8/2004: NCEP Updates Adult Treatment Panel Guidelines since 2001.
LDL cholesterol less than 100 md/dl and less than 70 for
those at high risk (DM, smokers, poorly controlled HBP,
metabolic syndrom, prvious MI)
8/2004: NYS DOH QARR measures:
Breast cancer screening
Chlamydia Screening (16-25)
Domprehensive DM Care for poorly controlled
Asthma: appropriate medications
Controlling HBP
2004: Quality Assurance Department: x3914, 6408
Josephine Jalandoni, QAC RN (Joji)
Lulu Ibanez, PA
Grand Rounds by Dr. Danial Buff (who interviewed Dr. Swango)
"Two wrongs don't make a right." The second wrong is in not reporting
the first)
IOM says there is something wrong with the system, in "To Err is Human"
indicates that there are 44-98,000 deaths/year.
A "Quality Gap" refers to standards not being met (e.g., didn't give
beta blocker to post-MI patient), versus an "error".
80% of errors result from communication problems (e.g., attending to
attending, resident to attending...)
Medication errors ("ADRs") - 28% of ADRs are errors.
ADRs are high in patients transferred from NH (dig, coumadin)
If survey patients about errors, they will report a high number.****
Six Sigma refers to error rate of less than 3.4 per 10*6
"Safety is job one." Lee Iacocca
A culture of safety that is non-punitive - do not focus on the
individual - focus on the system: standardize
In the 1980's, anesthesia solved the sandarization problem
and got to 5.4 errors per million (about 5 sigma).
JCAHO stepped in for patient safety; but now also need to
show performance improvement (e.g. restraints)
DOH-ACGME work hours (European Union: 56 -> 48 hrs/wk)
Feds: national guidelines for quality errors (electronic
record and Rx writing).
Now the government has Quality Measures via CMS (via IPRO) and
will publish the results for AMI-CHF-CAP, although we do not
yet know if this is decreasing errors. In other words, does
a "corrective action" cause a different, new problem?
For example, NYS DOH publishes Mortality Data since 1994
corrected for risk (but now angiographers are avoiding
risky patients). So improving quality but limiting care.
What is the process of Performance Improvement (PI)?
Identify a Department or Individual "issue".
Investigate cause of the Department or Individual issue.
That is, perform a Root Cause Analysis (RCA).
Develop a "corrective action".
Then monitor the corrective action.
Example of a PI project:
Clinic: pap, ppd, A1c, PSA > 50 yo
Handwashing by swabbing resident hands at GR!
Reviewing all deaths and looking for errors
Routine screen of 20 charts for compliance with
quality indicators (aiming for 90% compliance - but
we don't live near Walgreens).
Use of D50 as reflection of hypoglycemia
Use of Lovonox in patients with a decreased GFR.
Then need to monitor the corrective action (& "Think!)
4/2005: Dr. Simeon Schwartz
An alternative name for PBLI (practice-based learning & improvement)
is Clinical-based Practice Improvement
-When doctors heard the 1999 IOM Report they said, "Not me".
-Now is not unlike 1787 right before the French monarche fell.
-We are now seeing "Group-based Quality Improvement Projects".
-When we are paid, should quality be part of the product?
-We are moving from the solo physician craftsman (similar to
the gun maker craftsmen of the pre-industrial revolution)
to interlocking parts (e.g. new guns, fire department hoses).
The airplane that crashed in DC several years ago had only been
de-iced once. This lead to a change in the system. Imagine if
you had to think about how to prepare the plane each time you
took off. So NCQA has a checklist of items that need to be
reviewed: HbA1c, eye exam, podiatris, microalbumin, lipids, BP
At SBH, Dr Patel has developed a list for DM
When things do not go as they should, they are analyzed by
Root Cause Analysis (RCA). Deing the plane.
If a pilot uses a pre-flight checklist, can't we use a DM
checklist?
So we use a process with self-assessment and review. But what
about "human" factors, such as sleep, emotion?
So, in additiion of process/procedure factors, there are also
human factors.
What if the human makes a mistake? Should we use the 'castration'
method of improvement? For example, a resident gives chemo to
treat sickle cell disease, but the patient does not have S/S.
Perhaps we should modify the system (and include computer
decision support).
An example of process improvement in the blood bank is the use of
the wrist band. To look at "quality", we might need to
look at both Outcomes Measures as well as Process Measures.
Death post-CABG Pat. had mammogram
Tracked pts on Vioxx
This brings us to "Pay for Performance" and "six sigma".
Sigma is a standard deviation. Hewlett Packard noted that
1/1000 chips was defective and they wanted to improve this to
1/10000 However, Japanese manager mistook this for
1/100,000 (=six sigma)
To get from 1/1000 to 1/100,000, the Japanese engineer
realized he would need to take dust, temperature, a re-
engineer the whole chip making process.
-Quality does not begin with the workers
-quality begins from the leaders
-Quality needs to be measrued, eg. nosocomial ICU infections
on a montly basis
-Six Sigma is cultural and conceptural
-Physician is the captain of the health care team - need
more than an individual
-At Cornell, Antibiotics for CAP is only 70%: there are
preocess & leadershipp problems.
-Might look at mortality post CABG or readmit rates.
-Doing the right thing makes a difference:
Inhaled Steroids in asthma causes LOS to go from 7 -> 2
-So MD compensation based on performance:
Process, Quality, Outcomes (helped by technology)
Clinical Practice Guidelines
เฉ๊์ฅ Decision Support
(The Craftsman vs the Team)
