Do Clinical Quality Measures matter?

Do Clinical Quality Measures make a difference? Is healthcare actually improved by all this effort at capturing various reporting metrics, and reporting it? That is a question raised by practitioners and commenter’s regularly.

Measuring clinical quality has been an emerging science over the past 15-20 years, and incorporating the systematic measure of quality has become ingrained in the day-to-day processes in many healthcare settings – mostly institutional (hospitals, large group practices, and the like). Good performance measures are the key. The National Quality Forum (NQF), which serves to vet quality measures for scientific, evidence-based soundness, sums it up best: “How do patients know if their healthcare is good care? How do providers pinpoint the steps that need to be improved for better patient outcomes? And how do insurers and employers determine whether they are paying for the best care that science, skill, and compassion can provide? Performance measures give us a way to assess healthcare against recognized standards.”

Meaningful Use contains a set of Clinical Quality Measures (CQM) which must be reported on, although the first stage is fairly minimal – a provider, in order to attest for Meaningful Use, must submit 3 core measures (or alternate core measures, if the core ones don’t apply) plus 3 out of 38 menu measures, depending on applicability to one’s practice (or 0 if nothing applies).

To clinicians who are not used to reporting CQM, this may seem new – past CQM reporting was done by institutions, and carried out by administrative personnel, and not the clinician directly. Meaningful Use brings CQM into the practices of everyone – particularly small practices, scattered throughout the country, not necessarily involved with hospitals or large medical groups or IPAs that can do that reporting for them. It may seem like “just another hoop to jump through” in order to access federal Meaningful Use money – yet the linkage between that and actual change in clinical care might seem nebulous. It may seem like a burdensome “pay for reporting” exercise, rather than something truly meaningful.

Different kinds of measures
In order for CQM measures to have an impact, they must be linked with some kind of “action plan” that can be implemented by the clinical practice. So now you have a list, and you can report that list to CMS in order to check off that box needed for Meaningful Use attestation, but… so what? Are your patients getting better as a result? Answer: you need a method to do something with the data.

It might be helpful to separate CQM measures in order to understand them better: Process Measures, and Outcome Measures. “Action plans” for each type of measure would be different. We’ll use examples from the Meaningful Use CQM set here.

Process Measures
A Process Measure would be something like “percentage of patient visits with a diagnosis of hypertension with blood pressure recorded” (NQF 0013) – an obvious thing: measure blood pressures every time with patients who have the diagnosis of hypertension. The measure is simply a process – record the blood pressure.

The “action plan” for a process measure is to link the measure with a Clinical Decision Support (CDS) prompt. In the Electronic Health Record (EHR), one can display a prompt that will become active if the patient has the diagnosis of hypertension in their problem list, will be “red” if there is no blood pressure recorded on the current visit, and will become “green” once the blood pressure is recorded. Very simple, but a good example of a CDS rule linked to a CQM measure. (Bonus: implementing at least 1 CDS rule is also a Meaningful Use requirement)

Outcome Measures
Outcome Measures require something beyond simple process-prompts. An Outcome Measure would be something like “percentage of adult patients with diabetes who have blood pressures less than 140/90” (NQF 0061). Granted, this is a low bar, as other measures might measure blood pressures for diabetics that are below 130/80 – but for Meaningful Use, we’ll just use the looser threshold.

Sure, a CDS prompt can be included that will prompt the clinician to record blood pressures for diabetics (in addition to hypertensives). But the measure is to calculate the percentage of diabetics in the practice with good blood pressure control. Is your percentage 70%? Is it 85%? Is it 90%?

This is where some accountability structure is important. Within organizations that have cared for assigned populations (e.g. HMO patients) in a Pay-For-Performance environment, differential performance-based payments can be devised to reward the practices who achieve 90% more than those who achieve 70%. The envisioned Accountable Care Organizations (ACOs) – a proposed structure for Medicare fee-for-service patients – may devise similar methods for financially rewarding physicians for good performance on Outcome Measures.

However, financial incentives are simply a motivator, not a solution. The practice who performs at 70% may not know what the 90% practice is doing differently. That is where safe, internal forums for sharing experience is important. Within a healthcare delivery organization (like a hospital, a medical group, an IPA, or even a clinical practice), Peer Review activities are protected from discovery in malpractice litigation – thus, such settings are safe places where clinicians can learn from each other’s experience.

An online forum, on the other hand, does not have the same protection from discovery that a healthcare delivery organization has. Thus, Peer Review and collegial counseling are a bit more risky. Given how nascent collegial clinical forums are, there is not yet any legislative, administrative or legal framework which would protect these “non-traditional” settings from malpractice discovery, and therefore offer the same kind of safe harbor that is needed in order for clinicians to openly discuss quality-improvement methods. This is something we will advocate, so that the kinds of safe settings for quality improvement that have existed traditionally within hospitals, medical groups, and clinical practices can be extended to virtual on-line secure forums for the purpose of quality improvement.

Conclusions
Clinical Quality Measures have been embedded in healthcare institutions for years. Their use in ambulatory, small-doc practices is new – Meaningful Use will drive some of this. Process Measures lend themselves to Clinical Decision Support prompts which can be built into EHR systems. But Outcome Measures need some sort of accountability structure where clinicians can interact with each other and learn from collegial best-practices. These forums have traditionally been within Care Organizations – hospitals, medical groups, IPAs and clinical practices.

However, as technology evolves, and on-line clinical forums emerge (which can be rendered secure from a technology standpoint), new legal rules need to be created that protect such Peer Review forums (specifically, protection from malpractice discovery) in order that actual quality improvement can occur. The result? Over time, clinical quality scoring improves. Populations suffer less disease burden, and patients get better care. The move in that direction is unstoppable.