131

C OPYRIGHT Ó 2018 BY T HE J OURNAL OF B ONE AND J OINT S URGERY, I NCORPORATED

A commentary by Lars Engebretsen MD,

PhD, is linked to the online version of this
article at jbjs.org.

Low Prevalence of Hip and Knee Arthritis in Active

Marathon Runners
Danielle Y. Ponzio, MD, Usman Ali M. Syed, BS, Kelly Purcell, BS, Alexus M. Cooper, BS, Mitchell Maltenfort, PhD,
Julie Shaner, MD, and Antonia F. Chen, MD, MBA

Investigation performed at the Department of Orthopaedic Surgery, The Rothman Institute at Thomas Jefferson University,
Philadelphia, Pennsylvania

Background: Existing evidence on whether marathon running contributes to hip and knee arthritis is inconclusive. Our
aim was to describe hip and knee health in active marathon runners, including the prevalence of pain, arthritis, and
arthroplasty, and associated risk factors.
Methods: A hip and knee health survey was distributed internationally to marathon runners. Active marathoners who
completed ‡5 marathons and were currently running a minimum of 10 miles per week were included (n = 675). Questions
assessed pain, personal and family history of arthritis, surgical history, running volume, personal record time, and current
running status. Multivariable analyses identified risk factors for pain and arthritis. Arthritis prevalence in U.S. marathoners
was compared with National Center for Health Statistics prevalence estimates for a matched group of the U.S. population.
Results: Marathoners (n = 675) with a mean age of 48 years (range, 18 to 79 years) ran a mean distance of 36 miles
weekly (range, 10 to 150 miles weekly) over a mean time of 19 years (range, 3 to 60 years) and completed a mean of 76
marathons (range, 5 to 1,016 marathons). Hip or knee pain was reported by 47%, and arthritis was reported by 8.9% of
marathoners. Arthritis prevalence was 8.8% for the subgroup of U.S. marathoners, significantly lower (p < 0.001) than the
prevalence in the matched U.S. population (17.9%) and in subgroups stratified by age, sex, body mass index (BMI), and
physical activity level (p < 0.001). Seven marathoners continued to run following hip or knee arthroplasty. Age and family
and surgical history were independent risk factors for arthritis. There was no significant risk associated with running
duration, intensity, mileage, or the number of marathons completed (p > 0.05).
Conclusions: Age, family history, and surgical history independently predicted an increased risk for hip and knee arthritis
in active marathoners, although there was no correlation with running history. In our cohort, the arthritis rate of active
marathoners was below that of the general U.S. population. Longitudinal follow-up is needed to determine the effects of
marathon running on developing future hip and knee arthritis.
Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

A
lthough distance running is associated with numerous 78.4 million, or 26% of the adult population12-14. Although
health benefits, the impact on hip and knee joint the cause of arthritis is not fully understood and multifac-
health is inconclusive. Distance running has been torial, arthritis is associated with age, family history, female
associated with an increased prevalence of arthritis in some sex, race, obesity, limb alignment, muscle weakness, and
studies1-4, but other studies have shown an inverse association trauma12.
or no association5-11. Arthritis is the most common cause of Biomechanically, it seems intuitive that repetitive load-
physical disability in adults in the United States12. By 2040, the bearing cycles of running, long-term, over long distances, may be
number of adults with arthritis has been projected to reach detrimental to the hip and knee. Joint load forces while running

Disclosure: There was no source of external funding for this study. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the
online version of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical
arena outside the submitted work (http://links.lww.com/JBJS/E516).

J Bone Joint Surg Am. 2018;100:131-7 d http://dx.doi.org/10.2106/JBJS.16.01071

 132
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
L O W P R E VA L E N C E O F H I P AND KN E E ART H R I T IS IN AC T I V E
V O LU M E 1 00-A N U M B E R 2 J A N UA R Y 17, 2 018
d d
M A R AT H O N R U N N E R S

are approximately 8 times body weight at the knee and 5 times and were excluded. Knee pain was the most common reason
body weight at the hip15,16. A proposed explanation as to why (42%). Other reasons included a change in fitness goals (26%),
runners do not have high arthritis rates includes a relatively short hip and knee pain (11%), total hip arthroplasty (11%), hip pain
duration of ground contact and long stride lengths that blunt the (5%), and muscle imbalance (5%). Inclusion criteria required
high peak joint loads, such that per-unit distance loads are completion of ‡5 marathons and active running status covering
equivalent to walking15. Additionally, running-enhanced limb a minimum of 10 miles per week, which identified 675 mara-
compliance, muscle mass, bone density, and body weight thon runners for further analysis.
maintenance may counter forces through the joints2. One study Active marathoners originated from 31 countries: United
showed a 2.7% arthritis rate in runners, as well as reduced risk States (430), United Kingdom (50), Denmark (37), Canada
related to lower body mass index (BMI)17. (20), Australia (20), the Netherlands (18), Ireland (15), New
The relationship between distance running and hip and Zealand (14), Germany (11), Finland (9), Belgium (7), Sweden
knee arthritis is unclear. Prior studies have been centered on (7), India (5), France (4), South Africa (3), Mexico (3), Ma-
radiographs, small cohorts, and heterogeneous athletic popu- laysia (3), the Philippines (3), Singapore (3), Italy (2), Hong
lations2-7. Our large, cross-sectional study aims to describe hip Kong (1), Hungary (1), Norway (1), Trinidad & Tobago (1),
and knee health in active marathon runners, including the Austria (1), Belarus (1), Brazil (1), Iran (1), Japan (1), Korea
prevalence of pain, arthritis, and arthroplasty, and associated (1), and Russia (1).
risk factors. Also, we compare arthritis prevalence in U.S. Nineteen survey questions (see Appendix) were designed
marathoners to prevalence estimates for a matched U.S. adult to assess running history and hip and knee health. Demo-
population. graphic information included age, sex, height, weight, and
country of origin. Running history included running duration
Materials and Methods (years), weekly mileage, number of marathons completed,

T hrough an electronic survey distributed internationally to

marathon clubs, 953 marathon runners provided infor-
mation with regard to their hip and knee health and running
personal record time (an indicator of running intensity), and
current running status. Joint health questions inquired about
hip or knee pain in the past year, doctor-diagnosed hip or knee
history. The international scope enabled inclusion of a large arthritis, age of diagnosis, family history of arthritis (arthritis in
number of high-volume, high-intensity marathoners over a a parent or sibling), and hip or knee surgical history. The
range of demographic characteristics, with the hypothesis question pertaining to arthritis read as follows: “Have you ever
that running volume and intensity would relate to increased been diagnosed by a doctor with hip or knee arthritis (worn-
rates of pain and arthritis of the hip and knee. Responses out cartilage)?” The formulation of this question was based on
were collected between December 2015 and March 2016. epidemiological studies, including the National Health Inter-
Participation was on a voluntary basis, and participants were view Survey (NHIS) conducted by the U.S. National Center for
informed that their responses would be used in a scientific Health Statistics (NCHS), Centers for Disease Control and
publication. Prevention (CDC), which was used in our study as a reference
Target subjects were restricted to adult active mara- for arthritis prevalence in the U.S. population12,18,19. The NHIS
thoners (‡18 years of age) to avoid survival bias associated with is a health survey administered to adult U.S. citizens that serves
an unknown magnitude of runners who retire early because of as the principal source of information on the health of the
arthritis and did not participate in the survey. Of the 953 re- civilian household population of the United States18. The NHIS
spondents, 19 (2%) reported that they were no longer running survey question pertaining to arthritis read as follows18,19: “Have

TABLE I Demographic Characteristics of Marathon Runners in the Study

Demographic Characteristics All Marathoners (N = 675) U.S. Marathoners (N = 430)

Age* (yr) 47.9 ± 11.6 (18 to 79) 46.1 ± 11.7 (18 to 74)
Sex†
Male 392 (58.1%) 209 (48.6%)
Female 283 (41.9%) 221 (51.4%)
BMI* (kg/m2) 23.6 ± 3.4 (16.1 to 47.8) 23.6 ± 3.5 (17.1 to 47.8)
No. of marathons* 75.5 ± 116.6 (5 to 1,016) 48.2 ± 89.2 (5 to 1,016)
Duration of running* (yr) 18.8 ± 12.4 (3 to 60) 19.0 ± 12.2 (3 to 55)
Mileage* (miles/week) 36.4 ± 17.0 (10 to 150) 34.7 ± 16.2 (10 to 150)

*The values are given as the mean and the standard deviation, with the range in parentheses. †The values are given as the number of runners,
with the percentage in parentheses.
 133
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
L O W P R E VA L E N C E O F H I P AND KN E E ART H R I T IS IN AC T I V E
V O LU M E 1 00-A N U M B E R 2 J A N UA R Y 17, 2 018
d d
M A R AT H O N R U N N E R S

sufficient (‡150 minutes per week) according to the U.S.

Department of Health and Human Services 2008 Physical
Activity Guidelines21.

Statistical Analysis
Univariate and multivariable regression analyses were per-
formed to identify risk factors for hip and knee pain and ar-
thritis in marathoners. Multivariable analysis results are
reported as odds ratios (ORs) with 95% confidence intervals
(CIs). Significance was defined by p < 0.05. All analyses were
performed using R 3.1.1 (R Foundation for Statistical Com-
puting), utilizing the “rms” package to perform the logistic
regression.

Results
Fig. 1
Graph showing the rates of pain and arthritis across age in marathoners.
T he mean age of marathoners was 47.9 years (range, 18 to 79
years), and 41.9% of runners were female. Marathoners
ran a mean distance of 36.4 miles per week (range, 10 to 150
The shaded regions indicate the 95% CI.
miles per week) over a mean time of 18.8 years (range, 3 to 60
years) and completed a mean number of 75.5 marathons
you ever been told by a doctor or other health professional that (range, 5 to 1,016 marathons). Table I summarizes the demo-
you have some form of arthritis?” graphic characteristics.
Arthritis prevalence in U.S. marathoners was compared Hip and/or knee pain was reported by 317 marathoners
with prevalence estimates for the U.S. population overall and (47.0%), consisting of 22.2% with knee pain, 11.1% with hip
was stratified by selected characteristics (age, sex, BMI, and pain, and 13.6% with hip and knee pain; 35.4% reported a
physical activity level) based on the 2010 to 2012 NHIS12,17. positive family history of arthritis. The rate of arthritis in
Coarsened exact matching was employed to match the U.S. marathoners was 8.9% (60 of 675), involving the knee (5.8%),
population to the U.S. marathoner subset by age, sex, and hip (2.1%), and hip and knee (1.0%). Figure 1 depicts rates of
BMI20. Physical activity level within the U.S. population was pain and arthritis across age. Of marathoners with arthritis, the
reported by frequency and duration of moderate or vigorous mean age at the time of diagnosis was 46.8 years (range, 18 to
activity. Americans were categorized as inactive (0 minutes 68 years). Of marathoners who were ‡65 years of age, the
per week), insufficient (1 to 149 minutes per week), and arthritis rate was 24.5%.

Fig. 2
Bar graphs showing that arthritis prevalence in U.S. marathoners is significantly lower than the prevalence in a matched U.S. adult population (National
Health Interview Survey 2010 to 2012), both overall and when stratified by selected demographic characteristics of age (p < 0.0001) (Fig. 2-A), sex
(p < 0.0001) (Fig. 2-B), BMI (p < 0.0001) (Fig. 2-C), and physical activity level (p < 0.0001) (Fig. 2-D). The error bars indicate the 95% CI.
 134
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
L O W P R E VA L E N C E O F H I P AND KN E E ART H R I T IS IN AC T I V E
V O LU M E 1 00-A N U M B E R 2 J A N UA R Y 17, 2 018
d d
M A R AT H O N R U N N E R S

Fig. 3-A Fig. 3-B

Graphs showing the prevalence of arthritis in U.S. marathoners compared with the prevalence in a matched U.S. adult population (National Health
Interview Survey 2010 to 2012), stratified by physical activity level (inactive, insufficient, and sufficient activity levels) in relation to BMI (p < 0.0001)
(Fig. 3-A) and age (p < 0.0001) (Fig. 3-B). Despite increased arthritis rates with age and BMI, there is an inverse relationship between activity level and
arthritis rate. The shaded regions indicate the 95% CI.

Arthritis prevalence in U.S. marathoners (n = 430) was

compared with prevalence estimates for a matched group of the TABLE II Multivariable Analysis of Risk Factors for Pain and
Arthritis of the Hip and Knee in Marathon
U.S. adult population (n = 9,526). Similar to the U.S. popu- Runners
lation, the subset of U.S. marathoners showed increased ar-
thritis prevalence in female individuals and among those who Risk Factor OR* P Value
were overweight or obese. Arthritis prevalence in U.S. mara- Age, per year
thoners (8.8%) was significantly lower (p < 0.001) than the Pain 1.01 (0.99 to 1.03) 0.185
prevalence in the matched U.S. population (17.9%); it was also Arthritis 1.05 (1.02 to 1.08) <0.001
significantly lower (p < 0.001) than the prevalence in sub-
BMI, per kg/m2
groups stratified by age, sex, BMI, and physical activity level
Pain 1.00 (0.96 to 1.05) 0.867
(Fig. 2). Despite an increasing arthritis rate with BMI and age
Arthritis 1.07 (0.99 to 1.16) 0.090
(Fig. 3), the higher level of physical activity in marathoners was
related to diminished arthritis rates (p < 0.001). Although ar- Female sex
thritis rates for U.S. and international marathoners were Pain 1.39 (0.98 to 1.97) 0.063
comparable, we avoided direct comparison between the in- Arthritis 1.61 (0.85 to 3.06) 0.144
ternational marathoners and the U.S. population because of Family history
confounding variables of diet, attitude, and cultural differences. Pain 1.23 (0.89 to 1.71) 0.209
Multivariable analysis (Table II) identified history of hip Arthritis 2.52 (1.41 to 4.47) 0.002
or knee surgical procedures as most predictive of hip and knee History of hip or
pain. Hip and knee pain correlated with arthritis (OR, 8.56 knee surgery
[95% CI, 3.95 to 21.23]; p < 0.001). An increasing number of Pain 2.56 (1.58 to 4.17) <0.001
marathons was associated with decreased rates of joint pain Arthritis 5.43 (2.94 to 10.03) <0.001
(OR, 0.57 [95% CI, 0.40 to 0.82]; p = 0.002) (Fig. 4). There was No. of marathons
a slight upward trend in arthritis rate as the number of mara- (log10)
thons increased (Fig. 4), which was not a significant predictor Pain 0.57 (0.40 to 0.82) 0.002
of arthritis in the multivariable model (OR, 1.22 [95% CI, 0.64 Arthritis 1.22 (0.64 to 2.33) 0.551
to 2.33]; p = 0.55). An increasing number of marathons cor-
related with an upward trend in age, as older adults tended to *The values are given as the OR, with the 95% CI in parentheses.
run more marathons. Multivariable analysis identified risk
 135
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
L O W P R E VA L E N C E O F H I P AND KN E E ART H R I T IS IN AC T I V E
V O LU M E 1 00-A N U M B E R 2 J A N UA R Y 17, 2 018
d d
M A R AT H O N R U N N E R S

Among retired marathoners who were excluded from analy-

sis, there were 2 additional cases of total hip arthroplasty, and
these marathoners discontinued running after the surgical
procedure because of postoperative limb-length discrepancy
and surgeon recommendation. There was 1 individual run-
ning after a unicondylar knee arthroplasty who was excluded
from analysis for not meeting the running volume inclusion
criteria.

Discussion

A mong active marathoners, we found no correlation be-

tween running history and arthritis. Age, family history,
and surgical history were independent predictors for hip and
knee arthritis in marathoners. However, in the United States,
our cohort of marathoners demonstrated an arthritis rate be-
low that of the general population, both overall and across
subgroups stratified by age, sex, BMI, and physical activity
level.
Although we found no association between running
history and arthritis in marathoners, other studies have
Fig. 4
linked distance running with increased arthritis prevalence1-4.
Graph showing the rate of hip and knee pain and arthritis with increasing
Cheng et al. found that running 20 miles per week was as-
number of marathons. An increasing number of marathons was associated
sociated with self-reported arthritis among men <50 years of
with decreased rates of joint pain (OR, 0.57 [95% CI, 0.42 to 0.78]; p =
age1. Increased prevalence of radiographic hip and knee ar-
0.002). The number of marathons was not a significant predictor of arthritis
thritis was shown in small cohorts of former elite distance
(OR, 1.22 [95% CI, 0.64 to 2.33]; p = 0.551). The shaded regions indicate runners and tennis players3. Tveit et al. demonstrated in-
the 95% CI. creased prevalence of arthritis and arthroplasty in former
male elite athletes, driven by the inclusion of contact athletes
factors for arthritis in marathoners, including increasing age and athletes with a history of knee injury2. These studies
(OR, 1.05 [95% CI, 1.02 to 1.08] per year; p = 0.002) and were largely centered on radiographs, small cohorts, and
runners with a family history of arthritis (OR, 2.52 [95% CI, heterogeneous athletic populations. Alternatively, several
1.41 to 4.47]; p = 0.002) or a history of hip or knee surgical studies have documented similar rates of radiographic,
procedures (OR, 5.43 [95% CI, 2.94 to 10.03]; p < 0.001). clinical, or symptomatic arthritis between runners and non-
There was no positive relationship between pain or arthritis runners5-11. Williams reported a 2.7% arthritis rate in run-
and running duration, intensity, weekly mileage, or number of ners17. Sohn and Micheli showed that distance running was
marathons. not associated with arthritis in former college cross-country
A history of hip or knee surgical procedures, present in runners compared with former college swimmers over 25
12.6%, was the most pronounced risk factor for pain and years of follow-up8.
arthritis. In marathoners with a history of surgical proce- Several factors may counter the effect of mechanical
dures, 42.7% reported pain in the operative joint, 67.4% re- loading of the joints while running. The relatively short
ported generalized hip and/or knee pain, and 28.1% reported duration of ground contact and long stride lengths during
arthritis, predominantly involving the knee (76% of those running blunt the effect of high joint loads, such that the
with arthritis). Prior surgical procedures included meniscal per-unit distance loads are equivalent between running and
surgical procedures (34.8% of prior surgical procedures), walking15. Lane et al. found 40% more bone mineral content
knee arthroscopy (33.7%), anterior cruciate ligament (ACL) in runners compared with controls and no differences in
reconstruction (19.1%), and hip arthroscopy (3.4%). Other radiographic, clinical, or symptomatic arthritis7. Hyldahl
procedures included distal femoral tumor excision, hip cyst et al. showed a running-induced decrease in knee synovial
curettage and bone-grafting, open reduction and internal fluid pro-inflammatory cytokine concentration in healthy
fixation of slipped capital femoral epiphysis, hip fracture open young adults after 30 minutes of running, suggesting an
reduction and internal fixation, periacetabular osteotomy, pes anti-inflammatory chondroprotective effect22. Several studies
anserine bursectomy, multiligament knee reconstruction, and have indicated an inverse relationship between muscle mass
microfracture of a knee chondral defect. Six of the analyzed and arthritis and suggest that muscle fatigue, diminished
marathoners underwent arthroplasty, including 1 who un- proprioception, and age-related decreased tissue compli-
derwent total knee arthroplasty, 1 who underwent bilateral ance, as well as joint injury, disease, or surgical procedure,
total knee arthroplasty, 1 who underwent hip resurfacing may contribute to arthritis2,23-26. Our findings align with this
arthroplasty, and 3 who underwent total hip arthroplasty. theory as aging, insufficient physical activity, and joint
 136
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
L O W P R E VA L E N C E O F H I P AND KN E E ART H R I T IS IN AC T I V E
V O LU M E 1 00-A N U M B E R 2 J A N UA R Y 17, 2 018
d d
M A R AT H O N R U N N E R S

surgical procedures correlated with arthritis. We also between radiographic and clinical arthritis36-38. Several stud-
showed an inverse relationship between arthritis and phys- ies have indicated that self-reported diagnosis of arthritis has
ical activity level, and arthritis rates were further reduced in acceptable reliability for use in epidemiological studies12,39.
marathoners, supporting joint health benefits from physical NHIS arthritis estimates for the U.S. population also rely on
activity. self-reported diagnosis, and our survey question pertaining
A prior joint surgical procedure was the strongest to arthritis was consistent with that of the NHIS and other
predictor of pain and arthritis in marathoners. Disruption of population health surveys12,18,19. Second, participation in this
the cruciates, tibiofemoral osteochondral surface, or menisci survey required computer access. This could bias outcomes,
leads to changes in the path of the instantaneous center of the as a younger generation is more likely to engage in an elec-
knee and may result in arthritis. For example, several studies tronic survey. Third, it is difficult to isolate all factors that
have indicated that ACL reconstruction does not restore may affect arthritis (e.g., trauma, occupation, gait, and limb
normal kinematics during dynamic, functional loading, and alignment). Additionally, it is not possible to conclude that
graft degradation may occur over time27,28. An increased risk distance running is not a risk factor for arthritis, as runners
of arthritis has been found after ACL reconstruction, which diagnosed with early arthritis may refrain from running.
may be related to abnormal kinematics associated with ACL However, we included a large number of runners who com-
injury and reconstruction28-30. In a previously compromised pleted >100 marathons with the thought that if running
joint, our current study suggests a relationship between high- caused joint deterioration, we would likely detect it. It is
impact running cycles and arthritis. Teichtahl et al. found important to note that these individuals may have potential
that preexisting joint health determines the response to protective qualities such as low BMI, male sex, or other ge-
persistent vigorous physical activity over time9. Among older netic advantages that make it difficult to generalize results to
adults with clinically healthy knees followed longitudinally, the population at large. Ideally, this cohort will be followed to
participation in regular vigorous physical activity was asso- determine outcomes later in life.
ciated with deleterious knee cartilage changes only in those Although existing evidence on whether distance run-
with preexisting changes on magnetic resonance imaging ning potentiates arthritis is inconclusive, we found a low
(MRI)9. Runners with joint injury or those who had under- prevalence of arthritis in active marathon runners. Age, family
gone surgical procedures after joint injury may decide to vary history, and surgical history independently predicted hip and
their exercise practices to include lower-impact activities. knee arthritis in marathoners. There was no significant cor-
Research should examine the biomechanics and long-term relation between running history and arthritis, but an in-
outcomes of athletes following joint injury and surgical creasing number of marathons predicted decreased joint pain.
procedures. Longitudinal studies are needed to determine the impact of
Surprisingly, an increased number of marathons was marathon running on developing arthritis. We plan to pro-
associated with decreased joint pain. Although this may re- spectively follow the marathoners with redistribution of our
flect benefits associated with distance running, it may also survey to assess for changes in joint health over time.
reflect self-selection whereby runners with pain discontinue
running. Tesarz et al. found that athletes possessed higher Appendix
pain tolerance compared with normally active controls, pos- A table showing a health survey of hip and knee surgical
sibly related to genetics or conditioning31. Bruce et al. found procedures in marathon runners is available with the
that although musculoskeletal pain increases with age, there online version of this article as a data supplement at jbjs.org
was no increase in pain in older adults who participated in (http://links.lww.com/JBJS/E517). n
regular vigorous exercise32. There was 25% less pain and less
arthritis in runners compared with those who had never been
runners32. Chakravarty et al. found significantly divergent
physical disability levels later in life favoring runners com-
pared with non-runners over a 21-year longitudinal study33. Danielle Y. Ponzio, MD1
Usman Ali M. Syed, BS1
Even among aging adults with mild to moderate knee ar- Kelly Purcell, BS1
thritis, regular exercise reduced pain, whereas inactivity was Alexus M. Cooper, BS1
associated with greater pain32,34. Poor physical function and Mitchell Maltenfort, PhD1
weak thigh extensor strength were linked to frequent pain in Julie Shaner, MD1
older adults with bilateral knee osteoarthritis35. The current Antonia F. Chen, MD, MBA1
study and the U.S. population data support a protective role of 1Department of Orthopaedic Surgery, The Rothman Institute at Thomas
physical activity in joint health.
Jefferson University, Philadelphia, Pennsylvania
There were several limitations with our study. First,
doctor-diagnosed arthritis was self-reported. Clinical or ra- E-mail address for D.Y. Ponzio: danielle.ponzio@gmail.com
diographic diagnosis is not easily implemented in observa-
tional studies, and studies have suggested a discordance ORCID iD for D.Y. Ponzio: 0000-0002-6908-1107
 137
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
L O W P R E VA L E N C E O F H I P AND KN E E ART H R I T IS IN AC T I V E
V O LU M E 1 00-A N U M B E R 2 J A N UA R Y 17, 2 018
d d
M A R AT H O N R U N N E R S

References
1. Cheng Y, Macera CA, Davis DR, Ainsworth BE, Troped PJ, Blair SN. Physical 21. Office of Disease Prevention and Health Promotion. 2008 Physical activity
activity and self-reported, physician-diagnosed osteoarthritis: is physical activity a guidelines for Americans. 2008. https://health.gov/paguidelines/pdf/paguide.pdf.
risk factor? J Clin Epidemiol. 2000 Mar 1;53(3):315-22. Accessed 2017 Aug 25.
2. Tveit M, Rosengren BE, Nilsson JÅ, Karlsson MK. Former male elite athletes have 22. Hyldahl RD, Evans A, Kwon S, Ridge ST, Robinson E, Hopkins JT, Seeley MK.
a higher prevalence of osteoarthritis and arthroplasty in the hip and knee than Running decreases knee intra-articular cytokine and cartilage oligomeric matrix
expected. Am J Sports Med. 2012 Mar;40(3):527-33. Epub 2011 Nov 30. concentrations: a pilot study. Eur J Appl Physiol. 2016 Dec;116(11-12):2305-14.
3. Spector TD, Harris PA, Hart DJ, Cicuttini FM, Nandra D, Etherington J, Wolman RL, Epub 2016 Oct 3.
Doyle DV. Risk of osteoarthritis associated with long-term weight-bearing sports: a 23. Lohmander LS, Englund PM, Dahl LL, Roos EM. The long-term consequence of
radiologic survey of the hips and knees in female ex-athletes and population con- anterior cruciate ligament and meniscus injuries: osteoarthritis. Am J Sports Med.
trols. Arthritis Rheum. 1996 Jun;39(6):988-95. 2007 Oct;35(10):1756-69. Epub 2007 Aug 29.
4. Marti B, Knobloch M, Tschopp A, Jucker A, Howald H. Is excessive running pre- 24. Shrier I. Muscle dysfunction versus wear and tear as a cause of exercise
dictive of degenerative hip disease? Controlled study of former elite athletes. BMJ. related osteoarthritis: an epidemiological update. Br J Sports Med. 2004 Oct;38
1989 Jul 8;299(6691):91-3. (5):526-35.
5. Panush RS, Schmidt C, Caldwell JR, Edwards NL, Longley S, Yonker R, Webster E, 25. Slemenda C, Brandt KD, Heilman DK, Mazzuca S, Braunstein EM, Katz BP,
Nauman J, Stork J, Pettersson H. Is running associated with degenerative joint Wolinsky FD. Quadriceps weakness and osteoarthritis of the knee. Ann Intern Med.
disease? JAMA. 1986 Mar 7;255(9):1152-4. 1997 Jul 15;127(2):97-104.
6. Konradsen L, Hansen EM, Søndergaard L. Long distance running and osteoar- 26. Hurley MV. The role of muscle weakness in the pathogenesis of osteoarthritis.
throsis. Am J Sports Med. 1990 Jul-Aug;18(4):379-81. Rheum Dis Clin North Am. 1999 May;25(2):283-98, vi.
7. Lane NE, Oehlert JW, Bloch DA, Fries JF. The relationship of running to osteoar- 27. Tashman S, Collon D, Anderson K, Kolowich P, Anderst W. Abnormal rotational
thritis of the knee and hip and bone mineral density of the lumbar spine: a 9 year knee motion during running after anterior cruciate ligament reconstruction. Am J
longitudinal study. J Rheumatol. 1998 Feb;25(2):334-41. Sports Med. 2004 Jun;32(4):975-83.
8. Sohn RS, Micheli LJ. The effect of running on the pathogenesis of osteoarthritis of 28. Tashman S, Kolowich P, Collon D, Anderson K, Anderst W. Dynamic function of
the hips and knees. Clin Orthop Relat Res. 1985 Sep;198:106-9. the ACL-reconstructed knee during running. Clin Orthop Relat Res. 2007
9. Teichtahl AJ, Wluka AE, Wang Y, Forbes A, Davies-Tuck ML, English DR, Giles GG, Jan;454:66-73.
Cicuttini FM. Effect of long-term vigorous physical activity on healthy adult knee 29. Barenius B, Ponzer S, Shalabi A, Bujak R, Norlén L, Eriksson K. Increased risk of
cartilage. Med Sci Sports Exerc. 2012 Jun;44(6):985-92. osteoarthritis after anterior cruciate ligament reconstruction: a 14-year follow-up
10. Mork PJ, Holtermann A, Nilsen TIL. Effect of body mass index and physical exer- study of a randomized controlled trial. Am J Sports Med. 2014 May;42(5):1049-57.
cise on risk of knee and hip osteoarthritis: longitudinal data from the Norwegian HUNT Epub 2014 Mar 18.
Study. J Epidemiol Community Health. 2012 Aug;66(8):678-83. Epub 2012 Apr 17. 30. Kessler MA, Behrend H, Henz S, Stutz G, Rukavina A, Kuster MS. Function,
11. Ageberg E, Engström G, Gerhardsson de Verdier M, Rollof J, Roos EM, osteoarthritis and activity after ACL-rupture: 11 years follow-up results of conser-
Lohmander LS. Effect of leisure time physical activity on severe knee or hip osteo- vative versus reconstructive treatment. Knee Surg Sports Traumatol Arthrosc. 2008
arthritis leading to total joint replacement: a population-based prospective cohort May;16(5):442-8.
study. BMC Musculoskelet Disord. 2012 May 17;13:73. 31. Tesarz J, Schuster AK, Hartmann M, Gerhardt A, Eich W. Pain perception in
12. Centers for Diseases Control and Prevention. Prevalence of doctor-diagnosed athletes compared to normally active controls: a systematic review with meta-
arthritis and arthritis-attributable activity limitation — United States, 2010–2012. analysis. Pain. 2012 Jun;153(6):1253-62.
2013 Nov 8. http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6244a1.htm. 32. Bruce B, Fries JF, Lubeck DP. Aerobic exercise and its impact on musculo-
Accessed 2017 Aug 29. skeletal pain in older adults: a 14 year prospective, longitudinal study. Arthritis Res
13. U.S. Bone and Joint Initiative. The burden of musculoskeletal diseases in the Ther. 2005;7(6):R1263-70. Epub 2005 Sep 19.
United States. Prevalence of arthritic conditions. 2014. http://www. 33. Chakravarty EF, Hubert HB, Lingala VB, Fries JF. Reduced disability and mortality
boneandjointburden.org/2014-report/ivb0/prevalence-arthritic-conditions. among aging runners: a 21-year longitudinal study. Arch Intern Med. 2008 Aug
Accessed 2017 Aug 29. 11;168(15):1638-46.
14. Hootman JM, Helmick CG, Barbour KE, Theis KA, Boring MA. Updated projected 34. Salacinski AJ, Krohn K, Lewis SF, Holland ML, Ireland K, Marchetti G. The
prevalence of self-reported doctor-diagnosed arthritis and arthritis-attributable activity effects of group cycling on gait and pain-related disability in individuals with mild-to-
limitation among US adults, 2015-2040. Arthritis Rheumatol. 2016 Jul;68(7):1582-7. moderate knee osteoarthritis: a randomized controlled trial. J Orthop Sports Phys
15. Miller RH, Edwards WB, Brandon SCE, Morton AM, Deluzio KJ. Why don’t most Ther. 2012 Dec;42(12):985-95. Epub 2012 Aug 2.
runners get knee osteoarthritis? A case for per-unit-distance loads. Med Sci Sports 35. Vopat BG, Klinge SA, McClure PK, Fadale PD. The effects of fitness on the aging
Exerc. 2014 Mar;46(3):572-9. process. J Am Acad Orthop Surg. 2014 Sep;22(9):576-85.
16. van den Bogert AJ, Read L, Nigg BM. An analysis of hip joint loading during 36. Claessens AA, Schouten JS, van den Ouweland FA, Valkenburg HA. Do clinical
walking, running, and skiing. Med Sci Sports Exerc. 1999 Jan;31(1):131-42. findings associate with radiographic osteoarthritis of the knee? Ann Rheum Dis.
17. Williams PT. Effects of running and walking on osteoarthritis and hip replace- 1990 Oct;49(10):771-4.
ment risk. Med Sci Sports Exerc. 2013 Jul;45(7):1292-7. 37. Davis MA, Ettinger WH, Neuhaus JM, Barclay JD, Segal MR. Correlates of knee
18. National Center for Health Statistics. National Health Interview Survey, 2010- pain among US adults with and without radiographic knee osteoarthritis. J Rheu-
2013. 2013. http://www.cdc.gov/nchs/nhis/data-questionnaires-documentation. matol. 1992 Dec;19(12):1943-9.
htm. Accessed 2017 Aug 25. 38. Bedson J, Croft PR. The discordance between clinical and radiographic knee
19. Murphy LB, Cisternas MG, Greenlund KJ, Giles W, Hannan C, Helmick CG. osteoarthritis: a systematic search and summary of the literature. BMC Muscu-
Defining arthritis for public health surveillance: methods and estimates in 4 US loskelet Disord. 2008 Sep 2;9:116.
population health surveys. Arthritis Care Res (Hoboken). 2017 Mar;69(3):356-67. 39. Sacks JJ, Harrold LR, Helmick CG, Gurwitz JH, Emani S, Yood RA. Valida-
20. Iacus SM, King G, Porro G. Causal inference without balance checking: coars- tion of a surveillance case definition for arthritis. J Rheumatol. 2005
ened exact matching. Polit Anal. 2011. Feb;32(2):340-7.