Return to Work and Risk of Subsequent Detachment From Employment After Myocardial Infarction: Insights From Danish Nationwide Registries
Background Limited data are available on return to work and subsequent detachment from employment after admission for myocardial infarction (MI).
Methods and Results Using individual‐level linkage of data from nationwide registries, we identified patients of working age (30–65 years) discharged after first‐time MI in the period 1997 to 2012, who were employed before admission. To assess the cumulative incidence of return to work and detachment from employment, the Aalen Johansen estimator was used. Incidences were compared with population controls matched on age and sex. Logistic regression was applied to estimate odds ratios for associations between detachment from employment and age, sex, comorbidities, income, and education level. Of 39 296 patients of working age discharged after first‐time MI, 22 394 (56.9%) were employed before admission. Within 1 year 91.1% (95% confidence interval [CI], 90.7%–91.5%) of subjects had returned to work, but 1 year after their return 24.2% (95% CI, 23.6%–24.8%) were detached from employment and received social benefits. Detachment rates were highest in patients aged 60 to 65 and 30 to 39 years, and significantly higher in patients with MI compared with population controls. Predictors of detachment were heart failure (odds ratio 1.20 [95% CI, 1.08–1.34]), diabetes mellitus (odds ratio 1.13 [95% CI, 1.01–1.25]), and depression (odds ratio 1.77 [95% CI, 1.55–2.01]). High education level and high income favored continued employment.
Conclusions Despite that most patients returned to work after first‐time MI, about 1 in 4 was detached from employment after 1 year. Several factors including age and lower socioeconomic status were associated with risk of detachment from employment.
What Is New?
This nationwide study reports on detachment from employment after return to work in patients with myocardial infarction.
Of 20 415 patients who returned to work, 4938 (24.2%) were detached from employment within 1 year and supported by social benefits.
Important risk factors of detachment from employment were younger age (30–39 years) and low socioeconomic status.
What Are the Clinical Implications?
The findings in this study suggest that a rehabilitation strategy with focus on employment maintenance is warranted.
Through decades, prognosis after myocardial infarction (MI) has improved as new treatment regimens for acute management, primary and secondary prevention have evolved.1, 2, 3 Although fewer post‐MI complications, such as recurrent MI, heart failure, and death, are important benchmarks for the quality of care, returning to work and being able to remain employed are similarly important markers of functional status, which are associated with individual self‐esteem and societal costs. For example, evidence has indicated that quality of life is poorer in unemployed subjects and the risk of depression is greater among these individuals.4, 5, 6 While previous studies have focused on return to work after MI, with return rates from 57% to 90% within 1 year, continuation of employment following return to work remains unclear.7, 8, 9, 10, 11, 12 Also, a single study of patients who underwent percutaneous coronary intervention (PCI) suggested that return to work is an insufficient measure of the actual employment status, because a considerable number of subjects become detached from employment relatively quickly (ie, 76% of patients returned to work after PCI, but only 60% had no recurrent sick‐leave during 1 year of follow‐up).13
An improved understanding of return to work and the subsequent detachment from employment after MI can contribute to the development of strategies to promote return and maintenance of work and thereby increase patient's quality of life and reduce societal costs. Therefore, we investigated return to work and subsequent detachment from employment after MI in a nationwide cohort during a 16‐year period and assessed risk factors associated with not returning to work and/or subsequent detachment from employment.
Study Design and Study Population
The study was a nationwide retrospective cohort study comprising patients aged 30 to 65 years with a first‐time primary discharge diagnosis of MI (International Classification of Diseases 10th revision code I21) in the period between January 1, 1997 and December 31, 2012. Patients with a previous discharge diagnosis of MI were excluded. Only patients who were employed for at least 3 consecutive weeks in a 5‐week period before the day of admission with MI were included. The patients were included 30 days after admission.
In Denmark, all 5.6 million citizens are covered by publicly financed national health insurance, guaranteeing free access to healthcare services at all times. All residents hold a unique and permanent personal identification number that allows linkage between the administrative registers. The Danish Civil Registration System holds information on birth, sex, and death status. The Danish National Patient Registry holds information on all admissions since 1978.14 All admissions to hospitals and out of hospital contacts are registered by 1 primary diagnosis, and supplementary secondary diagnoses if appropriate, according to the International Classification of Diseases. The MI diagnosis has previously been validated in this registry with a sensitivity of 97%.15 Surgical and interventional procedures are registered in the Nordic Medical Statistics Committees Classification of Surgical Procedures.
The DREAM database was used to identify patients who received social benefits and employment was defined as being self‐supportive, which has been validated with a positive predictive value of 98.2%.16 All residents in Denmark who are not working (eg, unemployed, on sick leave, on maternity leave, or eligible for early retirement or disability pension because of severe impairments) are entitled to government‐financed social benefits. Since 1991, all social benefits are registered on a weekly basis in the DREAM database, making this database suitable for studies of employment history and follow‐up.16
Redemption of drug prescriptions was identified using the Danish Register of Medicinal Product Statistics that holds information on all prescriptions dispensed from pharmacies in Denmark since 1995, and the registry classifies each drug according to the Anatomic Therapeutical Chemical classification. The Danish healthcare system provides partial reimbursement of drug expenses, which ensures a high validity of the registry.17 Information on education level and personal income was available on an annual basis and was provided by Statistics Denmark, and we indexed income on that of 2009.18, 19
The study outcomes were return to work within 1 year after admission with first‐time MI and detachment from employment following return to work, respectively.
In the multivariable model, the covariates were selected before analysis and we included those that were considered most relevant based on current literature (ie, sex, age, living alone, baseline income and educational level, comorbidities [heart failure, arrhythmia, diabetes mellitus, cerebrovascular disease, chronic kidney disease, chronic obstructive pulmonary disease, and depression]), invasive coronary procedures (coronary angiography, PCI, and coronary artery bypass grafting), and time period of index admission.20, 21, 22, 23 We examined the comorbidity status based on discharge‐ and outpatient diagnoses 5 years before the date of inclusion. Use of medication was investigated 6 months before date of inclusion. Diabetes mellitus and depression were either defined from International Classification of Diseases codes from the National Patient Registry or use of antidiabetic medication or antidepressant, respectively. International Classification of Diseases‐ and Anatomic Therapeutical Chemical‐codes used in the study are found in Table S1.
Return to work.
Patients were included 30 days after admission. To assess the cumulative incidence rates of return to work, the Aalen‐Johansen estimator with competing risk of death was used.24 Return to work was defined as being self‐supportive for at least 1 week, and in a sensitivity analysis, patients who returned to work and maintained employment for at least 3 months were considered. Logistic regression was applied to estimate the odds ratios (ORs) with 95% confidence intervals (CIs) for risk factors associated with not returning to work, and patients who died during follow‐up were considered as not returning to work in this model. Estimates for age and sex were adjusted for comorbidities (heart failure, arrhythmia, diabetes mellitus, cerebrovascular disease, chronic kidney disease, chronic obstructive pulmonary disease, and depression), living alone, invasive coronary procedures (modeled as a categorical variable with 4 levels: no procedure, coronary angiography, PCI, coronary artery bypass grafting), education level, income, and time periods of index admission. Other estimates in the same figure were adjusted for the same variables, but because of interaction between age and numerous of the other variables, age was modeled as a restricted cubic spline, and furthermore income was stratified on sex because of interactions between these variables. Also, this analysis was repeated with return to work considered as maintenance of employment for at least 3 months.
Detachment from employment.
In analyses of detachment from employment, only patients who returned to work were included. Information on age, comorbidity, medication, invasive procedures, living alone, education, and income were updated until the date of return to work. Detachment from employment was assessed by cumulative incidence using Aalen‐Johansen estimator with competing risk of death, and detachment was defined as receiving social benefits for at least 1 or 3 months, respectively. Additionally, the analyses were stratified by age, using the age intervals 30 to 39, 40 to 49, 50 to 59, and 60 to 65 years. Logistic regression was applied to estimate ORs and 95% CIs for risk factors associated with detachment from employment. The logistic regression model presenting risk factors associated with detachment was based on detachment periods of at least 1 month and the model was adjusted in the same manner as the one for return to work as the same interactions were found. In the logistic regression models, patients who were unable to complete follow‐up because of death were considered as detached from employment. This method was feasible because of the very low mortality of patients in the study.
The rate of detachment for post‐MI patients was compared with 1:5 matched controls from the employed general Danish population. The controls had no prior history of MI and were matched by year of birth and sex. To test for differences in detachment, Aalen‐Johansen estimator rates between patients with MI and the matched population, Fine‐Gray models were applied.25 The percentages of patients receiving different social benefits (sick leave, disability pensions, subsidized jobs, unemployment benefits, early retirements, pensions, student grants, and parental leave) at the time of detachment were calculated and compared with controls using conditional logistic regression with age and sex stratas.
Data management and statistical analyses were performed using SAS (version 9.4 for Windows, SAS Institute, Cary, NC) and R (version 3.2.3 for Windows, R Foundation for Statistical Computing).
This study was approved by the Danish Data Protection Agency (local reference No. 2007‐58‐0015/GEH‐2014‐014 I‐suite 02732). Informed consent and ethical approval is not required for retrospective registry‐based studies in Denmark.
Patients and Characteristics
Of 39 296 patients of working age (30–65 years), 22 394 (56.9%) patients were employed before admission with first‐time MI (Figure S1). In the study population, a total of 18 120 (80.9%) were men and the median age was 55 (interquartile range 49–59) years (Table 1). The most prevalent comorbidities were heart failure (8.9%), arrhythmia (7.3%), and diabetes mellitus (9.3%).
Return to Work
Within 1 month, 9329 (41.7% [95% CI, 41.0%–42.3%]) of the post‐MI patients had returned to work and after 1 year this number was 20 415 (91.1% [95% CI, 90.7%–91.5%]), whereas 92 (0.4% [95% CI, 0.3–0.5]) died (Figure 1). When 3 months of employment was considered, 19 369 (86.4% [95% CI, 86.0%–86.9%]) returned to work.
Risk factors associated with not returning to work were female sex (OR 1.43 [95% CI, 1.27–1.60]), heart failure (OR 2.36 [95% CI, 2.07–2.69]), arrhythmia (OR 1.67 [95% CI, 1.44–1.94]), cerebrovascular disease (OR 1.83 [95% CI, 1.44–2.31]), and chronic kidney disease (OR 1.46 [95% CI, 1.03–2.04], Figure 2). Conversely, predictors that favored return to work were high education level, and high income favored return to work for men, but not for women (Figure 2). When return to work was defined as at least 3 months of employment, age 60 to 65 years was no longer associated with return to work, while depression and MI during calendar years 2009 to 2012 compared with 1997 to 2000 were associated with not returning to work (Figure S2).
Detachment From Employment
Of 20 415 patients who had returned to work, 5169 (25.3%) were detached from employment within 1 year (Table 2), of which 4938 (24.2% [95% CI, 23.6%–24.8%]) received social benefits and 231 (1.1% [95% CI, 1.0%–1.3%]) had died (Figure 3). Age‐stratified analyses showed that detachment was most pronounced in patients aged 60 to 65 years, followed by patients aged 30 to 39 years (Figure 4). The post‐MI patients had a higher comorbidity burden and lower education level than the population controls (Table 3), and the rate of detachment was significantly higher in post‐MI patients (all P<0.001; Figure 5).
In adjusted analyses, patients aged 40 to 49 years had a lower risk of detachment than all the other age groups (Figure 6). The youngest age group (30–39 years) had a higher risk of detachment (OR 1.26 [95% CI, 1.06–1.48]) as did patients aged 60 to 65 years (OR 1.84 [95% CI, 1.71–1.99]) compared with patients aged 50 to 59 years. Other predictors for detachment were heart failure (OR 1.20 [95% CI, 1.08–1.34]), diabetes mellitus (OR 1.13 [95% CI, 1.01–1.25]), arrhythmia (OR 1.17 [95% CI, 1.04–1.32]), cerebrovascular disease (OR 1.37 [95% CI, 1.13–1.66], chronic kidney disease (OR 2.04 [95% CI, 1.55–2.68]), and depression (OR 1.77 [95% CI, 1.55–2.01]). Also, a high education level and high income favored maintenance of employment (Figure 6).
Among post‐MI patients, 74.6% were still employed 1 year after employment, while this number was 91.4% for controls (Table 4). At the time of detachment, 10.9% of post‐MI patients received sick leave benefits and 5.7% received unemployment benefits versus 2.4% and 1.4% of the controls, respectively. Detachment because of early retirement and pension was higher in post‐MI patients (4.6% and 1.8%, respectively) than in controls (2.7% and 1.2%, both P<0.01).
In this nationwide register‐based study, we found that 91.1% of the patients employed before admission with first‐time MI returned to work within 1 year. Despite this high reemployment rate, 24.2% of the post‐MI patients were detached from employment and received social benefits within 1 year after return to work. Detachment was associated with being in the youngest (30–39 years) or oldest (60–65 years) age group, heart failure, arrhythmia, and depression, respectively, while high income and high education level favored maintenance of employment. The rate of detachment for post‐MI patients was almost 3‐fold higher than the population controls. These findings suggest that post‐MI patients with these characteristics including excess comorbidities and poorer socioeconomic status should receive increased focus on cardiac rehabilitation aimed at reducing detachment from employment.
In previous studies of return to work after MI, return to work rates varied from 57% to 90%.7, 8, 9, 10, 11, 12 In ST‐segment elevation MI patients, rates varied between 76% and 93%.26, 27 The potentially slightly higher return to work rates observed in our study could be caused, in part, by our register‐based design that minimized loss to follow‐up. Furthermore, compared with other countries, all Danish citizens have the possibility of a paid sick leave, which may contribute to resumption of work.28 High income and high education level were predictors of both returning to work and continuous employment. This finding is in accordance with previous studies showing that white collar workers with higher salaries were more likely to return to work than blue collar workers.12 Furthermore, individuals who are employed in low income jobs generally often have lower decision latitudes at work, which is a well‐described stressor, and also a risk factor of not returning to work and poor maintenance of employment.12, 29, 30
Patients aged 30 to 39 years also had a high risk of detachment from employment compared with those aged 50 to 59 years. The underlying explanations are not clear, although it is well described that younger patients with MI differ from their older counterparts, with a higher burden of smoking and high BMI, potentially indicating a poorer general lifestyle.31, 32 Furthermore, the employer's willingness to accommodate individual needs may be greater for employees who have senior positions and longer previous employment at the workplace. This result is, however, alarming because the youngest group of patients has many more potential years left on the labor market and since unemployment is an individual risk factor for poor quality of life and subsequent depression, this again entails a worsened prognosis.4, 5, 6, 33, 34
Of note, depression before MI is indicative of mental vulnerability and may reinforce, for example, augmented depression or anxiety symptoms leading to increased risk of detachment from employment as observed in the present study.
We found that women were less likely to return to work compared with men, which is in accordance with very recent findings from the United States, where this result, however, was not sustained after adjustments.11 Another US report from 1992 found that single living was a risk factor for poorer outcomes after MI,35 but this was only found in our unadjusted analyses, indicating that these patients may have a higher burden of comorbidities and lower socioeconomic status.
Remarkably, although in Denmark all citizens have access to both education and social benefits financed by the Danish welfare system, socioeconomic factors remained predictors of employment status and follow‐up and these results are likely to be even more significant in countries with more societal and economic inequalities.
While our work largely confirms prior knowledge about return to work after MI, the notion that post‐MI patients frequently become detached from employment shortly after re‐employment is new. This observation emphasizes that return to work may not be a valid measure of successful recovery of working capacity. While rehabilitation strategies generally focus on elderly and vulnerable cases, younger and more resourceful patients (eg, patients who are employed) may not be invited or may decline or drop out from such programs.36 Indeed, socially vulnerable patients with low socioeconomic status are less likely to enroll in these rehabilitation programs.37 This may be reflected in our findings in that patients with lower socioeconomic status are less likely to return to work and more likely to become detached from employment. Thus, our novel findings on a nationwide level suggest that a rehabilitation strategy with focus on employment maintenance is warranted, in particular for patients aged 30 to 39 years and those of lower socioeconomic status who showed a disconcerting trend for becoming detached from employment.
Strengths and Limitations
Strengths of this current study include the following: (1) the nationwide coverage of patients, minimizing selection bias on the basis of region; (2) all citizens had free access to standardized health care without insurance requirements and the Danish welfare system with free healthcare services, partial reimbursement of medication, and paid sick leave would appear to provide favorable conditions for work resumption following MI28, 38; (3) the nationwide register‐based design enabled complete follow‐up (until death, emigration, or end‐of‐study) and determination of comorbidities and drug treatment, which minimized selection bias, recall bias, and misclassification, respectively.
The study also had some limitations including the following: (1) it is observational, and the results represent associations and not necessarily causations; (2) important factors for both return to work and long‐term outcomes such as job satisfaction, job type, health‐ and job‐related quality of life, and physical capacity were not measured10, 12, 34, 39, 40, 41, 42; (3) other unmeasured potential confounders included, for example, smoking, drinking habits, physical activity levels, body mass index, and left ventricular ejection fraction; (4) the actual reason for not returning to work or becoming detached from employment was also unknown and is not possible to address in the registries, although we did find that these patients most often received sick leave benefits. Patients who were close to 65 years of age at time of admission may have retired because of eligibility for early retirement and pension and not because of poor functional status, but compared with the matched controls, the post‐MI patients still had higher retirement rates; (5) the standard of care changed during the relatively long study period, (eg, troponins measurements changed the definition of MI in 2000/2001 and primary PCI was implemented in Denmark in 2003 after the benefits were demonstrated in a national randomized study).43 However, excluding patients admitted during the first period (before 2001) did not change the results (Figures S3 and S4). Also, MI was evaluated as 1 entity because discrimination between MI subtypes was not valid in the registries; (6) the legislation on sick leave became more restrictive during the study period (eg, the period before the employer received reimbursement from the state because of employee's sick leave changed from 15 days to 21 days in 2008, and to 30 days in 2012). To address these changes, we investigated detachment from the workplace with duration of at least 30 days.
In this nationwide register‐based study of patients discharged after first‐time incident MI who were employed before hospital admission, we found that 91% returned to work, but that subsequently 24% of patients were detached from employment within the first year, corresponding to a 3‐fold higher detachment rate compared with population controls. Younger age (30–39 years) and lower socioeconomic status were among important predictors of detachment from employment. Not being part of the workforce represents a considerable individual and societal burden, and the results may inform new strategies for maintenance of work in post‐MI patients.
Sources of Funding
This work was supported by the Danish Agency for Science, Technology and Innovation and by the Danish Council for Strategic Research (EDITORS: Eastern Denmark Initiative to Improve Revascularisation Strategies, grant 09‐066994), Helsefonden, and the Danish Heart Foundation. None of these institutions had any influence on the design and conduct of the study; management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript for submission.
Smedegaard is supported by the Danish Agency for Science, Technology and Innovation, the Danish Council for Strategic Research, Helsefonden, and the Danish Heart Foundation. Numé is supported by P. Carl Petersen Foundation and Helsefonden. Kragholm reports having received grants from The Laerdal Foundation and The Danish Heart Foundation, as well as having received speaker's honorarium from Novartis, outside the submitted work. Gislason is supported by an unrestricted clinical research scholarship from the Novo Nordisk Foundation and has received research grants from Bayer, Pfizer, AstraZeneca, Boehringer‐Ingelheim, and Bristol Meyer Squibb (BMS) and speaker honoraria from Pfizer, AstraZeneca, and BMS. Hansen is recipient of an unrestricted research grant from the LEO Foundation.
Table S1. Codes Used to Define Comorbidity and Medication
Figure S1. Flowchart of study population.
Figure S2. Risk factors associated with not returning to work when only patients who maintained employment for at least 3 consecutive mo were considered as having returned to work.
Figure S3. Risk factors associated with not returning to work where patients admitted before 2001 were excluded from the analysis.
Figure S4. Risk factors associated with detachment from employment where patients admitted before 2001 were excluded from the analysis.
- ↵Fox CS, Evans JC, Larson MG, Kannel WB, Levy D. Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study. Circulation. 2004;110:522–527.
- ↵Dreyer RP, Xu X, Zhang W, Du X, Strait KM, Bierlein M, Bucholz EM, Geda M, Fox J, D'Onofrio G, Lichtman JH, Bueno H, Spertus JA, Krumholz HM. Return to work after acute myocardial infarction comparison between young women and men. Circ Cardiovasc Qual Outcomes. 2016;9:S45–S52.
- ↵Hjollund NH, Larsen FB, Andersen JH. Register‐based follow‐up of social benefits and other transfer payments: accuracy and degree of completeness in a Danish interdepartmental administrative database compared with a population‐based survey. Scand J Public Health. 2007;35:497–502.
- ↵Kragholm K, Wissenberg M, Mortensen RN, Fonager K, Jensen SE, Rajan S, Lippert FK, Christensen EF, Hansen PA, Lang‐Jensen T, Hendriksen OM, Kober L, Gislason G, Torp‐Pedersen C, Rasmussen BS. Return to work in out‐of‐hospital cardiac arrest survivors: a nationwide register‐based follow‐up study. Circulation. 2015;131:1682–1690.
- ↵Rørth R, Wong C, Sørensen KDK, Fosbøl E, Mogensen UM, Lamberts M, Petrie MC, Jhund PS, Gerds TA, Torp‐Pedersen C, Gislason GH, McMurray JJV, Køber L, Kristensen SL. Return to the workforce following first hospitalization for heart failure: a Danish nationwide cohort study. Circulation. 2016;134:999–1009.
- ↵Isaaz K, Coudrot M, Sabry MH, Cerisier A, Lamaud M, Robin C, Richard L, Da Costa A, Khamis H, Abd‐Alaziz A, Gerenton C. Return to work after acute ST‐segment elevation myocardial infarction in the modern era of reperfusion by direct percutaneous coronary intervention. Arch Cardiovasc Dis. 2010;103:310–316.
- ↵Laut KG, Hjort J, Engstrøm T, Jensen LO, Tilsted Hansen H‐H, Jensen JS, Pedersen F, Jørgensen E, Holmvang L, Pedersen AB, Christensen EF, Lippert F, Lang‐Jensen T, Jans H, Hansen PA, Trautner S, Kristensen SD, Lassen JF, Lash TL, Clemmensen P, Terkelsen CJ. Impact of health care system delay in patients with ST‐elevation myocardial infarction on return to labor market and work retirement. Am J Cardiol. 2014;114:1810–1816.
- ↵Earle A, Ayanian JZ, Heymann J. Work resumption after newly diagnosed coronary heart disease: findings on the importance of paid leave. J Womens Health. 2006;15:430–441.
- ↵Avezum A, Makdisse M, Spencer F, Gore JM, Fox KAA, Montalescot G, Eagle KA, White K, Mehta RH, Knobel E, Collet JP. Impact of age on management and outcome of acute coronary syndrome: observations from the Global Registry of Acute Coronary Events (GRACE). Am Heart J. 2005;149:67–73.
- ↵Waziri H, Jørgensen E, Kelbæk H, Stagmo M, Pedersen F, Lagerqvist B, James S, Køber L, Wachtell K. Short and long‐term survival after primary percutaneous coronary intervention in young patients with ST‐elevation myocardial infarction. Int J Cardiol. 2016;203:697–701.
- ↵Spertus JA, Jones P, McDonell M, Fan V, Fihn SD. Health status predicts long‐term outcome in outpatients with coronary disease. Circulation. 2002;106:43–49.
- ↵Xie J, Wu EQ, Zheng Z‐J, Sullivan PW, Zhan L, Labarthe DR. Patient‐reported health status in coronary heart disease in the United States. Circulation. 2008;118:491–497.
- ↵Mittag O, Kolenda KD, Nordman KJ, Bernien J, Maurischat C. Return to work after myocardial infarction/coronary artery bypass grafting: patients’ and physicians’ initial viewpoints and outcome 12 months later. Soc Sci Med. 2001;59:1441–1450.
- ↵Fiabane E, Argentero P, Calsamiglia G, Candura SM, Giorgi I, Scafa F, Rugulies R. Does job satisfaction predict early return to work after coronary angioplasty or cardiac surgery? Int Arch Occup Environ Health. 2013;86:561–569.
- ↵Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H, Thayssen P, Abildgaard U, Pedersen F, Madsen JK, Grande P, Villadsen AB, Krusell LR, Haghfelt T, Lomholt P, Husted SE, Vigholt E, Kjaergard HK, Mortensen LS. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med. 2003;349:733–742.