Load library and the data
# Load library
library(hcruR)
# Load sample data
data(hcru_sample_data)
head(hcru_sample_data)
#> # A tibble: 6 × 11
#> patient_id cohort encounter_id care_setting visit_date admission_date
#> <chr> <chr> <chr> <chr> <date> <date>
#> 1 P_1001 control E_645327 IP 2023-08-23 2023-08-23
#> 2 P_1001 control E_591096 ED 2023-08-27 NA
#> 3 P_1001 control E_864419 OP 2023-09-25 NA
#> 4 P_1001 control E_379584 IP 2023-10-03 2023-10-03
#> 5 P_1001 control E_617025 ED 2024-02-07 NA
#> 6 P_1001 control E_379047 ED 2024-03-01 NA
#> # ℹ 5 more variables: discharge_date <date>, encounter_date <date>,
#> # index_date <date>, period <chr>, cost_usd <dbl>Generate HCRU summary using dplyr (this can be used for create HCRU plots)
# Step 1: Estimate HCRU
hcru_summary <- estimate_hcru(data = hcru_sample_data,
cohort_col = "cohort",
patient_id_col = "patient_id",
admit_col = "admission_date",
discharge_col = "discharge_date",
index_col = "index_date",
visit_col = "visit_date",
encounter_id_col = "encounter_id",
setting_col = "care_setting",
cost_col = "cost_usd",
readmission_col = "readmission",
time_window_col = "period",
los_col = "length_of_stay",
custom_var_list = NULL,
pre_days = 180,
post_days = 365,
readmission_days_rule = 30,
group_var = "cohort",
test = NULL,
gt_output = FALSE)
hcru_summary
#> $`Summary by settings using dplyr`
#> # A tibble: 12 × 10
#> cohort care_setting period Patients Visits Cost Avg_visits_per_patient
#> <chr> <chr> <chr> <int> <int> <dbl> <dbl>
#> 1 control ED post 65 101 262128. 1.55
#> 2 control ED pre 50 57 150620. 1.14
#> 3 control IP post 78 127 1877868. 1.63
#> 4 control IP pre 53 66 1066097. 1.25
#> 5 control OP post 67 111 116931. 1.66
#> 6 control OP pre 50 61 62427. 1.22
#> 7 treatment ED post 78 126 335850. 1.62
#> 8 treatment ED pre 51 63 153831. 1.24
#> 9 treatment IP post 68 100 1523437. 1.47
#> 10 treatment IP pre 65 80 1333135. 1.23
#> 11 treatment OP post 76 126 133056. 1.66
#> 12 treatment OP pre 56 71 67378. 1.27
#> # ℹ 3 more variables: Avg_cost_per_patient <dbl>, Avg_LOS <dbl>,
#> # Readmit_30d_Rate <dbl>Generate HCRU summary using gtsummary (a publication ready output)
# Step 1: Estimate HCRU
hcru_summary_gt <- estimate_hcru(data = hcru_sample_data,
cohort_col = "cohort",
patient_id_col = "patient_id",
admit_col = "admission_date",
discharge_col = "discharge_date",
index_col = "index_date",
visit_col = "visit_date",
encounter_id_col = "encounter_id",
setting_col = "care_setting",
cost_col = "cost_usd",
readmission_col = "readmission",
time_window_col = "period",
los_col = "length_of_stay",
custom_var_list = NULL,
pre_days = 180,
post_days = 365,
readmission_days_rule = 30,
group_var = "cohort",
test = NULL,
gt_output = TRUE)
hcru_summary_gt$`Summary by settings using gtsummary`| Variable | N |
Group
|
p-value2 | ||
|---|---|---|---|---|---|
|
Overall N = 1,0891 |
control N = 5231 |
treatment N = 5661 |
|||
| cost_usd | 1,089 | 0.2 | |||
| Mean (SD) | 6,504 (8,063) | 6,761 (8,149) | 6,266 (7,982) | ||
| Median (Q1, Q3) | 2,522 (1,127, 9,433) | 2,775 (1,122, 9,649) | 2,276 (1,127, 8,534) | ||
| Q1 | 1,127 | 1,122 | 1,127 | ||
| Q3 | 9,433 | 9,649 | 8,534 | ||
| Min - Max | 104 - 29,990 | 111 - 29,779 | 104 - 29,990 | ||
| length_of_stay | 373 | 0.7 | |||
| Mean (SD) | 6.42 (2.91) | 6.36 (2.88) | 6.49 (2.96) | ||
| Median (Q1, Q3) | 6.00 (4.00, 9.00) | 6.00 (4.00, 9.00) | 6.50 (4.00, 9.00) | ||
| Q1 | 4.00 | 4.00 | 4.00 | ||
| Q3 | 9.00 | 9.00 | 9.00 | ||
| Min - Max | 2.00 - 11.00 | 2.00 - 11.00 | 2.00 - 11.00 | ||
| Unknown | 716 | 330 | 386 | ||
| readmission | 1,089 | 47 (4.3%) | 30 (5.7%) | 17 (3.0%) | 0.027 |
| 1 n (%) | |||||
| 2 Wilcoxon rank sum test; Pearson’s Chi-squared test | |||||
Generate the HCRU plot: average visits per patient by cohort and time-line
# Plot results
p_avg_visit <- plot_hcru(summary_df = hcru_summary$`Summary by settings using dplyr`,
x_var = "period",
y_var = "Avg_visits_per_patient",
cohort_col = "cohort",
facet_var = "care_setting",
facet_var_n = 3,
title = "Per patient average visits by domain and cohort",
x_lable = "Healthcare Setting (Domain)",
y_lable = "Average visits",
fill_lable = "Cohort"
)
p_avg_visit
Generate the HCRU plot for average costs per patient by cohort and time-line
# Plot results
p_avg_cost <- plot_hcru(summary_df = hcru_summary$`Summary by settings using dplyr`,
x_var = "period",
y_var = "Avg_cost_per_patient",
cohort_col = "cohort",
facet_var = "care_setting",
facet_var_n = 3,
title = "Per patient average cost by domain and cohort",
x_lable = "Healthcare Setting (Domain)",
y_lable = "Average costs",
fill_lable = "Cohort"
)
p_avg_cost