Data transparency and clinical quality

Transparency of cost and quality data in health care has been a rallying cry for some time. The Commonwealth Fund published a sentinel article in 2006 outlining how collecting outcome and quality data, as well as cost data, is critical for any meaningful healthcare reform and in bending the cost curve going forward.

Recently, US News and World Report issued their annual hospital ranking report, which includes some quality criteria, but also subjective and anecdotal “reputation scores” (weighted heavily) to result in a ranking of questionable value. The report was sharply criticized in a blog piece by Paul Levy for failing to be truly quality-based. This underscores how much is at stake for hospitals when it comes to true data transparency. Despite the call for such transparency in 2006, we’re still not there yet.

But what about data transparency for ambulatory, community physicians? Will any valid measure emerge that is more useful than the subjective ratings (e.g. Zagat or Yelp ratings) available now? The answer may come from widespread use of Clinical Quality Measures – a part of the federal Meaningful Use effort to move clinicians off paper and onto Electronic Health Records (EHRs).

Making Clinical Quality Measure (CQM) results available to external audiences is, quite naturally, a threat to practicing physicians. Fearing scoring poorly on things they have no control over, objections arise in the form of “how can I be held accountable for the behavior of patients, some of whom refuse to accept medical advice?” or “people simply don’t want to take responsibility for themselves – how can I be scored on that?” or “my patients are sicker.”

Experience with use of CQM in clinical practice
Some areas of the country have a 10-15 year history of collecting Quality Measures, and reporting them internally to clinical peers. California has been viewed as the “land of the delegated model,” having large risk-taking medical groups and IPAs manage commercial HMO patients, and being rewarded by health plans with Pay For Performance bonuses (in addition to their ordinary capitation payments) – these bonuses are based on Clinical Quality Measures (specifically, HEDIS scores). What is the experience with this? How might it help guide wider (e.g. federal) policy concerning data transparency, and the use of Clinical Quality Data?

I will speak from the experience of being involved in a CQM-using environment, practicing within a large Northern California IPA. The objections cited above were common when CQM was starting to be implemented. Nevertheless, Primary Care Physicians (PCPs) have been paid by a combination methodology (fee-for-service + performance scores), with performance measured in multiple domains: (1) Clinical Quality, (2) utilization/cost, (3) participation, and (4) patient satisfaction. These measures are shared quarterly within panels of physicians (local geographic PCP groupings), so that a given practitioner could see how she/he performed compared to local peers.

Everyone in clinical practice will take care of some patients who refuse medical advice – the numbers spread out among practices. No one has 100% compliance (despite the fact that physicians, by nature, want to score at the top of the heap – we are used to being straight-A students). So CQM measures take on value when used to compare neighboring practices. For example: one CQM item is “what percentage of your diabetic patients have LDL cholesterol levels less than 100?” If my score is, say, 72% – 72% of my diabetic patients have most-recent LDL levels below 100 – then I might feel good about that. Years ago, when CQM was first being looked at, average scores may have been around 50%, and over time clinical quality has improved as a result of looking at this measure.

However, let’s say that a peer in my community has a score on that same measure of 80% – they are doing something different, which is resulting in a better score. In some safe, collegial forum (like a local meeting of PCPs within the organization) I might want to ask them what they are doing differently than me, so I can learn a best-practices improvement. Are they being more aggressive with medications? Do they have a nutritionist? Do they have a call-back program to reach out to people who haven’t been seen in a while? Do they have a different response when patients no-show, or cancel?

All of these things are part of the quality-improvement mix. It is local. It is within communities of peers who can interact with each other in a safe way. But it is also driven by availability of data.

Will modern EHRs bring CQM “to the masses?”
Traditionally, collecting CQM data has required heavy Health IT investment by groups and institutions – something beyond the reach of small practices, unless they participate in those groups. It has also required a forum for local physicians to interact with each other, in order to do something with this data.

However, modern EHR technology may be changing that. Given that CQM data is a requirement for Meaningful Use, and given that some EHR technology is web-based, lightweight, portable and free (e.g. Practice Fusion) – having been adopted by a wide swath of small-sized practices, especially Primary Care Practices – then the availability of such measures may become part of an ordinary-workday experience. CQM measures can be reviewed on-demand by physicians (themselves, or a designated delegate, such as a practice manager), and viewed in some context – perhaps compared to other docs in the same community, or docs in the same health plan, or docs with the same specialty, or even oneself over time.

In addition, peer-to-peer forums (perhaps with a moderator) can develop where safe, collegial discussion can occur, allowing focused advice-seeking from practices who score better – what are you doing that I might learn from, in order to deliver better clinical quality results (better CQM scores)? Does this take a formal group, or can an online forum achieve the same result? It is an interesting question, which can only be answered once the technology platform matures, and its use becomes widespread.

Modern EHR technology has the potential to incorporate Clinical Quality Measures into everyday medical practice. It also has the potential to create forums, where communities of clinicians can interact with each other in a safe way. It is not theoretical – there is a body of experience already in place, which can help shape the discussion nationally. Data transparency – the stuff that really counts – might just be achievable in the near-term future.