Methods

Descriptive analysis of emergency admissions from care homes nationally

The IAU developed a linked dataset that allowed us to look at A&E attendances and hospital admissions for permanent residential and nursing home residents across England for the first time. This dataset is based on data from 17 April 2016 to 15 April 2017.

Our method relied on accessing data on addresses that were collected by general practices in England. These were cross-referenced with a list of the addresses of care homes from the Care Quality Commission (CQC). Once care home residents were identified in this way, we linked this information to data on hospital admissions obtained from the Secondary User Services which is a national, person-level database that is closely related to the widely used Hospital Episode Statistics.

All processing of address information, and subsequent linkage of patient information was carried out by Arden & Greater East Midlands Data Services for Commissioners Regional Office (Arden & GEM DSCRO). The IAU subsequently carried out the analysis of the linked dataset using ‘pseudonymised’ information in a secure environment hosted by the Health Foundation. All data accessed by the IAU are anonymised in line with the Information Commissioner’s Office’s code of practice on anonymisation. Arden & GEM DSCRO and the Health Foundation are Data Processors on behalf of NHS England.

One limitation of our method is that it relies on patients updating their address information with their general practice after they move to a care home. As a result, we are likely to have identified only people who move to a care home on a permanent basis, and assume that we have excluded people who move to a care home temporarily for respite care, rehabilitation, short stays or other purposes. We will also have missed the early parts of care home stays if individuals did not update their address with their general practice immediately after moving to the care home.

Correction

Our method identified 195,296 permanent care home residents aged 65 or older in England at any point in time in 2016/17, compared with 274,040 according to Census 2011. Validation work has shown this is underestimated due to the method used to automate the cleaning of address information, but we found no evidence this underestimate varies across different age bands or geographies. To correct for this difference, all estimates on the number of emergency admissions have been multiplied by a factor of 1.403. Note, this correction does not affect our estimates of admissions rates.

Although the number of permanent care home residents in England, according to the census, dates back to 2011, we think it is still appropriate. We compared the distribution of age and gender at national level as well as patients aged 65 or older at regional level and found no substantial differences in variation between our linked dataset and census data. Data on all care home residents (including people living in a care home temporarily for respite care, short breaks or other purposes) shows little increase in the total number of residents, going from 328,600 in 2011 to 337,500 residents in 2016 (a 2.7% increase in five years).

Measuring emergency hospital admissions for care home residents

An emergency admission is one where a patient is admitted to hospital urgently and unexpectedly (that is, the admission is unplanned). Emergency admissions often occur via A&E departments but can also occur directly via GPs or consultants in ambulatory clinics. We identified these admissions from the hospital data linked to information about care home residents.

We also examined a subset of emergency admissions for specific conditions that were potentially manageable, treatable or preventable outside a hospital setting (so-called ‘potentially avoidable’), namely:

  • acute lower respiratory tract infections, such as acute bronchitis
  • chronic lower respiratory tract infections, such as emphysema
  • diabetes
  • food and drink issues, such as abnormal weight loss and poor intake of food and water, possibly due to neglect
  • fractures and sprains
  • intestinal infections
  • pneumonia
  • pneumonitis (inflammation of lung tissue) caused by inhaled food or liquid
  • pressure sores
  • urinary tract infections

The list of conditions was originally developed by the CQC, as part of its analysis of older people receiving health and social care. These were unplanned admissions for conditions that were potentially manageable, treatable or preventable outside of a hospital setting, or for conditions that could be caused by poor care or neglect. For example, some fractures may be avoidable with appropriate risk assessment and falls prevention, and urinary tract infections may be treatable within the community or care home. However, the complexity of the patient group means that ‘potentially avoidable’ admissions are not, when they become acute, necessarily avoidable. Context is also a factor in determining whether a ‘potentially avoidable’ admission could in fact have been avoided. For example, if a residential care home resident had pneumonia, hospital admission may be the most effective way of eliminating the infection quickly, whereas the available nursing support in a nursing home may have been able to oversee treatment of the same condition within the nursing home. Admissions for these conditions cannot always be avoided, especially given the complex and interacting health needs of care home residents. However, the enhanced support available in care homes could be expected to have greater impact on admissions for these conditions than others.

Further research is needed to validate the appropriateness of these conditions as a marker of avoidable admissions for the care home population. In the meantime, the data presented can only be taken as an illustration of the range of health conditions for which care home residents are admitted to hospital, some of which are potentially avoidable.

We expressed the number of admissions as rates per person per year. The rates were calculated as the total number of hospital admissions for all care home residents during 2016/17 divided by the number of days spent in the care home across all residents. That was then multiplied by 365 to provide a rate per person per year. Thus, two care home residents who stay in the care home for six months each would have counted towards one year in the rate.

In addition to looking at how often care home residents are admitted to hospital, we also examined the average length of hospital stay. For patients who stayed in hospital overnight, we calculated the length of their hospital stay as the number of nights spent in hospital. Patients who were admitted and discharged from hospital on the same day were counted as having spent half a day in hospital.

Long-term conditions

To investigate the level of complexity of the health needs of the care home population, we calculated the prevalence of certain long-term conditions and markers of frailty. Using the International Classification of Diseases (ICD) diagnosis codes recorded in either primary or secondary diagnosis fields in inpatient hospital data, we calculated the Charlson Index,, as an aggregate measure of the burden of disease, and the long-term conditions listed by Elixhauser, , as well as conditions related to frailty. To calculate these, we used inpatient data from the three years prior to 2016/17 (or prior to the date a patient moved into a care home if in 2016/17). If a resident did not have an inpatient admission, it is not possible to determine if they had any of these conditions; therefore, these were only calculated for the subset of residents who had a prior hospital admission.

Review of four case studies

The IAU has to date evaluated four initiatives to improve health and care in care homes that were either EHCH vanguards or similar initiatives associated with the NHS’s New Care Models programme. A review was conducted to identify key themes and learnings from the evaluation reports related to these four initiatives. The aim of this analysis was to identify patterns in relation to the ‘inputs’ each site invested (that is, ‘what was done’ locally) with a view to identifying any common themes that may prove useful in understanding impact, as measured by the IAU in terms of secondary hospital use.

We examined the five IAU reports evaluating the impact of the four care home initiatives:

  • Two reports on Principia enhanced support in Rushcliffe between August 2014 and August 2016,
  • Sutton Homes of Care between January 2016 and April 2017
  • Wakefield Clinical Commissioning Group (CCG) between February 2016 and March 2017
  • Nottingham City CCG between September 2014 and April 2017.

In addition to the five studies carried out by the IAU, this briefing, where applicable, also draws on local evaluation reports for each of the sites and the EHCH framework itself to assess to what extent the elements of the EHCH framework were implemented. These reports were:

  • Cordis Bright (2018) NHS Nottingham City CCG. Enhanced Health in Care Homes vanguard evaluation: final report (available upon request from NHS Nottingham City CCG)
  • SQW (2018) Evaluation of Sutton Homes of Care Vanguard report
  • Wakefield Public Health Intelligence Team 2018 report on the evaluation of a holistic assessment approach to: supporting care home and independent living schemes
  • The NHS’s New Care Models framework for Enhanced Health in Care Homes.

All documents were examined using a thematic analysis approach with the pre-existing EHCH framework used to identify and code key themes. In addition to the pre-existing framework, the researchers familiarised themselves with the studies and local evaluation reports with a view to identifying additional factors and themes that could assist in understanding. The documents were coded into key themes to better enable the identification of patterns.

Pseudonymised datasets have been stripped of identifiable fields, such as name, full date of birth and address. A unique person identifier (such as NHS number) has been replaced with a random identifier. This random identifier is used to identify hospital activity from the same patient over time, but cannot be used to identify an individual.

§ Our method was based on the primary diagnosis code associated with an admission.

Previous Next