|
Linking Guidelines and Process Improvement
Several of the articles I published in the last year have dealt with the notion that clinical guidelines and process improvement measurements often fail to produce any outcome improvement for patients despite vigorous, and expensive efforts. One problem is that guideline developers often don’t work with the process improvement people. In a recent article, Piggott and associates, from the Guidelines International—McMaster University consortium, have addressed several aspects of the challenge.[1] “Practice guideline and quality assurance and improvement (QAI) schemes are two sides of the same coin for improving health outcomes. One frequently cited set of characteristics specifies that a good quality indicator should be relevant, understandable, measurable, behaviorable, and achievable (the RUMBA rule.) However, we suggest that the characteristics of candidate quality indicators should include the following: 1) certainty; also known as scientific soundness (high-certainty evidence supporting selection is at low risk of bias, precise, direct, relevant, consistent, and without publication bias); 2) responsiveness or sensitivity to change; and 3) feasibility to measure, monitor, and implement.” The authors then outline a draft framework including 19 areas of concern with 40 potential steps. If this sounds both complicated and an enormous amount of work, it is. It also exposes how little we know about what measurements matter either to providers or to patients. The accompanying editorial postulates opportunities related to more flexible digital clinical data leading to better linkages.[2] “By leveraging new digital health data, jointly created guidelines and quality metrics designed to fit within quality improvement initiatives have a better chance of improving patient outcomes.” A less sanguine perspective was articulated by Drs. Schwartz and Werner in an editorial accompanying an article on “residual confounding.”[3] “Rewarding performance will improve performance. This principle has driven a wide range of reforms aimed at fostering accountability for outcomes…A key to accurately measuring the performance of providers or insurers is accounting for differences in patient risk. Risk adjustment is the standard statistical tool for leveling the playing field…Does risk adjustment produce accurate measures of performance?” The short answer to the question is no, as currently done, risk adjustment does not improve accuracy. So, what can be done? “Consider three strategies. First, we could sharpen our risk-adjustment tools with more sophisticated statistics and better measures of patient characteristics. This approach has yielded only small improvements so far is unlikely to get use where we need to be. Second, we could study whether error from risk adjustment is predictable, like the time on a clock set 5 minutes late…However, this strategy cannot fix unpredictable errors, which are likely to be common…Third, we could step back and carefully consider what policy problems we are trying to solve.” They look at the issue of health plan payment, noting one policy goal is to adequately compensate the plan so they don’t take steps to penalize high-cost patients. But, they note, other goals could be incorporated like expanding access, narrowing disparities, or ensuring a stable insurance market. Risk adjustment can help, but cannot solve the problem by itself. Taking a macro view, Skinner and associates examine the value of stabilizing health care’s share of the gross domestic product, (GDP.)[4] They note that health care’s share of GDP has risen an average of 2.2% every decade since 1960 in the US, compared to 1.1% per decade for other “first-world” countries. They discuss why this is of concern, particularly now with high inflation affecting both the US and the worldwide economy. They also note the benefits to be obtained by stabilizing the rate of growth to keep health care’s share of GDP stable at 18%, where it is now. But why are we different from our comparator group? “A key reason for high rates of growth in this sector is the lack of institutional budgets to constrain spending. Medicare is an entitlement program without legislative limits on expenditures…Governments and private insurers, for example could provide capitated payments to hospitals and physician practices to help them paya their fixed costs of operation, with annual increases tethered to GDP growth.” They have several caveats, and note most other countries have health care at 12% of GNP, but believe growth stabilization should be a national priority. Of course, they don’t have space to account for the differences in general social spending policy, where other countries fund needs from other parts of the budget, rather than expecting health care to pay for it, as we have discussed before. And the politics of limiting an entitlement program are daunting. So, where are we now? I fear working on the difficult issues of improving health care delivery and linking guidelines and process improvement won’t stop us from ramming the iceberg created by current economic realities of high inflation, low un employment, wage growth, lack of qualified applicants for positions requiring credentials, and pent-up patient demand that was deferred by the pandemic. While it is obvious that hitting the iceberg is bad, and will cause avoidable suffering and death compared to a thoughtful planned approach to reducing utilization, experience suggests we won’t act until it is clear we have no other choice. 17 April 2022 [1] Piggott T, Langendam MW, Parnelli E., et. al. Integrating Quality Assurance and Quality Improvement with Guidelines: Systematic Stakeholder-Driven Development of an Extension of the Guidelines International—McMaster Guideline Development Checklist. Annals.org, 1 March 2022. doi:10.7326/M21-3977. [2] Burstin H, Schneider E. Building Connections Between Guidelines and Quality Improvement. Annals.org, 1 March 2022. doi: 10.7326/M22-0409. [3] Schwartz, A., Werner RM. The Imperfect Science of Evaluating Performance: How Bad and Who Cares? Ann Intern Med 2022;175(3):448-449. doi:10.7326/M21-4665. [4] Skinner J., Cahan E., Fuchs VR. Stabilizing Health Care’s Share of the GDP. N Engl J Med 2022;386(8):708-711. doi:10.1056/NEJMp2114227. |
Further Reading
CQI - The Good, the Bad, and the Ugly Are the uses and misuses of the continuous quality improvement method emblematic of our times? More Encouraging News Are we on the verge of Clinical Guidelines 2.0? More on Variation - Part 1 Variation is not peculiar to healthcare, but is a general issue with the way the people think, and occurs whenever judgment is needed and the data are fuzzy. More on Variation, Part 2 Malpractice and restriction of hospital privileges represent extreme variation, but remain thorny problems. Social Determinants of Health Social determinants of health trump many medical therapies, but can changes be made that result in better health? Value-Based Dialysis Care A recent flurry of reports describe real-world results of the paying for value trend. What conclusions can be drawn? |