How many patients were in the hospital at 10 AM yesterday?
How many were in during each 15 minute spell between 2pm and 6pm?
How many were in during the last week, by hour?
This package aims to make answering these questions easier and quicker.
No SQL? No problem!
If you have time in, time out, a unique patient identifier, and optionally, a grouping variable to track moves between departments, this package will tell you how many patients were ‘IN’ at any time, at whatever granularity you need.
patientcounter is not on CRAN yet, so install the development version from GitHub with:
# install.packages("remotes") # if not already installed
remotes::install_github("johnmackintosh/patientcounter")
Obtain data for each individual patient, by hour, for each hour of their
stay.
Note we are restricting the outputs to keep this readable
library(patientcounter)
patient_count <- interval_census(beds,
identifier = 'patient',
admit = 'start_time',
discharge = 'end_time',
group_var = 'bed',
time_unit = '1 hour',
results = "patient",
uniques = TRUE)
head(patient_count)
#> bed patient start_time end_time interval_beginning
#> 1: A 1 2020-01-01 09:34:00 2020-01-01 10:34:00 2020-01-01 09:00:00
#> 2: B 5 2020-01-01 09:45:00 2020-01-01 14:45:00 2020-01-01 09:00:00
#> 3: A 1 2020-01-01 09:34:00 2020-01-01 10:34:00 2020-01-01 10:00:00
#> 4: B 5 2020-01-01 09:45:00 2020-01-01 14:45:00 2020-01-01 10:00:00
#> 5: C 9 2020-01-01 10:05:00 2020-01-01 10:35:00 2020-01-01 10:00:00
#> 6: A 2 2020-01-01 10:55:00 2020-01-01 11:15:24 2020-01-01 10:00:00
#> interval_end base_date base_hour
#> 1: 2020-01-01 10:00:00 2020-01-01 9
#> 2: 2020-01-01 10:00:00 2020-01-01 9
#> 3: 2020-01-01 11:00:00 2020-01-01 10
#> 4: 2020-01-01 11:00:00 2020-01-01 10
#> 5: 2020-01-01 11:00:00 2020-01-01 10
#> 6: 2020-01-01 11:00:00 2020-01-01 10
To obtain summary data for every hour, for all combined patient stays:
library(patientcounter)
patient_count_hour <- interval_census(beds,
identifier = 'patient',
admit = 'start_time',
discharge = 'end_time',
group_var = 'bed',
time_unit = '1 hour',
results = "total",
uniques = TRUE)
head(patient_count_hour)
#> interval_beginning interval_end base_date base_hour N
#> 1: 2020-01-01 09:00:00 2020-01-01 10:00:00 2020-01-01 9 2
#> 2: 2020-01-01 10:00:00 2020-01-01 11:00:00 2020-01-01 10 5
#> 3: 2020-01-01 11:00:00 2020-01-01 12:00:00 2020-01-01 11 4
#> 4: 2020-01-01 12:00:00 2020-01-01 13:00:00 2020-01-01 12 3
#> 5: 2020-01-01 13:00:00 2020-01-01 14:00:00 2020-01-01 13 3
#> 6: 2020-01-01 14:00:00 2020-01-01 15:00:00 2020-01-01 14 3
Note that you also receive the base date and base hour for each interval to enable easier filtering of the results.
This example shows grouping results by bed and hour.
The first ten rows of the resulting grouped values are shown
library(patientcounter)
grouped <- interval_census(beds,
identifier = 'patient',
admit = 'start_time',
discharge = 'end_time',
group_var = 'bed',
time_unit = '1 hour',
results = "group",
uniques = FALSE)
head(grouped[bed %chin% c('A', 'B')],10)
#> bed interval_beginning interval_end base_date base_hour N
#> 1: A 2020-01-01 09:00:00 2020-01-01 10:00:00 2020-01-01 9 1
#> 2: B 2020-01-01 09:00:00 2020-01-01 10:00:00 2020-01-01 9 1
#> 3: A 2020-01-01 10:00:00 2020-01-01 11:00:00 2020-01-01 10 2
#> 4: B 2020-01-01 10:00:00 2020-01-01 11:00:00 2020-01-01 10 1
#> 5: B 2020-01-01 11:00:00 2020-01-01 12:00:00 2020-01-01 11 1
#> 6: A 2020-01-01 11:00:00 2020-01-01 12:00:00 2020-01-01 11 2
#> 7: B 2020-01-01 12:00:00 2020-01-01 13:00:00 2020-01-01 12 1
#> 8: A 2020-01-01 12:00:00 2020-01-01 13:00:00 2020-01-01 12 1
#> 9: B 2020-01-01 13:00:00 2020-01-01 14:00:00 2020-01-01 13 1
#> 10: A 2020-01-01 13:00:00 2020-01-01 14:00:00 2020-01-01 13 1
- You must ‘quote’ your variables, for the time being at least..
- Set results to ‘patient’ for 1 row per patient per interval for each interval in the patient stay.
- Set results to ‘group’ to get a count per group per interval.
Remember this will also be influenced by the ‘uniques’ argument - set it to FALSE to ensure each move in each location is counted. - Set results to ‘total’ for a summary of the data set - interval, base_hour and count.
- To count individual patients ONLY, leave ‘uniques’ at the default value of ‘TRUE’.
- To count patient moves between locations during intervals, set uniques to ‘FALSE’. Patients who occupy beds in different locations during each interval are accounted for in each location. They will be counted at least twice during an interval - both in their initial location and their new location following a move.
- Everything is easier if you use “UTC” by default. You can attempt to coerce the final results yourself using lubridate::force_tz()
To find your system timezone:
Sys.timezone()
See ? seq.POSIXt for valid values
E.G. ‘1 hour’, ‘15 mins’, ‘30 mins’
Want to count those in between 10:01 to 11:00? You can do that using ‘time_adjust_period’ - set it to ‘start_min’ and then set ‘time_adjust_interval’ to 1.
10:00 to 10:59?
Yes, that’s possible as well - set ‘time_adjust_period’ to ‘end_min’
and set ‘time_adjust_interval’ as before. You can set these periods to
any value, as long as it makes sense in relation to your chosen
time_unit.
Here we adjust the start_time by 5 minutes
library(patientcounter)
patient_count_time_adjust <- interval_census(beds,
identifier = 'patient',
admit = 'start_time',
discharge = 'end_time',
group_var = 'bed',
time_unit = '1 hour',
time_adjust_period = 'start_min',
time_adjust_value = 5,
results = "total",
uniques = TRUE)
head(patient_count_time_adjust)
#> interval_beginning interval_end base_date base_hour N
#> 1: 2020-01-01 09:05:00 2020-01-01 10:00:00 2020-01-01 9 2
#> 2: 2020-01-01 10:05:00 2020-01-01 11:00:00 2020-01-01 10 5
#> 3: 2020-01-01 11:05:00 2020-01-01 12:00:00 2020-01-01 11 4
#> 4: 2020-01-01 12:05:00 2020-01-01 13:00:00 2020-01-01 12 3
#> 5: 2020-01-01 13:05:00 2020-01-01 14:00:00 2020-01-01 13 3
#> 6: 2020-01-01 14:05:00 2020-01-01 15:00:00 2020-01-01 14 3
Valid values for time_adjust_period are ‘start_min’, ‘start_sec’, ‘end_min’ and ‘end_sec’
See my package juncture which does exactly the same thing, but is easier to type (the function is simply called 'juncture') and uses tinytest and checkmate for testing, so has even fewer dependencies