Can self-reported disability assessment behaviour of insurance physicians be explained? Applying the ASE model

Abstract

Background: Very little is known about the attitudes and views that might underlie and explain the variation in occupational disability assessment behaviour between insurance physicians. In an earlier study we presented an adjusted ASE model (Attitude, Social norm, Self-efficacy) to identify the determinants of the disability assessment behaviour among insurance physicians. The research question of this study is how Attitude, Social norm, Self-efficacy and Intention shape the behaviour that insurance physicians themselves report with regard to the process (Behaviour: process) and content of the assessment (Behaviour: assessment) while taking account of Knowledge and Barriers.

Methods: This study was based on 231 questionnaires filled in by insurance physicians, resulting into 48 scales and dimension scores. The number of variables was reduced by a separate estimation of each of the theoretical ASE constructs as a latent variable in a measurement model. The saved factor scores of these latent variables were treated as observed variables when we estimated a path model with Lisrel to confirm the ASE model. We estimated latent ASE constructs for most of the assigned scales and dimensions. All could be described and interpreted. We used these constructs to build a path model that showed a good fit.

Results: Contrary to our initial expectations, we did not find direct effects for Attitude on Intention and for Intention on self reported assessment behaviour in the model. This may well have been due to the operationalization of the concept of 'Intention'. We did, however, find that Attitude had a positive direct effect on Behaviour: process and Behaviour: Assessment and that Intention had a negative direct effect on Behaviour: process.

Conclusion: A path model pointed to the existence of relationships between Attitude on the one hand and self-reported behaviour by insurance physicians with regard to process and content of occupational disability assessments on the other hand. In addition, Intention was only related to the self reported behaviour with regard to the process of occupational disability assessments. These findings provide some evidence of the relevance of the ASE model in this setting. Further research is needed to determine whether the ASE variables measured for insurance physicians are related to the real practice outcomes of occupational disability assessments.

Figure 1

Research model. From Steenbeek et al. [27]. S = Scale; D = Dimension, the number refers to the number of constructed scales and dimensions (measures).

Figure 2

Final Lisrel model: direct effects of endogenous variables on endogenous variables, disturbance terms and correlations between disturbance terms. For all coefficients p ≤ 0.05; except for the marked coefficient (*) p = 0.10. Italic coefficients are disturbance terms; underscored coefficients are correlations between disturbance terms; the other coefficients (not italic and not underscored) are direct effects. All coefficients are standardized. Relations are positive unless marked with a minus sign (-). df = degrees of freedom; p = probability; RSMEA = Root Mean Square Error of Approximation; CFI = Comparative Fit Index; SRMR = Standardized Root Mean Square Residual