Estimating Intraclass Correlation Coefficient and Identifying Influential Observations Under One-Way Random Effects Model

In many biomedical investigations, parameters of interest, such as the intraclass correlation coefficient, are functions of higher-order moments reflecting finer distributional characteristics. One popular method to make inference for such parameters is through postulating a parametric random effects model. We relax the standard normality assumptions for both the random effects and errors through the use of the Fleishman distribution, a flexible four-parameter distribution which accounts for the third and fourth cumulants. We propose a Fleishman bootstrap method to construct confidence intervals for correlated data and develop a normality test for the random effect and error distributions. Recognizing that the intraclass correlation coefficient may be heavily influenced by a few extreme observations, we propose a modified, quantile-normalized intraclass correlation coefficient. We evaluate our methods in simulation studies and apply these methods to the Childhood Adenotonsillectomy Trial sleep electroencephalogram data in quantifying wave-frequency correlation among different channels.

Download Full-text

Biomechanical Measurement of Rabbit Cornea by a Modified Scheimpflug Device

Journal of Ophthalmology ◽

10.1155/2016/8271762 ◽

2016 ◽

Vol 2016 ◽

pp. 1-6

Author(s):

Bo Zhang ◽

Jianjun Gu ◽

Xiaoxiao Zhang ◽

Bin Yang ◽

Zheng Wang ◽

...

Keyword(s):

Bulk Modulus ◽

Correlation Coefficient ◽

Intraclass Correlation Coefficient ◽

Intraclass Correlation ◽

Anterior Segment ◽

Intraobserver Reliability ◽

Rabbit Cornea ◽

Deformation Response ◽

Biomechanical Measurements ◽

Biomechanical Measurement

Purpose. To explore the probability and variation in biomechanical measurements of rabbit cornea by a modified Scheimpflug device.Methods. A modified Scheimpflug device was developed by imaging anterior segment of the model imitating the intact eye at various posterior pressures. The eight isolated rabbit corneas were mounted on the Barron artificial chamber and images of the anterior segment were taken at posterior pressures of 15, 30, 45, 60, and 75 mmHg by the device. The repeatability and reliability of the parameters including CCT, ACD, ACV, and CV were evaluated at each posterior pressure. All the variations of the parameters at the different posterior pressures were calculated.Results. All parameters showed good intraobserver reliability (Cronbach’s alpha; intraclass correlation coefficient,α, ICC > 0.96) and repeatability in the modified Scheimpflug device. With the increase of posterior pressures, the ratio of CCT decreased linearly and the bulk modulus gradually reduced to a platform. The increase of ACD was almost linear with the posterior pressures elevated.Conclusions. The modified Scheimpflug device was a valuable tool to investigate the biomechanics of the cornea. The posterior pressure 15–75 mmHg range produced small viscoelastic deformations and nearly linear pressure-deformation response in the rabbit cornea.

Download Full-text

A preliminary study of intensivist-performed DVT ultrasound screening in trauma ICU patients (APSIT Study)

Annals of Intensive Care ◽

10.1186/s13613-020-00739-8 ◽

2020 ◽

Vol 10 (1) ◽

Author(s):

Lloyd Roberts ◽

Tom Rozen ◽

Deirdre Murphy ◽

Adam Lawler ◽

Mark Fitzgerald ◽

...

Keyword(s):

Intensive Care ◽

Deep Vein Thrombosis ◽

High Risk ◽

Correlation Coefficient ◽

Intraclass Correlation Coefficient ◽

Major Trauma ◽

Intraclass Correlation ◽

Trauma Patients ◽

Vein Thrombosis ◽

Deep Vein

Abstract Background Multiple screening Duplex ultrasound scans (DUS) are performed in trauma patients at high risk of deep vein thrombosis (DVT) in the intensive care unit (ICU). Intensive care physician performed compression ultrasound (IP-CUS) has shown promise as a diagnostic test for DVT in a non-trauma setting. Whether IP-CUS can be used as a screening test in trauma patients is unknown. Our study aimed to assess the agreement between IP-CUS and vascular sonographer performed DUS for proximal lower extremity deep vein thrombosis (PLEDVT) screening in high-risk trauma patients in ICU. Methods A prospective observational study was conducted at the ICU of Alfred Hospital, a major trauma center in Melbourne, Australia, between Feb and Nov 2015. All adult major trauma patients admitted with high risk for DVT were eligible for inclusion. IP-CUS was performed immediately before or after DUS for PLEDVT screening. The paired studies were repeated twice weekly until the DVT diagnosis, death or ICU discharge. Written informed consent from the patient, or person responsible, or procedural authorisation, was obtained. The individuals performing the scans were blinded to the others’ results. The agreement analysis was performed using Cohen’s Kappa statistics and intraclass correlation coefficient for repeated binary measurements. Results During the study period, 117 patients had 193 pairs of scans, and 45 (39%) patients had more than one pair of scans. The median age (IQR) was 47 (28–68) years with 77% males, mean (SD) injury severity score 27.5 (9.53), and a median (IQR) ICU length of stay 7 (3.2–11.6) days. There were 16 cases (13.6%) of PLEDVT with an incidence rate of 2.6 (1.6–4.2) cases per 100 patient-days in ICU. The overall agreement was 96.7% (95% CI 94.15–99.33). The Cohen’s Kappa between the IP-CUS and DUS was 0.77 (95% CI 0.59–0.95), and the intraclass correlation coefficient for repeated binary measures was 0.75 (95% CI 0.67–0.81). Conclusions There is a substantial agreement between IP-CUS and DUS for PLEDVT screening in trauma patients in ICU with high risk for DVT. Large multicentre studies are needed to confirm this finding.

Download Full-text

Test-Retest-Reliability of Computer-Based Metamorphopsia Measurement in Macular Diseases

Klinische Monatsblätter für Augenheilkunde ◽

10.1055/a-1252-2910 ◽

2020 ◽

Author(s):

Daniela Claessens ◽

Alexander K. Schuster ◽

Ronald V. Krüger ◽

Marian Liegl ◽

Laila Singh ◽

...

Keyword(s):

Macular Edema ◽

Correlation Coefficient ◽

Intraclass Correlation Coefficient ◽

Intraclass Correlation ◽

Measuring Method ◽

Repeated Measurements ◽

Retest Reliability ◽

Macular Diseases ◽

Computer Based ◽

Test Retest Reliability

AbstractIn this study, the test-retest-reliability as one aspect of reliability of metamorphopsia measurements using a computer-based measuring method was determined in patients with macular diseases. Metamorphopsia amplitude, position, and area were quantified using AMD – A Metamorphopsia Detector software (app4eyes GmbH & Co. KG, Germany) in patients with diabetic, myopic, or uveitic macular edema, intermediate or neovascular age-associated macular degeneration, epiretinal membrane, vitelliform maculopathy, Irvine-Gass syndrome, or macular edema due to venous retinal occlusion. The intraclass correlation coefficient (ICC) was calculated in order to determine the repeatability of two repeated measurements and was used as an indicator of the reliability of the measurements. In this study, metamorphopsia measurements were conducted on 36 eyes with macular diseases. Metamorphopsia measurements made using AMD – A Metamorphopsia Detector software were highly reliable and repeatable in patients with maculopathies. The intraclass correlation coefficient of all indices was excellent (0.95 – 0.97). For diseases of the vitreoretinal interface or macular diseases with intra- or subretinal edema, this metamorphopsia measurement represents a supplement for visual function testing in the clinic, as well as in clinical studies.

Download Full-text