Osteoarthritis disease progression model using six year follow‐up data from the osteoarthritis initiative

Abstract

The objective was to develop a quantitative model of disease progression of knee osteoarthritis over 6 years using the total WOMAC score from patients enrolled into the Osteoarthritis Initiative (OAI) study. The analysis was performed using data from the Osteoarthritis Initiative database. The time course of the total WOMAC score of patients enrolled into the progression cohort was characterized using non-linear mixed effect modeling in NONMEM. The effect of covariates on the status of the disease and the progression rate was investigated. The final model provided a good description of the experimental data using a linear progression model with a common baseline (19 units of the total WOMAC score). The WOMAC score decreased by 1.77 units/year in 89% of the population or increased by 1.74 units/year in 11% of the population. Multiple covariates were found to affect the baseline and the rate of progression, including BMI, sex, race, the use of pain medications, and the limitation in activity due to symptoms. A mathematical model to describe the disease progression of osteoarthritis in the studied population was developed. The model identified two sub-populations with increasing or decreasing total WOMAC score over time, and the effect of important covariates was quantified.

Discussing prognosis with patients with osteoarthritis: a cross-sectional survey in general practice

“…It is becoming accepted that prognosis is influenced by a complex array of biological, clinical and social factors [ 14 ], and it is, therefore, perhaps not surprising that our findings support the evidence that doctors find prognostication in general difficult [ 15 ]. Heterogeneity of OA pathology, fluctuation of pain and physical limitation due to OA over time, and a lack of consensus on measures of progression and endpoint definition (both structural and clinical), make OA prognostication particularly difficult [ 12 , 16 ]. It is possible that the difference between thinking prognostic discussions are important, and engaging in them, is the recognition that although important to patients, prognostication in OA is attempting to predict the unpredictable.…”

The Arthroplasty Candidacy Help Engine tool to select candidates for hip and knee replacement surgery: development and economic modelling

“…A total of 22 such studies were identified. [99][100][101][115][116][117][135][136][137][138][139][140][141][142][143][144][145][146][147][148][149][150] However, the reported results were ambiguous. Most studies focused only on changes in WOMAC subscores and only reported data over a 2-to 5-year follow-up period.…”

“…In practice, patients who are not deemed to warrant immediate surgery may have treatment later, after their condition has worsened. However, there are relatively limited data on how clinical tool scores change over time without surgery, [99][100][101][102] and modelling the referral pathway and outcomes for patients who undergo surgery at different times would have greatly complicated the analysis. Instead, by directly comparing immediate TJA with no arthroplasty over 10 years, we made the most of existing UK data sets and directly assessed the cost-effectiveness of arthroplasty.…”

A review of disease progression models of Parkinson's disease and applications in clinical trials

“…Longitudinal data from clinical trials and observational studies are used to build these models to better understand and predict disease trajectories, both in the absence and presence of pharmacologic treatment . In addition, disease progression models can be used to correlate clinical states with structural or chemical biomarkers that also change with disease and to facilitate the identification of risk factors, demographics, and other covariates that affect baseline disease status and the rate of disease progression …”

“…2 In addition, disease progression models can be used to correlate clinical states with structural or chemical biomarkers that also change with disease and to facilitate the identification of risk factors, demographics, and other covariates that affect baseline disease status and the rate of disease progression. [3][4][5] For chronic progressive diseases like PD, a more precise understanding of the changes in disease course as it relates to treatment effects and patient-level factors would help in the design and efficiency of clinical trials. In particular, trials designed to detect diseasemodifying effects could be made more informative through the application of clinical trial simulations, which require a quantitative understanding of disease progression.…”