3066 — How Do Common Chronic Coexisting Conditions Affect Quality of Care Measures for Hypertension?
Petersen LA (Division of Health Policy and Quality, Houston Center for Quality of Care and Utilization Studies) , Woodard LD
(Division of Health Policy and Quality, Houston Center for Quality of Care and Utilization Studies), Henderson L
(Division of Health Policy and Quality, Houston Center for Quality of Care and Utilization Studies), Urech TH
(Division of Health Policy and Quality, Houston Center for Quality of Care and Utilization Studies)
Objectives:
There is concern that pay-for-performance and performance measurement may penalize health care providers of patients with multiple co-existing chronic diseases, as these patients are the most challenging to manage and are at highest risk. What is the impact of such diseases on the measured quality of care for hypertension?
Methods:
We studied 237,202 patients who had a primary care visit in FY 2004 in one of 8 VAMCs. Using both clinical (e.g. blood pressure [BP] readings, medications) and ICD-9 codes, we identified patients with hypertension and their concordant comorbidities (diabetes, ischemic heart disease, dyslipidemia), and discordant comorbidities (osteoarthritis, depression, chronic obstructive pulmonary disease). We assessed overall quality by determining the proportion who either were at BP goal (<=140/90) or received appropriate care in a 6-month follow-up period (defined as good quality), and how this varied by the presence or absence of concordant and discordant comorbid conditions, controlling for severity of illness and age.
Results:
154,444 (65%) had hypertension. Of these, 13% had none of the comorbidities, 41% had only concordant comorbidities, 11% had only discordant comorbidities, and 35% had both. At follow-up, the proportion with good quality was 82% of those with no comorbidities, 88% of those with concordant-only comorbidities, 86% of those with discordant-only comorbidities, and 91% of patients with both (p<0.0001). In multivariate analyses, patients with discordant comorbidities were more likely than those with hypertension alone to receive good quality care (OR=1.24, 95%CI:1.17-1.31). The odds of good quality were higher for those with concordant comorbidities (OR=1.70, 95%CI:1.63-1.78), and highest for those with both (OR=2.04, 95%CI:1.94-2.14) as compared with hypertensive patients with no comorbid conditions. Furthermore, as the number of comorbidities increased, so did the odds of good quality.
Implications:
We found that hypertensive patients with the most complex medical conditions were more likely than those without such conditions to have higher quality of care for hypertension.
Impacts:
Performance measures must be improved in order to ensure that providers who care for complex patients are not unfairly penalized under pay-for-performance and other performance programs. Providers appear to be identifying those at highest risk and focusing their efforts upon hypertension control in these patients.
