Regression coefficient-based scoring system should be used to assign weights to the risk index. Mehta, H. B., Mehta, V., Girman, C. J., Adhikari, D., & Johnson, M. L. Journal of Clinical Epidemiology, 79:22–28, November, 2016. doi abstract bibtex OBJECTIVE: Some previously developed risk scores contained a mathematical error in their construction: risk ratios were added to derive weights to construct a summary risk score. This study demonstrates the mathematical error and derived different versions of the Charlson comorbidity score (CCS) using regression coefficient-based and risk ratio-based scoring systems to further demonstrate the effects of incorrect weighting on performance in predicting mortality. STUDY DESIGN AND SETTING: This retrospective cohort study included elderly people from the Clinical Practice Research Datalink. Cox proportional hazards regression models were constructed for time to 1-year mortality. Weights were assigned to 17 comorbidities using regression coefficient-based and risk ratio-based scoring systems. Different versions of CCS were compared using Akaike information criteria (AIC), McFadden's adjusted R(2), and net reclassification improvement (NRI). RESULTS: Regression coefficient-based models (Beta, Beta10/integer, Beta/Schneeweiss, Beta/Sullivan) had lower AIC and higher R(2) compared to risk ratio-based models (HR/Charlson, HR/Johnson). Regression coefficient-based CCS reclassified more number of people into the correct strata (NRI range, 9.02-10.04) compared to risk ratio-based CCS (NRI range, 8.14-8.22). CONCLUSION: Previously developed risk scores contained an error in their construction adding ratios instead of multiplying them. Furthermore, as demonstrated here, adding ratios fail to even work adequately from a practical standpoint. CCS derived using regression coefficients performed slightly better than in fitting the data compared to risk ratio-based scoring systems. Researchers should use a regression coefficient-based scoring system to develop a risk index, which is theoretically correct.
@article{mehta_regression_2016,
title = {Regression coefficient-based scoring system should be used to assign weights to the risk index},
volume = {79},
issn = {1878-5921},
doi = {10.1016/j.jclinepi.2016.03.031},
abstract = {OBJECTIVE: Some previously developed risk scores contained a mathematical error in their construction: risk ratios were added to derive weights to construct a summary risk score. This study demonstrates the mathematical error and derived different versions of the Charlson comorbidity score (CCS) using regression coefficient-based and risk ratio-based scoring systems to further demonstrate the effects of incorrect weighting on performance in predicting mortality.
STUDY DESIGN AND SETTING: This retrospective cohort study included elderly people from the Clinical Practice Research Datalink. Cox proportional hazards regression models were constructed for time to 1-year mortality. Weights were assigned to 17 comorbidities using regression coefficient-based and risk ratio-based scoring systems. Different versions of CCS were compared using Akaike information criteria (AIC), McFadden's adjusted R(2), and net reclassification improvement (NRI).
RESULTS: Regression coefficient-based models (Beta, Beta10/integer, Beta/Schneeweiss, Beta/Sullivan) had lower AIC and higher R(2) compared to risk ratio-based models (HR/Charlson, HR/Johnson). Regression coefficient-based CCS reclassified more number of people into the correct strata (NRI range, 9.02-10.04) compared to risk ratio-based CCS (NRI range, 8.14-8.22).
CONCLUSION: Previously developed risk scores contained an error in their construction adding ratios instead of multiplying them. Furthermore, as demonstrated here, adding ratios fail to even work adequately from a practical standpoint. CCS derived using regression coefficients performed slightly better than in fitting the data compared to risk ratio-based scoring systems. Researchers should use a regression coefficient-based scoring system to develop a risk index, which is theoretically correct.},
language = {eng},
journal = {Journal of Clinical Epidemiology},
author = {Mehta, Hemalkumar B. and Mehta, Vinay and Girman, Cynthia J. and Adhikari, Deepak and Johnson, Michael L.},
month = nov,
year = {2016},
pmid = {27181564},
keywords = {Charlson comorbidity score, Regression coefficient, Risk ratio, Scoring algorithm, Scoring system, risk index},
pages = {22--28},
}
Downloads: 0
{"_id":"6zXjhG5FZn9FqHfWs","bibbaseid":"mehta-mehta-girman-adhikari-johnson-regressioncoefficientbasedscoringsystemshouldbeusedtoassignweightstotheriskindex-2016","downloads":0,"creationDate":"2017-01-15T23:18:10.800Z","title":"Regression coefficient-based scoring system should be used to assign weights to the risk index","author_short":["Mehta, H. B.","Mehta, V.","Girman, C. J.","Adhikari, D.","Johnson, M. L."],"year":2016,"bibtype":"article","biburl":"http://bibbase.org/zotero/hbmehta","bibdata":{"bibtype":"article","type":"article","title":"Regression coefficient-based scoring system should be used to assign weights to the risk index","volume":"79","issn":"1878-5921","doi":"10.1016/j.jclinepi.2016.03.031","abstract":"OBJECTIVE: Some previously developed risk scores contained a mathematical error in their construction: risk ratios were added to derive weights to construct a summary risk score. This study demonstrates the mathematical error and derived different versions of the Charlson comorbidity score (CCS) using regression coefficient-based and risk ratio-based scoring systems to further demonstrate the effects of incorrect weighting on performance in predicting mortality. STUDY DESIGN AND SETTING: This retrospective cohort study included elderly people from the Clinical Practice Research Datalink. Cox proportional hazards regression models were constructed for time to 1-year mortality. Weights were assigned to 17 comorbidities using regression coefficient-based and risk ratio-based scoring systems. Different versions of CCS were compared using Akaike information criteria (AIC), McFadden's adjusted R(2), and net reclassification improvement (NRI). RESULTS: Regression coefficient-based models (Beta, Beta10/integer, Beta/Schneeweiss, Beta/Sullivan) had lower AIC and higher R(2) compared to risk ratio-based models (HR/Charlson, HR/Johnson). Regression coefficient-based CCS reclassified more number of people into the correct strata (NRI range, 9.02-10.04) compared to risk ratio-based CCS (NRI range, 8.14-8.22). CONCLUSION: Previously developed risk scores contained an error in their construction adding ratios instead of multiplying them. Furthermore, as demonstrated here, adding ratios fail to even work adequately from a practical standpoint. CCS derived using regression coefficients performed slightly better than in fitting the data compared to risk ratio-based scoring systems. Researchers should use a regression coefficient-based scoring system to develop a risk index, which is theoretically correct.","language":"eng","journal":"Journal of Clinical Epidemiology","author":[{"propositions":[],"lastnames":["Mehta"],"firstnames":["Hemalkumar","B."],"suffixes":[]},{"propositions":[],"lastnames":["Mehta"],"firstnames":["Vinay"],"suffixes":[]},{"propositions":[],"lastnames":["Girman"],"firstnames":["Cynthia","J."],"suffixes":[]},{"propositions":[],"lastnames":["Adhikari"],"firstnames":["Deepak"],"suffixes":[]},{"propositions":[],"lastnames":["Johnson"],"firstnames":["Michael","L."],"suffixes":[]}],"month":"November","year":"2016","pmid":"27181564","keywords":"Charlson comorbidity score, Regression coefficient, Risk ratio, Scoring algorithm, Scoring system, risk index","pages":"22–28","bibtex":"@article{mehta_regression_2016,\n\ttitle = {Regression coefficient-based scoring system should be used to assign weights to the risk index},\n\tvolume = {79},\n\tissn = {1878-5921},\n\tdoi = {10.1016/j.jclinepi.2016.03.031},\n\tabstract = {OBJECTIVE: Some previously developed risk scores contained a mathematical error in their construction: risk ratios were added to derive weights to construct a summary risk score. This study demonstrates the mathematical error and derived different versions of the Charlson comorbidity score (CCS) using regression coefficient-based and risk ratio-based scoring systems to further demonstrate the effects of incorrect weighting on performance in predicting mortality.\nSTUDY DESIGN AND SETTING: This retrospective cohort study included elderly people from the Clinical Practice Research Datalink. Cox proportional hazards regression models were constructed for time to 1-year mortality. Weights were assigned to 17 comorbidities using regression coefficient-based and risk ratio-based scoring systems. Different versions of CCS were compared using Akaike information criteria (AIC), McFadden's adjusted R(2), and net reclassification improvement (NRI).\nRESULTS: Regression coefficient-based models (Beta, Beta10/integer, Beta/Schneeweiss, Beta/Sullivan) had lower AIC and higher R(2) compared to risk ratio-based models (HR/Charlson, HR/Johnson). Regression coefficient-based CCS reclassified more number of people into the correct strata (NRI range, 9.02-10.04) compared to risk ratio-based CCS (NRI range, 8.14-8.22).\nCONCLUSION: Previously developed risk scores contained an error in their construction adding ratios instead of multiplying them. Furthermore, as demonstrated here, adding ratios fail to even work adequately from a practical standpoint. CCS derived using regression coefficients performed slightly better than in fitting the data compared to risk ratio-based scoring systems. Researchers should use a regression coefficient-based scoring system to develop a risk index, which is theoretically correct.},\n\tlanguage = {eng},\n\tjournal = {Journal of Clinical Epidemiology},\n\tauthor = {Mehta, Hemalkumar B. and Mehta, Vinay and Girman, Cynthia J. and Adhikari, Deepak and Johnson, Michael L.},\n\tmonth = nov,\n\tyear = {2016},\n\tpmid = {27181564},\n\tkeywords = {Charlson comorbidity score, Regression coefficient, Risk ratio, Scoring algorithm, Scoring system, risk index},\n\tpages = {22--28},\n}\n\n","author_short":["Mehta, H. B.","Mehta, V.","Girman, C. J.","Adhikari, D.","Johnson, M. L."],"key":"mehta_regression_2016","id":"mehta_regression_2016","bibbaseid":"mehta-mehta-girman-adhikari-johnson-regressioncoefficientbasedscoringsystemshouldbeusedtoassignweightstotheriskindex-2016","role":"author","urls":{},"keyword":["Charlson comorbidity score","Regression coefficient","Risk ratio","Scoring algorithm","Scoring system","risk index"],"metadata":{"authorlinks":{}}},"search_terms":["regression","coefficient","based","scoring","system","used","assign","weights","risk","index","mehta","mehta","girman","adhikari","johnson"],"keywords":["charlson comorbidity score","regression coefficient","risk ratio","scoring algorithm","scoring system","risk index"],"authorIDs":[],"dataSources":["cQfjHaDeLndB9EhYA"]}