Musculoskeletal healthcare: Have we over‐egged the pudding?

Full text

Int J Rheum Dis. 2019;22:1957–1960.  
|
1957
wileyonlinelibrary.com/journal/apl
Received:3September2019
|
Accepted:3Septem ber2019
DOI : 10.1111/1756 -185X .13710
EDITORIAL
Musculoskeletal healthcare: Have we over‐egged the pudding?
The phrase “to over-egg the pudding” is a terrific analogy for what
is now increasingly common in health care: medical overuse. Cooks
would know that if you skimp with the eggs a pudding won't hold
together and if you use too many eggs the pudding will go rubbery.
Andinmusculoskeletalhealthcarewealso need toget thebalance
right. People's health can suffer when they receive too little health
care and also if they receive too much health care.
The problem of too little health care is well recognized and it is
easy to understand that patients’ health can be put at risk by un-
deruse of proven healthcare services. However, the opposite prob-
lem is also possible but is less well recognized.1 In this editorial we
adopt some perspectives from the field of overdiagnosis to consider
overuse in musculoskeletal health care.
Overdiagnosis2 is an unwarranted diagnosis that leads to un-
necessary treatments that do not benefit patients and that wastes
health resources that could be better used elsewhere. Overdiagnosis
also may cause harms: direct effects, unintended/indirect conse-
quences, psychological impact, costs and resource implications, op-
portunity cost. We focus on 4 aspects of overdiagnosis that can lead
to medical overuse in musculoskeletal health care:
(i) Overtesting where patients receive unnecessary tests
(ii) Overdetection where clinicians act upon clinically unimportant
findings
(iii) Overdefinition where the boundaries between disease and
health are shifted to encourage more healthcare, and
(iv) Overtreatment where culture, industry and health systems en-
courage treatment with no net benefit.
We provide some examples that illustrate the nature and size of
the problem,3 and highlight potential drivers of overuse of musculo-
skelatal health services (Table 1).
1 | OVERTESTING
Acommonstarting point for overdiagnosisinmusculoskeletalhealth
careisovertesting; wherepatients receiveunnecessary tests.Agood
example is the uncritical interpretation of red flags to screen for serious
pathology. Some texts and guidelines inadvertently encourage medical
overuse by offering a long list of red flags and encouraging diagnostic
work-up and/or specialist referral if even a single red flag is positive.
Onestudy inAustralian generalpracticefoundthatof1172consecu-
tive patients with back pain, 80% recorded a +ve response to at least
1 of the 25 red flags that were considered by the study general practi-
tionerss.4 The irony here is that even though the clinicians were acting
in good faith and aiming to help their patients, they may have harmed
them through overdiagnosis. Other examples in musculoskeletal health
care include repeat vitamin D testing; in Australia Medical Benefit s
Scheme costs for vitamin D testing rose from $109.0 million in the
2009-2010 financial year to $151.1 million in 2012-2013.5 In the sports
medicine field it is common to hear of professional athletes who have
sustained an acute hamstring muscle strain injury undergoing magnetic
resonance imaging to guide management and predict return to sport,
but neither is supported by robust evidence.6 The concern here is the
possibility this practice leak may leak out into the wider communit y.
2 | OVERDETECTION
In the overdiagnosis literature overdetection refers to the identi-
fication of abnormalities that resolve spontaneously or would not
progress sufficiently to cause symptoms or harm during a person's
lifetime.7 In the musculoskeletal field most incidental findings are
picked up by over testing in people with symptoms; using tests
that commonly yield positive test findings in asymptomatic peo-
ple. The challenge is then determining if the finding is relevant or
not. A good example of medical overuse driven by overdetection
would be acting upon the incidental findings commonly found with
musculoskeletal imaging (eg lumbar disc degeneration, rotator cuff
tear, femoroacetabular impingement, heel spur) and initiating more
intensive treatment for the patient (eg specialist referral, surgery).
What compounds the problem is that many of the surgeries that are
encouraged (eg knee arthroscopy,8 subacromial decompression9)
are now known to be no more effective than placebo. In all these
cases the medical overuse is triggered by an unwarranted diagnosis.
Overdetection is not confined to the tests that would typically be
considered the domain of the medical profession. In physiotherapy,
podiatry and chiropractic, the treatments that characterize these
professions are primarily driven by assessment of factors such as
posture, range of motion, alignment, weakness, balance and coor-
dination.Aproblemcanariseifaclinicianmistakenlyjudgesaminor
variation in one of these factors as abnormal, and institutes interven-
tions to correct the presumed abnormality (eg sacroiliac dysfunction,
Thisisanop enaccessarticleundertheter msoftheCreativeCommonsAttributionLicense,whichpe rmitsuse,distributionandreproductioninanymedium,
provide d the original wor k is properly cited.
©2019TheAuthors.International Journal of Rheumatic DiseasespublishedbyAsiaPacificLeagueofAssociationsforRheumatologyandJohnWiley&Sons
Australia,Ltd .

1958
|
EDITORIAL
lumbarinstability,poorhipcontrol).Asomewhatrelatedissueisthe
use of these types of tests to predict risk of injury in sports people.
Abatteryofsuchtests,theFunctionalMovementScreen,hasbeen
recently studied and shown to perform no better than chance in pre-
dicting which professional soccer players would sustain an injury.10
3 | OVERDEFINITION
Overdefinition encourages medical overuse by changing disease
boundaries. This can happen by lowering the threshold that defines
a disease or by expanding disease definitions.7 It has been suggested
that before changing disease definitions it is necessary to consider
certain issues. How many people will be affected? Why is the change
necessary? Can the new disease label be reliably used with patients?
What is the balance of benefits and harms for patients and society
with the change?11 In musculoskeletal health care there are many
examples where disease boundaries have been changed and the
challenge is distinguishing when this change has led to medical over-
use, patient harm and waste; and where the change has provided
benefits to patients and society.
Theinitiative“PainastheFifthVitalSign”loweredthethreshold
for medical treatment of pain to any pain score >0. This change has
been suggested to be one of the important drivers of the current
opioid epidemic which has claimed hundreds of thousands of lives.12
However,itisimportanttorecognizethatthePainastheFifthVital
Sign initiative was motivated by compassion for patients and it was
events happening in parallel where opioid medicines were mislead-
ingly marketed to doctors, patients and patient advocacy groups
that enabled overdefinition to cause so much harm.
Anotherwaytochangediseaseboundariesistosubdividebroad
non-specific disease categories into subcategories that are more tar-
getedandprecise.Forexample,someinthebackpainfielddismiss
the label non-specific low back pain and instead argue for labels tar-
geting a specific structure (disc, facet joint) or mechanism (eg insta-
bility) and the use of similarly targeted personalized therapies. The
problem is that the diagnoses offered are nominal diagnoses13 that
drive more invasive, costly and ineffective therapies without pro-
viding benefit.14,15 The resulting overuse is substantial: for example
the most expensive surgical procedure (US$12.8 billion per annum)
intheUSA isspinefusion,a procedurethatismostcommonly per-
formed for degenerative conditions for which there is good evidence
of harm and poor evidence of benefit over cheaper and safer alter-
natives. Also intheUSA ,therearenearly1millionfacetjointinjec-
tions and half a million neurotomies performed annually at great cost
despite good evidence showing lack of effectiveness.16 The specific
diagnoses also provide a target site for injections of stem cells, blood
products and sclerosing agents which are unproven, expensive and
have the potential for harm.16
“Central sensitization” and “nociplastic” pain are both nominal
diagnoses which have arisen out of attempts to explain (and even-
tually treat) painful conditions but both have the potential to drive
medical overuse because there is no evidence that the label leads to
cost-effective treatment that provides more benefit than harm. The
labels sarcopenia and osteopenia are normal aspects of aging (es-
pecially when defined relative to young people) but they have now
become diseases to be prevented and treated. These are good ex-
amples of changing disease boundaries as they blur the line between
disease and aging. While the desire to reverse or prevent undesir-
able aspects of aging is understandable, these diagnoses have not
been shown to provide a net benefit to “sufferers”.
Anexampleofexpandingdiseasedefinitionswasborrowingthe
“neuropathic pain” label, typically used for less common conditions
such as post-herpetic neuralgia, and attaching it to a much more
common condition: low back pain. The new diagnosis “neuropathic
low back pain” and its potential treatment with gabapentionoids was
promoted to clinicians and reinforced by a consumer-facing disease
awareness program. This led to a marked increase in the use of pre-
gabalin in many countries.17 Subsequently we have learnt that prega-
balin is ineffective for sciatica and low back pain,18 but is associated
with serious harms including death and dependence.19
Broadening disease boundaries is not always harmful and in
some situations, it is yet not clear. Non-radiographic axial spon-
dyloarthritis is an example of lowering the threshold that defines
a disease, in this case by removing the requirement for imaging
evidence of thecondition. At presentthe balance of benefits and
harms is unclear. However, this change has arguably benefited
some patients because it opened access to cost-effective thera-
pies; but it may have caused societal harm through leakage of very
expensive biological treatments to people with non-specific low
back pain who have been labeled as having non-radiographic axial
spondyloarthritis.
4 | OVERTREATMENT
In an ideal healthcare system all patients would receive the right
care, that is “care that is tailored for optimizing health and wellbe-
ing by delivering what is needed, wanted, clinically effective, afford-
able, equitable, and responsible in its use of resources”.1 However,
culture, industry and health systems can disrupt this approach and
encourage overtreatment where patients are provided with treat-
ments that do not provide a net benefit.
Musculoskeletalhealthcarehasarichhistoryofinnovativetests
and treatments escaping into routine care before we completely un-
derstandthebalanceofbenefitsandharms.Arecentexampleisau-
tologous stem cell interventions20 which have grown in popularity in
many countries and tend to focus on treatment of musculoskeletal
conditions.Forexamplea2018studyof432USbusinessesengaged
in direct-to-consumer marketing of stem cell interventions found that
387 businesses offered services for musculoskeletal conditions (or-
thopedic conditions, pain relief, sports injuries).21 These high-cost
interventions have not been shown to be effective but have been
associated with significant harms.
The premature uptake of unproven therapies can in part be ex-
plained by prevailing beliefs. Clinicians and patients both tend to

|
1959
EDITORIA L
overestimate benefits and underestimate harms of tests and treat-
ments.22,23 This may reflect the cognitive biases that we are all sus-
ceptible to24 but is also likely to be influenced by inaccurate health
information that clinicians and patients are exposed to. News stories
related to health inter ventions can be inaccurate,25 wit h a key problem
being exaggerated health news,26 and interestingly this seems to be at
least partly driven by exaggeration in the academic press releases that
initially prompted the news story.2 7, 28 Health information developed
forconsumerscanalsobelacking.AstudyofAustralianconsumerin-
formation on knee arthroscopy for symptomatic osteoarthritis found
that only 6 of 93 document s cited research and only 8 of 93 advised
against the procedure.29Astudyofinformationonlowbackpainon
79 “trustworthy” (eg government, university, hospital) websites from
six countries found that less than the half the treatment recommen-
dations were accurate.30 There is also predisposition with regard to
health care to believe that more is better and that new is better.3
Many professional associations have marketing and advocacy
programs directed to the general public; but if these encourage care
for conditions with a good natural histor y, for example non-specific
low back pain and ankle sprains, there is a substantial risk of over-
treatment. In theUSA initiatives suchasGetPT1st, ChoosePT and
PTNow market the concept that early physical therapy is vital for re-
covery from musculoskeletal conditions and there is little or no men-
tion of the option of self-management and staged or stratified care.
Health systems can drive overtreatment through the approaches
taken to regulation, reimbursement and commissioning of health
services. The end result can be that patients may receive more compli-
cated ca re than is necessar y (eg image-guided steroid inje ctions at sites
where blinded injections are just as effective, inpatient rehabilitation
vs home rehabilitation following knee replacement which is just as ef-
fective, robotic surgery) or care that is ineffective (eg arthroscopy for
knee ost eoarthriti s, spinal fusion sur gery) or of unknow n value (eg stem
cellinjections).InAustralia,whilenolongerthecasefornewitems,the
PharmaceuticalBenefitsSchemeandMedicalBenefitsSchemereim-
burse or subsidize ineffective treatments for chronic musculoskeletal
pain (eg spinal surgery, pregabalin, opioids) but not treatments that are
endorsed in guidelines (eg psychological pain management).
5 | CONCLUDING REMARKS
In musculoskeletal health care overdiagnosis is widespread and leads
to unnecessary tests and treatments that do not benefit patients,
and may cause harms, and waste health resources that could be bet-
ter used elsewhere. While the media often portrays medical overuse
as being deliberate, driven by greed and dishonesty, often medical
overuse is well intentioned. Overuse has many potential drivers, in-
cluding educational deficiencies, commercial incentives and reluc-
tance to challenge the status quo.
We see there are two important steps that need to take place to
address the issue. The first is that there is a clear need to inform con-
sumers, clinicians, decision makers and the public about the extent
TABLE 1 Potential drivers of overuse of musculoskeletal health services
Driver Examples Impact
Overtesting
Ordering unnecessary tests
• Actinguponasingleredflagtotriggerdiag-
nostic work-up and/or specialist referral
• FrequentvitaminDtesting
• Up to 80% of patients with low back pain have at least
one positive red flag
• MedicalBenefitsSchemecostsforvitaminDtesting
rose from $109.0 million in the 2009-2010 financial year
to $151.1 million in 2012-20135
Overdetection
Clinicians act upon clinically
unimportant findings
• Incidental findings on imaging trigger un-
necessary treatment
• Judgingminorpos turalvariationsasabnor-
mal triggers interventions to correct the
abnormalities
• Arthroscopicproceduresfordegenerativekneedisease
costmorethan$3billionperyearintheUSA.
• Medicalizinginf ancybydiagnosingnotionalspinalle-
sions that require manipulative care31
Overdefinition
Changed disease boundaries
encourage more health care
• PromotingPainastheFifthVitalSignen-
couraged treatment of any level of pain
• Disease subcategories that are no more than
nominal diagnoses (eg instability) encourage
use of ineffective therapies
• Creating the label “neuropathic” low back
pain encouraged the use of pregabalin for
low back pain
• Contributed to the opioid crisis that has reduced life
expectancyintheUSA
• Spinal fusion is the most expensive surgical procedure in
theUSA(US$12.8billionannually)
• There has been a surge in the use of pregabalin for pain
and in parallel an increase in pregabalin poisonings,
abuse and deaths
Overtreatment
Culture, industry and health
systems encourage treatment
that does not provide a net
benefit
• There is a predisposition with regard to
health care to believe that more is better
and that new is better
• Professional associations encourage care
for musculoskeletal conditions with a good
natural history
• Health systems reimburse and/or com-
mission more complicated care than is
necessary
• Proliferation of stem cell clinics of fering treatments for
musculoskeletal conditions resulting in high costs, direct
and indirect harms
• Increased treatment rates based on belief of trusted
sources (professional societies and individual
professionals)
• Higher rates of procedures performed in regions where
reimbursement is higher resulting in unwarranted prac-
tice variation

1960
|
EDITORIAL
of, and consequences of, overdiagnosis-driven medical overuse in
musculoskeletal health care. In parallel we need a research program
to characterize the problem, identify causes and develop responses
to address it.2
ChristopherG.Maher1,2
MaryO’Keeffe1,2
RachelleBuchbinder3,4
I.A.Harris1,2,5
1Institute for Musculoskeletal Health, Sydney, NSW, Australia
2School of Public Health, The University of Sydney, Sydney, NSW,
Australia
3Monash Department of Clinical Epidemiology, Cabrini Institute,
Melbourne, Vic., Australia
4Department of Epidemiology and Preventive Medicine, School of
Public Health & Preventive Medicine, Monash University, Melbourne,
Vic., Australia
5South Western Sydney Clinical School, Ingham Institute for Applied
Medical Research, Sydney, NSW, Australia
Correspondence
Christopher G. Maher, School of Public Health, The University of
Sydney, Sydney, A us tr al ia .
Email: christopher.maher@sydney.edu.au
ORCID
Christopher G. Maher https://orcid.org/0000-0002-1628-7857
Mary O’Keeffe https://orcid.org/0000-0001-7104-9248
Rachelle Buchbinder https://orcid.org/0000-0002-0597-0933
REFERENCES
 1. Elshaug AG, Rosent hal MB, La vis JN, et al. L evers for ad dressing
medical underuse and overuse: achieving high-value health care.
Lancet. 2017;390:191-202.
 2. Moynihan R, Barratt AL, Buchbinder R, et al. Australia is re-
sponding to the complex challenge of overdiagnosis. Med J Aust.
2018;209:332-334.
 3. PathiranaT,ClarkJ,MoynihanR.Mappingthedriversofoverdiag-
nosis to potential solutions. BMJ. 2017 ;3 58: j38 79.
 4. HenschkeN,MaherCG,RefshaugeKM,etal.Prevalenceofandscreen-
ing for serious spinal pathology in patients presenting to primary care
settings with acute low back pain. Arthritis Rheum. 2009;60:3072-3080.
 5. Boyages SC .Vitamin D testing: new targeted guidelines stem the
overtesting tide. Med J Aust. 2016;204:18.
 6. HamiltonB, WangensteenA ,Whiteley R,et al. Cohen'sMRIscor-
ing system has limited value in predicting return to play. Knee Surg
Sports Traumatol Arthrosc. 2018;26:1288-1294.
 7. Brode rsen J, Schw artz LM , Heneghan C , O'Sull ivan JW, Arons on
JK,WoloshinS.Overdiagnosis:whatitisandwhatitisn't.BMJ Evid
Based Med. 2018;23:1-3.
 8. SiemieniukR,HarrisIA,AgoritsasT,etal.Arthroscopicsurgeryfor
degenerative knee arthritis and meniscal tears: a clinical practice
guideline. BMJ. 2017;357:j1982.
 9. VandvikPO,Lahdeoja T,Ardern C,etal.Subacromial decompres-
sion surgery for adults with shoulder pain: a clinic al practice guide-
line. BMJ. 2019;364:l294.
10. BakkenA,TargettS,BereT,etal.Thefunc tionalmovementtest9+
is a poor screening test for lower extremity injuries in professional
male football players: a 2-year prospective cohort study. Br J Sports
Med. 2018;52:1047-1053.
11. DoustJ,VandvikPO,QaseemA ,etal.Guidanceformodifyingthe
definition of diseases. JAMA Intern Med. 2017;177(7):1020.
12. Chisholm‐BurnsMA, SpiveyCA,SherwinE,WheelerJ,Hohmeier
K.Theopioidcrisis: origins,trends,policies,andtherolesof phar-
macists. Am J Health Syst Pharm. 2019;76:424-435.
13. HancockMJ,MaherCG,LatimerJ,etal.Systematicreviewoftests
toidentif ythedisc,SIJorfacetjointasthesourceoflowbackpain.
Eur Spine J. 2007;16:1539-1550.
14. JuchJ,MaasET,OsteloR,etal.Effectofradiofrequencydener va-
tion on pain intensity among patients with chronic low back pain:
theMintrandomizedclinicaltrials.JAMA . 2017;318:68-81.
15. HarrisIA,TraegerA,StanfordR,MaherCG,BuchbinderR.Lumbar
spine fu sion: what is the ev idence? Intern Me d J. 2018 ;48 :1430 -1434 .
16. Foster NE, AnemaJR, Cherkin D, et al.Prevention and treatment
of low back pain: evidence, challenges, and promising directions.
Lancet. 2018;391:2368-2383.
17. Goodman CW, Brett AS. Gabapentin and pregabalin for pain
- is increased prescribing a cause for concern? N Engl J Med.
2017;37 7:411- 414.
18. EnkeO,NewHA,NewCH,etal.Anticonvulsants inthetreatment
of low back pain and lumbar radicular pain: a systematic review and
meta-analysis. CMAJ. 2018;190:E786-E793.
19. MoleroY,LarssonH,D'OnofrioBM,SharpDJ,FazelS.Associations
between gabapentinoids and suicidal behaviour, unintentional
overdoses, injuries, road traffic incidents, and violent crime: popu-
lation based cohort study in Sweden. BMJ. 2019;365:l2147.
20. Rubin R. Unproven but profitable: the boom in US stem cell clinics.
JAMA. 2018;320:1421-1423.
21. TurnerL, The US. Direct‐to‐consumer marketplace for autologous
stem cell interventions. Perspect Biol Med. 2018;61:7-24.
22. Hoffmann TC, Del Mar C.Clinicians' expectations of the benefit s
and harms of treatments, screening, and tests: a systematic review.
JAMA Intern Med. 2017;177:4 07-419.
23. Hoffm ann TC, Del Mar C . Patients' e xpectat ions of the bene fits
and harms of treatments, screening, and tests: a systematic review.
JAMA Intern Med. 2014;175(2):274.
24. ScottIA ,SoonJ,ElshaugAG,LindnerR.Counteringcognitivebiases
in minimising low value care. Med J Aust. 2017;206:407-411.
25. Walsh‐Ch ilders K, B raddock J, R abaza C, Schw itzer G. On e step
forward, one step back: changes in news coverage of medical inter-
ventions. Health Commun. 2018;33:174-187.
26. BossemaFG,BurgerP,BrattonL,etal.Expertquotesandexagger-
ation in health news: a retrospective quantitative content analysis.
Wellcome Open Res. 2019;4:56.
27. SumnerP,Vivian‐GriffithsS,BoivinJ,etal.Theassociationbetween
exaggeration in health related science news and academic press re-
leases: retrospective observational study. BMJ. 2014;349:g7015.
28. Winter s M, Larsso n A, Kowalsk i J, Sundberg C J. The assoc iation
between quality measures of medical university press releases and
their corresponding news stories—Important information missing.
PLoS ONE. 2019;14(6):e0217295.
29. BuchbinderR,BourneA.Contentanalysisofconsumerinformation
aboutkneearthroscopyinAustralia.ANZ J Surg. 2018;88:346-353.
30. Ferreira G, Traeger AC , Machado G, O'Keeffe M , Maher CG.
Credibility, accuracy, and comprehensiveness of internet-based in-
formation about low back pain: a systematic review. J Med Internet
Res. 2019;21:e13357.
31. WalkerBF.Thenewchiropractic .Chiropr Man Therap. 2016;24:26.