This retrospective study analyzed over 116 million NIS discharge records from 2000–2014 to generate updated projections of primary THA and TKA volume through 2030. It directly challenges the widely cited Kurtz et al. exponential growth model by showing that TKA growth has slowed significantly since 2008.
For a decade, health policy discussions cited Kurtz et al.'s projection of 3.48 million annual TKAs by 2030 to forecast arthroplasty costs and workforce needs. That model was built on pre-2008 exponential growth and did not account for the significant deceleration that began after 2008.
When you encounter policy documents or budget projections citing 3–4 million TKAs per year by 2030, recognize those figures are derived from outdated exponential models. Sloan et al. Provide the updated benchmark: 1.26 million (linear) to 1.68 million (Poisson) — roughly half the prior estimate.
This paper also teaches a broader critical appraisal lesson: the same statistical method applied to different time windows yields projections that differ by nearly twofold. Always ask which data window a projection model uses before accepting its conclusions.
Only 55% of TJA patients in this dataset were Medicare-covered. A fact worth knowing when someone argues MEDPAR data captures the full arthroplasty burden.
This retrospective study analyzed over 116 million NIS discharge records from 2000–2014 to generate updated projections of primary THA and TKA volume through 2030. It directly challenges the widely cited Kurtz et al. exponential growth model by showing that TKA growth has slowed significantly since 2008.
For a decade, health policy discussions cited Kurtz et al.'s projection of 3.48 million annual TKAs by 2030 to forecast arthroplasty costs and workforce needs. That model was built on pre-2008 exponential growth and did not account for the significant deceleration that began after 2008.
When you encounter policy documents or budget projections citing 3–4 million TKAs per year by 2030, recognize those figures are derived from outdated exponential models. Sloan et al. Provide the updated benchmark: 1.26 million (linear) to 1.68 million (Poisson) — roughly half the prior estimate.
This paper also teaches a broader critical appraisal lesson: the same statistical method applied to different time windows yields projections that differ by nearly twofold. Always ask which data window a projection model uses before accepting its conclusions.
Only 55% of TJA patients in this dataset were Medicare-covered. A fact worth knowing when someone argues MEDPAR data captures the full arthroplasty burden.