Figure 3 illustrates the differences in predicted scores by the logistic regression method and the ML method in terms of frequency distribution.
The sensitivity of the high-risk designation by the logistic regression method was significantly higher (266/601 = 44%) compared with the attribute-matching method (9/601 = 1.5%; 95% CI for difference of 42.5% = 39 to 47%).
As a result, the specificity of the high-risk designation by the attribute matching method was significantly higher (99%, 95% CI = 99 100%) than the specificity of the high-risk designation by the logistic regression method (79%, 95% CI = 78 80%).
Fourteen of the 24 genes found by the logistic regression method were also present in the list of the 40 best scoring genes (P < 10−3) by KM analysis (Additional file 5: Table S5) and five (IGLJ3, IGKC, IGHD, GM2A, DTNB) were in the top ten.
Nine of the top 24 genes found by the logistic regression method (IGHM, GM2A, DTNB, INPPL1, CD27, TNFRSF1, LAX1, IGKV4-1, IGHD) were in the list of 24 best scoring genes (P < 10−4) by this modified KM analysis, whereas the remaining 15 were not apparently related to the immune system.
Only two (INPPL1 and GM2A, which are perhaps not immune-related, Additional file 2: Table S2) of the 24 genes found by the logistic regression method were present among the 40 top scoring genes (4*10−5 > P < 1.4*10−3) found by KM analysis.
Multivariate analysis of factors associated with sarcopenia was performed by the logistic regression (backward method), which included the six most significant variables: age, cognitive function, admission profile (clinical or surgical), smoking, age ≥80 years and reports of physical inactivity pre-hospitalization (less than 2x per week).
The best fit for both sexes was provided by the Logistic model (see Methods), which describes age-specific mortality as a function of three different parameters, baseline mortality, rate-of-senescence and late-life deceleration in mortality (s).
