Also, like many of you, I have been through numerous JC (formerly JCAHO) triennial inspections, and while there is clearly some sense of déjà vu each time around, there really were some differences and innovative challenges this year. Notably, this was our first “unannounced” survey, a protocol intended to test our continual readiness and excellence; this contrasts markedly with past surveys, which seemed structured simply to test our ability to prepare for an inspection. Second, the team of surveyors from the JC and the Institute for Medical Quality were as tough as we have ever experienced. Overall, it was a significant learning experience, and in keeping with our goal to continually improve and be ready, we must now start the process of change and improvement.
I’d like to share with all of you some of the JC’s “real world” recommendations, all of which make sense as tools for us to improve health care. Three of the findings involve medication management. These make the most sense when viewed through the filter of “crew resource management” (CRM), the paradigm I talked to you about last month.
Medication
The surveyors noted that the PRN (pro re nata) orders many of us write for medication to be given “as needed” were issued without noting what the prescription was to be used for. Often we believe the usage is obvious to ourselves, and more importantly, to the nurses who actually dispense the medications. For example, if we write an order for Tylenol, we assume — usually correctly — that nurses understand that this is for control of pain or fever as needed. Fair enough, but in the new paradigm of safety we must spell that out and not just assume it to be true.
Just as airline crews go through a checklist, we must learn to be equally rigorous about avoiding assumptions, no matter how logical or even how seemingly trivial. The flight crew doesn’t assume that the fuel tank has been filled — the crew checks and confirms. Consequently, we are fine tuning our procedures to benefit from the safety principles developed from CRM, initiated by NASA in 1979 to save lives in the sky.
Similarly, how many of us have written what we assumed were perfectly clear orders for, say, 5 to 10 milligrams of morphine to be given every 3 to 4 hours as needed for pain? Should we assume the nurse would start at 5 milligrams? That may be what we intended and what the nurses we regularly work with will do. However, patient safety dictates that we must put aside assumptions and instead specify the starting point and then spell out the escalation criteria. Similarly, when ordering medications with therapeutic duplications we must indicate how to use each medication (morphine for moderate pain and Dilaudid for severe pain).
Both PRNs and escalation orders will be made easier and clearer when Epic comes on line and builds in questions for us to confirm. But we can’t afford to wait. We need to change our practice now.
A third medication issue involves prelabeling syringes that are slated to be filled later. For proper medication safety we should label the syringe after each medication has been drawn. That reduces the risk that the syringe will contain the wrong medication, since labeling retrospectively requires us to note what’s IN the syringe, not what we expect someone to put in it later. I’ll admit that sounds like a subtle distinction bordering on bureaucratic red tape. However, the error isn’t so subtle on those admittedly rare occasions when an error is made, especially on procedural units where sedation or anesthesia may be delivered — and where retrospective labeling (fortunately correct) was found by the survey team. Again, CRM principles are in play.
Chart notes
The surveyors found some unapproved abbreviations in charts. Interestingly, they noted these abbreviations are like cockroaches — very difficult to eliminate completely. The visiting team was in fact complimentary, telling us that we probably had the fewest incidences of unapproved abbreviations encountered at any center our size. Still, we must be vigilant. We cannot assume this problem is endemic and unfixable.
The surveyors found that some of our notes lacked appropriate information, such as op notes that didn’t document blood loss during a procedure or specimens removed. They also found missing updated interval histories when procedures were performed more than 24 hours after the initial H & P. All that is required to fulfill the interval history requirement is to examine the patient briefly and then document if anything has changed since the last formal history.
Some of our medical records contained totally unreadable handwriting. This is a problem with which I can relate since my handwriting sometimes can be atrocious. Again, this is a problem that Epic and electronic documentation should attack, but meanwhile, I, like many of you, will need to be aware of penmanship.
And clarifying orders goes beyond good handwriting, of course. Any orders which are not complete will need to be clarified before they are carried out, and physicians will continue to hear from a nurse or pharmacy whenever clarification is needed.
‘Time outs’
We seem to be doing fairly well with the “time out” in the operating rooms. However, for some of our procedures occurring elsewhere in our hospital we are not always as compulsive about performing the “time out” and documenting it. To be meaningful, the time out should really be performed just before the procedure starts, not an hour before a long set up.
Staff Governance
We had three “RFI’s” (requirements for improvement) relating to Medical Staff issues that emerged after the surveyors reviewed our bylaws and interviewed individuals. (See Kent Garman’s President’s column). One of these focused on “self governance.” We have had several opinions about this recently, and we have already taken steps to ensure that our Medical Staff governance structure complies with JC requirements and state law. In June, the Medical Board approved a motion put forth by pathology chair Steve Galli, M.D., to mandate the Chief of Staff and the President of the Medical Staff to appoint an ad hoc committee to study all of the opinions of our counselors, both past and present. The committee will also review and use the Bylaws Committee’s recent and earlier proposals as a foundation to propose a governance structure that will meet the intent of the JC and state law and will also reflect the special needs of our Medical Center. This ad hoc committee has been given an aggressive timetable to come up with a structural plan by September.
The second RFI related to fact that we as a Medical Staff do not have our own elected representative on the governing board of the Hospital. The board includes four physician members, but they are chosen by the board and not the Medical Staff.
The third RFI noted that the hospital and Medical Staff bylaws relating to the leadership structure of the Medical Staff were not in synch. However, it should be easy to synchronize the two sets of bylaws once our Medical Staff bylaws are revised to reflect a revised governance structure.
Supplementals
We had some supplemental recommendations that did not rise to the severity of an RFI. The surveyors noted that sometimes a complete pain assessment was not performed after pain medications were given. This is an issue that can be addressed by nursing documentation.
In the clinics, some medical assistants appeared to perform specific procedures that may be outside their scope of practice, since a formalized policy authorizing them to perform the procedure was not in place.
• • •
All in all this was a very successful survey for Stanford Hospital & Clinics. Although our next unannounced survey should not occur for three years, we still must maintain continual readiness. We want to hold our gains and work on that which we know we need to improve.
If you have any questions about the survey or these lessons to be learned, please contact KIM PARDINI-KIELY, Director of Quality and Patient Safety (kpardinikiely@stanfordmed.org); KEVIN TABB, M.D., Chief Quality/Medical Information Officer (ktabb@stanfordmed.org); or myself (lshuer@stanford.edu).
