What can routine laboratory test results tell us?

An essential part of the clinical decision-making process depends on interpreting different laboratory test values. This process is usually performed by scanning the results and looking for abnormal values. One of the core limitations is that the scan for abnormalities is done marginally, where every test is considered independently of the results of the other laboratory tests. Such practice can miss information in higher dimensions where a result of a laboratory test is assessed given the results of other laboratory tests.
Using multidimensional laboratory test results, we defined a new measure of `abnormality`. We showed that this measure might be useful for alerting physicians for patients at risk for illnesses otherwise overlooked.
Moreover, we showed that routine laboratory test results could be used to estimate biological age. Biological age may be of higher importance than chronological age, yet biological age is not trivial to estimate. We trained a machine-learning-based model to predict age using routine laboratory tests.
We trained an XGBoost model using data from about 472,000 individuals aged 37–82 years old. The model achieved an RMSE of 6.67 years. Subjects whose the model predicted to be younger than their actual age were found to be healthier as they had fewer diagnoses, fewer operations, and a lower prevalence of specific diseases than age-matched controls. On the other hand, subjects predicted to be older than their chronological age had no significant differences in the number of diagnoses, number of operations, and specific diseases than age-matched controls.