As readmission rates continue to rise around the country, one trust’s study into the causes should help it avoid financial penalties - and paves the way for others to do the same, reports West Suffolk Hospitals Trust’s senior house officer Nina Wilson.

Unplanned readmission rates have been rising year on year across the country. The government has equated this with failing care and, as part of its focus on outcomes, is bringing in financial penalties to try to reduce readmissions.

To avoid being hit by these penalties trusts need to optimise care. This means identifying the causes of readmissions and implementing the most effective interventions where they will have the most impact.

A project at West Suffolk Hospitals Trust, a 547 bed district general hospital serving a population of 275,000 people, set out to identify the types of patients most likely to be readmitted and why.

Unplanned admissions into the medical department enter the hospital via the emergency admissions unit, regardless of whether they are referred in by GPs or come via accident and emergency. During May 2010, the unit care coordinator recorded details of all patients admitted who had been inpatients at the hospital during the previous 30 days. The hospital discharge summary was used to create a database detailing the reason for the original admission, the reason for readmission, original length of stay and the duration spent out of hospital in between.

To obtain an understanding of the more subtle factors leading to each readmission, consultants on the post-take ward round were then asked to write brief comments identifying precipitating factors or events which may have been linked to the readmissions. They were also asked to suggest how the readmission might have been prevented. Their comments were combined with the information from the hospital discharge summaries and how many times patients had been admitted to hospital over the last two years, to create the full data set.

Using the data set the readmissions landscape could be characterised according to age and problem, and it was possible to look at specific challenges such as failed discharges (readmission within 48 hours) and recurrent attenders.

Major outliers

Alongside this, Dr Foster data for the hospital was analysed for major outliers – those medical conditions that generate higher relative risk of readmission than the national average. These were easy to pick out as they were marked with a red alert symbol.

The hospital data revealed that a total of 121 patients were readmitted through the unit during May 2010: 11 per cent were readmitted with a psychiatric problem, 22 per cent were failed discharges returning within 48 hours and 11 per cent were failed discharges returning in 24 hours or less.

As might be expected, the reasons for readmission varied according to age group. The vast majority (64 per cent) of readmissions in the youngest age group of 16-39 years were attributed to psychiatric problems, such as recurrent overdose or alcoholism. These patients had been admitted under the medical team for observation while the effects of the drugs/alcohol wore off, before psychiatric review or discharge.

Similar psychiatric problems accounted for approximately one quarter (27 per cent) of medical readmissions in the next age group of 40-64 years. A further 27 per cent were due to complications of cancer such as neutropaenic sepsis or recurrent pleural effusion and one third were due to either recurrent chest pain or shortness of breath. One patient had four admissions within one month with recurrent troponin-negative chest pain. The final two of the 22 patients in this age group were admitted with apparently unrelated complaints (surgical admission returning with back pain and infective exacerbation of chronic obstructive pulmonary disease returning with allergy 15 days later).

In the older age groups, higher levels of comorbidity meant the reasons for readmissions became more complicated. In general, chronic problems were a dominant feature in the 65-84 year old category. A significant proportion (17 per cent) were due to chest pain, half of which was recurrent. Recurrent shortness of breath, particularly in COPD patients, also featured in this age category.

Many patients aged 85-100 had a complex mix of health and social difficulties relating to age and independence that combined to result in a readmission. Most importantly, 28 per cent of these were failed discharges.

With respect to recurrent attendance, 33 of the 121 patients had been admitted more than eight times in the last two years and eight had been admitted more than 15 times.

17%:

Recurrent readmissions in 65-84 year olds that were due to chest pain

Analysis of patient coding revealed that a proportion of the readmissions (11 per cent) were not true emergency readmissions. Six patients were attending for weekly chemotherapy or transfusions, two patients were transferred to and then returned from tertiary centres and another self discharged then returned. Four other patients had returned with unrelated problems, for example one patient admitted electively for bladder botox was later admitted as an emergency with a chest infection.

Six unscheduled readmissions related to palliative patients, for example multiple day admissions for drainage of recurrent pleural effusion. In these cases easy access to secondary care as and when necessary may well represent optimum care rather than failing care, so thought needs to be given to such pathways and whether these would be eligible for exception reporting from the proposed penalties.

The Dr Foster data flagged red alerts for readmissions relating to leg pain, anaemia and drainage of peritoneal cavity. These readmissions were related to the patient pathways for deep vein thrombosis, top-up blood transfusions and ascitic taps, which needed to be looked at.

At West Suffolk a new pathway for DVT was already in the pipeline. If a GP suspected DVT, instead of sending the patient to hospital for D-dimer and then again the following day for an ultrasound scan, point of care D-dimer testing has allowed the first part of the pathway to be done in the community. If positive, the first dose of treatment can be given and arrangements made with the unit by phone for day case admission and an ultrasound scan the following day.

Particular scrutiny

West Suffolk Hospital Trust’s experience reveals that readmission areas which may be worthy of particular scrutiny include psychiatry, recurrent chest pain, shortness of breath, failed discharges in the elderly and recurrent attendance. Also approximately 10 per cent of readmissions could relate to the way some non-emergency admissions are entered onto computer systems.

As well as undertaking similar exercises themselves, trusts will find software useful for highlighting areas where patient pathways or flow can generate increased readmission rates.

It is also worth giving consideration to patient centred pathways, which could avoid non-elective admission, be treated in an elective setting if necessary and be exempt from penalties.

Readmissions by age

Age GroupsNumber of Readmissions
16-3914
40-6422
65-7935
80-10050
Total121