Investigating the Impact of Cannabis Consumption on Hospital Outcomes in Patients With Primary Spontaneous Pneumothorax: A Nationwide Analysis
Aman GoyalMohammed QuaziRayika SyedHafiz Abdullah IkramFarooq Ahmad SheikhAsif FarooqSulaiman SultanAbu Baker Sheikh
1
Citation
20
Reference
10
Related Paper
Citation Trend
Abstract:
Introduction Existing data suggest an association between primary spontaneous pneumothorax (PSP) and cannabis consumption, although evidence remains controversial. Methods This study used the 2016-2019 National Inpatient Sample Database to examine inpatients with PSP, categorizing them as cannabis users and non-users. Multivariate regression analyzed continuous variables, chi-square assessed categorical variables, and logistic regression models were built. Propensity score matching (PSM) mitigated the confounding bias. Results A total of 399,495 patients with PSP were admitted during the study period (13,415 cannabis users and 386,080 non-cannabis users). Cannabis users were more likely to be younger (p<0.001) and male (p<0.001) with a lower risk of baseline comorbidities than non-users. Cannabis users had a lower risk of sudden cardiac arrest, vasopressor use, the development of acute kidney injury, venous thromboembolism, the requirement for invasive and non-invasive mechanical ventilation, hemodialysis, ventilator-associated pneumonia, and the need for a tracheostomy. Cannabis use was associated with a 3.4 days shorter hospital stay (p<0.001), as confirmed by PSM analysis (2.3 days shorter, p<0.001). Additionally, cannabis users showed a lower risk of in-hospital mortality (p<0.001), a trend maintained in the PSM analysis (p<0.001). Conclusions Our study revealed correlations suggesting that cannabis users with PSP might experience lower in-hospital mortality and fewer complications than non-cannabis users.High-dimensional propensity score analysis automatically selects independent variables for calculating propensity scores, using a vast amount of information from real-world health care databases. This technique can reduce confounding by indication or unmeasured confounders more precisely compared with conventional propensity score analysis. High-dimensional propensity score analysis assumes that proxy information for important unmeasured confounders can be obtained from the underlying data. The number of published studies using high-dimensional propensity score analysis has increased, with pharmacoepidemiology as the main area in which these studies have been published. This report explains the main assumption and the limitations of this analytical method and provides step-by-step procedures to implement the method.
Proxy (statistics)
Cite
Citations (2)
Stratification (seeds)
Cite
Citations (0)
This chapter explores the important issue of confounding in observational studies. The potential imbalances that result for not controlling assignment of treatment or exposure may lead to imbalance of variables that are associated with both treatment and intervention (or exposure) thus confounding results. Therefore, in this context, a potential relationship between an intervention and an outcome could be invalid. This chapter therefore explains basic definitions of confounding and presents some methods to control for confounders, highlighting the use of the propensity score, which is considered a robust method for this purpose. Different techniques of adjustment using propensity score (matching, stratification, regression, and weighting) are also discussed. This chapter concludes with a case discussion about confounding and how to address it.
Cite
Citations (0)
Cite
Citations (1)
Propensity score matching is a common tool for adjusting for observed confounding in observational studies, but is known to have limitations in the presence of unmeasured confounding. In many settings, researchers are confronted with spatially-indexed data where the relative locations of the observational units may serve as a useful proxy for unmeasured confounding that varies according to a spatial pattern. We develop a new method, termed Distance Adjusted Propensity Score Matching (DAPSm) that incorporates information on units' spatial proximity into a propensity score matching procedure. We show that DAPSm can adjust for both observed and some forms of unobserved confounding and evaluate its performance relative to several other reasonable alternatives for incorporating spatial information into propensity score adjustment. The method is motivated by and applied to a comparative effectiveness investigation of power plant emission reduction technologies designed to reduce population exposure to ambient ozone pollution. Ultimately, DAPSm provides a framework for augmenting a "standard" propensity score analysis with information on spatial proximity and provides a transparent and principled way to assess the relative trade offs of prioritizing observed confounding adjustment versus spatial proximity adjustment.
Proxy (statistics)
Cite
Citations (0)
Cite
Citations (0)
A propensity score is the probability of being treated or exposed, given measured confounders such as age. Propensity score methods can control for measured confounders in observational research, but not for unmeasured confounders. Study participants with the same propensity score will, on average, have a similar distribution of measured confounders. The propensity score can be used to control for confounding by means of stratification, matching, including the propensity score as a covariate in a multivariable regression model, or by weighting the study population using the propensity score. Propensity score methods can often control for more confounders than other methods, particularly in the case of a rare outcome. Reports on propensity score methods should mention which confounders were included in the propensity score and to what extent confounders were balanced across study groups, after stratification based on the propensity score.
Cite
Citations (7)
Abstract Confounding variables can affect the results from studies of children with Down syndrome and their families. Traditional methods for addressing confounders are often limited, providing control for only a few confounding variables. This study introduces propensity score matching to control for multiple confounding variables. Using Tennessee birth data as an example, newborns with Down syndrome were compared with a group of typically developing infants on birthweight. Three approaches to matching on confounders—nonmatched, covariate matched, and propensity matched—were compared using 8 potential confounders. Fewer than half of the newborns with Down syndrome were matched using covariate matching, and the matched group was differed from the unmatched newborns. Using propensity scores, 100% of newborns with Down syndrome could be matched to a group of comparison newborns, a decreased effect size was found on newborn birthweight, and group differences were not statistically significant.
Cite
Citations (24)
Purpose Propensity scores are used in observational studies to adjust for confounding, although they do not provide control for confounders omitted from the propensity score model. We sought to determine if tests used to evaluate logistic model fit and discrimination would be helpful in detecting the omission of an important confounder in the propensity score. Methods Using simulated data, we estimated propensity scores under two scenarios: (1) including all confounders and (2) omitting the binary confounder. We compared the propensity score model fit and discrimination under each scenario, using the Hosmer–Lemeshow goodness-of-fit (GOF) test and the c-statistic. We measured residual confounding in treatment effect estimates adjusted by the propensity score omitting the confounder. Results The GOF statistic and discrimination of propensity score models were the same for models excluding an important predictor of treatment compared to the full propensity score model. The GOF test failed to detect poor model fit for the propensity score model omitting the confounder. C-statistics under both scenarios were similar. Residual confounding was observed from using the propensity score excluding the confounder (range: 1–30%). Conclusions Omission of important confounders from the propensity score leads to residual confounding in estimates of treatment effect. However, tests of GOF and discrimination do not provide information to detect missing confounders in propensity score models. Our findings suggest that it may not be necessary to compute GOF statistics or model discrimination when developing propensity score models. Copyright © 2004 John Wiley & Sons, Ltd.
Statistic
Goodness of fit
Cite
Citations (92)
In this article, we presented the rationale and calculation procedures of the propcnsity score matching (PSM), and its application in the designing stage of an cpidcrniological study. Based on existing observational data, PSM can be used to select one or more comparable controls for each subject in 'treatment' group according to the propensity scores estimated by 'treatment' variable and main covariates. The results of an example analysis showed that the bias for main confounders between the treated and control samples declined more than 55% after PMS. Conclusion: PSM can reduce most of the confounding bias of the observational study, and can obtain approximate study effect to the randomized controlled trials when used in the designing of thc cpidcmiological study.
Key words:
Propensity score matching; Confounding bias; Epidemiology
Cite
Citations (2)