|
Medical Waste
No, this is not going to be about the trash, both contaminated and non-contaminated, although there is certainly a great deal of it generated in routine practice. I want to consider wasteful practices. Of course, everyone is against waste, but there continues to be controversy about how to define it. This is not a case of “one man’s waste is another’s treasure,” it is a problem of specifying waste in a meaningful way. An article done in the VA Health System looked at variation in kidney disease care in patients not yet on dialysis.[1] They looked at 281,223 patients with an eGFR of 15 ml/min to assess measurement of urine protein to creatinine ratios, use of an ACE or ARB, blood pressure at goal and hemoglobin level. They found only 13% of patients met the goal on all metrics, with measurement of proteinuria being the worst, at only 37%. There was “considerable” variation from facility to facility which could not be explained by the available data. Another article also from the VA system looked at whether there was a difference in rates for process indicators depending on the educational and licensure status of the healthcare provider.[2] They found 45% of patients achieved a hemoglobin A1c less than 7, 42% had a systolic blood pressure less than 130, and 72% had an LDL of less than 100 mg/dL. The licensure status of the provider did not associate with the results in a meaningful way. If you believe the guidelines are appropriate for the vast majority, then the meager results suggest there is a lot of waste or ineffective care. On the other hand, there is room to argue about the results. For instance, in the first study, patients were selected for having known CKD. I would argue routine proteinuria screening in patients with established CKD is “waste.” The intention of the guideline was to make sure early cases were detected in hopes progression could be prevented. The other guidelines are all based on the hope that disease progression (CKD or diabetes in these two studies) can be retarded by application of consistent process measures. One of the challenges, though, is proof of concept is hard to come by. The Joslin Clinic experience is often cited as proof that good control of diabetes works. It has been 40 years since I read their book, and don’t have a copy to hand, but as I recall what they observed is the rate of complications, particularly CKD, was lower in patient with well controlled diabetes, but complications occurred in patients who were well controlled, just as some complications, like CKD did not occur in patients who were not controlled. The problem then, is one of biological variability in both disease process and susceptibility to complications in ways we don’t yet understand or know how to identify. Woodell and Rifkin raised this issue in an editorial entitled “Still Asking ‘Which Rate is Right?’ Years Later.”[3] They noted that Wennberg, in his initial studies of geographic variation raised this problem. “He posited that such variability resulted not just from correct and incorrect medical decisions, but also from legitimate differences in opinion about tests’ and treatments’ safety and efficacy. We have made great strides in attempting to standardize what is means to be safe or efficacious, but the distinction between appropriate and inappropriate remains elusive.” Yet another recent study, this time from a private insurance database raises another issue. Does funding matter? The authors looked at patients with diabetes covered by a single insurer who switched from low-deductible to high-deductible health plans and compared them to those who did not switch. They found “switchers” had longer times before a claim for a first visit for macrovascular symptoms, tests, and procedures.[4] In an accompanying editorial, Pauly asked the question, did the delay matter?[5] He notes the answer can’t be found in the data, because there is no insight into why people switched plans or stayed put, but concludes there is reason to be cautious about high-deductible plans, because it might. So, what should we conclude from these studies? First, I think we must recognize clinical guidelines and clinical performance measurements lack enough evidence to be taken as a pass/fail mechanism for deciding care is either appropriate or not. We also need to recognize that there are unintended consequences to our current pass/fail approach. Unfortunately, the payers need some mechanism to pay less, and CPM offers a way to do this while avoiding decision-making on their part. Second, we need to recognize that even “big data,” of which these studies are examples, don’t necessarily provide the sort of guidance needed for sound decision-making. I am sometimes amused and sometimes exasperated by the fans of big data and “artificial intelligence” who seem to think all the data points can be objectively assembled into an array so outputs free from nuance and variability can be created. There is little enough “human intelligence” being applied to solve our challenges, (as opposed to finding ways to spend or save money.) Switching to a machine programmed by a human far removed from clinical realities is not likely to be an improvement in other than limited circumstances. But that is a topic for another day. Happy New Year. 1 January 2019 [1]Navaneethan SD, Akeryod JM, Ramsey D, Ahmed ST, Mishra SR, Petersen LA, et. al. Facility-Level Variations in Kidney Disease Care Among Veterans with Diabetes and CKD. Clin J Am Soc Nephrol 2018;13(12):1842-1850. doi: 10.2215/CJN.03830318. [2] Jackson GL, Smith VA, Edelman D, Woolson SL, Hendrixx CC, Everett, CM, et. al. Intermediate Diabetes Outcomes in Patients Managed by Physicians, Nurse Practitioners, or Physician Assistants. A Cohort Study. Ann Intern Med 2018;169(12):825-835. doi: 10.7326/M-17-1987. [3] Woodell TB, Rifkin DE. Still Asking “Which Rate is Right?” Years Later. Clin J Am Soc Nephrol 2018;13(12):1783-1784. doi: 10.2215/CJN.12371018. [4] Wharam JF, Lu CY, Zhang F, Callahan M, Xu X, Wallace J, et. al. High-deductible Insurance and Delay in Care for the Macrovascular Complications of Diabetes. Ann Intern Med 2018;169(12):169:845-854. doi: 10.7326/M17=3365. [5] Pauly MV. Switching to High-Deductible Health Plans: It is Going to Be a Bumpy Ride. Ann Intern Med 2018;169(12):879-880. doi: 10.7326/M18-2825. |
Further Reading
Another Look at the Value Proposition A review of published data show pay for performance programs have not impacted either cost of care or health outcomes. Measuring Teamwork Measuring Teamwork is difficult, but important if healthcare systems are to invest in their development. This article reviews the literature and provides suggestions for action now. Quality Metrics Uncertainty Dealing with uncertainty is at the core of practicing medicine. Have we tried to escape this reality? Variation in Health Care Is variation in health care good, bad, or inevitable? The answer may determine future medical practice. |