Author + information
- Received March 14, 2017
- Revision received May 15, 2017
- Accepted May 16, 2017
- Published online July 10, 2017.
- Madelon Minneboo, MDa,
- Sangeeta Lachman, MDa,
- Marjolein Snaterse, MScb,
- Harald T. Jørstad, MD, PhDa,
- Gerben ter Riet, MD, PhDc,
- S. Matthijs Boekholdt, MD, PhDa,
- Wilma J.M. Scholte op Reimer, PhDa,b,
- Ron J.G. Peters, MD, PhDa,∗ (, )
- on behalf of the RESPONSE-2 Study Group
- aDepartment of Cardiology, Academic Medical Center, Amsterdam, the Netherlands
- bACHIEVE Center of Applied Research, Faculty of Health, Amsterdam University of Applied Sciences, Amsterdam, the Netherlands
- cDepartment of General Practice, Academic Medical Center, Amsterdam, the Netherlands
- ↵∗Address for correspondence:
Dr. Ron J.G. Peters, Department of Cardiology, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands.
Background Among patients with coronary artery disease (CAD), improvement of lifestyle-related risk factors (LRFs) reduces cardiovascular morbidity and mortality. However, modification of LRFs is highly challenging.
Objectives This study sought to evaluate the impact of combining community-based lifestyle programs with regular hospital-based secondary prevention.
Methods The authors performed a randomized controlled trial of nurse-coordinated referral of patients and their partners to 3 widely available community-based lifestyle programs, in 15 hospitals in the Netherlands. Patients admitted for acute coronary syndrome and/or revascularization, with ≥1 LRF (body mass index >27 kg/m2, self-reported physical inactivity, and/or smoking) were included. All patients received guideline-based usual care. The intervention was based on 3 lifestyle programs for weight reduction, increasing physical activity, and smoking cessation. The primary outcome was the proportion of success at 12 months, defined as improvement in ≥1 qualifying LRF using weight (≥5% reduction), 6-min-walking distance (≥10% improvement), and urinary cotinine (200 ng/ml detection limit) without deterioration in the other 2.
Results The authors randomized 824 patients. Complete data on the primary outcome were available in 711 patients. The proportion of successful patients in the intervention group was 37% (133 of 360) compared with 26% (91 of 351) in the control group (p = 0.002; risk ratio: 1.43; 95% confidence interval: 1.14 to 1.78). In the intervention group, partner participation was associated with a significantly greater success rate (46% vs. 34%; p = 0.03).
Conclusions Among patients with coronary artery disease, nurse-coordinated referral to a comprehensive set of community-based, widely available lifestyle interventions, with optional partner participation, leads to significant improvements in LRFs. (RESPONSE-2: Randomised Evaluation of Secondary Prevention by Outpatient Nurse SpEcialists 2; NTR3937)
- comprehensive lifestyle intervention
- coronary artery disease
- health behavior
- improvement of physical activity
- nurse-coordinated care
- secondary prevention
- smoking cessation
- weight reduction
Patients with coronary artery disease (CAD) are at high risk of recurrent events and mortality. Improvement of lifestyle-related risk factors (LRFs), including overweight, physical inactivity, and smoking, is associated with a significantly lower risk of recurrent events (1,2). Therefore, guidelines on secondary prevention of CAD recommend medical treatment plus lifestyle interventions for all patients (3–5). However, a significant gap exists between guideline recommendations and daily practice. In particular, attempts at improving LRFs have been disappointing (6–8).
Most studies have focused on a single LRF, including counselling, support systems, or easy access (9–11). Nurse-coordinated referral to a comprehensive set of easily accessible, existing community-based programs has not been studied. In addition, most studies have not included patients’ partners, which may be essential to change a patient’s daily routines (12,13).
The RESPONSE-1 (Randomized Evaluation of Secondary Prevention by Outpatient Nurse Specialists) trial showed that nurse-led care was effective in reducing drug-treated cardiovascular risk factors and improving quality of life in patients with CAD (14,15). Guidelines now recommend the integration of nursing care into secondary prevention (16). However, the impact of nurse-led care on LRFs has been shown to be limited (11,15,17).
We hypothesized that a strategy of nurse-coordinated referral to a comprehensive set of ≤3 community-based, existing interventions to achieve weight loss, improvement of physical activity, and smoking cessation, on top of usual care, and including the patient’s partner, improves LRFs in patients with CAD.
The RESPONSE-2 trial was a randomized trial conducted in 15 hospitals in the Netherlands. Study methods have been published and are summarized later (18). The institutional review boards of all recruiting hospitals approved the protocol, and written informed consent was obtained from all patients. The protocol was registered at the Dutch trials register on April 8, 2013.
Adult patients were eligible <8 weeks after hospitalization for acute coronary syndrome, and/or coronary revascularization, if they had ≥1 of the following LRF: 1) body mass index (BMI) ≥27 kg/m2 (because a BMI only slightly >25 may not provide sufficient motivation, and minor improvement could be classified as success); 2) self-reported physical inactivity (<30 min of physical activity of moderate intensity 5 times per week, guideline based); or 3) self-reported smoking <6 months before hospital admission, and if motivated to attend ≥1 lifestyle program.
Exclusion criteria were planned revascularization after discharge; life expectancy ≤2 years; congestive heart failure New York Heart Association functional class III or IV; visits to outpatient clinic and/or lifestyle program not feasible; no Internet access; and anxiety or depressive symptoms (Hospital Anxiety and Depression Scale >14), because this was expected to hinder lifestyle changes (19).
After the baseline interview, patients were randomized by an automated online protocol to the intervention group or the control group, using randomly varying block sizes (4, 6, or 8 allocations), stratified by hospital (18).
All patients received usual care, including visits to the cardiologist and cardiac rehabilitation, according to national and international guidelines (4,5), and up to 4 visits to a nurse-led secondary prevention program. The nurse program addressed (counseling on) healthy lifestyles, drug-treated risk factors, and medication adherence (4,5,20). As per current guidelines, cardiac rehabilitation included ≤12 weeks of outpatient physical rehabilitation plus counselling on secondary prevention, psychological support, and work resumption.
Patients were seen by registered nurses, with experience in cardiovascular care and training in motivational interviewing.
Patients in the intervention group were referred by the nurse to ≤3 community-based lifestyle programs (18). The number and sequence of the lifestyle programs was determined by the patient’s risk profile and preference. Partners were offered free participation in the programs. Three lifestyle programs were used in their existing format, uniformly in all participants:
1. Weight Watchers offers a program that emphasizes a healthy diet, changing unhealthy behavior and regular physical activity, and uses group motivation, coordinated by a Weight Watchers’ coach. Access to this program was for the duration of 1 year.
2. Philips DirectLife offers an Internet-based program aimed at improving physical activity. An accelerometer measures physical activity and an online coach provides personalized feedback. Access to this program was for the duration of 1 year.
3. Luchtsignaal is a smoking cessation program in the Netherlands that uses telephone counselling based on motivational interviewing by trained professionals, for the duration of 3 months. Nicotine replacement or varenicline therapy was prescribed, as appropriate.
Data collection and measurements
Data were collected at baseline (first visit after discharge) and 12 months, including cardiovascular risk factors, cardiovascular history, physical activity, smoking status, and medication use. Blood pressure was measured twice by an automated sphygmomanometer and the average of the 2 was used. Body weight, height, and waist circumference were measured, and BMI was calculated. Body composition, including fat percentage, was analyzed using bioimpedance scales (Tanita scale SC-240-MA, Tokyo, Japan). Fasting blood samples were analyzed for lipid profiles and glucose. Urinary cotinine was measured (UltiMed one step, Dutch Diagnostic, Zutphen, the Netherlands; detection limit 200 ng/ml). The 6-min-walking distance (6MWD) was performed according to validated protocols (21,22). Partner participation was defined as a partner attending ≥1 visit to a lifestyle program.
The primary outcome was the proportion of successful patients at 12 months follow-up, defined as improvement of ≥1 qualifying LRF, without deterioration in the other 2. Improvement per LRF was defined as: 1) weight loss of ≥5%; 2) ≥10% increase in 6MWD; and 3) a urine cotinine level <200 ng/ml. Deterioration was defined as: 1) any weight gain in combination with a BMI >25 kg/m2; 2) any decrease in 6MWD compared with baseline; and 3) a positive cotinine test (>200 ng/ml) in nonsmokers at baseline.
Two exceptions were made: in patients who stopped smoking and/or improved their 6MWD, an increase of ≤2.5% in BMI was classified as no deterioration. Secondary outcomes included differences in isolated LRFs (weight, 6MWD, and smoking), attendance rates of lifestyle programs, blood pressure, lipid profiles, and hospital readmissions.
Sample size calculation
We estimated that a sample size of 790 patients, with a withdrawal rate of 7%, had 80% power at a 5% significance level in a 2-sided test to detect at least a 30% relative increase in the proportion of successful patients (18).
All participants were analyzed by initial group assignment, irrespective of attending a lifestyle program. The primary analysis compared the proportion of successful patients at 12 months between the 2 groups. The treatment effect was expressed as risk ratio (RR) and the corresponding 95% confidence intervals (CIs) based on chi-square test. Where appropriate, logistic regression analysis was used. Continuous variables are presented as means with standard deviation for normally distributed data, and as median with quartiles (Q1 and Q3) for non-normally distributed data. Categorical variables are presented as frequencies and percentages.
A predefined multiple imputation analysis was performed to analyze the impact of selective dropout (18) using iterative chained equations, separately for the intervention and control group (Online Table 1). RR was calculated for each of the 50 imputed sets and pooled using Rubin rule to derive the correct CI. Adjustment was made for the 7 possible subgroups of LRFs combinations (smoking only, BMI ≥27 kg/m2 only, physical inactivity only, smoking and BMI ≥27 kg/m2, smoking and physical inactivity, BMI ≥27 kg/m2 and physical inactivity and smoking, BMI ≥27 kg/m2 and physical inactivity).
All statistical tests were 2-tailed and a p value of <0.05 was considered significant. We used IBM SPSS Statistics version 23.0 (IBM, Armonk, New York) and Stata version 13.1 (Stata Corp., College Station, Texas).
From April 2013 to July 2015, a total of 2,031 patients with CAD were screened for enrollment. Based on exclusion criteria, 994 were ineligible (Figure 1). In total, 824 patients provided informed consent and were randomized, of whom 731 patients attended the 12-month follow-up visit. In 20 patients, outcome data for the primary analysis were incomplete, and 711 patients were included in the primary analysis (360 intervention, 351 control) (Figure 1).
Patients who did not attend the 1-year follow-up visit (36 intervention, 57 control subject; p = 0.02) were more frequently younger, had a higher education profile, and were more often smokers, as compared with those who completed the 1-year follow-up visit (55.0 vs. 58.7 years, p < 0.001; higher education profile, 50% vs. 39%, p = 0.05; smokers, 33% vs. 21%, p = 0.01).
Patients (overall) had a mean age of 58.7 ± 9.2 years and 21% were female (Table 1). Most patients (66%) had no history of cardiovascular disease before the index hospitalization. At the baseline visit, 21% of the patients were smoking, and 27% had quit at (or <6 months before) hospital admission. Overweight (BMI >25 kg/m2) was present in 87%, and 63% did not meet the target for adequate physical activity. A total of 36% of patients had 1 LRF, 45% had 2 LRFs, and 19% had 3 LRFs. At baseline (i.e., after discharge), the use of preventive medication was high (98% antiplatelet therapy, 85% beta-blockers, 74% angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and 97% lipid-lowering drugs). A cardiac rehabilitation program was followed by 91% of the patients in both groups. Overall, 82% (582 of 711) of patients were living with a partner.
Attendance to the lifestyle programs
In the intervention group, 85% (305 of 360) of patients followed ≥1 lifestyle program: 48% (174 of 360) followed 1 lifestyle program, a total of 34% (121 of 360) followed 2 lifestyle programs, and 3% (10 of 360) followed 3 lifestyle programs. The frequency and duration of attendance varied per lifestyle program and per patient (Online Figure 1, Online Table 2).
The proportion of successful patients, who improved ≥1 of the nonoptimal LRFs without deterioration in the other LRFs at 12 months, was 37% (133 of 360) in the intervention group compared with 26% (91 of 351) in the control group, a RR of 1.43 (95% CI: 1.14 to 1.78; p = 0.002) (Table 2, Central Illustration). Dropout was 8.8% and 13.8% in the intervention and control group, respectively. After multiple imputation and reanalysis of the primary outcome, the RR remained significant: 1.48 (95% CI: 1.18 to 1.86; p = 0.001). The RR remained significant after adjusting for LRF groups: 1.46 (95% CI: 1.17 to 1.82; p = 0.001).
Improvement in ≥2 LRFs (without deterioration in the third LRF) was seen in 47 (13%) in the intervention group compared with 20 (6%) in the control group (p = 0.001) (Table 2).
We observed a significantly higher rate of ≥5% weight reduction in the intervention group as compared with the control group (27% vs. 14%; p < 0.001), respectively. Weight reduction to a BMI ≤25 kg/m2 was achieved in 15% of patients in the intervention group compared with 11% in the control group (p = 0.10) (Table 2). Overall improvement in the 6MWD was seen in 45% and 40% (p = 0.13). Negative cotinine tests were found in 76% and 74%, respectively (p = 0.55) (Table 2).
The subgroup of patients eligible to receive each individual intervention program, and the effects on the relevant LRFs are shown in Online Table 3. The proportion of patients with 3 LRFs at baseline who attained the risk factor goal for each risk factor is shown in Online Table 4.
In both groups, living with a partner (irrespective of participation) was associated with a greater proportion of success (Figure 2). Among patients in the intervention group who had a partner (298 of 360; 82.8%), partner participation in a lifestyle program (137 of 298; 46%) was associated with a significantly greater success rate: (46% vs. 34%; p = 0.03).
Self-reported adherence to medication at 12 months was high in both groups: 92% and 93% were on antiplatelet therapy, 66% and 73% on beta-blockers, 69% and 67% used angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and 93% and 93% were on lipid-lowering drugs in the intervention and control group, respectively. A systolic blood pressure <140 mm Hg was observed in 70% in the intervention group compared with 62% in the control group (p = 0.04). In the intervention group, low-density lipoprotein cholesterol <70 mg/dl was 34% compared with 38% in the control group (p = 0.23). There were no significant differences in hospital readmission rates between intervention and control group (94 of 360 [26%] patients and 102 of 351 [29%] patients; p = 0.63).
The main finding of the RESPONSE-2 trial is that among patients with CAD, nurse-coordinated referral to a comprehensive set of up to 3 widely available community-based lifestyle programs, with encouragement of partner participation, on top of usual care, is more effective in improving LRF than usual care alone. One in 3 individuals in the intervention group successfully improved their LRFs without deterioration in the others; an absolute increase of 11%, and a relative increase of 42% compared with the control group. The number needed to treat was 10. This improvement in LRFs was seen in spite of a rigorous definition of success and against the background of a high standard of usual care based on contemporary guidelines, with all patients receiving care from cardiologists and specialized nurses, in addition to cardiac rehabilitation (4,5). The approach was highly practical, which suggests that wide application is feasible. Although the overall rates of success are modest, our findings do provide clinicians with evidence-based options in patients who may be suitable candidates.
Key to our intervention is the comprehensive approach in targeting LRFs. Nurses explained the interactions between the interventions and the importance of reducing in overall risk. Patients were offered a choice of 3 different lifestyle programs, depending on their risk profiles and motivation. According to individual preferences, patients were able to follow ≥1 program, sequentially or simultaneously. Partners were encouraged to participate in the lifestyle programs. Consistent with previous observations, patients with partner participation had the highest proportion of success (23,24). This tailored, comprehensive approach may have enabled more patients to change their daily routines.
Previous trials have reported modest rates of success in modifying single LRFs in patients with CAD (11,15,17,25–27). However, none of these trials have taken deterioration of other LRFs into account when analyzing success rates. Although it may be possible to achieve significant improvement in 1 LRF, this should not come at the cost of deterioration of other LRFs. Our primary outcome therefore had a stricter definition of success, which incorporates deterioration in other LRFs. Minimal weight gain was accepted with smoking cessation (because of a net benefit in overall risk) or with improvement of physical activity (allowing for a small increase in lean body mass). Even with this strict definition of success, a significantly larger success rate was seen in the intervention group as compared with the control group. For isolated LRFs, success rates were considerably higher: 50% versus 46% for smoking cessation, 33% versus 15% for weight reduction, and 45% versus 40% for improvement in the 6MWD (intervention vs. control).
The community-based lifestyle programs were offered uniformly, in their existing commercial format. Most patients followed ≥1 program (84%), and a 42% of those followed ≥2 programs. The effect of the weight reduction component was the most pronounced in our study. Almost twice as many patients achieved significant weight loss in the intervention group as in the control group. The effects of the intervention program on smoking cessation and on physical activity were modest. Although this suggests differences in effectiveness among the programs, we believe that a comprehensive approach is key. Nurses explained the concept of overall risk and the impact of multiple interventions. Moreover, the repeated and consistent attention to risk and lifestyle in the separate interventions may reinforce the information and support patients in their efforts to change their daily routines.
Against a background of very high levels of preventive drug therapy, we found no improvement of LDL cholesterol levels, and a small, albeit statistically significant difference, in systolic blood pressure control in favor of the intervention group. The proportion of patients who reached the low-density lipoprotein cholesterol target was disappointing in both groups, despite the high level of care offered by multiple caregivers, and a high prescription rate of lipid-lowering medication. Lifestyle interventions may not be expected to improve these risk factors, and possibly more potent statin therapy was required. However, this was not analyzed.
Patients were eligible based on European Society of Cardiology guideline criteria for LRFs, including self-reported smoking and physical activity. However, for outcome assessment we used objective measurements. Thus, although part of the primary outcome, urinary cotinine and 6MWD were not used as inclusion criterion.
Physical activity is preferably measured using an accelerometer. However, this instrument was part of the physical activity program and using it for outcome assessment in both groups would have reduced contrast, in addition to introducing a Hawthorne effect (28). Physical fitness is ideally assessed using maximal exercise performance (VO2max); however, this was not feasible. As self-report is unreliable (29), we selected the 6MWD as a practical and objective tool. Participation in the lifestyle programs was free of charge. In clinical practice, the costs of the interventions may limit the generalizability of our findings.
We collected outcome measurements at 12 months only. Therefore, we cannot analyze the dynamics of change during the first year. However, by design patients varied in the selection and the sequence of lifestyle programs and observations before 12 months would therefore be less meaningful. Using a 12-month outcome, we included some time for potential loss of effect after initial success. Clearly, to reduce the risk of adverse events, longer persistence of improvements is required. This will be addressed in a follow-up investigation.
The RESPONSE-2 trial demonstrates that nurse-coordinated referral of patients with CAD and their partners to a comprehensive set of lifestyle programs, using up to 3 community-based interventions, improves LRF significantly more than usual care alone. Partner participation was associated with a higher rate of success. Referral to these widely available programs requires minimal effort, and this strategy is easily implemented into daily practice to improve secondary prevention of CAD.
COMPETENCY IN SYSTEMS-BASED PRACTICE: Nurse-coordinated referral to comprehensive community-based lifestyle interventions with optional partner participation can enhance secondary prevention in patients with coronary artery disease.
TRANSLATIONAL OUTLOOK: Longer term follow-up studies are needed to confirm the persistence of improvements in lifestyle-related cardiovascular risk factors.
The authors are grateful to all patients participating in the RESPONSE-2 trial, and the contribution of all research nurses: C. de Jong, A. van Dullemen, W. van der Poel, J. Doornenbal, M. Smit, B. van der Linden, S. Tanovic, N. Tenbult-van Limpt, M. Leguit, E. Dijkstra, J. Fischer, H. Saarloos, H. van Lint, H. Groeneweg, I. Kremer, C. Coenjaerds, K. Lansink, A. van Goor, E. J. Wolf, E. de Haan, M. van Dijkhuizen, A. Meissner, M. van Steenbergen, Z. Aukema-Wouda, I. Sterk, M. Damen-de Vries, M. Zootjes-Mes, T. Eltink, W. Glas, A. Obbema, A. Reijenga, and E. de Jong.
For a list of the RESPONSE-2 Study Group members as well supplemental information on the lifestyle programs and imputation including tables and figures, please see the online version of this article.
The RESPONSE-2 trial was sponsored by Weight Watchers International, Inc. (New York, New York), Philips Consumer Lifestyle (the Netherlands), and an anonymous private fund (Amsterdam, the Netherlands). The sponsors had no role in the design, data collection, data analysis, data interpretation, and writing of the manuscript. Dr. Boekholdt has received personal fees from Pfizer. Dr. Peters has received personal fees from Sanofi, Boehringer Ingelheim, Amgen, and AstraZeneca. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Minneboo and Lachman contributed equally to this work.
- Abbreviations and Acronyms
- 6-min-walking distance
- body mass index
- coronary artery disease
- confidence interval
- lifestyle-related risk factor
- risk ratio
- Received March 14, 2017.
- Revision received May 15, 2017.
- Accepted May 16, 2017.
- 2017 American College of Cardiology Foundation
- Chow C.K.,
- Jolly S.,
- Rao-Melacini P.,
- Fox K.A.,
- Anand S.S.,
- Yusuf S.
- Iestra J.A.,
- Kromhout D.,
- van der Schouw Y.T.,
- Grobbee D.E.,
- Boshuizen H.C.,
- van Staveren W.A.
- Eckel R.H.,
- Jakicic J.M.,
- Ard J.D.,
- et al.
- Perk J.,
- De B.G.,
- Gohlke H.,
- et al.
- Smith S.C. Jr..,
- Benjamin E.J.,
- Bonow R.O.,
- et al.
- Kotseva K.,
- Wood D.,
- De Bacquer D.,
- et al.
- Wood D.A.,
- Kotseva K.,
- Connolly S.,
- et al.
- Sher T.,
- Braun L.,
- Domas A.,
- Bellg A.,
- Baucom D.H.,
- Houle T.T.
- Jorstad H.T.,
- Minneboo M.,
- Helmes H.J.,
- et al.
- Jorstad H.T.,
- von B.C.,
- Alings A.M.,
- et al.
- Authors/Task Force M,
- Piepoli M.F.,
- Hoes A.W.,
- et al.
- Allen J.K.,
- Dennison-Himmelfarb C.R.,
- Szanton S.L.,
- et al.
- Lachman S.,
- Minneboo M.,
- Snaterse M.,
- et al.
- Ijzelenberg W.,
- Hellemans I.M.,
- van Tulder M.W.,
- et al.
- Bittner V.,
- Bertolet M.,
- Barraza Felix R.,
- et al.
- Maron D.J.,
- Boden W.E.,
- O'Rourke R.A.,
- et al.
- Wanner M.,
- Richard A.,
- Martin B.,
- Faeh D.,
- Rohrmann S.