Clinical Research Oriented Workshop (CROW) Meeting: May 12, 20010

Present: Matt Bovee, Abby Crocker, Amanda Kennedy, Rodger Kessler, Ben Littenberg, Charlie MacLean, Carol McBride, Connie van Eeghen (by Skype)

1. Round Table: Rodger updated on the application asked what the reference re: “counsel” is related to. Per Ben, these are the super-reviewers, who make additional suggestions to create a balanced portfolio. They make a recommendation to staff; the director of the Institute decides.

2. Rodger’s Research:

a. This paper is the outcome of a work reviewed a year ago, which was split into two possible articles. The presentation expected this June has been delayed, but Rodger is pushing forward with the work anyway.
b. Abby: no much detail on costs, although this is one third of the evaluation model
c. Amanda: got lost on figure 1 (the model)
d. Charlie: why is a new paradigm needed for mental health, relative to the evaluation of other clinical care processes? Rodger: need a consistent set of terms to discuss the elements and evaluate the results of this somewhat unique challenge of integrating mental health into primary care practices.
i. Ben: PCMH (primary care medical home) is not focused on mental health.
ii. Rodger: mental health has a high impact on morbidity and morality; has larger obstacles with respect to access, identification, and use of evidence based treatment
iii. Ben/Charlie: The key topic of interest may be the hypothesis about collocation, but this article is about evaluation of the outcome of those services.
1. Argument 1: here’s what we need the health care system to do (mental health)
2. Argument 2: the current solution (collocation), is one of many solutions; we need a measurement system to evaluate any solution
iv. Big problems in MH; many attempts to change; how to identify how to measure success will transcend any solution put forward
1. Berwick’s framework works – three aims of patient experience, clinical care, and cost, which can be evaluated over a population and society’s choices (e.g. care of acute care elderly as a tradeoff to providing preventive care to a pregnant woman) – per Charlie
2. Peek’s framework works – clinical, operational, financial – or what you need to do
3. Rodger explains how to use them
e. The use of “model” is either a view of the world or an approach to solving a problem. Rodger’s “model” is an approach, which is different than how to evaluate. This is a model of how to think about quality in mental health as applied to a PCMH.
i. First, make the problem obvious to policy makers. Produce measurements that shame the world into taking MH seriously.
ii. The results of the measurement system will identify the solution, but don’t make that proposal in this paper. Invite other practices to measure this for you, and get good at collecting the data.
iii. In another paper, use the results of the measurement system to make this case.
f. Ben provided an 18 segment framework in the shape of a cube
i. Access, outcome, cost
ii. Population, person, practice
iii. Cost & quality

g. Amanda: pretty complex. Cut the Gordian knot: just pick three measures and state them as a starting point that strips away the complexity of all the past thinking.
i. Pick them to fit your conceptual framework, but don’t explain the framework
ii. Get away from cubes and trees and other complex constructs
iii. Stimulate people’s imaginations to understand how to use the measures, both to collect and to apply
1. Improve the precision with which the metrics are described
2. Don’t rely on only EMR; there are other methods… and EMRs are not always (maybe not usually?) simple data collection devices
iv. Remove everything else; there is too much stuff in this paper
h. Review of the metrics
i. Suggestion: start with a simple outline – what is the minimum necessary for the message
ii. Assertion: Behavioral care is in trouble; there is a lot of unmet need, quality of care is poor…
iii. The characteristics of ideal care are uncertain. Many proposals; a range of suggestions…
iv. Systematic improvement requires agreed upon measures and common goals/consensus
1. Berwick’s triple aims, as an example of selective focus on key components
2. Peek’s relates directly to the mental health world
3. Identify the deficits in current measures; don’t compare the two above but use them as examples of what good measures look like
4. Note that most/all of the measures proposed are process measures, not outcome measures
v. Measures need to be explicit, easily calculable, applicable across multiple settings, transparent (e.g. high face validity), and related to the goals of care
vi. Propose: attend to the following domains that are particularly notable trouble areas:
1. Access
2. Identification
3. Use of evidence based care
vii. Measures:
1. Access
a. 95% of patients seen in less than 1 day (constants – the percent and the number of days – may be placed under discussion)
b. # of 1 day patients/# total patients referred
c. Check the literature on measures of access; managed care has created great metrics to evaluate access
2. Identification (not covered)
3. Use of evidence based care (not covered)

3. Connie’s research request:
a. Connie has transcripts from interviews, surveys, and a field journal intended to identify characteristics related to a set of independent and dependent variables about conducting a quality improvement project in a provider practice.
b. She would like to set up a method of analyzing these data by asking a set of individuals (those who volunteer from this workshop) to sort these utterances into these categories (the two types of variables) and to ask what, if any, conclusions each of us would come to.
c. Proposal: Each volunteer receives a set of PowerPoint slides, with one utterance per slide, to be sorted (using the slide sorter) under a set of “variable” headings, with a conclusion from each person for each header.
d. The conclusions will be collected and discussed at a future workshop meeting
i. Ben, Charlie, Rodger, Abby, Carol McBride agreed to participate

4. Next Fellows Meeting(s): Wednesday, 2:00 – 3:30 p.m., at Given Courtyard Level 4
a. May 19: Matt’s Diabetic study for the next steps to achieve progress
b. May 26: Abby’s questions on statistical tests
c. June 4: (no Connie)
d. June 11: (no Connie or Matt)
e. June 18: (no Connie or Matt)
f. June 25:
g. Future agenda to consider:
i. Rodger’s mixed methods article
ii. How to predict medical events effectively (Ben)
iii. Future: Review of different types of journal articles (lit review, case study, original article, letter to editor…), when each is appropriate, tips on planning/writing (Abby)
iv. Future: Informed consent QI: Connie to follow up with Nancy Stalnaker, Alan Rubin will follow up with Alan Wortheimer or Rob McCauly

5. Fellows document – nothing this time – to be reviewed after trialing Wednesday meeting times

Recorder: Connie van Eeghen