Smoking is associated with a higher risk of unplanned medical visits among adult patients with diabetes, using retrospective electronic medical record data from 2014 to 2016

Tobacco Smoking or Nicotine Phenotype and Severity of Clinical Presentation at the Emergency Department (SMOPHED): Protocol for a Noninterventional Observational Study

BACKGROUND

In the last few years, several nicotine products have become available as alternatives to smoking tobacco. While laboratory and limited clinical studies suggest that these devices are less toxic compared to classic tobacco cigarettes, very little is known about their epidemiological impact. Visiting the emergency department (ED) often represents the first or even the only contact of patients with the health care system. Therefore, a study conducted at the ED to assess the impact of these products on health can be reliable and reflect a real-life setting.

OBJECTIVE

The aim of this noninterventional observational study (SMOPHED study) is to analyze the association between the severity of clinical presentation observed during ED visits among patients using various nicotine products and the subsequent outcomes, specifically hospitalization and mortality.

METHODS

Outcomes (hospitalization and mortality in the ED) will be examined in relation to various patterns of nicotine products use. We plan to enroll approximately 2000 participants during triage at the ED. These individuals will be characterized based on their patterns of tobacco and nicotine consumption, identified through a specific questionnaire. This categorization will allow for a detailed analysis of how different usage patterns of nicotine products correlate with the clinical diagnosis made during the ED visits and the consequent outcomes.

RESULTS

Enrollment into the study started in March 2024. We enrolled a total of 901 participants in 1 month (approximately 300 potential participants did not provide the informed consent to participate). The data will be analyzed by a statistician as soon as the database is completed. Full data will be published by December 2024.

CONCLUSIONS

There is substantial debate about the harm reduction potential of alternative nicotine products in terms of their smoking-cessation and risk-reduction potential. This study represents an opportunity to document epidemiological data on the link between the use of different types of nicotine products and disease diagnosis and severity during an ED visit, and thus evaluate the harm reduction potential claims for these products.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

PRR1-10.2196/54041.

Hospital length of stay and readmission after elective surgery: a comparison of current and former smokers with non-smokers

BACKGROUND

The purpose of this study was to investigate differences between non-smokers, ex-smokers and current smokers in hospital length of stay (LOS), readmission (seven and 28 days) and cost of readmission for patients admitted for elective surgery.

METHODS

A retrospective cohort study of administrative inpatient data from 24, 818 patients admitted to seven metropolitan hospitals in Western Australia between 1 July 2016 and 30 June 2019 for multiday elective surgery was conducted. Data included smoking status, LOS, procedure type, age, sex and Indigenous status. LOS for smoking status was compared using multivariable negative binomial regression. Odds of readmission were compared for non-smokers and both ex-smokers and current smokers using separate multivariable logistic regression models.

RESULTS

Mean LOS for non-smokers (4.7 days, SD=5.7) was significantly lower than both ex-smokers (6.2 days SD 7.9) and current smokers (6.1 days, SD=8.2). Compared to non-smokers, current smokers and ex-smokers had significantly higher odds of readmission within seven (OR=1.29; 95% CI: 1.13, 1.47, and OR=1.37; 95% CI: 1.19, 1.59, respectively) and 28 days (OR=1.35; 95% CI: 1.23, 1.49, and OR=1.53; 95% CI: 1.39, 1.69, respectively) of discharge. The cost of readmission for seven and 28-day readmission was significantly higher for current smokers compared to non-smokers (RR=1.52; 95% CI: 1.1.6, 2.0; RR=1.39; 95% CI: 1.18, 1.65, respectively).

CONCLUSION

Among patients admitted for elective surgery, hospital LOS, readmission risk and readmission costs were all higher for smokers compared with non-smokers. The findings indicate that provision of smoking cessation treatment for adults undergoing elective surgery is likely to produce multiple benefits.

Current tobacco use patterns associated with healthcare utilization among non-Hispanic Black and Hispanic men with chronic conditions

INTRODUCTION

Tobacco use disparities persist among U.S. adults who are male, racially/ethnically diverse, and have chronic conditions. This study assessed current tobacco use patterns associated with past year healthcare utilization among non-Hispanic Black and Hispanic men ≥40 years old with ≥1 chronic condition.

METHODS

Data were collected from a sample of 1,904 non-Hispanic Black and Hispanic men from across the U.S. using an internet-delivered survey. Participants were categorized into four tobacco use groups: nontobacco users, exclusive cigarette smokers, dualtobacco users (cigarettes + one other tobacco product), and polytobacco users (cigarettes + ≥2 other tobacco products). Logistic regression analyses were conducted to assess current tobacco use patterns with past year primary care visits, emergency department (ED) visits, and overnight hospital stays. Adjusted models included participants' age, race/ethnicity, education level, marital status, health insurance coverage, body mass index, and number of chronic conditions.

RESULTS

Relative to nontobacco users, exclusive cigarette smokers were at decreased odds of having a past year primary care visit (adjusted odds ratio [AOR] = 0.68, 95% confidence interval [CI] = 0.47-0.99). Exclusive cigarette smokers (AOR = 1.66, 95%CI = 1.25-2.19), dualtobacco users (AOR = 1.75, 95%CI = 1.23-2.50), and polytobacco users (AOR = 4.10, 95%CI = 2.46-6.84) were at increased odds of having a past year ED visit compared to nontobacco users. Additionally, polytobacco users were at increased odds of having a past year overnight hospital stay (AOR = 2.72, 95%CI = 1.73-4.29) compared to nontobacco users.

CONCLUSIONS

Findings suggest current tobacco use patterns are uniquely associated with past year healthcare utilization among non-Hispanic Black and Hispanic men, while taking into consideration important factors including complex disease profiles.

Predicting unplanned medical visits among patients with diabetes: translation from machine learning to clinical implementation

Background

Diabetes is a medical and economic burden in the United States. In this study, a machine learning predictive model was developed to predict unplanned medical visits among patients with diabetes, and findings were used to design a clinical intervention in the sponsoring healthcare organization. This study presents a case study of how predictive analytics can inform clinical actions, and describes practical factors that must be incorporated in order to translate research into clinical practice.

Methods

Data were drawn from electronic medical records (EMRs) from a large healthcare organization in the Northern Plains region of the US, from adult (≥ 18 years old) patients with type 1 or type 2 diabetes who received care at least once during the 3-year period. A variety of machine-learning classification models were run using standard EMR variables as predictors (age, body mass index (BMI), systolic blood pressure (BP), diastolic BP, low-density lipoprotein, high-density lipoprotein (HDL), glycohemoglobin (A1C), smoking status, number of diagnoses and number of prescriptions). The best-performing model after cross-validation testing was analyzed to identify strongest predictors.

Results

The best-performing model was a linear-basis support vector machine, which achieved a balanced accuracy (average of sensitivity and specificity) of 65.7%. This model outperformed a conventional logistic regression by 0.4 percentage points. A sensitivity analysis identified BP and HDL as the strongest predictors, such that disrupting these variables with random noise decreased the model’s overall balanced accuracy by 1.3 and 1.4 percentage points, respectively. These recommendations, along with stakeholder engagement, behavioral economics strategies, and implementation science principles helped to inform the design of a clinical intervention targeting behavioral changes.

Conclusion

Our machine-learning predictive model more accurately predicted unplanned medical visits among patients with diabetes, relative to conventional models. Post-hoc analysis of the model was used for hypothesis generation, namely that HDL and BP are the strongest contributors to unplanned medical visits among patients with diabetes. These findings were translated into a clinical intervention now being piloted at the sponsoring healthcare organization. In this way, this predictive model can be used in moving from prediction to implementation and improved diabetes care management in clinical settings.