Overstay2 scoring model for HSC
This page describes the Overstay2 scoring model for HSC for the Project Overstay2, one of the site-based Overstay2 scoring models. The model and the Data definition for factor candidates for the Overstay2 project it uses were developed in Re-analysis and generation of Overstay2 model and is used to generate the Overstay2 colour, which in turn drives the Overstay2 processes on the units to reduce discharge delay.
Data elements and assigned coefficients
The binary logistic regression Y (Overstay GE 10 Days) which gave the AUC C-index =0.789 implying better predicted model and Hosmer and Lemeshow Goodness of fit test Chi-Square value of 10.5443 (prob=0.2289) is :
LN(OS/1-OS) = Y
Y = Intercept Estimate + sum of all the rest of (the Parameter values multiplied by the Estimate) listed below:
| Parameter | Estimate | Prob > ChiSq |
|---|---|---|
| Intercept | -6.628750489 | <.0001 |
| Age | 0.020816662 | <.0001 |
| Homeless | 0.73542220 | 0.0048 |
| MB_Winnipeg | 0.462297834 | 0.0009 |
| Dementia | 0.444548081 | 0.0472 |
| Trach | 1.439221519 | 0.0003 |
| ADL_bath | 0.407661477 | <.0001 |
| ADL_feed | 0.066130137 | 0.0521 |
| ADL_continence | 0.09033481 | 0.0074 |
| ADL_Adlmean_nh | -0.095015407 | 0.0024 |
| ADL_Adlmean_age | -0.000470209 | 0.0190 |
Where OS is the likelihood to overstay 10 days and beyond.
How to calculate OS:
- OS = exp(Y) / (1 + exp(Y)
- The model gave a 0.043 optimum cut-off probability for OS GE 10 Days with sensitivity = 71.0%, the rate of identifying OS GE 10d out of the patients who actually have OS GE 10d and with specificity = 71.7%, the rate of identifying OS LT 10d out of the patients who actually have OS LT 10d. This cut-off point of 0.043 will result to 28.3% rate of identifying OS GE 10d out of the patients who actually have OS LT 10d. Increasing the cut off predicted probability of (Overstay >= 10 days) to 0.054 resulted to sensitivity = 66.6%, thus reducing the rate of identifying OS GE 10d out of the patients who actually have OS LT 10d to 24.2%. Thus a predicted probability of (Overstay >= 10 days) of 0.054 is going to be used to classify patients into two groups; Overstay >= 10 days and Overstay < 10 days. See more discussion on Overstay2 colour.
- The model has been validated using the data from the validation set of same period. The AUC C-index = 0.0.787, the Hosmer and Lemeshow Goodness of fit test Chi-Square value = 10.4037 (prob=0.2346) and the optimum cut-off probability for OS = 0.043 with sensitivity = 71.6% and specificity = 71.5%.
Model files
- files describing the model and records of discussion about it are at S:\MED\MED_CCMED\Julie\MedProjects\Overstay_Project_2025\3_HSCSTB_15Oct2025\HSC\FinalModel
Log
- 2025-10-15 - model 6 selected as final model with AUC=0.789 and HL goodness of fit =10.5443(prob=0.2289)
- 2025-08-11 - found good fit model in the training data set (Nov2020 to Dec2024) as well as in the validation data set1 (Nov2020 to Dec2024) but not in the validation dataset2 (2025 data). When discussed with Dr Roberts and Tina, it was agreed to redefine the period to include Nov 2020 to July 2025 and divide into training set and one validation set and from these data sets find the good fit model. The results presented in Oct 15, 2025.
- 2025-07-18 - found good fit model but was agreed to re-do to include more parameters like Chronic Addiction, Acute OD Toxicity, geographic locations, etc as well as clarify the definitions of some parameters.
Related articles
| Related articles: |