The cost of long hospital stays
Background
Inpatient investigation and treatment is expensive. In the current financial climate of the NHS, there is a drive to reduce inpatient admission times. Streamlining the inpatient process would be of benefit to patients, who generally prefer tests to be performed promptly, and are at higher risk of potential iatrogenic complications during prolonged hospital stays, such as hospital-acquired infection1 and venous thromboembolism.2
Hospital managers use systems to estimate the cost of a hospital admission to facilitate budgeting. Patients are assigned to groups based on demographic and disease characteristics. In many hospitals, a coding proforma is filled for each patient at the time of discharge, usually by the junior medical staff, summarising data such as age, primary and secondary co-morbid diagnoses and procedures performed. These data are combined to allocate each patient a health resource group (HRG) code.3 This code is associated with a tariff, which is an estimate of the cost-per-day of the inpatient admission. The tariff is linked to a trim-point, which is the maximum funded length of stay for that patient for that admission. If the trim-point is exceeded, a financial penalty is incurred. For the purposes of this study, an episode during which a patient exceeds the trim-point will be termed an ‘overstay’.
It was hypothesised that inpatient overstays would prove a significant source of departmental overspending. The aim of this study was to assess the magnitude of the problem, associated financial costs, and to identify the contributory factors that could be addressed in order to reduce length of stay and save money.
Methods
The study was approved by the local audit and service evaluation committee. Sixty sets of consecutive case notes were obtained from inpatients on the acute neurology ward at the Royal Hallamshire Hospital, the neurology tertiary referral centre for the South Yorkshire region, with a catchment population of 1.8 million people.4 For each patient, data were collected summarising demographics, diagnosis, procedures performed and time spent waiting for each investigation, opinion, transfer and report. Actual length of stay was compared to the assigned financial trim-point. The percentage of overstays and the associated financial costs for the sample were calculated. These data were extrapolated to calculate the total costs for the neurology unit per annum, by multiplying the mean overstay cost per patient by the number of inpatient admissions over the year.
Results
The patient group comprised 31 men and 29 women. Mean age was 53 years (standard deviation (SD) 18). Mean waiting times for each procedure are reported in Table 1. Mean length of stay was 10.4 days (SD 11.7, median 7, range 0–75). The mean allocated trim-point was 22.8 days (SD 16.2, median 15, range 3–57). However, 20% of patients exceeded their trim-point incurring financial penalties and, for these patients, the mean overstay was 6.75 days (SD 4.54, median 5.5, range 2–18), with a mean penalty of £1,385. Extrapolating these data to the population of 4,326 neurology inpatients seen in the unit last year resulted in estimated total costs of £1.2 million per annum, or 5.6% of the annual budget.
Discussion
Overstays are a significant problem, resulting in large financial penalties. Strategies for reducing these costs could focus on streamlining the inpatient investigation process, by investing in pressurised services, such as imaging and rehabilitation units; waiting for social assessment in an acute neurology bed is an inappropriate use of a costly resource.
Another strategy to reduce costs would involve addressing the system used to predict length of stay, which suffers from a logical flaw. The trim-point is derived following coding at the time of discharge (as it is influenced by diagnoses, procedures, etc). Therefore, overstays are defined after the event, so cannot be identified, monitored or prevented during the inpatient stay. The coding proformas used to allocate patients to HRGs appear a crude implement; there is considerable ambiguity in categorisation and they are usually filled by staff without specific training.
Addressing these issues has the potential to make large financial savings and improve the hospital inpatient process for patients.
Footnotes
Letters not directly related to articles published in Clinical Medicine and presenting unpublished original data should be submitted for publication in this section. Clinical and scientific letters should not exceed 500 words and may include one table and up to five references.
- © 2012 Royal College of Physicians
References
Article Tools
Citation Manager Formats
Jump to section
Related Articles
- No related articles found.
Cited By...
- No citing articles found.