Whether healthcare is viewed as a right or a privilege, one thing is clear: as science and technology continue to advance, healthcare is on the verge of becoming untenably expensive, and yet the science of improving the value of healthcare has lagged behind. The good news is that leading health systems and innovative organisations have begun to prove that the difficult is not impossible. Achieving a learning health system that delivers continuous improvement in the outcomes and value that matter to patients and payers alike is a serious and important challenge. I’d like to share our experience in tackling head on two of the biggest barriers that must be overcome: credible, actionable, measurement data; and engaging physicians in redesigning care.
It begins with good data
As consumers, we judge the value of a product according to a very simple equation, regardless of industry:
Value of product = quality of the product / cost of the product
In healthcare, that equation becomes:
Value of healthcare = quality of healthcare / cost of healthcare[1]
Measuring the quality of healthcare has, so far, been the measurement of secondhand proxies for quality, not quality itself. We measure processes of care and infer quality from those processes, but few would argue otherwise that those inferences are quite often wrong. Instead, most patients and physicians would argue that the quality of care should be measured more directly. Quality is defined by the combination of the patient’s perception of quality combined with the functional health outcome of that patient following their purchase and receipt of care. However, we do not collect functional outcomes of care. Two years ago, my orthopaedic surgeon repaired my torn ACL, MCL, and fractured tibial plateau. The surgery took about three hours to complete. I visited him once, post-operatively, per his protocol, and have not seen him again since. He has no clue as to whether my outcome has been better, worse, or average for a 55-year old man.
It was up to me to find, screen, and choose a physical therapist, and describe to her my injury and surgery, which I did. She was a PhD, US Army captain, who previously served the physical therapy and rehabilitation needs of 400 Army troops in Afghanistan for two years. When she accepted me as a patient, the first thing she asked was, “What is your definition of rehabilitation success? What activities do you want to be able to do and how soon?” I had no data to turn to for assistance in establishing realistic goals, so I took a naïve guess and told her that I wanted to ride a bicycle on a strenuous mountain road in six weeks, mountain bike in four months, ski in six months, and run sprints in nine months. With that input from me—aggressive as it was—she developed an equally aggressive and personalised protocol by working backwards from my patient-driven outcomes goals. Throughout my rehab, she constantly measured my muscle growth, flexibility, balance, strength, and agility. She entered those measurements into an algorithm that provided a composite indicator of my overall progress. With her coaching, we met or surpassed every goal.
That, my friends (thanks to my physical therapist) is protocol-driven, data-driven, outcomes-driven healthcare.
It is worth noting that my physical therapist entered my data into her version of an electronic health record (EHR), which had no interoperability with my surgeon’s EHR system. Her EHR was designed much differently The hard work of improving healthcare outcomes han the EHR used by my surgeon. The physical therapist’s EHR functioned more like a project management software application with the patient at the centre of the project. My surgeon’s EHR functioned like a typical billing encountercentric EHR.
In terms of complexity and outcomes-driven project management, my knee surgery and rehab were vastly simpler than a patient who suffers from cancer or comorbid chronic conditions.
It has been six years since the passage of the HITECH act in which the US federal government is funneling $US 48 billion into financial incentives for the implementation and meaningful use of EHRs and Health Information Exchanges (HIEs) by US healthcare providers. The promise is that these investments, which have proven as costly to hospitals and practices in time and energy as they are in dollars, are establishing a critical, essential data infrastructure to drive datadriven quality and efficiency improvements.
The sad truth is that agile, patient-tailored outcomes monitoring and data visibility have not progressed, despite the vast sums invested; the data problems are geometrically compounded in healthcare delivery systems that are utilising multiple, disparate EHRs.[2] Data warehouses can solve part of this problem by integrating and harmonising the disparate data for offline analysis,[3] but that analysis is not presented at the point of care, in real-time, tailored to meet the personal outcomes management of a specific patient. Fifth generation EHRs need to look and feel more like a project management system; more like that of my physical therapist’s EHR, and they need to be able to incorporate evidence-based outcomes measurement and tracking systems.
Data empowers physicians
While we commonly hear the phrase “first do no harm,” physicians aspire to do far better than that. As Daniel Pink describes,[4] physicians are motivated by the same needs as the rest of us: a desire for mastery, autonomy, and purpose. They want a chance to develop and master a skill in which they can be proud—practicing medicine and improving the health of their patients. They want to be surrounded in a supportive environment if they need help, but for the most part they want to be left alone to practice the skills that they’ve mastered, and not be micromanaged. And they want to feel as if they are serving a purpose in life which is larger and more important than themselves. Nearly all physicians are naturally driven to deliver the very best care to every patient, and to excel in their chosen field. Yet we put physicians in a position to pursue these aims largely in the blind, like a pilot without cockpit instruments flying in a thunderstorm. They are left to using their best judgment and consulting the best available evidence in the scientific literature to guide their decisions, but without any insight into the outcomes that their own patients are achieving or an evidence- and data-driven plan for optimising those outcomes.
When you give physicians the data they need in the cockpit at the point of care, and you place them in a position to practice their mastered skill without burdensome oversight, and, because they are better informed about the best options available for treating their patients, these physicians feel a greater sense of purpose, too. At Intermountain Healthcare, we provided physicians and nurses with this data-rich environment, including statistical predictions of outcomes based on treatment options.[5] Intermountain’s leadership culture was also very light-handed and communal, leaving it up to physicians to determine how best to reduce variability and improve outcomes. As a result, Intermountain achieves better healthcare outcomes at an average of 34% lower cost than the general US healthcare delivery system.[6] We need to learn from and propagate the Intermountain model for outcomes improvement.
Stepping up to the challenge of outcome measurement
This is an exciting time in healthcare. Empowering physicians and the health systems in which they work with the tools to pursue their patient’s desired outcomes is a challenge that can and must be overcome. Outcomes data on patient care is one of the single most valuable pieces of data missing in our healthcare ecosystem.[7] At the heart of this movement will be defining the outcomes that matter to patients, tailoring our outcomes project plan to the goals and abilities of the patient. Without outcomes data, healthcare is guessing at best practices, not actually practicing best practices.
References
1. Porter, M. What Is Value in Health Care? N Engl J Med 2010;363:2477-48.
2. Rahurkar, S., Vest, J., Menachami, N. Despite the Spread of Health Information Exchanges, There Is Little Evidence Of Its Impact on Cost, Use, and Quality of Care. Health Aff Mar 2015: 34;3477-483.
3. Sanders, D., Protti, D. Data Warehouses in Healthcare: Fundamental Principles. ElectronicHealthcare, 6(3) January 2008. http://www.longwoods.com/content/19510 (accessed 28 Mar 2015).
4. Pink, D. Drive: The Surprising Truth About What Motivates Us. New York, NY: Riverhead Books, 2009.
5. Evans R, Pestotnik S, Classen D, et al. A computer-assisted management program for antibiotics and other antiinfective agents. N Engl J Med 1998 Jan 22; 338(4):232-8.
6. Poulsen, G, Improving Quality, Lowering Costs: The Role of Health Care Delivery System Reform. Testimony to the Senate Health, Education, Labor and Pensions Committee, 10 Nov 2011.http://www.gpo.gov/fdsys/pkg/CHRG-112shrg87752/ html/CHRG112shrg87752.htm (accessed 28 Mar 2015)
7. Hostetter M, Klein S, Using Patient-Reported Outcomes to Improve Health Care Quality, Commonwealth Fund, Quality Matters, Dec 2011/Jan 2012 Issue. http://www.commonwealthfund.orgpublications/newsletters/qualitymatters/2011/december-january-2012/in-focus (accessed 28 Mar 2015)
No comments:
Post a Comment