Association chain graphs: modelling etiological pathways

Mechanisms of action of therapeutic exercise for knee and hip OA remain a black box phenomenon: an individual patient data mediation study with the OA Trial Bank

OBJECTIVES

To evaluate mediating factors for the effect of therapeutic exercise on pain and physical function in people with knee/hip osteoarthritis (OA).

METHODS

For Subgrouping and TargetEd Exercise pRogrammes for knee and hip OsteoArthritis (STEER OA), individual participant data (IPD) were sought from all published randomised controlled trials (RCTs) comparing therapeutic exercise to non-exercise controls in people with knee/hip OA. Using the Counterfactual framework, the effect of the exercise intervention and the percentage mediated through each potential mediator (muscle strength, proprioception and range of motion (ROM)) for knee OA and muscle strength for hip OA were determined.

RESULTS

Data from 12 of 31 RCTs of STEER OA (1407 participants) were available. Within the IPD data sets, there were generally statistically significant effects from therapeutic exercise for pain and physical function in comparison to non-exercise controls. Of all potential mediators, only the change in knee extension strength was statistically and significantly associated with the change in pain in knee OA (β -0.03 (95% CI -0.05 to -0.01), 2.3% mediated) and with physical function in knee OA (β -0.02 (95% CI -0.04 to -0.00), 2.0% mediated) and hip OA (β -0.03 (95% CI -0.07 to -0.00), no mediation).

CONCLUSIONS

This first IPD mediation analysis of this scale revealed that in people with knee OA, knee extension strength only mediated ±2% of the effect of therapeutic exercise on pain and physical function. ROM and proprioception did not mediate changes in outcomes, nor did knee extension strength in people with hip OA. As 98% of the effectiveness of therapeutic exercise compared with non-exercise controls remains unexplained, more needs to be done to understand the underlying mechanisms of actions.

Transitions to adulthood and psychological distress in young adults born 12 years apart: constraints on and resources for development

BACKGROUND

Later transitions to adult roles and responsibilities have been linked with better psychological well-being yet psychological distress has risen despite young people making the transition to adulthood at older ages over recent years.

METHOD

We examine the role of structural constraints and adolescent resources in the relationship between the timing of transitions and psychological distress in early adult life in the 1958 National Child Development Study and the 1970 British Cohort Study. Graphical chain models were used to examine the influences on timing of four key transitions and their relationship with psychological distress (Malaise Inventory). The role of structural factors at birth (gender, social class) and adolescent resources (psychosocial problems, exam grades) were modelled.

RESULTS

An earlier transition to adult roles was associated with an increased risk for psychological distress but so was failing to make some key transitions. Structural constraints had negative effects on successful development. Persistent social class and gender inequalities in psychological distress were evident in both cohorts. Social class constraints were mediated by educational resources whereas gender constraints were mediated by psychosocial resources. The influence of structural constraints on the timing of transitions to adult roles was more complex with evidence of positive and negative mediation and moderation effects.

CONCLUSIONS

Delaying transition to adulthood promotes psychological health but failure of transition to independent living is associated with psychological distress. Life-course transitions are constrained by social origin and gender and possibly economic environment. Adolescent resources help young adults to make timely transitions to adult roles.

Visualizing Multivariate Dependencies with Association Chain Graphs

In a recent paper, a new type of graph to visualize the results from graphical models was proposed. Association chain graphs (ACGs) provide a richer visualization than conventional graphs (directed acyclic and recursive regression graphs) if the data can be described with only a small number of parameters. ACGs display not only which associations reach statistical significance, but also the magnitude of associations (confidence intervals for statistical main effects) as the contrast color to the background color of the graph. In this paper, the ACG visualization is extended especially for the case where all variables are binary by illustrating their relative frequencies. This shows the degrees of associations not only on the individual (as expressed by odds ratios or other indexes of association) but also on the community level. We applied the approach to an extensive example of birth and childhood factors for the onset of affective mental disorders using data from the EDSP (Early Developmental Stages of Psychopathology) Study.