A Paradoxical Effect?

Discussion Tips

This article provides an opportunity to consider findings from one novel component of a patient-centered medical home intervention. The study provides an example of the increasingly commonly used interrupted time series design for studying natural experiments, and requires consideration of contextual factors in interpreting and transporting the findings.

Discussion Questions

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

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

• How strong is the study design for answering the question? What alternatives exist for learning from this kind of real world experiment?

• To what degree can the findings be accounted for by:

  1. How patients were selected, excluded, or lost to follow-up?

  2. How the main variables were measured?

  3. Confounding (false attribution of causality because 2 variables discovered to be associated actually are associated with a 3rd factor)?

  4. Chance?

  5. How the findings were interpreted?

• How were potential threats to validity dealt with in the study design and analysis approach?

• What are the main study findings?

• How do you interpret the differences between the crude and regression analyses? How do you interpret the attenuation over time of the association between office visits and telephone encounters?

• How comparable is the study sample to similar patients in your practice? What is your judgment about the transportability of the findings?

• What contextual factors are important for interpreting the findings (ie the nesting of this study within an ongoing patient-centered medical home initiative within an integrated health care system)?

• 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 researchable questions remain?