Productivity in HealthCare - Part 2
In a previous article, I suggested there was tension between productivity and efficiency in healthcare, even though the words are often used interchangeably. Productivity is defined as the shortest, quickest way to provide reimbursable care, while efficiency is defined as the shortest, quickest, most appropriate way to care for an individual patient. In that article I also suggested the tension is seen in three perspectives on health care: artisanal, economic, and scientific. I entered the practice of medicine just as medical technology was blossoming. ICU care, ventilators, dialysis machines, and heart monitors, automated laboratory testing, and then computerized imaging studies have clearly changed medical care in ways not always understood at the time, and perhaps still not well understood. What was standard when I started medical school has come to be disparaged.[1] Our current health care system is essentially a cottage industry of nonintegrated, dedicated artisans who eschew standardization. Services are often highly variable, performance is largely unmeasured, care is customized to individual patients, and standardized processes are regarded skeptically. Autonomy is hardwired into the system, because most physicians practice in small groups with limited oversight or coordination. This transformation was first dissected in detail by Paul Starr in 1983.[2] When I read this book, I could see that we were becoming “industrialized,” and that the trend was probably inevitable. I responded by thinking about deliberately shifting the way I organized dialysis care from a craft focused on individual patients to an industrial process designed to standardize the care delivered as much as possible. Using tools which are now routine such as continuous process control charts, we had success in ways I have documented elsewhere. At the time, many of my academic colleagues criticized the approach as “unscientific,” meaning they were not randomized, controlled trials. While I have no regrets about what I did then, I have come to appreciate that the law of unintended consequences is alive and well, and I now find myself trying to adapt the system to deal with them. What happened? When the government began funding dialysis treatments through Medicare in 1973, the number of available dialysis units increased rapidly. This was generally thought to be a good and necessary things. Some units were organized by hospitals or doctors, but some were organized as publicly traded corporations. By the time Starr’s book was published, the government had decided it was overpaying for services and made the first of a series of significant changes in the rules. About every 10 years or so since, the government has “recalibrated” its payment scheme, using the creation of more units as proof that the “market” is still attractive. However, the market has contracted substantially. In the 2013 United States Renal Data System (USRDS) Annual Report[3] showed in 2011 two providers, Fresenius and DaVita, had 60% of the units and 63% of the patients were dialyzing in those units.[4] As pay for performance has become the norm, the government has taken published guidelines and made them thresholds for payment. Naturally, the publicly traded corporations have a financial stake in making sure their units exceed these thresholds, even if it does not make sense from the patient’s point of view. But isn’t that what guidelines are supposed to do—improve care for everyone? Guidelines, as most physicians understand, represent the “state of the science” reviews regarding treatment of a specific illness. Recommendations are graded on the basis of the strength of the underlying evidence, using a now standardized rating process. Many outside the process appreciate how few recommendations are based on grade A or grade B evidence. (Grade C is generally a consensus view of the writers.) Furthermore, few guidelines contain performance standards, yet when the government and other payers get involved, suddenly performance standards become mandatory. This mandatory financial aspect brings us back to the law of unintended consequences. When I began using statistical process control methods, I understood what I was trying to do was to improve reliability by differentiating between common cause and special cause variation. When we made a change to our process, we looked for evidence of special cause variation in the desired direction. What we usually found was little evidence for anything other than common cause variation. Yet if near median values suddenly become the threshold for payment, the penalty for common cause variation becomes substantial. Suddenly, the financial systems overwhelm the scientific systems, and the patient-based artisanal view is extinguished. Consider the problems of vascular access in hemodialysis patients. The medical literature is clear that a well-functioning arteriovenous fistula[5] lasts longer and is more trouble free than any other option. It is the preferred option for virtually all patients.[6] When the government began its “Fistula First” initiative, the prevalence of fistulas as a primary access varied enormously, but were generally on the order of 30%. The initiative aimed to get the prevalence to 60% and now to 66%. Two things were done well to facilitate this goal. First, the allowable charge for a fistula was increased to make payment neutral as to which kind of access (fistula or graft) was placed. Second, payment was made for studying the anatomy of the arteries and veins, (venous mapping), prior to surgery, which led to increased success rates. The medical literature also shows that catheter vascular access is associated with dramatic increases in morbidity and mortality, and is the most undesirable type of access. Like the statement about a fistula, there is really no controversy about this. Clearly, then, the policy goal was to increase fistulas and decrease catheters, and is was generally assumed this is what Fistula First would do. Analysis of USRDS data, though, suggest what really happened was replacement of graft access with fistulas, with surprisingly little reduction in long-term catheter usage. Clearly, Fistula First has had some success, but has not had the outcome it was intended to have. Did it work in terms of reducing cost? At the macro level, the answer is clearly “No.” One unexpected consequence is the impact on the death rate, which has flattened since the Fistula First initiative began. While things other than vascular access may play a role, the timing is suggestive. Since the death rate has flattened, and the acceptance rate is higher, the effect has been a marked increase in the number of treated ESRD patients.[7] At a technical level, anecdotal evidence suggests the prevalence of “good” fistulas is still on the order of 30%, while the other 30% have secondary fistulas that require frequent interventions and revisions much like graft accesses. There may be some gain in reduced infections, but this has been harder to quantify. The problem, then, is that physician/surgeon attitudes and practices did respond to economic pressure, but patient selection and anatomy did not.[8] To some extent, we have been gaming the system and may not have improved outcomes for patients as much as we had hoped. The problem recently became even more difficult as CMS has started applying penalties for dialysis units who do not have at least a 50% prevalence of fistulas. So what are units doing? I am aware of some units that will not admit patients who do not have a fistula already constructed. This will certainly keep them off the hit list, but the economic ripple effects of this are fairly obvious. This is only one of many examples where medical evidence has been adapted to financial ends without recognizing all of the unintended consequences. Lost in all of this is the patient.[9] All of us have patients who have no option other than a catheter, and all of us have patients with limited life-expectancy, for whom a permanent access may not be worth the pain and suffering. Yet this artisanal approach has to fly in the face of the scientific evidence that a fistula is superior and a catheter inferior, and the financial imperative to avoid penalties for failure to achieve thresholds. For physicians and nurses who must see the suffering regularly, extinguishing the patient’s perspective is not possible, and having to push for corporate (financial and scientific) goals in the face of this creates tension. Is it any wonder that physician and nursing morale is poor? And if the morale of those primarily responsible for the delivery of care is poor, who do we expect to achieve superior results? Although any suggested answer has to be tentative, and thought of as a hypothesis, I will examine these questions in a subsequent article. 30 October 2014 [1] Swenson SJ, Meyer GS, Nelson EC, et. al. Cottage Industry to Postindustrial Care—the Revolution in Health Care Delivery. N Engl J Med 2010;362:e12. doi: 10.1056/NEJMp0911199. [2] Starr, Paul. The Transformation of American Medicine: The Rise of a Sovereign Profession and the Making of A Vast Industry. (New York: Basic Books, 1983.) This book won the 1984 Pulitzer Prize for General Nonfiction and the Bancroft Prize in American History, among other awards. [3] U.S. Renal Data System, USRDS 2013 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 2013, Vol. 2, Chapter 10: Providers. Accessed 10 October 2014 at http://www.usrds.org/2013/pdf/v2_ch10_13.pdf. [4] Dialysis Clinic, Inc., for whom I provide medical director services, had 215 units, or 3.2% of the total, but it was the third largest provider. [5] Brescia MJ, Cimino JE, Appel K, Hurwich BJ. Chronic Hemodialysis Using Venipuncture and a Surgically Created Arteriovenous Fistula. N Engl J Med 1966;275:1089-1092. [6] Winkelmayer WC. Tackling the Achilles’ Heel of Dialysis. N Engl J Med 2011;364:372-374. [7] In current dollars, the annual cost is estimated at $90,000 per patient per year. The number of treated dialysis patients has almost doubled since 2002. While this may be good for the patient, it has certainly not saved money. [8] DeSilva RN, Patibandla BK, Vin Y, Narra A, Chawla V, Brown RS, Goldfarb-Rumyantzev AS. Fistula First is Not Always the Best Strategy for the Elderly. J Am Soc Nephrol 2013;24:1297-1304. doi: 10.1681/ASN.2012060632. [9] O’Hare AM. Vascular Access for Hemodialysis in Older Adults: A “Patient-First” Approach. J Am Soc Nephrol 2013;24:1187-1190. doi 10.1681/ASH.20130550507. |
Further Reading
Clinical Microsystems Clinical microsystems are composed of front-line clinicians engaged in direct patient care. Despite a lack of formal authority, they are the key to successful healthcare reform. Nursing Staff Turnover If empowered teams of clinicians is the key to effective, efficient care, then staff turnover is Achilles' heel. Nationally, RN turnover exceeds the cap needed to maintain patient safety and quality of care. The problem and approaches to a solution are considered. Physician Engagement Is physician engagement a strategy to promote physician leadership, or a code word meaning how do we get the doctors to do what we want? Strategic Human Capital Healthcare organizations need to realize the economic value of experienced teams of clinicians able to provide highly reliable care and to recognize the importance of maintaining team integrity in times of surges in patient volumes. |