Patient-profiled treatment responses in a large hypertension trial: a posthoc analysis of the INSIGHT study.

OBJECTIVE

Guidelines on the treatment of hypertensive patients usually refer to 'average' patients. However, in clinical practice, individual patient characteristics may differ substantially from the average. Thus, it seems worthwhile to examine the relationship between comprehensive patient profiles and blood pressure responses.

METHODS

We divided the patient population from the INSIGHT trial into exploration and validation cohorts and constructed composite patient profiles based on predictors of blood pressure control (age, severity of hypertension, comorbidities, and previous treatment status). Next, we tested in the exploration cohort whether blood pressure control rates and adverse effects after 6 months of therapy across these profiles differed from those in the entire patient group. Finally, we explored whether the results from the exploration cohort could be validated using another cohort.

RESULTS

Logistic regression analysis showed that the odds of achieving blood pressure control differed substantially between patient profiles but not between treatment modalities. Patients with a less favorable profile (e.g. the combination of age above 60 years, baseline systolic pressure above 160 mmHg, and the presence of diabetes) did less well than patients with a low-risk profile (e.g. absence of organ damage). These results were confirmed in the validation cohort.

CONCLUSION

We conclude that responses to antihypertensive treatment vary in a clinically important manner depending on the composite patient profiles. When found in other trials as well, a priori knowledge about response rates of various patient-profile treatment regimens may help choose the best treatment in individual patients and improve overall blood pressure control rates.