MACMLA 2016: Using Data to Improve Clinical Outcomes — Examples and Lessons Learned from Cleveland OH

  • Started with EHR in 1999 (some patients have up to 17 years  worth of data in the EHR)
  • Broader use with 2012 for enterprise-wide EHR adoption – starting with Meaningful Use 1 and progressed forward
  • Paper health records:Hammer :: Electronic health records:Power nail gun — increased power due to tech, but also large potential for more problematic outcomes
  • Case #1 of EHR’s potential – identifying the significant underdiagnosis of hypertension in children — data was available but just not applied
    • Response: Put in an alert to better identify/highlight existence of data
    • Baking the evidence based guidelines into the EHR (CDSS)
      • 38% decrease in false positives
      • 100% increase in provider recognition of abnormal blood pressure
    • Answer: Alerting helps, but not the total fix
  • Case #2 – Immunizations
    • Over 300 immunization rules for children — how do you know if your patient has completed their schedule?
    • Messaging algorithm to reach out to patients using TeleVox about immunization follow up
    • Messages helped increase results by 1/4, but not a perfect fix for the other 3/4
    • Number needed to message 4 people to get 1 immunized
    • $5k in messaging costs led to $200k clinical revenue
    • Personal Health Records
      • Patients will be the biggest amount of EHR users in the future
      • PHR allows for reminders, health info exchange (ex: immunization)
  • Case #3 – Meeting referral drop-off between obesity clinic to specialist
    • 76% referred but never seen
    • So what is the actual appointment follow through within a month after referral? 48%
    • Solutions: self-scheduling + giving specialists a list of patients so they know who is referred and they can take over outreach
    • Moved to 61% after new interventions (6700/month initial consults = $1mil
  • Case #4 – Depression Screening
    • Advanced CDSS for subjective data
    • Use PHQ-9
  • Case #5 – Health Information Exchange
    • Who is likely to have their data exchanged?
      • Older people, female, African Americans, Medicare/Medicaid
    • +1 mil patient records a day exchanged among those on EPIC system
  • Case #6 – Longitudinal diabetes data
    • Synopsis report of data overtime

Q&A notes

  • Information is getting smarter to not just be one click away to Micromedex or UpToDate as a general resource, but one click away from the specific drug entry in Micromedex or UpToDate
  • People in health care do not value information and its integration well into clinical practice — how do we change this paradigm? Informaticists
  • What are the opportunities for librarians to offer education on informatics/EHR competencies
  • Re: alert fatigue – who should the alert be going to? Is adding more info to the 15 min visit overload? What would be a better time? How actionable are these alerts that are provided in the visit?
  • 100% of privacy/security is not realistic, but we need processes in place like credit reports to help manage the inevitable

Follow up readings


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s