The objective of this study is to verify the statistical methods used to assess the compliance of medical devices that measure the same continuous variable in the medical literature. The share of the different statistical methods found in this review will also reflect the proportion of medical devices validated by these specific statistical methods in current clinical practice. A.S. Hedayat is a distinguished professor of Statistics and Senior Scholar at the Department of Mathematics, Statistics, and Computer Science, University of Illinois at Chicago. Its main areas of research and statistical advice are the design of experiments, medical and pharmaceutical statistics, environmental statistics, medicinal statistics, monitoring strategies, evaluation of agreements and sampling. In addition to the traditional periodic publications of more than 160 articles, he has also co-authored three books on statistics. Professor Hedayat is a Fellow of the Institute of Mathematical Statistics, a Fellow of the American Statistical Association and an elected member of the International Statistical Institute. He has worked in the editorial boards of numerous international magazines, including The Annals of Statistics, American Statistical Association, The American Statisticians and the Bulletin of Iranian Mathematical Society. All quotes identified during the search were uploaded to The FinalNote X1 software. Quotes were organized, duplicates were identified and deleted.
We excluded all studies with qualitative or categorical data, studies with different outcome units and association studies. Unpublished articles were not included in this review. The selection was made by two independent researchers. There was no disagreement between the two critics at the study selection stage. The third most popular method found in this review is the comparison of the measured values of two instruments. Coupled T-test is usually used to test the significant differences between the means of two data sets in order to evaluate the agreement . People have interpreted that non-significant results do not mean differences, so there is agreement between the two groups and vice versa. However, the coupled t-test with a non-significant result does not indicate a match.
The reason is that the value of the average value is influenced by the value of each data, resulting in inappropriate influence by extremely large or extremely small values. It is possible that a poor agreement between the two instruments may remain hidden in the distribution of differences, and therefore the two methods may seem unanimous. In an agreement study, we are not interested in the average reading of each instrument, but we are interested in each reading. It is important that each reading of the standard instrument is repeated by the new instrument. In addition, the importance is related to the performance of the study. We collected information on the statistical methods used to assess the compliance of the methodology or statistical analysis section, as well as by identifying the statistical methods that influenced the author`s conclusion of the agreement. This package calculates several statistics for compliance measurement, z.B average square deviation (MSD), overall deviation index (TDI) or correlation coefficient (CCC).