Automatic identification of variables in epidemiological datasets using logic regression.

Pubmed ID: 28407816

Pubmed Central ID: PMC5390441

Journal: BMC medical informatics and decision making

Publication Date: April 13, 2017

Affiliation: Faculty of Computer Science and Engineering, Frankfurt University of Applied Sciences, Frankfurt/Main, Germany.

MeSH Terms: Humans, Algorithms, Logistic Models, Prognosis, Carotid Artery Diseases, Predictive Value of Tests, Databases, Factual, Carotid Intima-Media Thickness, Meta-Analysis as Topic, Data Mining, Epidemiologic Factors, Medical Informatics Applications

Grants: N01 HC015103, N01 HC035129, N01HC55222, N01HC85079, N01HC85086, R01 DE013094, R37 NS029993, U01 HL080295

Authors: Franco OH, Lorenz MW, Polak JF, Kavousi M, Völzke H, Tuomainen TP, Sander D, Plichart M, Catapano AL, Kiechl S, Kiechl S, Rundek T, Desvarieux M, Lind L, Bots ML, Srinivasan SR, Sacco RL, Hofman A, Agewall S, Sitzer M, Steinmetz H, Dörr M, Schminke U, Poppert H, Bickel H, Kauhanen J, Ronkainen K, Empana JP, Ducimetiere P, Norata GD, Grigore L, Willeit J, Bokemark L, Liu J, Zhao D, Rosvall M, Dekker JM, Nijpels G, Stehouwer CD, Chien KL, Engström G, Hedblad B, Kitagawa K, Ikram MA, Okazaki S, Price JF, Robertson C, Willeit P, Bergström G, Iglseder B, Papagianni A, Yanez DN, Bülbül A, Su TC, Xie W, Lin HJ, McLachlan S, Schmidt C, Friera A, Suarez C, Gabriel R, Scheckenbach F, Pflug A, Catapano A, Ezhov M, Berenson G, Lee MS, Abdi NA, Orth A, Amato M, Kato A, Dimitriadis C, Przewlocki T, Lazarevic T, Bevc S, Staub D, Sirtori CR, Tripepi G, Beloqui O, Veglia F, Jovanovic A, Hojs R, Uthoff H, Park HW, Tremoli E, Kablak-Ziembicka A, Stolic R, Ekart R, Frauchiger B, Castelnuovo S, Zoccali C, Landecho MF, Bae JH, Baldassarre D

Cite As: Lorenz MW, Abdi NA, Scheckenbach F, Pflug A, Bülbül A, Catapano AL, Agewall S, Ezhov M, Bots ML, Kiechl S, Orth A, PROG-IMT study group. Automatic identification of variables in epidemiological datasets using logic regression. BMC Med Inform Decis Mak 2017 Apr 13;17(1):40.

Studies:

Abstract

BACKGROUND: For an individual participant data (IPD) meta-analysis, multiple datasets must be transformed in a consistent format, e.g. using uniform variable names. When large numbers of datasets have to be processed, this can be a time-consuming and error-prone task. Automated or semi-automated identification of variables can help to reduce the workload and improve the data quality. For semi-automation high sensitivity in the recognition of matching variables is particularly important, because it allows creating software which for a target variable presents a choice of source variables, from which a user can choose the matching one, with only low risk of having missed a correct source variable. METHODS: For each variable in a set of target variables, a number of simple rules were manually created. With logic regression, an optimal Boolean combination of these rules was searched for every target variable, using a random subset of a large database of epidemiological and clinical cohort data (construction subset). In a second subset of this database (validation subset), this optimal combination rules were validated. RESULTS: In the construction sample, 41 target variables were allocated on average with a positive predictive value (PPV) of 34%, and a negative predictive value (NPV) of 95%. In the validation sample, PPV was 33%, whereas NPV remained at 94%. In the construction sample, PPV was 50% or less in 63% of all variables, in the validation sample in 71% of all variables. CONCLUSIONS: We demonstrated that the application of logic regression in a complex data management task in large epidemiological IPD meta-analyses is feasible. However, the performance of the algorithm is poor, which may require backup strategies.