This study was conducted to gain a quantitative understanding of how the importance of patient concerns change during the first six weeks following primary TKA. It showed that the importance of some concerns change over time while others do not. In the discussion that follows we suggest these changes reflect the clinical aspects of the delivery of care and the early recovery process. We also propose that the importance ratings provide direct evidence that supports the differentiation of the Activity and Participation components of the ICF in the clinical context of this study. To the best of our knowledge, this is the first study of this nature.
The Friedman's ANOVA revealed the importance rankings of all four ICF components changed over time. Post hoc testing showed that concerns in the ICF Body Function component were ranked higher at post-operative week two compared to before surgery. The WHO has defined Body Function  as physiological functions of the body systems including psychological functions. Following major joint surgery such as TKA, physiological functions perceived as important by patients, such as motion of the knee joint, strength in the lower extremities and sleeping at night are commonly impaired for the first month post-operatively . The sensation of pain and swelling in the surgical leg are also common occurrences during this phase of post-operative recovery. Presence of these impairments after surgery would naturally cause discomfort and limit normal daily activities. Consistent with this logic, patients in this study rated these Body Function concerns as being more important to them in the first month following surgery.
Going through surgery and recovery may also explain the pattern of change in the Activity and Participation importance ratings. Post hoc testing showed that patients ranked concerns in these components as being less important two weeks after surgery. Due to the nature of the surgery it is common for patients to experience impairments with certain body functions (increased pain and swelling, decreased knee range of motion, impaired sleep), which generally lead to limitations in the activities that they can perform. Researchers have reported a decline in physical function in the first month following TKA surgery . Consistent with this, patients in our study demonstrated significant decreases in the KOOS ADL and Sport/Rec subscales in the first two weeks after surgery. Due to this decline, concerns from the ICF Activity component (e.g. 'walking on uneven ground', 'cooking your own meals', 'doing your own housework') and concerns from the Participation component (e.g. 'driving a vehicle', 'shopping' or 'returning to hobbies') dropped in importance during the first two weeks post-operatively. Their subsequent increase in importance to pre-operative levels after surgery, may reflect the fact that patients are generally starting to feel better, are mobilizing with greater ease and are becoming less dependent on their caregivers. Indeed, Kennedy and colleagues  and Stratford and Kennedy  found that physical function improved to pre-operative levels by post-operative week six to eight. Again, consistent with this, patients in our study showed significant increases in the KOOS ADL and Sport/Rec subscales from two weeks through to six weeks post-operatively. The importance ratings for Body Function, Activity and Participation, combined with parallel changes in these KOOS subscale scores therefore, supports the validity of these concerns and their importance to patients because they mirror the surgical and early rehabilitation focus.
Just like the Body Function component, patient concerns that mapped to the Environmental Factors component were ranked higher two weeks after surgery. Due to the immediate decline in physical function , patients require the support of family members and health care workers to help them manage with daily activities and restore their physical function. Consistent with this, concerns such as 'having the support of your family members', 'receiving competent care from health care workers' and 'being less of a burden on your spouse or caregiver' were important to patients during this acute post-operative phase of their recovery. 'Having the support of your neighbours' was the only concern mapping to the Environmental Factors component that was of moderate to small importance at each post-operative evaluation session. This may relate to the fact that TKA is an elective surgery, which allows the patient time to plan for their post-operative needs using alternative resources (home care, family members etc.) and therefore, not requiring the support of their neighbours.
Two concerns that were not covered by the ICF components, 'receiving appropriate information regarding what to expect with rehabilitation following your surgery' and 'being independent' were of great importance to patients before surgery and remained at the same level of importance throughout the first six weeks of recovery. The first emphasizes the importance of education from the patient perspective. In this regard, Soever and MacKay  have documented that receiving information about their rehabilitation is important to patients and that this type of education improves patient satisfaction following total joint replacement surgery. The second highlights the consistently high value that patients place on independence. Hinojosa and Youngstrom have reported that "independence is defined by the individual's culture and values, support systems, and ability to direct his or her life" . Gignac and Cott  have reported that a loss of independence may have consequences on the quality of life and psychosocial well being of an individual. The cultural background of patients in this study and the support system in our society along with the consequences of losing independence, may explain why patients consistently rated the importance of this concern so high, throughout the first six weeks of recovery.
There is considerable debate regarding the need to distinguish between the Activity component and the Participation component of the ICF [12–14, 27]. While the WHO decided not to distinguish between these two components; others have stressed the importance of their differentiation if the ICF is to be widely used when describing health and health-related states of various populations across the world [12–14]. The results of this study highlight the challenge one faces in differentiating the ICF Activity and Participation components. On one hand, the average ranking of importance across time for these two components (Figure 2) reveals a similar pattern over time, suggesting there is little to be gained from separation of the ICF components when describing the importance of patient concerns about recovery. On the other hand, the relatively low values in the KOOS Sport/Rec subscale (Figure 1) may explain the pattern in Participation concerns after the two-week mark following surgery. Only those concerns in the ICF Participation component showed consistent increases in their level of importance through the first six weeks after surgery. We believe this highlights one tangible benefit from separating the two components in the context of this work: patients are starting to think about their return to participation in roles early in the recovery phase. As evidenced by the KOOS Sport/Rec subscale, they are also aware of their limitations in Participation-related activities. This is important for the clinician to note as researchers have shown that patient expectations are a predictor of functional outcome and satisfaction following total joint replacement surgery . Therefore, patient education should encompass return to participation roles very soon after surgery and in particular, should re-evaluate participation-related goals from two to six weeks of postoperative care. Future studies should determine if the increasing importance assigned to Participation concerns in this study continues to increase beyond the six-week mark post-operatively.
All participants in the study were recruited from a single tertiary care hospital. This may be viewed as decreasing the generalizability of this study to other settings. Even though the results may not be applicable to all TKA populations, the patient demographics and pre-operative functional status findings were comparable to that of other TKA studies [17, 29–31]. The rehabilitation setting (e.g. home care, in-patient or out-patient facilities) following surgery may influence what is important to patients. The majority of patients in this study received their initial therapy at home after acute care discharge until approximately three weeks following surgery. Then they continued therapy at an out-patient clinic of their choice. Therefore, it is possible that patients undergoing TKA, who receive therapy in a different setting than described above may have different priorities and concerns during the first six weeks following surgery. We note however that no subjects in this study provided additional concerns when given the opportunity to do so at four different time points of data collection. Furthermore, when the importance findings are viewed in combination with the KOOS data, changes in patient concerns mirrored the early recovery pattern from TKA surgery. This supports the validity of the concerns we investigated and combined with the finding that importance levels varied over time, suggests a temporal element that should be included in future work of this nature. As there was no test-retest component in this study, reliability of the importance ratings at each time point of data collection could not be confirmed. Another limitation of this study was the inability to include non-English speaking individuals. Therefore, the results are not applicable to non-English speaking recipients of TKA surgery. Finally, our sample size for this study was based on sample size calculations of a concurrent study of responsiveness of the WOMAC [unpublished]. This seemed acceptable because we had no previous importance ratings on which to base our sample size calculations. Where no significant differences were found, it may be possible the study was underpowered.