Detailed overview
This page describes the parameters for the cases that are of primary interest to developers who want to start using fastABF.
Note
Readers who wish to contribute and extend this library should read this section as well as the section for library developers & contributors.
Typical workflow
Normally users would like to calculate the ABF prices for episodes. In fastABF this is easily done by gathering various pieces of data required to invoke the corresponding pipeline for each of the ABF cases. In order to know which pieces of data are required and to construct them easily - here is a simple overview of the inputs for each of the four cases
Admitted acute ABF
Parameters required
class Admitted_Acute_Record:
Birth_Date: str, # ddmmyy
Admission_Date: str, # ddmmyy
Separation_Date: str, # ddmmyy
AR_DRG_v10: str,
# [METeOR id: 269976] True if admitted patient transferred from another hospital.
bool_transfer_status: bool,
sex: int, # [METeOR id: 635126]
# [METeOR id: 269986] True if admission occurred on an emergency basis
bool_is_emergency_admission: bool,
bool_foetal_distress_flag: bool,
bool_instrument_use_flag: bool,
bool_ppop_flag: bool,
bool_prima_flag: bool,
HAC1: bool,
HAC2: bool,
HAC3: bool,
HAC4: bool,
HAC6: bool,
HAC7: bool,
HAC8: bool,
HAC9: bool,
HAC10: bool,
HAC11: bool,
HAC12: bool,
HAC13: bool,
HAC14: bool,
HAC15p2: bool,
Charlson_Score: int,
patient_leave_days: float = 0,
ICU_hours_L3: float = -1,
ICU_hours_other: float = 0,
Psych_Days: float = 0, # METeOR id: 552375
Pat_Postcode: str = "", # Format PCNNNN, METeOR id: 429894
Pat_SA2: int = 0, # NNNNNNNNN (9 digit), METeOR id: 469909
# Establishment's Remoteness Category
EST_Remoteness_Cat: Remoteness_Category_RA16 = Remoteness_Category_RA16.Unknown,
Indigenous_Status: Indigenous_Status_Category = \
Indigenous_Status_Category.Unknown_or_not_stated,
Pat_Radiotherapy_Flag: bool = False,
# Set to True if the patient has any of the ACHI 11th Edition codes
# listed in Appendices B and C of the `national efficient
# price determination 2020-21` document for radiotherapy relevant codes
# else False
Pat_Dialysis_Flag: bool = False,
# Set to True if the patient has any of the ACHI 11th Edition codes
# listed in Appendices B and C of the `national efficient
# price determination 2020-21` document for dialysis relevant codes
# else False
# Set to True if the patient is an eligible private patient
Pat_private_Flag: bool = False,
Pat_Covid19_Flag: bool = False,
care_type: Care_Type = Care_Type.acute_care_admitted_care
Admitted subacute ABF
Parameters required
class Admitted_Subacute_Record:
Birth_Date: str, # mmddyy
Admission_Date: str,
Separation_Date: str,
AN_SNAP_v4: str,
# [METeOR id: 269986] True if admission occurred on an emergency basis
patient_leave_days: float = 0,
Pat_Postcode: str = "", # Format PCNNNN, METeOR id: 429894
Pat_SA2: int = 0, # NNNNNNNNN (9 digit), METeOR id: 469909
# Establishment's Remoteness Category
EST_Remoteness_Cat: Remoteness_Category_RA16 = Remoteness_Category_RA16.Unknown,
Indigenous_Status: Indigenous_Status_Category = \
Indigenous_Status_Category.Unknown_or_not_stated,
Pat_Radiotherapy_Flag: bool = False,
# Set to True if the patient has any of the ACHI 11th Edition codes
# listed in Appendices B and C of the `national efficient
# price determination 2020-21` document for radiotherapy relevant codes
# else False
Pat_Dialysis_Flag: bool = False,
# Set to True if the patient has any of the ACHI 11th Edition codes
# listed in Appendices B and C of the `national efficient
# price determination 2020-21` document for dialysis relevant codes
# else False
# Set to True if the patient is an eligible private patient
Pat_private_Flag: bool = False,
Pat_Covid19_Flag: bool = False,
care_type: Care_Type = Care_Type.acute_care_admitted_care
Emergency ABF
Parameters required
class Emergency_Record:
Birth_Date: str,
Admission_Date: str,
URG_1p4_or_UDG_v1p3: int,
Emergency_Care_Level: str,
Pat_Postcode: str = "", # Format PCNNNN, METeOR id: 429894
Pat_SA2: int = 0, # NNNNNNNNN (9 digit), METeOR id: 469909
# Establishment's Remoteness Category
EST_Remoteness_Cat: Remoteness_Category_RA16 = Remoteness_Category_RA16.Unknown,
Indigenous_Status: Indigenous_Status_Category = \
Indigenous_Status_Category.Unknown_or_not_stated,
Pat_Covid19_Flag: bool = False,
care_type: Care_Type = Care_Type.acute_care_admitted_care
Non admitted ABF
Parameters required
class Nonadmitted_Record:
Birth_Date: str,
Event_Service_Date: str,
Tier2_CV5: str,
patient_leave_days: float = 0,
Pat_Postcode: str = "", # Format PCNNNN, METeOR id: 429894
Pat_SA2: int = 0, # NNNNNNNNN (9 digit), METeOR id: 469909
# Establishment's Remoteness Category
EST_Remoteness_Cat: Remoteness_Category_RA16 = Remoteness_Category_RA16.Unknown,
Indigenous_Status: Indigenous_Status_Category = \
Indigenous_Status_Category.Unknown_or_not_stated,
Multiple_Healthcare_Provider_Indicator: bool = False,
Pat_Covid19_Flag: bool = False,
care_type: Care_Type = Care_Type.acute_care_admitted_care
fastabf
└── pipelines (the most important module- contains the main pipelines)
├── nwau_admitted_acute.py
├── nwau_admitted_sub_and_non_acute.py
├── nwau_emergency.py
└── nwau_nonadmitted.py
Convenience functions and interfaces
As you may observe from the input parameters in the cases above, they use specific classes/data types that were created for convenience. These convenience interfaces help avoid having to remember the METeOR parameter values thereby reducing errors and speeding up code development.
These user friendly classes and routines can be found at the following locations in the library.
Structure of the fastABF library
A simplified overview for users of this library is as follows
.
├── datatypes.py (user friendly interfaces to data types used - more info below)
├── Helpers
└── charlson_class.py (computes the charlson complexity, if it is not known)
└── helper_remoteness_mappings.py (computes remoteness categories)
Data types
The datatypes module contains user friendly mappings to several of the METeOR classes. Instead of having to remember formally designated numbers, you can use these convenient interfaces in your own code. For the details of the possible values for each category, please refer to the specific section on data types.
class ABF_Service_Category(Enum)
class Care_Type(Enum)
class Hosp_State_Category(Enum)
class Indigenous_Status_Category(Enum)
class MDC_Type(Enum)
class Remoteness_Category_RA16(Enum)
class Sex_Category(Enum)
class Stay_Category(Enum)
Helper functions
The Charlson class helps compute the Charlson complexity in case it is not known apriori. Similarly the remoteness mappings help automatically use the right order of priorities to assign a remoteness category to an episode.