Article Text

Download PDFPDF
065 Integrating breast cancer risk assessment into EHRs to support shared screening decisions
  1. Karen B Eden1,
  2. Benjamin E Orwoll1,2,
  3. Katherine L Bensching3,
  4. Annette Totten1,
  5. William Hersh1,
  6. Lindsey Watson1,
  7. Ashley Scherman4,
  8. Rongwei Fu1,5,
  9. Heidi D Nelson6
  1. 1Department of Medical Informatics and Clinical Epidemiology, Oregon Health and Science University, Portland, Oregon, USA
  2. 2Department of Pediatrics, Oregon Health and Science University, Portland, Oregon, USA
  3. 3Department of Medicine, Oregon Health and Science University, Portland, Oregon, USA
  4. 4Information Technology Group, Oregon Health and Science University, Portland, Oregon, USA
  5. 5School of Public Health, Oregon Health and Science University, Portland, Oregon, USA
  6. 6Department of Health Systems Science, Kaiser Permanente School of Medicine, Pasadena, California, USA


Introduction Shared decision-making for breast cancer prevention requires an accurate understanding of the patient’s cancer risk. Risk information in the electronic health record (EHR) can be difficult to locate or outdated. Periodic collection of patient-reported data can improve cancer risk assessment. Leveraging new, health data interoperability methods, we developed and evaluated an EHR-integration for a breast cancer risk-assessment tool and patient decision aid, MammoScreen (MS).

Methods We invited patients eligible for screening mammography aged 40–74 years from a general internal medicine clinic via their patient portal. MS pulled relevant risk data from the EHR via standard application programming interfaces, presented it to patients for verification/correction, and provided summary notes (average or above-average risk, symptoms, health history) to the EHR. Rates of MS uptake, breast cancer risk levels, and new or corrected information were determined. Patient and clinician experiences were summarized from semi-structured interviews.

Results 801 patients were invited and 676 (84%) opened either the initial invitation or a reminder message. Of these, 43% (294/676) completed MS; 13% of completers (39/294) were categorized as above-average risk and 79% (233/294) as average risk and would benefit from a shared decision-making discussion of preventive strategies. Another 8% (22/294) had unknown risk due to unavailable history information. Six percent (17/294) reported breast symptoms requiring referral. Clinicians and patients reported MS helped identify previously unrecognized risks and was easy to use.

Discussion Integration of a breast cancer screening decision aid, including evidence-based risk algorithms, with the EHR was well-received by patients and clinicians with high patient uptake.

Conclusion This integrated approach provides more complete capture of cancer risk information, ensuring each patient’s risk is accurately addressed in shared decision-marking discussions.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.