Today marked the close of the two-day conference, and tomorrow’s press conference will unveil the consensus statement drafted by members of the writing committee. Some interesting themes from today:
1. Nursing issues as a barrier to improving inpatient diabetes care:
• The first day detailed the immense need for blood glucose control in-hospital and set the stage for questions 4 and 5: what are the barriers, and what are effective strategies for overcoming these barriers? Nurse and CDE Linda Haas (VA Puget Sound HCS, Seattle) began the day with a presentation on this topic. In 1997, 52% of diabetes patients were in the hospital for a non-diabetes-related reason, meaning that their diabetes is secondary. Some of these patients were undiagnosed at time of admission (a topic addressed in the last presentation of the day by Dr. Vivian Fonseca). Haas stressed that diabetes is not the main issue on which HCPs are focused, and nursing attention is dedicated elsewhere. That isn't a problem in and of itself as long as diabetes is managed, we think - but it obviously often falls through the cracks.
• Nurses fear hypoglycemia: Haas said that it was “common in an inpatient setting” for nurses to skip basal and prandial insulin because of a two-digit BG reading—even a number like 80 or 90 well within the desired range. She suggested that ICU nurses are accustomed to elevated blood glucose levels and need to be reeducated about what is normal.
• Equally problematic, nurses do not have adequate education. In a survey, only 41% said that they believed they had access to adequate education on diabetes. While significant changes have occurred in diabetes care in the last five years, 28% of nurses reported no (no!) continuing education on diabetes in the past 2-15 years.
• A theme of the talk and of the day was the issue of timing: insulin is often given hours before food arrives, especially around shift changes, as tray delivery is unpredictable. Also complicating the picture, patients who are ambulatory eat food from vending machines or cafeterias, or visitors may bring food for patients.
2. Lack of a young generation of diabetes specialists:
• This is something we've been very focused on and concerned about - it was good to hear CDE Geralyn Spollett of the Yale Diabetes Center address what is becoming a crisis in endocrinology. She emphasized the need to train a new generation of CDEs and better diabetes training for nurses in general - we would add there is a big need to attract and retain a new generation! Spollett noted that while there are 20.8 million patients with diabetes, there were only 13,000 CDEs as of 2004. It's a difficult job and doesn't come even close to receiving the respect, pay, or reimbursement necessary to attract larger numbers - this is a problem that we hope to see industry and the associations begin to address.
• Spollett highlighted deficits in diabetes knowledge: a university-based hospital study found that, on the Diabetes Basic Knowledge Test, nurses scored a mean of 73%, a failing grade. Another study showed that the more staff nurses perceived they knew about diabetes, the less they actually knew (that's like, at Harvard Business School, 92% of the class believe they are in the top half in terms of grades, achievements, etc.).
• Spollett concluded that “We have a crisis in the delivery of diabetes care, across the board.”
3. Practical strategies and IV infusion protocols:
• The bulk of the sessions on day two were dedicated to question 5, “effective strategies for achieving improved diabetes management in hospitalized patients.” Our favorite was the lead presentation by Dr. Phil Goldberg of Yale, who discussed how his hospital had created a protocol after failures with the Van den Berghe and Furnary protocols (which led to what he termed “an epidemic of hypoglycemia.”) This is likely because the right infrastructure isn't available.
• Dr. Goldberg’s presentation succeeded in making the problem of inpatient glucose management very apparent: his first case study was “a 61 yo man with multiple myeloma, admitted with multilobar pneumonia. Required ventilator, broad-spectrum antibiotics, high-dose steroids, dopamine.” His point was that, while all of the diabetes specialists in the room were thinking about diabetes, in the intensive care setting there are many other competing priorities.
• Prior to 2001, Yale’s top academic tertiary care center used sliding scale, and presentations of real charts and the “sawtooth” blood glucose levels were incredible.
• From the point of view of continuous monitoring, Dr. Goldberg’s presentation was fascinating: in one case study, the BG dropped from 360 to 160, and the Van den Berghe protocol recommended raising insulin—this is a case where direction and rate of change is asbsolutely critical. Dr. Goldberg’s protocol takes into account “the previous BG” and the “velocity of BG change” -- undoubtedly, we'll see much more of this with the emergence of accurate, real-time continuous glucose monitors.
• Dr. Golberg also showed data on the use of CGMS in the MICU, reporting data with a 0.88 correlation coefficient.
• Dr. Goldberg also emphasized that gradual implementation is important to avoid massive hypoglycemia, and for this reason he does not think JCAHO should mandate certain levels; he believes this would result in a dangerously rapid implementation of protocols at hospitals. Hmm. We aren't sure - on some level, some JCAHO mandating will make things move faster no matter what. Possibly JCAHO could mandate levels that would be easier to reach at the start but would change annually - so hospitals could change somewhat gradually - but better than not at all! And, hospitals that change more quickly could be rewarded. Food for thought.
• Dr. Goldberg also spoke to the importance of educating nurses about what normal levels truly are. He related a story about asking the nurses to estimate their own BG and then using a glucose meter to take their actual measurements—most of the (non-diabetic) nurses estimated that they were 170 (the number they thought of as “normal”) and were shocked to learn that they were at 72, a number they thought of as hypoglycemic. CRAZY! If nothing else does, this really underscores the education problem.
4. Diabetes safety committees and hypoglycemia:
• Dr. Mary Korytkowski of the University of Pittsburgh delivered a presentation on the creation of a “diabetes inpatient safety committee” at her hospital, which was most interesting because of the data she included on hypoglycemia.
• Reviewing the charts in her hospital, she found several episodes of severe hypoglycemia among inpatients, one of which was associated with a poor outcome. She noted that hypoglycemia was a “frequent, dangerous event for which there was no reporting mechanism and no standard treatment approach.”
• A breakdown of the causes of inpatient hypoglycemia (defined as BG < 40 mg/dL) attributed 82% to a “divergence in course of care,” such as a patient being called away to an x-ray while on a feeding tube, after the delivery of insulin. Other factors included “inappropriate medication selection,” “inadequate monitoring,” “performance,” and “patient behavior.”
• A retrospective case control study of 60 patients older than 65, which defined hypoglycemia as < 50 mg/dL, found that only 23/60 (38%) of episodes were noted at the time of occurrence. Mortality in the group experiencing hypoglycemia was 48%.