Developing high‑performance prediction models for medical outcomes
Published in Biomedical Research and Statistics
A prediction model can be considered to have high performance when it gives correct prediction in at least 90% cases for qualitative outcome and at least 90% differences between the observed and predicted values are within clinical tolerance in the case of quantitative outcome. A model with an accuracy between 80% and 90% can be possibly tolerated in some situations, but any model with less accuracy implies an unacceptably large error in clinical applications. Most of the models developed so far do not meet these criteria. Moreover, prediction of the unknown is confused with classification of the known. Developing high‑performance models requires a lot of extra efforts that are rarely seen in the present endeavors. For this, it is imperative that a large number of known and suspected predictors are considered and complexity in terms of interactions and nonlinearity is accepted. Models with many predictors is not a big limitation now because of wide availability of enormous computing power. More rigorous validation is needed than done now. Artificial intelligence‑based models can be linked to the relevant literature so that they continuously update with the new development under the data science paradigm. This article (https://pmc.ncbi.nlm.nih.gov/articles/PMC13390955/) highlights the problems with many of the existing models and describes steps to develop a high‑performance model for medical outcomes. Many of the advices we give are rarely available in the literature.
Prof Indrayan is the only biostatistician in India as of 2025 to be honored with the fellowship of the Indian Academy of Sciences (of Dr CV Raman fame). He completed MS and PhD from the Ohio State University in a record time of three years. He established the first Department of Biostatistics and Medical Informatics in India – that too in a non-PG medical institution – in the year 1995, which was later emulated by SGPGI (Lko) and JIPMER (Py). He retired as Head of this department in 2010 and is currently (2025) working as a Biostatistics Consultant at Max Healthcare.
Prof Indrayan is respected as the top biostatistician in the country with more than 300 publications, including a large number of his own research on medical biostatistics. He has authored 6 books, three of which are published in the US for international audience, and more than 20 chapters. He has completed nearly 40 projects for the World Health Organization, the World Bank, and the UNAIDS. He has also been running an online course for an international audience for the past 20 years in collaboration with a US institute. Prof Indrayan has been a Visiting Faculty at the Ohio State University and a Visiting Research Scientist at the University of Massachusetts. He has presented his work in 17 international conferences.
Prof Indrayan is serving biostatistics in several ways. He has developed a medically integrated curriculum of biostatistics for medical under- and postgraduate students that medicalizes the subject of biostatistics. His publications on Personalized Statistical Medicine, medical uncertainties, and the conundrum of P-values are widely acclaimed. Among his original ideas are the smoking index, objective measurement of an individual’s positive health, scoring for assessing the quality of the process of medical research from choosing a problem to its publication, a simple and robust method of measurement of agreement without Bland-Altman limits, correct method for assessing predictive performance of a model, and ‘promote heart disease, stop cancer’ by research on choosing the cause of death in old age.
