Internal 2010 - Patient Satisfaction with Primary Care

The GP Patient Survey (GPPS)
Analysis
Resuls
Summary
Patient Satisfaction with Primary Care
An analysis of the 2007-08 GP patient survey data
Evangelos Kontopantelis1 Martin Roland2
David Reeves1
1National Primary Care Research and Development Centre
University of Manchester
2General Practice and Primary Care Research Unit
University of Cambridge
NPCRDC Internal Seminar, 9th March 2010
Kontopantelis GP patient satisfaction

Analysis
Resuls
Summary
Outline
1 The GP Patient Survey
(GPPS)
General Information
2 Analysis
Descriptive statistics
Objectives
The data
The method
3 Resuls
Model A
Model B
Model C
Overview
4 Summary
Conclusions
Discussion

Analysis
Resuls
Summary
General Information
Background
Annual national survey, starting in
2006/07.
Set out to measure patient
experience/satisfaction with access
to their GP.
Ipsos MORI conducted the survey
on behalf of the DoH.
Questionnaire mailed directly to
patients randomly selected from
practices’ registered lists.
Practicipating practices were/are
rewarded, based on the outcome of
the survey.

Analysis
Resuls
Summary
General Information
The second year: 2007/08
DoH budgeted around
£10million.
8,307 English general
practices were included (of
8,403 identiﬁed, participation
of 98.9%).
A total of 1,999,523 valid
responses was received from
a total sample of 5 million
people, a response rate of
41%.

Analysis
Resuls
Summary
General Information
The main outcomes
Five main outcomes on experience and opinions of the
service patients received from their general practice, over
the last 6 months:
Satisfaction with getting through to someone on the phone.
Ability to get an appointment on the same day or on the
next 2 days the surgery was open.
Ability to get an appointment more than 2 full days in
advance.
Ability to make an appointment with a particular doctor
(even if it meant waiting for longer).
Satisfaction with the hours the surgery was open.

Analysis
Resuls
Summary
Objectives
The data
The method
Patient information
Numbers of missing values
were low (1-4%).
57% of respondents were
female.
52% aged 55 or over.
78% described themselves
as White British.
40% of respondents
reported at least one
long-term condition.
6%
3%
6%
7%
Ethnicity
White British Other White Black Asian Other
78%
6%
3%
6%
7%
Ethnicity
White British Other White Black Asian Other

Analysis
Resuls
Summary
Objectives
The data
The method
Patient information
...continued
48% were working & 55% of
these worked typical ofﬁce
hours.
69% of the working
respondents could take time
away from work to visit their
GP surgery.
48% had visited their GP
surgery in the last 3 months.
Only 12% had not made any
appointments in the last 12
months.
9% 2%
Employment
Full‐time paid work Part‐time paid work
Full‐time education Unemployed
Permanently sick or disabled Fully retired from work
Looking after home Doing something else
37%
12%2%3%6%
29%
9% 2%
Employment
Full‐time paid work Part‐time paid work
Full‐time education Unemployed
Permanently sick or disabled Fully retired from work
Looking after home Doing something else

Analysis
Resuls
Summary
Objectives
The data
The method
Overall levels of satisfaction & positive experience
getting through on the phone, in the last 6 months (satisfaction item)
Patients were very
satisﬁed with how
easy it was to get
through to
someone on the
phone at their
practice (86%).
0
10
20
30
40
50
60
70
80
90
100
%
W
hite
British
O
th
erw
hite
Bla
ck
Asia
n
O
th
er
by ethnicity
In general, are you satisfied with how easy it is to get through
to someone on the phone at your doctor’s surgery?
No Yes
05001000
Frequency
20 40 60 80 100
Satisfied with getting through on the phone
practice means

Analysis
Resuls
Summary
Objectives
The data
The method
urgent appointment, in the last 6 months (experience item)
60% of
respondents had
tried to get an
appointment with a
doctor fairly
quickly and 85%
were successful in
their attempt.
0
10
20
30
40
50
60
70
80
90
100
%
W
hite
British
O
th
erw
hite
Bla
ck
Asia
n
O
th
er
by ethnicity
Think about the last time you tried to get an appointment with a doctor
fairly quickly: Were you able to get the appointment on the same day
or on the next 2 days the surgery was open?
No Yes
0200400600800
Frequency
20 40 60 80 100
Able to get the appointment on the same day or on the next 2 days
practice means

Analysis
Resuls
Summary
Objectives
The data
The method
advance appointment, in the last 6 months (experience item)
45% of
respondents
wanted to book
ahead for an
appointment with
any doctor in their
practice, with a
large percentage
being successful
(75%).
0
10
20
30
40
50
60
70
80
90
100
%
W
hite
British
O
th
erw
hite
Bla
ck
Asia
n
O
th
er
by ethnicity
Last time you wanted to, were you able to get an appointment
with a doctor more than 2 full days in advance?
No Yes
0200400600
Frequency
0 20 40 60 80 100
Able to get an appointment more than 2 full days in advance
practice means

Analysis
Resuls
Summary
Objectives
The data
The method
appointment with a particular doctor, in the last 6 months (experience item)
49% of
respondents
wanted to make an
appointment with a
particular doctor in
their practice, and
86% were
successful, even if
it meant waiting
longer.
0
10
20
30
40
50
60
70
80
90
100
%
W
hite
British
O
th
erw
hite
Bla
ck
Asia
n
O
th
er
by ethnicity
Last time you wanted to, were you able to make an appointment
with a particular doctor − even if it meant waiting longer?
No Yes
05001000
Frequency
0 20 40 60 80 100
Able to make an appointment with a particular doctor
practice means

Analysis
Resuls
Summary
Objectives
The data
The method
opening hours, in the last 6 months (satisfaction item)
Satisfaction with
practices’ opening
hours was high
(80%) with the
most prevalent
reasons for
dissatisfaction
being practice not
open on Saturdays
(32%) and not
open late enough
(23%).
0
10
20
30
40
50
60
70
80
90
100
%
W
hite
British
O
th
erw
hite
Bla
ck
Asia
n
O
th
er
by ethnicity
Over the last 6 months or so, were you satisfied with the hours
your GP surgery was open?
No Yes
0200400600800
Frequency
40 60 80 100
Satisfied with the hours your GP surgery was open in last 6 months
practice means

Analysis
Resuls
Summary
Objectives
The data
The method
opening hours, in the last 6 months (satisfaction item)
Satisfaction with
practices’ opening
hours was high
(80%) with the
most prevalent
reasons for
dissatisfaction
being practice not
open on Saturdays
(32%) and not
open late enough
(23%).
0
10
20
30
40
50
60
70
80
90
100
%
Full−tim
e
paid
w
ork
Part−tim
e
paid
w
ork
Full−tim
e
education
U
nem
plo
yed
Perm
anently
sic
k
ordis
able
dFully
re
tire
d
from
w
ork
Lookin
g
afterhom
eD
oin
g
som
eth
in
g
els
e
by employment status
Over the last 6 months or so, were you satisfied with the hours
your GP surgery was open?
No Yes
0
10
20
30
40
50
60
70
80
90
100
%
Full−tim
e
paid
w
ork
Part−tim
e
paid
w
ork
Full−tim
e
education
U
nem
plo
yed
Perm
anently
sic
k
ordis
able
dFully
re
tire
d
from
w
ork
Lookin
g
afterhom
eD
oin
g
som
eth
in
g
els
e
by employment status
I was dissatisfied with the opening hours of my practice, because...
Not open early enough Not open at lunchtimes
Not open late enough Not open on Saturdays
Not open on Sunday Of some other reason

Analysis
Resuls
Summary
Objectives
The data
The method
Aim of the analysis
Satisfaction was high but we really want to...
Understand the factors that drive patient satisfaction and
experience with access to primary care.
Measure the variability in practice performance that can be
explained by the characteristics of:
The patients.
The practices.
The region.

Analysis
Resuls
Summary
Objectives
The data
The method
Three levels of data
Patient: Practice: Region
The patient level data from the survey were combined with
practice and region level data, from other sources:
General Medical Services (GMS) database 2006.
Super Output Area Indices of Multiple Deprivation (SOA
IMD) 2004.
Quality and Outcomes Framework (QOF), results for
2006/07.
Census 2001.
Primary Care Trust data.

Analysis
Resuls
Summary
Objectives
The data
The method
Three levels of data
Patient: Practice: Region
Patient level:
Gender, age, number of appointments, parent, employment
status, duration of journey from home to work, typical
working hours, ability to take time away from work to visit
GP, limiting longterm conditions, carer, ethnicity.
Practice level:
Deprivation, rurality, list size, contract type, FTE ratio, QOF
year 3 reported achievement, total hours, extended
opening, emergency admissions, GP referrals, SMR, new
registrations.
Regional level:
PCT practice staff number, PCT walk in centre attendance
(for only 31.8% of the PCTs), Strategic Health Authority.

Analysis
Resuls
Summary
Objectives
The data
The method
Analysis approach
General information
Multilevel multivariate regression was used to investigate
relationships between each domain of
satisfaction-experience and patient, practice and regional
characteristics.
We began with univariate analyses, which examined each
predictor separately, and followed these up with a
multivariate analysis which controlled for inter-relationships
between predictors.
Since the outcome variables were binary (e.g. satisﬁed -
not satisﬁed) in nature, we adopted a logistic regression
approach.

Analysis
Resuls
Summary
Objectives
The data
The method
Analysis approach
...some details
Large sample allowed a listwise deletion approach for the
analysis, avoiding less robust approaches.
Removed variables where the variance inﬂation factor was
greater than 4.
Total practice opening hours and extended opening were
excluded (were only available for 53% of the sample).
The full hierarchical nature of the data could not be
modelled, therefore we compromised by using a two-level
model that took account of the nesting of respondents
within practices, and effectively assigned the regional
variables to the individual practices.

Analysis
Resuls
Summary
Objectives
The data
The method
Analysis approach
...the models
Three different multilevel regression models...
...on all respondends (model A).
...on respondends working full- or part-time (model B).
...on all respondents but including patient/practice level
interactions for the variables found to be the strongest
predictors in model A (model C).

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Logistic multilevel regression on all respondents
Patient-level predictors
Only three patient-level variables impacted all ﬁve domains
of satisfaction and experience to a notable degree: age,
employment status and ethnicity.
Satisfaction and positive experience...
increased in level with increasing age.
was lower amongst those working full-time than in any
other group.
was in most instances lower for ethnic minority groups, and
in particularly amongst the Asian community.

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Practice-level predictors
All domains of
satisfaction &
experience were
heavily inﬂuenced
by practice size
with the exception
of satisfaction with
practice surgery
hours.
0
10
20
30
40
50
60
70
80
90
100
%
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
6457
Practice list size
Satisfaction with the ability to get through to someone on the phone
Able to gen an appointment on the same or next 2 days
Able to get an appointment more than 2 full days in advance
Able to make an appointment with a particular doctor
all other regression parameters are assumed fixed to their mean values
vertical grid line corresponds to mean observed listsize
Using results from multivariate regression with all respondents
Satisfaction/experience as affected by practice listsize

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Practice-level predictors
All domains of
satisfaction &
experience were
heavily inﬂuenced
by practice size
with the exception
of satisfaction with
practice surgery
hours.
0
20
40
60
80
100
%ofpositiveresponses
<1
1−2
2−3
3−4
4−6
6−88−1010−12>=12
Phone access
0
20
40
60
80
100
<1
1−2
2−3
3−4
4−6
6−88−1010−12>=12
Appointment within 2 days
0
20
40
60
80
100
<1
1−2
2−3
3−4
4−6
6−88−1010−12>=12
Advance appointment (>2 days)
0
20
40
60
80
100
<1
1−2
2−3
3−4
4−6
6−88−1010−12>=12
Appointment with a particular GP
Practice size in 1,000s of patients
by practice list size
Satisfaction & positive experience

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Region-level predictors
Strategic Health
Authority was the only
predictor of note.
Patients in the NE were
the most satisﬁed and
with better experiences.
Variability was large for
getting through on the
phone and advance
appointments but small
for same or next day
appointments.

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Region-level predictors
Strategic Health
Authority was the only
predictor of note.
Patients in the NE were
the most satisﬁed and
with better experiences.
Variability was large for
getting through on the
phone and advance
appointments but small
for same or next day
appointments.
£2.26 on average
Lowest in the UK
£2.72 on average
Highest in the UK

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Logistic multilevel regression on working respondents
Erm...why?
Employment status was the most important patient-level
predictor in the main analysis.
The sub-analysis gives us the opportunity to explore the
importance of variables speciﬁc to working respondents,
which were necessarily excluded from the main analysis:
Travel time to work.
Work pattern (Typical working hours).
Can take time away from work to visit GP.

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Logistic multilevel regression on working respondents
Predictors
Patient-level:
The most notable predictor, in all domains, was the ability to
take time off work to visit the GP.
The effect of age and ethnicity remained undiminished.
Travel time to work and work pattern had a sizeable effect
on satisfaction with surgery hours.
Practice-level:
Practice size’s strong effect was unchanged.
Emergency admission rates were notably related to
satisfaction with telephone access and experience of
advance appointments.
Regional-level:
Strategic Health Authority again displayed a strong effect.

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
...including patient/practice-level interactions
Are patient responses related to the distribution of certain
characteristics in the practice population?
The size of the area non-white population reduced
satisfaction and experience ratings amongst both white
and non-whites, on all 5 domains. Greatest reduction for
non-white patients.
Ratings of phone access, advance appointments and
opening hours for both ’younger’ (<45) and ’older’ people
slightly higher in areas with mainly young population.
Rating of phone access,urgent appointments and opening
hours were reduced amongst people working FT, and
those not, in areas of high FT employment.

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Satisfaction/experience predictors (Models A&B)
Moderate, strong and very strong
Phone
Same,
next day
Advance
Particular
GP
Opening
hours
Age + + + ++ +
Ethnicity + + + ++ +
Employment ++ ++ ++ ++ +++
Time away +++ +++ +++ +++ +++
Travel time + + ++
Work hours +++
Practice size +++ ++ +++ +++
Emerg. adm. ++ ++
SHA +++ + +++ ++ ++

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Estimated % of satisfaction/experience (Model C)
Etnicity
77%
80%
67%
83%
76%
89% 88%
79%
89%
82%
White respondent
Practice 20% White Practice 80% White
phone urgent app advance app particular doc app opening hours

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Etnicity
73%
78%
64%
76%
73%
87% 86%
76%
83%
80%
non‐White respondent
Practice 20% White Practice 80% White

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Age
85% 85%
73%
83%
75%
89%
85%
77%
84%
80%
'Young' (<45y) respondent
Practice 20% 'young' Practice 80% 'young'

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Age
90% 90% 90%
84%
93%
90%
83%
91%
87%
'Old' (>=45y) respondent
Practice 20% 'young' Practice 80% 'young'
80%

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Employment
90%
85%
77%
85%
76%
85%
82%
73%
84%
73%
Working FT respondent
Practice 20% WFT Practice 80% WFT

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Employment
92%
91%
83%
91%
88%88%
89%
90%
87%
non‐WFT respondent
Practice 20% WFT Practice 80% WFT
83%
80%

Analysis
Resuls
Summary
Model A
Model B
Model C
Overview
Selected unadjusted percentages
...of satisfaction and positive experience
Phone
Same,
next d
Advance
Particular
GP
Opening
hours
Overall 86% 85% 75% 86% 80%
Ethnicity
White Br 87% 87% 77% 88% 83%
Asian 77% 80% 68% 77% 73%
Time away
Yes 89% 87% 77% 87% 83%
No 77% 74% 61% 76% 58%
Prac. size
< 2, 000 94% 92% 87% 89% 85%
(2K, 6K] 90% 87% 79% 88% 83%
(6K, 10K] 86% 85% 74% 86% 82%
≥ 10K 83% 85% 71% 85% 82%
SHA
NE 90% 87% 78% 88% 86%
London 84% 82% 75% 83% 78%

Analysis
Resuls
Summary
Conclusions
Discussion
To sum it up
Patients reported high levels of satisfaction and positive
experience, across all domains.
The patient-level variables with the most notable impacts
were age, ethnicity and employment status.
For those in employment, being able to take time off work
to visit the GP had a very large effect, and effectively
removed the disadvantage across all domains.
So did other factors that freed up time, such as working
only part-time or having a short commute time.
Satisfaction and experience did vary by geographical area,
as represented by Strategic Health Authority.

Analysis
Resuls
Summary
Conclusions
Discussion
To sum it up
Continued...
Practice size was the only practice-level factor of real note,
with a substantial impact in all domains bar surgery hours.
Patients at small practices were more satisﬁed and
reported easier access, especially in telephone access
and obtaining appointments in advance.
In the univariate analyses, single-handed practices
received the highest satisfaction and experience ratings on
most domains.
Opening & extended hours did not appear to inﬂuence
satisfaction/experience, even for patients working FT.
Individual patient responses were related to the
characteristics of the practice population, with the effect
being particularly strong for ethnicity.

Analysis
Resuls
Summary
Conclusions
Discussion
Some disadvantages involved with small practices...
...but bigger is not always better!
QOF year
Overallreportedachievement
04/05 05/06 06/07
020406080100
list size 1,000-1,999
2,000 - 2,999
3,000 - 3,999
4,000 - 5,999
6,000 - 7,999
8,000 - 9,999
10,000 - 11,999
12,000 or more
Practices split by list size in each year, (n=7502)
Comments: e.kontopantelis@manchester.ac.uk

Internal 2010 - Patient Satisfaction with Primary Care

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Internal 2010 - Patient Satisfaction with Primary Care

Similar to Internal 2010 - Patient Satisfaction with Primary Care (20)

More from Evangelos Kontopantelis

More from Evangelos Kontopantelis (20)

Recently uploaded

Recently uploaded (20)

Internal 2010 - Patient Satisfaction with Primary Care