Health Education Research Advance Access published January 7, HEALTH EDUCATION RESEARCH 2014 Pages 1 12







Full text


Evaluation of a standardized patient education program

for inpatient cardiac rehabilitation: impact on illness

knowledge and self-management behaviors up to 1 year

Karin Meng


*, Bettina Seekatz


, Gu¨nter Haug


, Gabriele Mosler



Bernhard Schwaab


, Ulrike Worringen


and Hermann Faller


1Department of Medical Psychology, Medical Sociology, and Rehabilitation Sciences, University of Wu¨rzburg, D-97070

Wu¨rzburg, Germany,2Rehabilitation Center Bayerisch Gmain, D-83457 Bayerisch Gmain, Germany,3Rehabilitation

Hospital Ho¨henried, D-82347 Bernried, Germany,4Curschmann Klinik, Rehabilitation Hospital, D-23669 Timmendorfer

Strand, Germany and5Department Rehabilitation, German Statutory Pension Insurance Scheme, D-10709 Berlin, Germany

*Correspondence to K. Meng. E-mail:

Received on March 13, 2013; accepted on November 25, 2013


Patient education is an essential part of the treat-ment of coronary heart disease in cardiac re-habilitation. In Germany, no standardized and evaluated patient education programs for coron-ary heart disease have been available so far. In this article, we report the evaluation of a patient-oriented program. A multicenter quasi-experimental, sequential cohort design study of patients with coronary heart disease (n¼434) in inpatient cardiac rehabilitation was conducted. Intervention patients received the new patient-oriented program, control patients a traditional lecture-based program (usual care). Primary outcome illness knowledge and secondary behav-ioral and health outcomes were assessed at admission, discharge and 6- and 12-months follow-up. We found a significant, small be-tween-group intervention effect in both patients’ medical illness and treatment knowledge and be-havior change knowledge at discharge (medical: g2¼0.013; behavior change: g2¼0.011) and after 12 months (medical: g2¼0.015). Furthermore, a significant, small effect was observed for physical activity after 12 months (g2¼0.011), but no effects on healthy diet and medication adherence emerged. Superiority of the patient-oriented educational program for pa-tients with coronary heart disease was partially

confirmed. The program produced improved ill-ness knowledge and physical activity compared with usual care after 1 year.


Patient education is an essential part of the medical rehabilitation of patients with coronary heart disease targeting self-management behavior to reduce risk factors and subsequent cardiac events [1–3]. International meta-analyses provide evidence for the effectiveness of secondary prevention programs, psychological interventions and educational inter-ventions regarding risk factors and mental health, whereas evidence for mortality, myocardial infarc-tion and quality of life is heterogeneous [4–7]. However, transfer of study results on individual pro-grams seems complex because of differences in both interventions and rehabilitation settings. Cardiac re-habilitation in Germany [8] is predominantly offered as comprehensive inpatient treatment lasting for 3 weeks. It is accessible for patients with myocardial infarction, surgery or catheter-based interventions after discharge from acute hospital care or patients with chronic course of disease and subsequent func-tional impairment. For inpatient cardiac rehabilita-tion, no standardized and evaluated educational group program for patients with coronary heart dis-ease has so far been available for routine use.


Moreover, many German patient education pro-grams still lack certain quality requirements, such as the use of manuals, patient-oriented didactics, small-group format and evaluation of effectiveness [9, 10]. Furthermore, theory-based techniques to foster health behaviors [11–13] are only rarely employed. In addition, few studies have compared different educational approaches applied within a multidisciplinary rehabilitation program in an inpa-tient setting [e.g. 14–16]. Overall, further studies are needed to explore the effects of patient education programs, intervention techniques and subgroups of patients who benefit most.

The aim of our study was to evaluate the short-, intermediate- and long-term effects of a new patient-oriented educational program as compared with a traditional lecture-based program (usual care) for patients with coronary heart disease receiving inpa-tient medical rehabilitation. As Seekatzet al. [17] reported, the program was developed by an interdis-ciplinary group of health professionals and scientists and was based on a health education curriculum of the German Statutory Pension Insurance Scheme, which comprises an educational framework and about 25 curricula for specific indications including cardiovascular disease. Both topics and didactics in-corporate research evidence, practice guidelines and quality criteria for educational programs and were based on theories of health and illness behavior (such as the Health Action Process Approach [18] and the Common-sense model of illness-representations [19]). Theories were used to design the intervention (determine target constructs and intervention techniques) and to define proximal out-comes. The patient orientation aspect of the program was defined by the active involvement of the pa-tients in the whole educational process. Short-term results reported elsewhere [17] have shown a super-ior treatment effect as compared with the lecture-based program (usual care) on primary outcome illness knowledge at discharge from inpatient re-habilitation. Furthermore, some short-term effects were observed among secondary outcomes, such as medication beliefs, planning of physical activity after rehabilitation and treatment satisfaction. In this article, we present effects of the program on primary

outcome as well as intermediate- and long-term ef-fects on behavioral outcomes up to 1 year. We hypothesized that the patient-oriented program was superior to usual care regarding illness know-ledge not only in short-term, but also in intermedi-ate- and long-term. In addition, we expected superior effectiveness of the new program regarding several self-management behaviors, such as phys-ical activity, healthy diet and medication adherence. Moreover, moderator effects of type of rehabilita-tion—cardiac rehabilitation within 14 days after an acute cardiac index event (aCR) versus cardiac re-habilitation during the chronic course of disease without recent acute index event (cCR)—were explored.

Materials and methods

Study design and data collection

A multicenter quasi-experimental, sequential cohort design study of patients with coronary heart disease (n¼434) in inpatient cardiac rehabilitation was per-formed to compare the effectiveness of the patient-oriented educational program with a lecture-based program (usual care). First, usual care treatment (control condition) was performed; afterwards the new program was implemented in the clinics and carried out thereafter (intervention condition). Inclusion criterion was a primary diagnosis of cor-onary heart disease (ICD-10-GM: I20–I25, Z95.1, Z95.5). Exclusion criteria were: inadequate German language ability, age<18 years or>70 years, severe visual or hearing impairment, severe co-morbid psy-chiatric disorder, cognitive impairment and/or sub-mitted pension application. Data were assessed at admission (t1), discharge (t2) and after 6- and 12-months (t3, t4) with patient-reported questionnaires. Sample size was powered to detect small to medium effects in the primary outcome (d¼0.3, two-sided

a¼0.05, 1b¼0.8). Therefore, 352 persons were required. Power to detect smaller between-group ef-fects on the secondary outcomes may be too low; thus, effect sizes were reported throughout. The study was approved by the Ethics Committee of the Faculty of Medicine, University of Wu¨rzburg.


Patients were consecutively recruited for the study in two cardiac rehabilitation hospitals between May 2010 and September 2011; 1 year follow-up was finished in September 2012.Figure Ishows the participant flow throughout the study. Inclusion cri-teria were evaluated at admission by a physician. In total, 679 patients eligible were asked to participate in the study and 471 participated (69%) and signed the informed consent form. About 37 persons had to be excluded (e.g. withdrew consent, did not com-plete the questionnaire at baseline), finally 434 per-sons comprised the initial study sample. Follow-up rates exceeded 77% at all time points. Analyses were performed for all persons who participated in at least one follow-up assessment. Non-responder analyses revealed no differences between partici-pants and non-participartici-pants for study group, gender and primary diagnosis, but rehabilitants admitted directly after an aCR were less willing to participate (P<0.001). Drop-out analyses for follow-up points

were performed for study group, socio-demographic variables, type of rehabilitation (aCR, cCR) and main outcome variables. Results indicated no sys-tematic sample bias because of patient drop-out over time. In particular, no significant differences be-tween completers and drop-outs were found at dis-charge. However, patients who had not provided data at the 6- and 12-months follow-ups, respect-ively, were more often single, had more often endorsed aCR, had lower illness knowledge, were more often smokers and showed less physical activ-ity at baseline. The proportions of drop-outs were similar for both study groups and there were no sys-tematic baseline differences between the follow-up samples (data not shown).


The inpatient multidisciplinary rehabilitation [8] included medical treatment, exercise


therapy/physical training, health education, psycho-logical support, relaxation and social counseling and covered a 3 weeks period on average. In this trial, two basic educational programs that differed in pro-gram content, didactics/methods of delivery as well as group format were compared.

Intervention group

Patients in the intervention condition received a new patient-oriented program ‘Curriculum Coronary Heart Disease’ that was determined by a manual and consisted of five patient-oriented interactive ses-sions of 45 min each, which were held in small groups of a closed format (15 participants or less). The program was interdisciplinary with sessions led by a physician, a psychologist and a physiotherapist, respectively. In each session, patients were actively involved in the educational process using a combin-ation of didactic methods (short lectures, group dis-cussion and individual work). Didactic materials included presentations, flipcharts and a patient booklet. Contents of the lessons included basic knowledge about coronary heart disease (e.g. func-tioning of the heart, etiology and symptoms, risk factors, coronary intervention/bypass graft surgery and medication and medication adherence) with regard to individual needs of the participants. To foster medication adherence, unintentional and in-tentional barriers and corresponding management strategies were discussed in addition to information on effects and side-effects. To promote health be-havior, particularly physical activity and healthy diet, a theory-based intervention (Health Action Process Approach; [18]) was applied. In the first session, patients were encouraged to select those cardiac risk factors that were relevant for them per-sonally. In the fourth session, recommendations for health-related behaviors for cardiac patients were discussed based on patients’ experience made during the inpatient rehabilitation and participants could reflect on their own goals with regard to their risk profile; goal setting was done as homework. In the last session, participants were instructed on how to make both action plans and coping plans and how to self-monitor behavior after rehabilitation

(intervention techniques according to Michieet al. [12]: Technique No. 1, 2, 5–8, 10, 16).

Control group

Control condition was the traditional lecture-based educational program as it had been practiced in the hospitals for years (usual care). It consisted of two to four lectures of medical education by a physician with an open format, no limitation of group size and an overall duration of about 180 min. Information was mostly presented in a vertical manner, but patients could ask questions. Contents included basic illness information on epidemiology, anatomy, cardiac disorders, risk factors and medical and behavioral treatment as lifestyle modification.

Outcomes and measurements

The primary outcome was patients illness know-ledge on coronary heart disease and its treatment. Secondary outcomes included behavioral determin-ants, health behavior (physical activity and healthy diet), beliefs about medication and medication adher-ence, health status and treatment satisfaction. This article focuses on the primary outcome and behav-ioral outcomes—physical activity, healthy diet and medication adherence—that were of main interest at the follow-up time points. Behavioral outcomes were assessed by well-validated measures. To assess ill-ness knowledge a new measure was developed.

Illness knowledge

Patients were asked to judge 34 statements about coronary heart disease and its treatment as true or false. Items were summed up to yield two scores: (i) medical illness and treatment knowledge (range: 0–22) and (ii) behavior change knowledge (range: 0–12) with higher scores corresponding to higher knowledge; missing answers were counted as wrong. Item examples: (i) ‘What are risk factors for myocardial infarction: diabetes? physical stress?’ (ii) ‘What are positive effects of physical activity for patients with coronary heart disease: de-crease of blood pressure? improvement of lipids?’ The questionnaire was developed based on a pre-test (108 items), with item selection according to item


difficulty (0.10P0.90), proportion of missings (<15%) and content (content validity).

Physical activity

Participants reported how often per week and how long per session they performed strenuous, moder-ate, and light physical exercise (modified version of the Godin Leisure-Time Exercise Questionnaire; [20]) outside of work duties. A total physical activity score (in minutes per week) was calculated by total number of sessions per week in each domain multi-plied by minutes per session in each domain.

Healthy diet

A nine-item modified version of the self-report Food List (LML Lebensmittelliste; [21]) was adminis-tered. Patients assessed how often per week they were eating healthy (e.g. fruits and salads) or un-healthy (e.g. fried potatoes and ‘fast food’) food on a 5-point scale (1¼seldom or never; 5¼several times a day). Total scores may range from 0 to 33, with higher scores indicating less healthy diet.

Medication adherence

Patients completed the German version of the Medication Adherence Report Scale (MARS-D; [22]). The MARS-D is a five-item measure with a 1 (always) to 5 (never) response format that assesses patients’ non-adherence behavior (e.g. forgetting to take a dose or altering the dose). Items were summed up to form a sum score ranging from 5 to 25 points, with higher scores indicating greater medication adherence.

Statistical analysis

All statistical analyses were performed using SPSS 18.0 for Windows. Missing data were imputed using a multiple imputation procedure. Missing values due to drop-out were analyzed by pair-wise deletion. Non-response analyses and drop-out analyses were carried out by independent group comparisons using t-tests for continuous variables and chi-square test for categorical variables. Treatment effects (between-group effects) were evaluated separately for each

follow-up time point using analysis of covariance (ANCOVA) adjusting for baseline values [23]. Statistical significance (P<0.05, two-sided) and effect sizes (2) were reported for all between-group differences [24]. Bonferroni corrections were done to adjust the significance level for multiple comparisons within outcome domains (illness know-ledge, two tests: P¼0.025). In addition, within-group effects, including standardized effect sizes (SES) and accompanying 95% confidence intervals (CIs) were reported for both study groups. Moderator analysis was performed by including the moderator variable as an additional factor in the ANCOVA and examining interaction effects. The significance level for interaction effects was not adjusted due to their exploratory nature. When trends to significant inter-actions emerged (P<0.10), simple effects analyses were performed by subgroups.


Sample characteristics are presented inTable I. The initial sample consisted of 434 rehabilitants with coronary heart disease. Patients’ mean age was 54 years (SD¼6.6), 94% were male and 83% reported to live with a partner. The sample mostly consisted of employed persons (82%). The number of sick-leave days within the last 12 months averaged 27 (SD¼57.1, median¼8). At the time of admission, 61% had been declared unable to work. Subjective health status as measured by the SF-12 [25] was reduced compared with the general population. As

Table Ishows, no systematic differences existed be-tween intervention and control group concerning socio-demographic and medical data as well as type of rehabilitation.

Primary outcome

Participants of the intervention group as compared with the control group showed superior illness knowledge (primary outcome) representing a sig-nificant, albeit small treatment effect at discharge from inpatient rehabilitation (medical illness and treatment knowledge: P¼0.024, 2¼0.013, SEScontrol group¼0.25, SESintervention group¼0.46;


behavior change knowledge:P¼0.039,2¼0.011, SEScontrol group¼0.06, SESintervention group¼0.21) as

well as after 12 months on medical illness and treat-ment knowledge (P¼0.015,2¼0.018) but not on health behavior change knowledge; there were no significant effects 6 months after rehabilitation (Tables IIandIII).

Intermediate-term effects on secondary

behavioral outcomes

Between-group effects as well as within-group ef-fects at 6 months are shown inTable II. Both inter-vention and control groups showed improvements in

healthy diet and the amount of physical activity. However, SESs indicated a rather small effect for physical activity that fails significance in the inter-vention group. There was no significant increase for medication adherence in both groups that may be due to high baseline scores. Accordingly, between-group analyses (ANCOVA) revealed no significant treatment effects on behavioral outcomes.

Long-term effects on secondary behavioral


Between-group effects at 12 months and within-group SES for baseline to 12-months follow-up Table I. Baseline characteristics of the sample

Control group (n¼214)

Intervention group

(n¼220) P-value

Male,n(%) 200 (93.5) 205 (93.2) 0.908 Age in years, mean (SD) 53.1 (6.6) 53.9 (6.5) 0.242

Marital status,n(%)a 0.800 Single 20 (9.4) 27 (12.4) Married 160 (75.1) 159 (72.9) Divorced/separated 31 (14.6) 30 (13.8) Widowed 2 (0.9) 2 (0.9) Education,n(%)a 0.587

Less than junior (<10 years; basic secondary school) 116 (54.5) 116 (53.2) Junior (10 yyears; middle-level secondary school) 58 (27.2) 53 (24.3) Senior (high-school graduate)b 30 (14.1) 41 (18.8)

Other 9 (4.2) 8 (3.7)

Working status,n(%)c 0.384

Employed 169 (79.7) 184 (83.6)

Retired 8 (3.8) 11 (5.0)

Unemployed 18 (8.5) 15 (6.8) Main diagnosis (ICD-10)

Ischemic heart diseases (I20–I25),n(%) 212 (99.1) 218 (99.1) 1.00 Presence of aortocoronary bypass graft (Z95.1),n(%) 45 (21.0) 37 (16.8) 0.263 Presence of coronary angioplasty implant and graft


127 (59.3) 136 (61.8) 0.598

Number of cardiac risk factors, mean (SD) 3.9 (1.4) 4.1 (1.5) 0.155 BMI, mean (SD)d 28.3 (4.2) 28.9 (5.1) 0.147 Days of sick leave in the last year, mean (SD)e 25.4 (50.0) 27.9 (63.4) 0.660 SF-12 physical componentf, mean (SD) 40.2 (9.9) 41.1 (9.8) 0.353 SF-12 mental componentf, mean (SD) 43.4 (13.1) 45.5 (11.9) 0.074 Inpatient rehabilitation,n(%) 0.234

Cardiac rehabilitation within 14 days after an acute cardiac index event (aCR)

144 (67.3) 136 (61.8)

Cardiac rehabilitation during the chronic course of disease without recent acute index event (cCR)

70 (32.7) 84 (38.2)


missing n¼3;bQualification for university or university of applied sciences entrance.cmissing n¼2;dmissing n¼4; emissing n¼11;fSF-12 short form health survey. Score ranges from 0 to 100, with high scores representing best health status. BMI, body mass index.


Table III. Long-term (12 months) within-group and between-group effects on primary and secondary outcomes


Twelve months

follow-up Within-group change

Between-group difference ANCOVA n Mean (SD) Mean (SD) SES (95% CI) P 2

Medical illness and treatment knowledge 0.015 0.018

Control 158 13.63 (3.22) 13.54 (3.88) 0.03 (0.20 to 0.14) Intervention 176 13.60 (3.53) 14.35 (3.32) 0.22(0.07 to 0.37)

Behavior change knowledge 0.309 0.003 Control 158 8.06 (1.80) 8.18 (1.91) 0.06 (0.14 to 0.26)

Intervention 176 7.93 (2.07) 8.09 (1.90) 0.08 (0.06 to 0.22)

Physical activity (minutes per week) 0.053 0.011

Control 158 214.72 (170.39) 237.47 (189.55) 0.13 (0.05 to 0.31) Intervention 176 215.85 (184.90) 278.14 (218.56) 0.35(0.19 to 0.51) Healthy diet 0.926 <0.001 Control 158 14.09 (4.70) 11.92 (4.33) 0.48 (0.30 to 0.66) Intervention 176 14.23 (4.31) 12.03 (4.41) 0.49 (0.35 to 0.63) Medication adherence 0.131 0.007 Control 158 23.82 (1.70) 23.80 (1.87) 0.01 (0.14 to 0.12) Intervention 176 24.09 (1.17) 24.18 (1.15) 0.06 (0.11 to 0.22)

Positive SES represents improvement. Effect size2(small:2¼0.0099, medium:2¼0.0588, large:2¼0.137). Bold figures indicate: (i) regarding within-group effects: SES of the control group is outside the 95% CI of SES of the intervention group or (ii) regarding between-group effects: significant between-group effect (ANCOVA:P<0.05). Italics bold indicate between-group effects of2

0.01 (P<0.10).

Table II. Intermediate-term (6 months) within-group and between-group effects on primary and secondary outcomes


Six months

follow-up Within-group change

Between-group difference ANCOVA n Mean (SD) Mean (SD) SES (95% CI) P 2

Medical illness and treatment knowledge 0.399 0.002 Control 167 13.47 (3.44) 14.44 (2.70) 0.28 (0.10 to 0.45)

Intervention 179 13.58 (3.60) 14.39 (3.42) 0.23 (0.08 to 0.38)

Behavior change knowledge 0.789 <0.001 Control 167 7.87 (1.99) 8.17 (1.68) 0.15 (0.04 to 0.33)

Intervention 179 7.96 (2.06) 7.98 (1.88) 0.08 (0.16 to 0.17)

Physical activity (minutes per week) 0.940 <0.001 Control 167 213.34 (178.36) 245.07 (174.43) 0.18 (0.01 to 0.34) Intervention 179 222.25 (182.07) 247.31 (191.31) 0.14 (0.02 to 0.30) Healthy diet 0.220 0.004 Control 167 13.99 (4.62) 11.08 (4.19) 0.64 (0.48 to 0.80) Intervention 179 14.12 (4.47) 11.61 (4.23) 0.55 (0.41 to 0.70) Medication adherence 0.918 <0.001 Control 167 23.88 (1.61) 24.02 (1.43) 0.10 (0.03 to 0.23) Intervention 179 24.09 (1.16) 24.15 (1.27) 0.04 (0.11 to 0.19)


are presented inTable III. One year after rehabilita-tion, significant increases in healthy diet were main-tained among patients of both groups. Furthermore, a significant increase in the patients’ physical activ-ity was evident for the intervention group; SESs indicate that the intervention group improved more than the control group. Again, medication adherence was unchanged at a high level.

ANCOVAs resulted in a nearly significant, small effect in favor of the intervention group for physical activity (P¼0.053). After adjustment for baseline imbalance, estimated mean physical activity per week in the intervention group was about 40 min longer than in the control group. No significant dif-ferences between the two programs were observed for healthy diet and medication adherence, though.

Moderator analysis

Rehabilitants after an aCR were significantly differ-ent from those without a preceding acute index event but with cCR regarding illness knowledge (aCR<cCR), health behavior (aCR<cCR) and medication adherence (aCR>cCR) as well as mental health (aCR>cCR) and age (aCR<cCR) at baseline.

Moderator analysis showed no significant inter-action effects for illness knowledge at any follow-up time (12-months follow-up: ANCOVA medical ill-ness and treatment knowledge: main effect treat-ment group: P¼0.007, 2¼0.022; main effect

type of rehabilitation: P¼0.889, 2¼<0.001; interaction effect treatment grouptype of rehabili-tation: P¼0.130, 2¼0.007. ANCOVA behavior change knowledge: main effect treatment group: P¼0.492,2¼0.001; main effect type of rehabili-tation: P¼0.874, 2¼<0.001; interaction effect treatment grouptype of rehabilitation: P¼0.190,2¼0.005).

However, type of rehabilitation showed a signifi-cant interaction for total physical activity after 12 months (ANCOVA;Fig. II). Simple effect analyses indicated a significant between-group effect on physical activity among rehabilitants with aCR, but not among rehabilitants with cCR (aCR:

P¼0.021, 2¼0.016. cCR: P¼0.521,

2¼0.001). The aCR rehabilitants taking part in the intervention group reported on average 70 min longer physical activity than their counterparts in the control group (adjusted means intervention group: 299.97, control group: 230.34). In contrast, the cCR rehabilitants participating in either educational pro-gram showed similar physical activity levels (ad-justed means intervention group: 244.46, control group: 250.18). No interaction effects were found for the other secondary outcomes (data not shown).


In this controlled study we evaluated the short, inter-mediate- and long-term effectiveness of a patient-oriented educational program compared with a lecture-based program (usual care) for patients with coronary heart disease receiving inpatient med-ical rehabilitation. We found a significant, but small treatment effect in patients’ illness knowledge at discharge of rehabilitation as well as after 12 months in favor of the new program. Furthermore, with respect to behavioral outcomes a small treat-ment effect was observed on physical activity after 1 year. There were no significant group differences in these outcomes after 6 months, which might be due to different trajectories of the study groups over time. However, there were no additional effects of the new program on healthy diet or medication ad-herence. Thus, our primary hypothesis and second-ary hypothesis regarding behavioral outcomes were partially confirmed.

We compared the new patient-oriented program with an active control condition that generally dif-fered only with regard to some contents but varied in methods of delivery (e.g. strategies for behavior change as action planning and self-monitoring stra-tegies and interactive didactics) and that comprised almost the same amount of time. Therefore, small effects were expected, as opposed to the larger effects that could have been expected if an active treatment would have been compared with a no-treatment condition. In line with this, all effect sizes were small at most. They were, however, in the typical range for self-management interventions


or other interventions designed to change health-related behaviors [e.g. 26–29]. Moreover, the inter-vention comprised only a small amount of treatment within additional multidisciplinary rehabilitation treatments that all patients received, including med-ical, physmed-ical, behavioral and vocational rehabilita-tion components. Thus, significant changes occurred in both groups and the incremented effect of the additional intervention may have been diluted.

Results on behavioral outcomes showed long-term effects on physical activity but not on healthy diet. Similar results have been shown for several interventions with focus on volitional strategies like action planning and self-monitoring aimed to facilitate physical activity initiation and mainten-ance after discharge from rehabilitation [30–35]. Also, implementation intention training had proven effective for saturated fat intake among pa-tients after myocardial infarction [36]. However, those interventions were only targeting a single be-havior. On the other hand, a small-group interactive

education targeting multiple behaviors—sports activities, healthy diet and relaxation—showed only effects on diet in long term [15]. Results of meta-analyses on interventions to change physical activity among chronically ill adults or cardiac pa-tients suggest that interventions that target only one behavior were more successful than those attempt-ing to change multiple behaviors [26,27]. Findings from studies targeting both physical activity and diet suggest that dietary behaviors may be more amen-able to change than physical activity behaviors and for combined interventions, change of a single be-havior is more frequent [37]. Furthermore, post-intervention effects 1 year after post-intervention were seldom reported [27, 37]. Hence, the changes found for physical activity in our trial may be judged as a success. Besides, the patient-oriented program was consistent with a ‘global health/behav-ioral category approach’ [38], promoting a healthy lifestyle for coronary patients as a route to multiple behavioral changes and putting the focus on the Fig. 2. Total physical activity (mean) at 12-months follow-up by treatment groups and type of rehabilitation [cardiac rehabilitation within 14 days after an aCR and cardiac rehabilitation during the chronic course of disease without recent acute index event (cCR)]. Analysis of covariance: main effect treatment group:P¼0.13,2¼0.007; main effect type of rehabilitation:P¼0.40,2¼0.002; interaction effect treatment grouptype of rehabilitation:P¼0.08,2


issue of choice. Thus, each participant had to decide which strategy of self-management behaviors suited him best and the specific behaviors addressed were heterogeneous between participants. Moreover, the dose of intervention components targeting mainten-ance of behavior change may not have been powerful enough to generate changes in multiple behaviors. Follow-up prompts or aftercare programs may be required to sustain results [34]. Overall, fur-ther research on the mechanisms of multiple behav-ior change is needed [38,39].

Neither study program resulted in differences in medication adherence as measured by the MARS-D. However, the measure produced ceiling effects [22] and therefore may not be sensitive to change. But it remains unclear whether high scores were due to high adherence or over-reporting. Though we found treatment effects for medication beliefs in the short- and long-term periods ([17], results not shown), which were moderately related to medica-tion adherence [40].

Treatment effect on physical activity was moder-ated by type of rehabilitation. In particular, the pro-gram worked better for rehabilitants immediate after an acute index event. Meta-analysis of psycho-logical treatment of cardiac patients indicate that effects on mortality were only strong in studies that initiated treatment late (at least 2 months after an event), whereas those that started soon after the event showed little benefit [6]. There were no studies for behavior change which highlight this issue. Our study results pose the question whether patients at different illness stages might benefit from different modes of interventions. It is possible that patients after an acute event with less than ideal health be-haviors at the beginning of rehabilitation have a greater need for specific strategies conveyed to them in the new program such as action planning and may apply these strategies in a more efficient manner. On the other hand, patients with late or re-current rehabilitation and poorer mental health may need more problem-focused programs to deal with special problems of long-term behavior change and coping with distress. However, systematic research into this issue is mostly lacking and future studies should focus on this aspect as well.

Several limitations need to be considered. First, randomization was not possible because of insuffi-cient training rooms, treatment time slots and insuf-ficient staff resources for parallel implementation of both groups in the clinics. But patient samples were comparable with regard to socio-demographic and medical data. Furthermore, baseline scores of out-come parameters were adjusted in the analyses throughout. The proportion of drop-outs was similar for both groups. Factors associated with drop-out were consistent with previous research. Second, the sample had a high proportion of men, as in one rehabilitation hospital only men were recruited (94% male in contrast to commonly 83% within cardiac rehabilitation; [41]). Mean age of 53 years and high proportion of employees is representative of rehabilitants by the German Statutory Pension Insurance [41] but limits generalization to other populations (e.g. older populations with more co-morbidity and populations with a higher proportion of women). Due to the sample distribution, no gender-sensitive analysis could be conducted [42]. Third, the primary outcome illness knowledge was assessed with a new study-specific questionnaire as there was no validated knowledge questionnaire for use with cardiac patients available in German lan-guage. The questionnaire may not accurately reflect comprehensive heart disease knowledge in other populations, however. Fourth, behavioral outcomes were measured by self-report, albeit with validated instruments. Objective measures would be prefer-able but would have been difficult to collect in this context. However, self-reports have been shown to be reliable and valid when tested against objective measures [e.g. 43–45]. Finally, the sample size cal-culation was powered to detect small-to-medium ef-fects, as could be assumed for the primary outcome. Power to detect even smaller effects in the second-ary outcomes may have been too low. Therefore, results have been reported based on effect sizes.


To the best of our knowledge, this is the first edu-cational program evaluated among coronary heart


disease patients in Germany. The controlled study demonstrated that a patient-oriented program may be more effective in certain outcomes than a lecture-based program in both short- and long-term periods. Therefore, the program should be considered for dissemination within cardiac rehabilitation.


The German Statutory Pension Insurance Scheme [grant number 8011 – 106 – 31/31.93].

Conflict of interest statement

None declared.


1. Balady GJ, Williams MA, Ades PAet al. Core components of cardiac rehabilitation/secondary prevention programs: 2007 update.Circulation2007;115: 2675–82.

2. Bjarnason-Wehrens B, Held K, Hoberg Eet al. Deutsche Leitlinie zur Rehabilitation von Patienten mit Herz-Kreislauferkrankungen (DLL-KardReha) (German guideline for rehabilitation of patients with heart diseases).Clin Res Cardiol Suppl2007;2: III/1–III54.

3. Deutsche Rentenversicherung Bund, Hrsg. Reha-Therapiestandards Koronare Herzkrankheit – Leitlinie fu¨r die medizinische Rehabilitation der Rentenversicherung [Evidence-based Standards for Rehabilitation of Coronary Heart Disease]. Berlin: DRV Bund, 2009.

4. Brown JPR, Clark AM, Dalal Het al. Patient education in the management of coronary heart disease.Cochrane Database of Syst Rev2011;12: CD008895.

5. Clark AM, Hartling L, Vandermeer Bet al. Meta-analysis: secondary prevention programs for patients with coronary artery disease.Ann Intern Med2005;143: 659–72. 6. Linden W, Phillips MJ, Leclerc J. Psychological treatment of

cardiac patients: a meta-analysis. Eur Heart J2007;28: 2972–84.

7. Whalley B, Rees K, Davies Pet al. Psychological interven-tions for coronary heart disease.Cochrane Database of Syst Rev2011;8: CD002902.

8. Karoff M, Held K, Bjarnason-Wehrens B. Cardiac rehabili-tation in Germany.Eur J Cardiovasc Prev Rehabil2007;14: 18–27.

9. Faller H, Reusch A, Stro¨bl Vet al. Patientenschulung als Element der Patientenorientierung in der Rehabilitation [Patient education as a constituent of a patient-oriented ap-proach in rehabilitation].Rehabilitation2008;47: 77–83. 10. Reusch A, Schug M, Ku¨ffner Ret al. Gruppenprogramme

der Gesundheitsbildung, Patientenschulung und Psychoedu-kation in der medizinischen Rehabilitation 2010 - Eine

Bestandsaufnahme [Medical rehabilitation group-pro-grammes concerning health promotion, patient education and psychoeducation – a 2010 national survey]. Rehabilita-tion2013;52: 226–33.

11. Michie S, Abraham C, Whittington Cet al. Effective tech-niques in healthy eating and physical activity interventions: A meta-regression.Health Psychol2009;28: 690–701. 12. Michie S, Ashford S, Sniehotta FFet al. A refined taxomomy

of behaviour change techniques to help people change their physical activity and health eating behaviours: The CALO-RE taxonomy.Psychol Health2011;26: 1479–98. 13. Williams SL, French DP. What are the most effective

inter-vention techniques for changing physical activity self-efficacy and physical activity behaviour – and are they the same?Health Educ Res2011;2: 308–22.

14. Faller H, Koch GF, Reusch Aet al. Effectiveness of educa-tion for gastric cancer patients. A controlled prospective trial comparing interactive vs. lecture-based programs.Patient Educ Couns2009;76: 91–8.

15. Reusch A, Stro¨bl V, Ellgring H et al. Effectiveness of small-group interactive education vs. lecture-based information-only programs on motivation to change and lifestyle behaviours. A prospective controlled trial of rehabilitation inpatients. Patient Educ Couns 2011; 82: 186–92.

16. de Vries U, Muhlig S, Waldmann HCet al. Patient satisfac-tion with different asthma-training variants.Patient Educ Couns2008;70: 266–75.

17. Seekatz B, Haug G, Mosler Get al. Entwicklung und kurz-fristige Effektivita¨t eines standardisierten Schulungspro-gramms fu¨r die Rehabilitation bei Koronarer Herzkrankheit [Development and short-term effects of a standardized pa-tient education program for in-papa-tient cardiologic rehabilita-tion].Rehabilitation2013;52: 344–51.

18. Schwarzer R. Modeling health behavior change: How to predict and modify the adoption and maintenance of health behaviors.Appl Psychol2008;57: 1–29.

19. Leventhal H, Meyer D, Nerenz D. The common sense rep-resentation of illness danger. In: Rachman S (ed).Medical Psychology. New York: Pergamon, 1980, 7–30.

20. Godin G, Shephard RJ. A simple method to assess exercise behavior in the community.Can J Appl Sport Sci1985;10: 141–6.

21. Keller S.Zur Validita¨t des Transtheoretischen Modells – Eine Untersuchung zur Vera¨nderung des Erna¨hrungsverhal-tens [On the validity of the Transtheoretical Model of Change – An empirical study on dietary behavior]. Dissertation at the Philipps-Universita¨t Marburg, 1998. Available at: 0303/html/frame.htm. Accessed 15 May 2012.

22. Mahler C, Hermann K, Horne Ret al. Assessing reported adherence to pharmacological treatment recommendations. Translation and evaluation of the Medication Adherence Report Scale (MARS) in Germany. J Eval Clin Pract 2010;16: 574–9.

23. Vickers AJ. Analysis of variance is easily misapplied in the analysis of randomized trials: a critique and discussion of alternative statistical approaches. Psychosom Med 2005;

67: 652–5.

24. Cohen J. Statistical Power Analysis for the Behavioral Science. Hillsdale: Lawrence Erlbaum Associates, 1988.


25. Bullinger M, Kirchberger I. SF-36 Fragebogen zum Gesundheitszustand [Short-form-36 Health Survey]. Go¨ttingen: Hogrefe, 1998.

26. Conn VS, Hafdahl AR, Brown SAet al. Meta-analysis of patient education interventions to increase physical activity among chronically ill adults.Patient Educ Couns2008;70: 157–72.

27. Conn VS, Hafdahl AR, Moore SMet al. Meta-analysis of interventions to increase physical activity among cardiac subjects.Int J Cardiol2009;133: 307–20.

28. Johnson BT, Scott-Sheldon LAJ, Carey MP. Meta-Synthesis of health behavior change meta-analyses. Am J Public Health2010;100: 2193–8.

29. Warsi A, Wang PS, LaValley MPet al. Self-management education programs in chronic disease: a systematic review and methodological critique of the literature.Arch Intern Med2004;164: 1641–9.

30. Fleig L, Lippke S, Pomp Set al. Intervention effects of exercise self-regulation on physical exercise and eating fruits and vegetables: a longitudinal study in orthopedic and cardiac rehabilitation.Prev Med2011;53: 182–7. 31. Fuchs R, Go¨hner W, Seelig H. Long-term effects of a

psy-chological group intervention on physical exercise and health: The MoVo concept.J Phys Act Health2011;8: 794–803.

32. Meng K, Seekatz B, Roßband Het al. Intermediate and long-term effects of a standardized back school for inpatient orthopedic rehabilitation on illness knowledge and self-management behaviors: A randomized controlled trial.Clin J Pain2011;27: 248–57.

33. Sniehotta FF, Scholz U, Schwarzer Ret al. Long-term effects of two psychological interventions on physical exercise and self-regulation following coronary rehabilitation.Int J Behav Med2005;12: 244–55.

34. Stro¨bl V, Knisel W, Landgraf Uet al. A combined planning and telephone aftercare intervention for obese patients: Effects on physical activity and body weight after one year.J Rehabil Med2013;45: 198–205.

35. Sudeck G, Ho¨ner O. Volitional interventions within cardiac exercise therapy (VIN-CET): Long-term effects on physical activity and health-related quality of life. Appl Psychol: Health Well-being2011;3: 151–71.

36. Luszczynska A, Scholz U, Sutton S. Planning to change diet: A controlled trial of an implementation intentions training intervention to reduce saturated fat intake among patients after myocardial infarction. J Psychosom Res 2007; 63: 491–7.

37. Fjeldsoe B, Neuhaus M, Winkler Eet al. Systematic review of maintenance of behavior change following physical activ-ity and dietary interventions. Health Psychol 2011; 30: 99–109.

38. Noar SM, Chabot M, Zimmermann RS. Applying health behavior theory to multiple behavior change: considerations and approaches.Prev Med2008;46: 275–80.

39. Prochaska JO. Multiple health behavior research represents the future of preventive medicine. Prev Med 2008; 46: 281–5.

40. Byrne M, Walsh J, Murphy AW. Secondary prevention of coronary heart disease: patient beliefs and health-related behaviour.J Psychosom Res2005;58: 403–15.

41. Deutsche Rentenversicherung Bund (Eds). Reha-Bericht 2010 (Reha-report 2010). Berlin: DRV Bund, 2010. 42. Bjarnason-Wehrens B, Grande G, Loewel Het al.

Gender-specific issues in cardiac rehabilitation: do women with is-chaemic heart disease need specially tailored programmes? Eur J Cardiovasc Prev Rehabil2007;14: 163–71. 43. Aadahl AM, Kjaer M, Kristensen JHet al. Self-reported

physical activity compared with maximal oxygen uptake in adults.Eur J Cardiovasc Prev Rehabil2007;14: 422–8. 44. Bowles HR, Fitzgerald SJ, Morrow JRet al. Construct

val-idity of self-reported historical physical activity. Am J Epidemiol2004;160: 279–86.

45. Garber MC, Nau DP, Erickson SRet al. The concordance of self-report with other measures of medication adherence: a summary of the literature.Med Care2004;42: 649–52.



Related subjects :