Transcript A comparison of health care costs in Southern Sydney, NSW

A comparison of health care costs
in Southern Sydney, NSW,
involved in treating persons with
Parkinson’s disease to those
without Parkinson’s disease
Southern Neurology
Introduction
Parkinson’s disease (PD) and related
parkinsonism are major public health
issues.
 The prevalence of PD in the United States
and Europe is reported to be between
150-300 per 100,000 population.
 The risk of PD increases with advancing
age, rising sharply after the sixth decade.
 A recent Australian study has reported a
crude prevalence rate of PD in Sydney for
patients aged > 55 years of 776 per
100,000, which suggests that Sydney has
one of the highest prevalence rates of PD
in the Western world.

Study
Year of
Survey
Li et al.
1985 (11)
1983
Wang YS et al.
1991(17)
1986
Wang SJ et al.
1994 (18)
1992
Wang SJ et al.
1996 (19)
1993
Chen et al. 2001
(20)
1993
De Rijk MC et al.
1997 (14)
1990’s
De Rijk MC et al.
2000 (15)
Morgante et
al.1992 (21)
Chan et al (5)
Chan et al
Age Specific Prevalence
(per 100,000)
1990’s
1987
1998-1999
2002-2003
Crude
Prevalence
Country
China
Urban population
China
29 provinces
Method
50-59
60-69
92
145
615‡
-
44
Door to door
2 stages
22.5
89.4
157.6
132
14.6
Door to door
2 stages
780
1750
2540
Taiwan
Farming population
Taiwan.
Kinmen
Farming population
Taiwan
Ilan.
France, Italy,
Netherlands, Spain.
Pooled results of five
studies.
European
Pooled results of
Sweden, France,
Netherlands, Italy, Spain
and Germany.
70-79
80+
Door to door
1 stage
273
535
565
1839
119
122.5
546.7
819.7
2197.8
367.9
Door to door
2 stages
Door to door
-
625
1768
3498
-
Door to door
-
565
1666
2813
-
115.6
430.2
1212.5
236.7
167.4
Sicily
Sydney, Australia
Pooled results of two
studies
Door to door
1 stage
1159
3951
9091
780
Door to door
2 stages
Door to door
2 stages
Table 1: Crude prevalence of PD per 100, 000 by age group and
overall for Bankstown (2003), Randwick (1998)
Bankstown
Age group
(years)
Randwick
Combined
N=
PD=
Prevalence/100,000
(95% CI†)
N=
PD=
Prevalence/100,000
(95% CI†)
N=
PD=
Prevalence/100,000
(95% CI†)
55-59
53
0
0 (0-3)
69
2
2899 (351-10464)
122
2
1639 (198-5918)
60-69
167
1
599 (15-3335)
178
3
1685 (347-4921)
345
4
1159 (317-2968)
70-79
203
8
3841 (1699-7764)
202
8
3960 (1707-7802)
405
16
3951 (2260-6400)
80
77
8
10390 (4478-20468
77
6
7792 (2860-16987)
154
14
9091 (4964-15272)
Overall‡
501
17
776 (452-1241)
527
19
775 (467-1209)
1028
36
780 (546-1077)
Prevalence of Parkinson’s disease in Sydney
A survey was carried out in the Bankstown local government area
between August 2002 and July 2003. A total of 4080 households
were letter dropped; 1238 households did not participate (i.e.
refused or were unable to be contacted); 307 households could
not speak English. The participation rate was 70%.
Sample size and characteristics of participants
There were 501 people aged 55 or above who participated in the
study. Overall, the mean age was 70.9 years (SD 8.2) with a
range of 55-100 years. Mean age for males was 71.3 years (SD
7.9) and for females was 70.5 years (SD 8.4); 44.7% of
participants were male (224/501); 71.7% of participants were
born in Australia. The male to female ratio was 0.82:1.
Of the 501 participants, 135 were screened positive (that is,
answered 'yes' to 2/4 screening questions for Parkinsonism, or
answered 'yes' only to the question which asked “Have you noticed
a tremor of your hands, arms, legs or head?”
Background to present study
The high prevalence of the disease, its
degenerative course and quality of life
implications all indicate that the economic
impact of PD in Australia is likely to be
significant.
 Furthermore, it is probable that the
utilisation of medical, allied health and
community support services is likely to be
significant.
 A number of overseas studies
investigating the costs of caring for
patients with PD have found that resource
utilisation correlates with the stage and
severity of PD

Aims
 determine
the direct health care
costs involved in treating a person
with PD in the community
 compare those costs to an age and
sex matched healthy control without
PD
Subjects and methods





A single centre prospective study, in which patients attending a
Parkinson’s disease clinic at Calvary Hospital, Kogarah, Sydney,
were invited to complete a 3-month diary detailing all medical and
allied health expenses from January 1st to March 31st 2003.
Data collected included all pharmaceutical expenses (both
prescription and over-the-counter), general practitioner and
medical specialist expenses, as well as allied health or
complementary medicine expenses incurred during this time
period.
Demographic data including age, gender, race, marital status,
relationship between patient and/or carer (where applicable),
living arrangements, disease duration and intercurrent illnesses
was also obtained.
12 patients completed their 3-month diaries. All PD patients
demonstrated levodopa responsiveness and fulfilled diagnostic
criteria for PD [with at least two of the following signs: resting
tremor, rigidity, bradykinesia and postural instability, in the
absence of other apparent causes of PD].
Twelve age and sex matched control subjects were also asked to
complete a 3-month diary detailing the same information during
the same time period.
Results




There were six male and six female subjects in
both the PD and control groups.
The mean age for the PD group was 71.3 years
(SD 5.9, range 62-82 years) and the mean age
for the control group was 73.2 years (SD 6.7,
range 63-83 years).
The mean disease duration of the PD group was
6.8 years (SD 3.6, range 2-14 years). The
median Hoehn and Yahr stage was 3 (range 1-3).
All 12 PD subjects had a MMSE ≥ 27.
Nine of the PD subjects referred to at least one
family member as a ‘carer’ (8 spouses, 1
daughter) although the relevant family member
was not a recipient of a Federal government
‘carers’ pension or allowance. Similarly, none of
the PD subjects were recipients of a ‘community
aged care package’.
Results (continued)



Three of the PD group and three of the control
subjects had no intercurrent problems. In the PD
group, three patients were taking levodopa
monotherapy and the remaining 9 patients were
taking a combination of two or more of the following:
levodopa, entacapone, cabergoline, pergolide and
amantadine.
Concurrent medications ‘unrelated’ to PD in the
remainder of PD as well as control subjects consisted
of drug therapies for hypertension, hyperlipidaemia,
gastro-oesophageal reflux or peptic ulcer disease,
osteoarthritis and/or osteoporosis.
Two patients in the PD group (one with PD ‘unrelated’
urological procedure and one with PD ‘related’ fall)
and one control subject (urological procedure) had
short stay hospital admissions during the 3-month
diary period.
Table 1. A comparison of the community health care
costs for the 3-month study period in the PD and
control subjects.
PD (n=12)
Control (n=12)
P value
Medication*
$A636
SD 226
Range $A392-1079
$A175
SD 233
Range $A0-705
p < 0.001
GP and Specialists*
$A564
SD 670
Range $A147-2038
$A205
SD 397
Range $A0-1354
p = 0.12
Allied Health Care*
(including
physiotherapy)
$A323
SD 178
Range $A83-613
$21
SD 43
Range $A0-150
p < 0.001
Other costs**
$A232
SD 453
Range $A0-1500
$A12.50
SD 43
Range $A0-150
p = 0.10
Total costs
$A1755
SD 1201
Range $A832-4575
$A413
SD 515
Range $A0-1678
p = 0.001
Table 2. A comparison of the community health care costs
for the 3-month study period ‘related’ to and ‘unrelated’ to
PD in the 12 PD subjects.
PD related (n=12)
Unrelated to PD P value
(n=12)
Medication
$A404
SD 225
Range $A135-895
$A232
SD 242
Range $A0-587
p = 0.17
GP and Specialists
$A293
SD 368
Range $A98-1442
$A271
SD 534
Range $A0-1909
p = 0.91
Allied Health Care
(including
physiotherapy)
$A304
SD 180
Range $A83-613
$19
SD 19
Range $A0-50
p < 0.0001
Other costs**
$A200
SD 457
Range $A0-1500
$A32
SD 100
Range $A0-351
p = 0.26
Total costs
$A1202
SD 820
Range
3563
$A553
SD 591
$A520- Range $A38-1924
p = 0.03
Discussion





The largest single health care expense in the present study was
medication cost, in part due to the treatment cost of comorbidities.
Pressley et al, also recognised pharmaceutical expenditure for comorbidities as a significant cost factor in United States of America
(US) patients with parkinsonism.
In a second US study, Whetten-Goldstein et al, determined the
mean societal burden of PD to be approximately $US6000 per
year, with the single most important factor being compensation
for earnings lost.
In a ‘resource use’ questionnaire of 444 PD patients attending
United Kingdom general practitioner practices, Findley et al, found
that the total mean annual cost of care per patient (for all
patients) was 5,993 pounds and that the patients’ Hoehn and Yahr
stage significantly influenced expenditure.
A Swedish study of 127 PD patients found that direct health care
costs averaged $US2900 per patient per year (with medication
found to be the most costly component) but nonmedical direct
costs were even higher averaging $US4300 and costs due to lost
productivity the highest, averaging $US5,200.
Conclusions





The mean 3-month total (both related and ‘unrelated’
to PD) health care cost for the PD group was
significantly higher than the ‘healthy’ control group
($A1755, SD 1201 versus $A413, SD 515, p=0.001).
The total direct health care cost of PD for our patients
with Hoehn and Yahr stage  3 was four times that of
age and sex matched ‘healthy’ controls.
Medication was the most costly component for both
groups (PD $A636, SD 226 versus controls $A175, SD
233, p<0.001).
The estimated annual cost ($A7020 per patient) in our
patient cohort was comparable to that reported in the
United States and Europe.
This would equate to $A5.5 million/yr/100,000
population age > 55 years (~$A300 million/yr) based
on statistics from PD prevalence study in Sydney.