Employment status before and after open heart valve surgery: A cohort study.

OBJECTIVE: Detachment from the workforce following open heart valve surgery is a burden for the patient and society. The objectives were to examine patterns of employment status at different time points and to investigate factors associated with a lower likelihood of returning to the workforce within six months.
METHODS: A cohort study of patients aged 18-63 undergoing valvular surgery at a Danish tertiary centre from 2013-2017. Return to the workforce was defined as being employed, unemployed (still capable of working) or receiving paid leave of absence. The association between demographic-, clinical characteristics (including a surgical risk evaluation, EuroScore), and return to the workforce were investigated with a multivariable logistic regression model.
RESULTS: In total, 1,395 consecutive patients underwent surgery, 347 were between 18 and 63 years and eligible for inclusion. Of those, 282 were attached to the workforce before surgery and included in the study. At the time of surgery, 79% were on paid sick leave. After six months, 21% of the patients (being part of the workforce before surgery), were still on sick leave. In the regression model, prolonged sick leave prior to surgery (OR 0.43, 95% CI 0.23-0.79) and EuroScore ≥ 2.3 (OR 0.39, 95% CI 0.21-0.74) significantly reduced the likelihood of returning to the workforce.
CONCLUSION: One-fifth of patients in the working-age were on sick leave six months after surgery. Prolonged sick leave prior to surgery and a EuroScore ≥2.3 were associated with a lower likelihood of returning to the workforce.

Introduction

Among patients capable of working and undergoing cardiac surgery, factors influencing sick leave, how to recover and how to return to work are essential parts of the pathway to recovery after discharge and thus, important outcomes to measure [1]. Prolonged sick leave may have individual patient consequences by impacting the ability to resume daily living in the period after surgery. Furthermore, sick leave and prolonged time to return to the workforce after a treatment is a genuine societal problem with substantial economic consequences for the patient as well as society [2].

Return to the workforce is influenced by a person’s health and workability [3], but likewise, other factors might impact the ability to return to the workforce following open heart valve surgery. As the early period after discharge is a seemingly vulnerable period, the patients have an increased risk of experiencing complications and subsequent unplanned readmissions [4-7]. Also, patients experience symptoms of anxiety, depression, reduced quality of life and a changed bodily awareness in the early period after surgery [8-11]. Together, these factors might also influence the return to the workforce after surgery.

Knowledge of specific factors associated with return to the workforce after open heart valve surgery is sparse: Unemployment one year before surgery has been associated with reduced likelihood of returning to the workforce [1], whereas participating in cardiac rehabilitation is not associated with returning to the workforce [12]. Some studies have investigated factors associated with the likelihood of returning to the workforce following coronary artery bypass grafting (CABG) and demonstrated how younger age, male sex, higher educational level, work status before surgery and higher income were associated with returning to the workforce [1, 13, 14]. Also, sick leave status before treatment and job type/strain are known to impact employment status following a heart disease [15]. Whether similar characteristics, including differences in age groups, are associated with returning to the workforce among patients undergoing open heart valve surgery, are currently unknown. This knowledge is vital to identify patients at higher risk and thereby to prevent detachment from the workforce to the extent possible.

Thus, in a population of patients undergoing open heart valve surgery, the objectives of the study were to i) examine patterns of employment status at different time points, including differences among age groups and to ii) investigate demographic and clinical factors associated with a lower likelihood of returning to the workforce within six months after surgery.

Materials and methods

The current study is an exploratory cohort study investigating employment status six months before and after open heart valve surgery.

Participants and setting and recruitment

Patients undergoing open heart valve surgery at a high volume tertiary centre, Odense University Hospital, Denmark, from August 2013 to November 2017 were consecutively included in the cohort study [7]. Open heart valve surgery was defined as one of the following surgical procedures (Nordic/NOMESCO Classification of Surgical Procedures [16]): Aortic (KFCA, KFMA, KFMC, KFMD), Mitral (KFKB, KFKC, KFKD, KDKW) and Tricuspid (KFGC, KFGE).

Excluded were: patients ≥ 64 years (n = 1,027), patients who died during index admission or within 180-days after discharge (n = 12), patients who required endocarditis treatment during the index admission due to prolonged hospitalisation (n = 8) and patients who developed perioperative stroke and received neurological rehabilitation (n = 1) and patients who did not have a Danish civil registration number due to lack of data on work status (n = 2). Thus, the current study was restricted to patients at the working age (between 18 and 63 years) at inclusion time to ensure full follow-up before the patients were offered state pension (at the age of 65 years during the study period).

Data collection

Demographic and clinical data

Demographic and clinical data were obtained from electronic medical records and the Western Denmark Heart Registry (WDHR) [17]. Living status, smoking status, alcohol consumption, body mass index (BMI) and length of stay were obtained from the electronic medical records, whereas the type of surgery, comorbidity and EuroScore II (surgical risk evaluation) were obtained from WDHR. The EuroScore II is a logistic surgical risk evaluation calculated before surgery including age, sex, renal impairment, extracardiac arteriopathy, poor mobility, chronic lung disease, active endocarditis, critical pre-operative state, angina status, recent myocardial infarction, pulmonary hypertension, urgency and weight of the procedure [18].

Sick leave and workforce attachment

In Denmark, all citizens are entitled to social security benefits during sick leave. Data on sick leave and labour marked absence were acquired from the Danish DREAM registry, administered by the Danish Ministry of Employment. The DREAM registry contains information on social security benefits of all individuals who have received social transfer payment and consists of more than 100 codes on benefits reported with weekly status [19, 20]. Only sick leave lasting for more than two weeks is included in the registry, meaning that the status of working citizens based on short-term sick leave is not registered [19, 20].

As all citizens in Denmark have a unique, national civil registration number, the population of the study were matched to the DREAM registry based on the civil registration number ensuring that data on each patient was matched to specific sick leave benefits. Variables in the DREAM registry were grouped based on recommendations and similar study designs [13, 21, 22] related to the workforce attachment into:

On sick leave

Working/part of the workforce (employed, unemployed (but still capable of working based on the coding) or received paid leave of absence and educational grants)

Out of the workforce (patients were considered to be out of the workforce if they were on early retirement of any kind, n = 65)

Employment status six months prior to admission for the open heart valve surgery was determined based on being capable of working during the period. To reduce misclassification, work status was evaluated based on the status of each week before surgery by ensuring that patients who had a short term sick leave period were not grouped as being detached from the workforce [13]. Patients who were part of the workforce for at least two weeks (consecutive) in the period before surgery were included in the study. Also, patients on prolonged sick leave prior to surgery were defined as patients ≥ 2 weeks of sick leave.

Outcomes

Study outcomes included the return to the workforce six months after surgery and employment status (sick leave, part of the workforce or out of the workforce) before and up to six months after open heart valve surgery.

Statistics

Baseline characteristics were presented as mean and standard deviation (SD) or as median and 25th to 75th percentiles (IQR) for continuous data and number and percentage for categorical data. Normality was tested with the Shapiro-Wilks test.

Distribution of sick leave six months before and after surgery were visualised with a probability distribution plot (proportion of the population) for the overall population and divided into age groups. Differences in the proportion of patients on sick leave, part of the workforce and out of the workforce at different timepoints were described.

To investigate factors associated with returning to the workforce within six months after surgery, logistic regression analyses were performed. A multiple logistic regression model was chosen, as only two patients died during the 180-days follow-up. The following variables were included in the model; sex, age groups, surgical procedure, prolonged sick leave before surgery (defined as sick leave ≥ 2 weeks before surgery), EuroScore log II, postoperative atrial fibrillation and length of stay. The included variables were assumed to be associated with the outcome, but with restricted numbers to avoid overfitting the model. Model fit was tested with the Likelihood Ratio.

As a sensitivity analysis, we removed patients receiving educational grants from the overall regression analysis.

A P-value of <0.05 was considered to be statistically significant. SPSS 24 (IBM Corp, Armonk, NY) and R 3.2.2 (R Foundation for Statistical Computing) were used for the analyses.

Ethics approval

The study was approved by the Danish Data Protection Agency (18/19152), the Danish Patient Safety Authority and conformed with the principles outlined in the Declaration of Helsinki [23]. Due to Danish Legislation, signed consent for register-based studies is not needed.

Results

In total, 1,395 patients underwent open heart valve surgery during the period of which 368 patients were between 18 and 63 years, and thus in a working age. Of those, 347 were eligible after the exclusion criteria and withdrawal of patients on early retirement, n = 282 were included in the study (Fig 1). Clinical and demographic characteristics among patients being part of the workforce are summarised in Table 1 and among the total population in S1 Table. The majority of the patients being part of the workforce were men (81%), the median age was 55 years (IQR 49–60), and 26% were living alone at the time of surgery. A total of 124 patients were diagnosed with aortic valve stenosis (44%), and concomitant coronary artery bypass grafting (CABG) was performed in 9%. In addition, prior to the surgery, 42% had a prolonged sick leave period (sick leave ≥ 2 weeks), Table 1.

Fig 1

Patient flowchart.

Flowchart of the patient population.

Table 1

Baseline characteristics of the patients being part of the workforce.

	Part of the workforce (n = 282)
Characteristics
Sex, male, n (%)	229 (81)
Age-groups, n (%)
18–45 years	42 (15)
46–50 years	46 (16)
51–55 years	55 (20)
56–63 years	139 (49)
Living alone, n (%)	72 (26)
Sick leave ≥ 2 weeks before surgery	120 (42)
Pre-operative information
Reduced pulmonary functiona, n (%)	93 (33)
EuroScore II (logistic), median (IQR)	1.15 (0.73–2.44)
EuroScoreII ≥2.3, n (%)	73 (26)
Hypertension (medical treatment)	192 (68)
Estimated glomerular filtration rate ml/min.b, median (IQR)	101 (83–123)
Atrial fibrillation, n (%)	46 (16)
Diabetesc, n (%)	24 (9)
Ejection fraction ≤50%, n (%)	78 (28)
NYHA class ≥3, n (%)	86 (31)
Body Mass Index, median (IQR)	26 (24–30)
Current or former smoker, n (%)	140 (50)
Alcohol intake above national recommendations, n (%)	30 (11)
Primary diagnosis, n (%)
Aortic valve stenosis	124 (44)
Aortic valve regurgitation	72 (26)
Mitral valve stenosis	<5 (1)
Mitral valve regurgitation	83 (29)
Surgical information, n (%)
Type of valve procedured
Aortic valve, biological	37 (13)
Aortic valve, mechanical	152 (54)
Aortic valve, repair	6 (2)
Mitral valve, replacemente	21 (7)
Mitral valve, repair	65 (23)
Concomitant CABG	24 (9)
Post-procedure related, n (%)
Re-operation	15 (5)
Prolonged length of stayf, intensive care unit	28 (10)
Postoperative atrial fibrillation	132 (47)
New-onset postoperative atrial fibrillation	99 (35)
Length of stay
4–7 days	103 (37)
8–12 days	106 (38)
≥13 days	73 (26)

IQR, interquartile range, 25th to 75th quartile. NYHA, New York Heart Association Class.

* The total population of patients between 18–63 years.

a Patients with forced expiratory volume, % ≤80% of predicted value and / or a history of chronic obstructive pulmonary disease.

b Estimated glomerular filtration rate estimated by the Cockcroft-Gault Equation.

c Patients with diabetes; insulin, peroral and non-pharmacological treatment.

d One patient had surgery on the tricuspid valve and are not shown in the table, but included in the analyses.

e Both biological and mechanical mitral valve replacement.

f Admission at the intensive care unit for more than one day.

The pattern of employment status and return to the workforce after open heart valve surgery

The pattern of the employment status (being on sick leave or working) before and after surgery among the total population of patients ≤ 63 years are showed in Fig 2.

Fig 2

The proportional distribution of employment status six months before and after surgery among patients between 18–63 years.

The figure illustrates the status of the patients at different time points before and after surgery. Status of patients who are part of the workforce and patients on sick leave are visualised.

Patterns of sick leave among the patients being part of the workforce (n = 282) varied at different time points, with n = 7 (2%) being on paid sick leave six months before surgery and n = 223 (79%) receiving sick leave benefits at the time of surgery. The proportion of patients being on paid sick leave at the time of surgery varied across the different age groups, with 67% of the patients in the age group from 18–45 years being on sick leave versus 84% of the patients in the age group from 56–63 years (S2 Table).

At six months after surgery, most patients being part of the workforce before surgery had returned to the workforce, although n = 59 (21%) were still on sick leave (S2 Table). The highest proportion of patients who did not return to the workforce were found in the youngest age group (age 18–45) with 29% still being on sick leave six months after surgery (Fig 3 and S2 Table). Also, n = 6 (2%) had left the workforce after six months. Median time on sick leave during the total period was 15 weeks (IQR 8;24).

Fig 3

The proportion of patients being on sick leave at different time points, divided into age groups.

The figure illustrates the status of the patients (on sick leave) at different time points before and after surgery, divided into age groups.

The pattern of the overall employment status (being on sick leave, working or being out of the workforce) before and after surgery among the total population of patients ≤ 63 years are shown in S1 Fig and S3 Table. At the time of surgery, 19% of the total population were out of the workforce (S2 Table).

Factors associated with return to the workforce within six months after discharge

In the multiple logistic regression model, prolonged sick leave prior to surgery (OR 0.43, 95% CI 0.23–0.79) and EuroScore ≥ 2.3 (OR 0.39, 95% CI 0.21–0.74) significantly reduced the likelihood of returning to the workforce (Fig 4).

Fig 4

Factors associated with returning to the workforce.

The figure illustrates the logistic regression model of factors associated with returning to the workforce, with time to return to work (weeks) as the underlying time scale.

The sensitivity analysis where patients receiving educational grants were removed did not change the overall results (S2 Fig).

Discussion

In this cohort study, we investigated return to the workforce and factors associated with returning to the workforce within six months after open heart valve surgery. At the time of surgery, a lower proportion of patients in the youngest age group were on paid sick leave compared to the other age groups. Also, one-fifth of the patients were still on paid sick leave six months after surgery. Prolonged sick leave prior to surgery and a EuroScore ≥ 2.3 significantly reduced the likelihood of returning to the workforce.

Earlier studies have focused on different aspects of return to the workforce following open heart valve surgery: work status before surgery and participation in cardiac rehabilitation and the likelihood of returning to the workforce [1, 12]. Thus, to our knowledge, this is the first study to investigate clinical characteristics and the association with return to the workforce. As expected, the proportion of patients being on paid sick leave reached its maximum level at the time of surgery, but unexpected, fever patients in the younger age group (from 18–45 years) were on paid sick leave at that time point. One possible explanation for this might be due to more patients in the young group being studying/receiving educational grants, as previously demonstrated in a young population of patients with endocarditis [22]. Students in Denmark receive a paid monthly fee (educational grants) by the government, but do not receive paid sick leave. Similarly, more patients in the young group received immigration fees which also meant, that they did not receive paid sick leave. As the total number of patients in the young group is low, the few patients receiving educational grants or immigration fee (14% of the patients in the young group) might have an impact on the number on paid sick leave. Removing these patients did not change the results of the regression analysis.

Interestingly, 17% of the total population did not receive any paid sick leave at the time of the surgery. This does not necessarily imply that the patients had a working status, but only, that they did not receive paid sick leave as part of the national social security system. The results comply with an earlier Danish register-based study by Fonager et al., where 16% of a similar study population did not receive paid sick leave at the time of surgery [1]. Fonager et al. speculate how some patients might be economically funded by their spouses (although this might only be a seemingly low proportion), but also how the registration in the Danish DREAM database with requirements of two weeks sick leave might impact their results [1]. In addition, patients who are self-employed are not necessarily being on sick leave, depending on the status of their business.

At the six months follow-up, we demonstrated that one-fifth of the patients were still on sick leave which is similar to earlier studies among patients undergoing open heart valve surgery or patients undergoing CABG [1, 13].

In the regression analyses, prolonged sick leave prior to surgery and EuroScore ≥2.3 were associated with a lower likelihood of returning to the workforce. Thus, as expected, returning to the workforce was dependent on pre-operative morbidity (seen as prolonged sick leave) and surgical risk. Although tested in a small population in the current study, the results comply with a study among patients undergoing CABG; a high comorbidity burden is associated with a lower likelihood of returning to the workforce [13].

In general, research investigating the pattern of sick leave and factors of return to the workforce after both open heart valve surgery and comparable populations are sparse and needs to be investigated more thoroughly as there are several personal- and economic advantages for both patients and society in returning to the workforce. Similarly, to prevent detachment from the workforce to the extent possible, a better understanding of the associated factors is warranted. Currently, research investigating interventions aiming at increasing the likelihood of returning to the workforce is sparse with current studies lacking statistical power or reporting conflicting results [12, 24].

Strengths and limitations

The use of the DREAM register is both the main strength and a limitation of the study; A strength as it covers all entries on sick leave compensations. As the entries are dependent on employers claiming compensation refund, the incentive for registration is high–ensuring high accuracy of data. The use of the register though, also includes limitations as only sick leave spells of >2 weeks are included. Thus, the register does not include short-term absence, and similarly, the register does not distinguish between full- and part-time sick leave [19, 20]. Similarly, as the register only include persons who receive a paid benefit, it can be difficult to conclude on the status of persons who are not registered. Type of job, job strain (e.g. physical strenuous fields), self-employment and similar knowledge would potentially also be relevant data of the population, but not available through the DREAM registry. This is also a limitation by the use of the DREAM register. The number of available jobs during the study period could potentially have had an impact on returning to work, why the main focus of the study was returning to the workforce. Classification of the groups; sick leave, workforce and out of the workforce could have caused a risk of classification bias but were performed as a method to classify DREAM data as previously described [13, 21, 22]. In addition, the analyses were performed based on a small population which might cause a problem in the regression analyses due to possible lack of statistical power.

Exclusion of patients with endocarditis and patients requiring neurological rehabilitation cause another limitation of the study. Patients with infective endocarditis have a hospital course very different from elective heart valve surgery; a common need for prolonged antibiotic treatment and often complications related to the systemic infection contrary to the surgery. Thus, these patients were excluded. In addition, patients with an extended need for neurological rehabilitation per se precluded a return to the workforce were excluded. Even though the overall number of these patients were small, they could potentially have influenced our results with a lower likelihood of returning to the workforce. Thus, the current results represent a conservative estimate of a surgical population.

The study was also limited to patients between 18–63 years. We acknowledge, though, how several patients above the age of 63 were working and how they could have been included in the study. However, the age limit was set to ensure that no patients received state pension during the follow-up period.

Finally, the follow-up period of six months might be too short a period of follow-up, and with one-fifth of the patients still being on paid sick leave at the time of follow-up, future studies are encouraged to incorporate a more extended period of follow-up.

Conclusion

In this cohort study, one-fifth of patients in the working-age were still on sick leave six months after open heart valve surgery. Age had little influence on the pattern of sick leave, except for the time of surgery. In regression analyses, prolonged sick leave prior to surgery and EuroScore ≥2.3 were associated with a lower likelihood of returning to the workforce. This knowledge might help identify patients at higher risk of detachment from the workforce and might be targeted during rehabilitation. In addition, the study adds to knowledge about what to expect following open heart valve surgery.

Probability distribution of employment status six months before and after surgery among the total population of patients between 18–63 years.

(PDF)

Click here for additional data file.

(PDF)

Click here for additional data file.

Baseline characteristics of the total population (including patients on early retirement).

(PDF)

Click here for additional data file.

Patterns of sick leave and return to the workforce among the population being part of the workforce and divided by age groups.

(PDF)

Click here for additional data file.

Patterns of employment status before and after open heart valve surgery.

(PDF)

Click here for additional data file.

5 Aug 2020

PONE-D-20-20822

Employment status before and after open heart valve surgery: A cohort study

PLOS ONE

Dear Dr. Borregaard,

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.

Please submit your revised manuscript by Sep 19 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're 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.

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). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

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

We look forward to receiving your revised manuscript.

Kind regards,

Corstiaan den Uil

Academic Editor

PLOS ONE

Journal Requirements:

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

1. 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

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study. Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

3. 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.

4. Your ethics statement must appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please also ensure that your ethics statement is included in your manuscript, as the ethics section of your online submission will not be published alongside your manuscript.

[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: Partly

**********

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

Reviewer #1: No

**********

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: Yes

**********

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

**********

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: Thank you for the opportunity to review this interesting manuscript. The research described in this paper seems to be a retrospective cohort study looking at the clinical predictors associated with returning to work after heart valve surgery in patients aged 18 to 63 years. The data used for the research come from hospital records and registry data (secondary data sources). The manuscript is written in a clear English that is easy to understand.

Overall, the study finds that poorer cardiovascular health (based on EuroScore-II score) and a longer hospital stay after the heart valve operation lowers the chance of returning to work. The authors claim that this is the first study to focus on clinical characteristics and return to work. I can believe this, because it is hard to see how this information, which is rather apparent and often a secondary finding of papers looking at rehabilitation interventions, merits research. Nevertheless, I have some comments and questions regarding the manuscript itself and the methods chosen that I hope will be helpful.

Abstract

Line 40: Maybe you could add a very brief description mention EuroScore as clinical characteristic in parentheses. I think a brief description of EuroScore would be helpful as well.

Line 47-48: I think it would be easier to read if the OR for of longer length of hospital stay was written in descending order.

Background

Lines 61-62: Other factors that might impact the ability to return to work after heart surgery are mentioned, without listing them. This is fine, but one factor that seems to be neglected in this paper is the type of work a patient will be returning to. Someone with a physical job may find it harder to return to work than someone with a desk job after heart surgery. This could be mentioned here and must be considered later in the statistical analysis.

Line 79: The second objective of the study is rather vague. The factors being investigated should be named. This seems to be an explorative study, as the objectives are formulated in a way that does not sound as if a concrete hypothesis is being tested. The fact that this is an explorative study should be stated in the paper.

Lines 91-97: Reading the list of exclusions, I got the impression these exclusions were decided while reviewing the medical records and not based on an a priori list of exclusions. Is this the case? Was an a priori study protocol prepared for this research (even one that was not published)?

Line 103: EuroScore is mentioned here very briefly. Later the tables state EuroScore-II. Please describe the EuroScore used, how it is calculated and the meaning of the index values here. Was the log of EuroScore-II really included in the regression models or the binary categories ≥2.3 vs <2.3?

As I understand it, the EuroScore is calculated prior to the operation. Are there any other clinical values, beside atrial fibrillation that describe the health of patients directly after the operation and can be considered?

Lines 116-121: This was the one part of the text I did not understand. I do not see how the three items listed in lines 116-118 line up with the descriptions provided in 119-120. Do items 1 and 2 correspond to the workforce described in lines 119? I think this needs to be rewritten to be clearer.

Who determined if someone was “still capable of working” for the DREAM registry? Is there a legal definition of “capable of working” that can be described here?

Lines 134-135: I do not see the use of doing testing for normality with the Shapiro-Wilks test. The regression model being used, does not require normality.

What is the time scale unit for return to work? Days? Weeks?

Logistic regression would not be my first or second choice for this type of data. Ideally this should be evaluated as time-to-event with either a Cox regression model. This allows for censoring of data, which means people dying or having a perioperative stroke after hospital discharge can be included into the analysis until the time of their death/stroke (at which time they are censored). If the time scale for return to work is not that exact, an alternative would be to calculated incidence rate ratios (IRR) using Poisson-regression and person-months as a time scale.

The longer one is away from work, the harder it may be to return. I think it would be interesting if the time on sick-leave (in days or weeks) just prior to the OP could be included in the analysis as a variable. While this might be a proxy for the severity of the underlying illness, prolonged time away from work before the OP may make it harder to imagine going back afterwards.

How was the model selection conducted?

Results

Line 152: Early retirement: I do not understand why the patients who were already retired before their operation are just now being excluded. Even though they are described as being excluded, they are included in table 2, which is confusing. I do not understand why their exclusion cannot be mentioned already in the methods and why they are not consistently excluded in the results. If you want to see if someone comes out of early retirement post-OP, this should be considered a new research question and considered separately in only the subgroup that was in early retirement pre-OP.

Table 1:

Was there any data on arterial blood pressure/ hypertension for the population that could be included in this table?

Did any of the patients take part in post-OP rehabilitation programs? If so, the prevalence of such rehabilitation should be described, and perhaps considered in the regression models as this might also impact return to work.

The type or field of work conducted pre-OP should be described. At least the proportion working in physically strenuous fields of work should be mentioned. Job strain would also be an interesting factor to consider. High job strain (e.g. demand control model) may also prevent return to work.

Were any of the patients self-employed? Could this partially explain the number of patients not on sick leave at the time of their operation? People who are self-employed tend to return to work sooner. This might also need to be considered in the regression model.

The youngest age range is wide: 18-45 years. I suspect the indications for the OP were different for the youngest patients, while coronary artery disease was probably more common in older patients. Can some information regarding the indications in the different age groups be provided?

Younger patients who were also students seem make it difficult to interpret the results for this youngest age group, as they would not be categorized as being on sick leave. Could students or the youngest patients (<30 years) be excluded from the analysis to check this? This could either replace the main analysis or be a sensitivity analysis.

Figures 2: This is an interesting depiction of the proportion on sick leave or working. Would it also be possible to depict the proportion of patients who go into early retirement post-OP?

Figure 3: The stratification of Figure 2 into the different age groups takes up a lot of space and adds little to the paper. I find it hard to compare the groups. think it might be nicer if the curves could be included in one single graphic. Maybe as Kaplan-Meier curves post-OP.

Figure 4: Although age is otherwise considered in age-groups, age seems to have been included as a continuous variable in the regression model. While this is legitimate, I think it would make it add to the interpretation of the descriptive analysis of return-to-work if the same age-categories were used in the regression model. I think suspect the chance of returning to work might also be significant for some age-groups, while the increased chance of returning to work is not significant for an increase in a single year of age. Also, increased chance of returning to work might not be linear for age. It might be lower for younger age groups, higher in the middle and lower again for the oldest age group. Using the age categories will show if this is the case.

Line 244 (and Figure 4): I do not think the term “co-morbidity” is the best descriptor to describe the EuroScore and post-OP atrial fibrillation. A better term for these indicators of cardiovascular health is needed. A co-morbidity would be additional diseases, such as diabetes or kidney disorders. Why were these not considered in the model?

**********

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

[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.]

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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

14 Sep 2020

Comment

Abstract

Line 40: Maybe you could add a very brief description mention EuroScore as clinical characteristic in parentheses. I think a brief description of EuroScore would be helpful as well.

Line 47-48: I think it would be easier to read if the OR for of longer length of hospital stay was written in descending order.

Answer

Thank you for the comments. We have changed the abstract according to the reviewer suggestion, but with the new results (see below)

Comment

Background

Lines 61-62: Other factors that might impact the ability to return to work after heart surgery are mentioned, without listing them. This is fine, but one factor that seems to be neglected in this paper is the type of work a patient will be returning to. Someone with a physical job may find it harder to return to work than someone with a desk job after heart surgery. This could be mentioned here and must be considered later in the statistical analysis.

Answer

Thank you to the expert reviewer for this comment. It is both an excellent and highly relevant suggestion. We have added a line about this in the background section. Unfortunately, the DREAM database used in the current study does not contain information on job type, why this data are not available.

Knowledge of specific factors associated with return to the workforce after open heart valve surgery is sparse: Unemployment one year before surgery has been associated with reduced likelihood of returning to the workforce [1], whereas participating in cardiac rehabilitation is not associated with returning to the workforce [12]. Some studies have investigated factors associated with the likelihood of returning to the workforce following coronary artery bypass grafting (CABG) and demonstrated how younger age, male sex, higher educational level, work status before surgery and higher income were associated with returning to the workforce [1, 13, 14]. Also, sick leave status before treatment and job type/strain are known to impact employment status following a heart disease [15]. Whether similar characteristics, including differences in age groups, are associated with returning to the workforce among patients undergoing open heart valve surgery, are currently unknown. This knowledge is important to identify patients at higher risk and thereby to prevent detachment from the workforce to the extent possible.

Furthermore, we have elaborated in the limitations section. See below.

Comment

Line 79: The second objective of the study is rather vague. The factors being investigated should be named. This seems to be an explorative study, as the objectives are formulated in a way that does not sound as if a concrete hypothesis is being tested. The fact that this is an explorative study should be stated in the paper.

Answer

Thank you for pointing this out – a very good perspective. Given the retrospective design based on national registries, no causalities can be demonstrated, and as the expert reviewer points out, the study should be interpreted as an explorative and instead hypothesis-generating. We have added this in the methods section

Comment

Lines 91-97: Reading the list of exclusions, I got the impression these exclusions were decided while reviewing the medical records and not based on an a priori list of exclusions. Is this the case? Was an a priori study protocol prepared for this research (even one that was not published)?

Answer

The population was pre-defined and based on the following study (ClinicalTrials.gov NCT03053778). Thus, the exclusion criteria were given.

The current study was designed after the original study, but using the same cohort of patients and linked to register-based data.

Comment

Line 103: EuroScore is mentioned here very briefly. Later the tables state EuroScore-II. Please describe the EuroScore used, how it is calculated and the meaning of the index values here. Was the log of EuroScore-II really included in the regression models or the binary categories ≥2.3 vs <2.3?

As I understand it, the EuroScore is calculated prior to the operation. Are there any other clinical values, beside atrial fibrillation that describe the health of patients directly after the operation and can be considered?

Answer

Thank you to the expert reviewer for this comment. We have elaborated on the section about the EuroScore II in the methods section.

The EuroScore II is a logistic surgical risk evaluation calculated before surgery. We used the binary categories of this score in the regression models.

The text has been changed accordingly;

Demographic and clinical data

Demographic and clinical data were obtained from electronic medical records and the Western Denmark Heart Registry (WDHR) (16). Living status, smoking status, alcohol consumption, body mass index (BMI) and length of stay were obtained from the electronic medical records, whereas the type of surgery, co-morbidity and EuroScore II (surgical risk evaluation) were obtained from WDHR. The EuroScore II is a logistic surgical risk evaluation calculated before surgery including age, sex, renal impairment, extracardiac arteriopathy, poor mobility, chronic lung disease, active endocarditis, critical preoperative state, angina status, recent myocardial infarction, pulmonary hypertension, urgency and weight of the procedure (17).

As the EuroScore captures several other clinical variables / risk factors, we believed that this covers many of the suspected variables to be assumed related to our outcome. Also, due to lack of statistical power, we omitted further co-variates in the regression model. Thus, we restricted the number of variables in the multivariable models to avoid overfitting of models. This has been added to the statistics section.

Comment

Lines 116-121: This was the one part of the text I did not understand. I do not see how the three items listed in lines 116-118 line up with the descriptions provided in 119-120. Do items 1 and 2 correspond to the workforce described in lines 119? I think this needs to be rewritten to be clearer.

Who determined if someone was “still capable of working” for the DREAM registry? Is there a legal definition of “capable of working” that can be described here?

Answer

Thank you to the expert reviewer for pointing this out. We have re-written the section, so hopefully, it is now more transparent and easier to follow:

Variables in the DREAM registry were grouped based on recommendations and similar study designs related to the workforce attachment into:

1. On sick leave

2. Working/part of the workforce (employed, unemployed (but still capable of working based on the coding) or received paid leave of absence and educational grants)

3. Out of the workforce (patients were considered to be out of the workforce if they were on early retirement of any kind, n=65)

Comment

Lines 134-135: I do not see the use of doing testing for normality with the Shapiro-Wilks test. The regression model being used, does not require normality.

Answer

The Shapiro-Wilks test was used to test for normality on continuous variables reported in the baseline table, and not, as the expert reviewer states, for the regression model.

Comment

What is the time scale unit for return to work? Days? Weeks?

Answer

The time scale used for return to work was weeks, as the DREAM database includes changes in employment status weekly. This has not been highlighted.

Comment

Logistic regression would not be my first or second choice for this type of data. Ideally this should be evaluated as time-to-event with either a Cox regression model. This allows for censoring of data, which means people dying or having a perioperative stroke after hospital discharge can be included into the analysis until the time of their death/stroke (at which time they are censored). If the time scale for return to work is not that exact, an alternative would be to calculated incidence rate ratios (IRR) using Poisson-regression and person-months as a time scale.

Answer

Thank you to the expert reviewer for letting us explain our choice of regression model.

As no patients died during the six-month follow-up (among patients being alive at discharge) and only a few patients changed their status from the workforce to out of the workforce, we did not find the Cox proportional the appropriate choice. This was discussed with the statistician in the group, and it was investigated whether there was a censoring problem (not a problem, a Cox did not change any results). Thus, the logistic regression model was chosen as the most appropriate model.

Comment

The longer one is away from work, the harder it may be to return. I think it would be interesting if the time on sick-leave (in days or weeks) just prior to the OP could be included in the analysis as a variable. While this might be a proxy for the severity of the underlying illness, prolonged time away from work before the OP may make it harder to imagine going back afterwards

Answer

Thank you for pointing this out. This is a great suggestion. We have incorporated patients with prolonged sick leave prior to surgery in the overall regression model. This has been added to the statistics section and results.

Comment

Results

Line 152: Early retirement: I do not understand why the patients who were already retired before their operation are just now being excluded. Even though they are described as being excluded, they are included in table 2, which is confusing. I do not understand why their exclusion cannot be mentioned already in the methods and why they are not consistently excluded in the results. If you want to see if someone comes out of early retirement post-OP, this should be considered a new research question and considered separately in only the subgroup that was in early retirement pre-OP

Answer

This is a very interesting comment, and we acknowledge how this can be confusing.

To accommodate the reviewer comment, we have moved table 2 to supplementary materials S3 Table.

Also, we have re-structured the results sections to avoid further confusion.

Comment

Table 1:

Was there any data on arterial blood pressure/ hypertension for the population that could be included in this Table?

Did any of the patients take part in post-OP rehabilitation programs? If so, the prevalence of such rehabilitation should be described, and perhaps considered in the regression models as this might also impact return to work.

The type or field of work conducted pre-OP should be described. At least the proportion working in physically strenuous fields of work should be mentioned. Job strain would also be an interesting factor to consider. High job strain (e.g. demand control model) may also prevent return to work.

Answer

We have added information about hypertension (a binary variable from our registry) to the Table.

Unfortunately, we do not have information regarding participation in rehabilitation post OP, as this is only a requirement to report among patients with ischemic heart disease.

Also, an excellent comment about the type of work / job strain among the patients as seen above. This is, unfortunately, not information we have in our registry of employment status (DREAM). Instead, we have added something about this in the limitations section:

The use of the DREAM register is both the main strength and a limitation of the study; A strength as it covers all entries on sick leave compensations. As the entries are dependent on employers claiming compensation refund, the incentive for registration is high – ensuring high accuracy of data. The use of the register though, also includes limitations as only sick leave spells of >2 weeks are included. Thus, the register does not include short-term absence, and similarly, the register does not distinguish between full- and part-time sick leave [19, 20]. Similarly, as the register only include persons who receive a paid benefit, it can be difficult to conclude on the status of persons who are not registered. Type of job, job strain (e.g. physical strenuous fields), self-employment and similar knowledge would potentially also be relevant data of the population, but not available through the DREAM registry. This is also a limitation by the use of the DREAM register. The number of available jobs during the study period could potentially have had an impact on returning to work, why the main focus of the study was returning to the workforce. Classification of the groups; sick leave, workforce and out of the workforce could have caused a risk of classification bias but were performed as a method to classify DREAM data as previously described [13, 21, 22].

Comment

Were any of the patients self-employed? Could this partially explain the number of patients not on sick leave at the time of their operation? People who are self-employed tend to return to work sooner. This might also need to be considered in the regression model.

Answer

See the answer above. This is not available information. Instead, we have added something about that in the limitations section.

Comment

The youngest age range is wide: 18-45 years. I suspect the indications for the OP were different for the youngest patients, while coronary artery disease was probably more common in older patients. Can some information regarding the indications in the different age groups be provided?

Younger patients who were also students seem make it difficult to interpret the results for this youngest age group, as they would not be categorized as being on sick leave. Could students or the youngest patients (<30 years) be excluded from the analysis to check this? This could either replace the main analysis or be a sensitivity analysis

Answer

Thank you for letting us explain this more clearly.

We acknowledge the perspective of the reviewer regarding surgical indications among different age groups – and yes, CAD were more common among the elderly patients compared to the younger groups. This perspective is partly captured in the EuroScore (weight of the surgery and recent MI).

The age groups were chosen to capture differences in the groups, and yes, although the range 18-45 is wide, the group size is similar to the next two.

We have performed a sensitivity analysis, as suggested by the reviewer, where we have investigated the results of the regression analysis without patients on educational grants.

This has been added to both the method section, the results section and a supplementary figure 2.

Comment

Figures 2: This is an interesting depiction of the proportion on sick leave or working. Would it also be possible to depict the proportion of patients who go into early retirement post-OP?

Answer

We agree with the expert reviewer. This would be valuable information. But, as only a few patients go “out of workforce” during the follow-up period (see answer regarding the choice of regression model) it would be problematic to include those patients in the figure due to the Danish Law of sensitive personal data.

Instead, we have added an overall figure of the total population, including follow-up as supplementary material (S1 Fig)

Comment

Figure 3: The stratification of Figure 2 into the different age groups takes up a lot of space and adds little to the paper. I find it hard to compare the groups. think it might be nicer if the curves could be included in one single graphic. Maybe as Kaplan-Meier curves post-OP

Answer

Thank you to the expert reviewer for the great suggestion of a combined figure. As we want to capture the difference among age groups before surgery as well, we have combined the groups in a figure illustrating proportions of patients being on sick leave at different time points divided in age groups (new fig 3)

Comment

Figure 4: Although age is otherwise considered in age-groups, age seems to have been included as a continuous variable in the regression model. While this is legitimate, I think it would make it add to the interpretation of the descriptive analysis of return-to-work if the same age-categories were used in the regression model. I think suspect the chance of returning to work might also be significant for some age-groups, while the increased chance of returning to work is not significant for an increase in a single year of age. Also, increased chance of returning to work might not be linear for age. It might be lower for younger age groups, higher in the middle and lower again for the oldest age group. Using the age categories will show if this is the case.

Answer

Thank you for letting us explain this. We acknowledge how age divided into age group would give us a more accurate picture of the changes in odds among patients at different age groups.

Also, the reviewer is right; the change of returning to work is not linear with age.

We have discussed the issue with the statistician of the group and changed the variables in the model: age is included as groups, as suggested, and instead, length of stay is included as a continuous variable (where the increased risk is linear). The text has been changed throughout the manuscript accordingly.

Comment

Line 244 (and Figure 4): I do not think the term “co-morbidity” is the best descriptor to describe the EuroScore and post-OP atrial fibrillation. A better term for these indicators of cardiovascular health is needed. A co-morbidity would be additional diseases, such as diabetes or kidney disorders. Why were these not considered in the model?

Answer

Thank you to the expert reviewer for letting us change the wording of co-morbidity and the two variables. This is now changed in both the text and the figure.

As stated earlier in the review, we wanted to avoid over-fitting the model, why we included EuroScore as a combined marker of the surgical risk, instead of including the specific variables separately.

23 Sep 2020

Employment status before and after open heart valve surgery: A cohort study

PONE-D-20-20822R1

Dear Dr. Borregaard,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Corstiaan den Uil

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

28 Sep 2020

PONE-D-20-20822R1

Employment status before and after open heart valve surgery: A cohort study

Dear Dr. Borregaard:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Corstiaan den Uil

Academic Editor

PLOS ONE