Health Catalyst, Inc · NASDAQ:HCAT
Health Catalyst — Conference Presentation (September 2024)
18 slides from Health Catalyst's conference presentation, September 2024
Slide 1: AI in Healthcare: Hype, Hope, and Missed Opportunities
AI in Healthcare: Hype, Hope, and Missed Opportunities Jason Jones, Jessica Curran © Health Catalyst. Confidential and Proprietary.
Slide 2: Agenda
• Sample explorations with generative AI • Purposeful AI for everyday decisions Agenda • Improving human-computer interactions • Next step? Purposeful AI for large levers and high stakes © Health Catalyst. Confidential and Proprietary.
Slide 3: Let’s Ask ChatGPT…
Let’s Ask ChatGPT… DischargeMonth LoS 6/30/2020 3.041666 7/31/2020 2.833333 8/31/2020 2.875 9/30/2020 2.833333 … 4/30/2023 2.833333 5/31/2023 2.791666 6/30/2023 3 Here is my length of stay data by month. What is my baseline performance? Has anything changed? © Health Catalyst. Confidential and Propr
Slide 4: The Answer…
The Answer… I apologize for the inconvenience. Since I'm an AI language model, I can't directly execute code with graphics or visualizations. However, I can show you the numerical results from the code that calculates the baseline performance. Here are the results: Baseline Average Length of Stay: 2
Slide 5: Can you give me R code…
Can you give me R code… # Length of Stay data los_data <- c(3.041666, 2.833333, 2.875, 2.833333, 2.791666, 2.791666, 2.791666, 2.791666, 2.666666, 2.75, 2.75, 2.75, 2.75, 2.875, 2.875, 2.791666, 2.75, 2.75, 2.791666, 2.916666, 2.75, 2.75, 2.833333, 2.916666, 3, 3.333333, 2.916666, 2.833333, 2.916666
Slide 6: © Health Catalyst. Confidential and Proprietary.
© Health Catalyst. Confidential and Proprietary.
Slide 7: © Health Catalyst. Confidential and Proprietary.
© Health Catalyst. Confidential and Proprietary.
Slide 8: © Health Catalyst. Confidential and Proprietary.
© Health Catalyst. Confidential and Proprietary.
Slide 9: Poll Question #3
Poll Question #3 How would you compare the current and possible new version of the charts? © Health Catalyst. Confidential and Proprietary.
Slide 10: © Health Catalyst. Confidential and Proprietary.
© Health Catalyst. Confidential and Proprietary.
Slide 11: © Health Catalyst. Confidential and Proprietary.
© Health Catalyst. Confidential and Proprietary.
Slide 12: Next step? Causal Analysis
Next step? Causal Analyisis © Health Catalyst. Confidential and Proprietary.
Slide 13: Large Levers, High Stakes
Large Levers, High Stakes • Program evaluation: Should we continue/expand/alter/stop X (with a vendor)? • Drug/device selection: Is X benefiting patients/people relative to the cost? • Process change: How well is our innovative approach working? © Health Catalyst. Confidential and Proprietary.
Slide 14: Should we expand our work training program?
Should we expand our work training program? Typical “comparison group” is not a good comparison (apples and oranges) Love Plot shows important “imbalances” (potential confounding): • Both treatment group (propensity score) and outcome (prognostic score) can be predicted by other characteristics • Im
Slide 15: Poll Question #4
Poll Question #4 How interested are you in a follow-up session focused on causal analysis? © Health Catalyst. Confidential and Proprietary.
Slide 16: Questions?
Questions? AI in Healthcare: Hype, Hope, and Missed Opportunities Jason Jones, Jessica Curran © Health Catalyst. Confidential and Proprietary.
Slide 17: Would you like to win a complimentary pass to HAS 24?
Would you like to win a complimentary pass to HAS 24? © Health Catalyst. Confidential and Proprietary.
Slide 18: Questions?
Questions? © Health Catalyst. Confidential and Proprietary.
Health Catalyst — Conference Presentation (September 2024) on 6ix