CI | [1.902, 2.229] | [1.169, 1.461] | [0.949, 1.019] | [1.210, 1.540] | [0.921, 1.197] | [0.948, 1.023] |
| P-value | <.0001 | <.0001 | 0.354 | <.0001 | 0.470 | 0.437 |
| ED visit | ||||||
| Odds Ratio | 1.297 | 0.970 | 1.470 | 1.149 | 1.115 | 1.251 |
| 95% CI | [1.146, 1.468] | [0.844, 1.116] | [1.420, 1.523] | [1.012, 1.306] | [0.980, 1.270] | [1.206, 1.298] |
| P-value | <.0001 | 0.670 | <.0001 | 0.030 | 0.098 | <.0001 |
Note: No risky alcohol or smoking is the reference group for all analyses;
Estimate for risky alcohol and smoking group is significantly different (p <0.05) than estimate for smoking only group;
Estimate for risky alcohol and smoking group is significantly different (p <0.05) than estimate for risky alcohol only group
Hospital Readmission
In the propensity score weighted model, the odds of hospital readmission were higher in the risky alcohol and smoking group (OR 1.28, 95% CI 1.12–1.46) relative to the no risky alcohol/smoking group. The odds of hospital readmission were significantly higher in the smoking only group (OR 1.08, 95% CI 1.04–1.128) and the risky alcohol and smoking group (OR 1.46, 95% 1.28–1.66) in the unadjusted model only.
Surgical complications
In the propensity score weighted model, the odds of surgical complications were higher in the risky alcohol and smoking group (OR 1.37, 95% CI 1.21–1.54) relative to the no risky alcohol/smoking group. The estimate for the risky alcohol and smoking group was also significantly different from either of the estimates for the smoking only or risky alcohol only groups. The odds of surgical complications were significantly higher in the risky alcohol only group (OR 1.31, 95% CI 1.17 – 1.46) and risky alcohol and smoking group (OR 2.091, 95% CI 1.90–2.30) in the unadjusted model only.
ED visit
In the propensity score weighted model, the odds of ED visit were significantly higher in the risky alcohol and smoking group (OR 1.15, 95% CI 1.01–1.30) relative to the no risky alcohol or smoking group. In the propensity score weighted model, the odds of a ED visit were significantly higher in the smoking only group (OR 1.25, 95% CI 1.21–1.30) relative to the no risky alcohol or smoking group.
Discussion
In this investigation of more than 200,000 surgical patients, the combination of both risky alcohol use (> 2 drinks/day) as well as a history of smoking cigarettes in the past year was associated prospectively with increased odds of 30-day adverse events after surgery. This is the first study to evaluate the likelihood of adverse post-operative outcomes based on the combination of both risky levels of alcohol use (> 2 drinks/day) as well as a history of smoking cigarettes in the past year; prior research has only examined these two health behaviors in isolation, despite their high co-occurrence. We found increased odds of reoperation, readmission, surgical complications, and ED visits among those with co-occurring risky alcohol use and smoking prior to surgery relative to those who do not drink alcohol at risky levels or smoke even after applying propensity score weights that included a host of relevant covariates. In the propensity score weighted models, odds of reoperation were 77% higher among those with co-occurring risky alcohol use and smoking relative to those who do neither, while the odds of complications were 37% higher, the odds of readmission were 28% higher, and the odds of ED visits were 15% higher.
After applying propensity score weights that included relevant clinical variables, those with co-occurring risky alcohol use and smoking also had the highest odds of reoperation, readmission, and surgical complications relative to those who only smoked or only drank alcohol at risky levels, highlighting the compounding risks of these two behaviors on surgical health. Risky alcohol use and smoking are already well-documented surgical risk factors,13,31–37 associated with increased healthcare utilization and costs after surgery,38 and even post-operative mortality.13,14 This study extends this literature by demonstrating the combined risks of these two behaviors. Careful consideration of surgical candidacy should precede scheduling elective surgeries for those who drink at risky levels and smoke. Those with both risk factors should be offered preoperative education and intervention to reduce both risk factors prior to surgery. This should ideally include counseling and pharmacotherapy, leveraging surgery as a unique point for behavior change to address acute, salient surgical health risks.39–41 Preoperative alcohol and smoking screening and intervention are important and often overlooked tools in surgical healthcare.17,42,43 Key surgical and anesthesia groups recommend alcohol and tobacco screening prior to surgery,44 yet alcohol screening in particular is often sub-optimal and biased.43,45–47 Furthermore, empirically-supported alcohol and tobacco interventions decrease the likelihood of surgical complications when delivered at appropriate time points,17,39 but are rarely implemented, particularly in the case of alcohol interventions. Interventions that address both risky alcohol and smoking concurrently prior to surgery appear to increase cessation of both substances, but their efficacy in preventing surgical complications has yet to be demonstrated.48 The broader literature supports the efficacy of behavior change interventions that address these two behaviors simultaneously.49 Additional attention and research is needed on multiple behavior interventions in the context of surgical health optimization.
It is important to note that only 2.8% of patients in this sample were identified as drinking more than 2 drinks/day on average, which is likely a substantial under-ascertainment. The population prevalence of past-year alcohol use disorders in the United States is approximately 14%,1 and is generally elevated in surgical populations.3 Past research highlights sub-optimal alcohol screening practices in surgical settings that results in under identification of risky alcohol use.47,50 All of this suggests under-ascertainment of alcohol use in the present sample, which would bias our findings towards the null. It is likely that many patients who drink alcohol at risky levels were included in the reference group in analysis. Thus, the estimated effect of risky alcohol use on surgical health outcomes might be significantly higher than our findings suggest, and further detailed research on this topic is warranted.
This study has several limitations in addition to the potential misclassification of risky alcohol use discussed previously. The sample was from a single state in the United States, and was predominantly White and non-Hispanic. Surgical outcomes were evaluated through 30 days after surgery, thus adverse outcomes occurring in subsequent months could not be evaluated. The data on ED visits and readmission outcomes could have included events unrelated to the surgery of record, as MSQC did not track whether ED visits and readmission were definitively linked to the surgery until 2020. Under ascertainment of alcohol likely biased findings towards the null. In addition, the time from for smoking and alcohol-related risks overlap but differed in length (past year smoking vs past two week alcohol use). Despite limitations this study has many strengths including a large sample from a large midwestern state including rural and urban hospitals. Variables were extracted from medical records by trained chart abstractors with standardized protocols and oversight.
Conclusions
Co-occurring risky alcohol use and smoking appear to confer the highest risks of adverse post-surgical events, more so than either alcohol use or smoking alone. Additional attention to co-occurring alcohol and tobacco use in patients undergoing general surgery is warranted, including improvements to screening and intervention practices.
Supplementary Material
Conflicts of Interest and Source of Funding:
ACF effort was partially supported by a career development award from the National Institute of Alcohol Abuse and Alcoholism (K23 AA023869) and reports other grants from National Institutes of Health and Michigan Department of Health and Human Services outside the current work. KMB reports salary support from the National Institutes of Health, Centers for Disease Control, and Veterans Affairs Health Administration. MB reports grants from National Institutes of Health, Centers for Disease Control, Alosa Health, and Michigan Department of Health and Human Services outside the current work, and past consultation for Axial Healthcare and Alosa Health outside of the current work. MJE reports salary support from the National Institutes of Health, Centers for Disease Control, Michigan Department of Health and Human Services, and Blue Cross and Blue Shield of Michigan. For the remaining authors none were declared.
Funding/Support:
The Michigan Surgical Quality Collaborative is funded as part of the Blue Cross Blue Shield of Michigan Value Partnership program
References
- 1.Grant BF, Goldstein RB, Saha TD, et al. Epidemiology of DSM-5 Alcohol Use Disorder. JAMA Psychiatry. 2015;72:757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Grant BF, Chou SP, Saha TD, 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;74:911–923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Harris AHS, Frey MS, Debenedetti AF, et al. Alcohol Misuse Prevalence and Associations with Post-Operative Complications in US Surgical Patients : A Review. Open J Surg. 2008;50–58. [Google Scholar]
- 4.Howard R, Singh K, Englesbe M. Prevalence and Trends in Smoking Among Surgical Patients in Michigan, 2012–2019. JAMA Netw Open. 2021;4:e210553–e210553. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Khuri SF, Henderson WG, Daley J, et al. Successful implementation of the Department of Veterans Affairs’ National Surgical Quality Improvement Program in the private sector: the Patient Safety in Surgery study. Ann Surg. 2008;248:329–36. [DOI] [PubMed] [Google Scholar]
- 6.Durand WM, DePasse JM, Bokshan SL, et al. Tobacco Use and Complications Following Spinal Fusion: A Comparison of the National Surgical Quality Improvement Program and National Inpatient Sample Datasets. World Neurosurg. 2019;123:e393–e407. [DOI] [PubMed] [Google Scholar]
- 7.Hatcher JL, Sterba KR, Tooze JA, et al. Tobacco use and surgical outcomes in patients with head and neck cancer. Head Neck. 2016;38:700–706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jackson KL, Devine JG. The Effects of Smoking and Smoking Cessation on Spine Surgery: A Systematic Review of the Literature. Glob Spine J. 2016;6:695–701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hawn MT, Houston TK, Campagna EJ, et al. The Attributable Risk of Smoking on Surgical Complications. Ann Surg. 2011;254:914–920. [DOI] [PubMed] [Google Scholar]
- 10.Borad NP, Merchant AM. The effect of smoking on surgical outcomes in ventral hernia repair: a propensity score matched analysis of the National Surgical Quality Improvement Program data. Hernia. 2017;21:855–867. [DOI] [PubMed] [Google Scholar]
- 11.Turan A, Mascha EJ, Roberman D, et al. Smoking and perioperative outcomes. Anesthesiology. 2011;114:837–846. [DOI] [PubMed] [Google Scholar]
- 12.Kubasiak JC, Landin M, Schimpke S, et al. The effect of tobacco use on outcomes of laparoscopic and open ventral hernia repairs: a review of the NSQIP dataset. Surg Endosc. 2017;31:2661–2666. [DOI] [PubMed] [Google Scholar]
- 13.Eliasen M, Grønkjær M, Skov-Ettrup LS, et al. Preoperative Alcohol Consumption and Postoperative Complications: A Systematic Review and Meta-analysis. Ann Surg.;258. Epub ahead of print May 31, 2013. DOI: 10.1097/SLA.0b013e3182988d59. [DOI] [PubMed] [Google Scholar]
- 14.Rubinsky AD, Bishop MJ, Maynard C, et al. Postoperative risks associated with alcohol screening depend on documented drinking at the time of surgery. Drug Alcohol Depend. 2013;132:521–527. [DOI] [PubMed] [Google Scholar]
- 15.Bradley KA, Rubinsky AD, Sun H, et al. Alcohol screening and risk of postoperative complications in male VA patients undergoing major non-cardiac surgery. J Gen Intern Med. . Epub ahead of print 2011. DOI: 10.1007/s11606-010-1475-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mulligan RP, McCarthy KJ, Grear BJ, et al. Preoperative Risk Factors for Complications in Elective Ankle and Hindfoot Reconstruction. Foot Ankle Spec. 2017;193864001770615. [DOI] [PubMed] [Google Scholar]
- 17.Egholm JW, Pedersen B, Møller AM, et al. Perioperative alcohol cessation intervention for postoperative complications. Cochrane Database Syst Rev. . Epub ahead of print November 8, 2018. DOI: 10.1002/14651858.CD008343.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Thomsen T, Villebro N, Møller AM. Interventions for preoperative smoking cessation. Cochrane Database of Systematic Reviews.;2014. Epub ahead of print March 27, 2014. DOI: 10.1002/14651858.CD002294.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jiang N, Lee YO, Ling PM. Young adult social smokers: Their co-use of tobacco and alcohol, tobacco-related attitudes, and quitting efforts. Prev Med (Baltim). 2014;69:166–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Falk DE, Yi HY, Hiller-Sturmhöfel S. An Epidemiologic Analysis of Co-Occurring Alcohol and Tobacco Use and Disorders: Findings From the National Epidemiologic Survey on Alcohol and Related Conditions. Alcohol Res Heal. 2006;29:162. [PMC free article] [PubMed] [Google Scholar]
- 21.McKee SA, Falba T, O’Malley SS, et al. Smoking status is a clinical indicator for alcohol misuse in US adults. Arch Intern Med. 2007;167:716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pelucchi C, Gallus S, Garavello W, et al. Alcohol and tobacco use, and cancer risk for upper aerodigestive tract and liver. Eur J Cancer Prev. 2008;17:340–344. [DOI] [PubMed] [Google Scholar]
- 23.Benowitz NL, Jones RT, Jacob P. Additive cardiovascular effects of nicotine and ethanol. Clin Pharmacol Ther. 1986;40:420–424. [DOI] [PubMed] [Google Scholar]
- 24.Taylor B, Rehm J. When risk factors combine: The interaction between alcohol and smoking for aerodigestive cancer, coronary heart disease, and traffic and fire injury. Addict Behav. 2006;31:1522–1535. [DOI] [PubMed] [Google Scholar]
- 25.Liu B, Balkwill A, Roddam A, et al. Separate and Joint Effects of Alcohol and Smoking on the Risks of Cirrhosis and Gallbladder Disease in Middle-aged Women. Am J Epidemiol. 2009;169:153–160. [DOI] [PubMed] [Google Scholar]
- 26.Campbell DA, Kubus JJ, Henke PK, et al. The Michigan Surgical Quality Collaborative: a legacy of Shukri Khuri. Am J Surg. 2009;198:S49–S55. [DOI] [PubMed] [Google Scholar]
- 27.Share DA, Campbell DA, Birkmeyer N, et al. How A Regional Collaborative Of Hospitals And Physicians In Michigan Cut Costs And Improved The Quality Of Care. https://doi.org/101377/hlthaff20100526. 2017;30:636–645. [DOI] [PubMed] [Google Scholar]
- 28.Leslie S, Thiebaud P. Using Propensity Scores to Adjust For Treatment Selection Bias Available from: http://support.sas.com/documentation/onlinedoc/91pdf/sasdoc_91/stat_ug_7313.pdf. Accessed April 20, 2022.
- 29.Austin PC. An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies. Multivariate Behav Res. 2011;46:399–424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70:41–55. [Google Scholar]
- 31.Noh SH, Cho PG, Kim KN, et al. Risk factors for reoperation after lumbar spine surgery in a 10-year Korean national health insurance service health examinee cohort. Sci Rep. 2022;12:4606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Badiee RK, Chan AK, Rivera J, et al. Smoking Is an Independent Risk Factor for 90-Day Readmission and Reoperation Following Posterior Cervical Decompression and Fusion. Neurosurgery. 2021;88:1088–1094. [DOI] [PubMed] [Google Scholar]
- 33.Tischler EH, Ko LM, Chen AF, et al. Smoking Increases the Rate of Reoperation for Infection within 90 Days After Primary Total Joint Arthroplasty. J Bone Joint Surg Am. 2017;99:295–304. [DOI] [PubMed] [Google Scholar]
- 34.Bydon M, Macki M, De La Garza-Ramos R, et al. Smoking as an independent predictor of reoperation after lumbar laminectomy: a study of 500 cases. J Neurosurg Spine. 2015;22:288–293. [DOI] [PubMed] [Google Scholar]
- 35.Andersen SB, Smith EC, Støttrup C, et al. Smoking Is an Independent Risk Factor of Reoperation Due to Recurrent Lumbar Disc Herniation. Glob spine J. 2018;8:378–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Campenfeldt P, Al-Ani A, Hedström M, et al. Low BMD and high alcohol consumption predict a major re-operation in patients younger than 70 years of age with a displaced femoral neck fracture—A two –year follow up study in 120 patients. Injury. 2018;49:2042–2046. [DOI] [PubMed] [Google Scholar]
- 37.Bradley KA, Rubinsky AD, Sun H, et al. Alcohol screening and risk of postoperative complications in male VA patients undergoing major non-cardiac surgery. J Gen Intern Med. 2011;26:162–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Warner DO, Borah BJ, Moriarty J, et al. Smoking Status and Health Care Costs in the Perioperative Period: A Population-Based Study. JAMA Surg. 2014;149:259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Thomsen T, Villebro N, Møller AM. Interventions for preoperative smoking cessation. Cochrane Database Syst Rev. . Epub ahead of print 2010. DOI: 10.1002/14651858.CD002294. [DOI] [PubMed] [Google Scholar]
- 40.Oppedal K, Møller A, Pedersen B, et al. Preoperative alcohol cessation prior to elective surgery. Cochrane database Syst Rev. 2012;CD008343. [DOI] [PubMed] [Google Scholar]
- 41.Lauridsen SV, Thomsen T, Kaldan G, et al. Smoking and alcohol cessation intervention in relation to radical cystectomy: A qualitative study of cancer patients’ experiences. BMC Cancer. 2017;17:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Fernandez AC, Claborn KR, Borsari B. A systematic review of behavioural interventions to reduce preoperative alcohol use. Drug Alcohol Rev. 2015;34:508–520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Shourie S, Conigrave KM, Proude EM, et al. Pre-operative screening for excessive alcohol consumption among patients scheduled for elective surgery. Drug Alcohol Rev. 2007;26:119–25. [DOI] [PubMed] [Google Scholar]
- 44.American Society of Anesthesiologists. ASA Physical Status Classification System Available from: https://www.asahq.org/standards-and-guidelines/asa-physical-status-classification-system. 2014.
- 45.Kleinwächter R, Kork F, Weiss-Gerlach E, et al. Improving the detection of illicit substance use in preoperative anesthesiological assessment. Minerva Anestesiol. 2010;76:29–37. [PubMed] [Google Scholar]
- 46.Bradley KA, Rubinsky AD, Sun H, et al. Prevalence of alcohol misuse among men and women undergoing major noncardiac surgery in the Veterans Affairs health care system. Surgery. 2012;152:69–81. [DOI] [PubMed] [Google Scholar]
- 47.Fernandez AC, Guetterman TC, Borsari B, et al. Gaps in Alcohol Screening and Intervention Practices in Surgical Healthcare: A Qualitative Study. J Addict Med. 2020;15:113–119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Lauridsen SV, Thomsen T, Jensen JB, et al. Effect of a Smoking and Alcohol Cessation Intervention Initiated Shortly Before Radical Cystectomy—the STOP-OP Study: A Randomised Clinical Trial. Eur Urol Focus.;0. Epub ahead of print March 2022. DOI: 10.1016/J.EUF.2022.02.005. [DOI] [PubMed] [Google Scholar]
- 49.Prochaska JJ, Prochaska JO. A Review of Multiple Health Behavior Change Interventions for Primary Prevention. DOI: 10.1177/1559827610391883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Kip MJ, Neumann T, Jugel C, et al. New strategies to detect alcohol use disorders in the preoperative assessment clinic of a German university hospital. Anesthesiology. 2008;109:171–179. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.