Advertisement

Substantial rise in the lifetime risk of primary total knee replacement surgery for osteoarthritis from 2003 to 2013: an international, population-level analysis

Open ArchivePublished:November 14, 2016DOI:https://doi.org/10.1016/j.joca.2016.11.005

      Summary

      Objective

      To estimate and compare the lifetime risk of total knee replacement surgery (TKR) for osteoarthritis (OA) between countries, and over time.

      Method

      Data on primary TKR procedures performed for OA in 2003 and 2013 were extracted from national arthroplasty registries in Australia, Denmark, Finland, Norway and Sweden. Life tables and population data were also obtained for each country. Lifetime risk of TKR was calculated for 2003 and 2013 using registry, life table and population data.

      Results

      Marked international variation in lifetime risk of TKR was evident, with females consistently demonstrating the greatest risk. In 2013, Finland had the highest lifetime risk for females (22.8%, 95%CI 22.5–23.1%) and Australia had the highest risk for males (15.4%, 95%CI 15.1–15.6%). Norway had the lowest lifetime risk for females (9.7%, 95%CI 9.5–9.9%) and males (5.8%, 95%CI 5.6–5.9%) in 2013. All countries showed a significant rise in lifetime risk of TKR for both sexes over the 10-year study period, with the largest increases observed in Australia (females: from 13.6% to 21.1%; males: from 9.8% to 15.4%).

      Conclusions

      Using population-based data, this study identified significant increases in the lifetime risk of TKR in all five countries from 2003 to 2013. Lifetime risk of TKR was as high as 1 in 5 women in Finland, and 1 in 7 males in Australia. These risk estimates quantify the healthcare resource burden of knee OA at the population level, providing an important resource for public health policy development and healthcare planning.

      Keywords

      Introduction

      Knee osteoarthritis (OA) represents a significant public health challenge internationally. The increasing burden of knee OA worldwide is evident from the results of the Global Burden of Disease Study
      • Cross M.
      • Smith E.
      • Hoy D.
      • Nolte S.
      • Ackerman I.
      • Fransen M.
      • et al.
      The global burden of hip and knee osteoarthritis: estimates from the Global Burden of Disease 2010 study.
      . This landmark study highlighted a major shift in the global burden of disease over the past 20 years from infectious diseases to non-communicable diseases including musculoskeletal conditions that are associated with significant disability
      • Murray C.J.L.
      • Vos T.
      • Lozano R.
      • Naghavi M.
      • Flaxman A.D.
      • Michaud C.
      • et al.
      Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010.
      . This is supported by data from a range of developed countries that show steady growth in the rate of knee replacement surgeries performed predominantly for severe knee OA over the past two decades
      • Kurtz S.
      • Mowat F.
      • Ong K.
      • Chan N.
      • Lau E.
      • Halpern M.
      Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002.
      • Singh J.A.
      Epidemiology of knee and hip arthroplasty: a systematic review.
      • Australian Orthopaedic Association National Joint Replacement Registry
      Annual Report – Hip and Knee Arthroplasty.
      . Total knee replacements (TKRs) represent the majority of procedures performed, with only a small proportion of patients receiving unicompartmental knee replacement (UKR)
      • Australian Orthopaedic Association National Joint Replacement Registry
      Annual Report – Hip and Knee Arthroplasty.
      • Robertsson O.
      • Bizjajeva S.
      • Fenstad A.M.
      • Furnes O.
      • Lidgren L.
      • Mehnert F.
      • et al.
      Knee arthroplasty in Denmark, Norway and Sweden: a pilot study from the Nordic Arthroplasty Register Association.
      . While joint replacement surgery is cost-effective
      • Losina E.
      • Walensky R.P.
      • Kessler C.L.
      • Emrani P.S.
      • Reichmann W.M.
      • Wright E.A.
      • et al.
      Cost-effectiveness of total knee arthroplasty in the United States: patient risk and hospital volume.
      , planning for future healthcare demand is critical and requires robust population-level data on disease burden and healthcare utilisation.
      Estimating the lifetime risk of joint replacement surgery is an evolving area within musculoskeletal epidemiology. This statistical approach is commonly used in the cardiovascular and cancer fields
      • Leening M.J.G.
      • Ferket B.S.
      • Steyerberg E.W.
      • Kavousi M.
      • Deckers J.W.
      • Nieboer D.
      • et al.
      Sex differences in lifetime risk and first manifestation of cardiovascular disease: prospective population based cohort study.
      • Ahmad A.S.
      • Ormiston-Smith N.
      • Sasieni P.D.
      Trends in the lifetime risk of developing cancer in Great Britain: comparison of risk for those born from 1930 to 1960.
      . The lifetime risk of TKR refers to the probability of having this surgical procedure over an individual's lifetime. Lifetime risk estimates provide a complementary approach to quantifying population-level disease burden and related use of healthcare services, and can be easily interpreted by health policymakers, clinicians and patients (as they are expressed as percentages). A key advantage of the lifetime risk statistic is that it provides a cumulative measure of risk that incorporates population life expectancy and all-cause mortality.
      Data on the lifetime risk of TKR surgery are limited. Research from the United Kingdom found that the lifetime risk of TKR had increased markedly over a 15-year period from 1991 to 2006, particularly for women
      • Culliford D.J.
      • Maskell J.
      • Kiran A.
      • Judge A.
      • Javaid M.K.
      • Cooper C.
      • et al.
      The lifetime risk of total hip and knee arthroplasty: results from the UK General Practice Research Database.
      . In the United States, Weinstein et al.
      • Weinstein A.M.
      • Rome B.N.
      • Reichmann W.M.
      • Collins J.E.
      • Burbine S.A.
      • Thornhill T.S.
      • et al.
      Estimating the burden of total knee replacement in the United States.
      used national health survey data to estimate the cumulative lifetime risk of TKR, although changes in risk over time were not evaluated. Most recently, Bohensky and colleagues used hospital administrative data to estimate the lifetime risk of TKR in the state of Victoria, Australia
      • Bohensky M.
      • Ackerman I.
      • de Steiger R.
      • Gorelik A.
      • Brand C.
      Lifetime risk of total knee replacement and temporal trends in incidence by health care setting, socioeconomic status and geographic location.
      . A clear increase in the lifetime risk of TKR was evident over a nine-year period (1999–2008), most notably for females. Previous studies investigating the lifetime risk of TKR have all obtained data on joint replacement utilisation from observational studies or health system administrative datasets, which have known limitations around generalisability, completeness and accuracy. The use of population-based procedure data from national arthroplasty registries with almost complete coverage would enable more precise estimates of the lifetime risk of TKR.
      While a number of earlier studies have compared TKR incidence rates or utilisation rates between countries
      • Singh J.A.
      Epidemiology of knee and hip arthroplasty: a systematic review.
      • Robertsson O.
      • Bizjajeva S.
      • Fenstad A.M.
      • Furnes O.
      • Lidgren L.
      • Mehnert F.
      • et al.
      Knee arthroplasty in Denmark, Norway and Sweden: a pilot study from the Nordic Arthroplasty Register Association.
      • Kurtz S.M.
      • Ong K.L.
      • Lau E.
      • Widmer M.
      • Maravic M.
      • Gomez-Barrena E.
      • et al.
      International survey of primary and revision total knee replacement.
      • Ravi B.
      • Croxford R.
      • Reichmann W.M.
      • Losina E.
      • Katz J.N.
      • Hawker G.A.
      The changing demographics of total joint arthroplasty recipients in the United States and Ontario from 2001 to 2007.
      • Pabinger C.
      • Lothaller H.
      • Geissler A.
      Utilization rates of knee-arthroplasty in OECD countries.
      , an international comparison of the lifetime risk of TKR has not been undertaken. The present study aimed to:
      • estimate and compare the lifetime risk of primary TKR for OA in five countries;
      • describe change in lifetime risk over a ten-year period (2003–2013); and
      • examine changes in utilisation rates of primary TKR and UKR performed for OA over time.

      Methods

      Study design

      A multi-national, population-level retrospective analysis was undertaken.

      Data sources

      We obtained data on all primary TKR and UKR procedures performed for OA from 1 January 2003 to 31 December 2003 and 1 January 2013 to 31 December 2013 in Australia, Denmark, Finland, Norway and Sweden. These countries were selected for their longstanding and comprehensive national arthroplasty registries. The years 2003 and 2013 were chosen to align with the most recent life table data available across all five countries. De-identified, aggregate data on the number of surgical procedures and the number of patients receiving TKR and UKR in each year were obtained from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR), the Danish Knee Arthroplasty Register, the Finnish Arthroplasty Register (Finnish UKR data were obtained from the Finnish Hospital Discharge Register), the Norwegian Arthroplasty Register, and the Swedish Knee Arthroplasty Register. These registries collect data from public and private hospitals and all report over 95% registration completeness for primary joint replacement procedures. Approval for accessing AOANJRR data was obtained from The University of Melbourne Human Research Ethics Committee and the AOANJRR Data Review Committee. The study was also approved by the Nordic Arthroplasty Register Association. Ethics approval was not required for Denmark, Finland, Norway or Sweden, in line with local legislation.
      Extracted registry data for each country included:
      • sex;
      • age; and
      • operation type: TKR and UKR.
      Life table data for 2003 and 2013 (stratified by gender) were obtained online from the Australian Bureau of Statistics, Statistics Denmark, Statistics Norway, and Statistics Sweden. Life table data for Finland were obtained online from Eurostat, the statistical office of the European Union. Life tables use all-cause mortality rates to estimate the number of people alive at each year of age (age range 0–100 years) for a hypothetical cohort of 100,000 people. Data on the population of each country (by age and sex) and life expectancy for 2003 and 2013 were obtained from the above sources and OECD.Stat

      OECD.Stat. Health status. Available from: https://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT. [Accessed 26 September 2016].

      , respectively.

      Data analysis

      Data were categorised into pre-specified age groups for analysis: <40 years, 40–49 years, 50–59 years, 60–69 years, 70–79 years and ≥80 years. A ‘standardised lifetime risk’ calculation incorporating age-specific rates
      • Sasieni P.D.
      • Adams J.
      Standardized lifetime risk.
      was used to calculate the lifetime risk of primary TKR, accounting for potential differences in population size and life expectancy between countries (Supplementary material). Simultaneous bilateral TKR was counted as one TKR procedure to avoid potential over-estimation of lifetime risk. Where staged (non-simultaneous) bilateral TKR procedures were performed within the same year, only the first procedure was included in the dataset.
      The lifetime risk of TKR was calculated for each age group by dividing the total number of people having TKR procedures in that year (obtained from registry data) by the age group-specific and sex-specific population, and then multiplying these rates by the total number of people expected to be alive at the beginning of the interval (obtained from life table data). Lifetime risk of TKR was calculated for 2003 and 2013, with separate calculations undertaken for males and females due to known gender differences in knee OA prevalence and surgery rates
      • Kurtz S.
      • Mowat F.
      • Ong K.
      • Chan N.
      • Lau E.
      • Halpern M.
      Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002.
      • Ravi B.
      • Croxford R.
      • Reichmann W.M.
      • Losina E.
      • Katz J.N.
      • Hawker G.A.
      The changing demographics of total joint arthroplasty recipients in the United States and Ontario from 2001 to 2007.
      • Srikanth V.K.
      • Fryer J.L.
      • Zhai G.
      • Winzenberg T.M.
      • Hosmer D.
      • Jones G.
      A meta-analysis of sex differences prevalence, incidence and severity of osteoarthritis.
      . Confidence intervals (95%CI) were estimated using Poisson models
      • Sasieni P.D.
      • Adams J.
      Standardized lifetime risk.
      . Changes in lifetime risk of TKR over time and comparison of lifetime risk estimates between countries were analysed descriptively, using calculated confidence intervals. Lifetime risk of UKR was not calculated due to the small number of procedures performed. Instead, a sensitivity analysis (Supplementary material) was undertaken to estimate the combined lifetime risk of TKR and UKR in 2003 and 2013 in each country (using the same methods as for the TKR-only analyses).
      Similar to previous methods
      • Pabinger C.
      • Lothaller H.
      • Geissler A.
      Utilization rates of knee-arthroplasty in OECD countries.
      • Pabinger C.
      • Geissler A.
      Utilization rates of hip arthroplasty in OECD countries.
      , overall and age-based utilisation rates for TKR were calculated for each country in 2003 and 2013 by summing the count of procedures from each registry and dividing by the relevant population (with regard to gender and age group) for that year. These are reported as TKR utilisation rates per 100,000 population, with separate calculations for males and females. Where bilateral TKRs were performed, these were counted as two procedures to avoid underestimating the true utilisation of TKR. For UKR, only overall utilisation rates were calculated due to the relatively small numbers of procedures performed.

      Results

      Population characteristics and demographics of TKR

      Table I summarises the population characteristics for each country. While population size varied substantially across the five countries, the gender distribution was similar. Life expectancy was comparable across the countries and all countries experienced an increase in life expectancy from 2003 to 2013. Demographic data relating to primary TKR use are also presented in Table I. In 2003 and 2013, the majority of TKR procedures in each country were undertaken for females. The proportion of TKRs performed for people aged ≤60 years increased over time for all countries, from 15.8% to 17.1% in Australia, from 17.2% to 17.7% in Denmark, from 12.7% to 17.4% in Finland, from 11.3% to 16.4% in Norway, and from 13.1% to 16.1% in Sweden. The proportion of TKRs performed for the oldest individuals (those aged ≥80 years) decreased over the 10-year period in all countries except Finland (Table I). In 2003, the majority of TKR procedures in each country were performed for the 70–79 age group (Table I). In 2013, this was still evident for Denmark, Finland and Sweden, although TKR was most frequently performed for the 60–69 age group in Australia and Norway at this time point.
      Table IPopulation characteristics and TKR demographics
      CountryPopulation dataTKR data from registries
      Population size% FemaleLife expectancy
      Data on population life expectancy at birth were obtained from OECD.Stat13.
      Number of primary TKR
      Bilateral procedures performed within the same year were counted as two TKRs.
      % Female
      Proportion of those who received primary TKR at each time point.
      % Aged <40 years
      Proportion of those who received primary TKR at each time point.
      % Aged 40–49 years
      Proportion of those who received primary TKR at each time point.
      % Aged 50–59 years
      Proportion of those who received primary TKR at each time point.
      % Aged 60–69 years
      Proportion of those who received primary TKR at each time point.
      % Aged 70–79 years
      Proportion of those who received primary TKR at each time point.
      % Aged ≥80 years
      Proportion of those who received primary TKR at each time point.
      Australia
       200319,720,73750.480.3 years20,98657.20.21.913.730.840.113.3
       201323,125,86850.282.2 years42,91956.60.11.815.238.332.811.8
      Denmark
       20035,387,17450.577.4 years2,90865.70.32.214.831.736.814.4
       20135,605,83650.480.4 years6,10761.20.22.714.836.036.59.8
      Finland
       20035,219,73251.178.5 years6,09069.80.11.111.529.147.011.3
       2013
      Three TKR patients (n = 4 TKR procedures) from Finland were excluded from these analyses due to missing data on gender.
      5,451,27050.881.1 years9,56964.30.21.715.533.835.913.0
      Norway
       20034,552,25250.479.6 years2,16871.70.01.310.026.244.418.0
       20135,051,27549.881.8 years4,01062.30.12.413.937.434.611.5
      Sweden
       20038,975,67050.580.3 years6,65661.40.11.012.029.542.914.5
       20139,644,86450.182.0 years12,12456.80.12.213.836.237.010.7
      Data on population life expectancy at birth were obtained from OECD.Stat
      • Kurtz S.M.
      • Ong K.L.
      • Lau E.
      • Widmer M.
      • Maravic M.
      • Gomez-Barrena E.
      • et al.
      International survey of primary and revision total knee replacement.
      .
      Bilateral procedures performed within the same year were counted as two TKRs.
      Proportion of those who received primary TKR at each time point.
      § Three TKR patients (n = 4 TKR procedures) from Finland were excluded from these analyses due to missing data on gender.

      Comparison of lifetime risk of TKR between countries

      Table II presents the lifetime risk of TKR for males and females in each country in 2003 and 2013. Overall, lifetime risk varied considerably across the countries. In 2003, the lifetime risk of TKR for females ranged from 5.84% (in Denmark) to 19.21% (in Finland), and the lifetime risk for males ranged from 2.76% (in Norway) to 9.77% (in Australia). Across all five countries, females had a consistently higher lifetime risk of surgery. This was most evident in Finland, where lifetime risk of TKR for females was more than double the risk for males in 2003 (19.21% vs 7.91%; P < 0.05).
      Table IIBetween-country variation in lifetime risk of TKR
      CountryLifetime risk (95%CI)
      FemalesMales
      Australia
       200313.63 (13.40–13.86)9.77 (9.58–9.97)
       201321.13 (20.85–21.42)15.37 (15.13–15.61)
      Denmark
       20035.84 (5.69–5.99)3.10 (2.99–3.21)
       201310.85 (10.65–11.06)6.76 (6.60–6.93)
      Finland
       200319.21 (18.94–19.49)7.91 (7.74–8.09)
       2013
      Three TKR patients from Finland were excluded from the 2013 analyses due to missing data on gender.
      22.79 (22.49–23.08)11.68 (11.47–11.89)
      Norway
       20036.59 (6.43–6.75)2.76 (2.66–2.87)
       20139.70 (9.50–9.89)5.78 (5.63–5.93)
      Sweden
       20037.70 (7.53–7.87)4.93 (4.79–5.07)
       201311.81 (11.59–12.02)8.87 (8.69–9.06)
      Data are presented as percentages.
      Simultaneous bilateral TKR was counted as one TKR procedure to avoid potential over-estimation of lifetime risk. Where staged bilateral TKR procedures were performed within the same year, only the first procedure was included in the dataset.
      Three TKR patients from Finland were excluded from the 2013 analyses due to missing data on gender.
      In 2013, the lowest lifetime risk for females was seen in Norway and the greatest lifetime risk for females was in Finland, closely followed by Australia (Table II). For males, the lowest lifetime risk of TKR was in Norway and the highest lifetime risk was in Australia. Similar to the 2003 data, females consistently demonstrated a higher lifetime risk of TKR across all countries in 2013. The difference in lifetime risk between sexes was greatest in Finland, where the risk for females in 2013 was almost double the risk for males (22.79% vs 11.68%; P < 0.05).

      Changes in lifetime risk of TKR over time

      Each country demonstrated a significant increase in the lifetime risk of TKR from 2003 to 2013 for both females and males. For females, the greatest absolute increases in lifetime risk over time were evident for Australia and Denmark (Fig. 1), while Finland and Norway had the smallest absolute change. All five countries also demonstrated significant increases in the lifetime risk of TKR for males over time (Fig. 2). Australia had the greatest absolute increase, while the other countries showed smaller absolute increases.
      Fig. 1
      Fig. 1Changes in lifetime risk of TKR over time for females. Whiskers indicate 95%CI; P < 0.05 for all 2003–2013 comparisons. For these estimates, simultaneous bilateral TKR was counted as one TKR procedure to avoid potential over-estimation of lifetime risk. Where staged bilateral TKR procedures were performed within the same year, only the first procedure was included in the dataset.
      Fig. 2
      Fig. 2Changes in lifetime risk of TKR over time for males. Whiskers indicate 95%CI; P < 0.05 for all 2003–2013 comparisons. For these estimates, simultaneous bilateral TKR was counted as one TKR procedure to avoid potential over-estimation of lifetime risk. Where staged bilateral TKR procedures were performed within the same year, only the first procedure was included in the dataset.
      Sensitivity analyses incorporating both TKR and UKR data produced similar results (Supplementary material), with marked between-country variation and significant increases over time in the combined lifetime risk of TKR and UKR observed for both sexes in all countries.

      Age-specific utilisation rates for primary TKR

      For all countries, the greatest TKR utilisation rates were observed for people aged between 70 and 79 years and this was evident for both sexes and at both time points (Table III). Across the countries, females aged 70–79 years in Finland experienced the highest rate of TKR (1770 procedures per 100,000 population in 2013). This rate was over 1.5 times higher than the utilisation rate for similarly-aged females in Australia and approximately three times higher than the rate for 70–79 year old females in Norway, Denmark, and Sweden. While utilisation rates for people aged ≤50 years were low, each country demonstrated an increase in TKR rates over time for females and males aged 40–49 years, 50–59 years, 60–69 years, 70–79 years and ≥80 years (Table III).
      Table IIIComparison of age-specific utilisation rates for TKR
      CountryUtilisation rate
      The overall utilisation rate was calculated using the total number of procedures for females (or males) as the numerator and the number of females (or males) in the population as the denominator. Age-specific utilisation rates were calculated using the number of procedures for each age group as the numerator and the age-specific population as the denominator. Bilateral procedures performed within the same year were counted as two TKRs for calculating utilisation rates to avoid underestimating the true utilisation of TKR.
      per 100,000 people
      Overall<40 years40–49 years50–59 years60–69 years70–79 years≥80 years
      Australia
       Females 2003121015128450789427
       Females 20132090272517661135575
       Males 200392012108371656447
       Males 2013162021193650907565
      Denmark
       Females 20037001074204381206
       Females 2013132024153369612279
       Males 2003370640146208164
       Males 20138511795261439222
      Finland
      Four TKR procedures from Finland were excluded from the 2013 analyses due to missing data on gender.
       Females 2003159091184411720218
       Females 20132221252385331770301
       Males 2003720759249917121
       Males 20131270231573391081210
      Norway
       Females 2003680549209414222
       Females 201399019108323537235
       Males 2003270425102189129
       Males 2013600769224361164
      Sweden
       Females 2003900874241491216
       Females 2013143023172398633265
       Males 2003580354193347180
       Males 2013109017114348525256
      The overall utilisation rate was calculated using the total number of procedures for females (or males) as the numerator and the number of females (or males) in the population as the denominator. Age-specific utilisation rates were calculated using the number of procedures for each age group as the numerator and the age-specific population as the denominator. Bilateral procedures performed within the same year were counted as two TKRs for calculating utilisation rates to avoid underestimating the true utilisation of TKR.
      Four TKR procedures from Finland were excluded from the 2013 analyses due to missing data on gender.

      Overall utilisation of UKR surgery

      Compared to the number of TKR procedures performed, utilisation of UKR in each country was relatively low. In 2003, the number of UKR procedures ranged from 426 (in Norway) to 4070 (in Australia). In 2013, the number of procedures ranged from 276 (in Finland) to 2056 (in Australia). As a proportion of all knee replacement procedures, UKR utilisation decreased substantially in all countries from 2003 to 2013; from 16.2% to 4.6% in Australia, from 15.6% to 5.5% in Denmark, from 7.0% to 2.8% in Finland, from 16.0% to 9.7% in Norway, and from 12.5% to 3.7% in Sweden.
      When population size was taken into account, a reduction in UKR utilisation rate over time was also evident for all countries. The greatest reductions were observed for Australia (from 20.6 UKR procedures per 100,000 population in 2003 to 8.9 procedures per 100,000 in 2013) and Sweden (from 10.6 to 4.9 UKR procedures per 100,000). Smaller reductions in utilisation rates were seen for Denmark (from 9.9 to 6.4 UKR procedures per 100,000), Finland (from 8.8 to 5.1 UKR procedures per 100,000) and Norway (from 9.4 to 8.5 UKR procedures per 100,000).

      Discussion

      This study is the first to use population-based arthroplasty registry data to estimate the lifetime risk of TKR at the national level, and to compare lifetime risk between countries and over time. We used data from five well-validated registries to obtain the most accurate information on TKR utilisation. We found a marked increase in the lifetime risk of primary TKR for OA in all countries over the ten-year study period, and substantial variation between countries in the utilisation of TKR. These lifetime risk estimates advance our understanding of population-level knee OA disease burden and healthcare utilisation, beyond data from the Global Burden of Disease Study that were modelled using systematic reviews of OA prevalence and incidence
      • Cross M.
      • Smith E.
      • Hoy D.
      • Nolte S.
      • Ackerman I.
      • Fransen M.
      • et al.
      The global burden of hip and knee osteoarthritis: estimates from the Global Burden of Disease 2010 study.
      , and beyond published TKR incidence or utilisation rates that do not consider life expectancy, age-specific mortality, or whether individuals have multiple surgical procedures
      • Kurtz S.M.
      • Ong K.L.
      • Lau E.
      • Widmer M.
      • Maravic M.
      • Gomez-Barrena E.
      • et al.
      International survey of primary and revision total knee replacement.
      • Pabinger C.
      • Lothaller H.
      • Geissler A.
      Utilization rates of knee-arthroplasty in OECD countries.
      .
      The observed international variation in lifetime risk is unlikely to be explained purely by differences in knee OA prevalence, given the overlap in prevalence data for the five included countries
      • National Public Health Institute
      Musculoskeletal Disorders and Diseases in Finland: Results of the Health 2000 Survey.
      • Grotle M.
      • Hagen K.
      • Natvig B.
      • Dahl F.
      • Kvien T.
      Prevalence and burden of osteoarthritis: results from a population survey in Norway.
      • Ackerman I.N.
      • Osborne R.H.
      Obesity and increased burden of hip and knee joint disease in Australia: results from a national survey.
      • van der Pas S.
      • Castell M.V.
      • Cooper C.
      • Denkinger M.
      • Dennison E.M.
      • Edwards M.H.
      • et al.
      European project on osteoarthritis: design of a six-cohort study on the personal and societal burden of osteoarthritis in an older European population.
      • Turkiewicz A.
      • Petersson I.F.
      • Bjork J.
      • Hawker G.
      • Dahlberg L.E.
      • Lohmander L.S.
      • et al.
      Current and future impact of osteoarthritis on health care: a population-based study with projections to year 2032.
      • Laxafoss E.
      • Jacobsen S.
      • Gosvig K.K.
      • Sonne-Holm S.
      Case definitions of knee osteoarthritis in 4,151 unselected subjects: relevance for epidemiological studies; the Copenhagen Osteoarthritis Study.
      . It is possible that differences in OA severity distributions may have contributed to our findings, although country-level severity data are not available to confirm this hypothesis. Variation in obesity rates between countries
      • Pabinger C.
      • Lothaller H.
      • Geissler A.
      Utilization rates of knee-arthroplasty in OECD countries.

      OECD.Stat. Non-medical determinants of health. Available from: https://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT. [Accessed 26 September 2016].

      and changes over time may also influence the knee OA burden in individual countries. According to national health survey data collated by the Organisation for Economic Co-operation and Development (OECD)

      OECD.Stat. Non-medical determinants of health. Available from: https://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT. [Accessed 26 September 2016].

      , the prevalence of self-reported obesity in Finland increased from 12.8% in 2003 to 15.7% in 2013, and from 9.8% in Sweden in 2003 to 11.7% in 2013. Only single year obesity prevalence estimates are available for Denmark and Norway during the period 2003–2013 (14.2% in Denmark in 2013, and 10.0% in Norway in 2012). For Australia, the national prevalence of obesity increased from 24.6% in 2007 to 28.3% in 2011 but these estimates are based on measured height and weight data rather than self-reported data (the latter tend to under-report obesity). As high-income countries, life expectancy in Australia and the Nordic countries was similar and unlikely to have contributed to the international variation in lifetime risk. Longer life expectancy for females is likely, however, to have contributed to the higher lifetime risk of TKR seen for females in all countries.
      The most plausible explanation for the between-country differences in lifetime risk of TKR is international variation in health system factors. These include (but are not limited to) differences in local indications for surgery, access to surgery, healthcare funding and health workforce issues. Earlier research has shown significant international variation in the pre-operative status of people undergoing joint replacement for OA
      • Ackerman I.
      • Dieppe P.
      • March L.
      • Roos E.
      • Nilsdotter A.
      • Brown G.
      • et al.
      Variation in age and physical status prior to total knee and hip replacement surgery: a comparison of centres in Australia and Europe.
      • Dieppe P.
      • Judge A.
      • Williams S.
      • Ikwueke I.
      • Guenther K.-P.
      • Floeren M.
      • et al.
      Variations in the pre-operative status of patients coming to primary hip replacement for osteoarthritis in European orthopaedic centres.
      , suggesting differing clinical thresholds for performing surgery. The higher lifetime risk in Australia could also relate to increased access to surgery within the private healthcare system. In contrast, orthopaedic surgeons in the Nordic countries might be more likely to consider non-surgical management, given the availability of region-based OA prevention and management programs that actively encourage people to trial physiotherapy, disease education and exercise prior to considering surgery. These include the ‘Better management of patients with OsteoArthritis’ (BOA) program in Sweden
      • The BOA-register
      Better Management of Patients with Osteoarthritis: Annual Report 2013.
      , the ‘Good Life with osteoArthritis in Denmark’ (GLA:D) program in Denmark
      • Skou S.T.
      • Simonsen M.E.
      • Roos E.M.
      Good Life with Arthritis in Denmark (GLA: D) – implementation of evidence-based care for knee and hip osteoarthritis in clinical practice.
      , and ‘AktivA’ in Norway

      AktivA. ActiveOA – Active living with osteoarthritis. Available from: http://aktivmedartrose.no/english. [Accessed 26 September 2016].

      . However, while conservative management programs might improve OA symptoms and delay the need for TKR surgery
      • Skou S.T.
      • Roos E.M.
      • Laursen M.B.
      • Rathleff M.S.
      • Arendt-Nielsen L.
      • Simonsen O.
      • et al.
      A randomized, controlled trial of total knee replacement.
      , whether they can ultimately reduce an individual's lifetime risk is not known. Personal factors could also play a role in promoting the uptake of TKR in individual countries and increasing lifetime risk; for example, greater acceptance of joint replacement surgery in the community, cultural factors, more exposure to successful outcomes among peers, and access to paid leave or injury compensation schemes. The high lifetime risk in Finland might relate to local patient preferences for surgery, with registry research suggesting that Finnish baby-boomers elect to undergo TKR when their OA symptoms are relatively mild
      • Leskinen J.
      • Eskelinen A.
      • Huhtala H.
      • Paavolainen P.
      • Remes V.
      The incidence of knee arthroplasty for primary osteoarthritis grows rapidly among baby boomers: a population-based study in Finland.
      . It is not clear why Norway had the lowest lifetime risk of TKR for both sexes in 2013 but this could relate to their relatively high utilisation of UKR and the comparatively good patient-reported outcomes for UKR in that country
      • Lygre S.H.L.
      • Espehaug B.
      • Havelin L.I.
      • Furnes O.
      • Vollset S.E.
      Pain and function in patients after primary unicompartmental and total knee arthroplasty.
      , although a recent systematic review reported higher revision rates than for TKR
      • Pabinger C.
      • Lumenta D.B.
      • Cupak D.
      • Berghold A.
      • Boehler N.
      • Labek G.
      Quality of outcome data in knee arthroplasty.
      .
      The significant increases in lifetime risk over time for each country are also unlikely to relate simply to growth in OA prevalence. In contrast, Global Burden of Disease data showed that worldwide, the age-standardised prevalence of knee OA did not change significantly from 1990 to 2010
      • Cross M.
      • Smith E.
      • Hoy D.
      • Nolte S.
      • Ackerman I.
      • Fransen M.
      • et al.
      The global burden of hip and knee osteoarthritis: estimates from the Global Burden of Disease 2010 study.
      while in Finland, national health surveys have shown that the prevalence of knee OA among women has actually decreased over a 20-year period
      • National Public Health Institute
      Musculoskeletal Disorders and Diseases in Finland: Results of the Health 2000 Survey.
      . Our data showed that Australia experienced the greatest absolute change in lifetime risk over time (for both sexes), and this probably reflects ‘catch up’ of previous unmet need following the introduction of government financial incentives in 1999–2000 to promote the uptake of private health insurance cover. Finland also experienced considerable unmet need for TKR prior to 2005, with patients experiencing long delays in accessing surgery. New Finnish legislation introduced in 2005 specified maximum waiting times for orthopaedic consultation and TKR, and hospitals received additional resources to meet these requirements. Although this cannot be quantified, these macro-level initiatives would undoubtedly have contributed to the rise in lifetime risk of TKR in both Australia and Finland. While detailed information on policy changes in each country was not available, it is possible that changes to government healthcare policies in the other countries over the study period may have contributed to the growth observed.
      Our calculated utilisation rates showed the greatest burden of TKR was borne by the 70–79 age group in 2003 and 2013. Younger patients (those aged 40–59 years) demonstrated only a small absolute increase in utilisation rates over the ten-year period, and perhaps this reflects awareness of the relatively high TKR revision rates for younger individuals
      • Julin J.
      • Jamsen E.
      • Puolakka T.
      • Konttinen Y.T.
      • Moilanen T.
      Younger age increases the risk of early prosthesis failure following primary total knee replacement for osteoarthritis. A follow-up study of 32,019 total knee replacements in the Finnish Arthroplasty Register.
      • W-Dahl A.
      • Robertsson O.
      • Lidgren L.
      Surgery for knee osteoarthritis in younger patients.
      . These utilisation rates cannot be directly compared to other studies examining TKR incidence or utilisation rates between countries, predominantly due to differences in data sources and methods. The study by Kurtz et al.
      • Kurtz S.M.
      • Ong K.L.
      • Lau E.
      • Widmer M.
      • Maravic M.
      • Gomez-Barrena E.
      • et al.
      International survey of primary and revision total knee replacement.
      used a combination of inpatient hospital administrative data and arthroplasty registry data. Their reported incidence rates were not stratified by age or sex and were calculated at different time points (2007–2010) than those used for our study. The authors also acknowledged hospital coding limitations, where it was not possible to consistently determine primary TKR from revision TKR, or TKR from UKR. Most recently, Pabinger et al.
      • Pabinger C.
      • Lothaller H.
      • Geissler A.
      Utilization rates of knee-arthroplasty in OECD countries.
      compared TKR utilisation rates in OECD countries but separate analyses for males and females were not reported.
      Our research design has uniquely generated burden of knee OA estimates using national data from five countries. Combined, the five countries had a population of almost 49 million people and performed over 74,000 primary TKRs in 2013. A major strength of this study is our use of robust arthroplasty registry data to ensure accurate estimations of lifetime risk and enable fair international comparisons. The Nordic countries have led the world with regard to implementing and maintaining high-quality national arthroplasty registries and the five included registries have near-complete TKR capture at the population level. We counted all TKR procedures when calculating utilisation rates but were careful to avoid erroneously inflating our lifetime risk estimates by only counting bilateral TKR procedures at the patient-level for these analyses. Given that provision of TKR is highly age-related, the standardised lifetime risk approach was important for dealing with changes to a country's age structure over time (for example, growth in older age groups due to population ageing). In this way, standardised lifetime risk calculations are likely to be more accurate for monitoring changes in lifetime risk and undertaking between-country comparisons than non-standardised methods. We also acknowledge the limitations of this research. We included all patients who received a primary TKR for OA in 2003 or 2013 (regardless of whether they had previously received a contralateral primary TKR), as from a clinical perspective these patients are still ‘at risk’ of having surgery in the years of interest. This method also accounts for the different establishment years for each registry, and reflects the challenges of estimating lifetime risk for conditions that can have multiple occurrences over time
      • Sasieni P.D.
      • Shelton J.
      • Ormiston-Smith N.
      • Thomson C.S.
      • Silcocks P.B.
      What is the lifetime risk of developing cancer?: the effect of adjusting for multiple primaries.
      or conditions that can affect more than one joint, in the case of knee OA. Annual lifetime risk was not calculated as annual life tables were not consistently available for all countries, and it is possible that fluctuations may have occurred over the ten-year study period. Finally, we acknowledge that there may be some variation in the coding of diagnoses and classification of knee replacement procedures between the national registries that cannot be accounted for in our analyses.
      In conclusion, this study has identified significant increases in the lifetime risk of primary TKR performed for OA in Australia, Denmark, Finland, Norway and Sweden over a ten-year period. There was substantial variation in lifetime risk across all countries, with females consistently demonstrating the highest risk. These data augment our understanding of the population burden of knee OA, and can be used by individual countries to inform public health policy and resource planning.

      Author contributions

      Study conception and design: INA, MAB, RDS, CAB and GG; data acquisition: all other authors; data analysis: MAB and INA; manuscript drafting: INA with input from all authors. All authors have approved the final version of the manuscript.

      Competing interests

      All authors have completed the ICMJE disclosure form and declare: no support from any organisation for the submitted work; AE has received payment for lectures from DePuy and Stryker and research and travel grants from DePuy outside the submitted work; no other relationships or activities that could appear to have influenced the submitted work.

      Role of the funding source

      A/Prof Ackerman was supported by a National Health and Medical Research Council of Australia Public Health (Australian) Early Career Fellowship (#520004). The Swedish Knee Arthroplasty Register receives support from the Swedish Medical Research Council (MFR 09509), Stiftelsen för bistånd åt vanföra i Skåne, the National Board of Health and Welfare, the Swedish Association of Local Authorities and Regions, and the Medical Faculty. These institutions had no role in the study design, collection, analysis and interpretation of data, in the writing of the manuscript, or in the decision to submit the manuscript for publication.

      Acknowledgments

      The authors wish to thank Professor Stefan Lohmander for facilitating this international collaboration and for his valuable advice regarding study design. We are also grateful to Professor Ewa Roos for her assistance in initiating our collaboration with the Danish registries.

      Appendix A. Supplementary data

      The following is the supplementary data related to this article:

      References

        • Cross M.
        • Smith E.
        • Hoy D.
        • Nolte S.
        • Ackerman I.
        • Fransen M.
        • et al.
        The global burden of hip and knee osteoarthritis: estimates from the Global Burden of Disease 2010 study.
        Ann Rheum Dis. 2014; 73: 1316-1322
        • Murray C.J.L.
        • Vos T.
        • Lozano R.
        • Naghavi M.
        • Flaxman A.D.
        • Michaud C.
        • et al.
        Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010.
        Lancet. 2012; 380: 2197-2223
        • Kurtz S.
        • Mowat F.
        • Ong K.
        • Chan N.
        • Lau E.
        • Halpern M.
        Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002.
        J Bone Joint Surg Am. 2005; 87: 1487-1497
        • Singh J.A.
        Epidemiology of knee and hip arthroplasty: a systematic review.
        Open Orthop J. 2011; 5: 80-85
        • Australian Orthopaedic Association National Joint Replacement Registry
        Annual Report – Hip and Knee Arthroplasty.
        Australian Orthopaedic Association, Adelaide2015
        • Robertsson O.
        • Bizjajeva S.
        • Fenstad A.M.
        • Furnes O.
        • Lidgren L.
        • Mehnert F.
        • et al.
        Knee arthroplasty in Denmark, Norway and Sweden: a pilot study from the Nordic Arthroplasty Register Association.
        Acta Orthop. 2010; 81: 82-89
        • Losina E.
        • Walensky R.P.
        • Kessler C.L.
        • Emrani P.S.
        • Reichmann W.M.
        • Wright E.A.
        • et al.
        Cost-effectiveness of total knee arthroplasty in the United States: patient risk and hospital volume.
        Arch Intern Med. 2009; 169: 1113-1121
        • Leening M.J.G.
        • Ferket B.S.
        • Steyerberg E.W.
        • Kavousi M.
        • Deckers J.W.
        • Nieboer D.
        • et al.
        Sex differences in lifetime risk and first manifestation of cardiovascular disease: prospective population based cohort study.
        BMJ. 2014; 349: g5992
        • Ahmad A.S.
        • Ormiston-Smith N.
        • Sasieni P.D.
        Trends in the lifetime risk of developing cancer in Great Britain: comparison of risk for those born from 1930 to 1960.
        Br J Cancer. 2015; 112: 943-947
        • Culliford D.J.
        • Maskell J.
        • Kiran A.
        • Judge A.
        • Javaid M.K.
        • Cooper C.
        • et al.
        The lifetime risk of total hip and knee arthroplasty: results from the UK General Practice Research Database.
        Osteoarthritis Cartilage. 2012; 20: 519-524
        • Weinstein A.M.
        • Rome B.N.
        • Reichmann W.M.
        • Collins J.E.
        • Burbine S.A.
        • Thornhill T.S.
        • et al.
        Estimating the burden of total knee replacement in the United States.
        J Bone Joint Surg. 2013; 95: 385-392
        • Bohensky M.
        • Ackerman I.
        • de Steiger R.
        • Gorelik A.
        • Brand C.
        Lifetime risk of total knee replacement and temporal trends in incidence by health care setting, socioeconomic status and geographic location.
        Arthritis Care Res. 2014; 66: 424-431
        • Kurtz S.M.
        • Ong K.L.
        • Lau E.
        • Widmer M.
        • Maravic M.
        • Gomez-Barrena E.
        • et al.
        International survey of primary and revision total knee replacement.
        Int Orthop. 2011; 35: 1783-1789
        • Ravi B.
        • Croxford R.
        • Reichmann W.M.
        • Losina E.
        • Katz J.N.
        • Hawker G.A.
        The changing demographics of total joint arthroplasty recipients in the United States and Ontario from 2001 to 2007.
        Best Pract Res Clin Rheum. 2012; 26: 637-647
        • Pabinger C.
        • Lothaller H.
        • Geissler A.
        Utilization rates of knee-arthroplasty in OECD countries.
        Osteoarthritis Cartilage. 2015; 23: 1664-1673
      1. OECD.Stat. Health status. Available from: https://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT. [Accessed 26 September 2016].

        • Sasieni P.D.
        • Adams J.
        Standardized lifetime risk.
        Am J Epidemiol. 1999; 149: 869-875
        • Srikanth V.K.
        • Fryer J.L.
        • Zhai G.
        • Winzenberg T.M.
        • Hosmer D.
        • Jones G.
        A meta-analysis of sex differences prevalence, incidence and severity of osteoarthritis.
        Osteoarthritis Cartilage. 2005; 13: 769-781
        • Pabinger C.
        • Geissler A.
        Utilization rates of hip arthroplasty in OECD countries.
        Osteoarthritis Cartilage. 2014; 22: 734-741
        • National Public Health Institute
        Musculoskeletal Disorders and Diseases in Finland: Results of the Health 2000 Survey.
        National Public Health Institute, Helsinki2007
        • Grotle M.
        • Hagen K.
        • Natvig B.
        • Dahl F.
        • Kvien T.
        Prevalence and burden of osteoarthritis: results from a population survey in Norway.
        J Rheumatol. 2008; 35: 677-684
        • Ackerman I.N.
        • Osborne R.H.
        Obesity and increased burden of hip and knee joint disease in Australia: results from a national survey.
        BMC Musculoskelet Disord. 2012; 13: 254
        • van der Pas S.
        • Castell M.V.
        • Cooper C.
        • Denkinger M.
        • Dennison E.M.
        • Edwards M.H.
        • et al.
        European project on osteoarthritis: design of a six-cohort study on the personal and societal burden of osteoarthritis in an older European population.
        BMC Musculoskelet Disord. 2013; 14: 138
        • Turkiewicz A.
        • Petersson I.F.
        • Bjork J.
        • Hawker G.
        • Dahlberg L.E.
        • Lohmander L.S.
        • et al.
        Current and future impact of osteoarthritis on health care: a population-based study with projections to year 2032.
        Osteoarthritis Cartilage. 2014; 22: 1826-1832
        • Laxafoss E.
        • Jacobsen S.
        • Gosvig K.K.
        • Sonne-Holm S.
        Case definitions of knee osteoarthritis in 4,151 unselected subjects: relevance for epidemiological studies; the Copenhagen Osteoarthritis Study.
        Skeletal Radiol. 2010; 39: 859-866
      2. OECD.Stat. Non-medical determinants of health. Available from: https://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT. [Accessed 26 September 2016].

        • Ackerman I.
        • Dieppe P.
        • March L.
        • Roos E.
        • Nilsdotter A.
        • Brown G.
        • et al.
        Variation in age and physical status prior to total knee and hip replacement surgery: a comparison of centres in Australia and Europe.
        Arthritis Care Res. 2009; 61: 166-173
        • Dieppe P.
        • Judge A.
        • Williams S.
        • Ikwueke I.
        • Guenther K.-P.
        • Floeren M.
        • et al.
        Variations in the pre-operative status of patients coming to primary hip replacement for osteoarthritis in European orthopaedic centres.
        BMC Musculoskelet Disord. 2009; 10: 19
        • The BOA-register
        Better Management of Patients with Osteoarthritis: Annual Report 2013.
        BOA, Sweden2013
        • Skou S.T.
        • Simonsen M.E.
        • Roos E.M.
        Good Life with Arthritis in Denmark (GLA: D) – implementation of evidence-based care for knee and hip osteoarthritis in clinical practice.
        Osteoarthritis Cartilage. 2014; 22: S457-S458
      3. AktivA. ActiveOA – Active living with osteoarthritis. Available from: http://aktivmedartrose.no/english. [Accessed 26 September 2016].

        • Skou S.T.
        • Roos E.M.
        • Laursen M.B.
        • Rathleff M.S.
        • Arendt-Nielsen L.
        • Simonsen O.
        • et al.
        A randomized, controlled trial of total knee replacement.
        N Engl J Med. 2015; 373: 1597-1606
        • Leskinen J.
        • Eskelinen A.
        • Huhtala H.
        • Paavolainen P.
        • Remes V.
        The incidence of knee arthroplasty for primary osteoarthritis grows rapidly among baby boomers: a population-based study in Finland.
        Arthritis Rheum. 2012; 64: 423-428
        • Lygre S.H.L.
        • Espehaug B.
        • Havelin L.I.
        • Furnes O.
        • Vollset S.E.
        Pain and function in patients after primary unicompartmental and total knee arthroplasty.
        J Bone Joint Surg. 2010; 92: 2890-2897
        • Pabinger C.
        • Lumenta D.B.
        • Cupak D.
        • Berghold A.
        • Boehler N.
        • Labek G.
        Quality of outcome data in knee arthroplasty.
        Acta Orthop. 2015; 86: 58-62
        • Julin J.
        • Jamsen E.
        • Puolakka T.
        • Konttinen Y.T.
        • Moilanen T.
        Younger age increases the risk of early prosthesis failure following primary total knee replacement for osteoarthritis. A follow-up study of 32,019 total knee replacements in the Finnish Arthroplasty Register.
        Acta Orthop. 2010; 81: 413-419
        • W-Dahl A.
        • Robertsson O.
        • Lidgren L.
        Surgery for knee osteoarthritis in younger patients.
        Acta Orthop. 2010; 81: 161-164
        • Sasieni P.D.
        • Shelton J.
        • Ormiston-Smith N.
        • Thomson C.S.
        • Silcocks P.B.
        What is the lifetime risk of developing cancer?: the effect of adjusting for multiple primaries.
        Br J Cancer. 2011; 105: 460-465