Stimulated by Yang et al 2020.[1]

Photo by Kindel Media on

Acromioplasty – removal of parts of the subacromial bursa and the anterior-inferior surface of the acromion
NHIS – National Health Insurance Service (Korea)
CCI – Charlson comorbidity index

key to terms and acronyms

I am really late to comment on this paper, and I’m not sure how I managed to miss it as it was published in Acupuncture in Medicine in August last year. I definitely saw the proofs because I remember spotting an error on the cover image, so supplied one of my own as a replacement.

I came across it on searching for research on acupuncture in shoulder pain recently. It is another one of the large retrospective cohort studies, and this one comes from Korea. We are now up to about 19 of these trials, with 12 so far from Taiwan, and this one being the 5th from Korea, leaving one each from the US and Thailand.

We have published 5 of these in Acupuncture in Medicine from 2019, and I have seen another very interesting looking one going through review in our system.

This paper used data from the Korean National Health Insurance Service – National Sample Cohort (NHIS–NSC) between the years 2002 and 2013. This NHIS database comprises a random sample of approximately 1 million individuals, comprising 2.2% of the entire population.

The lifetime prevalence of shoulder disorders is 66.7%

Shoulder disorders are the third most prevalent musculoskeletal disorder, with a lifetime prevalence of around 2/3,[2] and the use of surgical approaches has increased dramatically in recent years. We are told that the use of acromioplasty in particular has increased nearly 10-fold over a 5-year period from 2007 in Korea.

Yang et al first searched the database for common shoulder disorders and found 204 448 cases, of whom, 133 108 had received acupuncture at least twice within a period of 6 weeks, and 71 340 had not.

The shoulder disorders selected from the NHIS–NSC database were:

  • adhesive capsulitis (M750)
  • rotator cuff syndrome (M751)
  • shoulder impingement syndrome (M754)
  • sprain and strain of shoulder joints (S134)

After propensity score matching based on age, sex, income and CCI, they were left with two cohorts of 70 811 with which to compare the frequency of acromioplasty. I have described the CCI in a previous blog: Can acupuncture prevent neck surgery?

The adjusted hazard ratio for surgery in the acupuncture cohort was 0.264, meaning that this group was 4 times less likely to get operated on than the matched comparator cohort that did not receive acupuncture (at least twice within 6 weeks of diagnosis).

Acromioplasty was 4 times less frequent in the acupuncture cohort

The absolute rate of surgery in the matched acupuncture cohort was 0.25%, which equated to 180 cases, and 0.96% in the control cohort, which equated to 679 cases.

A quick bit of maths later and I estimate that in the whole country acupuncture could be responsible for saving 250 000 procedures over a roughly 10-year period. At a few thousand dollars each procedure, that is a saving of close to the annual running cost of the UK NHS… and that is just one surgical procedure. We have already seen similar results for neck surgery and knee arthroplasty.


1          Yang G, Jung B, Kim M-R, et al. Acromioplasty rates in patients with shoulder disorders with and without acupuncture treatment: a retrospective propensity score-matched cohort study. Acupunct Med 2020;38:255–63. doi:10.1177/0964528419895529

2          Luime JJ, Koes BW, Hendriksen IJM, et al. Prevalence and incidence of shoulder pain in the general population; a systematic review. Scand J Rheumatol 2004;33:73–81. doi:10.1080/03009740310004667

Declaration of interests MC