Skip to contents

Screening Outcomes and Predictors with MakeUnivariateRegressionTable()

Source: vignettes/MakeUnivariateRegressionTable.qmd

1 Overview

MakeUnivariateRegressionTable() screens outcome-predictor relationships and returns a report-facing table, a detail table, a tidy results dataframe, fitted models, and metadata.

Use it when you want to quickly ask:

  • Which predictors are associated with one or more outcomes?
  • Do associations persist after covariate adjustment?
  • Are the displayed labels and logistic event levels what I think they are?

The function chooses the model family automatically:

  • Numeric outcomes use linear regression.
  • Two-level factor, character, or logical outcomes use logistic regression.
  • Logistic estimates are exponentiated by default and reported as odds ratios.

Note

MakeUnivariateRegressionTable() was previously named UnivariateRegressionTable(). The old name still works as a backwards-compatible synonym, so existing scripts continue to run.

2 Load packages

3 Load example data

The examples use SampleData and SampleVariableTypes, then apply labels and recoding with RevalueData().

data("SampleData")
data("SampleVariableTypes")

RevaluedObj <- RevalueData(
  SampleData,
  SampleVariableTypes
)

df_Revalued <- RevaluedObj$RevaluedData

4 Basic linear screening

This example screens several labelled biomarkers against two predictors.

vars_Outcomes <- c(
  "Calbindin",
  "Ferritin",
  "MMP7",
  "Sortilin"
)

vars_Predictors <- c(
  "Diagnosis",
  "age"
)
Reg_Obj_Un <- MakeUnivariateRegressionTable(
  data = df_Revalued,
  outcome_vars = vars_Outcomes,
  predictor_vars = vars_Predictors,
  Standardize = TRUE
)

5 Formatted and detailed tables

FormattedTable is the report-facing wide table: each outcome has its own column spanner with an estimate (95% CI) and p-value. Significant effects carry stars and are bolded.

Reg_Obj_Un$FormattedTable
Calbindin
Ferritin
Matrix metalloproteinase 7
Sortilin
Estimate (95% CI) p-value Estimate (95% CI) p-value Estimate (95% CI) p-value Estimate (95% CI) p-value
Diagnosis : Impaired 0.223 (-0.0179, 0.464) 0.069 0.259 (0.0187, 0.5)* 0.035 0.551 (0.316, 0.786)**** <0.001 0.42 (0.182, 0.658)*** <0.001
Age 0.0517 (-0.0581, 0.161) 0.36 0.0303 (-0.0799, 0.14) 0.59 -0.0464 (-0.156, 0.0632) 0.41 0.0668 (-0.0433, 0.177) 0.23

LargeTable uses the same wide outcome layout but keeps N, estimate, standard error, confidence limits, and p-value as separate columns for QC.

Reg_Obj_Un$LargeTable
Calbindin
Ferritin
Matrix metalloproteinase 7
Sortilin
N Estimate SE 95% CI Low 95% CI High p-value N Estimate SE 95% CI Low 95% CI High p-value N Estimate SE 95% CI Low 95% CI High p-value N Estimate SE 95% CI Low 95% CI High p-value
Diagnosis : Impaired 333.000 0.223 0.123 −0.018 0.464 0.069 333.000 0.259 0.122 0.019 0.500 0.035 333.000 0.551 0.119 0.316 0.786 0.000 333.000 0.420 0.121 0.182 0.658 0.001
Age 322.000 0.052 0.056 −0.058 0.161 0.355 322.000 0.030 0.056 −0.080 0.140 0.589 322.000 −0.046 0.056 −0.156 0.063 0.405 322.000 0.067 0.056 −0.043 0.177 0.234

6 Tidy results dataframe

Results holds the underlying numbers as a plain dataframe: one row per estimated term, with the estimate, standard error, confidence interval, p-value, and the labels used in the tables.

Reg_Obj_Un$Results
    Outcome               OutcomeLabel OutcomeFamily EffectType Predictor
1 Calbindin                  Calbindin        linear   Estimate Diagnosis
2 Calbindin                  Calbindin        linear   Estimate       age
3  Ferritin                   Ferritin        linear   Estimate Diagnosis
4  Ferritin                   Ferritin        linear   Estimate       age
5      MMP7 Matrix metalloproteinase 7        linear   Estimate Diagnosis
6      MMP7 Matrix metalloproteinase 7        linear   Estimate       age
7  Sortilin                   Sortilin        linear   Estimate Diagnosis
8  Sortilin                   Sortilin        linear   Estimate       age
  PredictorLabel              Term    Level            TermLabel   N
1      Diagnosis DiagnosisImpaired Impaired Diagnosis : Impaired 333
2            Age               age     <NA>                  Age 322
3      Diagnosis DiagnosisImpaired Impaired Diagnosis : Impaired 333
4            Age               age     <NA>                  Age 322
5      Diagnosis DiagnosisImpaired Impaired Diagnosis : Impaired 333
6            Age               age     <NA>                  Age 322
7      Diagnosis DiagnosisImpaired Impaired Diagnosis : Impaired 333
8            Age               age     <NA>                  Age 322
     Estimate   StdError     ConfLow   ConfHigh       PValue Significant
1  0.22318256 0.12254159 -0.01787596 0.46424108 6.946687e-02       FALSE
2  0.05166760 0.05580595 -0.05812530 0.16146050 3.552248e-01       FALSE
3  0.25934156 0.12232632  0.01870650 0.49997662 3.474297e-02        TRUE
4  0.03027304 0.05597431 -0.07985109 0.14039718 5.889954e-01       FALSE
5  0.55090515 0.11937344  0.31607888 0.78573143 5.630656e-06        TRUE
6 -0.04643922 0.05571349 -0.15605022 0.06317177 4.051640e-01       FALSE
7  0.41966369 0.12097458  0.18168773 0.65763966 5.914370e-04        TRUE
8  0.06679376 0.05597311 -0.04332801 0.17691553 2.336285e-01       FALSE
  ReferenceValue
1              0
2              0
3              0
4              0
5              0
6              0
7              0
8              0

Because it is an ordinary dataframe, you can filter, sort, or export it directly, and pass it (or a subset of it) straight to PlotForestFromTable().

Reg_Obj_Un$Results %>%
  filter(Significant) %>%
  arrange(PValue)
   Outcome               OutcomeLabel OutcomeFamily EffectType Predictor
1     MMP7 Matrix metalloproteinase 7        linear   Estimate Diagnosis
2 Sortilin                   Sortilin        linear   Estimate Diagnosis
3 Ferritin                   Ferritin        linear   Estimate Diagnosis
  PredictorLabel              Term    Level            TermLabel   N  Estimate
1      Diagnosis DiagnosisImpaired Impaired Diagnosis : Impaired 333 0.5509052
2      Diagnosis DiagnosisImpaired Impaired Diagnosis : Impaired 333 0.4196637
3      Diagnosis DiagnosisImpaired Impaired Diagnosis : Impaired 333 0.2593416
   StdError   ConfLow  ConfHigh       PValue Significant ReferenceValue
1 0.1193734 0.3160789 0.7857314 5.630656e-06        TRUE              0
2 0.1209746 0.1816877 0.6576397 5.914370e-04        TRUE              0
3 0.1223263 0.0187065 0.4999766 3.474297e-02        TRUE              0

7 Covariate adjustment

Use covariates when each screen should adjust for the same covariates.

Reg_Obj_Un_Covar <- MakeUnivariateRegressionTable(
  data = df_Revalued,
  outcome_vars = vars_Outcomes,
  predictor_vars = vars_Predictors,
  covariates = "sex",
  Standardize = TRUE
)

Reg_Obj_Un_Covar$FormattedTable
Calbindin
Ferritin
Matrix metalloproteinase 7
Sortilin
Estimate (95% CI) p-value Estimate (95% CI) p-value Estimate (95% CI) p-value Estimate (95% CI) p-value
Diagnosis : Impaired 0.278 (0.0401, 0.516)* 0.022 0.22 (-0.0203, 0.46) 0.073 0.522 (0.287, 0.758)**** <0.001 0.429 (0.189, 0.669)*** <0.001
Age 0.0325 (-0.076, 0.141) 0.56 0.0464 (-0.063, 0.156) 0.4 -0.034 (-0.144, 0.0754) 0.54 0.0662 (-0.0446, 0.177) 0.24

8 Many predictors for one outcome

The same function also works when the question is one outcome against many candidate predictors.

Reg_Obj_ManyPredictors <- MakeUnivariateRegressionTable(
  data = df_Revalued,
  outcome_vars = "Sortilin",
  predictor_vars = c("Diagnosis", "age", "Calbindin", "Ferritin", "MMP7"),
  Standardize = TRUE
)

Reg_Obj_ManyPredictors$FormattedTable
Sortilin
Estimate (95% CI) p-value
Diagnosis : Impaired 0.42 (0.182, 0.658)*** <0.001
Age 0.0668 (-0.0433, 0.177) 0.23
Calbindin 0.484 (0.39, 0.579)**** <0.001
Ferritin 0.617 (0.532, 0.702)**** <0.001
Matrix metalloproteinase 7 0.175 (0.0689, 0.282)** 0.0013

9 Logistic outcomes

Binary outcomes use logistic regression automatically. By default, estimates are exponentiated and shown as odds ratios.

Reg_Obj_Logistic <- MakeUnivariateRegressionTable(
  data = df_Revalued,
  outcome_vars = "Diagnosis",
  predictor_vars = c("age", "Calbindin", "Ferritin", "MMP7"),
  Standardize = TRUE
)

Reg_Obj_Logistic$FormattedTable
Diagnosis
Odds ratio (95% CI) p-value
Age 0.929 (0.728, 1.18) 0.55
Calbindin 1.25 (0.981, 1.6) 0.071
Ferritin 1.3 (1.02, 1.66)* 0.036
Matrix metalloproteinase 7 1.8 (1.38, 2.35)**** <0.001

Always check the metadata for logistic models. It records the reference level and event level.

Reg_Obj_Logistic$Metadata$Outcomes
    Outcome OutcomeLabel OutcomeFamily ReferenceLevel EventLevel
1 Diagnosis    Diagnosis      logistic        Control   Impaired

10 Access fitted models

Fitted models are returned by outcome, then predictor.

names(Reg_Obj_Un$ModelSummaries)
NULL
names(Reg_Obj_ManyPredictors$ModelSummaries$Sortilin)
NULL
summary(Reg_Obj_ManyPredictors$ModelSummaries$Sortilin$Ferritin)
Length  Class   Mode
     0   NULL   NULL

11 Reproducibility 

# save.image("MakeUnivariateRegressionTable_workspace.RData")
print(sessionInfo())
R version 4.6.1 (2026-06-24)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 24.04.4 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.26.so;  LAPACK version 3.12.0

locale:
 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8
 [4] LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8    LC_MESSAGES=C.UTF-8
 [7] LC_PAPER=C.UTF-8       LC_NAME=C              LC_ADDRESS=C
[10] LC_TELEPHONE=C         LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C

time zone: UTC
tzcode source: system (glibc)

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] dplyr_1.2.1           SciDataReportR_21.0.0

loaded via a namespace (and not attached):
 [1] gtable_0.3.6           xfun_0.60              bayestestR_0.18.1
 [4] ggplot2_4.0.3          insight_1.5.2          rstatix_1.1.0
 [7] lattice_0.22-9         paletteer_1.7.0        vctrs_0.7.3
[10] tools_4.6.1            generics_0.1.4         datawizard_1.3.1
[13] tibble_3.3.1           pkgconfig_2.0.3        RColorBrewer_1.1-3
[16] correlation_0.8.8      S7_0.2.2               RcppParallel_6.2.0
[19] gt_1.3.0               lifecycle_1.0.5        compiler_4.6.1
[22] farver_2.1.2           carData_3.0-6          snakecase_0.11.1
[25] sass_0.4.10            htmltools_0.5.9        yaml_2.3.12
[28] Formula_1.2-6          pillar_1.11.1          car_3.1-5
[31] tidyr_1.3.2            statsExpressions_2.0.0 abind_1.4-8
[34] tidyselect_1.2.1       sjlabelled_1.2.0       digest_0.6.39
[37] mvtnorm_1.4-2          gtsummary_2.5.1        purrr_1.2.2
[40] rematch2_2.1.2         forcats_1.0.1          ggstatsplot_1.0.0
[43] labelled_2.16.0        fastmap_1.2.0          grid_4.6.1
[46] cli_3.6.6              magrittr_2.0.5         patchwork_1.3.2
[49] dichromat_2.0-1        broom_1.0.13           withr_3.0.3
[52] scales_1.4.0           backports_1.5.1        estimability_2.0.0
[55] rmarkdown_2.31         emmeans_2.0.4          otel_0.2.0
[58] hms_1.1.4              coda_0.19-4.1          evaluate_1.0.5
[61] knitr_1.51             haven_2.5.5            parameters_0.29.2
[64] rstantools_2.7.0       rlang_1.3.0            xtable_1.8-8
[67] glue_1.8.1             xml2_1.6.0             jsonlite_2.0.0
[70] effectsize_1.0.3       R6_2.6.1               fs_2.1.0