Patient-reported outcomes predict return to work and health-related quality of life six months after cardiac rehabilitation: Results from a German multi-centre registry (OutCaRe)

Annett Salzwedel; Iryna Koran; Eike Langheim; Axel Schlitt; Jörg Nothroff; Christa Bongarth; Markus Wrenger; Susanne Sehner; Rona Reibis; Karl Wegscheider; Heinz Völler; for the OutCaRe investigators

doi:10.1371/journal.pone.0232752

. 2020 May 5;15(5):e0232752. doi: 10.1371/journal.pone.0232752

Patient-reported outcomes predict return to work and health-related quality of life six months after cardiac rehabilitation: Results from a German multi-centre registry (OutCaRe)

Annett Salzwedel ^1,^*, Iryna Koran ², Eike Langheim ³, Axel Schlitt ⁴, Jörg Nothroff ⁵, Christa Bongarth ⁶, Markus Wrenger ⁷, Susanne Sehner ⁸, Rona Reibis ⁹, Karl Wegscheider ⁸, Heinz Völler ^1,²; for the OutCaRe investigators^¶

Editor: Stephanie Prince Ware¹⁰

PMCID: PMC7199966 PMID: 32369514

Abstract

Background

Multi-component cardiac rehabilitation (CR) is performed to achieve an improved prognosis, superior health-related quality of life (HRQL) and occupational resumption through the management of cardiovascular risk factors, as well as improvement of physical performance and patients’ subjective health. Out of a multitude of variables gathered at CR admission and discharge, we aimed to identify predictors of returning to work (RTW) and HRQL 6 months after CR.

Design

Prospective observational multi-centre study, enrolment in CR between 05/2017 and 05/2018.

Method

Besides general data (e.g. age, sex, diagnoses), parameters of risk factor management (e.g. smoking, hypertension), physical performance (e.g. maximum exercise capacity, endurance training load, 6-min walking distance) and patient-reported outcome measures (e.g. depression, anxiety, HRQL, subjective well-being, somatic and mental health, pain, lifestyle change motivation, general self-efficacy, pension desire and self-assessment of the occupational prognosis using several questionnaires) were documented at CR admission and discharge. These variables (at both measurement times and as changes during CR) were analysed using multiple linear regression models regarding their predictive value for RTW status and HRQL (SF-12) six months after CR.

Results

Out of 1262 patients (54±7 years, 77% men), 864 patients (69%) returned to work. Predictors of failed RTW were primarily the desire to receive pension (OR = 0.33, 95% CI: 0.22–0.50) and negative self-assessed occupational prognosis (OR = 0.34, 95% CI: 0.24–0.48) at CR discharge, acute coronary syndrome (OR = 0.64, 95% CI: 0.47–0.88) and comorbid heart failure (OR = 0.51, 95% CI: 0.30–0.87). High educational level, stress at work and physical and mental HRQL were associated with successful RTW. HRQL was determined predominantly by patient-reported outcome measures (e.g. pension desire, self-assessed health prognosis, anxiety, physical/mental HRQL/health, stress, well-being and self-efficacy) rather than by clinical parameters or physical performance.

Conclusion

Patient-reported outcome measures predominantly influenced return to work and HRQL in patients with heart disease. Therefore, the multi-component CR approach focussing on psychosocial support is crucial for subjective health prognosis and occupational resumption.

Trial registration

The study was registered at the German Clinical Trial Registry and the International Clinical Trials Registry Platform (ICTRP) of the World Health Organization (DRKS00011418; http://www.drks.de/DRKS00011418, http://apps.who.int/trialsearch/Trial2.aspx?TrialID=DRKS00011418).

Introduction

According to recommendations by the European Association of Preventive Cardiology, cardiac rehabilitation (CR) should include medical examination, exercise training, cardiovascular risk factor management (e.g. lipid and blood pressure monitoring, smoking cessation), lifestyle counselling in terms of physical activity and nutrition, as well as vocational support and psychosocial management. [1] Through this multimodal approach, an improved clinical prognosis, high quality of life and social participation should be achieved. [1]

While the positive effect of CR on the clinical course and health-related quality of life (HRQL) has been sufficiently proven, [2–6] the evidence regarding the impact on occupational outcomes and return to work rates remains inconclusive. [7] However, CR programmes based on a multi-modal approach seem to be at an advantage over exercise-based programmes in terms of both clinical and occupational outcomes. [3,7] Nevertheless, a significant proportion of patients (30% of patients in Germany) [8] of employable age fail to return to work after CR, resulting in a psychosocial burden on the patient with reduced HRQL [9] and high costs to society. To overcome this gap, we need optimised CR programmes adapted to the individual needs of patients. Knowledge of modifiable parameters predicting return to work and HRQL, which reflect the holistic interprofessional therapy options of the multi-modal CR, could provide important clues. In this regard, patient-reported outcome measures must be considered, as they have become increasingly important as a measure of patient-centred care and as an integral part of high-quality healthcare over the last decade. [10,11]

In general, return to work of patients with cardiovascular disease depends on several clinical and contextual factors, such as the patient’s physical performance, depression, anxiety, expectations of professional reintegration and perceived health, with psychosocial parameters appearing more influential than classical cardiovascular risk factors. [12–16] Even for HRQL in patients living with heart disease, influencing factors such as exercise, coronary artery bypass surgery, socioeconomic parameters and a poor physician-patient relationship have been investigated. [17–19] Since return to work and HRQL after a cardiac event are closely associated, [15] several common predictors can be assumed.

However, to date there are no systematic and prospective research data on the associations between a multitude of modifiable parameters presenting the most important core components of multi-modal CR and return to work, as well as HRQL over the mid-term course (three months to one year) after CR discharge. In the Outcome of Cardiac Rehabilitation (OutCaRe) study, we aimed to identify predictors of occupational reintegration and HRQL among the same set of patient-reported outcome measures, clinical parameters, cardiovascular risk factors and physical performance.

We hypothesised that modifications of patient-reported outcome measures during CR independent of cardiovascular risk factors and physical performance would predict return to work, as well as HRQL six months after CR.

Methods

Study design

The OutCaRe study was designed to identify and evaluate performance measures of CR. The study was conducted following a stepwise concept. First, a Delphi expert survey of 70 cardiologists, psychologists, sports therapists and physiotherapists extracted potential indicators of rehabilitation success, including parameters of the four key areas of CR: cardiovascular risk factors; physical performance; social medicine and subjective health. [20]

Subsequently, a national multi-centre register study was performed to evaluate the predefined indicators regarding feasibility and modifiability of the CR routine. [21] This paper analyses these parameters in terms of their mid-term predictive value for return to work and HRQL after CR.

Patients and cardiac rehabilitation

In 12 German rehabilitation centres, eligible patients aged up to 65 years were enrolled at admission to CR, regardless of their primary allocation diagnoses (e.g. acute myocardial infarction, coronary artery bypass grafting, coronary artery disease, valvular disease, vascular disease, implantation of active electrical devices (implantable cardioverter defibrillator or cardiac resynchronisation therapy)). Patients with insufficient German language skills, early retirement or missing consent were excluded.

All patients performed a standardised comprehensive CR programme with a regular duration of three to four weeks according to the specifications of the German pension insurance. [21] The programme could be performed in either an inpatient or outpatient setting and consisted of all-day activities including counseling by a cardiologist, risk-factor modification strategies (e.g. patient education on nutritional habits, smoking cessation, physical activity and medication adherence), physician-supervised exercise training and sports therapy (e.g. training on a bicycle ergometer, outdoor walking, resistance training, gymnastics), psychosocial interventions (health education and counseling, psychotherapy, stress management in single or group sessions), and vocational assessment and physician and social worker counseling. [22,23] On average, patients performed 12 training and sports therapy units per week with a duration, depending on the training group and physical performance, of up to 30 minutes and 45 minutes for outdoor walking, respectively, and 8 additional counselling sessions. [24]

Predictors, data capturing

The data capture was realised via a web-based electronic case report form provided by SecuTrial^® (interActive Systems, Berlin).

Upon admission to CR, sociodemographic data (e.g. age, sex, educational level), diagnoses, procedures and relevant comorbidities (diabetes mellitus, musculoskeletal disease, psychological diagnosis) and—as potential confounders of rehabilitation outcomes—perceived occupational stress and incisive life events using the INTERHEART stress scale [25] were recorded. Additionally, modifiable cardiovascular risk factors, parameters of physical performance, social medicine and subjective health operationalised by a multitude of patient-reported performance measures were captured at both CR admission and discharge. The majority of captured data were taken from patients’ records, while social medicine and subjective health were assessed by means of specific patient-reported outcome measures used in the study:

Cardiovascular risk factors: smoking behaviour, systolic/diastolic blood pressure, low density lipoprotein (LDL) cholesterol.
Physical performance: maximum exercise capacity on the bicycle stress test, endurance training load in ergometry, 6-min walking distance.
Social medicine and subjective health in patient-reported outcome measures:
- depression on the patient health questionnaire (PHQ-9), [26]
- heart-focussed anxiety using the cardiac anxiety questionnaire with the aspects fear, avoidance and attention (revised German version: Herzangstfragebogen, HAF-17), [27]
- subjective well-being on the World Health Organization questionnaire (WHO-5), [28]
- quality of life on the short-form health survey (SF-12) with the physical and mental health component summary scales, [29]
- subjective health using indicators of rehabilitation status (IRES-24) with the subscales pain, somatic health and mental health, [30]
- general self-efficacy expectations on the short scale for measuring general self-efficacy beliefs (ASKU; range 1–5 points), [31]
- lifestyle change motivation (single choice question: Can you imagine changing your lifestyle due to your condition? Answer options: definitely; probably; uncertain; probably not; no way),
- self-assessed health prognosis (single choice question: Please estimate what state of health you expect after the next 6 months. Answer options: excellent; very good; good; less good; poor),
- pension desire and self-assessed occupational prognosis using the Würzburger Screening identifying occupational issues and the need for vocational rehabilitation. [32]

A more detailed description of parameter operationalisation is published elsewhere. [21]

Outcome measures

A follow-up survey conducted by mail (paper-pencil procedure) or e-mail with a direct link to the electronic case report platform regarding patients’ preferences 6 months after CR, return to work status and HRQL were assessed as outcome measures by the physical and mental component summary scales in the SF-12.

Ethics approval and study registration

All patients were informed about the contents of the study and provided written informed consent prior to enrolment. The study was approved by the Ethics Committee of the State Medical Association of Brandenburg (approval number S 4(a)/2017) as the responsible institutional review board of the principal investigator (HV, Klinik am See, Rüdersdorf, Germany) and, additionally, by the local institutional review board for each participating CR centre. OutCaRe is registered with the German Register of Clinical Trials and the International Clinical Trial Agency (ICTRP, WHO; registration number DRKS00011418).

Statistics

For description, continuous variables are presented as means ± standard deviation, and categorical variables as absolute values and percentages. Differences in variables between admission to CR, discharge from CR and follow-up were tested for statistical significance using Wilcoxon tests for continuous variables and McNemar tests for categorical variables. For metric variables, the standardised effect size (SES) was additionally calculated (ratio of mean value differences and pretest standard deviation), [33] which was interpreted according to Cohen, who defined an effect size of 0.20 as small, 0.50 as moderate and 0.80 or greater as large. [34] In addition, the changes in patient-reported outcome measures (metric variables) during CR were assessed using the minimal important difference (MID) if available. This concerns the IRES-24 (MID 0.5 points) [35] and the SF-12 (MID two points for the physical and three points for the mental component summary). [36] For the WHO-5, we anticipate a MID of 10 percentage -points. [28] No MID was considered for PHQ-9 because the MID reported in the literature focusses on the change of PHQ values in the acute phase of depression treatment in affected patients, [37] which is of subordinate relevance in our investigation. There is no MID available for the HAF-17.

While the descriptive analyses of baseline and rehabilitation data used the original available data without imputation, an imputation model was developed to allow more complex analyses with many predictors despite missing values. Missing values were imputed using chained equations (MICE) with 20 imputations. All psychological scores were imputed by predictive mean matching with the ten nearest neighbours; for all other variables, a parametric approach depending on the measurement level of the variable was chosen. For all further analyses, only patients with a non-missing 6-month outcome in the original dataset were included. Return to work was analysed regardless of the employment status at baseline, as it is a goal of CR in Germany to support return to work even in patients who were on sick leave or unemployed before CR. Predictors of patients’ return to work and HRQL were identified using multiple linear regression models with the imputed baseline and discharge data. With this full model as the starting model (see supplemental material, S1–S3 Figs), variable selections were performed as described by Wood et al. [38] A backward selection, using the likelihood-ratio test with a significance level for removal of p = 0.05, was executed with a weighted linear regression for the stacked imputed data set. The weights were calculated with respect to the number of imputations and the average fraction of missing data across all variables. The selected models were fitted for each imputed dataset separately and the parameter estimates were combined using Rubin’s Rule. [39] To control the reason for CR, we simplified the variable to ‘acute coronary syndrome’ vs ‘no acute coronary syndrome’. This variable was selected for the final model (the marginal means of RTW by reason for CR are shown in the supplement, S4 Fig). For the complete set of potential regressors, potential multicollinarity was studied by calculating the variance inflation factors, values below 10 were considered acceptable.

Results are presented as forest plots with 95% confidence intervals (95% CI) and p-values. Effects with a p-value of less than 0.05 were considered statistically significant. Calculations were carried out using STATA 15.0 and SPSS Version 25.0.

Results

Patient characteristics and cardiac rehabilitation

Out of a total of 1,586 enrolled patients, 1,262 participants (80%) responded to the follow-up questionnaires and could be analysed (Fig 1). The majority of these patients (mean age 54±7 years, 77% male) had completed at least 10 years of education (78%), lived in a family or with a partner (79%) and were employed before CR (90%), though 72% were on sick leave. The patients were assigned to CR primarily due to an acute coronary syndrome (40%) or coronary artery disease without myocardial infarction (18%), followed by heart valve diseases in 12% of patients and coronary artery bypass grafting (8%). CR was most often conducted after an acute event (83%) and in an inpatient setting (92% of patients). Patients stayed an average of 23.7±4.5 days in CR and most were discharged normally. Eighteen patients (1.4%) were discharged prematurely at their own request, and nine (0.7%) were transferred to acute hospital care. Patient characteristics, diagnoses, risk factors and comorbidities are presented in Table 1.

Table 1. Baseline characteristics (n = 1,262).

Characteristics (1)	mean±SD or n (%)
Sociodemographic data
Age (years)	54.2±7.0
Sex (male)	968 (76.7)
Education n = 1,249
<10^th grade	209 (16.8)
Secondary school	691 (55.3)
College/university	288 (23.1)
Living situation n = 1,248
Family/partner	987 (79.1)
Living alone	218 (17.5)
Other	43 (3.4)
Occupation (employed) n = 1,248	1127 (90.3)
Sick leave before CR n = 1,248	892 (71.5)
Cardiac rehabilitation
Admission to CR
After an acute event	1044 (82.7)
For a chronic disorder	218 (17.3)
Setting of CR
Inpatient	1158 (91.8)
Outpatient/day care	104 (8.2)
Indication for referral to CR
ACS	500 (39.6)
Stable CAD	229 (18.1)
Heart valve disease	153 (12.1)
CABG	96 (7.6)
Venous disease	49 (3.9)
Cardiac arrhythmia	49 (3.9)
Aortic diseases	47 (3.7)
Arterial hypertension	46 (3.6)
Chronic heart failure	36 (2.9)
Atherosclerosis (incl. PAD)	24 (1.9)
Intervention (PCI, ICD/CRT)	19 (1.5)
Myo-/endo-/pericarditis	7 (0.6)
Other	7 (0.6)
Comorbidities/ risk factors	3.4±1.6^*
Arterial hypertension	840 (66.6)
Hyperlipidaemia	772 (61.2)
Diabetes mellitus	206 (16.3)
Atrial fibrillation	111 (8.8)
PAD	64 (5.1)
Depression	55 (4.4)
COPD	52 (4.1)
Kidney disease	51 (4.0)

Open in a new tab

*number of comorbidities; ACS, acute coronary syndrome; CABG, coronary artery bypass graft; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CR, cardiac rehabilitation; CRT, cardiac resynchronisation therapy; ICD, implantable cardioverter defibrillator; PAD, peripheral artery disease; PCI, percutaneous coronary intervention; SD, standard deviation

During the CR program, several cardiovascular risk factors, physical performance and patient-reported outcome measures were improved. The proportion of current smokers was significantly reduced from 35% to 15%. The mean endurance training load was increased by 21 watts. The percentage of lifestyle change-motivated patients and those who reported excellent or very good self-assessed health prognoses was improved (p<0.001 for all). All scores on subjective health questionnaires (e.g. PHQ-9, HAF-17, WHO-5, SF-12 and IRES-24) were significantly improved with SES to a mild to moderate extent. The changes in the physical and mental scales of the SF-12 (+5.7/+6.1 points) and IRES-24 (physical health +1.2, mental health +1.4, pain 1.0) and in the WHO-5 (+18.3%-points) exceeded the MID. However, the proportion of patients who rated their occupational prognosis negatively increased significantly from 41% to 45% (Table 2).

Table 2. Functional parameters and risk factors and patient-reported outcome measures at baseline and discharge from cardiac rehabilitation (N = 1,262; mean ± standard deviation; n and percentage).

Parameter	CR admission	CR discharge	SES	P-value
Cardiovascular risk factors
Smoking behaviour^ǂ (smoker) n = 1,234	427 (34.6)	187 (15.2)	--	<0.001
Systolic blood pressure (mmHg) n = 1,258	128.8±18.5	121.6±13.8	0.39	<0.001
Diastolic blood pressure (mmHg) n = 1,258	80.3±11.4	75.2±9.1	0.44	<0.001
LDL-Cholesterol (mmol/l) n = 981	4.6±2.5	3.8±2.2	0.32	<0.001
Physical performance
Maximum exercise capacity (Watt) n = 790	111.4±37.6	130.9±41.5	0.52	<0.001
Endurance training load (Watt) n = 1,204	48.6±20.6	69.7±26.3	1.03	<0.001
6-min walking distance (m) n = 812	451.8±90.6	526.6±90.8	0.82	<0.001
Patient-reported outcome measures
Depression (PHQ-9) n = 1,149	6.4±4.8	4.4±4.0	0.43	<0.001
Heart-focussed anxiety (HAF-17) n = 1,102	1.5±0.6	1.3±0.6	0.32	<0.001
HAF-17 heart-focussed fear n = 1,146	1.6±0.7	1.4±0.7	0.27	<0.001
HAF-17 heart-focussed avoidance n = 1,163	1.5±1.0	1.0±0.8	0.44	<0.001
HAF-17 heart-focussed attention n = 1,158	1.3±0.7	1.3±0.6	0.02	0.308
Quality of life/subjective well-being
WHO-5 n = 1,180	51.2±25.6	69.5±21.1	0.71	<0.001
SF-12 PCS n = 1,072	38.9±10.6	44.6±9.5	0.53	<0.001
SF-12 MCS n = 1,072	48.2±11.9	54.3±8.9	0.51	<0.001
IRES-24 physical health n = 1,173	5.8±2.7	7.0±2.4	0.43	<0.001
IRES-24 mental health n = 1,190	6.4±2.5	7.8±2.1	0.58	<0.001
IRES-24 pain n = 1,190	6.3±2.6	7.3±2.4	0.40	<0.001
General self-efficacy expectations (ASKU)	4.1±0.7	4.1±0.7	0.10	<0.001
Lifestyle change motivation (certain/fairly certain) n = 1,187	939 (79.1)	1037 (87.4)	--	<0.001
Self-assessed health prognosis (excellent/very good) n = 1,187	509 (42.9)	605 (51.0)	--	<0.001
Pension desire (yes) n = 1,170	205 (17.5)	182 (15.6)	--	0.028
Self-assessed occupational prognosis (negative) n = 1,137	463 (40.7)	507 (44.6)	--	<0.001

Open in a new tab

^‡Patients who quit smoking due to the acute event before subsequent cardiac rehabilitation were classified as smokers upon admission to rehabilitation.

ASKU, Allgemeine Selbstwirksamkeit Kurzskala (short scale for measuring general self-efficacy beliefs); CR, cardiac rehabilitation; HAF-17, Herzangstfragebogen (German version of the Cardiac Anxiety Questionnaire); IRES-24, indicators of rehabilitation status; LDL, low density lipoprotein; PHQ-9, Patient Health Questionnaire; SES, standardised effect size according Cohen’s d; SF-12, Short-Form health survey with PCS, physical component summary and MCS, mental component summary; WHO-5, World Health Organization well-being index

Follow-up data

In the follow-up survey, 864 patients (68.5%) returned to work and 67 patients (5.3%) reported having retired, while 79 (6.3%) had applied for pension. Eighty nine patients (7.1%) were unemployed six months after CR and 190 (15.1%) were still on sick leave. Furthermore, 194 (15.4%) patients had experienced readmission to acute hospital care for any cardiac event.

Data for HRQL were available for 1,181 patients (94%). The mean score was 45.5±10 points on the physical component summary scale and 48.9±11 points on the mental component summary scale. Scores for HRQL 6 months after CR were significantly correlated to return to work (age and sex adjusted OR = 1.08/1.06 per point on the physical/mental component summary scale; p<0.001). In general, patients reported a moderately or substantially improved health status due to CR in 935 (74.7%) cases and an enhanced occupational or physical capacity in 876 (69.9%) cases.

Predictors of occupational resumption and quality of life after cardiac rehabilitation

Acute coronary syndrome as an indication diagnosis and comorbid heart failure upon CR admission were negatively associated with return to work, while a higher endurance training load, HRQL, a high educational level and work stress detected at CR admission had a positive impact on return to work. Patients’ pension desire and self-assessed negative occupational prognoses at discharge from CR reduced the probability of returning to work six months after discharge from CR by 67% and 66%, respectively. Additionally, an improvement in mental HRQL by 10 points between CR admission and discharge resulted in an improved probability of returning to work by 30%, whereas an enhancement of heart-focussed anxiety during CR reduced the chances of returning to work by 29% per point difference on the HAF-17. However, the latter association achieved no statistical significance (Fig 2).

Fig 2 — The forest plot shows the final model after backward selection. The following variables were taken into account in the starting model (see S3 Fig): at admission to rehabilitation: sex, smoking, lifestyle change motivation, pension desire, self-assessed occupational prognosis, self-assessed health prognosis, living situation, educational level, rehabilitation referral for chronic disorder, comorbidities (diabetes mellitus, depression, peripheral artery disease, heart failure, diseases of the back and spine, stress at work, stress from major life events, age, systolic/diastolic blood pressure, body mass index, endurance training load, depression (PHQ-9), heart-focussed anxiety (HAF-17), well-being (WHO-5), physical/mental component summary on the SF-12, physical/mental health and pain in the IRES-24, self-efficacy (ASKU); at discharge from rehabilitation: smoking, lifestyle change motivation, pension desire, self-assessed occupational prognosis, self-assessed health prognosis; changes during rehabilitation: systolic/diastolic blood pressure, body mass index, endurance training load, depression (PHQ-9), heart-focussed anxiety (HAF-17), well-being (WHO-5), physical/mental component summary on the SF-12, physical/mental health and pain on the IRES-24, self-efficacy (ASKU). T0 –baseline measurement at CR admission, T1 –CR discharge, T2 –follow-up six months after CR discharge. CI, confidence interval; HAF-17, Herzangstfragebogen (German version of the Cardiac Anxiety Questionnaire); OR, odds ratio; pt(s), point(s); RTW, return to work; SF-12, Short-Form health survey with PCS, physical component summary and MCS, mental component summary.

Patients’ HRQL after six months was primarily predicted by patient-reported outcome measures, which were mostly assessed at admission to CR. In addition, modifications during CR in endurance training load, heart-focussed anxiety, physical quality of life and health were predictive of physical component summary scores, while changes in anxiety, well-being, mental health and general self-efficacy expectations were significant for the mental component summary score at follow-up. For example, an improvement in mental health (IRES-24) of two points during CR resulted in an enhanced MCS of 2.54 points at follow-up (Fig 3).

Fig 3 — The forest plots show the final models after backward selection. The following variables were taken into account in the starting models (see S1 and S2 Figs): at admission to rehabilitation: sex, smoking, lifestyle change motivation, pension desire, self-assessed occupational prognosis, self-assessed health prognosis, living situation, educational level, rehabilitation referral for chronic disorder, comorbidities (diabetes mellitus, depression, peripheral artery disease, heart failure, diseases of the back and spine, stress at work, stress from major life events, age, systolic/diastolic blood pressure, body mass index, endurance training load, depression (PHQ-9), heart-focussed anxiety (HAF-17), well-being (WHO-5), physical/mental component summary on the SF-12, physical/mental health and pain on the IRES-24, self-efficacy (ASKU); at discharge from rehabilitation: smoking, lifestyle change motivation, pension desire, self-assessed occupational prognosis, self-assessed health prognosis; changes during rehabilitation: systolic/diastolic blood pressure, body mass index, endurance training load, depression (PHQ-9), heart-focussed anxiety (HAF-17), well-being (WHO-5), physical/mental component summary on the SF-12, physical/mental health and pain on the IRES-24, self-efficacy (ASKU). T0 –baseline measurement at CR admission, T1 –CR discharge, T2 –follow-up 6 months after CR discharge. ACS, acute coronary syndrome; ASKU, Allgemeine Selbstwirksamkeit Kurzskala (short scale for measuring general self-efficacy beliefs); CI, confidence interval; HAF-17, Herzangstfragebogen (German version of the Cardiac Anxiety Questionnaire); IRES-24, indicators of rehabilitation status; OR, odds ratio; PHQ-9, Patient Health Questionnaire; pt(s), point(s); RTW, return to work; SF-12, Short-Form health survey with PCS, physical component summary and MCS, mental component summary; WHO-5, World Health Organization well-being index.

Discussion

This investigation found that 69% of patients successfully returned to work 6 months after CR. HRQL was significantly improved by seven percent (physical component) and one percent (mental component) compared to CR admission. The OutCaRe study demonstrated that occupational reintegration, as well as HRQL are predominantly determined by patient-reported outcome measures, more than by clinical parameters, cardiovascular risk factors or physical performance. Specifically, pension desire, negative patient expectations of occupational resumption, acute coronary syndrome, comorbidity (i.e. heart failure) and heart-focussed anxiety represented barriers to occupational reintegration, while an enhanced endurance training load and quality of life on the SF-12, high educational level and work stress facilitated return to work. HRQL in its physical dimension was primarily influenced by physical health and quality of life, heart-focussed anxiety at CR admission and changes during CR, as well as self-assessed health prognosis and pension desire at CR discharge. Psychoemotional parameters, for instance the mental health scale of the IRES-24, perceived stress, comorbid depression, general self-efficacy and well-being assessed by the WHO-5, predicted the mental component of HRQL on the SF-12 as might be expected. Heart-focussed anxiety, pension desire and mental and physical HRQL at different measurement times in CR were the only predicting parameters for both return to work and HRQL after CR.

Heart-focussed anxiety as measured by the HAF-17 comprises the specific fear of recurrent cardiac events, heightened attention to cardiac-related stimuli and symptoms, as well as the aspect of avoidance. [27] These conditions may lead to a reduced adherence to exercise and medication intake and consequently to increased cardiac events.[40] Van Beek et al. found the prognostic impact particularly driven by avoidance behaviour and identified the special need to address cardiac anxiety in CR programs.[40] In fact, the sum score of heart-focussed anxiety and especially the avoidance subscale showed significant improvements to a small extent in the investigated population during the three- to four-week CR programme, which is in line with other recent studies.[41,42]

‘Pension desire’ was assessed in our study by a single question on the intention to apply for retirement without any causal differentiation. Thus, this parameter may implicate aspects of non-captured occupation-related factors, such as non-motivation to return to work, occupational self-efficacy, job satisfaction, job control or expectations of retirement, as they are identified to predict occupational resumption in other publications.[13,43] However, patients with pension desire and negative self-assessed occupational prognoses in CR suffer from a high burden of predominantly social and mental disease consequences. On the other hand, applied coping strategies and support services are mainly focussed on physical aspects of the cardiovascular disease. [44] Especially the essential adaptation of everyday life including the work environment after a cardiac event and subsequent CR could be overtaxing, triggering the desire for early retirement.

In this context, social workers in CR play a crucial role as they advise patients, for example on retraining or gradual reintegration as needed. The support by social workers in CR programmes is mandatory according to the specifications of the German pension insurance. Nevertheless, in accordance with previous studies and official data, return to work rates after a cardiac event were only 69% in our study. [8,13,45] An original randomised clinical trial evaluating an extended social therapy counseling and training program during CR still had no effect on return to work and HRQL in patients at risk of occupational reintegration. [16] This finding suggests that the regular therapy density in the standardised German intensive short-term CR program as described in the methods section is at maximum volume. Continuous support by a social worker in the further course after a cardiac event could remedy this.

The impact of pension desire on HRQL on both the physical and mental scales after CR in our study is less intuitive. It can be interpreted in the context of patients’ expected (likely occurring) difficulties in everyday life and working environment, whereas a mediating role of pension desire for return to work and HRQL is conceivable. Van Cauter et al. accordingly evidenced the significant association between return to work and an enhanced HRQL in the EUROASPIRE IV study. [15]

Also, the counterintuitive effects of perceived stress on occupational reintegration are remarkable. Thus, perceived stress at work facilitated patients’ occupational reintegration. Our findings go along with an Iranian study of 248 patients after coronary bypass surgery, where the probability of returning early to work was 2.3 times higher in patients with job stress. [46] We assume that patients who report a higher perception of work stress may hold a leading position with a higher degree of personal identification and recognition, resulting in a greater motivation to return to work. This interpretation is quite speculative, but may be supported by the fact that the proportion of patients who returned to work was significantly higher by 14 percentage points in highly educated versus less-educated patients (78% vs. 64%). However, perceived stress at work was not associated with the educational level and, in addition, predicted a diminished mental quality of life in the present investigation.

Furthermore, attention should be paid to changes during CR in the parameters independently affecting the analysed mid-term outcome of CR. Besides heart-focused anxiety as mentioned above, the means that mental and physical quality of life as well as subjective health (IRES-24) and well-being on the WHO-5 were significantly improved, with small to moderate effect sizes. The notified changes are considered clinically relevant since they substantially exceed the MID by 1.8 (WHO-5) to 2.9 times (physical component of SF-12).

For the modification of this multiplicity of influencing parameters, in particular patient-reported outcome measures, CR should be composed as a multi-modal programme based on the holistic approach of the biopsychosocial model of the International Classification of Functioning, Disability and Health by the World Health Organization (WHO) [47] and in accordance with current national and international guidelines. [1,48,49] In particular, close cooperation in an interprofessional CR team including cardiologist, social worker and sports therapist or physiotherapist is needed. The offered CR programme should be individualised in the sense of patient-shared decision making to achieve the best possible outcomes.

Limitations

This work is limited in terms of various methodological aspects. First, our recruitment quota was 32%, which can reduce the generalisability of our results to the general CR population. However, our participants were similar to our target population (patients in CR aged up to 65 years) according to statistics on the rehabilitation services of the German pension insurance with regard to age and sex. [50] Second, there is a selection bias due to the incomplete follow up. Non-responding study participants were younger, less educated, more likely to smoke and less frequently employed. At discharge from CR, they reported a lower health status. These differences were taken into account by the imputation model. However, there may be reasons for the absence that were not covered by our data. This has to be kept in mind when interpreting the study results. Furthermore, we investigated a heterogeneous population including patients of all diagnoses in CR. The main objective of the OutCaRe project was to identify and evaluate performance measures for the entire spectrum of CR as generally offered in Germany. Therefore, specific statements pertaining to individual indication groups cannot be derived from the current study. In addition, the feasibility of a parameter as a performance measure was assessed in terms of data availability. [21] Patients were allowed to refrain from answering parts of the questionnaires or individual questions. Missing values were imputed in the statistical models to balance the data. However, it may be that these modeling approaches do not compensate for the selection effects. Finally, the SF-12 used to operationalise HRQL is a generic instrument. Probably, a disease-specific screening tool (eg. HeartQoL questionnaire) [51] could achieve a higher responsiveness to change and acceptance in the investigated heart-disease population. Nevertheless, the SF-12 is well established and commonly used, which ensures the comparability of our results with other studies and populations.

Multicollinearity may limit the interpretability of model coefficients. However, the variance inflation factors of the studied predictors were distinctly below the generally accepted limit of 10 with one exception: mental health at admission had a variance inflation factor of 10.48. Altogether, the multivariate models were sufficiently stable to allow clinical interpretations.

Conclusion

Return to work and physical, as well as mental HRQL half a year after CR were predominantly determined by patient-reported outcome measures, whereas patients’ pension desire and heart-focussed anxiety had a dominant impact on all investigated endpoints. Changes in patient-reported outcome measures during CR affected the occupational and health-related prognosis, underscoring the importance of the multi-component approach in CR. Therefore, questionnaires assessing subjective health should be applied consequently upon commencing rehabilitation with findings subsequently affecting the individual therapy content and rehabilitation goals.

Supporting information

S1 Fig. Full model-multiple imputation(n = 1181).

SF—12, MCS—T2.

(TIF)

Click here for additional data file.^{(894.5KB, tif)}

S2 Fig. Full model-multiple imputation(n = 1181).

SF—12, PCS—T2.

(TIF)

Click here for additional data file.^{(883.9KB, tif)}

S3 Fig. Full model-multiple imputation(n = 1262).

Return to work(RTW).

(TIF)

Click here for additional data file.^{(894.5KB, tif)}

S4 Fig. Full model-multiple imputation(n = 1262).

No significant different probabilities for subdiagnoses (p = 0.127).

(TIF)

Click here for additional data file.^{(109KB, tif)}

Acknowledgments

We thank all participating patients, the frontline staff of the cooperating CR centres and the OutCaRe investigators: Heinz Völler, Klinik am See, Rüdersdorf (principal investigator; e-mail: heinz.voeller@klinikamsee.com); Johannes Glatz and Eike Langheim, Reha-Zentrum Seehof, Teltow; Axel Schlitt, Paracelsus-Harz-Klinik, Quedlinburg; Jörg Nothroff, MediClin Reha-Zentrum Spreewald, Burg; Klaus Schröder, ZAR Stuttgart, Stuttgart; Ronja Westphal, Segeberger Kliniken, Bad Segeberg; Christa Bongarth, Klinik Höhenried, Bernried; Sieglinde Spörl-Dönch and Gerhard Alexander Müller, Frankenklinik, Bad Neustadt; Markus Wrenger, Caspar Heinrich Klinik Bad Driburg GmbH & Co. KG, Bad Driburg; Roger Marx, MediClin-Fachklinik Rhein/Ruhr, Essen; Rainer Schubmann, Dr. Becker Klinik Möhnesee, Möhnesee; Martin Schikora, Brandenburgklinik, Bernau; all centres are located in Germany. We gratefully acknowledge the German Society of Prevention and Rehabilitation of Cardiovascular Diseases (DGPR) for their cooperation. We thank Klaus Balzer for his help in creating the plots and tables.

Data Availability

Data are available (https://zenodo.org/badge/DOI/10.5281/zenodo.3605618.svg) and citable as Salzwedel, Annett, Balzer, Klaus, Koran, Iryna, Langheim, Eike, Schlitt, Axel, Nothroff, Axel, … Völler, Heinz. (2020). Patient-reported outcomes predict return to work and health-related quality of life 6 months after cardiac rehabilitation: Results from a German multi-centre registry (OutCaRe): dataset [Data set]. Zenodo. http://doi.org/10.5281/zenodo.3605618.

Funding Statement

This work was supported by the German Federal Pension Insurance (grant No. 8011-106-31/31.114.1). Funding was granted to the University of Potsdam. URL: https://www.deutsche-rentenversicherung.de/DRV/DE/Experten/Reha-Forschung/Forschungsfoerderung/Forschungsfoerderung.html The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Piepoli MF, Corrà U, Benzer W, Bjarnason-Wehrens B, Dendale P, Gaita D, et al. Secondary prevention through cardiac rehabilitation: from knowledge to implementation. A position paper from the Cardiac Rehabilitation Section of the European Association of Cardiovascular Prevention and Rehabilitation. Eur J Cardiovasc Prev Rehabil. 2010; 17: 1–17. 10.1097/HJR.0b013e3283313592 [DOI] [PubMed] [Google Scholar]
2.Anderson L, Taylor RS. Cardiac rehabilitation for people with heart disease: an overview of Cochrane systematic reviews. Cochrane Database Syst Rev. 2014: CD011273 10.1002/14651858.CD011273.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.van Halewijn G, Deckers J, Tay HY, van Domburg R, Kotseva K, Wood D. Lessons from contemporary trials of cardiovascular prevention and rehabilitation: A systematic review and meta-analysis. Int J Cardiol. 2017; 232: 294–303. 10.1016/j.ijcard.2016.12.125 [DOI] [PubMed] [Google Scholar]
4.Kabboul NN, Tomlinson G, Francis TA, Grace SL, Chaves G, Rac V, et al. Comparative Effectiveness of the Core Components of Cardiac Rehabilitation on Mortality and Morbidity: A Systematic Review and Network Meta-Analysis. J Clin Med. 2018; 7 10.3390/jcm7120514 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Candelaria D, Randall S, Ladak L, Gallagher R. Health-related quality of life and exercise-based cardiac rehabilitation in contemporary acute coronary syndrome patients: a systematic review and meta-analysis. Qual Life Res. 2019. 10.1007/s11136-019-02338-y [DOI] [PubMed] [Google Scholar]
6.Salzwedel A, Jensen K, Rauch B, Doherty P, Metzendorf M-I, Hackbusch M, et al. Effectiveness of comprehensive cardiac rehabilitation in coronary artery disease patients treated according to contemporary evidence based medicine: Update of the Cardiac Rehabilitation Outcome Study (CROS-II). Eur J Prev Cardiolog. 2020; 1: 204748732090571. 10.1177/2047487320905719 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Hegewald J, Wegewitz UE, Euler U, van Dijk JL, Adams J, Fishta A, et al. Interventions to support return to work for people with coronary heart disease. Cochrane Database Syst Rev. 2019; 3: CD010748 10.1002/14651858.CD010748.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Deutsche Rentenversicherung Bund. Reha-Bericht 2018.
9.Norlund F, Lissåker C, Wallert J, Held C, Olsson EM. Factors associated with emotional distress in patients with myocardial infarction: Results from the SWEDEHEART registry. Eur J Prev Cardiol. 2018; 25: 910–920. 10.1177/2047487318770510 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Anker SD, Agewall S, Borggrefe M, Calvert M, Jaime Caro J, Cowie MR, et al. The importance of patient-reported outcomes: a call for their comprehensive integration in cardiovascular clinical trials. Eur Heart J. 2014; 35: 2001–2009. 10.1093/eurheartj/ehu205 [DOI] [PubMed] [Google Scholar]
11.Rumsfeld JS, Alexander KP, Goff DC, Graham MM, Ho PM, Masoudi FA, et al. Cardiovascular health: the importance of measuring patient-reported health status: a scientific statement from the American Heart Association. Circulation. 2013; 127: 2233–2249. 10.1161/CIR.0b013e3182949a2e [DOI] [PubMed] [Google Scholar]
12.Bhattacharyya MR, Perkins-Porras L, Whitehead DL, Steptoe A. Psychological and clinical predictors of return to work after acute coronary syndrome. Eur Heart J. 2007; 28: 160–165. 10.1093/eurheartj/ehl440 [DOI] [PubMed] [Google Scholar]
13.Salzwedel A, Reibis R, Heidler M-D, Wegscheider K, Völler H. Determinants of Return to Work after Multi-Component Cardiac Rehabilitation. Archives of Physical Medicine and Rehabilitation. 2019. 10.1016/j.apmr.2019.04.003 [DOI] [PubMed] [Google Scholar]
14.Biering K, Nielsen TT, Rasmussen K, Niemann T, Hjollund NH. Return to work after percutaneous coronary intervention. The predictive value of self-reported health compared to clinical measures. PLoS ONE. 2012; 7: e49268 10.1371/journal.pone.0049268 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.van Cauter J, Bacquer D, Clays E, Smedt D, Kotseva K, Braeckman L. Return to work and associations with psychosocial well-being and health-related quality of life in coronary heart disease patients: Results from EUROASPIRE IV. Eur J Prev Cardiol. 2019: 2047487319843079. 10.1177/2047487319843079 [DOI] [PubMed] [Google Scholar]
16.Salzwedel A, Wegscheider K, Schulz-Behrendt C, Dörr G, Reibis R, Völler H. No impact of an extensive social intervention program on return to work and quality of life after acute cardiac event: a cluster-randomized trial in patients with negative occupational prognosis. International Archives of Occupational and Environmental Health. 2019. 10.1007/s00420-019-01450-3 [DOI] [PubMed] [Google Scholar]
17.Müller-Nordhorn J, Kulig M, Binting S, Völler H, Gohlke H, Linde K, et al. Change in quality of life in the year following cardiac rehabilitation. Qual Life Res. 2004; 13: 399–410. 10.1023/B:QURE.0000018473.55508.6a [DOI] [PubMed] [Google Scholar]
18.Farin E, Meder M. Personality and the physician-patient relationship as predictors of quality of life of cardiac patients after rehabilitation. Health Qual Life Outcomes. 2010; 8: 100 10.1186/1477-7525-8-100 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Weeger S, Farin E. The effect of the patient-physician relationship on health-related quality of life after cardiac rehabilitation. Disabil Rehabil. 2017; 39: 468–476. 10.3109/09638288.2016.1146360 [DOI] [PubMed] [Google Scholar]
20.Salzwedel A, Haubold K, Barnack B, Reibis R, Völler H. Indikatoren der Ergebnisqualität kardiologischer Rehabilitation. Rehabilitation (Stuttg). 2019; 58: 31–38. [DOI] [PubMed] [Google Scholar]
21.Zoch-Lesniak B, Dobberke J, Schlitt A, Bongarth C, Glatz J, Spörl-Dönch S, et al. Performance Measures for Short-Term Cardiac Rehabilitation in Patients at Working Age: Results of the Prospective Observational Multicenter Registry OutCaRe. Archives of Rehabilitation Research and Clinical Translation. 2019; 1: submitted. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Karoff M, Held K, Bjarnason-Wehrens B. Cardiac rehabilitation in Germany. Eur J Cardiovasc Prev Rehabil. 2007; 14: 18–27. 10.1097/HJR.0b013e3280128bde [DOI] [PubMed] [Google Scholar]
23.Rauch B, Riemer T, Schwaab B, Schneider S, Diller F, Gohlke H, et al. Short-term comprehensive cardiac rehabilitation after AMI is associated with reduced 1-year mortality: results from the OMEGA study. Eur J Prev Cardiol. 2014; 21: 1060–1069. 10.1177/2047487313486040 [DOI] [PubMed] [Google Scholar]
24.Benzer W, Rauch B, Schmid J-P, Zwisler AD, Dendale P, Davos CH, et al. Exercise-based cardiac rehabilitation in twelve European countries results of the European cardiac rehabilitation registry. Int J Cardiol. 2017; 228: 58–67. 10.1016/j.ijcard.2016.11.059 [DOI] [PubMed] [Google Scholar]
25.Rosengren A, Hawken S, Ounpuu S, Sliwa K, Zubaid M, Almahmeed WA, et al. Association of psychosocial risk factors with risk of acute myocardial infarction in 11119 cases and 13648 controls from 52 countries (the INTERHEART study). Case-control study. Lancet. 2004; 364: 953–962. 10.1016/S0140-6736(04)17019-0 [DOI] [PubMed] [Google Scholar]
26.Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. Validity of a Brief Depression Severity Measure. J Gen Intern Med. 2001; 16: 606–613. 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Eifert GH, Thompson RN, Zvolensky MJ, Edwards K, Frazer NL, Haddad JW, et al. The Cardiac Anxiety Questionnaire. Development and preliminary validity. Behaviour Research and Therapy. 2000; 38: 1039–1053. 10.1016/s0005-7967(99)00132-1 [DOI] [PubMed] [Google Scholar]
28.Topp CW, Østergaard SD, Søndergaard S, Bech P. The WHO-5 Well-Being Index. A systematic review of the literature. Psychother Psychosom. 2015; 84: 167–176. 10.1159/000376585 [DOI] [PubMed] [Google Scholar]
29.Ware JE, Kosinski M, Keller SD, Lincoln RI. Sf-12: How to score the sf-12 physical and mental health summary scales. Quality Metric incorporated. Boston MA. 1998.
30.Wirtz M, Farin E, Bengel J, Jäckel WH, Hämmerer D, Gerdes N. IRES-24 Patientenfragebogen: Entwicklung der Kurzform eines Assessmentinstrumentes in der Rehabilitation mittels der Mixed-Rasch-Analyse. Diagnostica. 2005; 51: 75–87. 10.1026/0012-1924.51.2.75 [DOI] [Google Scholar]
31.Beierlein C, Kovaleva A, Kemper CJ, Rammstedt B. Ein Messinstrument zur Erfassung subjektiver Kompetenzerwartungen: Allgemeine Selbstwirksamkeit Kurzskala (ASKU). Mannheim: GESIS; 2012. http://nbn-resolving.de/urn:nbn:de:0168-ssoar-292351. Accessed 16 October 2019.
32.Löffler S, Wolf HD, Vogel H. Das Würzburger Screening zur Identifikation von beruflichen Problemlagen–Entwicklung und Validierung. Gesundheitswesen. 2008; 70 10.1055/s-0028-1086302 [DOI] [Google Scholar]
33.Kazis LE, Anderson JJ, Meenan RF. Effect Sizes for Interpreting Changes in Health Status. Medical Care. 1989; 27: 178–189. [DOI] [PubMed] [Google Scholar]
34.Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed Hillsdale, NJ: Erlbaum; 1988. [Google Scholar]
35.Röhrig B, Nosper M, Linck-Eleftheriadis S, Strandt G, Salzwedel A, Völler H. Ergebnismessung und -bewertung kardiologischer Rehabilitation anhand von Qualitätsindikatoren-eine Methodenbeschreibung. Rehabilitation (Stuttg). 2014; 53: 31–37. [DOI] [PubMed] [Google Scholar]
36.Berg SK, Rasmussen TB, Thrysoee L, Lauberg A, Borregaard B, Christensen AV, et al. DenHeart: Differences in physical and mental health across cardiac diagnoses at hospital discharge. Journal of Psychosomatic Research. 2017; 94: 1–9. 10.1016/j.jpsychores.2017.01.003 [DOI] [PubMed] [Google Scholar]
37.Löwe B, Unützer J, Callahan CM, Perkins AJ, Kroenke K. Monitoring depression treatment outcomes with the patient health questionnaire-9. Medical Care. 2004; 42: 1194–1201. 10.1097/00005650-200412000-00006 [DOI] [PubMed] [Google Scholar]
38.Wood AM, White IR, Royston P. How should variable selection be performed with multiply imputed data. Stat Med. 2008; 27: 3227–3246. 10.1002/sim.3177 [DOI] [PubMed] [Google Scholar]
39.Rubin DB. Multiple Imputation for nonresponse in surveys. New York, NY: Wiley; 1987. [Google Scholar]
40.van Beek MHCT, Zuidersma M, Lappenschaar M, Pop G, Roest AM, van Balkom AJLM, et al. Prognostic association of cardiac anxiety with new cardiac events and mortality following myocardial infarction. Br J Psychiatry. 2016; 209: 400–406. 10.1192/bjp.bp.115.174870 [DOI] [PubMed] [Google Scholar]
41.Muschalla B, Glatz J, Linden M. Bibliotherapy on coping with illness improves health literacy but not heart-related anxiety of patients in cardiological rehabilitation. Psychother Psychosom. 2013; 82: 349–350. 10.1159/000350449 [DOI] [PubMed] [Google Scholar]
42.Uhlig J, Schmitz C, Langheim E, Rademacher W, Köllner V. Effektivität von psychokardiologischer Versorgung in der stationären Rehabilitation—ein Pilotprojekt. Psychotherapie · Psychosomatik · Medizinische Psychologie. 2019; 69. [DOI] [PubMed] [Google Scholar]
43.Gragnano A, Negrini A, Miglioretti M, Corbière M. Common Psychosocial Factors Predicting Return to Work After Common Mental Disorders, Cardiovascular Diseases, and Cancers. A Review of Reviews Supporting a Cross-Disease Approach. Journal of Occupational Rehabilitation. 2017. 10.1007/s10926-017-9714-1 [DOI] [PubMed] [Google Scholar]
44.Schulz-Behrendt C, Salzwedel A, Rabe S, Ortmann K, Völler H. Aspekte beruflicher und sozialer Wiedereingliederung aus Sicht kardiovaskulär erkrankter Rehabilitanden in besonderen beruflichen Problemlagen—Ergebnisse einer qualitativen Erhebung. Rehabilitation (Stuttg). 2017; 56: 181–188. [DOI] [PubMed] [Google Scholar]
45.Salzwedel A, Reibis R, Wegscheider K, Eichler S, Buhlert H, Kaminski S, et al. Cardiopulmonary exercise testing is predictive of return to work in cardiac patients after multicomponent rehabilitation. Clin Res Cardiol. 2016; 105: 257–267. 10.1007/s00392-015-0917-1 [DOI] [PubMed] [Google Scholar]
46.Mehrdad R, Ghadiri Asli N, Pouryaghoub G, Saraei M, Salimi F, Nejatian M. Predictors of early return to work after a coronary artery bypass graft surgery (CABG). Int J Occup Med Environ Health. 2016; 29: 947–957. 10.13075/ijomeh.1896.00798 [DOI] [PubMed] [Google Scholar]
47.World Health Organization. International classification of functioning, disability and health. ICF. Geneva: WHO; 2001.
48.Bjarnason-Wehrens B, Held K, Hoberg E, Karoff M, Rauch B. Deutsche Leitlinie zur Rehabilitation von Patienten mit Herz-Kreislauferkrankungen (DLL-KardReha). Clin Res Cardiol Suppl. 2007; 2: 1–54. 10.1007/s11789-007-0001-0 [DOI] [Google Scholar]
49.Balady GJ, Williams MA, Ades PA, Bittner V, Comoss P, Foody JM, et al. Core components of cardiac rehabilitation/secondary prevention programs: 2007 update: a scientific statement from the American Heart Association Exercise, Cardiac Rehabilitation, and Prevention Committee, the Council on Clinical Cardiology; the Councils on Cardiovascular Nursing, Epidemiology and Prevention, and Nutrition, Physical Activity, and Metabolism; and the American Association of Cardiovascular and Pulmonary Rehabilitation. Circulation. 2007; 115: 2675–2682. 10.1161/CIRCULATIONAHA.106.180945 [DOI] [PubMed] [Google Scholar]
50.Deutsche Rentenversicherung Bund. Rehabilitation 2018. Statistik der Deutschen Rentenversicherung Leistungen zur medizinischen Rehabilitation, sonstige Leistungen zur Teilhabe und Leistungen zur Teilhabe am Arbeitsleben der gesetzlichen Rentenversicherung im Jahre 2018. 216th ed.; 2018.
51.Oldridge N, Höfer S, McGee H, Conroy R, Doyle F, Saner H. The HeartQoL: part II. Validation of a new core health-related quality of life questionnaire for patients with ischemic heart disease. Eur J Prev Cardiol. 2014; 21: 98–106. 10.1177/2047487312450545 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0232752.r001

Decision Letter 0

Stephanie Prince Ware

21 Nov 2019

PONE-D-19-29345

Patient-reported outcomes predict return to work and health-related quality of life 6 months after cardiac rehabilitation: Results from a German multi-centre registry (OutCaRe)

PLOS ONE

Dear Dr. Salzwedel,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Return to work (RTW) and quality of life following a cardiac event are important patient outcomes and this paper looks to identify factors that predict these outcomes following cardiac rehabilitation. The paper has been assessed by two experts in the field of cardiac rehabilitation and they have raised some excellent points. While the paper identifies an important area of research, the paper requires major revisions before it can be considered eligible for publication. Below you will find comments from the reviewers, as well as some points of my own. The authors should also consider all of the work done on predictors of HRQL in the CR setting and describe how this work adds to the pre-existing literature. If you feel that you can make the required changes, we would be happy to re-review the paper.

We would appreciate receiving your revised manuscript by Jan 05 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Stephanie Prince Ware, PhD

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

3. One of the noted authors is a group or consortium: OutCaRe investigators

In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

Additional Editor Comments:

1. Abstract: The methods should state more clearly that you are looking to predict RTW with the variables collected at baseline and discharge from cardiac rehabilitation. Also, the results should present data.

2. Line 64: what is the advantage?

3. Please use acronyms consistently. For example RTW is used on line 64, but spelled out fully on 67. HRQL is spelled out in full on line 80. Lines 95-96 have both of these terms spelled out fully.

4. Methods: Consider using headings for all of the parameters and describing them under each heading to help improve readability. It is difficult to follow all of the measures.

5. Reason for CR should also be controlled for in the model as the severity of event is likely to impact an individual’s desire to RTW.

6. It sounds like the CR program is only 3-4 weeks in duration which appears quite short. The authors describe 12 training sessions per week, but this does not appear to be feasible. Can you clarify?

7. What activities are the participants undertaking when they are asked for their BORG rating?

8. The SF-12 is a generic measure of HRQL for the outcome following CR. This is likely a limitation as it does not assess disease-specific HRQL.

9. Is there an ethics approval number associated with this project?

10. Were the RTW analysis only conducted on those who were working at baseline?

11. Statistics: Please change “metric” with “descriptive”.

12. Line 152: Respectively to what?

13. Lines 197-198: In the methods clinically important differences in these measures should be outlined and results should be discussed in terms of clinical and statistical significance.

14. Why are the ORs for RTW age-adjusted? There is a limitation of age on the participants already.

15. Include all outcomes in figures 2 and 3.

16. Line 285, what does “opp. RTW motivation” mean?

17. Discussion: There is discussion that greater reported stress at work may be related to earlier RTW due to those individuals holding higher positions. The study collected respondent education; this should be looked at in relation to stress and RTW.

18. Line 311: this appears to be an error: “physis-focussed”

19. It is intuitive that physical health measures would predict the PCS of the SF-12 while mental health measures would predict the MCS of the SF-12. The authors should comment on which aspects are the most important. Also, there is likely multicollinearity occurring in the models, was this assessed?

20. Discussion: the authors should consider discussing the benefit of Social Workers as part of the interdisciplinary treatment in CR.

21. Discussion: Were the data missing at random? Were those with complete data different than those with incomplete data?

22. Reference 1 and 35 are the same.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This paper has hallmarks of good clinical and ecological validity in light of its links with a patient outcome registry. It would be helpful if the Abstract results included actual data, as opposed to statements which are more like conclusions

Whilst desire to retire was an interesting and important covariate to RTW, would not age also be key, in respecting the Methods clearly had a cut-off of 65 years.

There are a number of risk factor data reported and mention of the Borg scale. However, their relevance to all this isn't clear. Furthermore, the changes in most risk factors, were these actually related to the Rehab process or merely functions of medications prescribed.

Line 105 mentions attendance of 12 training units per week? What is a training unit? In terms of programme fidelity and in addition to programme uptake and completion, how compliant were the participants with attending? (e.g. participation rates per week). The Exercise Capacity was measured but not factored into the relationship with RTW? A way to check on whether exercise capacity had changed was to see if patients heart rate or RPE was lower for any give workmate in Watts (submit or max), which in fact you have reported

More clarity required whether RTW data included those already unemployed and or on sick leave prior to commencement of CR (apologies for my oversight if this has been stated)

There seem to be lots of abbreviations; can these be reduced? or at least make sure they are clearly defined in legends for all Tables and Figures

Reviewer #2: This paper aimed to identify predictors of return to work (RTW) and quality of life (QOL) 6 months after cardiac rehabilitation. Although the paper describes an interesting topic, I do have quit some concerns and comments that need attention. Some major revisions need to be made in the manuscript before this manuscript can be reviewed in detail.

Major comments:

1. The authors should be more clear what is the novelty of their paper. In the introduction the authors describe that several factors are known to influence RTW (such as depression, anxiety, expectations) in lines 69-72. However, in line 77 it is said that few is known about the association between patient centered outcomes and RTW. Are the factors mentioned in lines 69-72 not also patient centered outcomes? And in line 62, it is described that the effect of CR on QOL is well-known, so why is QOL choses as an outcome in this paper?

2. The methods section describing the study outcomes/ predictors is extremely difficult to follow (lines 116-136), partly due to the large amount of parameters. Could the authors try to re-organize this paragraph and make clear what exactly is measured and which instrument is used (for instance with a table or using bullet points).

3. Could the authors give some more information on the content of the CR program (duration, training sessions and counselling sessions). Twelve training sessions per week seem quite high.

4. With regard to the used statistics I have some questions:

a. Were all predictors added to the same multiple linear regression model? I can imagine that there might be some overlap and correlation between some of the predictors. Did the authors test for multicollinearity?

b. Could the authors add some information about the power of their study to test for this large amount of predictors?

c. Why was chosen for only one imputation and not for multiple imputations?

5. With regard to generalizability of the data: could the authors say something about patients that were not willing to participate in this study? Is the sample that was willing to participate representative of the whole population?

6. Please clarify in your paper what is seen as a PROM and what is seen as subjective health (and what is the difference between these two concepts).

7. In the paper there is a lot of attention focused on the role of pension desire in RTW. What is the rationale behind investigating this? If patients have the desire and possibility to retire, should that be seen as a negative outcome of CR? In my opinion, it should not be seen like that. It can be a good and well thought out choice. Reading your results, one could conclude that to improve RTW we should try to change someone’s wish to retire during CR. What is the vision of the authors on this?

8. The authors chose to look for predictors of successful return to work. On top of patients still on sick leave, the group of patient with unsuccessful return to work included patients that chose to retire, patient that applied for pension and patient that were already unemployed before CR. Wouldn’t it be more interesting to look for predictors of patients that failed to return to work as a result of their cardiac disease, even though they would have wanted to (the group of 15% still on sick leave)? This would be the group that might benefit of extra support during or after CR.

9. The percentages in lines 215-217 add up to 102.3%.

10. Lines 228-251 in the results are difficult to follow. Could the authors clarify if the mentioned predictors are measured at start or completion of CR or concern an improvement during CR?

11. If I understand well, figures 2 and 3 only show outcomes of significant results (predictors). This gives a misleading representation of all results. I would suggest to add all variables that were tested to this figure so that it is clear which predictors were significant and which were not.

12. The discussion is quite difficult to follow. When are results of other authors discussed and when does it concern the author’s own results? Also, it is sometimes difficult to follow if the outcomes with regard to QOL or RTW are discussed. And how can these results be used in daily clinical practice. I suggest a re-organization of this section (first discuss own results (split RTW and QOL), second how does this compare to results of other studies, third what can be done differently with this information in daily CR practice) and a “deeper” discussion of the work.

Smaller remarks:

1. Please try to be consistent in your manuscript with the use of abbreviations.

2. What do the authors mean with “mid-term associations”?

3. Could the authors try to rephrase the aim in the abstract? It is a complex sentence that I had to read several times to understand.

4. What do the authors mean with occupational study? (line 29)

5. In Table 1 :

a. living situation does not add up to 100%.

b. What is meant with the number 3.4 +/- 1.6 behind comorbidities?

c. Is AF a diagnosis or comorbidity?

6. In Table 2: could the author add p-values?

7. Social integration is not the same as return to work (it also includes return to leisure time activities etc). I would suggest to not mix these terms.

8. How many centres were involved in this study?

9. The description of the figures has a lot of overlap with the text.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2020 May 5;15(5):e0232752. doi: 10.1371/journal.pone.0232752.r002

Author response to Decision Letter 0

17 Jan 2020

Please see the attached response to the reviewers letter, that includes a formatted detailed point-by-point reply.

At first, we thank the editor and reviewers for reading the manuscript as well as for their helpful comments, with which we have dealt intensively. We reviewed and revised the manuscript accordingly. Please find our point-by-point response below. Modifications in the manuscript are highlighted in grey.

Sincerely,

Annett Salzwedel

for the authors of the manuscript.

Editor

…

*****We tried to analyse all available studies on the same topic. In more than 30 relevant original research publications and additional reviews, which were evaluated in detail, there were no studies, that systematically investigated modifiable parameters in the CR setting. We rephrased and extended the second part of the introduction to describe the purpose of our work.

Additional Editor Comments:

*****We rewrote the abstract including the rephrasing of the study aim. The results include now several data. Due to the word restriction, we present only a selection of the most important findings.

2. Line 64: what is the advantage?

*****In contrast to exercise-based rehab, comprehensive rehabilitation programs reduce not only cardiovascular mortality and myocardial infarction but also all-cause mortality and stroke rate according to the review and meta-analysis by Halewijn et al. The multi-modal approach also increases return to work up to 6 months (Hegewald et al.). We modified the sentence as follows:

“However, CR programmes based on a multi-modal approach seem to be at an advantage over exercise-based programmes regarding both clinical and occupational outcomes.[4,6]”

3. Please use acronyms consistently. For example RTW is used on line 64, but spelled out fully on 67. HRQL is spelled out in full on line 80. Lines 95-96 have both of these terms spelled out fully.

*****We revised the manuscript regarding the consistent use of abbreviations. Apart from the abstract, we also omit the abbreviation of return-to-work and patient reported outcome measures in the manuscript for better readability.

4. Methods: Consider using headings for all of the parameters and describing them under each heading to help improve readability. It is difficult to follow all of the measures.

*****We re-organized the paragraph using bullet points, as suggested by Reviewer # 2, hopefully this will make it easier to read the manuscript.

5. Reason for CR should also be controlled for in the model as the severity of event is likely to impact an individual’s desire to RTW.

*****The CRF distinguished between 12 different reasons for CR. Some of these reasons were rare. Correspondingly, this variable was not selected for the final RTW model. To control the reason for CR, we simplified the variable to ‚ACS‘ vs ‚no ACS‘. This variable was selected for the final model. Although not significant, we added the marginal means of RTW by reason for CR to the supplement (figure S4).

6. It sounds like the CR program is only 3-4 weeks in duration which appears quite short. The authors describe 12 training sessions per week, but this does not appear to be feasible. Can you clarify?

*****We extended the description of the conducted rehab program as provided by the German pension insurance and we hope for a better understanding:

“All patients performed a comprehensive standardised CR programme with a regular duration of three to four weeks according to the specifications of the German pension insurance.[16] The programme can be performed in either an inpatient or outpatient setting and consists of all-day activities including counseling by a cardiologist, risk-factor modification strategies (e.g. patient education on nutritional habits, smoking cessation, physical activity, and medication adherence), physician-supervised exercise training and sports therapy (e.g. training on a bicycle ergometer, outdoor walking, resistance training, gymnastics), psychosocial interventions (health education and counseling, psychotherapy, stress management in single or group sessions), and vocational assessment and physician and social worker counseling. [17,18] On average, patients perform 12 training and sports therapy units per week with a duration –depending on the training group and physical performance – up to 30 Minutes and 45 Minutes for outdoor walking, respectively, and 8 additional counselling sessions.[19]”

7. What activities are the participants undertaking when they are asked for their BORG rating?

*****BORG rating was done during the endurance training on the bicycle ergometer. However, the BORG rating is of no relevance for the presented results – we removed this information from the text.

8. The SF-12 is a generic measure of HRQL for the outcome following CR. This is likely a limitation as it does not assess disease-specific HRQL.

*****We added this to the limitations:

“The SF-12 used to operationalize HRQL is a generic instrument. Probably, a disease-specific screening tool (eg. HeartQoL questionnaire) [40] could achieve a higher responsiveness to change and acceptance in the investigated heart-disease population. Nevertheless, the SF-12 is well established and commonly used, which ensures the comparability of our results with other studies and populations.”

9. Is there an ethics approval number associated with this project?

*****The approval number (S 4(a)/2017) is added to the Ethics section.

10. Were the RTW analysis only conducted on those who were working at baseline?

*****The RTW analysis was conducted to all patients as the occupational resumption is the main goal of CR for the German pension insurance, which is the funding institution for the most CR patients before the retirement age of 65. Therefore, the support by social workers in CR is mandatory according to the specifications of the German pension insurance. Social workers advise for example on retraining or gradual reintegration as needed. We modified the regarding sentence in the statistics: “For all further analyses, only patients with a non-missing 6 month outcome in the original dataset were included, regardless of the employment status at baseline.” (line 201-203)

11. Statistics: Please change “metric” with “descriptive”.

*****We have changed the corresponding sentence to ‚For description, continuous variables are presented as means ± standard deviation, and categorical variables as absolute values and percentages.‘ to emphasize the different presentation of continuous and categorical variables.

12. Line 152: Respectively to what?

*****We modified the sentence as follows: “Differences in variables between admission to CR (reference), discharge from CR, and follow-up were tested for statistical significance using Wilcoxon for continuous variables and McNemar tests for categorical variables, respectively.”

13. Lines 197-198: In the methods clinically important differences in these measures should be outlined and results should be discussed in terms of clinical and statistical significance.

*****Minimal (clinically) important difference (MID or MCID) is a concept depending value, mostly calculated to evaluate the longitudinal change of a measure in an individual based on the need for intervention. There are several methods to calculate a MID without a consense for appropriate use within the scientific community. We added available and appropriate MIDs to the statistic section, results and discussion as follows:

“In addition, the changes in patient-reported outcome measures during CR were assessed using the minimal important difference (MID). This concerns to the IRES-24 (MID 0.5 points) [30] and the SF-12 (MID 2 points for the physical and 3 points for the mental component summary, respectively).[31] For the WHO-5, we anticipate a MID of 10%-points.[23] No MID was considered for PHQ-9 because the MID reported in the literature focuses the change of PHQ-values in the acute phase of depression treatment in affected patients,[32] which is of subordinate relevance in our investigation. There is no MID available for the HAF-17.”

“The changes in the physical and mental scales of SF-12 (+5.7/+6.1 points) and IRES-24 (physical health +1.2, mental health +1.4, pain 1.0) and in the WHO-5 (+18.3%-points) exceeded the MID.”

“Furthermore, attention should be paid to changes during CR in the parameters affecting the analysed mid-term outcome of CR. Besides heart-focused anxiety as mentioned above, the means in mental and physical quality of life as well as subjective health (IRES-24) and well-being in the WHO-5 were significantly improved with small to moderate effect sizes. The notified changes are considered clinically relevant since they substantially exceed the MID by 1.8 (WHO-5) to 2.9 times (physical component of SF-12).”

However, as stated in the methods section (line 107-110) this paper focuses on the predictive value of parameters for return-to-work and HRQL. The modifiability of the parameters during CR is expressly subject of previous publication and is comprehensively discussed there. (Zoch-Lesniak B. et al. Performance Measures for Short-Term Cardiac Rehabilitation in Patients at Working Age: Results of the Prospective Observational Multicenter Registry OutCaRe. Archives of Rehabilitation Research and Clinical Translation. 2019. In press.).

14. Why are the ORs for RTW age-adjusted? There is a limitation of age on the participants already.

*****That is right, patients at pension age (>65 years) were excluded from the study. Nevertheless, according to experience, age and sex have an impact on RTW as younger patients have a higher opportunity returning to work. Therefore, we calculated the bivariate association between HRQL 6 months after CR and RTW adjusted for age and sex.

Age was also taken into account conducting the multivariable predictor analysis of RTW, but achieved no statistical significance and was eliminated from the model.

15. Include all outcomes in figures 2 and 3.

*****We now clarify in the statistics section that we present only final models after backward selection. These models give more precise estimates of regression coefficients than the models that include all potential regressors which are overfitted to the data. However, we show the analysis of all measurements in the supplement (figures S1-S3).

16. Line 285, what does “opp. RTW motivation” mean?

*****It meant ‘the opposite of motivation to return to work’. We removed this misleading phrase.

*****High educated patients returned to work in 78%, while the proportion in less educated patients was 64%. There was no association between stress and education. We added the information to the responding paragraph in discussion (line 366-372):

“We assume that patients who report a higher perception of work stress may hold a leading position with a higher degree of personal identification and recognition, resulting in a higher intention to return-to-work. This interpretation is quite speculatively but may be supported by the fact that the proportion of patients returned to work was significantly higher by 14%-points in high educated than in less educated patients (78% vs. 64%). However, perceived stress at work was not associated with the educational level and also predicted a diminished mental quality of life in our study.”

18. Line 311: this appears to be an error: “physis-focussed”

*****We removed the term.

*****We restructured the entire discussion section since there were items suggested to add. Hereby, we have also realigned the focus of HRQL related predictors.

We studied potential multicollinarity by calculation of the variance inflation factors (VIF) for the complete set of potential regressors. VIF were distinctly below the generally accepted limit of 10 with one exception: mental health at baseline as assessed by use of the IRES-24 had an VIF of 10.48 which means that the corresponding coefficients might be slightly compromised by multicollinearity. We added to the limitations section: “Multicollinearity may limit the interpretability of model coefficients. However, the variance inflation factors of the studied predictors were distinctly below the generally accepted limit of 10 with one exception: the mental health at admission had a variance inflation factor of 10.48. Altogether, the multivariate models were sufficiently stable to allow clinical interpretations.”

20. Discussion: the authors should consider discussing the benefit of Social Workers as part of the interdisciplinary treatment in CR.

*****We added a discussion of the benefit of social workers in line 345-354: “In this context, social workers in CR play a crucial role as they advise for example on retraining or gradual reintegration as needed. The support by social workers in CR program is mandatory according to the specifications of the German pension insurance. Nevertheless, in accordance with previous studies and official data, return-to-work rates after a cardiac event were only 69% in our study. [7,9,32] An own randomized clinical trial evaluating an extended social therapy counseling and training program during CR remained without an effect on return-to-work and HRQL in patients at risk of occupational reintegration. [12] This finding suggests that the regular therapy density in the standardized German intensive short-term CR program as described in the methods section is of maximum volume. A continuous support by a social worker in the further course after a cardiac event could remedy.”

21. Discussion: Were the data missing at random? Were those with complete data different than those with incomplete data?

*****There were systematic differences between patients with with complete and with incomplete data. These differences were taken into account by the chosen imputation model.

The following statements are included in the limitation section:

- …. Second, there is a selection bias due to the incomplete follow up. Non-responding study participants were younger, less educated, more often smokers, and less frequently employed. At discharge from CR, they reported a lower health status. These differences were taken into account by the imputation model. However, there may be reasons for missingness that were not covered by our data. This has to be kept in mind when interpreting the study results.

- … Patients were allowed to refrain from answering parts of the questionnaires or individual questions. Missing values were imputed in the statistical models to balance the data. However, it may be that these modeling approaches do not compensate the selection effects.

22. Reference 1 and 35 are the same.

*****Thank you. We corrected this mistake.

Reviewer #1

This paper has hallmarks of good clinical and ecological validity in light of its links with a patient outcome registry. It would be helpful if the Abstract results included actual data, as opposed to statements which are more like conclusions

*****Thank you very much. We revised the abstract and present now a selection of the most important findings.

Whilst desire to retire was an interesting and important covariate to RTW, would not age also be key, in respecting the Methods clearly had a cut-off of 65 years.

*****Age was included as potential covariate for each of the endpoints, but was only selected for explanation of MCS.

*****There are several parameters including sociodemographic, rehabilitation indication, comorbidities and cardiovascular risk factors, which are predominantly reported to describe the investigated patient population. This is a prerequisite to ensure transparency regarding the comparability of our population to others of interest. However, relevant parameters, in particular modifiable risk factors, were taken into account in the statistic modeling. The statistic section was modified to fully describe the modeling process.

The risk factor management including the prescription of medications for secondary prevention is an essential part of the German standardized CR program. The effect of this program on risk factors cannot be differentiated for single program components.

The BORG rating is irrelevant to the presented results – we removed this information from the text.

*****A training unit consists of up to 30 Minutes endurance or resistance training or gymnastics, or up to 45 Minutes outdoor walking. The duration and intensity depends on the assigned training group and physical performance of the individual patient. The CR program was conducted mostly as an inpatient intervention with mandatory all-day activities, which were specified for the patient. We did not capture patients’ compliance, but we generally assume that it was high due to the obligatory character of the intervention. We extended the description of the conducted rehab program as provided by the German pension insurance and hope for better understanding:

The Exercise Capacity was measured but not factored into the relationship with RTW? A way to check on whether exercise capacity had changed was to see if patients heart rate or RPE was lower for any give workmate in Watts (submit or max), which in fact you have reported

*****The change of exercise capacity is considered in the variable ‘endurance training load [watt]’. This changes during CR achieved no significance for the final model, while the endurance training load at admission to CR turned out to be a predictor of RTW.

More clarity required whether RTW data included those already unemployed and or on sick leave prior to commencement of CR (apologies for my oversight if this has been stated)

*****At baseline, 1127 patients (90.3%) were employed, 94 (7.5%) were unemployed (others were homemaker or in temporary annuity). 892 patients (71.5%) were on sick leave before CR (table 1). RTW was analysed for all enrolled patients. We added this information into the section “Statistics”:

“For all further analyses, only patients with a non-missing outcome in the original dataset were included, regardless of the employment status at baseline.” (line 201-203)

There seem to be lots of abbreviations; can these be reduced? or at least make sure they are clearly defined in legends for all Tables and Figures

*****The most abbreviations result from the use of several questionnaires. We revised all table and figure legends and the manuscript regarding the consistent use of abbreviations. Apart from the abstract, we also omit the abbreviation of return-to-work and patient reported outcome measures in the manuscript. We hope it is easier to read now.

Reviewer #2

This paper aimed to identify predictors of return to work (RTW) and quality of life (QOL) 6 months after cardiac rehabilitation. Although the paper describes an interesting topic, I do have quit some concerns and comments that need attention. Some major revisions need to be made in the manuscript before this manuscript can be reviewed in detail.

Major comments:

*****The introduction was extensively revised. We hope for a better understanding of our intention.

*****According to your suggestion, we re-organized the paragraph using bullet points, hoping it is now a more structured concept.

3. Could the authors give some more information on the content of the CR program (duration, training sessions and counselling sessions). Twelve training sessions per week seem quite high.

*****We extended the description of the conducted rehab program as provided by the German pension insurance and hope, that it is clearer now:

4. With regard to the used statistics I have some questions:

*****We studied potential multicollinarity by calculation of the variance inflation factors (VIF) for the complete set of potential regressors. VIF were distinctly below the generally accepted limit of 10 with one exception: mental health at baseline as assessed by use of the IRES-24 had an VIF of 10.48 which means that the corresponding coefficients might be slightly compromised by multicollinearity. We added to the limitations section: “Multicollinearity may limit the interpretability of model coefficients. However, the variance inflation factors of the studied predictors were distinctly below the generally accepted limit of 10 with one exception: the mental health at admission had a variance inflation factor of 10.48. Altogether, the multivariate models were sufficiently stable to allow clinical interpretations.”

b. Could the authors add some information about the power of their study to test for this large amount of predictors?

*****We assume that despite the lack of multicollinearity the models with the full set of covariates (which are now shown in the appendix) are overfitted which results in a lack of power. However, after variable selection the models were robust and performed quite well to identify several predictors of return to work and HRQL

c. Why was chosen for only one imputation and not for multiple imputations?

*****Thank you for this valuable comment. We recalculated all models and performed the backward selections with 20 imputations instead of one as described in detail in the statistics paragraph.

*****Patients without an informed consent were excluded from this study. Due to privacy, it was not allowed to capture data for these patients. However, our participants were similar to our target population (cardiac patients ≤65 years of age) according to statistics on the rehabilitation services of the German pension insurance with regard to age and sex. We added this information including a reference to the limitations (line 390-392).

6. Please clarify in your paper what is seen as a PROM and what is seen as subjective health (and what is the difference between these two concepts).

*****‚Subjective health‘ is a key area of cardiac rehabilitation (line 106) meaning that the interprofessional CR team strives to improve the perceived health status of a patient in a holistic treatment approach. We used PROMs to operationalize subjective health, as they are “questionnaires completed by patients to assess the effects of disease or treatment (or both) on symptoms, functioning, and health-related quality of life from their perspective (Calvert M et al. 2019). In the methods section, we stated now: “… The majority of captured data were taken from patients’ records, while social medicine and subjective health were assessed by means of specific patient-reported outcomes measures used in the study: …” (line 139-141).

*****As described in the methods section, this paper is part of the OutCaRe study, that started with a Delphi expert survey of 70 cardiologists and other professionals. These experts extracted and consented potential indicators of rehabilitation success. All chosen parameters including pension desire were analyzed in terms of their predictive value for return-to-work and HRQL after CR according to the predefined methodological approach. In this analysis, pension desire as it was one of only two parameters, that predict both return-to-work and HRQL (discussion, first paragraph). For this reason, pension desire should be taken into account for any predictive modelling of return to work, at least as a control variable.

Pension desire is an item of the Würzburger Screening, a valid instrument identifying occupational issues and the need for vocational rehabilitation.

However, patients’ pension desire is not synonymous with the possibility to retire. Retirement before pension age requires a proven permanent significant limitation of occupational capacity. Patients in our study with pension desire at admission to CR were retired in only 16%, 33% were at sick leave, while 37% had returned to work within a half year after CR. Pension desire may be a surrogate parameter, that implicates aspects of non-captured occupation-related factors as well as disease-related aspects (see discussion section). Indeed, the occupational resumption is the main goal of CR for the German pension insurance, which is the funding institution for the most CR patients before the retirement age of 65

*****Since successful return to work is a main goal of CR by law for all patients of employable age, it was per protocol the predefined aim of this study to systematically investigate a multitude of modifiable clinical parameters and patient-reported outcome measures regarding their impact on return-to-work. Of course, the group still on sick leave is of special interest. However, this group comprises only 190 patients. In the study, in total 1586 patients were enrolled, 1262 patients responded to the follow-up-survey. Thus, it is a high selected patient population. Results of such an analysis would be difficult to translate into clinical practice because patients who are on sick leave at follow up cannot be identified at CR admission.

9. The percentages in lines 215-217 add up to 102.3%.

*****Multiple answers were possible. Patients could be employed or unemployed or at sick leave if they had applied for pension or could be unemployed and at sick leave.

10. Lines 228-251 in the results are difficult to follow. Could the authors clarify if the mentioned predictors are measured at start or completion of CR or concern an improvement during CR?

*****We revised the regarding paragraphs and included consistently the respective capturing times of the parameters.

*****The models with the full set of covariates are now shown in the supplement (figures S1-S3). However, these models are not very powerful and thus not appropriate for identification of relevant predictors. We thus preferred to keep the results of the models with variable selection in the main part of the paper.

*****We are grateful for this suggestion. For easier readability, we rewrote the discussion. We start with the description of main results, followed by predictors of RTW, HRQL and both with a deeper discussion of the latter, subsequently discussing non-intuitive results. Finally, we discuss the impact of modifications in PROMs during CR and the usefulness of an interprofessional team and finish with the limitations of the study.

Smaller remarks:

1. Please try to be consistent in your manuscript with the use of abbreviations.

*****We revised the manuscript regarding the consistent use of abbreviations.

2. What do the authors mean with “mid-term associations”?

*****We meant associations … in the mid-term course after CR. We rephrased this sentence accordingly (line 91).

3. Could the authors try to rephrase the aim in the abstract? It is a complex sentence that I had to read several times to understand.

*****We extensively revised the abstract as well as the the aim of our investigation: “… Out of a multitude of variables collected at CR admission and discharge, we aimed to identify predictors of returning to work (RTW) and HRQL 6 months after CR.”

4. What do the authors mean with occupational study? (line 29)

*****We corrected this mistake. The correct term is “observational”

5. In Table 1 :

a. living situation does not add up to 100%.

*****The category „others“ with 3.4% was missing (e.g. commune). We added it into the table 1.

b. What is meant with the number 3.4 +/- 1.6 behind comorbidities?

*****This is the mean and standard deviation of number of comorbidities. We added the information into the table legend.

c. Is AF a diagnosis or comorbidity?

*****AF (atrial fibrillation) is a comorbidity, see table 1.

6. In Table 2: could the author add p-values?

*****Yes, indeed. We added a column with the p-values.

7. Social integration is not the same as return to work (it also includes return to leisure time activities etc). I would suggest to not mix these terms.

*****Thank you for this hint. We substituted ‘social integration’ by ‘occupational resumption’.

8. How many centres were involved in this study?

*****12 rehabilitation centers were involved. Please see line 112.

9. The description of the figures has a lot of overlap with the text.

*****We removed the redundant text from the figure legends.

Attachment

Submitted filename: 2020-01-17_OutCaRe_RTW-QoL_ReviewerResponse.docx

Click here for additional data file.^{(65.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0232752.r003

Decision Letter 1

Stephanie Prince Ware

14 Feb 2020

PONE-D-19-29345R1

Patient-reported outcomes predict return to work and health-related quality of life 6 months after cardiac rehabilitation: Results from a German multi-centre registry (OutCaRe)

PLOS ONE

Dear Dr. Salzwedel,

We would appreciate receiving your revised manuscript by Mar 30 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Stephanie Prince Ware, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

I thank the authors for their detailed consideration of the points previously raised by myself and the two reviewers. Both original reviewers re-assessed the manuscript and have considerations that remain in order for this manuscript to be considered for publication. Please find their feedback and some points raised by myself below:

-See my previous comment #10 concerning the RTW analysis: I am still not completely clear why those who were not working at baseline were not excluded.

-Previous comment #15 concerning figures 2 and 3: in their description please list all variables in the models.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: My one main concern continues to be that since this included a strong for on an exercise-based intervention, the authors still seem to be confident on their results on changes (or in fact no change in functional capacity) purely based on the submaxmial "exercise training dose" not significantly being changed over time. This represents only a psychological desire for whatever the reason is to who and how the intensity should have been progressed over time as a very fundamental element of any exercise training programme. Unless there is a change in some measured physiological parameter like a reduction in HR for a given HR or other physiological changes demonstrating a change, the authors can not be certain that no change has occurred.

Reviewer #2: I would like to compliment the authors on their revisions. The document has greatly improved. Nevertheless, I still have some remarks that I hope the authors are willing to take into consideration. I mainly feel that the discussion section could be further improved.

Major remarks

1. The introduction greatly improved as compared to the previous version. Nevertheless, I am still not convinced of the added value of investigating predictors of HRQL. The introduction states that there are mainly inconclusive results with regard to return to work with suboptimal outcomes after CR. Lines 85-88 seem to suggest that predictors for HRQL are already known. In line 73 the authors state that return to work influences HRQL, but why do the authors decide to separately investigate predictors of return to work AND HRQL and not just return to work?

2. A clear research aim/goal is missing in the introduction

3. In the statistics sessions the MID is mentioned for part of the patient-reported outcomes, but not for all. How did you deal with the other outcomes, such as the WHO-5, ASKU and Wurzburger list?

4. The discussion sessions has greatly improved with your extra information, but is now quit long. I suggest to shorten this section and also rearrange a bit the paragraphs. Start with discussing the predictors that you found and how you think (based on literature) these predictors could influence RTW/QOL (this last information is for instance missing for heart-focused anxiety). Move the paragraph lines 349-358 to the end and combine this paragraph with lines 382-389. Describe in this paragraph not only what is NOT working for RTW & HRQL, but also give clear suggestions (based on your outcomes) what should be done to improve CR.

5. There seems to be some grammar errors in the paper. I recommend consulting a native speaker/ professional to check the paper.

Smaller remarks

1. In the abstract ‘lifestyle change motivation’ is defined as risk factor management, shouldn’t this be part of patient-reported outcomes?

2. I advise to split lines 42-46 (abstract) in two sentences to increase readability

3. Please make clear in your abstract what predictors were measured at discharge and which predictor concern changes during CR.

4. Could you add a reference to line 73 that return to work influences HRQL?

5. Line 74-77 it is unclear whether outcomes or predictors are discussed (e.g. functional status is a predictors and HRQL an outcome?). I advise to delete this sentence.

6. Please also add to your methods section that you check for multicollinearity.

7. Sex and gender are both used in your document. Please pick one (I prefer sex).

8. I would recommend adding a subheading “Predictors” to page 6 to increase readability.

9. Nice summary at the start of the discussion. Could you shortly add (one sentence) how RTW and QOL were improved at 6 months follow-up?

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Attachment

Submitted filename: Review_R1.docx

Click here for additional data file.^{(14.1KB, docx)}

PLoS One. 2020 May 5;15(5):e0232752. doi: 10.1371/journal.pone.0232752.r004

Author response to Decision Letter 1

19 Mar 2020

We thank the editor and the reviewers for re-reviewing our manuscript. Please find our responses to the comments below. Changes in the manuscript text are highlighted in grey.

Sincerely,

Annett Salzwedel

for the authors of the manuscript.

Editor:

-See my previous comment #10 concerning the RTW analysis: I am still not completely clear why those who were not working at baseline were not excluded.

>>>A large percentage of those undergoing rehabilitation generally work, but have been on sick leave since their cardiac event and the corresponding hospitalization until rehabilitation and beyond. Cardiac rehabilitation in Germany has the mission per law to support RTW for patients in employable age even if they were at sick leave or unemployed before rehab. Our choosen method represents this real world goal of rehabilitation in Germany. We tried to clarify this fact in the methods: Return-to-work was analyzed regardless of the employment status at baseline as it is a goal of CR in Germany to support return-to-work even in patients at sick leave or unemployed before CR. (line 201-203)

-Previous comment #15 concerning figures 2 and 3: in their description please list all variables in the models.

>>>As requested, we now listed all variables in the description of the figures as follows:

The forest plot shows the final model after backward selection. The following variables were taken into account in the starting model (see supplemental material, figure S1-S3): at admission to rehabilitation: sex, smoking, lifestyle change motivation, pension desire, self-assessed occupational prognosis, self-assessed health prognosis, living situation, educational level, rehabilitation referral for chronic disorder, comorbidities (diabetes mellitus, depression, peripheral artery disease, heart failure, diseases of the back and spine, stress at work, stress by major life events, age, systolic/diastolic blood pressure, body mass index, endurance training load, depression (PHQ-9), heart-focused anxiety (HAF-17), well-being (WHO-5), physical/mental component summary in the SF-12, physical/mental health and pain in the IRES-24, self-efficacy (ASKU); at discharge from rehabiliation: Sm smoking, lifestyle change motivation, pension desire, self-assessed occupational prognosis, self-assessed health prognosis; changes during rehabilitation: systolic/diastolic blood pressure, body mass index, endurance training load, depression (PHQ-9), heart-focused anxiety (HAF-17), well-being (WHO-5), physical/mental component summary in the SF-12, physical/mental health and pain in the IRES-24, self-efficacy (ASKU).

>>>In fact, the measured parameters of physical performance (maximum exercise capacity in the bycicle ergometry stress test, endurance training load and 6-min walking distance) showed the largest enhancements during cardiac rehabilitation with p-values <0.001 and standardized effect sizes up to >1 (see table 2) compared with the other measured parameters. Therefore, in our opinion, these parameters of physical performance suggested by experts as indicators of rehabilitation success were adequately confirmed by our study. Moreover, the changes in the endurance training load increase the probability of return-to-work statistically significantly, but to a small extend (OR 1.09 per 5 Watts). The other parameters of physical performance were not taken into account in the multivariable modeling due to the content overlap and less available data.

Major remarks

>>>The relationship between return-to-work and HRQL is probably bi-directional and not causal in only one direction. There are some studies investigating predictors for occupational reintegration and HRQL, but usually not in the same study with the same set of parameters for both outcomes. Moreover, the multitude of parameters that can be modified by CR and in particular patient reported outcomes was not systematically taken into account. We added the following connecting reasoning in line 87-88: Since return-to-work and HRQL after a cardiac event are closely associated, [12] several common predictors can be assumed. (line 87-88)

2. A clear research aim/goal is missing in the introduction

>>>We added the following aim: In the OutCaRe-study we aimed to identify predictors of occupational reintegration and HRQL among the same set of patient-reported outcome measures, clinical parameters, cardiovascular risk factors and physical performance. (line 92-94)

3. In the statistics sessions the MID is mentioned for part of the patient-reported outcomes, but not for all. How did you deal with the other outcomes, such as the WHO-5, ASKU and Wurzburger list?

>>>The use of MID was a complementary method (in addition to statistical significance and determination of effect size) for evaluating changes in patient reported outcomes, which was added as requested by the editor. MIDs are not available for all measures. We modified the regarding sentence in the methods section as follows: In addition, the changes in patient-reported outcome measures (metric variables) during CR were assessed using the minimal important difference (MID) if available. (line 190-191) The MID of WHO-5 is 10% as stated below (line 193-194). Statistical significance and standardized effect size were used for the evaluation of ASKU (no MID available). The Wurzburger screening generates categorical variables, no MID exists.

>>>Thank you for your suggestion. In the first revision, we added several items to the discussion section as requested by the editor and reviewers (e.g. MID, educational level and return to work, HRQL related predictors, multicollinearity, benefit of social workers, missing data). Taking into account these several issues, we rearranged and rewrote the discussion section.

From our point of view, the paragraph discussing the benefit of social workers in CR is placed in a logical order to the paragraph above. Moving the paragraph would take the content out of the context.

In the last paragraph above the limitation section we recommend to provide multiprofessional CR programms including cardiologists, social workers and sports therapists or physiotherapists on the basis of an individualized treatment approach. For this we recommend the consistent implementation of questionnaires assessing subjective health (see conclusion). More recommendations are not supported by our data.

5. There seems to be some grammar errors in the paper. I recommend consulting a native speaker/ professional to check the paper.

>>>The paper has been revised by a professional proof reader.

Smaller remarks

>>>We considered all following remarks.