Our Medical Executive Committee recently received a report from our Critical Care Committee. I cannot be more proud of our staff and the progress they have made to reduce harm and improve quality of care in our ICUs. I include two of the charts.Let me translate the implications of the reduction in Ventilator Associated Pneumonia (VAP). Preventing 744 cases over three years -- at a treatment cost of about $20,000 per case -- translates into a societal savings of $14.9 million during this period.
The rate of central line infections also dropped from 4.14 to 0.52 cases per 1000 patient days between FY2003 and FY2009, a reduction of 83%.
This probably reflects lost revenue for the hospital under the fee-for-service reimbursement system. So why do we do it? First, because it is the right thing to do and saves lives.
Hundreds of lives.
On the business front, it has contributed to a reduction in length of stay in our ICUs. We were able to avoid the multi-million dollar capital cost of expanding our ICU capacity. Indeed, we were able to create capacity out of the existing facilities and improve throughput.
I hope that those who argue that global payments (i.e., capitation) are a necessary condition to create societal cost savings and improve patient care will read this. I do not deny that such a payment methodology may be worth implementing for other reasons, but there is a lot that can and should be done under the current payment system.
While the state debate goes on about cost control, why can't we get all of the hospitals in Boston to release information like this about their quality improvement efforts to provide the public and public officials with a sense of confidence that we care about these matters and are willing to be held accountable.
It is worth highlighting that the keystone of these interventions was the reflexive impulse developed by leaders to put the patient at the center of work. These improvements happened in a context of greater participation of families and patients, and fresh ways of thinking about possibilities rather than only constraints. It is a model worth sharing.
ReplyDeletePaul,
ReplyDeleteI applaud your efforts to improve quality, but lets be frank; most of these cases I suspect are under Meidicare, which pays your hospital on a DRG basis, so shorter lengths of stay result in decreased use of resources and increased net profit.
Secondly, Medicare is giving close scrutiny to these indicators and will be using them shortly in a punitive manner in its attempt to limit payment for shoddy care. This is driving more hospitals to collect this data (which I believe they have to under current Medicare rules) and nobody wants to be on the lower end of the quality care scale when this data is released.
Third, I would hope you are using this data as your bargaining chip with the commercial carriers when negotiating contracts. Hopefully they are willing to pay more to steer patients to hospitals with good quality data instead of the biggest health care system with the best name recognition (can we say Partners?).
I am curious whether you have an ethics committee or ethics consultant that could have helped with decreasing the length of stay in you ICU. Do you have any way of tracking the impact the consultant or committee had on the length of stay in the ICU? If so, how did you measure it?
ReplyDeleteWhether or not global payment ensues, the hospital which is most efficient and patient centered will be well positioned, particularly given Keith's observation concerning CMS withholding payment for an increasing list of "never events."
ReplyDeleteSo you are doing the right thing in lots of ways. Congrats!
nonlocal
This is really really good - maybe even better than you think, if severity adjusted...
ReplyDeleteBravo! It's a shame that when a horrible complication is avoided, no one notices because the achievement is a nonevent.
ReplyDeleteHuzzah for nonevents!
I'd like to learn more about the processes that lead to these reductions. great work!
ReplyDeleteGreat work, both with regards to the reduction and the transparency. I'm currently completing my MHA and patient safety and quality improvement is one of my passions. I was wondering if you would be willing to share more on this reduction. I would be interested in learning about some of the interventions that were undertaken to promote this change.
ReplyDeleteJust wondering if the ICU is also using the Pronovost checklist and, if so, if your outcomes jive with those in Michigan?
ReplyDeleteIntersting that you posted the VAP data. We just took a look at the 2009 VAP data at our hospital. Our infection rate is very low (which we are proud of) but what was more interesting was the dramatic decrease in the number of device days. While the total number of vent patients was about the same we saw our days on a vent drop by 40% from the previous year. We posted this on our intranet and shared it with the medical staff.
ReplyDeleteStefani - we have been using checklist-driven care for at least a decade, now integrated into our critical care information and doumentation system (which all of the ICU docs write our notes in), so we did not change to a different checklist based on Pronovost's work. Ours turns out to be very similar.
ReplyDeleteAnon 1:46 -- completely agree that the conventional metric (pneumonias per 1,000 ventilator days) accounts inappropriately for the phenomenon that you've described here. We have also seen a drop in our ventilator days, which ironically (because of the mathematical construction of the metric) can cause your rate to go up! On the research side, we are doing work (funded by the Robert Wood Johnson Foundation) trying to better quantify this issue using a mix of real data and stochastic simulation.
- michael, one of the ICU docs at BIDMC