Logistic regression frequently outperformed propensity score methods, especially for large datasets: a simulation study.

OBJECTIVES: In observational studies, researchers must select a method to control for confounding. Options include propensity score (PS) methods and regression. It remains unclear how dataset characteristics (size, overlap in PSs, and exposure prevalence) influence the relative performance of the methods. STUDY DESIGN AND SETTING: A simulation study to evaluate the role of dataset characteristics on the performance of PS methods, compared to logistic regression, for estimating a marginal odds ratio was conducted. Dataset size, overlap in PSs, and exposure prevalence were varied. RESULTS: Regression showed poor coverage for small sample sizes, but with large sample sizes was relatively robust to imbalance in PSs and low exposure prevalence. PS methods displayed suboptimal coverage as overlap in PSs decreased, which was exacerbated at larger sample sizes. Power of matching methods was particularly affected by a lack of overlap, low exposure prevalence, and small sample size. The advantage of regression for large data size was reduced in sensitivity analysis with a complementary log-log outcome generation mechanism and unmeasured confounding, with superior bias and error but inferior coverage to matching methods. CONCLUSION: Dataset characteristics influence performance of methods for confounder adjustment. In many scenarios, regression may be the preferable option.

A Multicenter Study of the Validity and Reliability of Responses to Hand Cold Challenge as Measured by Laser Speckle Contrast Imaging and Thermography: Outcome Measures for Systemic Sclerosis-Related Raynaud's Phenomenon.

OBJECTIVE: Reliable and objective outcome measures to facilitate clinical trials of novel treatments for systemic sclerosis (SSc)-related Raynaud's phenomenon (RP) are badly needed. Laser speckle contrast imaging (LSCI) and thermography are noninvasive measures of perfusion that have shown excellent potential. This multicenter study was undertaken to determine the reliability and validity of a hand cold challenge protocol using LSCI, standard thermography, and low-cost cell phone/mobile phone thermography (henceforth referred to as mobile thermography) in patients with SSc-related RP. METHODS: Patients with RP secondary to SSc were recruited from 6 UK tertiary care centers. The patients underwent cold challenge on 2 consecutive days. Changes in cutaneous blood flow/skin temperature at each visit were imaged simultaneously using LSCI, standard thermography, and mobile thermography. Measurements included area under the curve (AUC) for reperfusion/rewarming and maximum blood flow rate/skin temperature after rewarming (MAX). Test-retest reliability was assessed using intraclass correlation coefficients (ICCs). Estimated latent correlations (estimated from multilevel models, taking values between -1 and 1; denoted as rho values) were used to assess the convergent validity of LSCI and thermography. RESULTS: In total, 159 patients (77% with limited cutaneous SSc) were recruited (84% female, median age 63.3 years). LSCI and standard thermography both had substantial reliability, with ICCs for the reperfusion/rewarming AUC of 0.67 (95% confidence interval [95% CI] 0.54, 0.76) and 0.68 (95% CI 0.58, 0.80), respectively, and ICCs for the MAX of 0.64 (95% CI 0.52, 0.75) and 0.72 (95% CI 0.64, 0.81), respectively. Very high latent correlations were present for the AUCs of LSCI and thermography (ρ = 0.94; 95% CI 0.87, 1.00) and for the AUCs of standard and mobile thermography (ρ = 0.98; 95% CI 0.94, 1.00). CONCLUSION: This is the first multicenter study to examine the reliability and validity of cold challenge using LSCI and thermography in patients with SSc-related RP. LSCI and thermography both demonstrated good potential as outcome measures. LSCI, standard thermography, and mobile thermography had very high convergent validity.