Common pitfalls in statistical analysis: Measures of agreement

Ranganathan, Priya; Pramesh, C S; Aggarwal, Rakesh

doi:10.4103/picr.picr_123_17

Cited by 321 publications

(224 citation statements)

References 5 publications

Supporting

Mentioning

220

Contrasting

Unclassified

Order By: Relevance

“…The inter-rater agreement 49 was moderate for CD90 (Cohen's kappa = 0.506) and PDGFRb (kappa = 0.599), substantial for ASMA (kappa = 0.663) and PDGFRa (kappa = 0.620) and excellent for FAP (kappa = 0.890) and CD8a (kappa = 0.809) ( Supplementary Figs. 1 and 2).…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Fibroblasts in urothelial bladder cancer define stroma phenotypes that are associated with clinical outcome

Mezheyeuski

Segersten

Leiss

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Little attention was given to the interaction between tumor and stromal cells in urothelial bladder carcinoma (UBC). While recent studies point towards the existence of different fibroblast subsets, no comprehensive analyses linking different fibroblast markers to UBC patient survival have been performed so far. Through immunohistochemical analysis of five selected fibroblast markers, namely alpha smooth muscle actin (ASMA), CD90/Thy-1, fibroblast activation protein (FAP), platelet derived growth factor receptor-alpha and-beta (PDGFRa,-b), this study investigates their association with survival and histopathological characteristics in a cohort of 344 UBC patients, involving both, muscleinvasive and non-muscle-invasive cases. The data indicates that combinations of stromal markers are more suited to identify prognostic patient subgroups than single marker analysis. Refined stromamarker-based patient stratification was achieved through cluster analysis and identified a FAPdominant patient cluster as independent marker for shorter 5-year-survival (HR(95% CI)2.25(1.08-4.67), p = 0.030). Analyses of interactions between fibroblast and CD8a-status identified a potential minority of cases with CD90-defined stroma and high CD8a infiltration showing a good prognosis of more than 80% 5-year-survival. Presented analyses point towards the existence of different stroma-cell subgroups with distinct tumor-modulatory properties and motivate further studies aiming to better understand the molecular tumor-stroma crosstalk in UBC. The estimated incidence of urinary bladder cancer (UBC) worldwide was about 550.000 new cases and 200.000 deaths in 2018 1 , with highest incidences indicated in developed areas 2. Given that the prevalence of UBC is several times higher than the incidence, it represents a major burden for health care systems. Carcinomas originating from the urothelium, the inner surface of the bladder, represent the most common type of bladder cancer 3,4. Approximately 70% of all newly diagnosed cases are classified as non-muscle-invasive bladder cancer disease, while about 30% of patients have already muscle-invasive tumors with 10-15% of them being metastatic 5-8. Reported 5-year recurrence rates of non-muscle invasive cases range between 40-70% among the highest recurrence rates of solid tumors 3,9. Multidisciplinary approaches have been established in order to improve patient management and to decrease the mortality rate, but still treatment strategies for UBC remain controversial. A better understanding of the mechanisms underlying tumor progression is crucial and novel prognostic and predictive markers as well as therapeutic targets need to be identified 10,11. Recent research efforts on UBC aimed mostly to decipher molecular drivers of an invasive tumor phenotype through description of genetic varieties, and with the introduction of checkpoint inhibitory immunotherapy, awareness has also risen for tumor infiltrating leukocytes 12,13. However, little attention was given the interaction of tumor cells and non-leukoc...

show abstract

Section: Resultsmentioning

confidence: 99%

“…To assess the degree that raters provided consistency in their scorings the inter-rater agreement was calculated with Cohen's kappa (squared weightage) 49 using the irr package for R. For subsequent analyses, the average score between the two raters was calculated resulting in a data range from 0 to 9 with 0.5 intervals (see also Supplementary Figs. 1 and 2).…”

mentioning

confidence: 99%

Fibroblasts in urothelial bladder cancer define stroma phenotypes that are associated with clinical outcome

Mezheyeuski

Segersten

Leiss

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…We used Fleiss's kappa statistics (MAGREE.SAS) to examine the agreement amongst all observers (interoperator agreement). 13,14 We defined the level of agreement according to kappa value as follows: poor agreement, k ≤ 0; slight agreement, k 0 to 0.2; fair agreement, k 0.2 to 0.4; moderate agreement, k 0.4 to 0.6; substantial agreement, k 0.6 to 0.8; almost perfect agreement, k 0.8 to 1. SEs and 95% confidence intervals (CI) for the stomach position grades are also presented.…”

Section: Discussionmentioning

confidence: 99%

Standardization and reproducibility of sonographic stomach position grades in fetuses with congenital diaphragmatic hernia

Ibirogba

Novoa

Sutton

et al. 2019

J of Clinical Ultrasound

View full text Add to dashboard Cite

Objectives The purpose of this study was to evaluate the reproducibility of stomach position grading in congenital diaphragmatic hernia (CDH) as proposed by Cordier et al and Basta et al after standardization of the methods at our center. Methods We collected sonographic images from 23 fetuses with left‐sided CDH at our center from 2010 to 2018. Nine operators (one maternal fetal medicine expert and eight sonographers) reviewed the selected images and graded the stomach position according to the methods of Cordier et al and Basta et al. We assessed the interoperator agreement with Fleiss's kappa statistics. Results Overall agreement amongst all operators was moderate for both methods proposed by Cordier et al (k = 0.60, SE 0.07, 95% CI 0.47‐0.73, P < .0001) and Basta et al (k = 0.60, SE 0.06, 95% CI 0.47‐0.73, P < .0001). Interoperator agreement was moderate for grade 3 with the method by Cordier et al (k = 0.45, SE 0.09, 95% CI 0.27‐0.64, P < .0001) and fair for grade 4 with the method by Basta et al (k = 0.33, SE 0.08, 95% CI 0.18‐0.49 P < .0001). Conclusions Our study demonstrates a fair to moderate interoperator agreement of the stomach position grading methods proposed in the literature after standardization of the methods at our center. Further multicenter studies are needed to confirm our results.

show abstract

“…Meanwhile, when agreement was analyzed according to the K index, the modified Friedewald equation showed the highest agreement in the overall analysis and the subgroup of patients with TG below 150 mg/dL, aged over 65 years, female subjects, and with diabetes (Table 3). However, there could be differences in findings between two statistical methods based on scale measurements, as reported previously 23 . The high K index of the Friedewald equation can be attributed to the definition of the study population.…”

Section: Discussionmentioning

confidence: 72%

Comparison of three equations for estimating low‐density lipoprotein‐cholesterol in the rural northeastern region of Thailand

Sonsok¹,

Sarakarn

Sanchaisuriya

2020

Clinical Laboratory Analysis

View full text Add to dashboard Cite

Background Cardiovascular disease is the most common cause of death worldwide, and the detection of LDL‐C contributes to reducing risks. However, the LDL‐C is rarely evaluated according to the gold standard method because it is costly and time‐consuming. This study aimed to determine the agreement of LDL‐C among three equations, namely Friedewald's equation, Puavilai's equation, and Dansethakul's equation. Methods A cross‐sectional descriptive study. Results Using the data of lipid measurement from a specific group of people in the remote rural area, we found that the Thai equations have more superior agreement with direct measurement than the Friedewald equation (ICC = 0.870, 95% CI = 0.857‐0.882) when the agreement of continuous data was used for total analysis. Although the categorical analysis that gave better agreement was from Friedewald equation (K index = 0.730, 95% CI = 0.720‐0.751), the findings from this study confirmed the population‐specific use of Pauvilai's equation and Dansethakul's equation for determining the LDL‐C. Conclusion Pauvilai's equation showed better agreement with direct measurement for LDL‐C. Thus, it could be considered as an alternative for the direct method, particularly in laboratories in rural areas in Thailand.

show abstract

Common pitfalls in statistical analysis: Measures of agreement

Cited by 321 publications

References 5 publications

Fibroblasts in urothelial bladder cancer define stroma phenotypes that are associated with clinical outcome

Fibroblasts in urothelial bladder cancer define stroma phenotypes that are associated with clinical outcome

Standardization and reproducibility of sonographic stomach position grades in fetuses with congenital diaphragmatic hernia

Comparison of three equations for estimating low‐density lipoprotein‐cholesterol in the rural northeastern region of Thailand

Contact Info

Product

Resources

About