I wrote an article about perceived waiting time a few months ago. It is my top 5 most viewed post in 2019. I must say I am amazed at the number of readers who are interested to read about waiting time. On top of that, there were so many helpful comments about that post on my Facebook page after it was published. It was humbling and it made me realise that there is a growing demand from my readers to have more articles related to the subject. So, I have decided to write this one right here, just for you. I hope it’ll open up your eyes to what happens behind waiting time data collection and measurement, further allowing you to leverage on the hidden improvement potentials from the basic act of waiting time data manipulation.
Waiting time is and will continue to be a huge concern for everyone in the healthcare industry. The fact that it is a constantly moving target which indirectly relates to patient satisfaction makes it one of the most intriguing topics, always to be discussed in high-level stakeholders meetings the world over. People (especially healthcare administrators) have been implementing the latest management tools on it ever since the birth of management theories, monitor it with complex systems, find rooms for improvement and repeat the whole cycle, all in the name of greater efficiency. Others conduct research on the subject and publish it in prestigious journals. It seems that no one can ever say enough about it and I am not spared.
However, what I want to present about waiting time in this post may be slightly controversial. It is based on what I have encountered while I was in charge of monitoring the Outpatient Clinics Waiting Time KPI in a local hospital years ago, coupled with similar experiences resonated by my fellow colleagues in other healthcare facilities around the nation. Please note that this is by no means an intent to tarnish the image of organisations that I have worked at or the ones I’m affiliated with. It is a sincere offer of improvement suggestions based on careful observations of the way the reported KPI data have been creatively manipulated. I believe that the way waiting time data has been manipulated could point towards some really interesting measures to improve actual waiting time.
Before I begin I would like to briefly clarify several things about the waiting time KPI measured in almost all outpatient clinics nationwide.
- Technically, it is measured from the time of registration/ appointment (whichever is later) to the time the patient is first seen by the doctor.
- The final data is presented in percentages of patients who waited less than 90 minutes. The standard set for it is no less than 90%.
- Exclusion is made for patients who request to see a specific doctor, who walked in without an appointment, who requires a procedure prior to seeing a doctor, who have multiple appointments on the same day or who are slotted in for special consultation.
Explicit KPI technical specifications have been written in order to standardise the start and endpoints, which also made clear the data collection method and the data analysis to be employed in the process. But, despite all the well-meaning attempts, the waiting time KPI is still open to various interpretations. This has made it particularly vulnerable to manipulations. Although some may argue that manipulating the data is wrong on all levels, I believe it is more productive to look at it from an improvement standpoint. Even though I agree that figuring out why people manipulate the data in the first place may be helpful (maybe I’ll write another post about this) I’m currently more interested on what waiting time data manipulation could teach us about how to make it better.
Mistakes are the portals of discovery.
Manipulation 1: Start from clinic registration instead of central registration.
In the effort to make waiting time duration appears shorter, many starts the measurement from the time patients registers at clinic counters instead of from the central registration counter, even if in reality the patients have spent a long time waiting at the latter. Some argue that the term ‘from the time of registration’ may mean registration at clinic registration counter too, and it is only fair to do so because the amount of time patient waited at the central registration counter is not part of the time they spend waiting for consultations in the clinic of any particular discipline. The fact that the waiting time KPI is assigned to each discipline probably had fuelled this type of data manipulation. Alas, the KPI has unintentionally created mental silos within the complex healthcare system.
Improvement 1: Remove central registration counters
Since the redundancy of the central registration counter can be recognised and have been identified as a non-value added step in the patient journey from arrival to consultation, why don’t we just remove it? Instead of manipulating data and discounting the time patients actually wait at the central registration counter in the name of achieving a KPI, isn’t it better if we really cut off the part where patients have to wait there? It is time we reflect the true/ absolute waiting time that patients have to endure when they turn up to outpatient clinics for their appointments. I’m aware that there may be policies, processes and systems that may be obstacles to creating this drastic change at this present moment, but I’m positive that there are steps that can be taken to overcome this.
Manipulation 2: Start seconds before doctor ready to call patient in
This is by far the most creative yet complex way of waiting time data manipulation that I’ve ever encountered. It is so cleverly engineered that even doctors who participate in the process may be oblivious to it. This is how it is done (Disclaimer: I’m not teaching you how to do it, just merely demonstrating the process. I shall not be responsible of what you do with the following information):
- Patients are given call numbers at the clinic counter and asked to wait. Their arrival is not acknowledged in the queue management system yet.
- When the doctors are ready to call a patient in, he/she will punch in a pre-specified call number, for example, 0000, that will appear on the queue management system’s display.
- When a nurse at the counter sees this number (0000) on the display, he/she will immediately click on a patient’s name in the system to indicate that the patient has ‘just arrived’, even though the patient has waited for hours in front of the clinic as instructed earlier. This marks the start of the waiting time measurement.
- Then, the nurse would quickly push the patient’s name and his/her corresponding call number into the doctor’s queue list.
- The doctor would then be able to punch in the patient’s actual call number and start a consultation, marking the end of the waiting time measurement.
What the process above does is to cut the waiting time to just a few seconds. After all, it only takes a few seconds to move from step 3 to step 5 and that’s how it will eventually appear in the waiting time data collection sheet – 30 seconds of waiting time. This injustice to patients is often justified as attempts of practising the Just In Time (JIT) concept. You know, push patients into doctors’ queue list just in time when the doctors request them. But that doesn’t justify why the patients’ arrival is not acknowledged just in time in step 1, now does it?
Improvement 2: Slot patients accordingly and acknowledge arrival as it happens
If we want to be serious about cutting our patients’ waiting time to a fraction of a minute just like the above data manipulation tried desperately to achieve, then we should take appointment slotting very seriously. If we put patients into properly designated appointment slots beforehand (read What Appointment Time? for more information on this process), we could just measure the start time from the pre-allocated appointment time as stated within the KPI’s technical specification. We should simply acknowledge patient arrivals as they occur, and need not worry about over-engineering such complex data manipulation systems that not only confuses everyone (doctors and patients alike) but also defeats the purpose of measuring the waiting time in the first place.
Manipulation 3: End measurements at Vital Signs Room
This is a great example of creative interpretation of the definition stipulated within the KPI technical specifications. It is stated that the measurement of the waiting time ends when a patient is seen by a doctor. So, a few house officers may have been stationed in the Vital Signs Room, where blood pressure, heart rate, oxygen saturation and temperature are recorded prior to seeing another doctor for consultations just to fit into the definition and hence ‘justified’ as the endpoint of the waiting time measurement. This creates the illusion that patients did not have to wait for a long time to see a doctor when in reality they do have to. The patients have to wait a while longer after having their vitals signs taken before actual consultations with their doctors take place.
Improvement 3: Convert the Vital Signs Room into consultation room
Proponents of the above manipulation method often insist that having house officers stationed in the vital signs room is essential as it helps allay patients growing concerns about waiting time. I get it. It is a way in which they try to influence patients’ perception of the waiting time. But, they are missing the point of the argument altogether. It is not the effect of placing house officers in the vital signs room that we are arguing about, but rather its role as the endpoint of waiting time measurement. Instead, I’d like to propose that all vital signs room be reassigned as consultation rooms (given that the infrastructure is permissible) so that its place as the end of waiting time measurement is not construed, while also helping increase the number of rooms available for doctors to consult.
Perhaps there are more manipulations of waiting time data from which improvement measures could be drawn upon. But, I would like to end the list as it is for now. Note that the list above is not exhaustive, nor would it fit into the processes of every outpatient clinics. Each clinic is unique in its processes and operations, hence requiring unique improvement solutions. Whatever you decide to do, just remember that data misrepresentation is an injustice to patients and have the potential to make one unintentionally blinded to the reality that patients go through. Never discount the amount of time patients spend waiting for consultations, for if we do we are fooling none other than ourselves. Please don’t miss out on the opportunity to make patient experience better.
As always, thank you for reading. This is my lengthiest post by far. Please share it if you find it beneficial. And don’t forget to subscribe to my mailing list so that you’ll receive an email every week when I publish new posts up on this blog.