Evaluating Quality Care
This week we are taking a brief respite from the economics of surgical care to think about quality of care. The underlying point is, access without quality is not access, and the economics becomes a bit moot.
We first briefly discuss the measurement of quality. The question of “What is quality care?” followed by “How do we measure that?” could be considered a bit of a conundrum. The first paper reasonably defines the former in its elucidation of the latter. In Surgical quality indicators in low-resource settings: A new evidence-based tool, Surgery 164 (2018) 946–952, by Citron, Saluja, Amundson, et al, the authors point out that while there are a number of efforts to capture quality in high income settings, these are not generally adaptable to the resource- limited setting and can be quite resource-intensive to collect, often (usually) involving expensive software as well as equipment. To evaluate quality, they used two well-known frameworks: the structure/process/outcome (Donabedian) approach, and the Institute of Medicine’s mandate of care being (1) safe, (2) effective, (3) patient-centered, (4) timely, (5) efficient, and (6) equitable. By overlaying these, the authors created a matrix from which they derived fifteen quality measures deemed appropriate for the low-resource settings.
It was possible to categorize these into ones that measured within the domains of the Donabedian and IOM frameworks. Table 2 from the paper is shown below, which provides each of the measures and how those measures fit into the overall quality categories.
For these measures, tools for data collection were then developed in order for them to be of practical value. These include (1) a modified operating room logbook; (2) a modified intra-operative checklist; (3) a patient questionnaire; and (4) a modified patient admission and discharge book. The simplicity and the current existence in some form of three of the four of these is perhaps the aspect of this particular paper that makes it of most value. The authors conclude with the statement that these measures are expected to be modified and evolve in order to refine further what constitutes quality care and how it may be improved in the low-resource setting.
Next we very briefly mention Watters et al, Perioperative Mortality Rate (POMR): A Global Indicator of Access to Safe Surgery and Anesthesia, World J Surg (2015) 39:856–864. This paper called for the acceptance of the POMR as a standard measure of surgical and anesthetic safety and quality. POMR has now gained that acceptance. It is originally described as death before discharge from the hospital or within 30 days of a surgical procedure, whichever is sooner. While it is understood that it may often be difficult to obtain information about death after discharge in many LMIC settings, it cannot be disputed that including these deaths creates a more robust assessment of quality of care. The authors note this, and suggest using the term POMR 30 for data the includes post-discharge deaths, and note that the death-prior-to-discharge number is used as a proxy for the more robust information. Also of critical importance in interpreting the POMR is risk stratification, this paper suggesting that that include urgency (emergency vs. elective), age, the condition being treated or the procedure being performed, and American Society of Anesthesiologists (ASA) status.
In the interest of brevity (aka, keeping this blog short enough to entice the reader to continue reading it), and with the confession that we cannot resist bringing in economics even when we said we were taking a break from that, we conclude with the recent paper by Alkire, Peters, Shrime, and Meara titled The Economic Consequences Of Mortality Amenable To High-Quality Health Care In Low- And Middle-Income Countries, Health Affairs 37, no. 6 (2018): 988–996; doi:10.1377/hlthaff.2017.1233. This paper contains a very nice summary of the entire concept of health as a generator of wealth, and if the reader is unfamiliar, s/he is encouraged to read the Introductory section of this paper simply for the background.
The authors had first to determine an estimate of the number of deaths due to amenable causes by country, then convert that into economic terms. The first of these was done by estimating the preventable and unavoidable deaths using global burden of disease data from 2015 and subtracting this from the total number of deaths. The second of these was done using two methods that have different interpretations of economic loss. As such, they measure different things and cover different time periods.
The Value of Lost Output approach uses a WHO model that projects the effect of mortality (disease-specific) on the labor force and physical capital; i.e., the two main aspects of economic production in a country. The authors investigated what this relationship was in the decade prior to the study and projected this onto the years 2015-2030. They then modeled the same relationship using the counterfactual that there was no mortality due to amenable causes. By subtracting the latter from the former they could estimate the difference in labor and physical capital, and therefore the economic loss in terms of GDP, over this time period due to this mortality.
While the Value of Lost Output approach looks at the concrete aspects of the economy embodied in labor and capital, the Value of Lost Welfare approach is based on the Value of a Statistical Life concept. Thus, it attempts to capture the value placed by humans on life, including leisure time and general well-being. The authors used this method to estimate the economic loss in the year of the study (2015).
While the original premise of this paper is quality of care, the end result is a function of both quality of care (how many deaths amenable) and economic welfare of the country/region. I.e., an area with abysmal quality of care but with a very attenuated economy might still have a relatively small economic loss. Conversely, an area with very little mortality amenable to higher quality of care but high economic output might have higher loss using these measures. This point is made simply to point out that lower economic loss does not equate with diminished need for improved quality of care. The ultimate point is that access to care is a necessary, but far from sufficient, entity if over seven billion people are going to have appropriate care. High-quality care must also be an integral part of universal health coverage. This paper reiterates that quality and access are not only humanitarian imperatives, but economic ones as well. For policy makers moved by economic realities or preferences, quality health care for all is a good investment.