@BinglingICL has correctly noted that the ApptType IPAdmission should also be applied on the event of an in-patient being discharged. (The appointment type - despite its name - is relevant to the work entitled in admission and also discharge.) This is not done currently and there is no special accounting for the work of discharging a patient. The is an important consideration for when we are calibrating the model to counts of appointments.
Adding these footprints, should be automated through the BedDays class. A few ideas about how to do so, from simplest to hardest are given below.
Simplest; let the admission event be associated with 2 * IPAdmission (one for admission and one for discharge). Note that even if the patient dies, they are still discharged from care. In this case, the appointment related to this charge will, erroneously, be seen to occur on the day of admission - but this is probably not a big problem!
We could schedule an "empty" HSI for that person (e.g. 'HSI_Discharge'), for which the APPT_FOOTPRINT is just 1 * IPAdmission. The HSI would occur on the expected day of discharge.
The BedDay function that is called every day could find the numbers of persons being discharged each day from each facility and create an overall footprint, similar to how the computation is done for in-patients.
@BinglingICL has correctly noted that the ApptType
IPAdmission
should also be applied on the event of an in-patient being discharged. (The appointment type - despite its name - is relevant to the work entitled in admission and also discharge.) This is not done currently and there is no special accounting for the work of discharging a patient. The is an important consideration for when we are calibrating the model to counts of appointments.Adding these footprints, should be automated through the
BedDays
class. A few ideas about how to do so, from simplest to hardest are given below.Simplest; let the admission event be associated with 2 *
IPAdmission
(one for admission and one for discharge). Note that even if the patient dies, they are still discharged from care. In this case, the appointment related to this charge will, erroneously, be seen to occur on the day of admission - but this is probably not a big problem!We could schedule an "empty" HSI for that person (e.g. 'HSI_Discharge'), for which the
APPT_FOOTPRINT
is just 1 *IPAdmission
. The HSI would occur on the expected day of discharge.The
BedDay
function that is called every day could find the numbers of persons being discharged each day from each facility and create an overall footprint, similar to how the computation is done for in-patients.