Healio
Healio
February 01, 2005
14 min read
Save

Epidemiology of Osteoarthritis

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

ABSTRACT

The majority of adults >55 years of age show radiographic evidence of osteoarthritis; however, determining definitive prevalence and incidence rates of knee osteoarthritis and hip osteoarthritis poses methodologic problems. Several factors play a role in osteoarthritis risk, including age, gender, genetics, behavioral influences, and ethnicity. Studies cited compare data from the Framingham Osteoarthritis Study with data from studies performed in Asian and European populations. A number of broad trends stand out, including higher prevalence in advanced age for both hip and knee osteoarthritis, with higher prevalence and incidence in women than in men.

Osteoarthritis is probably the leading cause of functional disability among adults in developed countries.1 Reliable estimates of the incidence and prevalence of osteoarthritis are difficult to obtain for a number of reasons, including differences in diagnostic criteria and variability in techniques used for data collection.

A number of difficulties arise in epidemiological studies of osteoarthritis. As Lawrence et al2,3 note in a study of arthritis patterns in the United States, the first challenge in assessing osteoarthritis prevalence is the widely recognized disparity between objective diagnosis, based on radiographic evidence, and patient-reported symptoms of pain or disability. It is not clear which of these measures is more “meaningful.” Patient-perceived pain intensity, duration, and frequency are likely strong determinants of the degree of disability in osteoarthritis and, in turn, of disease impact on the health care system and society at large. However, objective diagnoses (based on expert review of radiological evidence) provide more reproducible results and facilitate more accurate longitudinal or comparative studies.

The second challenge is the need to rely on clinical evaluation for broad-based population estimates because of the inability (on cost and possibly ethical grounds) to conduct multiple-joint radio-graphic assessment. Finally, whether for patient-reported, clinician-diagnosed, or radiography-based surveys, the degree of severity arises.2 Few adults aged >55 years have no joint pain, and most have some radiographic evidence of osteoarthritis.4 The decision to include or exclude mild osteoarthritis, as well as moderate-to-severe osteoarthritis, when estimating prevalence can profoundly affect the resulting data.

Even when radiographic evaluation is the basis for confirming the presence or absence of osteoarthritis, differences in diagnostic criteria can create disparities. For example, the radiological grading system for osteoarthritis developed by Kellgren and Lawrence4 is based on the presence of osteophytes, joint space narrowing, subchondral sclerosis, and bony cysts, whereas diagnostic criteria used by Ahlbäck5 for knee osteoarthritis rely more heavily on joint space narrowing, as well as bone attrition. Additional radiological criteria specifically intended for epidemiological studies have been developed for knee osteoarthritis6 and hip osteoarthritis.7

In addition to these methodological problems, in recent decades a profound shift in the pathophysiology of osteoarthritis has occurred. Osteoarthritis is no longer viewed as a disease of “wear and tear,” but is instead recognized as a result of an active disease process characterized by derangements in process involved in the maintenance of healthy cartilage.8,9 This shift has been accompanied by, and to some extent driven by, changes in imaging techniques (such as magnetic resonance imaging). For the most part, changing views of the osteoarthritis disease process have not yet been assimilated into diagnostic criteria used for epidemiological studies.8

Although the prevalence of knee osteoarthritis, especially hip osteoarthritis, appears to be lower than osteoarthritis in other locations such as the hand and wrist, these conditions have a disproportionate impact on patient disability and on the health care system.9,10 For that reason, this brief review will focus on knee and hip osteoarthritis.

Prevalence Estimates

Knee Osteoarthritis

Nearly all studies of knee osteoarthritis prevalence have focused on the tibio-femoral joint, rather than the patellofemoral joint.8 One of the hallmark studies of osteoarthritis prevalence is the Framing-ham Osteoarthritis Study, an outgrowth of the Framingham Heart Study, which was initiated in 1948 and is based on biennial examination of a large study population. Between 1983 and 1985, an evaluation of the cohort was undertaken that incorporated patient responses to a standardized question about knee osteoarthritis, physician assessment, and radiographic evidence.11

Based on the results of the radiographic survey, using Kellgren’s and Lawrence’s criteria,4 the trends shown are generally illustrative of other population-based surveys (Table 1). A clear relationship between knee osteoarthritis prevalence and age exists, with the overall prevalence rising from 11.5% in patients aged <70 years to 19.4% in those aged >80 years.11 This trend is similar to those reported in other surveys of osteoarthritis; in the study by Lawrence et al,2 whole-body osteoarthritis prevalence based on clinical evaluation is 3 to 4 times higher among the subset of patients aged ≥60 years than in the overall adult population aged >20 years.

Table 1
Prevalence of Knee Osteoarthritis Based on Radiographic Survey
Group NAll Knee Osteoarthritis (%)Grade 3-4 Knee Osteoarthritis (%)
All patients142033.015.7
<70 years53027.4*11.5*
70-79 years68434.117.8
≥80 years20643.719.4
Women83134.415.3
<70 years20325.1*10.6*
70-79 years41436.317.6
≥80 years11452.621.1
Men58930.916.0
<70 years22730.412.8
70-79 years27030.718.2
≥80 years9232.617.4
*P<.01 for linear trend showing increased prevalence with age.
Adapted from: Felson DT, Naimark A, Anderson J, Kazis L, Castelli W, Meenan RF. The prevalence of knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study. Arthritis Rheum. 1987; 30:914-918.11

Another trend revealed in the Framingham study is a higher prevalence of radiographic knee osteoarthritis among women than among men (34.4% vs. 30.9%), although the difference disappears when only grade 3 to 4 radiographic changes are considered (15.5% vs. 16.0%).11 However, the age dependence of knee osteoarthritis is much more pronounced among women than men. In women, the prevalence rises from 25.1% in women aged <70 years to 52.6% in women aged >80 years; the comparable rates in men are 30.4% and 32.6%, respectively. For grade 3 to 4 radiographic changes, the same trend persists, with prevalence rising from 10.6% in women aged <70 years to 21.1% in women aged >l80 years (an increase of 99%), and from 12.8% to 17.4% (an increase of 36%) among men.11

Generally, the gender disparity for knee osteoarthritis in the Framingham study is substantially less than that found in other studies.2,8 However, the trend toward increased age dependence among women is a consistent feature of knee osteoarthritis.8 This has led to speculation that estrogen exerts a chondroprotective effect that recedes during and following menopause.12 In support of this hypothesis, a follow-up to the Framingham study showed that the risk of knee osteoarthritis incidence and progression was 60% lower among elderly women who had used or were using estrogen replacement therapy than among women who had never taken estrogen, although the difference was not statistically significant.13 However, other studies have failed to demonstrate a relationship between hormone supplementation and reduced osteoarthritis risk.14

Another important contribution of the Framingham study was to delineate the degree of discrepancy between radio-graphic evidence of osteoarthritis and self-reported symptoms. The percentage of patients reporting symptoms rises from 7.6% among patients with no radiographic osteoarthritis to 40% among those with grade 3 to 4 disease (Table 2). Women report symptomatic osteoarthritis more frequently than men with a similar degree of radiographic osteoarthritis, although the gender difference becomes less significant with increasing radiographic severity.11

The relatively low rate of symptomatic knee osteoarthritis among patients with radiographic osteoarthritis, less than half even for the most severe disease by radiography, leads to reduced prevalence estimates when the combination of radio-graphic evidence and symptoms is used to establish the presence of disease. The rates of patients with both radiographic and symptomatic evidence of knee osteoarthritis are typically <15%, even among elderly populations; these figures are generally regarded as gross underestimates. In a US population study, underreading of radio-graphic evidence may also have contributed to lower prevalence estimates, along with symptomatic criteria (“daily pain”) that may be considered unreasonably high.2

The Framingham study has been mirrored by one conducted in Beijing, China, that used an identical protocol. The two studies permit comparisons across cultural and ethnic boundaries. For example, the Beijing study revealed a greater gender disparity than the Framingham study. Among Chinese men, the prevalence of radiographic osteoarthritis rises from 10% at ages 60 to 64 to 45.7% in Chinese men aged >80 years; comparable rates for Chinese women are 39.6% and 59.1%, respectively.15

The prevalence of knee osteoarthritis among Chinese men was similar to that in the Framingham study, whereas the prevalence among Chinese women was substantially higher than that in the Framingham study. This trend was observed both for radiographic and symptomatic osteoarthritis, and at roughly the same degree; prevalence among Chinese women is about 40% to 45% higher than among women in the United States.15

Table 2
Association Between Radiographic Grade of Knee Osteoarthritis and Reported Symptoms
Osteoarthritis Grade Reporting Symptoms (overall) (%)Men Reporting Symptoms (%)Women Reporting Symptoms (%)
0 (no disease)7.64.49.6
1 (questionable)10.88.513.0
2 (osteophytes)19.29.124.8
3 or 4 (definite)40.034.843.4
Adapted from: Felson DT, Naimark A, Anderson J, Kazis L, Castelli W, Meenan RF. The prevalence of knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study. Arthritis Rheum. 1987; 30:914-918.11

To some extent, the patterns of knee osteoarthritis in Beijing are puzzling, given the relative leanness of elderly patients in China compared with those in the United States. A more detailed collaborative review of radiographic evidence from both studies revealed that Chinese patients were more than twice as likely as US patients to experience lateral knee osteoarthritis, while the rates of medial knee osteoarthritis were roughly comparable. It is suggested that racial differences in anatomic alignment may explain all or some of these observed differences.16

In a follow-up to the Beijing study, a strong association was demonstrated between the risk of knee osteoarthritis and squatting — a far more common daily posture in China than in the United States. Patients who reported squatting >2 hours a day at age 25 were more than twice as likely to demonstrate radiographic osteoarthritis as those who reported squatting <30 minutes per day. Squatting >1 hour a day at age 25 was much more common among Chinese women than Chinese men (68% vs. 40%), which may partially explain the more pronounced gender disparity revealed in the Beijing epidemiologic study.15,17

Hip Osteoarthritis

There is substantial evidence that hip osteoarthritis and knee osteoarthritis are distinct pathological entities; one important observation being that patients requiring total hip replacement (THR) rarely require total knee replacement, and vice versa.18 In addition, population surveys have shown a lower prevalence for hip osteoarthritis than for knee osteoarthritis.8

Lawrence et al2 report that the US prevalence of moderate to severe hip osteoarthritis among adults aged ≥55 years, based on data from the first National Health and Nutrition Examination Survey (NHANES I) normalized to the 1990 population, was 1.4% among both men and women. When mild osteoarthritis was included, the prevalence increased to 3.5% and 2.8% for men and women, respectively.

In a study of 6321 patients aged >40 years in Denmark, an overall hip osteoarthritis prevalence estimate of 4.7% was derived from radiographic analyses of colon-screening radiographs. Although mild hip osteoarthritis occurred at about the same rate in both sexes, severe osteoarthritis was almost twice as prevalent among women as among men, with total prevalence rates of 5.6% and 3.7%, respectively. Nearly half (44%) of patients with hip osteoarthritis demonstrated bilateral involvement.19

A Dutch study of 6585 village residents illustrated the sharp age-dependent rise in hip osteoarthritis prevalence, as well as the pattern for increased prevalence among women. Among women, hip osteoarthritis prevalence rose from 2.6% at ages 55 to 59 to 14.8% at ages 75 to 79; the corresponding figures for men were 5.9% and 10.2%, respectively.20

Analyses of hip osteoarthritis prevalence across national and ethnic boundaries have revealed interesting, and occasionally surprising, patterns of occurrence. The prevalence of hip osteoarthritis in Iceland, based on colon radiographs, appears to be about 5 times higher than that observed in surveys of southern Scandinavia, although the two populations are closely related genetically. Despite similar criteria for disease, 8% of Icelandic people aged >35 years demonstrated radiographic evidence of hip osteoarthritis, compared with 1.2% in a similar population from Malmo, Sweden.21

The historic isolation of Iceland, as well as its well-documented family genealogies, provides fertile opportunity to dissect the contribution of hereditary, environmental, and behavioral influences on the development of hip osteoarthritis. A study of THR patterns in Iceland illustrated strong familial associations. In addition, genomic analysis of a family with high rates of hip osteoarthritis identified a locus on chromosome 16p associated with elevated osteoarthritis risk. An association between the same genetic locus and hip osteoarthritis had previously been demonstrated in an English study, which suggests that a common genetic basis may underlie the development of hip osteoarthritis in some patients.22

Inoue et al23 conducted a comparative study of hip osteoarthritis among 401 French and 782 Japanese adults aged 20 to 79 years, based on radiographic evidence from consecutive admissions for intravenous urography. The prevalence of hip osteoarthritis was highest among French men, followed in order by Japanese women, French women, and Japanese men; French men had a risk for hip osteoarthritis nearly 4 times that of Japanese men.

In a review of hip osteoarthritis studies, Hoaglund and Steinbach18 conclude that the prevalence of moderate-to-severe hip osteoarthritis is significantly higher among Caucasians than among East Indians, Hong Kong Chinese, and Native Americans. They also report remarkable constancy of observed rates over 4 decades, inferring that little change has occurred with regard to genetic and environmental factors involved in osteoarthritis development during that time.

Incidence Estimates

Few studies of incidence rates in osteoarthritis exist. The methodological issues described above with regard to disease definition are made even more difficult when combined with attempts to define disease onset. Longitudinal radio-graphic studies usually involve widely separated time points, which makes extensive interpolation necessary. In addition, mortality among the elderly patients most profoundly affected by osteoarthritis can limit the evaluable patient set and can introduce a confounding influence if mortality risk is associated in any way with the presence of osteoarthritis.

Despite these difficulties, effort has been made to characterize osteoarthritis incidence in elderly populations. As with prevalence estimates, some of the best data are derived from the Framingham study of knee osteoarthritis. Longitudinal comparison of radiographic and symptomatic evidence from 1983 to 1985 and 1992 to 1993 (mean interval 8.1 years) showed that among women, there was a mean incidence of ~2% per year for radiographic knee osteoarthritis and ~1% for symptomatic knee osteoarthritis. The same evidence demonstrated a risk of osteoarthritis progression of 4% per year. The overall rate of incident knee osteoarthritis was ~1.7 times greater among women than men, which is consistent with patterns illustrated in the Framingham prevalence studies. The incidence rate did not appear to vary with age.24

A longitudinal study of members of an HMO, using radiographic evidence (Kellgren-Lawrence grade 2+) plus symptomatic evidence as disease criteria, produced incidence estimates of 240/100,000 person-years (0.24%) for knee osteoarthritis and 88/100,000 person-years (0.088% per year) for hip osteoarthritis. The stringent disease criteria (radiographic plus symptomatic evidence) and broad age range may help explain some of the variance with rates reported in the Framingham study; in women aged 70 to 89 years, the incidence rate approached 1% per year. As in the Framingham results, the incidence of osteoarthritis among women was consistently higher than among men; for both sexes, incidence increased with age but leveled off around age 80.25

One potential advantage of incidence studies is that they may provide more accurate assessment of the contribution of various risk factors to osteoarthritis because the presence of a specific factor is more closely associated temporally with the development of disease. In this regard, data from the Framingham study have demonstrated strong independent positive correlations between several risk factors and the risk of incident disease. The most compelling associations were found for body mass index (odds ratio [OR]: 1.6 per 5 kg/m2); weight change (OR: 1.4 per 10-lb increase); and physical activity (OR: 3.3 for highest vs. lowest quartile).26

Another possible approach to estimating incidence rates for osteoarthritis is to extrapolate from treatment rates for end-stage disease, such as for total hip or knee replacements.8 A study of THR rates in Hawaii was particularly revealing in this regard, as it enabled cross-ethnic comparisons. As with prevalence estimates for hip osteoarthritis, the risk of THR was significantly higher among whites than among Asian ethnic groups (including Japanese, Chinese, Filipino, and Hawaiian). Among women aged 40 to 84, THR risk was 4.4% for whites vs. 1.1% and 1.7% for age-matched Japanese and Chinese women, respectively. In addition, hip osteoarthritis was found to be a more important contributor to THR rates among whites than among other ethnic groups: hip osteoarthritis accounted for 59% and 66% of THR in white women and men, respectively, compared with 36% and 30% for Japanese women and men.27

Conclusion

Despite the difficulties involved in assessing the prevalence and incidence of knee osteoarthritis and hip osteoarthritis on a population basis, a number of broad trends stand out in review studies that have addressed these issues. These include a marked age dependence among both sexes for prevalence and higher prevalence rates among elderly women than men of the same age. In addition, the degree of age dependence for hip and knee osteoarthritis, reflected in both prevalence and incidence studies, is greater among women than men. The risks for both knee and hip osteoarthritis vary extensively between ethnic groups and even between subpopulations of the same broad ethnic group; these differences appear to arise from both genetic and behavioral influences.

The need to more fully understand the drivers of osteoarthritis risk in general, and of knee and hip osteoarthritis specifically, has never been greater. The number of people and the proportion of the US population affected by arthritis have been projected to grow significantly, from a 1995 estimate of 40 million (15% of the population) to a projected 59.4 million (18.2% of the population) in 2020, an increase of nearly 50% in patient load.2 Better understanding of the factors contributing to the development of these age-dependent disease states, in part facilitated by epidemiologic studies, is critical to ameliorating their impact on patients, families, and society.

References

  1. Prevalence of disabilities and associated health conditions among adults-United States, 1999. MMWR Morb Mortal Wkly Rep. 23 2001; 50:120-125.
  2. Lawrence RC, Helmick CG, Arnett FC, et al. Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Rheum. 1998; 41:778-799.
  3. Kean WF, Kean R, Buchanan WW. Osteo-arthritis: symptoms, signs and source of pain. Inflammopharmacology. 2004; 12:3-31.
  4. Kellgren JH, Lawrence JS. Osteo-arthrosis and disk degeneration in an urban population. Ann Rheum Dis. 1958; 17:388-397.
  5. Ahlbäck S. Osteoarthrosis of the knee. A radiographic investigation. Acta Radiol Diagn (Stockh). 1968; (suppl 277):7-72.
  6. Spector TD, Hart DJ, Byrne J, Harris PA, Dacre JE, Doyle DV. Definition of osteoarthritis of the knee for epidemiological studies. Ann Rheum Dis. 1993; 52:790-794.
  7. Croft P, Cooper C, Wickham C, Coggon D. Defining osteoarthritis of the hip for epidemiologic studies. Am J Epidemiol. 1990; 132:514-522.
  8. Petersson I, Jacobsson L. Osteoarthritis of the peripheral joints. Best Pract Res Clin Rheumatol. 2002; 16:741-760.
  9. American Academy of Orthopaedic Surgeons. Improving Musculoskeletal Care in America. Osteoarthritis of the Hip. Available at: www3.aaos.org/research/imca/OAHip_Overview.htm. Accessed September 24, 2004.
  10. American Academy of Orthopaedic Surgeons. Improving Musculoskeletal Care in America. Osteoarthritis of the Knee. Available at: www3.aaos.org/research/imca/oa_knee_overview_final.htm. Accessed September 24, 2004.
  11. Felson DT, Naimark A, Anderson J, Kazis L, Castelli W, Meenan RF. The prevalence of knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study. Arthritis Rheum. 1987; 30:914-918.
  12. Felson DT, Nevitt MC. The effects of estrogen on osteoarthritis. Curr Opin Rheumatol. 1998; 10:269-272.
  13. Zhang Y, McAlindon TE, Hannan MT, et al. Estrogen replacement therapy and worsening of radiographic knee osteoarthritis: the Framingham Study. Arthritis Rheum. 1998; 41:1867-1873.
  14. Erb A, Brenner H, Gunther KP, Sturmer T. Hormone replacement therapy and patterns of osteoarthritis: baseline data from the Ulm Osteoarthritis Study. Ann Rheum Dis. 2000; 59:105-109.
  15. Zhang Y, Xu L, Nevitt MC, et al. Comparison of the prevalence of knee osteoarthritis between the elderly Chinese population in Beijing and whites in the United States: the Beijing Osteoarthritis Study. Arthritis Rheum. 2001; 44:2065-2071.
  16. Felson DT, Nevitt MC, Zhang Y, et al. High prevalence of lateral knee osteoarthritis in Beijing Chinese compared with Framingham Caucasian subjects. Arthritis Rheum. 2002; 46:1217-1222.
  17. Zhang Y, Hunter DJ, Nevitt MC, et al. Association of squatting with increased prevalence of radiographic tibiofemoral knee osteoarthritis: the Beijing Osteoarthritis Study. Arthritis Rheum. 2004; 50:1187-1192.
  18. Hoaglund FT, Steinbach LS. Primary osteoarthritis of the hip: etiology and epidemiology. J Am Acad Orthop Surg. 2001; 9:320-327.
  19. Jorring K. Osteoarthritis of the hip. Epidemiology and clinical role. Acta Orthop Scand. 1980; 51:523-530.
  20. Van Saase JL, van Romunde LK, Cats A, Vandenbroucke JP, Valkenburg HA. Epidemiology of osteoarthritis: Zoetermeer survey. Comparison of radiological osteoarthritis in a Dutch population with that in 10 other populations. Ann Rheum Dis. 1989; 48:271-280.
  21. Ingvarsson T, Hagglund G, Lohmander LS. Prevalence of hip osteoarthritis in Iceland. Ann Rheum Dis. 1999; 58:201-207.
  22. Ingvarsson T. Prevalence and inheritance of hip osteoarthritis in Iceland. Acta Orthop Scand Suppl. 2000; 298:1-46.
  23. Inoue K, Wicart P, Kawasaki T, et al. Prevalence of hip osteoarthritis and acetabular dysplasia in French and Japanese adults. Rheumatology (Oxford). 2000; 39:745-748.
  24. Felson DT, Zhang Y, Hannan MT, et al. The incidence and natural history of knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study. Arthritis Rheum. 1995; 38:1500-1505.
  25. Oliveria SA, Felson DT, Reed JI, Cirillo PA, Walker AM. Incidence of symptomatic hand, hip, and knee osteoarthritis among patients in a health maintenance organization. Arthritis Rheum. 1995; 38:1134-1141.
  26. Felson DT, Zhang Y, Hannan MT, et al. Risk factors for incident radiographic knee osteoarthritis in the elderly: the Framingham Study. Arthritis Rheum. 1997; 40:728-733.
  27. Oishi CS, Hoaglund FT, Gordon L, Ross PD. Total hip replacement rates are higher among Caucasians than Asians in Hawaii. Clin Orthop. 1998; 353:166-174.

Author

From the University of Colorado Health Sciences Center, Denver, Colo.

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.