How CPOE can reduce length of stay
Computerized provider order entry, along with bedside bar coding and electronic health records has had a tremendous impact on patient safety and outcomes.
In their poster presentation at the 2015 HIMSS Annual Conference & Exhibition in Chicago, "CPOE Associated with Shorter Length of Stay in a Community Hospital," Richard Schreiber, MD, intends to help attendees understand that is a strong statistical correlation between the increases in rates of CPOE with a drop in length of stay.
In a July 2014 study published in Applied Clinical Informatics, Schreiber, chief medical informatics officer at Holy Spirit Hospital in Camp Hill, Pa., and co-author Steven Shaha, noted the correlation came out to be in the vicinity of about 63 percent.
"That is, about 63 percent of the reduction in length of stay correlates with the rise in CPOE rates. This was true overall, and was statistically significant for 13 of the 19 disciplines for which we had data," Schreiber told Healthcare IT News.
Since they were able to control on a per patient, per visit basis, Schreiber noted, they could demonstrate that CPOE was another independent factor.
The vast majority of the participants in the study were in-patient physicians caring for patients who were observed or admitted.
Researchers looked at the data for the last two years 2013 and 2014, and noticed that length of stay had crept back up a little bit.
"Indeed, in certain areas where the length of stay crept up, the CPOE rates diminished," Schreiber said.
Even though the case mix index – which is the overall assessment of the severity and complexity of patients – has stayed steadier or actually gone up a little bit, and even though there are all of those numerous other efforts still in place, to shorten length of stay, there is a mathematical correlation between CPOE rates and length of stay.
"In this case it happened to be inverse to what we had demonstrated before," Schreiber added. "I think that's supportive of a cause and effect relationship."
Mathematically speaking, said Schreiber, the adoption rate of CPOE fit a sigmoid shaped curve like with all other adoptions in human behavior – and length of stay demonstrated in the study showed a sigmoid-shaped reduction in duration of hospitalization – and the critical inflection point was not achieved until CPOE rates hit about 60 percent.
Stage 2 meaningful use, commented Schreiber, requires 60 percent of the medication orders to be CPOE and 30 percent of the RAD and LAB, which is not ahead of the inflection point.
"So if there is a site that has 'only' achieved the 30 percent and 60 percent numbers of meaningful use, they are not necessarily going to see a ROI in reduced length of stay,” said Schreiber. "I hope that those who are skeptical about the return of the technologies will look at it and say this whole discussion should really be in our past about whether or not to go electronic, that's number one."
The second goal would be that others get the idea to start looking at their data in this way and to see if you can prove what works and what doesn't work.
"The third thing is that our method is a little bit different – we looked at a per patient, per visit basis, and controlled for different CPOE rates by different clinicians in various disciplines."
The session “CPOE Associated with Shorter Length of Stay in a Community Hospital” is slated for Sunday April 12 as part of the Physician’s IT Symposium from 8 a.m.-4:30 pm.