Annals of Family Medicine encourages readers to develop a learning community to improve health and health care through enhanced primary care. With the Annals Journal Club, we encourage diverse participants—particularly students, trainees, residents, and interns—to think critically about and discuss important issues affecting primary care, and even consider how their discussions might inform their practice.

Annals provides discussion tips and questions related to one original research article in each issue. We welcome you to post a summary of your conversation to our eLetters section, a forum for readers to share their responses to Annals articles. Further information and links to previous Annals Journal Club features can be found on our website.

Tuan W-J, Yan Y, Abou Al Adat B, et al. Predicting missed appointments in primary care: a personalized machine learning approach. Ann Fam Med. 2025; 23(4): 294-301. doi: https://doi.org/10.1370/afm.240316.

Primary care visit no-shows and late cancellations are a frustrating problem in the management of primary care practices. Past studies have largely used electronic health records to predict these outcomes with variable success. This study, from a single primary care network in Pennsylvania, incorporates clinical records with location and weather data to investigate the robustness of predicting no-shows and late cancellations using various modeling techniques.

What question is asked by this study and why does it matter?

What methods were used in the study? What measures were used to determine how well the predictive models worked?

Define: gradient boost, random forest, neural network, LASSO logistic regression, area under the receiver operating curve (AUROC), F1 score, and SHapley Additive exPlanations (SHAP).

How does this study advance beyond previous research and clinical practice on this topic?

How strong is the study design for answering the question?

To what degree can the findings be accounted for by:

Study sample

Variables included in the model and how they were categorized

Outcome variables (no-show, late cancellation, etc)

Time frame from when the study data were collected

Interaction effects—different effects of individual characteristics by predictors (ie, weather)

What are the main study findings?

How comparable is the study sample to similar patients in your practice or region?

What is your judgment about the transportability of the findings?

Would you use the included model in your population without further study?

What contextual factors are important for interpreting the findings?

How might this study change your practice? Policy? Education? Research?

Who are the constituencies for the findings, and how might they be engaged in interpreting or using the findings?

What are the next steps in interpreting or applying the findings?

What are the trade-offs between more complex models and simpler less predictive models?

How important is the accuracy of the predictive model compared to the design of an intervention?

What types of interventions might someone trial or use in practice to decrease no-shows and/or late cancellations?

What researchable questions remain?