[Background] Osteoarthritis (OA) is a common chronic disease in the elderly. It is a degenerating change of knee. When knee joints got a pathological change, they will suffer pain and the daily activities will be limited. It will lead to reduce their social activity and present low self-esteem, social isolation and dependence. The best choice of treatment for severe OA is receiving total knee replacement (TKR). They may get more difficult to accept the necessary of surgery treatment. The patients’ acceptance of disability will influence their willingness of surgery. The issue is worth exploring. Purpose：The study was to explore the acceptance of disability among patients with severe OA of knee and the related factors. [Methods] This was a longitudinal and investigative study design. In outpatient department interviewing 210 patient before surgery, and then for 108 patient undergoing surgery, at postoperative 5-6 days, postoperative two months and postoperative six months interview further. Using structured questionnaire to collect data. In addition, after outpatient department three to four months, the 102 patient did’t undergoing surgery, conducted a telephone interview, explore its causes. The tools include Linkowski’s Acceptance of Disability Scale (score range from 50 to 300), Chinese’s version, Instrumental Activities of Daily Living Scale (IADL), Goniometer to measure the degree of knee flexion, Sit-to-stand cycles in sixty seconds (STS-60) and VAS pain score (0-10). [Result] Participants were more females (81.4%), mean age was 70.04 years, BMI mean was 27.13 kg/m2, there were more low level of education (71%), more spouses (65.2%), more living with family (91.9%), more no woker (75.7%). In demography, women than men had significant poor operative kne flexion, non-operative knee flexion, STS-60 and IADL. Age was negatively correlation with BMI, operative knee flexion, non-operative knee flexion, STS-60 and IADL. The older was lower levels of education, without a spouse or no work. BMI was negatively correlation with non-operative kne flexion. Low levels of education than hign levels of education had significant poor non-operative kne flexion and IADL. No worker had significant poor STS-60, IADL and acceptance of disability. In preoeration, operative knee flexion was negatively correlation with pain score. Non-operative knee flexion was negatively correlation with duration of symptoms, postively correlation with operative knee flexion. STS-60 was negatively correlation with pain score, postively correlation with operative knee flexion and non-operative knee flexion. IADL was negatively correlation with pain score, postively correlation with operative knee flexion, non-operative knee flexion and STS-60. Acceptance of disability was negatively correlation with pain score, postively correlation with operative knee flexion, non-operative knee flexion, STS-60 and IADL. In the surgery-related data, the postoperative ambulation first time was postively correlation with pain score, negatively correlation with operative knee flexion, STS-60, IADL and acceptance of disability. In postoperation, operative knee flexion was negatively correlation with pain score. STS-60 was negatively correlation with pain score, postively correlation with operative knee flexion. IADL was negatively correlation with pain score, except post operative 5-6 days, postively correlation with operative knee flexion and STS-60. Acceptance of disability was negatively correlation with pain score, postively correlation with operative knee flexion, STS-60 and IADL. Between operative knee flexion and non-operative knee flexion distance was negatively correlation with acceptance of disability. For preoperation and postoperation change in pain score, preoperative and postoperative 5-6 days were no significant differences, at postoperative two months began more and more drups, until postoperative six months the pain had relief. In operative knee flexion and STS-60 change, preoperative than postoperative 5-6 days was significantly poor, but at postoperative two months than preoperative was significantly good, even postoperative six months better than preoperative. In IADL change, preoperative than postoperative 5-6 days and postoperative two months were significantly poor, until postoperative six months better than preoperative.In acceptance of disability change, postoperation each stage was significant to rise gradually. According Mixed model to preoperative, postoperative 5-6 days, postoperative two months and postoperative six months four point overall data analysis. Higher pain score lower acceptance of disability. Higher operative knee flexion, STS-60 and IADL higher acceptance of disability. Using the statistical method of Multiple Logistic regression to predicted undergoing surgery probability, the non-operative knee flexion increase wound increase 6.6%. The acceptance of disability score increase wound decrease 2.7%. The duration of symptoms increase wound increase 0.7%. Using the statistical method of Multile linear regression to acceptance of disability explanatory R2=0.341, adjusted R2=0.321. [Conclusion] Resluts revealed this study can help nurses understand osteoarthritis of knee and influence postoperative status, to advocacy reference in clinical health education. In addition, according to pre- and post-operative period, patient may face many problems, given the individual education and psychological suppor. Let nurses play the uniqueness of care, but also let patient and family prepare preoperation and postoperative care konwledge better.