Individualized No-Show Predictions: Effect on Clinic Overbooking and Appointment Reminders

Yutian Li, Sammi Tang, Joseph Johnson, David A. Lubarsky

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Patient no-shows and late cancellations lead to clinic inefficiency, high clinic costs and low patient satisfaction. The two main strategies clinics employed to alleviate the adverse effects of no-shows are overbooking and patient appointment reminders. Developing effective overbooking schedules depends on accurately predicting each patient’s no-show probability, while developing effective reminder systems requires a patient-level estimate of communication sensitivity. Current methods of estimating no-show probabilities do not produce such patient-level predictions. To remedy this, we develop a Bayesian nested logit model which utilizes appointment confirmation data and estimates individual-level coefficients for patient-specific predictors. A log-likelihood comparison of model fit on 12 months of appointment data shows that the Bayesian model outperforms the standard logit model by about 30% improvement in model fit. Additionally, our Bayesian model allows categorization of patients based on their appointment confirmation behavior. Finally, using patient-specific no-show probabilities as an input to a simulated appointment scheduler we find that the Bayesian model improves clinic profit over the standard logit model. The benefit comes mainly from waiting cost reduction when no-show probability is low and from physician overtime and idle time cost reduction when no-show probability is high. Our study has two managerial implications. First, the Bayesian method allows customizing appointment reminder effort based on patient’s confirmation behavior. Second, the Bayesian method also allows improved overbooking scheduling especially in clinics that experience large patient throughput.

Original languageEnglish (US)
JournalProduction and Operations Management
DOIs
StatePublished - Jan 1 2019

Fingerprint

Cost reduction
Prediction
Overbooking
Profitability
Scheduling
Throughput
Communication
Costs
Bayesian model
Bayesian methods
Logit model
Predictors
Schedule
Inefficiency
Nested logit model
Cancellation
Patient satisfaction
Profit
Coefficients
Time costs

Keywords

  • appointment reminder
  • Bayesian method
  • nested logit model
  • no-shows
  • overbooking

ASJC Scopus subject areas

  • Management Science and Operations Research
  • Industrial and Manufacturing Engineering
  • Management of Technology and Innovation

Cite this

@article{fae06aeeedba425c931636bb7b51b45e,
title = "Individualized No-Show Predictions: Effect on Clinic Overbooking and Appointment Reminders",
abstract = "Patient no-shows and late cancellations lead to clinic inefficiency, high clinic costs and low patient satisfaction. The two main strategies clinics employed to alleviate the adverse effects of no-shows are overbooking and patient appointment reminders. Developing effective overbooking schedules depends on accurately predicting each patient’s no-show probability, while developing effective reminder systems requires a patient-level estimate of communication sensitivity. Current methods of estimating no-show probabilities do not produce such patient-level predictions. To remedy this, we develop a Bayesian nested logit model which utilizes appointment confirmation data and estimates individual-level coefficients for patient-specific predictors. A log-likelihood comparison of model fit on 12 months of appointment data shows that the Bayesian model outperforms the standard logit model by about 30{\%} improvement in model fit. Additionally, our Bayesian model allows categorization of patients based on their appointment confirmation behavior. Finally, using patient-specific no-show probabilities as an input to a simulated appointment scheduler we find that the Bayesian model improves clinic profit over the standard logit model. The benefit comes mainly from waiting cost reduction when no-show probability is low and from physician overtime and idle time cost reduction when no-show probability is high. Our study has two managerial implications. First, the Bayesian method allows customizing appointment reminder effort based on patient’s confirmation behavior. Second, the Bayesian method also allows improved overbooking scheduling especially in clinics that experience large patient throughput.",
keywords = "appointment reminder, Bayesian method, nested logit model, no-shows, overbooking",
author = "Yutian Li and Sammi Tang and Joseph Johnson and Lubarsky, {David A.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1111/poms.13033",
language = "English (US)",
journal = "Production and Operations Management",
issn = "1059-1478",
publisher = "Wiley-Blackwell",

}

TY - JOUR

T1 - Individualized No-Show Predictions

T2 - Effect on Clinic Overbooking and Appointment Reminders

AU - Li, Yutian

AU - Tang, Sammi

AU - Johnson, Joseph

AU - Lubarsky, David A.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Patient no-shows and late cancellations lead to clinic inefficiency, high clinic costs and low patient satisfaction. The two main strategies clinics employed to alleviate the adverse effects of no-shows are overbooking and patient appointment reminders. Developing effective overbooking schedules depends on accurately predicting each patient’s no-show probability, while developing effective reminder systems requires a patient-level estimate of communication sensitivity. Current methods of estimating no-show probabilities do not produce such patient-level predictions. To remedy this, we develop a Bayesian nested logit model which utilizes appointment confirmation data and estimates individual-level coefficients for patient-specific predictors. A log-likelihood comparison of model fit on 12 months of appointment data shows that the Bayesian model outperforms the standard logit model by about 30% improvement in model fit. Additionally, our Bayesian model allows categorization of patients based on their appointment confirmation behavior. Finally, using patient-specific no-show probabilities as an input to a simulated appointment scheduler we find that the Bayesian model improves clinic profit over the standard logit model. The benefit comes mainly from waiting cost reduction when no-show probability is low and from physician overtime and idle time cost reduction when no-show probability is high. Our study has two managerial implications. First, the Bayesian method allows customizing appointment reminder effort based on patient’s confirmation behavior. Second, the Bayesian method also allows improved overbooking scheduling especially in clinics that experience large patient throughput.

AB - Patient no-shows and late cancellations lead to clinic inefficiency, high clinic costs and low patient satisfaction. The two main strategies clinics employed to alleviate the adverse effects of no-shows are overbooking and patient appointment reminders. Developing effective overbooking schedules depends on accurately predicting each patient’s no-show probability, while developing effective reminder systems requires a patient-level estimate of communication sensitivity. Current methods of estimating no-show probabilities do not produce such patient-level predictions. To remedy this, we develop a Bayesian nested logit model which utilizes appointment confirmation data and estimates individual-level coefficients for patient-specific predictors. A log-likelihood comparison of model fit on 12 months of appointment data shows that the Bayesian model outperforms the standard logit model by about 30% improvement in model fit. Additionally, our Bayesian model allows categorization of patients based on their appointment confirmation behavior. Finally, using patient-specific no-show probabilities as an input to a simulated appointment scheduler we find that the Bayesian model improves clinic profit over the standard logit model. The benefit comes mainly from waiting cost reduction when no-show probability is low and from physician overtime and idle time cost reduction when no-show probability is high. Our study has two managerial implications. First, the Bayesian method allows customizing appointment reminder effort based on patient’s confirmation behavior. Second, the Bayesian method also allows improved overbooking scheduling especially in clinics that experience large patient throughput.

KW - appointment reminder

KW - Bayesian method

KW - nested logit model

KW - no-shows

KW - overbooking

UR - http://www.scopus.com/inward/record.url?scp=85065476898&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065476898&partnerID=8YFLogxK

U2 - 10.1111/poms.13033

DO - 10.1111/poms.13033

M3 - Article

AN - SCOPUS:85065476898

JO - Production and Operations Management

JF - Production and Operations Management

SN - 1059-1478

ER -