As many people know, it is once again election time in Israel, and the newspapers are filled with commentary about the various parties. In addition, there is an endless list of critical analyses, indicating how every party seems to have totally failed in its initially set goals. One would think that the last time the Israeli government passed an effective policy was decades ago, when it decided to include Elite chocolate in the rations for the soldiers.
In one of the Israeli newspapers, they recently had a table that listed the various promises that the many parties had made before the last election. Of course, the parties only had two years versus, the hoped for, four years to implement their programs. Nevertheless, there was a checkmark next to each promise, indicating the various levels of success.
In this summary table, there were three columns, indicating a status of “failed”, “partially succeeded”, and “succeeded”. I was actually impressed by how much information this single table provided. By reviewing the topics that it assessed, I was reminded of a whole slew of previous pre-election promises that had been made by the various parties. In this single table, one was able to see very easily and very quickly just how effective each party was. I consider this type of information, in this specific format, to be critically important. More so, it struck me that such an evaluation of the various parties should have been something that was being updated on a monthly basis.
Imagine that we had such a table, easily retrievable, going back to the establishment of the State of Israel. We could look at every single election and every single party and track the ratio between pre-election promises and successful implementation. In general, people’s memories tend to be limited to the last few years, if that much. More so, people tend to focus on present day economics and security when choosing their candidates. But if everyone spent the time to look at these longitudinal charts, following them just as the nation follows the level of the Kinneret, I believe that there would be a far improved level of discussion about the quality of government.
There is absolutely no reason that a similar type of table could not be produced for each hospital, each department in each hospital and even every single staff member. There is no question that there would need to be a special new position created in order to oversee the generation of all of these data tables. And of course, there would have to be some standardized method of measuring quality and success, in order to fill in each cell of the overall quality table.
The information that is used to fill in each column and row would likely be a combination of patients’ reviews, staff reviews, colleague reviews and other more objective measures [such as total patients examined, rates of complications, infection rates and so on]. No system of measurement that is not entirely quantitative is ever going to be accurate. And even when using a quantitative scale, it will only be accurate if the designation of each score is by some objective system [like a computer, versus a human being].
Setting up such a system and then implementing it across the world would be an enormous undertaking. On the other hand, considering how much money is spent on healthcare, and how many programs are being pushed forward to improve healthcare and increase efficiency, this type of quality measurement is not by any means a luxury. Whatever funds it would take and however long it would be until such a system was standard, would be worth it.
With such a system in place, patients could finally truly compare overall quality of care as well as the quality of individual healthcare providers amongst different hospitals, even in different countries. Having said all of this, I appreciate that there will be countless physicians who will argue that there is no way to create a standardized way to compare physician performance. And I would agree that this was true until the last few years. Today however, as more and more information is recorded digitally, it is possible to have advanced computer systems do the most intricate data analysis.
For example, imagine that a patient of advanced age presents to an emergency room with a hip fracture. Today, it is possible to quantify every step of the assessment. One can measure
- the time at initial intake
- total time in triage
- time from nurse assessment to first physician assessment
- time from physician assessment to the performance of laboratory and radiology studies
- time between radiology studies
- time until interpretation of various studies both by the on-site physician and the senior remote specialists
- time until surgery
- total time of surgery
- all checklists recorded pre-, during and post the operation
- time of recovery
- status of the patient every day after the surgery
- presence of infection
- time until discharge
- status of patient on first post operative visit
- status of patient at one month, three month and one year postop
This is not an exhaustive list. Also, you must include all of the “regular” medical information about the patient, such as patient gender, medications that the patient takes, family history of illness, known genetic propensities for various diseases, and the like.
There is more. You must include any and all data that is collected by various sensors throughout the pre-op period until one year after the surgery. Patient pulse rate, blood pressure, oxygen levels in the blood, levels of hydration and any and all other physiological parameters that are recorded by wearable sensors would be stored in the patient’s electronic medical record and become part of the analysis of success (or failure) of the hip fracture treatment operation.
I have purposely detailed all of these data points to be clear about just some of the parameters that can be assessed and compared between patients, to compare medical care efficiency and success. Imagine now that ALL of this information and collected data gets multiplied by the hundreds of thousands of patients undergoing hip fracture surgery every year. It is simply impossible for any group of human beings to fully process all of the data and thus assess the quality of the overall care.
Fortunately, we now have computer systems that can analyze this kind and quantity of data. More so, learning systems will soon come to understand what is considered a reasonable amount of time or quality output for each stage of the care. Such systems will be able to compare patients and match them on the basis of all of their initial parameters. Therefore, it will be possible to say that of the 500 patients with hip fractures cared for at a particular hospital with a particular set of initial parameters, their outcomes are either better or worse in comparison to another institution with a similar group of 500 patients. And for the first time in the history of medicine, this information, which will be made public, will allow patients to truly compare which hospitals do better. On the basis of this information, individuals, companies and insurance providers will all decide where they wish to have their health care provided.
Whether we are talking about an election or about medical care, ultimately it is all about the data. Collecting the data is a technical chore that can be rendered far easier by the use of better and better sensor technology, as well as better EMRs with more clinically oriented interfaces. We are truly on the verge of a transformation in the measurement of quality of care. I for one very much look forward for such advancements.
Thanks for listening