Statistics from Altmetric.com
Implications for practice and research
Increasingly, there is recognition that women with heart disease have greater psychological morbidity than men.
Owing to increased rates of depression and other factors, such as older age, women are less likely to adhere to behaviour change strategies and attend cardiac rehabilitation (CR).
A tailored and targeted approach, using a gender-based framework and a theoretically derived behaviour change strategy, can improve depressive symptoms in women following an acute cardiac event.
The worth of tailoring and targeting strategies to women's readiness to change behaviours as well as the likely benefit of providing a gender-based approach to care delivery.
The need to replicate this study in other study settings and in women from cultural and linguistically diverse groups.
The importance of measuring process measures and ensuring monitoring of intervention fidelity in research.
In developed countries, coronary heart disease (CHD) contributes to death and disability among women.1 Gender is a dynamic and socially defined construct comprising role enactment, values and beliefs of men and women. Gender-based roles and health outcomes are shaped by social, economic, political and cultural factors, rather than by biology alone.2 These are important considerations in ensuring health equity and access to evidence-based interventions.
Depression is a predictor of adverse health outcomes in CHD, and women have higher rates than do men.3 The low participation rates and poor adherence of women in CR have been well recognised.4 Adverse health outcomes for women are compounded by these higher rates of depression. People who are depressed are less likely to attend and complete CR, and engage in important behaviour change strategies, such as exercise. Therefore many women potentially do not achieve the benefits of CR in morbidity and mortality reduction, particularly in the context of a dose-dependent effect.5 These factors contribute to health disparities and adverse health outcomes. This compels researchers and clinicians to explore tailored and targeted strategies.
This study used a two-group, single-blind, randomised clinical trial methodology to compare physiological and psychosocial outcomes in women recruited from an outpatient CR programme in the USA following an acute cardiac event (medical and surgical admissions). Women participated in either a traditional (usual care) 12-week CR programme (n=111) or a tailored programme guided by the transtheoretical model (n=141) of behaviour change augmented by motivational interviewing techniques and a dedicated exercise programme for women. Usual care and intervention groups adopted a case-management model. Random sequence generation occurred using a computerised number sequence, and allocation concealment was accomplished using sequentially numbered opaque envelopes.
Women in the usual care arm participated in a supervised, mixed-sex, exercise programme 3 days per week for 12 weeks and attended education classes on risk factor modification. Those in the tailored group received usual care components plus an assessment of their readiness to change. Healthy eating, physical activity and stress levels were assessed using a computerised system producing tailored reports, stage-specific strategies and re-enforcement at baseline, 13 weeks and 37 weeks. Participants in the tailored group also received individualised psychotherapeutic sessions, using motivational interviewing techniques, provided by trained personnel (either a clinical psychologist or clinical nurse specialist). In addition, intervention participants received 10 sessions on gender-specific topics including behaviour change and relaxation.
The primary outcome of depressive symptoms was assessed by the 20-item Center for Epidemiological Studies Depression Scale (CES-D).6 Physiological measures such as blood pressure, body weight and exercise tolerance were also collected by an assessor blinded to treatment allocation. Analysis of variance was used to compare changes in CES-D6 scores over time.
A total of 1681 were screened, 1429 excluded, 252 randomised (n=111 to traditional, n=141 to tailored). At the 37-week end point, 83% of participants in the traditional and 94% in the tailored group were included in the analysis. Less individuals in the tailored group, 4.3% at 13 weeks and 2.8% at 37 weeks, were lost to follow-up compared with 11% and 6.3%, respectively, in the traditional group. The female-specific, tailored, intervention group showed significantly decreased CES-D scores from baseline to post-test (F (1, 223)=50.34, p<0.001); this was in spite of an increase of scores following the programme to the 6-month follow-up. Scores on the CES-D scores remained lower than baseline (F (1, 223)=19.25, p<0.001).
This is an important study designed to address the well-documented barriers to CR participation and depressive symptoms in women following an acute cardiac event. There have only been a handful of studies that have explored these issues, and few have used the systematic processes of a randomised controlled trial as undertaken in this study.7 There is no sample size calculation provided in this study in order to determine an a priori effect size and should be considered in future studies, particularly related to clinical significance. This study implemented strategies to have blinded outcome assessment and to address potential confounders such as smoking which could influence the level of depressive symptoms. The CES-D6 is a valid and reliable instrument, is appropriate in screening for depressive symptoms in this population and can readily be adopted in clinical practice.
As the attendance rate was higher in the tailored group, compared with the traditional group, this was investigated. In order to examine the influence of attendance on depressive symptom scores, a multiple regression model was constructed for predicting postintervention depression scores from baseline depressive symptoms, group assignment, education attendance and exercise attendance. Controlling for these factors, the influence of group assignment on depressive symptoms remained significant, suggesting the value of this gender-based approach.
Beckie and colleagues suggest that this is a multifaceted intervention where it is difficult to tease out what elements of the intervention contributed to the benefit observed. In future studies, methods of complex intervention analysis may assist to explain these factors.8 The applicability and utility of the intervention for clinical practice would benefit from a health economic intervention to describe the incremental benefit of this gender-based approach compared with traditional approaches. Further research should focus on providing gender-based approaches to decrease health disparities for women with heart disease.
Competing interests PMD has previously undertaken funded research on women and heart disease.