Transcript Social care costs

How Obesity Affects the Need
for Social Care: Preliminary
Findings from HSE
Vicky Copley, PHE Risk Factor Intelligence
Outline
•
Background
•
Data
•
Methods
•
Results
•
Summary of work so far
•
Future considerations
2
How obesity affects the need for social care
PHE Adult Weight Management Tool
•
Version 1 launched April 2014
•
Designed to help the public health community make informed decisions
about the commissioning of obesity interventions by providing a practical
assessment of costs and benefits.
•
Uses population-level data to estimate the health impact of weight loss in
any group of people who have participated in an intervention or programme.
•
Designed to be used to compare the costs and benefits of interventions for
which there are outcome data, as well as to test theoretical or planned
programmes where data are absent.
•
Developed with an expert advisory group and tested with practitioners
•
Available to download from
http://www.noo.org.uk/visualisation/economic_assessment_tool
3
How obesity affects the need for social care
Simple model of obesity and health
Impact on mortality and
disease:
Study
population
BMI
•
•
•
•
•
diabetes
heart disease
stroke
colorectal cancer
breast cancer
healthcare
costs
4
How obesity affects the need for social care
Modified by an intervention
5
How obesity affects the need for social care
Feedback
•
Post-launch feedback mainly minor technical issues
•
Some more fundamental challenges (“why aren’t any of my programmes
cost-effective?”)
•
Survey conducted in October 2014
•
Wants:
• Young people
• Clearer assumptions
• Social care costs
6
How obesity affects the need for social care
Possible approaches to social care costs
•
•
Published estimates of social care costs
•
By BMI
•
By disease category
Estimate our own
•
7
Health Survey for England (HSE)
How obesity affects the need for social care
Data
• Health Survey for England (HSE)
• Has included social care questions for people aged 65 and over since 2011
• Pooled three years of data 2011-2013
• HSE does not cover care homes
• Obtains self-reported need for, and receipt of, social care
• Activities of Daily Living (ADLs)
• Instrumental Activities of Daily Living (IADLs)
• Height and weight measured during interviewer visit
• Other variables which may have independent association with need for social
care
• Limiting long term illness
• Age
• Sex
• Ethnicity
• Deprivation
• etc.
8
• How
etc.obesity affects the need for social care
ADLs and IADLs
9
How obesity affects the need for social care
Sources of help
•
Informal
• Formal
• Spouse / partner
• Home care worker*
• Family member
• 2nd homecare worker*
• Friend
• 3rd homecare worker*
• etc
• Reablement team*
• Occupational therapist/physiotherapist
• Voluntary helper
• Warden/sheltered housing manager*
• Cleaner
• Council handyman*
* denotes potential LA-funded help
10
How obesity affects the need for social care
Some
descriptive
statistics
11
Unweighted
count
Body mass index (kg/m2)
<18.5
18-25
25-30
30-40
40+
Missing and unreliable
Limiting long term illness
No
Yes
Sex
Female
Male
Age (years)
65-69
70-74
75-79
80-84
85+
IMD quintile
1 least deprived
2
3
4
5 most deprived
How obesity affects the need for social care
% or % of those with self-reported
mean need for social care (95% CI)
52
1328
2152
1405
108
1417
0.8
20.8
33.2
21.7
1.7
21.8
0.8 (0.3 to 1.3)
13.8 (11.9 to 15.7)
19.5 (17.5 to 21.5)
21.3 (19.1 to 23.5)
2.5 (1.7 to 3.3)
42.1 (39.4 to 44.9)
3632
2830
55.7
44.3
12.2 (10.4 to 13.9)
87.8 (86.1 to 89.6)
3500
2962
45.3
54.7
62.9 (60.9 to 65.0)
37.1 (35.0 to 39.1)
2004
1609
1262
904
683
30.4
24.3
20.0
14.4
11.0
16.9 (14.9 to 18.8)
18.3 (16.3 to 20.3)
19.5 (17.4 to 21.7)
21.5 (19.4 to 23.6)
23.8 (21.5 to 26.1)
1474
1592
1405
1105
886
22.6
24.9
21.9
17.1
13.4
15.4 (13.2 to 17.6)
22.4 (20.0 to 24.9)
21.4 (19.1 to 23.8)
20.8 (18.5 to 23.2)
19.9 (17.4 to 22.4)
Method
•
Two-step approach
1. model the probability of self-reported need for social care
•
•
•
univariable regression
multivariable regression with adjustment for all potential determinants excluding
limiting long term illness
multivariable regression with adjustment for all potential determinants including
limiting long term illness
2. combine model-predicted probability of need for social care with the mean hours
of help received
-> calculate the expected hours of help received per adult
12
How obesity affects the need for social care
Results
• 6462 observations of adults aged 65 or over available for analysis
•
Considered missing values as separate category in categorical variables
•
Imputed missing height and weight and used continuous BMI in multivariable
regression models
• Conditional on receiving at least some help, the mean hours of care received
per week, per person is
• 20.9 hours from informal sources
• 1.6 from formal local authority sources
• 22.4 hours in total
13
How obesity affects the need for social care
Results from multivariable models
14
How obesity affects the need for social care
Expected hours of care by BMI
Expected hours of community-based care per week, per person by source of care
and BMI in population aged 65 and over
15
How obesity affects the need for social care
Add in cost
The expected hours of help received can be combined with the average hourly
cost of a social care worker to provide estimates of the cost of social care
by BMI for use in economic models.
The total cost of providing a home care worker in England is approximately £24
per hour and on this basis, using model 2, the annual cost of local authority
funded social care for an individual with a BMI of 40 kg/m2 would be £653
(£24*52 weeks*0.52 hours per week).
The same cost for an individual with a BMI of 18 kg/m2 is £354.
For a local authority with 30,000 adults aged over 65 with a morbid obesity
prevalence of 2.1% this equates to an annual excess social care cost of
£188,000.
Corresponding annual costs estimated using model 1 are £948 and £415 for
BMIs of 40 and 18 respectively, equivalent to an annual excess social care
cost to the local authority of £336,000.
16
How obesity affects the need for social care
Summary
•
Social care costs make up an important part of the overall cost to society of
obesity
•
There is a paucity of robust UK data on the association between BMI and
social care usage
•
Our approach uses a transparent methodological approach combined with
Health Survey for England data to provide an estimate of community-based
social care costs associated with obesity in the over 65s
• Adjusted for other factors which affect social care need
•
Estimates will be incorporated into the PHE cost effectiveness tool
•
This is a first attempt at estimating these costs…
• Quite conservative
• Only over 65s
• Non-care home population
17
How obesity affects the need for social care
To consider
•
Is definition of formal care appropriate
• e.g. should occupational therapists be included in definition
•
Other sources of data
• care home population
• population aged under 65
•
Model might consider wider cost to society
•
18
Total hours of care rather than LA-funded care
How obesity affects the need for social care