HLTH 8750 Week 1 Trend Detection Framing Post Example

Week 1 Discussion: Initial Post and Replies

A Bad Quarter or a Trend? Look at Who Is Deciding Differently

Northgate Health, a composite three-hospital system, saw its hospital outpatient surgery volume fall for one quarter. The easy readings run in both directions. It might be noise: a busy orthopedic surgeon on leave, a contract dispute that held up one payer's authorizations, a slow summer. Or it might be the start of something: cases leaving for the ambulatory surgery centers that have opened in the region. One quarter cannot tell us which, and an executive who acts on either reading now is guessing.

Ansoff's argument about weak signals explains why the guess still matters. He described how strategic surprises are usually preceded by faint, ambiguous information that grows clearer over time, while the time left to respond shrinks; by the time the signal is strong enough to be certain, the best responses may no longer be available (Ansoff, 1975). Waiting for three more quarters of volume data is therefore not neutral. It trades certainty for time.

So I propose a test that does not wait for volume. Temporary causes reverse on their own; structural causes change who decides where a case goes. A trend shows first in the decisions of the people who direct cases, surgeons and payers, and only later in a volume report. Three observations at Northgate would indicate a structural change: surgeons on its medical staff requesting privileges at a new ambulatory center; a major payer announcing a site-of-service rule that steers eligible procedures away from hospital outpatient departments; and a physician group announcing an investment in a center. None of these appears in a volume report, and each would precede the volume shift by months. Two strands of research support looking at decisions first. Diffusion research finds that new practices spread through social systems in a sequence, starting with a few early adopters whose choices the rest of the group observes (Rogers, 2003); in a medical staff, the first surgeons to take privileges at a new center are those early adopters. And physician ownership of surgery centers has been linked to higher surgical volume at those centers (Hollingsworth et al., 2010), which means an investment announcement predicts where cases will go.

My reading for Northgate: two of the three have already appeared. Two orthopedic surgeons have requested privileges at a center that opened last spring, and one commercial payer has announced a site-of-service policy for certain procedures. I would treat the quarter as the early edge of a trend, not as noise, and begin planning on that basis while watching for the third signal.

Reply to Erin

Erin, you read your system's falling emergency visits as a trend toward urgent care. What would you expect to see if it were a mild flu season instead? If both explanations predict the same volume report, the volume cannot settle it. Is there a decision someone is making differently, such as an insurer changing copays for emergency visits, that would separate the two?

Reply to Sam

Sam, your post lays out the possible causes of your hospital's falling inpatient census well, but it ends by recommending another two quarters of monitoring. Given Ansoff's point that the response window narrows as the signal clarifies, what would you be willing to act on now? Even a provisional reading, stated with the observation that would reverse it, gives your executives more than another report.

References

Ansoff, H. I. (1975). Managing strategic surprise by response to weak signals. California Management Review, 18(2), 21-33. https://doi.org/10.2307/41164635

Hollingsworth, J. M., Ye, Z., Strope, S. A., Krein, S. L., Hollenbeck, A. T., & Hollenbeck, B. K. (2010). Physician-ownership of ambulatory surgery centers linked to higher volume of surgeries. Health Affairs, 29(4), 683-689. https://doi.org/10.1377/hlthaff.2008.0567

Rogers, E. M. (2003). Diffusion of innovations (5th ed.). Free Press.