Variation in ambulance call rates for care homes in Torbay, UK.

TLDR: The pattern of ambulance call rates from care homes is described and factors predicting those homes calling for an ambulance most frequently are identified to determine the extent to which call rate variability reflects the different needs of resident populations or differences in care home policies and practice.

Abstract: Emergency ambulance calls represent one of the routes of emergency hospital admissions from care homes. We aimed to describe the pattern of ambulance call rates from care homes and identify factors predicting those homes calling for an ambulance most frequently. We obtained data from South Western Ambulance Service NHS Foundation Trust on 3138 ambulance calls relating to people aged 65 and over from care homes in the Torbay region between 1 April 2012 and 31 July 2013. We supplemented this with data from the Care Quality Commission (CQC) website on home characteristics and outcomes of CQC inspections. We used descriptive statistics to identify variation in ambulance call rates for residential and nursing homes and fitted negative binomial regression models to determine if call rates were predicted by home type (nursing versus residential), the five standards in the CQC reports, dementia care status or travel time to hospital. One hundred and forty-six homes (119 residential and 27 nursing) were included in the analysis. The number of calls made ranged from 1 to 99. The median number (IQR; range) of calls per resident per year was 0.51 (0.21-0.89; 0.03-2.45). Nursing homes had a lower call rate than residential homes [adjusted rate ratio (ARR) 0.29; 95% CI: 0.22-0.40; P < 0.001]; care homes failing the quality and suitability of management standard had a lower call rate compared to those who passed (ARR 0.67; 95% CI: 0.50-0.90; P = 0.006); and homes specialising in dementia had a higher call rate compared to those not specialising (ARR 1.56; 95% CI: 1.23-1.96; P < 0.001). These findings require replication in other regions to establish their generalisability and further investigation is required to determine the extent to which call rate variability reflects the different needs of resident populations or differences in care home policies and practice.

Figures

Table 2: Results from negative binomial regression models of number of calls

Table 1: Descriptive statistics for home size and ambulance calls from residential and nursing homes

Figure 1: Observed vs predicted number calls over the 16 month period by home

Full text

ORE Open Research Exeter
TITLE
Variation in ambulance call rates for care homes in Torbay, UK
AUTHORS
Hancock, J; Matthews, JN; Ukoumunne, OC; et al.
JOURNAL
Health and Social Care in the Community
DEPOSITED IN ORE
26 July 2016
This version available at
http://hdl.handle.net/10871/22730
COPYRIGHT AND REUSE
Open Research Exeter makes this work available in accordance with publisher policies.
A NOTE ON VERSIONS
The version presented here may differ from the published version. If citing, you are advised to consult the published version for pagination, volume/issue and date of
publication

Title: Variation in ambulance call rates for care homes in Torbay, UK
Authors: Jason HANCOCK, MBChB, MRCPsych, Academic Clinical Fellow and Higher
Specialist Trainee in Old Age Psychiatry, Mental Health Research Group, University of
Exeter Medical School, Room 1.15, College House, St Luke’s Campus, Heavitree Road,
Exeter, EX1 2LU, jason.hancock@nhs.net, 07850490726.
Justin MATTHEWS, BSc, MSc, Associate Research Fellow in Medical statistics, NIHR
CLAHRC South West Peninsula (PenCLAHRC), University of Exeter Medical school, Room
1.14, South Cloisters, St Luke's Campus, Exeter, EX1 2LU, J.N.Mathews@exeter.ac.uk, +44
(0) 1392 727418.
Obioha C UKOUMUNNE, BSc, MSc, PhD, Associate Professor in Medical Statistics, NIHR
CLAHRC South West Peninsula (PenCLAHRC), University of Exeter Medical school, Room
1.05, South Cloisters, St Luke's Campus, Exeter, EX1 2LU, O.C.Ukoumunne@exeter.ac.uk,
+44 (0) 1392 726070.
Iain LANG, MFPH, DPhil, MSc, MA, MA, NIHR Knowledge Mobilisation Research Fellow
and Clinical Senior Lecturer in Public Health, University of Exeter, South Cloisters,
University of Exeter, Heavitree Rd, Exeter EX1 2LU, i.lang@exeter.ac.uk, +44 (0) 1392
726087.
David SOMERFIELD, MBChB, FRCPsych, Chief Operating Officer & Consultant
Psychiatrist, Devon Partnership NHS Trust, Wonford House Hospital, Dryden Road, Exeter,
Devon, EX2 5AF, d.somerfield@nhs.net, 01392 208654.
James WENMAN, BSc, Clinical Development Manager, South Western Ambulance Service
NHS Foundation Trust, Abbey Court, Eagle Way, Exeter EX2 7HY,
james.wenman@swast.nhs.uk, 01803 615010.
Chris DICKENS, MBBS, MRCP, MRCPsych, MSc, PhD, Professor of Psychological
Medicine, Mental Health Research Group Lead Member, University of Exeter Collaboration
for Academic Primary Care (APEx), University of Exeter Medical school, Room 1.04,
College House, St Luke’s Campus, Heavitree Road, Exeter, EX1 2LU,
C.M.Dickens@exeter.ac.uk, 01392 726013.


1

Title: Variation in ambulance call rates for care homes in Torbay, UK
Abstract:
Emergency ambulance calls represent one of the routes of emergency hospital admissions from
care homes. We aimed to describe the pattern of ambulance call rates from care homes and
identify factors predicting those homes calling for an ambulance most frequently. We obtained
data from South Western Ambulance Service NHS Foundation Trust on 3138 ambulance calls
relating to people aged 65 and over from care homes in the Torbay region between 1/4/12 and
31/7/13. We supplemented this with data from the Care Quality Commission (CQC) website
on home characteristics and outcomes of CQC inspections. We used descriptive statistics to
identify variation in ambulance call rates for residential and nursing homes and fitted negative
binomial regression models to determine if call rates were predicted by home type (nursing
versus residential), the five standards in the CQC reports, dementia care status or travel time to
hospital. One hundred and forty-six homes (119 residential and 27 nursing) were included in
the analysis. The number of calls made ranged from 1 to 99. The median number (IQR; range)
of calls per resident per year was 0.51 (0.21 to 0.89; 0.03 to 2.45). Nursing homes had a lower
call rate than residential homes (adjusted rate ratio (ARR) 0.29; 95% CI: 0.22 to 0.40 ;
p<0.001); care homes failing the quality and suitability of management standard had a lower
call rate compared to those who passed (ARR 0.67; 95% CI: 0.50 to 0.90; p=0.006); and homes
specialising in dementia had a higher call rate compared to those not specialising (ARR 1.56;
95% CI: 1.23 to 1.96; p<0.001). These findings require replication in other regions to establish
their generalisability and further investigation is required to determine the extent to which call-
rate variability reflects the different needs of resident populations or differences in care home
policies and practice.

2

Keywords: Ambulance, Care homes, Dementia, CQC
What is already known about the subject:
 The number of calls received by ambulance services are increasing, as are the number
of emergency hospital admissions.
 Residents living in an area with a care home have a higher chance of being admitted to
hospital as an emergency compared to an age-matched population.
What this paper adds:
 This is the first study to investigate factors associated with variation in ambulance call
rates from care homes.
 We found marked variation in the number of emergency ambulance calls made by care
homes.
 More ambulance calls were made by residential homes, dementia specialist homes, and
homes passing the Care Quality Commission ‘quality and management’ standard.

3

Introduction
Recent increases in the number of emergency hospital admissions have led to substantial effort
to reverse this trend(Bardot et al 2013, The Kings Fund 2013). Residents living in an area with
a care home have a higher chance of being admitted to hospital as an emergency compared to
an age-matched population (Smith et al 2015). Since 400,000 people currently live in a care
home in the UK (Age UK 2015), reducing the number of emergency hospital admissions from
this setting is a priority but we know little about the drivers of emergency admission from care
homes. One route to admission is via an emergency ambulance and understanding the variation
and factors driving ambulance call rates from care homes is important.
Our aims in this study were to describe the pattern and variation of ambulance call-outs for
people aged 65 and over living in residential or nursing homes in the Torbay area of south-west
England and to identify home-level factors associated with higher ambulance call-out rates.
Methods
We obtained data from the South Western Ambulance Service NHS Foundation Trust
(SWAST) for all calls from care homes in the Torbay region for people aged 65 and over, over
a 16 month period between 1
st
April 2012 and 31
st
July 2013. Torbay is a largely urban area
with a population of 130,000. There are a significant number of care homes in the region,
reflecting the high proportion of older people in the region. Torbay has a higher proportion of
people over 50 than the UK average (Torbay council 2011). The Care Quality Commission
(CQC) recognises two different types of care homes, residential and nursing. While both offer
care and support for residents, nursing homes also have 24 hour medical care from a qualified
nurse (Care Quality Commission 2015). We considered both types of home in this analysis.
Data were available on date and time of call, reason for the call, care home name, and postcode.
These data were summarised at the level of the home and supplemented by data we obtained

Frequently asked questions

Q1. What are the future works in this paper?

Clearly future studies would benefit from a more detailed analysis of the reasons why homes failed each of the CQC standards. Further work across the UK is needed to understand the extent to which call-rate variability reflects the different needs of resident populations or differences in care home policies and practice. 

Q2. What are the contributions in this paper?

The authors aimed to describe the pattern of ambulance call rates from care homes and identify factors predicting those homes calling for an ambulance most frequently. The authors used descriptive statistics to identify variation in ambulance call rates for residential and nursing homes and fitted negative binomial regression models to determine if call rates were predicted by home type ( nursing versus residential ), the five standards in the CQC reports, dementia care status or travel time to hospital. These findings require replication in other regions to establish their generalisability and further investigation is required to determine the extent to which callrate variability reflects the different needs of resident populations or differences in care home policies and practice. 

Q3. How many of the 119 residential homes were registered as specialising in dementia?

Seventy-four of the 119 (62%) residential homes and 16 of the 27 (59%) nursing homes were registered as specialising in dementia. 

Q4. How many calls were made from residential homes?

Calls following a fall or injury accounted for 40% (1163) of all calls from residential homes and 22% (52) of all calls from nursing homes. 

Q5. What is the main purpose of this study?

While much emphasis is often placed on the need to reduce the number of ambulance calls and emergency hospital admissions (Bardot et al 2013, The Kings Fund 2013) this study suggests that those homes with higher call rates may have a higher quality of management. 

Q6. What was the median number of calls per resident per year?

The median number (IQR; range) of ambulance calls per resident per year was 0.58 (0.34 to 0.99; 0.05 to 2.45) for residential homes and 0.12 (0.08 to 0.25; 0.03 to 1.00) for nursing homes. 

Q7. What was the rate of calls made by care homes?

Of the CQC standards only quality and suitability of management was a significant predictor of call rate: homes that failed this standard made fewer calls than those that passed (rate ratio 0.67; 95% CI: 0.50 to 0.90; p=0.006). 

Q8. What is the average call rate for care homes?

Nursing homes had a lower call rate than residential homes (adjusted rate ratio (ARR) 0.29; 95% CI: 0.22 to 0.40 ; p<0.001); care homes failing the quality and suitability of management standard had a lower call rate compared to those who passed (ARR 0.67; 95% CI: 0.50 to 0.90; p=0.006); and homes specialising in dementia had a higher call rate compared to those not specialising (ARR 1.56; 95% CI: 1.23 to 1.96; p<0.001). 

Q9. What were the reasons for the number of calls made by residential and nursing homes?

Models were fitted to the number of ambulance calls (outcome) using eight predictors: home type (residential or nursing), home5   dementia status (whether the home specialises in dementia care), whether the homes failed each of the CQC five standards and the log of predicted journey time to hospital by road.