Rheumatic conditions and their sequelae, including rheumatoid arthritis (RA), psoriatic arthritis, and knee osteoarthritis (OA), constitute a tremendous disease burden in the US and worldwide. Unlike the results of studies on risk factors for incident rheumatic conditions, findings on risk factors for disease sequelae among individuals with rheumatic conditions have been contradictory or paradoxical. Although biological explanations for these unexpected results may exist, an enticing alternative methodological explanation is a type of selection bias called collider-stratification bias (hereaftr referred to as collider bias), which occurs when one evaluates the effect of a risk factor on disease sequelae among those in an intermediate stage of disease (e.g., studying smoking effects in people with RA on the risk of a CV event). Little research has methodologically investigated the paradoxical phenomena of risk factors for sequelae events in the context of rheumatic conditions, leaving a crucial gap in knowledge on this important topic. Unless an appropriate design or analytic method is used to ascertain the true impact of suspected risk factors in these major rheumatic conditions, much research funds, time, and effort may be depleted without providing useful evidence for clinical recommendations. To overcome the methodological challenges associated with assessing the risk conferred by purported risk factors for sequela events among people with rheumatic diseases, we will investigate paradoxical findings in 3 key rheumatic disease contexts (smoking and risk of CV events in RA;smoking and risk of psoriatic arthritis in psoriasis;and obesity and risk of disease progression i knee OA). We will use three large databases: a prospective cohort (the Norfolk Arthritis Register [NOAR]), a general population database (The Health Improvement Network [THIN]), and a prospective OA cohort (the Multicenter Osteoarthritis Study [MOST]).
The specific aims of proposed research are as follows: (1) to determine and quantify collider bias in the paradoxical phenomena of 3 key rheumatic conditions of interest (smoking and risk of CV events in RA;smoking and risk of psoriatic arthritis in people with psoriasis;and obesity and risk of disease progression in people with knee OA);(2) to determine the true impact of smoking status and BMI on the disease sequelae by evaluating the change in exposure status measured after the diagnosis of the 3 key rheumatic conditions. The proposed research will help the candidate develop expertise in rheumatic disease research and gain hands-on research experience in advanced study design and analytic methods. Further, the training component of this K01 award will help advance the candidate's career development. The proposed training experiences have five aims: (1) to develop advanced knowledge of rheumatic diseases through coursework and national conference attendance;(2) to obtain relevant didactic training in rheumatology and clinical epidemiology through Rheumatology Grand Rounds and journal clubs;(3) to gain knowledge and experience in advanced analytic techniques through coursework in methods;(4) to build the credentials to become an independent investigator including first- authored publications and presentations;(5) to submit an NIH R01 proposal by the end of the proposed training period to pursue related research. The experience and expertise of the mentoring team will help ensure that the goals and objectives of this research and training will be achieved within the proposed timeframe. Further, the organizational structure of the research environment provides a nurturing balance of independent research with intellectual and resource support and infrastructure. By leveraging the expertise of the research mentors, the available databases, the novel design, and the analytic methods of this study, the expected results will fill crucial gaps i our understanding of the true impact of modifiable risk factors on key outcomes of interest, and provide relevant evidence-based clinical recommendations. The rigorous training in the methods and content areas of this proposed research project will help lay the foundation for future studies of risk factors and sequelae of other rheumatic diseases.
The burden of arthritis involves not only the morbidity from arthritis, but also its associated sequelae and premature mortality. Our ability to prevent the debilitating sequelae and costly outcomes of rheumatic conditions depends on relevant knowledge of the risks of these events, and particularly on an accurate understanding of modifiable risk factors for the development of these sequelae. Our aims would help move the field forward in several major ways: 1) evidence obtained from this project regarding the true impact of these key factors will directly inform practice recommendations to prevent and manage these major outcomes;2) the approach and results from the proposed research could help shift the paradigms in the field of rheumatic disease outcomes research;3) the application of methods from this research could also be relevant to research on many non-rheumatic diseases including studies on risk factors for recurrent cardiovascular outcomes among people with heart attacks.
|Choi, Hyon K; Nguyen, Uyen-Sa; Niu, Jingbo et al. (2014) Selection bias in rheumatic disease research. Nat Rev Rheumatol 10:403-12|
|Nguyen, U-S D T; Felson, D T; Niu, J et al. (2014) The impact of knee instability with and without buckling on balance confidence, fear of falling and physical function: the Multicenter Osteoarthritis Study. Osteoarthritis Cartilage 22:527-34|
|Nguyen, Uyen-Sa D T; Niu, Jingbo; Choi, Hyon K et al. (2014) Commentary: effect of obesity on mortality: comment on article by Banack and Kaufman. Epidemiology 25:2-3|