smdi_diagnose()
- the flagship functionThe smdi
main function is smdi_diagnose()
which calls all three group diagnostics, all of which are also
accessible individually.
smdi_diagnose()
builds on theoretical concepts developed
and validated in a comprehensive simulation study based on the workstream:
Approaches to Handling Partially Observed Confounder Data From Electronic Health Records (EHR) In Non-randomized Studies of Medication Outcomes.
A most minimal example could look like this (if you want to accept all of the default parameters).
smdi_diagnose(
data = smdi_data,
covar = NULL, # NULL includes all covariates with at least one NA
model = "cox",
form_lhs = "Surv(eventtime, status)"
) %>%
smdi_style_gt()
Covariate | ASMD (min/max)1 | p Hotelling1 | AUC2 | beta univariate (95% CI)3 | beta (95% CI)3 |
---|---|---|---|---|---|
ecog_cat | 0.030 (0.003, 0.071) | 0.783 | 0.522 | -0.06 (95% CI -0.16, 0.03) | -0.06 (95% CI -0.16, 0.03) |
egfr_cat | 0.216 (0.010, 0.485) | <.001 | 0.613 | 0.06 (95% CI -0.03, 0.15) | -0.01 (95% CI -0.10, 0.09) |
pdl1_num | 0.056 (0.016, 0.338) | <.001 | 0.517 | 0.12 (95% CI 0.01, 0.23) | 0.11 (95% CI -0.00, 0.22) |
p little: <.001, Abbreviations: ASMD = Median absolute standardized mean difference across all covariates, AUC = Area under the curve, beta = beta coefficient, CI = Confidence interval, max = Maximum, min = Minimum | |||||
1 Group 1 diagnostic: Differences in patient characteristics between patients with and without covariate | |||||
2 Group 2 diagnostic: Ability to predict missingness | |||||
3 Group 3 diagnostic: Assessment if missingness is associated with the outcome (univariate, adjusted) |