Objectives: This study sought to make the Brief Illness Perception Questionnaire (BIPQ) to be more informative about illness representation among older adults with hypertension. The authors developed categories for coding the open-ended question regarding cause of illness in the BIPQ--a pervasive quantitative measure for illness representation. Methods: Using inductive thematic analysis, the authors described categories which emerged from analysing the open-ended question of the BIPQ applied to patients with hypertension. Then using deductive thematic analysis, we applied known categories in the literature to causes for hypertension elicited by the BIPQ. Our categories are: behavioural, natural, physical, psychosocial, supernatural, and other. Results: We established inter-coder reliability by applying the coding scheme to a sample of suburban central Illinois (n = 197) and urban Tucson, Arizona (n = 299) older adults with hypertension (initial kappa = 0.61, revised kappa after decision rules = 0.987). To demonstrate the utility of the coding scheme we found that both Illinois and Arizona nonwhite patients differ from white patients in how they understand the cause of their hypertension. Conclusion: A more complete illness representation can now be quantified and statistically analysed through the BIPQ using the categories we revised for cause of hypertension. Assessment of causal beliefs through the BIPQ and the authors' coding scheme may enable tailored and efficient patient education, resulting in more effective provider-patient relationships.