Tulva AI for outpatient care
Abstract
Do you struggle to satisfy the demand for care with the medical personnel at your disposal? Do you feel that it is increasingly difficult to recruit sufficient number of trained medical professionals? You are not alone. Due to aging populations, the world is facing a shortage of medical personnel that is unprecedented in developed countries. In this article we tackle the problem of resourcing in private outpatient clinics and tell how Tulva AI can help in making the existing workforce to suffice.
Increasing demand, shrinking supply
Currently, there is a massive global deficit of medical personnel. In 2019, a study published in Lancet estimated a global shortage of 6.5 million physicians, 30.6 million nurses and midwives, 3.3 million dentistry personnel and 2.9 million pharmaceutical personnel [1]. At the same time, the population in developed countries is steadily aging with no end in sight: for example the population of Europe has been projected to age up until year 2100 [2]. Due to combination of low supply and increasing demand for care, WHO has projected that there will be a global shortfall of 10 million health workers by 2030 [3] and Europe has been suggested to be is facing "a crippling shortage of nearly 1.8 million healthcare workers" [4].
The unit cost of a medical professionals appointment is by default very high and the year by year increasing gap between supply and demand will drive prices even higher. This development obviously risks steadily increasing health disparity, where health care is mostly accessible to those who can afford it. For these reasons, no country can afford wasting these appointments due to suboptimal resource allocation.
Tulva AI for rescue
This is where Tulva AI comes in. With Tulva AI, a health care service provider can make sure that the limited resources meet the demand. Let us explain what this means in private outpatient care.
Imagine a moderate sized private healthcare clinic which has six doctor’s offices and and a pool of independent practitioners (see Figure 1). The goal is to allocate the practitioner’s time as efficiently as possible meaning that 1) all the appointments will be utilized and 2) every patient get’s the appointment they require. Let’s also imagine for the sake of the argument that this kind of allocation would happen on Sunday for the next 7 days, although we know that planning is often done on monthly resolution at best. The most common way to allocate the physicians time is simple heuristics and personal experience.
Figure 1. Conceptual model of current state of weekly resource allocation. On Sunday, the number of daily practitioners (yellow squares) for every day of the following week is determined based on short historical window and educated guesswork.
When the week goes by and we look at what actually happened (see Figure 2), we see that on Friday and Sunday, there was an excess of two practitioners and on Saturday an excess of one practitioner (red squares). On the other hand, on Monday and Tuesday there was a deficit of one practitioner and on Tuesday two practitioners (light green squares). Moreover, there were almost daily attempts to correct for these mismatches, either by calling in additional practitioners or by practitioners independently closing up shop earlier than initially intended.
Figure 2. Results of the heuristic resourcing. Five practitioners’s daily contribution is wasted (red squares) whereas there is a deficit of four practitioners (light green squares). Reactive readjustments are attempted almost every day (red triangles).
This is of course wasteful and inefficient. If the clinic was using Tulva AI, they could have foreseen these mismatches and encouraged the four physicians to move their appointments from periods of low demand to periods of high demand as shown in Figure 3. Moreover, one physician could have gone to work someplace else without negatively affecting the economics of the polyclinic and contribute somewhere else where demand exists.
Figure 3. Intelligent resourcing with Tulva AI (T). Accurate predictions of the service demand for the whole week would enable proactive resourcing once per week or whenever needed.
Come see the future with us
With Tulva AI it is this simple: no over-resourcing, no under-resourcing, just great care. Contact us to make this happen at your polyclinic!
References