This study will investigate the associations among lay beliefs about coronary heart disease (CHD), modifiable risk behaviors and quality of life (QoL) in a diverse sample of veterans recovering from myocardial infarction (MI). A recent study by the PI found that post-MI patients often do not incorporate personally relevant behavioral risk factors into their representations of CHD; such non-veridical beliefs predicted non-adherence to recommendations regarding diet, exercise, and smoking cessation. One goal of the proposed research is to extend the investigation of lay beliefs and post-MI health behaviors to a more diverse sample of veterans; this is important because cardiac risk factors (e.g., smoking, hypertension) are very common among veterans in general and African American veterans in particular. Another goal is to prospectively follow the natural evolution of patients' CHD beliefs for 6-months during post-MI recovery in order to contribute to our understanding of the optimal timing for post-MI interventions. The proposed study would comprise a critical step in the PI's ongoing development of an intervention that targets the modification of non-veridical beliefs in patients' CHD representations (i.e., idiosyncratic beliefs about CHD symptoms, etiology, timeline, control/cure, and consequences). The long-term goal of this program of research is to facilitate healthy lifestyle behaviors, improve QoL, and reduce morbidity/mortality among post-MI patients.
The project objectives are to: 1) describe and compare the content and trajectory of CHD representations for 6-months after hospital discharge in African American and Caucasian post-MI VHA patients and 2) model the implications of CHD representations for health outcomes, including health behaviors, QoL, and health services utilization for 6-months after discharge in African American and Caucasian post-MI VHA patients.
A prospective, correlational design will be used; 430 post-MI patients (50% African American) will be recruited from three clinical sites. IRB approval has been received at the Houston TX VAMC clinical site; IRB review is in process for clinical sites at the Durham NC and Cleveland OH sites. An IRB modification will be submitted to add the Durham and Cleveland sites after IRB approval has been received at those sites. 300 of these subjects are expected to complete each and every round of data collection. Data collection will be coordinated and performed from the central study site at the Iowa City IA VAMC. Telephone interviews and self-report questionnaires will be used to measure CHD representations, cognitive (cardiac self-efficacy, sense of control), affective (depression, anxiety, negative affect), and sociocultural variables (social support, ethnicity, SES), cardiac behavioral intentions, CHD health behaviors (exercise, diet, smoking, medication adherence), and CHD outcomes (QoL, health services utilization) at four time points during the 6-months following hospital discharge. Additional data regarding medical status, health services use and prescription refills (as a measure of adherence) will be obtained from participants' medical records.
Analysis was based on data from 317 Caucasians and 93 African Americans. Among the 410 patients recruited, 314 (77%) completed at least the first wave (one-week) of follow up data collection. Two-hundred fifty four patients (62%) completed all four waves (one-week, 4- or 8-week, 3-month, and 6-month) of data collection.
Aim 1. The principal variable for our analysis is the extent to which each subject's common sense model is veridical vs. maladaptive. We gauged this measure in two ways. First, we created an absolute maladaptive composite score. This was calculated as the sum of risk factors relevant to a particular subject that were not recognized as such in that subject's representation. However, since not all subjects have the same number of risk factors, we also created a relative maladaptive composite score defined as the number of maladaptive representations divided by the total number of personally relevant cardiac risk behaviors for that subject. That were a total of nine risk factors considered: Age and gender (Males aged 45 or older were considered at risk, as were females aged 55 or older), current smoking status, high blood pressure, diabetes, other comorbidities, cholesterol, BMI, personal cardiac history, and family cardiac history.
Results of the mixed-effects models showed no significant relationship between the absolute maladaptive composite score and Race (p=0.3307), no significant change over Time (p=0.7933), and no interaction between Race and Time (p=0.7687). Similarly, parameter estimates from the model based on relative maladaptive composite scores indicated no significant effect of Race or Time.
Aim 2. We used a linear mixed-effect model for Time 1 through Time 3, with the maladaptive representation score (MA) as an outcome and Time as a predictor. We used each of these linear regression models with each health outcome at Time 4 as the dependent variable. The absolute MA at Time 1 - Time 3 is associated with 8 of 13 health outcomes at Time 4; however, the relative MA score (which is tailored to the individual's risk profile) is significantly associated with only 2 of the 13 health outcomes.
This study will make a crucial contribution to the development of a larger intervention trial of a nursing intervention that targets patients' CHD representations to facilitate health behaviors, improve Quality of Life, and reduce post-MI morbidity and mortality.
External Links for this Project
- Howren MB, Suls J, Martin R. Depressive symptomatology, rather than neuroticism, predicts inflated physical symptom reports in community-residing women. Psychosomatic medicine. 2009 Nov 1; 71(9):951-7. [view]
- Wakefield BJ, Dellsperger K, Mehr D, Wakefield D, Boren S. Development of a Decision Support Intervention to Improve Symptom Reporting in Persons with Chronic Heart Failure. Poster session presented at: Midwest Nursing Research Society Annual Conference; 2013 Mar 8; Chicago, IL. [view]
- Sperber N, Sandelowski M, Voils CI. Process evaluation of a dyadic lifestyle change intervention to improve low-density lipoprotein cholesterol. Poster session presented at: VA HSR&D / QUERI National Meeting; 2012 Jul 18; National Harbor, MD. [view]
- Wakefield BJ, Alexander A, Dellsperger K, Erdelez S. Usability of a Web-Based Symptom Monitoring Tool for Heart Failure. Paper presented at: Midwest Nursing Research Society Annual Conference; 2013 Mar 9; Chicago, IL. [view]