Fig 1.
Flow diagram of the participants eligible for the trajectory analyses.
K-L grade, Kellgren-Lawrence grade; KOOS QoL, Quality of Life subscale of the Knee injury and Osteoarthritis Outcome Score; MIC, Minimal Important Change; OAI, Osteoarthritis Initiative.
Table 1.
Baseline demographics and clinical characteristics of participants eligible for data analyses.
Fig 2.
Estimated HRQoL change (MIC) trajectory groups and proportion of individuals in the groups.
MIC, a minimally important reduction, i.e. a decrease of at least 10 points, in KOOS QoL. Solid lines represent proportion of individuals classified into the HRQoL trajectory groups and dashed lines represent the model estimate size of the HRQoL groups (lines should be closely corresponding). No change in KOOS QoL, 62.9% (green solid line), predicted 59.5% (lighter green dashed line); improving KOOS QoL after decline, 10.4% (blue solid line), predicted 11.9% (lighter blue dashed line); slowly worsening KOOS QoL, 17.1% (yellow solid line), predicted 19.1% (lighter orange dashed line); rapidly worsening KOOS QoL, 9.5% (red solid line), predicted 9.5% (lighter red dashed line).
Table 2.
Odds ratios (OR) and 95% confidence intervals (CI) of multinomial logistic regression analysis (trajectory of ‘no change in KOOS QoL’ as reference).
Fig 3.
Cox cumulative hazard function for knee replacement during extended follow-up by HRQoL trajectory groups.
The time to KR (months between the dates of baseline visit and the date of KR) in relation to membership in the trajectory groups. Participants without a recorded KR event were censored. The Cox regression was carried out in an unadjusted mode. HRQoL change (MIC, a minimally important reduction in KOOS QoL) trajectory groups: no change in KOOS QoL (green line); improving KOOS QoL after decline (blue line); slowly worsening KOOS QoL (yellow line); rapidly worsening KOOS QoL (red line).
Fig 4.
Frequent pain medication users (prevalence, %) by HRQoL trajectory groups.
Pooled answers regarding the use of strong pain medications, acetaminophen and nonsteroidal anti-inflammatory drugs. Answers were dichotomized (frequent pain medication user / nonuser of pain medication). HRQoL change (MIC, a minimally important reduction in KOOS QoL) trajectory groups: no change in KOOS QoL (green line); improving KOOS QoL after decline (blue line); slowly worsening KOOS QoL (yellow line); rapidly worsening KOOS QoL (red line).