Albumin

Ketone bodies

## boundary (singular) fit: see help('isSingular')

IL-1R2

GTT Glucose Day 13-15

GTT Insulin Day 13-15

## fixed-effect model matrix is rank deficient so dropping 2 columns / coefficients
## fixed-effect model matrix is rank deficient so dropping 2 columns / coefficients

est sd cv se cv_mean endpoint
1.1505718 0.3165826 0.2751524 0.1827790 0.1588593 albumin
0.5793882 0.5539152 0.9560347 0.3198031 0.5519669 ketone
4.8093635 0.5041378 0.1048242 0.3564793 0.0741219 il1r2
0.1416424 0.0038564 0.0272264 0.0027269 0.0192520 glucose
-1.6498099 0.3614241 -0.2190701 0.2555654 -0.1549060 insulin

Interstudy reproducibility metrics

In this chunk, two metrics are computed across all studies for each given condition:

Simple Coefficient of Variation (CV) - i.e. no accounting for technical factors

Intra-Class Correlation Coefficient (ICC) - I used ICC2 from psych package (not ICC2k) as it seemed the most appropriate.

Note: data were not log-transformed.

## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
condition endpoint ICC CV
High HCT-HG albumin 0.5532544 0.3991602
High HCT-HG ketone 0.5658850 0.3049145
High HCT-HG il1r2 0.9235520 0.2457077
High HCT-HG insulin 0.3841673 0.8958944
High HCT-HG glucose 0.7534388 0.0482676
High HCT-NG albumin 0.5763549 0.2572888
High HCT-NG ketone 0.0030842 0.4971109
High HCT-NG il1r2 0.0501867 0.4131444
High HCT-NG insulin 0.3857564 1.5858973
High HCT-NG glucose 0.4457201 0.1167323
Low HCT-HG albumin 0.1402081 0.5324163
Low HCT-HG ketone 0.2091522 0.2342297
Low HCT-HG il1r2 0.7427219 0.5331425
Low HCT-HG insulin 0.5004977 0.7510045
Low HCT-HG glucose 0.8757985 0.0627144
Low HCT-NG albumin 0.2772667 0.3460743
Low HCT-NG ketone 0.0000000 0.3206010
Low HCT-NG il1r2 0.7392075 0.5451335
Low HCT-NG insulin 0.4985921 0.9275471
Low HCT-NG glucose 0.8607399 0.0707763

Here, we find the ICC and CV, but calculate it per-study and per-condition. I used ICC2 from psych package (not ICC2k) as it seemed the most appropriate.

Note: data were not log-transformed.

## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
## boundary (singular) fit: see help('isSingular')
condition endpoint study ICC CV
High HCT-HG albumin Study 1 0.5806151 0.3654457
High HCT-HG albumin Study 2 0.4794971 0.3705305
High HCT-HG albumin Study 3 0.8900719 0.2067851
High HCT-HG ketone Study 1 0.8434458 0.1630355
High HCT-HG ketone Study 2 0.4589309 0.3818775
High HCT-HG ketone Study 3 0.2938115 0.3800412
High HCT-HG il1r2 Study 2 0.9389784 0.1450800
High HCT-HG il1r2 Study 3 0.9772931 0.0733322
High HCT-HG insulin Study 1 0.9053309 0.5063891
High HCT-HG insulin Study 2 0.9461248 0.4809310
High HCT-HG glucose Study 1 0.7244151 0.0346589
High HCT-HG glucose Study 2 0.8819179 0.0513563
High HCT-NG albumin Study 1 0.5345070 0.2782865
High HCT-NG albumin Study 2 0.7913464 0.2378688
High HCT-NG albumin Study 3 0.6400615 0.3379235
High HCT-NG ketone Study 1 0.6181710 0.2052161
High HCT-NG ketone Study 2 0.3935942 0.1884120
High HCT-NG ketone Study 3 0.8358665 0.3039191
High HCT-NG il1r2 Study 2 0.0000000 0.2595635
High HCT-NG il1r2 Study 3 0.3579468 0.2225435
High HCT-NG insulin Study 1 0.9406554 0.1880961
High HCT-NG insulin Study 2 0.8130398 1.5858973
High HCT-NG glucose Study 1 0.9649046 0.0240061
High HCT-NG glucose Study 2 0.6443449 0.0635424
Low HCT-HG albumin Study 1 0.4997950 0.5885357
Low HCT-HG albumin Study 2 0.4514274 0.2486935
Low HCT-HG albumin Study 3 0.0675256 0.3445829
Low HCT-HG ketone Study 1 0.0930372 0.1366018
Low HCT-HG ketone Study 2 0.0970900 0.1859814
Low HCT-HG ketone Study 3 0.9355984 0.0762750
Low HCT-HG il1r2 Study 2 0.9870501 0.3957020
Low HCT-HG il1r2 Study 3 0.9799020 0.2463113
Low HCT-HG insulin Study 1 0.9866203 0.2293909
Low HCT-HG insulin Study 2 0.8584641 0.4050092
Low HCT-HG glucose Study 1 0.9275354 0.0540907
Low HCT-HG glucose Study 2 0.9266640 0.0343019
Low HCT-NG albumin Study 1 0.8050722 0.2518356
Low HCT-NG albumin Study 2 0.7433340 0.1708972
Low HCT-NG albumin Study 3 0.3490273 0.2626604
Low HCT-NG ketone Study 1 0.4219087 0.0761203
Low HCT-NG ketone Study 2 0.4636688 0.1321250
Low HCT-NG ketone Study 3 0.3546682 0.1540906
Low HCT-NG il1r2 Study 2 0.9687175 0.4633230
Low HCT-NG il1r2 Study 3 0.6647262 0.4986884
Low HCT-NG insulin Study 1 0.9534604 0.1995606
Low HCT-NG insulin Study 2 0.9274458 0.6163073
Low HCT-NG glucose Study 1 0.9077396 0.0935303
Low HCT-NG glucose Study 2 0.9601137 0.0287145

Appendix

Reproducibility

Analyses were conducted using the R Statistical language (version 4.3.1; R Core Team, 2023) on Windows 10 x64 (build 19045), using the packages magrittr (version 2.0.3; Bache S, Wickham H, 2022), lme4 (version 1.1.35.3; Bates D et al., 2015), Matrix (version 1.5.4.1; Bates D et al., 2023), lubridate (version 1.9.3; Grolemund G, Wickham H, 2011), plotrix (version 3.8.4; J L, 2006), lmerTest (version 3.1.3; Kuznetsova A et al., 2017), emmeans (version 1.10.1; Lenth R, 2024), report (version 0.5.8; Makowski D et al., 2023), tibble (version 3.2.1; Müller K, Wickham H, 2023), ggplot2 (version 3.5.1; Wickham H, 2016), forcats (version 1.0.0; Wickham H, 2023), stringr (version 1.5.1; Wickham H, 2023), tidyverse (version 2.0.0; Wickham H et al., 2019), readxl (version 1.4.3; Wickham H, Bryan J, 2023), dplyr (version 1.1.4; Wickham H et al., 2023), purrr (version 1.0.2; Wickham H, Henry L, 2023), readr (version 2.1.5; Wickham H et al., 2024), tidyr (version 1.3.1; Wickham H et al., 2024) and psych (version 2.4.3; William Revelle, 2024).

References