The cost-effectiveness of traditional acupuncture for low back pain:

1
The cost-effectiveness of traditional acupuncture
for low back pain a pragmatic randomised
controlled trial  
Hugh MacPherson Senior Research
Fellow Department of Health Sciences University
of York Research Director Foundation for
Traditional Chinese Medicine York
Kate Thomas Deputy DirectorMedical Care
Research Unit Faculty of MedicineUniversity of
Sheffield
2
Relevance
16 of the adult UK population consult their GP
with back pain in a year Cost of back pain to
the NHS is 480 million p.a. Acupuncture
increasingly used but under-evaluated Some
evidence to suggest that acupuncture may help
Medical Care Research Unit University of Sheffield
3
Testing the hypothesis that.

• . primary care patients with persistent low
  back pain, when given access to a primary care
  acupuncture service, gain more relief from pain
  than those offered usual management only, for
  equal or less cost.

4
Pragmatic research design - evaluating
comparative effectiveness

• Set in York with LREC approval
• Patients with low back pain referred by their
  general practitioner
• Randomised to either
• acupuncture plus usual GP care
• (up to 10 sessions provided by one of 6
  acupuncturists usual GP care)
• or usual GP care only

5
Inclusion Criteria

• Patients aged 20 to 65 presenting with low back
  pain
• Assessed as suitable for primary care management
  according to Clinical Standards Advisory Group
  for Back Pain guidelines (CSAG)
• A current episode of low back pain of at least 4
  weeks duration

6
Exclusion criteria

• A current episode of back pain of more than 12
  months duration
• Possible serious spinal pathology or severe or
  progressive motor weakness.
• Past spinal surgery (e.g. laminectomy)
• Patients with litigation pending

7
Sample size and primary outcome

• Clinical outcomes measured at 3, 12 and 24 months
• Primary outcome SF-36 Bodily Pain at 12 and 24
  months range 0 worst to100 best
• Sample size
• 240 patients expected to give a 90 chance of
  detecting a difference of 10 points in SF-36
  Bodily Pain score at 12 months

8
Other outcomes

• Secondary outcomes
• Other pain measures (Oswestry, McGill PPI)
• SF-36 (physical functioning, mental health etc),
• Satisfaction, 'worry about back pain, pain-free
  months
• Economic outcomes at 24 months
• Costs
• EQ-5D
• SF-6D

9
GP referral

• 43 GPs from 19 York practices referred patients
• Mean number of referrals per GP 7 patients
• Recruitment period 18 months
• 289 patients referred by GPs
• 17 attrition following identification by GP
• Patients choosing not to come into the study
• Patients excluded
• Back pain resolved immediately
• 241 patients recruited

10
Did patients represent the full range of primary
care patients?

• GPs did not report systematic exclusion of
  particular patients
• GPs estimated that approximately 50 of possible
  patients were invited
• 5 of invited patients declined to come into the
  trial

11
Randomisation process

• Concealed randomisation to acupuncture or
  control group, based on ratio 21
• unequal so as to explore acupuncturist effect
• Allocation to acupuncturist based on convenience

12

Randomised (N 241)



Usual Management

Acupuncture offer

N81 (-1)

(-1)
N160
3
-
month follow
-
up

3
-
month
-
up

follow
Responses

146 (92)
Responses

71 (92)



12 month follow-up
12
-
month follow
-
up


Responses

68 (85)
Responses

147 (92)

24
-
month follow
-
up

24
-
month follow
-
up

Responses

123 (77)
Responses

59 (73)


13
Demographic profiles (at baseline)
14
Demographic profiles (at baseline)
15
Treatments received during first three months
from recruitment
16
Acupuncture treatments provided

• Average 10 needles per treatment, and 8
  treatments per patient, usually weekly
• Commonly selected points
• BL points BL-23, BL-26, BL-53, BL-40
• GB points GB-30, GB-34
• Local points Huatou at L3, L4 and L5, AhShi
  points
• Syndrome points KID-3
• Auxiliary techniques and advice

17
Treatment acceptability at 3 months

• Satisfied with treatment received
• acupuncture group 74.2
• usual care group 60.3

18
Treatment acceptability at 3 months

• Satisfied with treatment received
• acupuncture group 74.2
• usual care group 60.3
• Further evidence
• 90 completed acupuncture course
• 86 willing to try acupuncture again
• 86 would recommend acupuncture to a friend

19
SF-36 Bodily Pain score adjusted for baseline
Diff5 pts p0.13
20
SF-36 Bodily Pain score adjusted for baseline
Diff6 pts P0.11
Diff5 pts
21
SF-36 Bodily Pain score adjusted for baseline
Diff 5 pts
Diff6 pts
Diff 8 pts P0.03
Pop norm.
22
Sensitivity analysis - effect of intervention on
pain scores at 24 months
23
Sensitivity analysis - effect of intervention on
pain scores at 24 months
24
Sensitivity analysis - effect of intervention on
pain scores at 24 months
25
At randomisationDo you believe that acupuncture
can help your low back pain?
26
At randomisationDo you believe that acupuncture
can help your low back pain?
27
Acupuncture or acupuncturist?
28
Secondary outcomes

• At 24 months
• 81 in acupuncture group reported that their
  allocated treatment had helped their back pain,
  compared with 52 in the usual care group.
• 13 reported 12 months pain-free, compared with
  3 in the usual care group
• Worry profile

29
Worry about back pain at 24 months (compared to
baseline)
Diff. between groups Plt0.001
30
Clinical summary

• Primary outcome
• Better pain scores (clinical and statistical
  significance at 24 months)
• Secondary outcomes
• Trend in favour of acupuncture some statistical
  significance
• Cost-effectiveness .. ?

31
NHS and total social costs (mean cost/patient
over 24 months)
32
Cost utility (NHS perspective)

• NHS costs ()
• Generic health utilities gained over time (QALYs)
• Cost per QALY gained

33
Cost utility over 24 months

• Using the EQ-5D Estimated cost per QALY 3,156
• Using the SF-6D Estimated cost per QALY 2,436
• If 30,000 is taken as the maximum threshold for
  what the NHS can afford to pay (NICE guidelines),
  then acupuncture for low back pain appears highly
  cost-effective.

34
Conclusions

• It is possible to conduct a pragmatic RCT of
  individualised acupuncture in primary care.
• A short course of acupuncture confers long-term
  clinical benefits
• Results are unlikely to be due to belief
• Acupuncture for low back pain is cost-effective
  in the longer term, relative to usual care.

35
Implications for healthcare

• Based on the study findings..
• Commissioners of musculoskeletal services
  would be justified in considering making GP
  referral to a short course of traditional
  acupuncture available for a typical population of
  primary care attendees with persistent
  non-specific low back pain.

36
Further research

• Optimum timing for an acupuncture treatment
• Why continued improvement over time?
• Variability between acupuncturists
• Meaning and value to patients of reduction in
  worry about back pain.
• Distillation of protocol for acupuncture
  treatment of LBP.

37
Acknowledgements

• NCCHTA
• Patients
• Acupuncturists
•  
• Patient representative
• David Laverick
• Advisory Board
• Trevor Sheldon
• Sally Bell?Syer

• Research Team
•  
• Lucy Thorpe
• Mark Roman
• Julie Ratcliffe
• John Brazier
• Mike Fitter
• Mike Campbell
• Ann Morgan
• Liz Oswald
• Helen Wilkinson
• Jon Nicholl