Home   About Kurtosis   Courses   Course Calendar   Booking Information   Ideas   Technique   Clients   Contact

Under pressure

Do we discharge inpatients from hospital when they are ready to be discharged? Or do weinsteaddischarge them when we need their beds?

It’s the 2nd of January and it’s the ninth day of Christmas. Acute hospitals across the land will be starting to fill up nicely now. And a lot of people think the reason why hospitals fill up at this time of year is less to do with there being more admissions than usual butinsteadbecause there are fewer discharges than usual. The days following Christmas Eve tend to see markedly fewer inpatient discharges than usual.

And it's not just that fewer patients get discharged when the hospital is quiet. The reverse is also true. When hospitals start to fill up, the call goes out from the bed managers: “Discharge more patients!” So you’d sort of expect there to be a relationship between how full the hospital is and how many patients get discharged. On busy days we’d discharge more patients; on quiet days, not so many.

So, is this true? Is the number of inpatient hospital discharges we get on any particular day determined primarily by the pressure on beds? Let’s have a look at one hospital’s inpatient workload over a period of nearly three years. For each of the 1,000 days in that period (from 1st January 2009 to 27th September 2011) we measured how many inpatients were occupying beds at midday to see how full the hospital was on each day.

And then we measured how many discharges there were on each of those 1,000 days.

And then we summarised these measurements.

To do this, we first of all split the data into three specialty-based chunks: Orthopaedics, General Surgery and Medical.

Secondly, we asked, for each specialty, how many dischargeson averageoccurred on the quiet days? How many occurred on the not-so-quiet days? How many occurred on the busy days? In fact, we calculated the average number of discharges that occurred at every level of “full-ness”.

So, taking Orthopaedics as our example (see the chart below), the quietest days were those when there were only 29 patients occupying beds at midday. There were two such days in the 1,000-day period. And the average number of discharges on those two days was five per day. Then we looked at the days when there were 30 patients occupying beds at midday (there was only one such day, and there were six discharges on that day). And so on, all the way up to the day when there were 86 patients occupying beds when there were 23 discharges.

And when you draw a scatterplot of this data you see that there is a close-to-perfect relationship (r = 0.83) between bed occupancy and the number of patients discharged. The fuller the hospital, the more likely patients are to be discharged.

And it's not just Orthopaedics where we see this relationship. In General Surgery we see an even closer relationship (r = 0.96):

On days when there are 50-60 beds occupied in the surgical wards, about seven or eight patients get discharged. By contrast, on days when there are 90-100 beds occupied, about eighteen patients get discharged.

And in the largest specialty groupingthe medical specialtieswe see the same relationship:

On quiet days, roughly 10-20 patients get discharged; on busy days, roughly 25-35 patients get discharged.

Correlation is not causation. But these scatterplots do appear to be painting a bleak picture of acute hospital discharge planning. As pressure on beds grows, more patients get discharged. This can't be what the NHS means by "criteria-led discharge".

[2 January 2013]


Subscribe to the Kurtosis Mailing List

Email address:

We'll use your email address only to send you updates about blogs, ideas, techniques, new courses and new course dates,  We'll not share your email address with anyone else.

Make a comment on this article:


Job Title:


Email address:

© Kurtosis 2013. All Rights Reserved.