Discharge destination and length of stay: differences between US and English hospitals for people aged 65 and over

Discharge destination and length of stay: differences between US and English hospitals for people aged over 65

[posted as supplied by Dr Foster]

In England, people aged 65 and over comprise 16% of the population but occupy

nearly two-thirds of general and acute hospital beds. Because of the pressure on capacity and concern about bed blocking, particularly by elderly patients, interest in the use of intermediate care facilities, such as hospices and nursing homes, has grown. The health system in the US provides an alternative model for the coordinated use of hospital beds and intermediate care facilities.

Extra tables

Table 1a. Discharge destinations in England for year April 2001 – March 2002, for people aged 65 and over with length of stay >0 days
 

Table 1b. Discharge destinations in US (Medicare data) for year October 2000-March 2001 for people aged 65 and over with length of stay >0 days
 

Methods

Overview

This analysis was produced by comparing English hospital episode statistics (HES) data with US Medicare data. For both countries the analysis was based on hospital admissions for patients aged 65 and over, and excluded admissions for which the length of stay was recorded as 0 days (in order to exclude day cases). Destination of discharge and the mean and median lengths of stay were calculated for all diagnoses, and for four specific common diagnoses. The Medicare data cover the period 1 October 2000 to 30 September 2001 and the HES data cover the period 1 April 2001 to 31 March 2002. Military hospitals were not included. Acute and psychiatry diagnostic data were included (maternity data were excluded by the 65 and over age limitation).

The hospital data include a wide range of details such as the patient’s age, sex and diagnosis and any operations they had. The Medicare data covered hospital admissions regardless of how many specialists/consultants were seen but the HES data were in the form of consultant episodes (the continuous period during which the patient is under the care of one consultant), which needed to be linked into "spells" or admissions. About 10% of spells comprised more than one episode, and the patient’s method and date of discharge needed to be derived from the final episode in the spell. Episodes were assumed to be duplicates and excluded if they had the same combination of provider, date of birth, sex, postcode, date of admission and episode number. Some spells had the same date of admission but different dates of discharge. This was not valid unless the patient was discharged and readmitted on the same day, so the spell with the earliest admission date was arbitrarily taken to be the valid one. Only hospital admissions with known discharge destinations were included (less than 0.1% of admissions were excluded for this reason).

Coverage of the population

The HES data cover all National Health Service hospital admissions in English hospitals – including admissions to private wards in NHS hospitals, but do not cover admissions to independent hospitals in England, which account for only a few percent of all hospital admissions. Medicare is the publicly funded health service in the US that covers approximately 88% of hospital admissions for people aged 65 or more. Medicare patients account for about 70% of all deaths in US hospitals.

Individual diagnoses

The US Clinical Classification System(CCS codes – each code representing a group of clinically related ICD9 diagnoses) developed by the US Agency for Healthcare Research and Quality (AHRQ), was used to define the diagnostic groups in the Medicare dataset and admissions were ranked in descending order of total length of stay by CCS code. Four of the seven top diagnoses that together cover about a sixth of all bed usage - acute myocardial infarction, fractured hip, stroke and heart failure – were chosen and analysed individually. The combination of ICD10 codes corresponding to the CCS groups were similarly analysed. The codes included were:
 

References
 

1.   Department of Health National Beds Inquiry: 1998/9 Local Variation Analysis Project Report 11 December 2000. http://www.doh.gov.uk/nbi/lvafinalreport.pdf
2.   Elixhauser A, Andrews RM, Fox, S. Clinical classifications for health policy research: Discharge statistics by principal diagnosis and procedure. Provider Studies Research Note 17. Rockville, MD: Agency for Health Care Policy and Research; 1993. AHCPR Pub. No. 93-0043.