Factors associated with therapeutic inertia in hypertension: validation of a predictive model

Objective To study factors associated with therapeutic inertia in treating hypertension and to develop a predictive model to estimate the probability of therapeutic inertia in a given medical consultation, based on variables related to the consultation, patient, physician, clinical characteristics, and level of care. Methods National, multicentre, observational, cross-sectional study in primary care and specialist (hospital) physicians who each completed a questionnaire on therapeutic inertia, provided professional data and collected clinical data on four patients. Therapeutic inertia was defined as a consultation in which treatment change was indicated (i.e., SBP >= 140 or DBP >= 90 mmHg in all patients; SBP >= 130 or DBP >= 80 in patients with diabetes or stroke), but did not occur. A predictive model was constructed and validated according to the factors associated with therapeutic inertia. Results Data were collected on 2595 patients and 13 792 visits. Therapeutic inertia occurred in 7546 (75%) of the 10 041 consultations in which treatment change was indicated. Factors associated with therapeutic inertia were primary care setting, male sex, older age, SPB and/or DBP values close to normal, treatment with more than one antihypertensive drug, treatment with an ARB II, and more than six visits/year. Physician characteristics did not weigh heavily in the association. The predictive model was valid internally and externally, with acceptable calibration, discrimination and reproducibility, and explained one-third of the variability in therapeutic inertia. Conclusion Although therapeutic inertia is frequent in the management of hypertension, the factors explaining it are not completely clear. Whereas some aspects of the consultations were associated with therapeutic inertia, physician characteristics were not a decisive factor. J Hypertens 28: 1770-1777 (c) 2010 Wolters Kluwer Health vertical bar Lippincott Williams & Wilkins.