Devon Referral to Treatment (RTT) Model in R

Author: Richard Blackwell

Email: [email protected]

Date: 2023-08-04

---

Variables

General:

Variables ending in nonadm refer to the non-admitted pathway and those ending in adm refer to the admitted pathway. A variable displayed as variable_name_... in this README signifies that both the variable_name_nonadm and variable_name_adm variables are being referred to.

---

Simulation variables

• sim_name - Name of the simulation for use in output directory and filenames.
• sim_trials - Number of simulation runs.
• sim_periods - Number of time periods each simulation will run for.
• sim_bins - Number of bins to use for the waiting list i.e. sim_bins = 52 will monitor waits from >=0 weeks and <1 week through to >=52 weeks. NB: Bin zero will always be added so we do not need to add that into the number required
• sim_time_unit - Unit of time for the model i.e. year, month, week, day.

NB: the same unit of time will be used across the entire model, i.e. if we use week then the demand is expressed in weekly demand, the capacity in weekly capacity, the waiting list bins as weekly bins.

---

Profile Variables

Clock Stop Profile variables

• csprof_type_nonadm & csprof_type_adm - Type of clock stop profile input, either actual (read direct from input file) or synthetic (created from beta-binomial distribution using parameters supplied in input file)
• csprof_param_nonadm & csprof_param_adm - If the csprof_type_... is synthetic this is an array of [sim_period, 2] in which to store shape parameter 1 (also know as $\alpha$ ) and shape parameter 2 (also know as $\beta$ ) used to define the shape of the beta-binomial distribution of the clock stop profile for each simulation period.
• csprof_nonadm & csprof_adm - Array of dimensions [sim_periods, sim_bins+1] in which to store the clock stop profile in bins for each simulation period.

Demand Profile variables

• demprof_type_nonadm & demprof_type_adm - Type of demand profile input, either actual (read direct from input file) or synthetic (created from beta-binomial distribution using parameters supplied in input file)
• demprof_param_nonadm & demprof_param_adm - If the demprof_type_... is synthetic this is an array of [sim_period, 2] in which to store shape parameter 1 (also know as $\alpha$ ) and shape parameter 2 (also know as $\beta$ ) used to define the shape of the beta-binomial distribution of the demand profile for each simulation period.
• demprof_nonadm & demprof_adm - Array of dimensions [sim_periods, sim_bins+1] in which to store the demand profile in bins for each simulation period.

Removal Other Than Treatment (ROTT) Profile variables

• rottprof_type_nonadm & rottprof_type_adm - Type of ROTT profile input, either actual (read direct from input file) or synthetic (created from beta-binomial distribution using parameters supplied in input file)
• rottprof_param_nonadm & demprof_param_adm - If the rottprof_type_... is synthetic this is an array of [sim_period, 2] in which to store shape parameter 1 (also know as $\alpha$ ) and shape parameter 2 (also know as $\beta$ ) used to define the shape of the beta-binomial distribution of the ROTT profile for each simulation period.
• rottprof_nonadm & rottprof_adm - Array of dimensions [sim_periods, sim_bins+1] in which to store the ROTT profile in bins for each simulation period.

Conversion Profile variables

• convprof_type_nonadm & convprof_type_adm - Type of conversion profile input, either actual (read direct from input file) or synthetic (created from beta-binomial distribution using parameters supplied in input file)
• convprof_param_nonadm & convprof_param_adm - If the rottprof_type_... is synthetic this is an array of [sim_period, 2] in which to store shape parameter 1 (also know as $\alpha$ ) and shape parameter 2 (also know as $\beta$ ) used to define the shape of the beta-binomial distribution of the conversion profile for each simulation period.
• convprof_nonadm & convprof_adm - Array of dimensions [sim_periods, sim_bins+1] in which to store the conversion profile in bins for each simulation period.

---

Process Variables

Waiting List variables

• wl_type_nonadm & wl_type_adm - Type of waiting list input, either actual (read direct from input file) or synthetic (created from beta-binomial distribution using parameters supplied in input file).
• wl_size_nonadm & wl_size_adm - If the wl_type_... is synthetic this is size of the waiting list to be created using the beta-binomial distribution.
• wl_shape1_nonadm & wl_shape1_adm - If the wl_type_... is synthetic this is shape parameter 1 (also know as $\alpha$ ) used to define the shape of the beta-binomial distribution.
• wl_shape2_nonadm & wl_shape2_adm - If the wl_type_... is synthetic this is shape parameter 2 (also know as $\beta$ ) used to define the shape of the beta-binomial distribution.
• wl_nonadm & wl_adm - Array of dimension [sim_periods+1, sim_bins+1, sim_trials] in which to store the waiting list in bins for each simulation period and run. NB: Dimensions are sim_periods+1 as the starting waiting list (period 0) will also be recorded, and sim_bins+1 as the waiting list bins run from zero through to sim_bins.

Clock Stop variables

• cs_nonadm & cs_adm - Array of dimensions [sim_periods, sim_bins+1, sim_trails] in which to store the clock stops by bin for each simulation trial and simulation period.

Demand variables

• dem_param_nonadm & dem_param_adm - Array of dimensions [sim_periods, 2] in which to store the mean and standard deviations parameters for each simulation period used to create the dem_vol_... from a normal distribution.
• dem_vol_nonadm & dem_vol_adm - Array of dimensions [sim_periods, sim_trials] used to store the demand volume for the non-admitted and admitted pathways for each period and simulation run.
• dem_nonadm & dem_adm - Array of dimensions [sim_periods, sim_bins+1, sim_trials] in which to store the demand delivery in bins for each simulation period and run.

Capacity variables

• cap_param_nonadm & cap_param_adm - Array of dimensions [sim_periods, 2] in which to store the mean and standard deviations parameters for each simulation period used to create the cap_vol_... from a normal distribution.
• cap_vol_nonadm & cap_vol_adm - Array of dimensions [sim_periods, sim_trials] used to store the capacity volume for the non-admitted and admitted pathways for each period and simulation run.

Non-RTT variables

• nonrtt_param_nonadm & nonrtt_param_adm - Array of dimension [sim_periods] in which to store the probability parameter for each simulation period used to create the nonrtt_vol_... from a binominal distribution
• nonrtt_vol_nonadm & nonrtt_vol_adm - Array of dimensions [sim_periods, sim_trials] used to store the non-RTT volume for the non-admitted and admitted pathways for each period and simulation run.

Removal Other Than Treatment (ROTT) variables

• rott_param_nonadm & rott_param_adm - Array of dimension [sim_periods] in which to store the probability parameter for each simulation period used to create the rott_vol_... from a binominal distribution
• rott_vol_nonadm & rott_vol_adm - Array of dimensions [sim_periods, sim_trials] used to store the ROTT volume for the non-admitted and admitted pathways for each period and simulation run.
• rott_nonadm & rott_adm - Array of dimensions [sim_periods, sim_bins+1, sim_trials] in which to store the ROTT activity in bins for each simulation period and run.

Conversion variables

• conv_param_nonadm & conv_param_adm - Array of dimension [sim_periods] in which to store the probability parameter for each simulation period used to create the conv_vol_... from a binominal distribution
• conv_vol_nonadm & conv_vol_adm - Array of dimensions [sim_periods, sim_trials] used to store the conversion volume for the non-admitted (to admitted) and admitted (to non-admitted) pathways for each period and simulation run.
• conv_nonadm & conv_adm - Array of dimensions [sim_periods, sim_bins+1, sim_trials] in which to store the conversion activity in bins for each simulation period and run.

---

Simplified Flowchart

flowchart TB
    sim_init[Initialise Simulation:\nCreate empty arrays\nCreate all time independent data ] --> trial_init
    trial_init[Initilise Trial:\nSet initial WL] --> period_increment
    period_increment[Period increment:\nInitialise WL with previous period] --> process_period
    process_period[Process period:\nRemove ROTT\nRemove CS\nRight shift WL\nAdd DEM\nAdd CONV] --> check{Check Trial End}
    check -- Y --> trial_init
    check -- N --> period_increment

Loading