The National Quality Foundation is a non-profit founded in 1999 as a collaborative effort between representatives of many healthcare entities. Its description of itself from its website reads: NQF’smore than 375 members represent virtually every sector of the healthcare system: consumers, purchasers, health plans, providers, healthcare professionals, suppliers, community health organizations, and quality alliances.
The NQF was founded in part from a desire of payers – governmental and industry – to be convinced that they were receiving high quality healthcare for their dollars. Also from its website, NQF’s current vision is:
- To be the convener of key public and private sector leaders to establish national priorities and goals to achieve healthcare that is safe, effective, patient-centered, timely, efficient, and equitable;
- That NQF-endorsed standards will be the primary standards used to measure and report on the quality and efficiency of healthcare in the United States; and
- To be a major driving force for and facilitator of continuous quality improvement of American healthcare quality.
Also woven into NQF statements is a belief that EMRs will be the tools that will seamlessly and effortlessly deliver accurate information to the payers on these quality measures.
I have nothing against measuring quality in healthcare. In fact I believe there is fair evidence that measuring and feedback has made some care processes better, both at local levels and nationally. But I do not believe stupid measurements help anything either. Let me now acknowledge that the following example of pharyngitis means not much to the overall cost of healthcare in the U.S. However, it is a simple example that illustrates a larger point.
One of the 743 quality measures of the NQF reads as follows:
Percentage of patients who were diagnosed with pharyngitis, prescribed an antibiotic, and who received a group A streptococcus test for the episode.
This measure says every single patient prescribed an antibiotic must be tested to meet their standard. However, this is not the most cost-effective way to manage sore throats.
Clinical decision rules for diagnosing Strep pharyngitis have been developed to guide testing and treatment approaches. The most commonly used criteria involve a scoring system based on the following factors: fever, absence of cough, tender lymph nodes in one area of the neck, swollen tonsils, tonsils covered in a substance that looks kind of like old mayonnaise, and age. These rules take into account the reality that no test is perfect, especially as can be obtained in real clinical practice. Some clinics have a hard time collecting a good throat swab specimen. They’re unwilling to be aggressive enough. They don’t like making their patient gag, which is pretty much what has to happen. If the strep throat prediction score is high enough, the guidelines say it’s reasonable to treat without testing, thus saving a lot of $15-ish rapid strep tests. (And this example assumes American assumptions on the necessity of antibiotics in the first place, whichseveral European countries don’t agree with.)
All of these factors should be entered somewhere in the medical record, so why couldn’t a computer program sort this out? Electronic medical records (EMRs) can’t do this. They can’t extract clinical information to apply to a scoring system for use in a quality measure. The EMR can report accounts receivable for each insurance company a practice accepts. It can run a report of diagnoses a practice codes. But it can’t tell me how many patients with a sore throat had swollen tonsils. The NQF quality measure was designed around the limits of the EMR.
Now fast forward to a possible day when I am paid a bonus for providing “high quality care.” I see a child with a relatively high likelihood of having a sore throat caused by a Strep bacteria. I think the best and most cost-effective treatment approach is to just give him a prescription for penicillin (or a shot in the office), but I have a clear incentive to order an unnecessary test.
What would you do if you were in my shoes?
Best I can tell, the NQF has become a facilitator of quality measures in part because it has convinced the payers it can define for them whether or not they’re receiving quality care for their high-cost health plans. Many of their measures are reasonable, but just as in this example, high quality cost-effective healthcare is often more complicated than it first appears. When asking the question, “Should a patient with X condition receive Y test or Z treatment?,” the best answer is most often, “It depends.” EMRs usually can’t answer the “it depends” question.
Beyond sore throats, more costly issues include the fact that diagnosing heart failure with certainty in many cases is difficult. Another measure counts the delivery of Pap smears. I see nothing in their description that says they take into account if a women had a hysterectomy including the cervix, which makes Pap smears pointless in most cases.
Maybe one day quality measures and EMRs will understand this complexity. In the meantime, those of us on the front lines are forced to choose between the most cost-effective delivery of healthcare or the delivery that will most please the bureaucrats. This only serves to undermine the delivery of complex family medicine and potentially forces us to turn off our comfort with complexity to simply check a box on a form. Family medicine deserves better.