MaryAnn McBurnie

Psychosocial status predicts mortality in patients with life-threatening ventricular arrhythmias

BACKGROUND Quality-of-life (QoL) instruments evaluate various aspects of physical, mental, and emotional health, but how these psychosocial characteristics impact long-term outcome after cardiac arrest and ventricular tachycardia (VT) is unknown. OBJECTIVE The purpose of this study was to evaluate the relationship of baseline QoL scores with long-term survival of patients enrolled in the Antiarrhythmics Versus Implantable Defibrillators (AVID) trial. METHODS Formal QoL measures included SF-36 mental and physical components, Patient Concerns Checklist, and Ferrans and Powers Quality-of-Life Index-Cardiac Version. Multivariate Cox regression was used to assess the association of survival and these measures, adjusting for index arrhythmia type, gender, race, age, ejection fraction, history of congestive heart failure, antiarrhythmic therapy, and beta-blocker use. RESULTS During mean follow-up of 546 +/- 356 days, 129 deaths occurred among 740 patients. Higher baseline SF-36 physical summary scores (P <.001), higher baseline QoL Index summary scores (P = .015), and lower baseline Patient Concerns Checklist summary scores (P = .047) were associated with longer survival, even after adjustment for clinical variables. When QoL measures were examined simultaneously, only the SF-36 physical summary score remained significant (P = .002). CONCLUSION During recovery after sustained VT or cardiac arrest, formal baseline QoL assessment provides important prognostic information independent of traditional clinical data.

PS1-1a: Use of the CER Hub to Identify Out-of-Control Asthma and Compare Therapeutic Classes of Step-Up Asthma Medications in Clinical Practice

Background/Aims Asthma is a chronic inflammatory condition that imposes a substantial burden on patients and society. A major target in asthma care is guideline adherence to disease control assessment and therapy. Our aim was to develop an electronic medical record (EMR) based measure of provider determination of asthma control and use it to assess different treatment modalities employed in out-of-control asthma to allow observational comparative effectiveness research (CER) on different types of step-up therapy. Methods We developed EMR-based abstraction rules to allow automated determination of asthma control during clinical encounters, a construct that indicates need for treatment intensification. The EMR-based measure operationalizes components in the Expert Panel Report-3 recommendations for assessing a patient’s level of asthma control across the domains of risk and impairment. We used manual chart abstraction on samples of encounter records provided by six diverse health systems participating in the CER Hub project, to develop and validate the EMR-based measure of asthma control. Results We identified over 185,000 patients diagnosed with asthma across CER-Hub during 2006–2010. Provider documentation (predominantly text clinical notes) was rich in data related to asthma control including aspects of impairment (patient-reported symptom frequency, nighttime awakenings, interference with activity, frequency of rescue inhaler use, and lung function) and risk to patient well-being such as asthma exacerbations and use of systemic corticosteroids. Using the automated medical record classifier MediClass, which enables access to both coded and free-text components of the record, we will assess patients on low-dose inhaled corticosteroid therapy whose asthma is not well controlled. We are using the EMR-based measure to investigate the comparative effectiveness of the following step-up therapies (1) addition of a leukotriene modifier, (2) addition of a long-acting beta-agonist, and (3) increase to higher dose inhaled corticosteroids. Conclusions Traditional large database studies have been unable to assess elements of asthma control, such as symptom frequencies or activity limitations, because these clinical data are typically only available within free-text progress notes documenting the patient visit. The CER Hub asthma control measure provides new capacity to evaluate the comparative effectiveness of asthma interventions across diverse healthcare settings and in large real-world populations.

C-C4-04: Development of a Measure Set for Routine, Comprehensive, Automated Assessment of Obesity Care Quality

Background We have developed a technology platform for scalable and routine measurement of care quality using comprehensive electronic medical record (EMR) data, including providers’ free-text notes documenting clinical encounters, and are applying this technology to assess the care delivered to obese and overweight patients in two distinct health systems. NHLBI’s evidence-based clinical guidelines for overweight and obesity provide a clear set of patient care procedures for the primary care setting. Using these treatment guidelines, we have developed a set of measures for automated assessment of obesity care quality using EMRs. Methods Development started with an iterative process to identify key quality measures for obesity care. This process was guided by project aims to (1) target primary care, (2) ensure scalable application of the measure set to multiple health systems and EMR implementations, (3) assess feasibility of using natural language processing technology to allow inclusion of information recorded in the free-text notes, and (4) prioritize existing NHLBI efforts to define best clinical practices for obese and overweight patients. Our development process involved a multi-disciplinary team (including data specialists, medical records technicians, clinicians, and obesity treatment experts) reviewing, vetting, and reaching consensus on translating each clinical step in the NHLBI guideline to measurable clinical events documented in the EMR. Results A comprehensive set of process measures have been identified and are in the process of being operationalized for routine automated assessment of obesity care in two distinct health systems caring for diverse patient populations. These measures provide capacity to assess actual care practices for their adherence to recommendations that patients (a) be assessed both for weight and waist circumference as well as for readiness to lose weight, (b) be advised to lose weight if they are overweight or obese, (c) be assisted with goal-setting and plans for diet and exercise activities, and (d) receive follow-up from their primary care clinicians regarding these activities. Conclusions For health information technology to impact obesity care, EMR-based automated quality measures must be subjected to a repeatable and rigorous process of refinement, revision, and validation.