The 358-bed O’Connor Hospital in San Jose, Calif., is in many ways a typical community hospital. One of three Daughters of Charity Health System hospitals in Northern California, O’Connor sees its share of surgical volume, with clinicians performing about 6,000 procedures a year in its 13 OR suites (11 regular ORs, plus two endoscopy suites).
For busy hospitals like O’Connor (with dozens of surgeons regularly performing procedures there, and 125 altogether in the perioperative area), optimizing perioperative processes is becoming increasingly important. What’s more, linking any perioperative information system to the hospital’s core electronic health record (EHR) is also an essential step. In that regard, O’Connor leaders have been working with both their perioperative IS vendor (the Alpharetta, Ga.-based Surgical Information Systems, or SIS) and their core clinical vendor (the Reston, Va.-based QuadraMed Corporation) on process optimization recently.
Myrna Chang, R.N., D.H.A., is O’Connor’s nursing director for its perioperative services division and for its wound clinic. It is Chang who, as chair of the Daughters of Charity Perioperative Services Team for the five-hospital Northern California Region of that health system, has been helping to lead an initiative to cut costs and improve efficiency at those five hospitals. In that context, O’Connor went live with the scheduling and billing components of the SIS product in December 2008, and then with nursing documentation in June 2009. Then last year, O’Connor went live with SIS’ anesthesia clinical documentation system, and with its clinical analytics system; the hospital also implemented its patient tracking module. Finally, later last year, the hospital went live with its QuadraMed EHR.
Myrna Chang spoke recently with HCI Editor-in-Chief Mark Hagland regarding O’Connor’s journey forward in clinical information systems in the perioperative area and more broadly. Below are excerpts from that interview.
You and your colleagues made the decision to wait to go live with your core EHR until after you had implemented your perioperative information system. Can you explain the strategy involved in making that choice?
Yes, we established the tone, and were very successful with the implementation, with the perioperative information system; and within three months of clinical documentation implementation, people felt very comfortable with that system. And in August 2010, we started using the vendor’s analytics component more actively; and we’ve been making extensive use of that component ever since then. All those elements not only created success within the perioperative area, but also set us up well for success with our core EHR implementation.
Overall, what benefits and gains have you realized from implementing and leveraging the perioperative IS?
Let me tell you a story about how the analytics component in particular has made a difference in our hospital. Several months ago, we were holding a surgery executive committee meeting, and I was showing the members of the committee some reports coming out of the analytics component of our perioperative system, and one of the reports, which the SIS people had helped me to customize for us, had provided average interoperative durations by individual surgeon, with each surgeon named. And, on the chart, one surgeon’s average interoperative time was highlighted.
And what we call his “cut-to-close” time was 166 minutes, which was significantly longer than that of any of the surgeons on that chart. And no one said a word about that physician’s performance specifically; but he was in the room when that slide was shown. And one month later, his cut-to-close time had gone down to an average of 149 minutes, which was in line with those of the other surgeons. And that was within one month—with no prodding from anyone!
So that physician essentially immediately changed behaviors on his own?
Yes, because he didn’t want to be an outlier. That’s how powerfully peer pressure works with this group.
So simply by sharing data, it makes physicians change what they’re doing?
Yes, exactly. And that physician introspected, and probably made certain adjustments; perhaps he became less verbal in his cases. But I can only imagine what he was feeling when he saw the graph showing him as the most extreme outlier. And it’s true—they want knowledge, they want to know what’s going on, and how they make change. They’re human, too!
This all takes place within a context, of course. You and your colleagues are having to consider the potential for provider reimbursement cuts under Medicare, in the context of the current federal budget situation, correct?
Yes, that’s right. And it means continually looking at data in order for us to improve clinical, operational, and financial performance. Let me give you an example. For laparoscopic cholecystectomy, the average cost per case in 2010 for all our surgeons performing that procedure was $1,240; and of course, when we initially presented them with the data, they challenged its accuracy. So I said to them, OK, maybe this isn’t precisely accurate, but this is our only source of data. And within the next year, it had gone down to $742 per case.
That’s about a 40 percent drop in cost per case in one year (2010 to 2011). And we have 327 cases per year, so that translated to $162,846 in annual savings. This savings amount represents supply costs, and those costs are not reimbursed. The DRG (diagnosis-related group/prospective) payment is around $1,600 per case—it’s about a 40-percent increase in revenues. These are hard costs—supply costs—they don’t include labor or hospitalization on lap-choly.
You’ve had good results with knee replacements, too?
Yes, that’s right; our cost per case with lap-choly surgeries went down from $9,261 to $8,506 per case in terms of supply costs, from 2010 to 2011. In the past, I wouldn’t have known how our negotiations with supply vendors affected this, without our analytics capability.
What lessons have been learned in your organization around efficiency and effectiveness?
Before we implemented computerization, on average, I would have three or four physicians knocking at my door every day because of instances in which their patients not ready for surgery. As of the last eight months, I’ve only had three complaints in the last eight months, versus three per day. And I look at the speed at which they process patients now, because that’s another thing that’s trackable, and we can show nurses who’s performing well and not—and we’re seeing a significant reduction in the length of time needed to flow patients through the perioperative phase. In sum, we’ve reduced the number of complaints on the part of physicians, from 3-4 per day to 3-4 every eight months.
What would your advice be to CIOs and CMIOs?
Implement computerized documentation, because it not only improves efficiency, but also allows for easy access to data that will help physicians in their decision-making processes. And the value of IS-facilitated cannot be underestimated. And future quality improvement and performance improvement will come out of some of these processes.
Are you planning on doing any analytics around quality of care and outcomes?
Yes, we’re starting to build that capability, but we’re not there yet. I did a presentation on making beta blocker administration a required field. We had started at about 33 percent [of the recommended administration of beta blockers within one hour of admission following a heart attack], and leapt up to 99 percent.
What was the cause of the change?
We made beta blocker administration a required field in the nursing/clinical documentation. We’re still working on sustainability… but that change took place within one month, back in June 2009.