A ModelsCloud wrapper for the ACCEPT package — ACute COPD Exacerbation Prediction Tool.
This package wraps the resplab/accept package for deployment on the ModelsCloud platform. It supports all versions of ACCEPT including ACCEPT 3.0-CPRD, a UK primary-care specific recalibration derived from the Clinical Practice Research Datalink (CPRD).
For direct use, install the main ACCEPT package instead:
remotes::install_github("resplab/accept")- ACCEPT 3.0-CPRD (
country = "GBR-primary") — UK primary care recalibration using CPRD data. Includes automatic imputation of missing optional predictors. - ACCEPT 3.0 (
version = "accept3") — Country-specific recalibration based on the NOVELTY study. Requires a country code. - ACCEPT 2.0 (
version = "accept2") — Updated model improving accuracy for patients without exacerbation history. - ACCEPT 1.0 (
version = "accept1") — Original ACCEPT model.
| Variable | Description |
|---|---|
ID |
Patient identifier |
age |
Age in years |
male |
TRUE/FALSE |
FEV1 |
FEV1 % predicted (10-120) |
LastYrExacCount |
Total exacerbations last year |
LastYrSevExacCount |
Severe exacerbations last year |
mMRC or SGRQ or CAT |
Symptom score (at least one required) |
When using country = "GBR-primary", these predictors are optional — if missing
they are automatically imputed using a UK-specific sequential regression
imputation model derived from CPRD data:
LABA, oxygen, ICS, LAMA, statin, BMI, smoker
For all other versions, all predictors should be provided. For requirements for other versions, see resplab/accept.
library(modelscloud)
connect_to_model("resplab/acceptpexa", access_key = "YOUR_API_KEY")
# Get sample patients
patients <- get_sample_input()
# UK primary care - ACCEPT 3.0-CPRD (all predictors present)
model_run(patients, country = "GBR-primary")
# UK primary care - with missing optional predictors (auto-imputed)
patients_missing <- patients
patients_missing$BMI <- NULL
patients_missing$statin <- NULL
patients_missing$LABA <- NULL
model_run(patients_missing, country = "GBR-primary")
# UK specialty care
model_run(patients, country = "GBR-specialty")
# Other countries (ACCEPT 3.0)
model_run(patients, country = "CAN")
model_run(patients, country = "USA")
# ACCEPT 2.0
model_run(patients, version = "accept2")
# ACCEPT 1.0
model_run(patients, version = "accept1")
# Single patient
model_run(get_default_input(), country = "GBR-primary")| Function | Description |
|---|---|
model_run(model_input, version, country) |
Run ACCEPT predictions |
get_sample_input(n) |
Get sample patient data (optional: n patients) |
get_default_input() |
Get a single default patient |
ARG, AUS, BRA, CAN, COL, DEU, DNK, ESP, FRA, ITA, JPN, KOR, MEX, NLD, NOR, SWE, USA
For UK: use "GBR-primary" (primary care) or "GBR-specialty" (specialty care).
For unsupported countries, provide obs_modsev_risk:
model_run(patients, version = "accept3", country = "XXX", obs_modsev_risk = 0.2)- resplab/accept — main ACCEPT package
- resplab/accept3_UK — manuscript analysis code
- ModelsCloud platform