The Search for Causal Inferences: Using Propensity Scores Post Hoc to Reduce Estimation Error With Nonexperimental Research

Abstract

While experimental designs are regarded as the gold standard for establishing causal relationships, such designs are usually impractical owing to common methodological limitations. The objective of this article is to illustrate how propensity score matching (PSM) and using propensity scores (PS) as a covariate are viable alternatives to reduce estimation error when experimental designs cannot be implemented. To mimic common pediatric research practices, data from 140 simulated participants were used to resemble an experimental and nonexperimental design that assessed the effect of treatment status on participant weight loss for diabetes. Pretreatment participant characteristics (age, gender, physical activity, etc.) were then used to generate PS for use in the various statistical approaches. Results demonstrate how PSM and using the PS as a covariate can be used to reduce estimation error and improve statistical inferences. References for issues related to the implementation of these procedures are provided to assist researchers.

Laparoscopic versus open surgery for pheochromocytoma: a meta-analysis

“…When disputes arose during data extraction, a consensus was reached through discussion with a third party. Data concerning the following were extracted: (1) general characteristics, including first author, publication year, country, study type, sex and age of patients, and body mass index; and oncology characteristics, including tumour size, tumour laterality, postoperative pathology, perioperative mortality, and postoperative recurrence; (2) intraoperative outcomes, including intraoperative haemodynamic instability (IHD), operative time, intraoperative blood loss and the number of blood transfusions; IHD was defined as intra-operative systolic blood pressure > 200 mmHg or mean arterial pressure < 60 mmHg requiring drugs or blood transfusions to main normal blood pressure intra-operatively [ 12 ]; (3) postoperative outcomes, including time to ambulation and diet, indwelling time of drainage tube, postoperative hospital stay, overall rates and severity (Clavien-Dindo grade ≥ II) of postoperative complications, rates of postoperative hypotension and cardiovascular disease (CVD) morbidity, and postoperative blood pressure control during follow-up. CVD morbidity was defined as complications related to the cardiovascular system, such as postoperative hypotension requiring drugs or blood transfusion, myocardial ischaemia, arrhythmia, and pulmonary embolism/cerebral infarction/deep vein thrombosis [ 13 ].…”

Effect of team training on efficiency of trauma care in a Chinese hospital

“… 20 Propensity-score analysis is a non-parametric method that can help to reduce the model dependence. 21 – 23 This was performed by regressing the team group variable on variables with P < 0.2 in a bivariate analysis. Propensity score or distance was calculated for each subject and represented the probability of assignation to the team group conditional on relevant variables.…”

Can propensity score matching be applied to cross-sectional data to evaluate Community-Based Rehabilitation? Results of a survey implementing the WHO’s Community-Based Rehabilitation indicators in Vietnam

“…Another limitation of PSM is that it leads to reduced sample size, which could limit generalizability, though this is partly addressed through the provided sensitivity analysis. The reduced sample size also increases the risk of type II error,36 but the sample size of this study met the commonly recommended minimum sample size of 10(p+1), where p is the number of matching variables 37. This study presents a starting point to encourage the generation of quantitative CBR research and demonstrates one possible method for reducing bias when analysing cross-sectional CBR data.…”