Purpose <p>Total hip arthroplasty (THA) and total knee arthroplasty (TKA) as treatments for osteoarthritis increasingly affect working-age adults. However, little is known about the development of work ability after THA and TKA. This study investigated different trajectory classes of work ability after THA and TKA.</p> Methods <p>This prospective study included 146 working patients (THA n = 51, TKA n = 95). They received THA/ TKA treatment as usual, including surgery, rehabilitation, and optional post-rehabilitation programmes. The primary outcome work ability was assessed five times (before surgery, end of rehabilitation, 3-month, 6-month, and 12-month follow-up) with the Work Ability Score. Additionally, the questionnaires included sociodemographic, health-, rehabilitation-, and work-related data. We applied a latent class growth analysis to identify distinct trajectory classes of work ability over 12&#xa0;months, and examined associated factors using univariable multinominal logistic regression analyses in MPlus.</p> Results <p>We found three distinct trajectory classes of work ability within one year after THA/ TKA: <i>no improvement</i> (24.5%), <i>delayed improvement</i> (36.6%), and <i>early improvement in work ability</i> (38.9%). Preoperative predictors for work ability trajectory classes were sex, pain intensity, functional ability in everyday life, and risk of work-related disabilities. Furthermore, work ability trajectory classes were associated with participation in the T-RENA post-rehabilitation programme, and with pain intensity one year after having started rehabilitation.</p> Conclusion <p>Despite the fact that most patients showed improvements in work ability after THA and TKA, there is a noteworthy proportion who did not. Further studies are needed to enhance identification of these individuals, and to offer targeted interventions.</p> Trial registration <p>The study is registered in the German Clinical Trials Register (DRKS00029630).</p>