ArticlePDF Available

Hospital readmission among people experiencing homelessness in England: A cohort study of 2772 matched homeless and housed inpatients

Authors:

Abstract and Figures

Background Inpatients experiencing homelessness are often discharged to unstable accommodation or the street, which may increase the risk of readmission. Methods We conducted a cohort study of 2772 homeless patients discharged after an emergency admission at 78 hospitals across England between November 2013 and November 2016. For each individual, we selected a housed patient who lived in a socioeconomically deprived area, matched on age, sex, hospital, and year of discharge. Counts of emergency readmissions, planned readmissions, and Accident and Emergency (A&E) visits post-discharge were derived from national hospital databases, with a median of 2.8 years of follow-up. We estimated the cumulative incidence of readmission over 12 months, and used negative binomial regression to estimate rate ratios. Results After adjusting for health measured at the index admission, homeless patients had 2.49 (95% CI 2.29 to 2.70) times the rate of emergency readmission, 0.60 (95% CI 0.53 to 0.68) times the rate of planned readmission and 2.57 (95% CI 2.41 to 2.73) times the rate of A&E visits compared with housed patients. The 12-month risk of emergency readmission was higher for homeless patients (61%, 95% CI 59% to 64%) than housed patients (33%, 95% CI 30% to 36%); and the risk of planned readmission was lower for homeless patients (17%, 95% CI 14% to 19%) than for housed patients (30%, 95% CI 28% to 32%). While the risk of emergency readmission varied with the reason for admission for housed patients, for example being higher for admissions due to cancers than for those due to accidents, the risk was high across all causes for homeless patients. Conclusions Hospital patients experiencing homelessness have high rates of emergency readmission that are not explained by health. This highlights the need for discharge arrangements that address their health, housing and social care needs.
Content may be subject to copyright.
1
LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
Hospital readmission among people experiencing
homelessness in England: a cohort study of 2772
matched homeless and housedinpatients
Dan Lewer ,1,2,3 Dee Menezes,1 Michelle Cornes,4 Ruth M Blackburn,1
Richard Byng,5 Michael Clark,6 Spiros Denaxas,1,7 Hannah Evans,1 James Fuller,4
Nigel Hewett,8 Alan Kilmister,4 Serena April Luchenski,1 Jill Manthorpe,4
Martin McKee ,9 Joanne Neale,10 Alistair Story,11 Michela Tinelli,6
Martin Whiteford,12 Fatima Wurie,3 Alexei Yavlinsky,1 Andrew Hayward,2,3
Robert Aldridge1
To cite: LewerD, MenezesD,
CornesM, etal. J Epidemiol
Community Health Epub
ahead of print: [please
include Day Month Year].
doi:10.1136/jech-2020-
215204
Additional material is
published online only. To view
please visit the journal online
(http:// dx. doi. org/ 10. 1136/
jech- 2020- 215204).
For numbered affiliations see
end of article.
Correspondence to
Dan Lewer, Institute of
Epidemiology and Healthcare,
University College London,
London, UK; d. lewer@ ucl. ac. uk
The analysis presented in this
paper was outlined in our
previously published study
protocol.
Received 23 July 2020
Revised 6 November 2020
Accepted 4 December 2020
© Author(s) (or their
employer(s)) 2021. Re- use
permitted under CC BY.
Published by BMJ.
ABSTRACT
Background Inpatients experiencing homelessness
are often discharged to unstable accommodation or the
street, which may increase the risk of readmission.
Methods We conducted a cohort study of 2772
homeless patients discharged after an emergency
admission at 78 hospitals across England between
November 2013 and November 2016. For each
individual, we selected a housed patient who lived in a
socioeconomically deprived area, matched on age, sex,
hospital, and year of discharge. Counts of emergency
readmissions, planned readmissions, and Accident and
Emergency (A&E) visits post- discharge were derived from
national hospital databases, with a median of 2.8 years
of follow- up. We estimated the cumulative incidence of
readmission over 12 months, and used negative binomial
regression to estimate rate ratios.
Results After adjusting for health measured at the
index admission, homeless patients had 2.49 (95% CI
2.29 to 2.70) times the rate of emergency readmission,
0.60 (95% CI 0.53 to 0.68) times the rate of planned
readmission and 2.57 (95% CI 2.41 to 2.73) times the
rate of A&E visits compared with housed patients. The
12- month risk of emergency readmission was higher for
homeless patients (61%, 95% CI 59% to 64%) than
housed patients (33%, 95% CI 30% to 36%); and the
risk of planned readmission was lower for homeless
patients (17%, 95% CI 14% to 19%) than for housed
patients (30%, 95% CI 28% to 32%). While the risk
of emergency readmission varied with the reason for
admission for housed patients, for example being
higher for admissions due to cancers than for those
due to accidents, the risk was high across all causes for
homeless patients.
Conclusions Hospital patients experiencing
homelessness have high rates of emergency readmission
that are not explained by health. This highlights the need
for discharge arrangements that address their health,
housing and social care needs.
INTRODUCTION
Homelessness is an enduring social problem that is
associated with poor health outcomes, with cohort
studies showing mortality rates of three to six times
the general population.1–4 Although the size and
structure of the homeless population are difficult to
estimate, data in England suggest steep increases in
recent years. The number of people sleeping rough
identified by official counts increased from 1353 in
2010 to 4266 in 2019,5 with actual numbers likely
to be greater than this. The same period also saw
a steep increase in hospital attendances for people
with ‘no fixed abode’.6
Leaving hospital is often a traumatic experience
for people without a fixed address, and surveys
suggest that 30%–70% of homeless inpatients are
‘discharged to the street’ (ie, sleeping rough imme-
diately after discharge).7 8 Recovery after a hospital
admission may be spent sleeping rough or in insecure
accommodation such as hostels or sofa- surfing. In
addition, access to ongoing community healthcare
may be poor. Qualitative research has identified
barriers including stigmatisation when accessing
health services, primary healthcare practitioners
refusing to register homeless people, and priorities
that compete with health such as arranging accom-
modation.9–11 As a result, outcomes after hospital
discharge may be poor, and studies in the USA have
shown that homeless inpatients are more likely to
be readmitted than housed inpatients.12–15
In response to concerns about poor discharge
arrangements for homeless inpatients, the UK
government set up the ‘Homeless Hospital
Discharge Fund’,16 which funded partnerships of
National Health Service (NHS) and non- profit
organisations to develop methods of supported
discharge. These schemes operated between 2013
and 2016 and used a range of models. Most
included a housing specialist who helped patients
access community health and housing services. In
some schemes, discharge was managed by a multi-
disciplinary team including general practitioners
(GPs), nurses, therapists, and housing workers.
Some had accompanying intermediate care facili-
ties providing accommodation and clinical support.
Data collected by the schemes give a detailed insight
into the outcomes of homeless inpatients after
discharge from hospital.
Our previous analysis of linked death records for
people attending the homeless hospital discharge
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
2LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
schemes found that deaths after discharge were related to a
wide range of physical health problems, with alcohol, drugs or
suicide (sometimes considered the main health problems in this
population) the primary cause for only one- third of deaths.17 In
this present analysis we compare the risk of hospital readmis-
sion among homeless inpatients with housed inpatients living in
socioeconomically deprived areas.
METHODS
We conducted a retrospective cohort study of the rate of hospital
readmission after an emergency hospital admission, comparing
homeless patients with housed patients living in socioeconomi-
cally deprived areas.
Data source
The data were collected as part of an evaluation of hospital
discharge schemes established through the Homeless Hospital
Discharge Fund.16 We worked with 17 schemes covering
78 hospitals across England between November 2013 and
November 2016. Schemes did not adopt a common definition
of homelessness and had various ways of identifying eligible
patients. Some relied on referrals from clinicians and others did
ward rounds. Services primarily focused on people living on the
streets or in night shelters and hostels for single homeless people,
but may also have worked with people who meet broader defi-
nitions of homelessness such as people who are sofa- surfing, or
at risk of losing an existing tenancy. Teams helped plan discharge
and helped patients access housing, intermediate care, and other
services that were available locally. Full details of the datasets are
available in a prepublished protocol.18
Data cleaning and linkage
We collected patient identifiers from the 17 homeless discharge
schemes. We excluded patients under the age of 18 years and
those with insufficient data for record linkage. We sent the iden-
tifiers to NHS Digital for deterministic linkage19 with national
hospital and mortality databases (‘Hospital Episode Statistics’
and the Office for National Statistics mortality database respec-
tively). Linked data were available until 31 March 2018. We
considered the ‘index’ admission as the first time a homeless
patient was seen by a hospital discharge service. The admission
dates provided by the homeless discharge schemes sometimes
varied from those recorded in the national hospital database,
and we used an algorithm to select the index admission (see
online supplemental information). For comparison purposes,
we requested data on hospital episodes and deaths for a sample
of housed inpatients admitted to the same 78 hospitals between
November 2013 and November 2016, with a home address in
the most deprived quintile of neighbourhoods based on the Index
of Multiple Deprivation,20 and who were not seen by a homeless
discharge scheme. We processed the hospital episode data into
‘spells’ (or ‘admissions’), because English hospital data are struc-
tured such that a single hospital spell is sometimes divided into
several episodes of care led by different doctors or departments.
A spell represents a continuous period of time in hospital.
Matching
For this analysis, we only included patients who were admitted
in an emergency. Planned admissions were rare in the homeless
group (8% of all admissions) and often represent healthcare
that can be conducted on a single day, such as dialysis or phys-
ical rehabilitation, and readmission may not be related to poor
discharge arrangements or ongoing care. We used a matched
design to compare homeless and housed patients. For each
homeless patient we selected at random one housed patient of
the same sex and age group (using age groups 18–24, 25–34
and then 10- year age groups) who was discharged alive from the
same hospital in the same year. We have published our matching
algorithm at https:// github. com/ danlewer/ homeless- discharge/.
We used this design to allow definition of an ‘index’ admission
for housed patients, while avoiding potential biases in choosing
an index date that may result from other approaches (such
as selecting an admission at random, which is likely to select
patients during periods of poor health). The derivation of the
study cohort is shown in figure 1.
Outcomes
Outcomes were the counts of planned hospital readmissions,
emergency hospital readmissions, and Accident and Emergency
(A&E) visits, as defined in the study protocol.18
Covariates
We defined the number of comorbidities as the number of
ICD-10 (10th revision of the International Statistical Classifica-
tion of Diseases and Related Health Problems) chapters that were
present in the primary diagnosis field for hospital admissions in
the 4 years prior to the index date. We used 4 years because all
participants had at least 4 years of prior data). We only counted
comorbidities from ICD-10 chapters 2–14 and 17,21 excluding
chapters such as infections where an admission may not repre-
sent a long- term condition. We defined the reason for index
admission using the ICD-10 chapter of the primary diagnosis,
grouping chapters accounting for fewer than 100 index admis-
sions as ‘other’ (see table 1 for a list of ICD-10 chapters included
in the analysis). For descriptive purposes, we also reported
whether discharge from the index admission was against medical
advice, and whether patients died during follow- up.
Statistical analysis
We estimated the probability of patients having at least one of
each outcome (emergency readmission, planned readmission and
A&E visit) in the 12 months after discharge, with each outcome
stratified by the ICD-10 chapter of the index admission. We used
the Kaplan- Meier method to estimate cumulative incidence at 12
months, because some participants had less than 12 months of
follow- up, for example due to death. The SEs of the cumulative
incidence were clustered by hospital to account for differences
in clinical practice.
To estimate rate ratios comparing homeless and housed
patients, we used a mixed negative binomial model for each
outcome, with homeless/housed status as the main independent
variable; adjustment for the matching variables age, sex, year of
discharge and a random intercept for hospital site; and an offset
for the log time- at- risk. We then additionally adjusted for the
count of comorbidities and the reason for admission. We used
negative binomial models because the outcomes were dispersed
(ie, the variance of the count of readmissions was greater than
the mean).
Analysis was conducted using R V.3.6.2.
RESULTS
A total of 3894 homeless patients were supported by the 17
specialist discharge services. Of these patients, 3309 were
admitted in an emergency and eligible for inclusion. We matched
2772 (84%) of these patients to a unique housed patient. The
remaining 537 could not be matched as there was no remaining
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
3
LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
eligible housed patient in our data. Unmatched homeless patients
were slightly younger and more likely to be men than matched
patients (online supplemental information).
Homeless patients had more comorbidities than housed
patients and were more likely to be admitted for mental health
problems or ‘external’ causes (including accidents). Table 1
shows the characteristics of homeless patients and matched
housed patients.
We plotted all readmissions and A&E visits for every patient
in the study (figure 2). Visual inspection of this plot suggested
that emergency readmissions and A&E visits occurred more
frequently for patients experiencing homelessness at the index
admission, while there was a similar rate of planned readmis-
sions. There were more series of repeat planned admissions
within the group experiencing homelessness (visible on the plot
as solid horizontal lines). Consequently, we conducted an explor-
atory analysis where we defined a ‘series’ as weekly (or more
frequent) planned admissions for at least 8 consecutive weeks.
In the homeless group, there were 16 such series involving 9
patients. There were 1201 admissions within these series, of
which 949/1201 (79%) had a primary diagnosis of renal failure
and most had procedure codes for dialysis. This means that one-
third (1201/3400; 35%) of the planned admissions in the home-
less cohort related to these series. In the housed group, there
were 11 series involving 7 patients. There were 571 admissions
in these series and 174 (30%) had a primary diagnosis of renal
failure. This means that one in eight (571/4769; 12%) of the
planned admissions in the housed cohort related to these series.
The 12- month risk of emergency readmission was 61% (95%
CI 59% to 64%) for homeless patients and 33% (95% CI 30% to
36%) for housed patients; for planned readmission it was 17%
(95% CI 14% to 19%) for homeless patients and 30% (95% CI
28% to 32%) for housed patients; and for A&E visits it was 94%
(95% CI 93% to 95%) for homeless patients and 84% (95%
CI 81% to 86%) for housed patients (figure 3). Among housed
patients, the risk of emergency readmission varied substantially
according to the cause of the index admission. For example,
patients admitted with a primary cause of cancer had a 56%
(95% CI 45% to 68%) risk of emergency readmission over the
following 12 months, compared with 25% (95% CI 20% to
30%) for patients admitted following an accident. In contrast,
the 12- month risk of an emergency readmission was greater
than 50% regardless of the reason for the index admission, with
limited variation.
After adjusting for comorbidities and the ICD-10 chapter of
the index admission, patients experiencing homelessness had
2.49 (95% CI 2.29 to 2.70) times the rate of emergency read-
missions, 0.60 (95% CI 0.53 to 0.68) times the rate of planned
readmissions and 2.57 (95% CI 2.41 to 2.73) times the rate
of A&E visits (table 2). As an exploratory sensitivity analysis
(not prespecified), we fit the model for planned readmissions
excluding ‘series’ of admissions (defined above), which gave
a rate ratio of 0.51 (95% CI 0.45 to 0.57) when adjusting for
matching variables only, and a fully adjusted ratio of 0.55 (95%
CI 0.49 to 0.62), that is, a wider relative difference between
homeless and housed patients after removing series of planned
admissions.
DISCUSSION
Hospital patients who are experiencing homelessness have
high rates of emergency readmission and A&E visits after
discharge. Following an acute illness, most patients are
expected to recover in their own home, and the rate of read-
mission is relatively low. Our results show this is not the case
for homeless patients.
Figure 1 Derivation of the study cohort. HES, Hospital Episode Statistics; NHS. National Health Service.
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
4LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
Similar to our study, studies of hospital admissions among
homeless populations in North America have reported diverse
physical and mental health problems on admission and high
rates of emergency readmission.8 12–15 Results are difficult to
compare directly due to differences in the populations, health-
care systems, and study methodologies. A study in the USA found
that 17% of inpatients with hospital records indicating homeless-
ness were readmitted within 90 days; only slightly higher than
14% of housed patients.15 Another study, also using homeless-
ness recorded in routine hospital data, found that homelessness
was associated with 1.4 times the odds of readmission.13 These
studies both found modest differences in readmission risk. Use
of routinely captured data to identify homelessness may have led
to inclusion of people experiencing less severe forms of home-
lessness, such as unstable or temporary housing. In contrast,
two other studies in the USA measured hospital readmission for
patients who live in homeless shelters, and are likely to have
longer histories of homelessness and sleeping rough.12 13 These
studies found that readmission risk was approximately three
times that of housed patients; more similar to our estimates. It is
also important to remember that the housed comparison group
in our study live in areas of high socioeconomic deprivation.
The difference in readmission risk between homeless individ-
uals and the general population is likely to be wider, because
socioeconomic deprivation is associated with morbidity and
readmission.22
As far as we are aware, ours is the largest study of the
outcomes of homeless patients after discharge from hospital
in the UK. National data regarding inpatients with ‘no fixed
abode’ show that the median age was 43, three- quarters were
men and 9 out of 10 were admitted in an emergency.23 Our
sample has similar characteristics, supporting generalisability
between our results and homeless patients discharged from
hospital nationally.
Table 1 Characteristics of hospital patient experiencing homelessness, compared with housed patients living in socioeconomically deprived areas
Variable Level Homeless, n (%) Housed, n (%)
Total 2772 (100) 2772 (100)
Age at index admission (matched) Mean (SD) 44.21 (14.15) 44.34 (14.56)
Median (IQR) 43.64 (33.37–53.75) 43.67 (33.05–53.51)
Sex (matched) Female 768 (28) 768 (28)
Male 2004 (72) 2004 (72)
Year of index admission (matched) 2013 76 (3) 76 (3)
2014 769 (28) 769 (28)
2015 948 (34) 948 (34)
2016 979 (35) 979 (35)
Number of comorbidities, based on prior hospital admissions
(ICD-10 chapters 2–14 and 17)
0 926 (33) 1238 (45)
1 769 (28) 784 (28)
2 541 (20) 424 (15)
3 307 (11) 183 (7)
4+ 229 (8) 143 (5)
Mean (SD) 1.38 (1.41) 1.02 (1.24)
Median (IQR) 1 (0–2) 1 (0–2)
ICD-10 chapter of index admission Accidents and other external 695 (25) 452 (16)
Digestive diseases 223 (8) 298 (11)
Circulatory diseases 226 (8) 228 (8)
Mental health 347 (13) 90 (3)
Respiratory diseases 189 (7) 236 (9)
Skin problems 206 (7) 106 (4)
Genitourinary diseases 89 (3) 210 (8)
Musculoskeletal problems 112 (4) 127 (5)
Infections 75 (3) 78 (3)
Cancers 63 (2) 80 (3)
Other* 547 (20) 867 (31)
Discharge method for index admission Self (without clinical consent) 253 (9) 100 (4)
With clinical consent 2519 (91) 2672 (96)
Years of follow- up after index admission 6753 6987
Emergency readmissions (rate per 1000 person- years) 12 472 (1847) 4525 (648)
Planned readmissions (rate per 1000 person- years) 3400 (503) 4769 (683)
A&E visits (rate per 1000 person- years) 43 808 (6487) 14 186 (2030)
Died during the study (%) 451 (16) 311 (11)
*‘Other’ includes ICD-10 chapters (II) diseases of the blood and blood- forming organs; (IV) endocrine, nutritional and metabolic diseases; (VI) diseases of the nervous system; (VII) diseases of
the eye; (VIII) diseases of the ear.
A&E, Accident and Emergency; ICD-10, 10th revision of the International Statistical Classification of Diseases and Related Health Problems.
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
5
LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
In our protocol18 we planned to analyse data from a ‘control’
group of homeless patients who were not seen by a specialist
discharge scheme and were instead identified through linkage
to an external service that supports people experiencing
homelessness. This control group was intended to allow esti-
mation of the effect of the specialist discharge schemes on
readmission rates. However, we subsequently found problems
with these data; most importantly that we could not confirm
whether patients were homeless at the point of hospital admis-
sion (unlike those included in this analysis, whom specialist
discharge teams identified as homeless). We therefore limited
the present analysis to a comparison of readmission for home-
less and housed patients. This means that our analysis primarily
provides insight into readmissions for homeless people after
hospital discharge, rather than an evaluation of the discharge
schemes.
We did not account for competing risks in our analysis. Death
is likely to be a competing risk for hospital readmission, on the
assumption that patients who died would have an increased
risk of readmission if they had survived. Homeless patients in
our study had a higher risk of death (16% of homeless patients
died, compared with 11% of housed patients) and the effect
of competing risks is therefore likely to be that rate ratios are
slightly understated. We did not account for competing risks to
maximise the simplicity of the analysis, and because competing
risks are unlikely to have an important bearing on the results.
Figure 2 Plot of hospital spells and A&E visits after discharge, with patients arranged in rows and ordered by follow- up duration. Dots indicate
readmission dates. Patients with no readmissions are represented as an empty (white) row. A&E, Accident and Emergency.
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
6LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
We found that homeless inpatients had a lower rate of planned
readmissions than housed inpatients. Our post- hoc visual inspec-
tion of the readmissions data (figure 2) suggested that the distri-
bution of planned readmissions may be different in the two
groups, with more ‘series’ of planned admissions for homeless
patients. We found that these series related to a small number of
individuals with large numbers of admissions, mainly for renal
failure and dialysis. After removing these individuals from both
groups, the difference between homeless and housed patients
was wider. This low rate of planned care reflects barriers that
have been observed in qualitative research,9–11 and suggests that
long- term conditions are often managed in crises. The relatively
large number of homeless patients with renal failure was not
expected and is an avenue for further research. It may represent
different patterns of healthcare use (for example dialysis being
undertaken by housed patients in other settings), or a higher risk
of renal failure related to use of drugs or alcohol, or cardiovas-
cular disease. The importance of renal failure in terms of the
quantity of healthcare would not have been identified from the
index admission alone and reflects the strengths of using longi-
tudinal data. Nephrologists in the USA have previously observed
the difficulty of providing dialysis to homeless patients with end-
stage renal disease.24
A range of intermediate care services has been developed in
England for older people. These services are not easily acces-
sible to homeless patients younger than 55 years who may
also be frail or in need of rehabilitation or palliative care.25 26
The effectiveness of these services can be limited by shortages
of longer term care, leading to intermediate care becoming
‘blocked’27 and contributing to bed shortages. In this context it
is unsurprising that people experiencing homelessness, many of
whom are middle- aged and do not meet eligibility criteria for
services for older people, struggle to access existing intermediate
care provision. Our results show a need for community ‘step-
down’ services that provide ongoing care. Observational studies
have found that step- down services are associated with reduced
readmissions,28–31 and a trial of GP- led management of discharge
in the UK found reduced street homelessness and improved
quality of life.32
Hospital patients who are experiencing homelessness have
extreme rates of emergency readmission after discharge,
reflecting poor housing and ongoing community care that is
designed around the needs of people who have stable housing.
Figure 3 The 12- month risk of readmission, stratified by the ICD-10 chapter of index admission. A&E, Accident and Emergency; ICD-10, 10th
revision of the International Statistical Classification of Diseases and Related Health Problems.
Table 2 Incidence rate ratios of readmissions and A&E visits,
comparing homeless patients with housed patients living in
socioeconomically deprived areas (results of negative binomial
regression)
Adjusted on matching
variables
Further adjusted for
comorbidities and ICD-10
chapter of index admission
Emergency
readmissions
2.92 (2.67–3.19) p<0.001 2.49 (2.29–2.70) p<0.001
Planned
readmissions
0.63 (0.55–0.72) p<0.001 0.60 (0.53–0.68) p<0.001
A&E visits 3.06 (2.86–3.27) p<0.001 2.57 (2.41–2.73) p<0.001
A&E, Accident and Emergency; ICD-10, 10th revision of the International Statistical
Classification of Diseases and Related Health Problems.
Box 1 Interpretation by an expert with lived experience
For many people who are street homeless, hospital is an
inhospitable, if not hostile environment. A single visit, or
even street lore alone, can be enough to cause one to make
inventive efforts to disguise one’s homelessness in order to
receive less visceral and judgemental handling. Certainly, it is
not an experience anyone would rush to embrace, hence the
(potentially fatal) avoidance and delay before seeking treatment.
Conversely, particularly for some of us with poor mental health,
a hospital represents a building with ‘indoor’ comforts and
facilities like heat, light and hot water and crucially, a place
populated by people who are perceived to have a duty to play
nicely. Perhaps this cohort of ‘regulars’ is partly responsible for
the medical profession’s distaste of us as a whole. In any case,
there are deep differences in how people who are homeless
approach and are received by healthcare services, as opposed
to those who are housed. This may explain the results under
Discussion. If hospital Trusts were to adopt a less ‘gatekeeping’
approach to homeless patients, try not to refer to us as
bed- blockers, at least not in our hearing, and provide timely
treatments for our multiple morbidities, the costs of our care
could be reduced dramatically. (JF)
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
7
LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
Ultimately, the inequality between housed and homeless people
can only be addressed by preventing homelessness (Box 1).
Author affiliations
1Institute of Health Informatics, University College London, London, UK
2Collaborative Centre for Inclusion Health, University College London, London, UK
3Institute of Epidemiology and Health Care, University College London, London, UK
4NIHR Policy Research Unit in Health and Social Care Workforce, King’s College
London, London, UK
5Community and Primary Care Research Group, University of Plymouth, Plymouth, UK
6Care Policy and Evaluation Centre, The London School of Economics and Political
Science, London, UK
7Alan Turing Institute, British Library, London, UK
8Pathway Charity, London, UK
9Department of Health Services Research and Policy, London School of Hygiene &
Tropical Medicine, London, UK
10National Addiction Centre, Institute of Psychiatry, Psychology & Neuroscience,
King’s College London, London, UK
11Find & Treat, University College London Hospitals NHS Foundation Trust, London,
UK
12Department of Nursing & Community Health, Glasgow Caledonian University,
Glasgow, UK
Twitter Dan Lewer @danlewer, Martin McKee @martinmckee and Robert Aldridge
@rob_aldridge
Contributors MCo, AH and RA—conceptualisation. DL, DM, RMB, HE and RA—
methodology. DL and DM—software. DL—formal analysis. DM, RMB, HE, AY and
RA—data curation. DL and JF—writing (original draft). DL, DM, MCo, RMB, RB,
MC, SD, HE, JF, NH, AK, SAL, JM, MM, JN, AS, MT, MW, FW, AY, AH and RA—writing
(review and editing). DL—visualisation. MCo, AH and RA—supervision. DM and
RA—project administration. MCo, AH and RA—funding acquisition.
Funding This study was supported by the National Institute for Health Research
(NIHR) (Project number: 13/156/10 to HS & DR). We also acknowledge the support
from the Health Data Research (HDR) UK which receives its funding from HDR UK
funded by the UK Medical Research Council, Engineering and Physical Sciences
Research Council, Economic and Social Research Council, Department of Health
and Social Care (England), Chief Scientist Office of the Scottish Government Health
and Social Care Directorates, Health and Social Care Research and Development
Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart
Foundation (BHF) and the Wellcome Trust. AH’s salary is provided by Central and
North West London NHS Community Trust. AS is funded by UCLH Foundation Trust.
DL is funded by an NIHR Doctoral Research Fellowship (DRF-2018-11- ST2-016).
JN is part- funded by the NIHR Biomedical Research Centre for Mental Health at
South London and Maudsley NHS Foundation Trust and King’s College London.
RMB is supported by a UK Research and Innovation Fellowship funded by a grant
from the Medical Research Council (MR/S003797/1). SL is funded by NIHR (ICA-
CDRF-2016-02-042). RB is supported by the NIHR Applied Research Collaboration
(ARC) South West Peninsula and JM is supported by the NIHR ARC South London.
RWA is supported by Wellcome through a Wellcome Clinical Research Career
Development Fellowship (206602). This article is based on independent research
commissioned and funded by the NIHR Health Service and Delivery Programme.
Disclaimer The views expressed in the publication are those of the authors and
not necessarily those of the NHS, the NIHR, the Wellcome Trust, the Department of
Health and Social Care, Public Health England or its arm’s length bodies or other
government departments.
Competing interests NH is medical director, and AH is a trustee of the charity
’Pathway: Healthcare for homeless people’. AS is clinical lead and manager for the
’Find and Treat’ service at University College London Hospitals.
Patient consent for publication Not required.
Ethics approval To undertake this study, collection of patient identifiers and
performed data linkage were performed without explicit consent from participants.
This research was therefore undertaken following approval (reference 16/CAG/0021)
from the Secretary of State for Health and Social Care through the Confidentiality
Advisory Group (CAG). Health Research Authority Research Ethics Committee
approval was also sought and received (REC 16/EE/0018). In addition, local R&D
approvals were obtained prior to local data collection at each of the Homeless
Hospital Discharge Fund sites.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data may be obtained from a third party and
are not publicly available. As part of the approvals and information governance
frameworks, we are unable to share the underlying data for this research study. Our
approval only allowed researchers involved in this specific project to access data for
the prespecified and approved analyses. Therefore, data collection and linkage would
have to be repeated with new approvals sought by anyone wanting access to the
underlying data used in this analysis. Application for access should be directed to the
CAG of the Health Research Authority. Information regarding the application process
and relevant links for applications are available from the CAG website.
Supplemental material This content has been supplied by the author(s). It
has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have
been peer- reviewed. Any opinions or recommendations discussed are solely those
of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and
responsibility arising from any reliance placed on the content. Where the content
includes any translated material, BMJ does not warrant the accuracy and reliability
of the translations (including but not limited to local regulations, clinical guidelines,
terminology, drug names and drug dosages), and is not responsible for any error
and/or omissions arising from translation and adaptation or otherwise.
Open access This is an open access article distributed in accordance with the
Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits
others to copy, redistribute, remix, transform and build upon this work for any
purpose, provided the original work is properly cited, a link to the licence is given,
and indication of whether changes were made. See: https:// creativecommons. org/
licenses/ by/ 4. 0/.
ORCID iDs
DanLewer http:// orcid. org/ 0000- 0003- 3698- 7196
MartinMcKee http:// orcid. org/ 0000- 0002- 0121- 9683
REFERENCES
1 Morrison DS. Homelessness as an independent risk factor for mortality: results from a
retrospective cohort study. Int J Epidemiol 2009;38:877–83.
2 Stenius- Ayoade A, Haaramo P, Kautiainen H, etal. Mortality and causes of death
among homeless in Finland: a 10- year follow- up study. J Epidemiol Community Health
2017;71:841–8.
3 Nordentoft M, Wandall- Holm N. 10 year follow up study of mortality among users of
hostels for homeless people in Copenhagen. BMJ 2003;327:81–0.
4 Beijer U, Andreasson S, Ågren G, etal. Mortality and causes of death among homeless
women and men in Stockholm. Scand J Public Health 2011;39:121–7.
5 Ministry of Housing, Communities & Local Government. Rough sleeping snapshot in
England: autumn 2019, 2020. Available: https://www. gov. uk/ government/ statistics/
rough- sleeping- snapshot- in- england- autumn- 2019 [Accessed 27 Oct 2020].
6 Iacobucci G. Homeless people’s A&E visits treble in seven years. BMJ 2019;364.
7 Link H. The unhealthy state of homelessness: health audit results 2014, 2014.
Available: https:// homeless. org. uk/ sites/ default/ files/ site- attachments/ The%
20unhealthy% 20state% 20of% 20homelessness% 20FINAL. pdf [Accessed 27 Oct
2020].
8 Doran KM, Ragins KT, Iacomacci AL, etal. The revolving Hospital door: Hospital
readmissions among patients who are homeless. Med Care 2013;51:767–73.
What is already known on this subject
Many people experiencing homelessness are discharged from
hospital without secure accommodation. Rough sleeping and
sofa- surfing are common during recuperation from an acute
illness. Homelessness is also associated with poor access to
community health and social care services. Studies in North
America show that homelessness is associated with high risk
of readmission.
What this study adds
Hospital inpatients in England who are experiencing
homelessness have a high risk of emergency readmission
and Accident and Emergency visits after discharge, when
compared with housed patients. Unlike housed patients,
the risk of readmission among homeless inpatients is high
regardless of the original cause of admission. Homeless
inpatients are less likely to have planned readmissions, which
may reflect poor ongoing care. Hospitals need specialist
discharge schemes that work alongside community support
services, including housing and social care.
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
8LewerD, etal. J Epidemiol Community Health 2021;0:1–8. doi:10.1136/jech-2020-215204
Original research
9 Wen CK, Hudak PL, Hwang SW. Homeless people’s perceptions of welcomeness and
unwelcomeness in healthcare encounters. J Gen Intern Med 2007;22:1011–7.
10 Gunner E, Chandan SK, Marwick S, etal. Provision and accessibility of primary
healthcare services for people who are homeless: a qualitative study of patient
perspectives in the UK. Br J Gen Pract 2019;69:e526–36.
11 Gelberg L, Browner CH, Lejano E, etal. Access to women’s health care: a qualitative
study of barriers perceived by homeless women. Women Health 2004;40:87–100.
12 Saab D, Nisenbaum R, Dhalla I, etal. Hospital readmissions in a community- based sample of
homeless adults: a matched- cohort study. J Gen Intern Med 2016;31:1011–8.
13 Buck DS, Brown CA, Mortensen K, etal. Comparing homeless and domiciled patients’
utilization of the Harris County, Texas public hospital system. J Health Care Poor
Underserved 2012;23:1660–70.
14 Titan A, Graham L, Rosen A, etal. Homeless status, Postdischarge health care
utilization, and readmission after surgery. Med Care 2018;56:460–9.
15 Khatana SAM, Wadhera RK, Choi E. Association of Homelessness with Hospital
Readmissions—an Analysis of Three Large States. J Gen Intern Med (Published Online
First: 17 June 2020).
16 Department of Health and Social Care. Homeless hospital discharge fund 2013 to
2014, 2013. Available: https://www. gov. uk/ government/ publications/ homeless-
hospital- discharge- fund- 2013- to- 2014 [Accessed 27 Oct 2020].
17 Aldridge RW, Menezes D, Lewer D, etal. Causes of death among homeless people: a
population- based cross- sectional study of linked hospitalisation and mortality data in
England. Wellcome Open Res 2019;4:49.
18 Blackburn RM, Hayward A, Cornes M, etal. Outcomes of specialist discharge
coordination and intermediate care schemes for patients who are homeless: analysis
protocol for a population- based historical cohort. BMJ Open 2017;7:e019282.
19 Ray D, Roebuck C, Smith O. Delivering linked datasets to support health and care delivery
and research, 2018. Available: https:// digital. nhs. uk/ binaries/ content/ assets/ website- assets/
services/ dars/ linked- datasets- in- nhs- digital- final. pdf [Accessed 27 Oct 2020].
20 Ministry of Housing, Communities & Local Government. English indices of deprivation
2015, 2015. Available: https://www. gov. uk/ government/ statistics/ english- indices- of-
deprivation- 2015 [Accessed 27 Oct 2020].
21 World Health Organisation. ICD-10. Version:2016. 2016.
22 Lyratzopoulos G, Havely D, Gemmell I, etal. Factors influencing emergency medical
readmission risk in a UK district general Hospital: a prospective study. BMC Emerg
Med 2005;5:1.
23 Office of the Chief Analyst. Healthcare for single homeless people, 2010. Available:
https:// webarchive. nationalarchives. gov. uk/ 20100713061219uo_/ http:// www. dh. gov.
uk/ prod_ consum_ dh/ groups/ dh_ digitalassets/@ dh/@ en/@ ps/ documents/ digitalasset/
dh_ 114369. pdf [Accessed 27 Oct 2020].
24 Novick TK, Gadegbeku CA, Crews DC. Dialysis for patients with end- stage renal
disease who are homeless. JAMA Intern Med 2018;178:1581.
25 Rogans- Watson R, Shulman C, Lewer D. Premature frailty, geriatric conditions
and multimorbidity among people experiencing homelessness: a cross- sectional
observational study in a London hostel. Hous Care Support 2020.
26 Cornes M, Whiteford M, Manthorpe J, etal. Improving hospital discharge
arrangements for people who are homeless: a realist synthesis of the intermediate
care literature. Health Soc Care Community 2018;26:e345–59.
27 et alBarton P, Stirling B, Glasby J. A national evaluation of the costs and outcomes of
intermediate care for older people, 2006. Available: https://www. birmingham. ac. uk/
Documents/ college- social- sciences/ social- policy/ HSMC/ research/ intermediate- care-
older- people. pdf [Accessed 27 Oct 2020].
28 Hewett N, Halligan A, Boyce T. A general practitioner and nurse led approach to
improving hospital care for homeless people. BMJ 2012;345:e5999.
29 Gazey A, Vallesi S, Martin K. The cottage: providing medical respite care in
a home- like environment for people experiencing homelessness. Hous Care
Support
2019;22:54–64.
30 Buchanan D, Doblin B, Sai T, etal. The effects of respite care for homeless patients: a
cohort study. Am J Public Health 2006;96:1278–81.
31 Doran KM, Ragins KT, Gross CP, etal. Medical respite programs for homeless patients:
a systematic review. J Health Care Poor Underserved 2013;24:499–524.
32 Hewett N, Buchman P, Musariri J. Randomised controlled trial of GP- led in- hospital
management of homeless people (’Pathway). Clin Med 2016;16:223–9.
on January 6, 2021 by guest. Protected by copyright.http://jech.bmj.com/J Epidemiol Community Health: first published as 10.1136/jech-2020-215204 on 5 January 2021. Downloaded from
... Problems arising from homelessness was an emergent finding from our research. According to Lewer et al. (2021), the UK's homeless patients are more than twice as likely to be admitted to hospital via acute care than those with stable housing, and twice as likely to present to ED than housed patients. Repeated presentations are not, according to Lewer et al, necessarily due to having more health problems, but associated with unmanaged chronic conditions. ...
... This patient group often only present to services at a late stage, and often reluctantly, and, as reported by our acute clinical stakeholders, can discharge themselves before any follow-up care can be organized. Lewer et al. (2021) and our informants suggest this is because they have experienced stigmatised attitudes on previous visits. van Boekel et al. (2013) highlight that health professionals' stigmatizing attitudes can stem from perceived threat of violence and poor motivation, and lack of empathy. ...
... As older people live with more complex health problems, including disabilities, chronic diseases, and geriatric symptoms [21], this adds a further layer of disadvantage to this demographic who may find it more challenging to access the specialist health services they require, both economically and socially, for best-practice preventive or chronic disease management [22,23]. Consequently, their health management is often crisis based, 'managed' by frequent unplanned attendances at after-hours medical care and hospital EDs which can result in potentially avoidable ward admissions [8,9,24], adding to the mortality gap between those housed and those experiencing homelessness due to inadequate healthcare [6,25]. ...
Article
Full-text available
People experiencing, or at risk of, homelessness face challenges that result in poorer health outcomes compared to those in stable housing. This study provides the results of over 40 health measures that capture the health status of a group of people in temporary accommodation due to experiencing homelessness, aged 22 to 84 years, in an inner-city location, invited to participate in a comprehensive assessment of physical and psychological health. Evidence of accelerated ageing was found, with 44.2% of people being clinically frail, 63% having poor functional movement, and 36% having pain associated with oral health. Additionally, 90.6% of participants showed health risks due to nutritional deficiencies, over half reported poor sleep quality, 55.3% reported having psychological distress, and almost half reported fair or poor overall dental health. This study suggests a pathway to providing a relatively easily implemented series of health assessments to help respond to a group of underlying causes for accelerated ageing among a group of inner-city people experiencing homelessness. This work can be used to inform the prioritisation and development of community-based health services to address functional deficits that may contribute to accelerated ageing.
... Given that there was an absence of A&E re-admissions, it was assumed that this rate was 3.244 over 18 months. 19 For patients who did not use dental services, they had 1.32 higher odds for A&E admission. 20 ...
Article
Full-text available
Aim The study aims to conduct economic evaluation of the Peninsula Dental Social Enterprise (PDSE) programme for people experiencing homelessness over an 18-month period, when compared to a hypothetical base-case scenario (‘status quo'). Methods A decision tree model was generated in TreeAge Pro Healthcare 2024. Benefit-cost analysis and cost-effectiveness analysis were performed using data informed by the literature and probabilistic sensitivity analysis (Monte Carlo simulation with 1,000 cycles). The pre-determined willingness-to-pay threshold was estimated to be £59,502 per disability-adjusted life year (DALY) averted. Costs (£) and benefits were valued in 2020 prices. Health benefits in DALYs included dental treatment for dental caries, periodontitis and severe tooth loss. Results The hypothetical cohort of 89 patients costs £11,502 (SD: 488) and £57,118 (SD: 2,784) for the base-scenario and the PDSE programme, respectively. The health outcomes generated 0.9 (SD: 0.2) DALYs averted for the base-case scenario, and 5.4 (SD: 0.9) DALYs averted for the PDSE programme. The DALYs averted generated £26,648 (SD: 4,805) and £163,910 (SD: 28,542) in benefits for the base-scenario and the PDSE programme, respectively. The calculated incremental benefit-cost ratio was 3.02 (SD: 0.5) and incremental cost-effectiveness ratio was £10,472 (SD: 2,073) per DALY averted. Uncertainty analysis demonstrated that the PDSE programme was 100% cost-effective. Conclusions Funding a targeted dental programme from the UK healthcare perspective that provides timely and affordable access to dental services for people experiencing homelessness is cost-effective.
... As older people live with more complex health problems, including disabilities, chronic diseases, and geriatric symptoms [17], this adds a further layer of disadvantage to this demographic who may find it more challenging to access the specialist health services they require, both economically and socially, for best-practice preventive or chronic disease management [18,19]. Consequently, their health management is often crisis based, 'managed' by frequent unplanned attendances at after-hours medical care and hospital EDs which can result in potentially avoidable ward admissions [7,8,20], adding to the mortality gap between those housed and those experiencing homelessness due to inadequate healthcare [21]. ...
Preprint
Full-text available
People experiencing homelessness face challenges that result in poorer health outcomes compared to those in stable housing. This study provides the results from over 40 health measures that capture the general physical and mental health of a group of people experiencing homelessness aged 22 to 84 years, in an inner-city location, invited to participate in a comprehensive assessment of physical and psychological health. We found evidence of accelerated ageing, with 44.2% of people being clinically frail, 63% had poor functional movement, and 36% had pain associated with oral health. Additionally, 90.6% of participants showed health risks due to nutrition deficiencies, over half reported poor sleep quality, 55.3% reported having psychological distress, and almost half reporting fair or poor overall dental health. This study presents a path to providing a relatively easily implemented group of health assessments to help respond to a group of underlying causes for accelerated ageing, among a group of inner-city people experiencing homelessness. This work can be used to inform the prioritisation and development of community-based health services to address functional deficits that may contribute to accelerated ageing.
... Studies examining the social gradient of health reveal that individuals experiencing rough sleeping face significantly poorer health outcomes than those who are sheltered (Nagy-Borsy et al., 2021;Ra et al., 2023). Providing shelter can enhance access to healthcare and increase the utilization of nonurgent care services (Trummer et al., 2020), addressing the issue of neglected health prevention and early treatment that leads to the inefficient use of healthcare services (Davies and Wood, 2018;Lewer et al., 2021;Reilly et al., 2022). ...
... Among the 14 research papers (Table 2), nine papers were qualitative studies (Armstrong et al., 2021;Bell et al., 2022;Csikar et al., 2019;Gunner et al., 2019;Jagpal et al., 2020;Mc Conalogue et al., 2021;Paisi et al., 2020;Ungpakorn & Rae, 2020;Warren et al., 2021), three were quantitative studies (Elwell-Sutton et al., 2017;Field et al., 2019;Lewer et al., 2021), and two were mixed methods studies (Dawes et al., 2017;Heaslip et al., 2022). The qualitative studies explored access and utilization of services among PEH related to general health and social care, primary health care services, stakeholders' engagement in PEH services, nursing services, oral and dental health services, health-related street outreach services, mental health and substance misuse, and nurse-led homeless health services. ...
... Given that there was an absence of A&E re-admissions, it was assumed that this rate was 3.244 over 18 months. 19 For patients who did not utilise dental services, they had 1.32 higher odds for A&E admission. 20 ...
Preprint
Full-text available
Aim The study aims to conduct economic evaluation of the Peninsula Dental Social Enterprise (PDSE) programme for people experiencing homelessness over an 18-month period, when compared to a hypothetical base-case scenario (‘status quo’). Methods A decision tree model was generated in Treeage Pro Healthcare 2024. Benefit-cost analysis and cost-effectiveness analysis were performed using data informed by the literature and probabilistic sensitivity analysis (Monte-Carlo simulation with 1,000 cycles). The predetermined willingness to pay threshold was estimated to be £59,502 per disability-adjusted life year (DALY) averted. Costs (£) and benefits were valued in 2020 prices. Health benefits in DALYs included dental treatment for dental caries, periodontitis and severe tooth loss. Results The hypothetical cohort of 89 patients costs £11,502 (SD 488) and £57,118 (SD 2,784) for the base-scenario and the PDSE programme, respectively. The health outcomes generated 0.9 (SD 0.2) DALYs averted for the base-case scenario, and 5.4 (SD 0.9) DALYs averted for the PDSE programme. The DALYs averted generated £26,648 (SD 4,805) and £163,910 (SD 28,542) in benefits for the base-scenario and the PDSE programme, respectively. The calculated incremental benefit-cost ratio was 3.02 (SD 0.5) and incremental cost-effectiveness ratio was £10,472 (SD 2,073) per DALY averted. Uncertainty analysis demonstrated that the PDSE programme was 100% cost-effective. Conclusions Funding a targeted dental programme from the UK healthcare perspective that provides timely and affordable access to dental services for people experiencing homelessness is cost-effective.
... 4 Yearly rates of admission to accident and emergency departments of a hospital are up to fourfold higher in PEH when compared with the general population. [5][6][7] Within this, many experience trimorbidity: the co-occurrence of mental ill health, substance use disorder and physical illness 7 8 often on a background of complex trauma and adverse childhood experiences. This can result in a complex presentation that requires trauma-informed, multidisciplinary support. ...
Article
Full-text available
Background Due to the recognition that people experiencing homelessness (PEH) often die young and unsupported, a growing number of initiatives focusing on palliative care and homelessness are emerging across the UK. However, there has been no systematic exploration of the nature and landscape of this work. Aims To understand the range, aims, successes and challenges of current initiatives within the field of palliative care and homelessness in the UK, by exploring existing projects and initiatives. Method An online survey was distributed to members of an Extension for Community Healthcare Outcomes network focusing on palliative care and homelessness for a mixed professional audience. The survey collated the aims, successes and challenges of initiatives aiming to improve palliative care for PEH. Responses were summarised using descriptive statistics, and free-text responses were analysed using thematic analysis. Results 162 professionals completed the survey. Of these, 62% reported involvement in at least one palliative care and homelessness initiative. Initiatives focused on service delivery (59%), training (28%) and research (28%). Themes for success included improved service engagement, relationship formation, housing provision, honouring end-of-life wishes, upskilling staff and enabling safe hospital discharge. The main challenges included stigma around substance misuse, securing funding, staff capacity, equipment and facilities, and engaging communities. Conclusion The number and scope of initiatives aiming to support PEH with advanced ill health and palliative care needs across the UK is growing, with a range of professionals engaging in the field. Future research may benefit from exploring initiatives in more detail to understand the specific drivers of impact on PEH and the staff and services supporting them.
Article
Full-text available
Background There is a high prevalence of health problems among single people who are homeless. Specialist primary health care services for this population have been developed in several locations across England; however, there have been very few evaluations of these services. Objectives This study evaluated the work of different models of primary health care provision in England to determine their effectiveness in engaging people who are homeless in health care and in providing continuity of care for long-term conditions. It concerned single people (not families or couples with dependent children) staying in hostels, other temporary accommodation or on the streets. The influence on outcomes of contextual factors and mechanisms (service delivery factors), including integration with other services, were examined. Data from medical records were collated on participants’ use of health care and social care services over 12 months, and costs were calculated. Design and setting The evaluation involved four existing Health Service Models: (1) health centres primarily for people who are homeless (Dedicated Centres), (2) Mobile Teams providing health care in hostels and day centres, (3) Specialist GPs providing some services exclusively for patients who are homeless and (4) Usual Care GPs providing no special services for people who are homeless (as a comparison). Two Case Study Sites were recruited for each of the specialist models, and four for the Usual Care GP model. Participants People who had been homeless during the previous 12 months were recruited as ‘case study participants’; they were interviewed at baseline and at 4 and 8 months, and information was collected about their circumstances and their health and service use in the preceding 4 months. Overall, 363 participants were recruited; medical records were obtained for 349 participants. Interviews were conducted with 65 Case Study Site staff and sessional workers, and 81 service providers and stakeholders. Results The primary outcome was the extent of health screening for body mass index, mental health, alcohol use, tuberculosis, smoking and hepatitis A among participants, and evidence of an intervention if a problem was identified. There were no overall differences in screening between the models apart from Mobile Teams, which scored considerably lower. Dedicated Centres and Specialist GPs were more successful in providing continuity of care for participants with depression and alcohol and drug problems. Service use and costs were significantly higher for Dedicated Centre participants and lower for Usual Care GP participants. Participants and staff welcomed flexible and tailored approaches to care, and related services being available in the same building. Across all models, dental needs were unaddressed and staff reported poor availability of mental health services. Limitations There were difficulties recruiting mainstream general practices for the Usual Care GP model. Medical records could not be accessed for 14 participants of this model. Conclusions Participant characteristics, contextual factors and mechanisms were influential in determining outcomes. Overall, outcomes for Dedicated Centres and for one of the Specialist GP sites were relatively favourable. They had dedicated staff for patients who were homeless, ‘drop-in’ services, on-site mental health and substance misuse services, and worked closely with hospitals and homelessness sector services. Funding This project was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (HSDR 13/156/03) and will be published in full in Health and Social Care Delivery Research ; Vol. 11, No. 16. See the NIHR Journals Library website for further project information.
Article
Full-text available
Purpose The purpose of this paper is to assess frailty, geriatric conditions and multimorbidity in people experiencing homelessness (PEH) using holistic evaluations based on comprehensive geriatric assessment (CGA) and draw comparisons with general population survey data. Design/methodology/approach Cross-sectional observational study conducted in a London-based hostel for single PEH over 30 years old in March–April 2019. The participants and key workers completed health-related questionnaires, and geriatric conditions were identified using standardised assessments. Frailty was defined according to five criteria in Fried’s phenotype model and multimorbidity as the presence of two or more long-term conditions (LTCs). Comparisons with the general population were made using data from the English Longitudinal Study of Ageing and the Health Survey for England. Findings A total of 33 people participated with a mean age of 55.7 years (range 38–74). Frailty was identified in 55% and pre-frailty in 39%. Participants met an average of 2.6/5 frailty criteria, comparable to 89-year-olds in the general population. The most common geriatric conditions were: falls (in 61%), visual impairment (61%), low grip strength (61%), mobility impairment (52%) and cognitive impairment (45%). All participants had multimorbidity. The average of 7.2 LTCs (range 2–14) per study participant far exceeds the average for even the oldest people in the general population. Originality/value To the best of authors’ knowledge, this is the first UK-based study measuring frailty and geriatric conditions in PEH and the first anywhere to do so within a CGA-type evaluation. It also demonstrates the feasibility of conducting holistic evaluations in this setting, which may be used clinically to improve the health outcomes for PEH.
Article
Full-text available
Background : Homelessness has increased by 165% since 2010 in England, with evidence from many settings that those affected experience high levels of mortality. In this paper we examine the contribution of different causes of death to overall mortality in homeless people recently admitted to hospitals in England with specialist integrated homeless health and care (SIHHC) schemes. Methods : We undertook an analysis of linked hospital admission records and mortality data for people attending any one of 17 SIHHC schemes between 1st November 2013 and 30th November 2016. Our primary outcome was death, which we analysed in subgroups of 10th version international classification of disease (ICD-10) specific deaths; and deaths from amenable causes. We compared our results to a sample of people living in areas of high social deprivation (IMD5 group). Results : We collected data on 3,882 individual homeless hospital admissions that were linked to 600 deaths. The median age of death was 51.6 years (interquartile range 42.7-60.2) for SIHHC and 71.5 for the IMD5 (60.67-79.0). The top three underlying causes of death by ICD-10 chapter in the SIHHC group were external causes of death (21.7%; 130/600), cancer (19.0%; 114/600) and digestive disease (19.0%; 114/600). The percentage of deaths due to an amenable cause after age and sex weighting was 30.2% in the homeless SIHHC group (181/600) compared to 23.0% in the IMD5 group (578/2,512). Conclusion : Nearly one in three homeless deaths were due to causes amenable to timely and effective health care. The high burden of amenable deaths highlights the extreme health harms of homelessness and the need for greater emphasis on prevention of homelessness and early healthcare interventions.
Article
Full-text available
Purpose Co-existing health conditions and frequent hospital usage are pervasive in homeless populations. Without a home to be discharged to, appropriate discharge care and treatment compliance are difficult. The Medical Respite Centre (MRC) model has gained traction in the USA, but other international examples are scant. The purpose of this paper is to address this void, presenting findings from an evaluation of The Cottage, a small short-stay respite facility for people experiencing homelessness attached to an inner-city hospital in Melbourne, Australia. Design/methodology/approach This mixed methods study uses case studies, qualitative interview data and hospital administrative data for clients admitted to The Cottage in 2015. Hospital inpatient admissions and emergency department presentations were compared for the 12-month period pre- and post-The Cottage. Findings Clients had multiple health conditions, often compounded by social isolation and homelessness or precarious housing. Qualitative data and case studies illustrate how The Cottage couples medical care and support in a home-like environment. The average stay was 8.8 days. There was a 7 per cent reduction in the number of unplanned inpatient days in the 12-months post support. Research limitations/implications The paper has some limitations including small sample size, data from one hospital only and lack of information on other services accessed by clients (e.g. housing support) limit attribution of causality. Social implications MRCs provide a safe environment for individuals to recuperate at a much lower cost than inpatient admissions. Originality/value There is limited evidence on the MRC model of care outside of the USA, and the findings demonstrate the benefits of even shorter-term respite post-discharge for people who are homeless.
Article
Full-text available
Introduction People who are homeless often experience poor hospital discharge arrangements, reflecting ongoing care and housing needs. Specialist integrated homeless health and care provision (SIHHC) schemes have been developed and implemented to facilitate the safe and timely discharge of homeless patients from hospital. Our study aims to investigate the health outcomes of patients who were homeless and seen by a selection of SIHHC services. Methods and analysis Our study will employ a historical population-based cohort in England. We will examine health outcomes among three groups of adults: (1) homeless patients seen by specialist discharge schemes during their hospital admission; (2) homeless patients not seen by a specialist scheme and (3) admitted patients who live in deprived neighbourhoods and were not recorded as being homeless. Primary outcomes will be: time from discharge to next hospital inpatient admission; time from discharge to next accident and emergency attendance and 28-day emergency readmission. Outcome data will be generated through linkage to hospital admissions data (Hospital Episode Statistics) and mortality data for November 2013 to November 2016. Multivariable regression will be used to model the relationship between the study comparison groups and each of the outcomes. Ethics and dissemination Approval has been obtained from the National Health Service (NHS) Confidentiality Advisory Group (reference 16/CAG/0021) to undertake this work using unconsented identifiable data. Health Research Authority Research Ethics approval (REC 16/EE/0018) has been obtained in addition to local research and development approvals for data collection at NHS sites. We will feedback the results of our study to our advisory group of people who have lived experience of homelessness and seek their suggestions on ways to improve or take this work further for their benefit. We will disseminate our findings to SIHHC schemes through a series of regional workshops.
Article
Full-text available
Background: Homelessness is associated with increased mortality, and some predictors of mortality have been previously identified. We examined the overall and cause-specific mortality among homeless men in Helsinki and the associations of social background and health service use with mortality. Methods: To assess cause-specific mortality in a competing risks framework, we performed a register-based, case-control study of 617 homeless men and an age-matched control group of 1240 men from the general population that were followed for 10 years between 2004 and 2014. Cox proportional hazards model was used to calculate HR for death and a competing risks model to calculate sub-HRs (sHR) for cause-specific death. Results: During the follow-up, 45.0% of the homeless died compared with 10.5% of controls (HR 5.38, 95% CI 4.39 to 6.59). The risk of death was particularly elevated for the homeless aged≤50 years (HR 10.3, 95% CI 7.0 to 15.2). External causes caused 34% of the deaths (sHR 11.2, 95% CI 6.8 to 18.2), but also deaths from medical causes were common (sHR 3.6, 95% CI 2.9 to 4.6). Age and somatic hospitalisation were significant predictors of death both among homeless and controls. Educational attainment, marital status, employment and psychiatric hospitalisation were significant predictors of mortality among the controls, but not among the homeless. Conclusions: Homelessness is associated with a fivefold mortality compared with the controls, and especially homeless aged ≤50 years have an increased risk of death. Being homeless eliminates the protective effects of marriage, employment and education on mortality risk.
Article
Background Individuals experiencing homelessness have higher hospitalization and mortality rates compared with the housed. Whether they also experience higher readmission rates, and if readmissions vary by region or cause of hospitalization is unknown.Objective Evaluate the association of homelessness with readmission rates across multiple US states.DesignRetrospective analysis of administrative claimsPatientsAll inpatient hospitalizations in Florida, Massachusetts, and New York from January 2010 to October 2015Main MeasuresThirty- and 90-day readmission ratesKey ResultsOut of a total of 23,103,125 index hospitalizations, 515,737 were for patients who were identified as homeless at the time of discharge. After adjusting for cause of index hospitalization, state, demographics, and clinical comorbidities, 30-day and 90-day readmission rates were higher for index hospitalizations in the homeless compared with those in the housed group. The difference in 30-day readmission rates between homeless and housed groups was the largest in Florida (30.4% vs. 19.3%; p < 0.001), followed by Massachusetts (23.5% vs. 15.2%; p < 0.001) and New York (15.7% vs. 13.4%; p < 0.001) (combined 17.3% vs. 14.0%; p < 0.001). Among the most common causes of hospitalization, 30-day readmission rates were 4.1 percentage points higher for the homeless group for mental illness, 4.9 percentage points higher for diseases of the circulatory system, and 2.4 percentage points higher for diseases of the digestive system.Conclusions After adjusting for demographic and clinical characteristics, homelessness is associated with significantly higher 30- and 90-day readmission rates, with a significant variation across the three states. Interventions to reduce the burden of readmissions among individuals experiencing homelessness are urgently needed. Differences across states point to the potential of certain public policies to impact health outcomes for individuals experiencing homelessness.
Article
Background: Anecdotal reports of people who are homeless being denied access and facing negative experiences of primary health care have often emerged. However, there is a dearth of research exploring this population's views and experiences of such services. Aim: To explore the perspectives of individuals who are homeless on the provision and accessibility of primary healthcare services. Design and setting: A qualitative study with individuals who are homeless recruited from three homeless shelters and a specialist primary healthcare centre for the homeless in the West Midlands, England. Method: Semi-structured interviews were audiorecorded, transcribed verbatim, and analysed using a thematic framework approach. The Theoretical Domains Framework (TDF) was used to map the identified barriers in framework analysis. Results: A total of 22 people who were homeless were recruited. Although some participants described facing no barriers, accounts of being denied registration at general practices and being discharged from hospital onto the streets with no access or referral to primary care providers were described. Services offering support to those with substance misuse issues and mental health problems were deemed to be excluding those with the greatest need. A participant described committing crimes with the intention of going to prison to access health care. High satisfaction was expressed by participants about their experiences at the specialist primary healthcare centre for people who are homeless (SPHCPH). Conclusion: Participants perceived inequality in access, and mostly faced negative experiences, in their use of mainstream services. Changes are imperative to facilitate access to primary health care, improve patient experiences of mainstream services, and to share best practices identified by participants at the SPHCPH.
Article
End-stage renal disease (ESRD) affects more than 700 000 individuals in the United States and accounts for $33 billion in annual Medicare spending.¹ Since 1973, the Medicare program has provided health coverage for most individuals with ESRD regardless of their age. About 34% of people with newly diagnosed ESRD live in areas where more than 1 in 5 households are below the federal poverty level.² Although the prevalence of homelessness among patients with ESRD is unknown, such patients would benefit from safe and stable housing. Homelessness imposes substantial barriers to good medical care.³ We highlight the challenges faced by patients with ESRD who are homeless, recommend measures to determine the prevalence and outcomes of homelessness among Medicare beneficiaries with ESRD, and propose approaches, including a federal demonstration project, to potentially improve outcomes through supportive housing.
Article
Introduction: Homeless Veterans are vulnerable to poor care transitions, yet little research has examined their risk of readmission following inpatient surgery. This study investigates the predictors of surgical readmission among homeless relative to housed Veteran patients. Methods: Inpatient general, vascular, and orthopedic surgeries occurring in the Veterans Health Administration from 2008 to 2014 were identified. Administrative International Classification of Diseases, Ninth Revision, Clinical Modification codes and Veterans Health Administration clinic stops were used to identify homeless patients. Bivariate analyses examined characteristics and predictors of readmission among homeless patients. Multivariate logistic models were used to estimate the association between homeless experience and housed patients with readmission following surgery. Results: Our study included 232,373 surgeries: 43% orthopedic, 39% general, and 18% vascular with 5068 performed on homeless patients. Homeless individuals were younger (56 vs. 64 y, P<0.01), more likely to have a psychiatric comorbidities (51.3% vs. 19.4%, P<0.01) and less likely to have other medical comorbidities such as hypertension (57.1% vs. 70.8%, P<0.01). Homeless individuals were more likely to be readmitted [odds ratio (OR), 1.43; confidence interval (CI), 1.30-1.56; P<0.001]. Discharge destination other than community (OR, 0.57; CI, 0.44-0.74; P<0.001), recent alcohol abuse (OR, 1.45; CI, 1.15-1.84; P<0.01), and elevated American Society Anesthesiologists classification (OR, 1.86; CI, 1.30-2.68; P<0.01) were significant risk factors associated with readmissions within the homeless cohort. Conclusions: Readmissions are higher in homeless individuals discharged to the community after surgery. Judicious use of postoperative nursing or residential rehabilitation programs may be effective in reducing readmission and improving care transitions among these vulnerable Veterans. Relative costs and benefits of alternatives to community discharge merit investigation.