Excluded from sickness insurance: rejected sickness cash benefit claims after 180 days of sick leave in the Swedish rehabilitation chain

doi:10.21203/rs.3.rs-3781060/v1

Purpose—to study factors predicting rejection of prolonged sickness cash benefit claims among persons on compensated sick leave for 180 days.

Methods—all 246 872 claims for employed persons on sick leave recorded in the Swedish official statistics register between 2018.01.01 − 2021.06.30 were analysed. Claim outcomes were evaluated using logistic regression with odds-ratios recalculated to relative risks (RR) with 95-per cent confidence intervals (CI), adjusted for sociodemographic, work and health related factors.

Results—Overall 46 611 (23 per cent) of the claims were rejected with slightly lower rates among women (0.98; 0.97 − 0.99). Musculoskeletal diseases had the highest rates (RR 1.41; CI 1.35 − 1.47) followed by injuries (1.32; 1.27 − 1.36) and symptoms (1.29; 1.26 − 1.33). Mental disorders also had above average rates (1.13; 1.10 − 1.17). The lowest rates were found among pregnancy related diagnoses (0.22; 0.20 − 0.26) and neoplasms (0.30; 0.30 − 0.31). Higher rates were also found among immigrants (1.19; 1.18 − 1.21), those with only primary education (1.05; 1.04 − 1.07) and blue collar workers (RR:s around 1.06). The regional variation was substantial (RR-range 0.48 − 1.82). Rejection rates declined substantially from 2021 due to changes in sickness insurance regulations.

Conclusions—High rejection rates were found for complex diagnoses and among groups with precarious positions at the labour market. Systematic unexplained differences in rates were identified between geographic regions.

sick leave

sickness insurance

claim

rejection

work ability

The cost of sick leave and disability in developed countries is high [1], and in Sweden a particular concern have been directed towards extensive sick leave due to mental disorders [2, 3]. The policy response from the former right wing government in order to reduce sick leave and increase labour supply, was to introduce time limits in sickness insurance and stricter criteria for assessing work ability according to the “rehabilitation chain” in 2008 [4–7]. In practice this was a shift in responsibility, from professionals representing organisations, e.g. case managers, to individuals experiencing reduced work ability [8, 9]. In 2015, when sickness insurance expenditures had been rising again for some years, the left-wing government established a specific goal for the “sickness rate” to be met at the end of 2020 [7, 10]. Henceforth, the Swedish Social Insurance Agency (SSIA) applied stricter assessments of work ability and insurance eligibility without contemporary changes in legislation [7, 10, 11], which subsequently decreased public trust in the sickness insurance system and the social insurance administration [7, 12, 13]. As a consequence of the stricter work ability assessments, the number of rejected sickness cash benefit claims rose substantially between 2015 and 2020, as depicted in Fig. 1. This was especially accentuated for those still on sick leave after 180 days when their ability to work were evaluated towards available jobs at the entire labour market rather than their current employment [6, 7, 10, 13, 14]. The left-wing government finally responded to the public distrust in sickness insurance with changes in sickness insurance legislation increasing eligibility, and from 2021 and onwards rejection rates have been declining substantially [14, 15].

FIGURE 1 ABOUT HERE

Figure 1 Proportion of rejected sickness cash benefit claims during day 180 − 365

in year 2010 − 2022 by diagnosis (ICD-10)

The quite few previous studies on rejected sickness or disability benefit claims for those with reduced work capacity due to impaired health, have indicated that a fairly limited fraction actually return to gainful employment in contrast with policy expectations [4, 6, 7, 16–18]. On the contrary, rejection of benefits claims have been reported to worsen mental health among already vulnerable groups in society [13, 19–24]. The shift towards individual responsibility in the return to work process [8, 9], could also be problematized in a health equity perspective since impaired health and reduced work capacity is more common among the less educated in lower socioeconomic strata [25, 26]. Lower socioeconomic groups are also less well equipped regarding health literacy, which is an important resource for maintaining or regaining health and work ability in complex societies and health care systems [27]. In addition, the ability to navigate the social insurance system, i.e. social insurance literacy, is a more challenging task for those with less formal education and lower socioeconomic status [28, 29]. Hence, it could be expected that lower status groups at the labour market have a higher risk of rejected sickness cash benefit claims. Up to 2019 the SSIA also requested objective medical findings in the medical certificates to support the reduction in work ability [11]. Since objective medical findings are more often at hand for some medical conditions than others, higher rejection rates may also be expected for diseases with a more complex anamnesis and less objective findings such as mental disorders and for symptom diagnoses as reflected in official statistics [7, 14]. In addition, it could be expected that rejection rates are higher for persons with injuries or diseases in the musculoskeletal system, because according to the interpretation of the legislation at the time of the study, they could be considered to have work ability in less physically strenuous work [7, 10, 11].

In Sweden, compulsory sickness insurance was introduced in 1955 and it took more than six decades before official statistics on rejection rates for sickness cash benefit were published in 2019 [14]. In addition, surprisingly few studies have been conducted in Sweden or internationally regarding rejected sickness cash benefits claims. Therefore, the current study aiming at investigating individual and contextual factors predicting rejection of prolonged sickness cash benefit claims among persons on compensated sick leave for 180 days, is a valuable contribution to the field.

Box 1: Swedish sickness insurance

The compulsory Swedish sickness insurance scheme potentially covers all Swedish or foreign citizens working in Sweden [30]. The prime requisite for being eligible for sickness cash benefit is that the person in question has a disease or injury that reduces his/her ability to work by at least 25 per cent. In practice the first day of sick leave is a qualifying day without compensation and sick pay is provided by the employer for the first 14 days. A medical certificate from a physician is required from day eight. If the reduction in work ability persists after 14 days, the employee receives sickness cash benefit from the national social insurance system. In 2008 specific time limits was introduced regarding the assessment of work ability which gradually becomes stricter over time and after 180 days of sick leave you are only entitled to the benefit if you are unable to perform any job at the labour market. But several exceptions to this rule applies and have been made more lenient over time especially since 2021 [30].

Study design and population

Cases of sick leave compensated by compulsory Swedish sickness insurance were retrieved from the SToRE-database, with data originating from administrative social insurance registers held by the SSIA. All spells exceeding 180 days with a claim and decision of prolonged sickness absence between 1st January 2018 and 31st June 2021 were included in the study. The total number of cases was 246 872, with 162,388 for women and 84,484 for men.

Outcome variable

A rejected sickness cash benefit claim after day 180 in the rehabilitation is defined as a rejected claim between day 181–365 in the sick leave period, in short: rejected benefit or rejection rate. This definition is identic with the definition applied in Swedish official statistics [14].

Predictors

To each case of sick leave, predictor variables were matched using the Swedish personal identity number [31]. Predictor variables were sex, age, civil status, diagnosis, occupation, branch (i.e. industry), employment sector, education, occupational status, origin of birth (Sweden or abroad), phase in time in relation to Covid-19 in Sweden (pre-pandemic 2018.01-2020.02, pandemic 2020.03-2021.02 and post-pandemic 2021.03-2021.06) and finally county of residence. Occupation were measured according to the Swedish version of ISCO-08, Swedish Standard Classification of Occupations [32], at the one-digit level with one exception. The major group 5 “Service, care and shop sales workers” were analysed at the 2-digit level since group 5.3 “Personal Care Workers” have essentially different work characteristics compared to the other more commercial occupations within major group 5. Branches were measured according to the Swedish version of NACE rev 2, Swedish Standard Industrial Classification at the one-digit level [33]. Personal information about educational level (according to ISCED 1997), economic sector of employer (private, state, regional or municipal authority) and occupational status (employed or not) have also been retrieved from Statistics Sweden. All other variables originate from the National Social Insurance Registers held by the Swedish Social Insurance Agency. Diagnosis were analysed at the chapter level, and were the latest primary sick leave diagnosis recorded on the medical certificate according to the Swedish version of ICD-10 [34]. For all other variables the data used refers to the status at the start of the sick leave spell as recorded in the Swedish social insurance registers.

Statistical analysis

All statistical analyses were performed using SAS EG (release 7.1). Logistic regression was used to analyse the odds of rejected benefit claim after day 180 and corresponding 95% confidence intervals (CI). Since rejected claims were fairly common the odds-ratios (OR) were recalculated to relative risks (RR) according to the formula RR = OR/(1 + OR). Missing values for covariates constitute distinct categories in the analysis, but their results are not presented since they lack meaningful interpretation.

Study population frequencies and multiple logistic regression outcomes are presented in Table 1. A majority of claimants are women and they have a slightly lower, 2 per cent, rejection rate than men. Those in ages 20 − 24 have a higher rejection rate and those in ages 60 − 64 have a lower rejection rate. Divorced persons have a somewhat higher rejection rate than married persons.

Table 1 ABOUT HERE

Table 1

Frequencies of covariates in the study population and logistic regression outcomes for rejected sickness cash benefits claims after day 180
Covariates	Study population		Rejected benefit claim
Covariates	n	%	RR	95%-CI
Sex
Women	162 388	65.8	0.98	0.97–0.99
Men (Reference)	84 484	34.2	1.00	-
Age in years
20–24	8 450	3.4	1.05	1.02–1.08
25–29	19 486	7.9	1.01	0.98–1.03
30–34	23 371	9.5	1.02	0.99–1.04
35–39	26 528	10.8	1.00	0.98–1.03
40–44 (Reference)	30 620	12.4	1.00	-
45–49	33 183	13.4	1.00	0.98–1.02
50–54	37 524	15.2	0.99	0.97–1.01
55–59	36 430	14.8	1.01	0.99–1.03
60–64	29 539	12.0	0.90	0.88–0.93
65-	1 741	0.7	1.00	0.93–1.07
Civil status
Married (Reference)	107 463	43.5	1.00	-
Unmarried	89 681	36.3	1.01	0.98–1.03
Divorced	45 375	18.4	1.03	1.00-1.06
Widow/Widower	4 353	1.8	0.98	0.90–1.07
Diagnosis chapter ICD-10. Reference is the unweighted mean across diagnosis-categories 1.00
I A00-B99 Certain infectious & parasitic diseases	1 183	0.5	1.02	1.00-1.04
II C00-D48 Neoplasms	19 015	7.7	0.30	0.30–0.31
III D50-D89 Diseases of the blood & blood-forming organs	500	0.2	1.02	0.96–1.08
IV E00-E90 Endocrine, nutritional & metabolic diseases	1 792	0.7	1.15	1.15–1.15
V F00-F99 Mental & behavioural disorders	117 555	47.6	1.13	1.10–1.17
VI G00-G99 Diseases of the nervous system	8 426	3.4	0.98	0.97-1.00
VII H00-H59 Diseases of the eye & adnexa	799	0.3	1.12	1.09–1.14
VIII H60-H95 Diseases of the ear & mastoid process	947	0.4	1.26	1.25–1.26
IX I00-I99 Diseases of the circulatory system	8 173	3.3	0.95	0.94–0.96
X J00-J99 Diseases of the respiratory system	3 415	1.4	1.01	1.00-1.01
XI K00-K93 Diseases of the digestive system	2 812	1.1	0.93	0.92–0.94
XII L00-L99 Diseases of the skin & subcutaneous tissue	890	0.4	1.22	1.21–1.22
XIII M00-M99 Diseases of the musculoskeletal system & connective tissue	46 496	18.8	1.41	1.35–1.47
XIV N00-N99 Diseases of the genitourinary system	1 165	0.5	0.77	0.72–0.82
XV O00-O99 Pregnancy, childbirth & the puerperium	3 944	1.6	0.22	0.20–0.26
XVIII R00-R99 Symptoms, signs & abnormal clinical & laboratory findings, not elsewhere classified	5 132	2.1	1.29	1.26–1.33
XIX S00-T98 Injury & poisoning	14 411	5.8	1.32	1.27–1.36
Other (chapter XVI, XVII, XX & XXI)	4 102	1.7	0.90	0.90–0.90
Unknown diagnosis	6 115	2.5	-	-

Table 1 cont.
Occupation (ISCO-08). Reference is the unweighted mean across occupations 1.00
0 Armed forces	116	0.1	1.08	0.86–1.27
1 Managers	9 866	4.0	0.90	0.89–0.92
2 Professionals	56 327	22.8	0.91	0.88–0.93
3 Technicians and associate professionals	25 872	10.5	0.89	0.87–0.91
4 Clerical support workers	18 893	7.6	0.87	0.85–0.89
5.1 Personal Services Workers	6 507	2.6	1.01	0.99–1.03
5.2 Sales Workers	9 995	4.1	0.99	0.98–1.01
5.3 Personal Care Workers	43 846	17.8	1.03	1.01–1.07
5.4 Protective Services Workers	1 741	0.7	1.03	1.02–1.04
6 Skilled agricultural and fishery workers	2 123	0.9	1.10	1.10–1.11
7 Craft and related trades workers	16 420	6.6	1.05	1.02–1.08
8 Plant and machine operators and assemblers	14 331	5.8	1.07	1.05–1.10
9 Elementary occupations	13 991	5.7	1.06	1.04–1.09
Occupation unknown	26 844	10.9	-	-
Branch (NACE Rev. 2). Reference is the unweighted mean across branches 1.00
Land management (A)	2 183	0.9	0.97	0.96–0.99
Manufacturing (B, C, D, E)	23 813	9.6	0.88	0.87–0.89
Construction (F)	13 813	5.6	1.03	1.02–1.03
Trade (G)	22 212	9.0	0.97	0.97–0.97
Transportation (H)	10 412	4.2	1.09	1.08–1.10
Hotel, restaurant, entertainment (I, R)	8 971	3.6	1.03	1.02–1.03
Business services (J, K, L, M, N, S)	43 942	17.8	0.97	0.97–0.97
Public administration (O)	16 862	6.8	1.00	1.00–1.00
Education (P)	32 762	13.3	1.03	1.03–1.04
Social services (Q)	57 673	23.4	1.03	1.03–1.04
Branch unknown	14 229	5.8	-	-
Employment sector
Private (Reference)	126 728	51,3	1,00	-
Municipality	66 381	26,9	0,97	0,96-0.99
Region	19 917	8,2	0,87	0,84 − 0.90
State	20 832	8,4	0,82	0,80-0.85
Employment sector unknown	13 014	5,3	-	-
Primary education	27 584	11.17	1.05	1.04–1.07
Secondary education (Reference)	120 516	48.82	1.00	-
Post-secondary education < 2 years	34 893	14.13	1.00	0.98–1.01
Post-secondary education ≥ 2 years	62 252	25.22	1.00	0.98–1.02
Education unknown	1 627	0.66	-	-
Occupational status
Employed (work income)	229 791	93.1	1.00	-
Not employed (no work income)	17 081	6.9	0.93	0.89–0.97
Country of birth
Sweden (Reference)	202 707	82.1	1.00	-
Abroad	44 165	17.9	1.19	1.18–1.21
Phase in time in relation to Covid-19
Pre-pandemic 2018.01-2020.02 (Reference)	162 539	65.8	1.00	-
Pandemic 2020.03-2021.02	69 618	28.2	1.35	1.34–1.36
Post-pandemic 2021.03-2021.06	14 715	6.0	0.69	0.66–0.72

Table 1 cont.
County of residence. Reference is the unweighted mean across counties 1.00
Stockholm	54 324	22.0	0.87	0.86–0.88
Uppsala	10 369	4.2	1.00	1.00-1.01
Södermanland	7 498	3.0	0.50	0.49–0.53
Östergötland	9 364	3.8	0.48	0.46–0.51
Jönköping	9 132	3.7	0.96	0.95–0.97
Kronoberg	5 113	2.1	0.93	0.92–0.94
Kalmar	5 818	2.4	0.60	0.59–0.63
Gotland	1 417	0.6	0.86	0.83–0.88
Blekinge	3 576	1.4	1.07	1.06–1.07
Skåne	28 246	11.4	0.72	0.70–0.74
Halland	7 587	3.1	0.96	0.95–0.97
Västra Götaland	47 095	19.1	0.72	0.71–0.74
Värmland	6 799	2.8	1.27	1.25–1.28
Örebro	7 043	2.8	1.23	1.21–1.23
Västmanland	6 891	2.8	0.87	0.86–0.89
Dalarna	7 409	3.0	1.22	1.20–1.22
Gävleborg	7 490	3.0	1.13	1.12–1.13
Västernorrland	6 323	2.6	1.34	1.31–1.36
Jämtland	2 957	1.2	1.82	1.73–1.90
Västerbotten	7 204	2.9	1.04	1.04–1.04
Norrbotten	5 086	2.1	1.42	1.38–1.45
Residence unknown	131	0.0	-	-

There are substantial differences in rejection rates between different diagnosis categories. Lower rejection rates were present for neoplasms, pregnancy related diagnoses, genitourinary diseases and circulatory diseases. Higher rejection rates were found for musculoskeletal diseases, injuries, symptoms, diseases of ear and eye, skin diseases, endocrine diseases and mental disorders. The other disease categories have rejections rates around the average.

The socioeconomic pattern is evident with lower rejection rates among white collar occupations in group 1 to 4 and higher rejection rates among blue collar occupations in group 5.3 to 9. A similar pattern was found between branches with higher than average rejection rates for those working within construction, transportation, hotel-restaurant-entertainment, education and social services, compared to the lower than average rejection rates among those working within land management, manufacturing, trade and business services.

Those working in private companies have higher rejection rates than those working for public employers such as municipalities, and in particular those working for employers at the regional or state level. As for education, those with solely primary education have higher rejection rates than those with secondary education or higher. Those who lack employment have lower rejection rates than those with employment. Another evident difference is the higher rejection rates for those born abroad compared to native swedes.

The higher rejection rates initiated from 2015 and onwards culminated during the pandemic with 35 percent higher levels during the pandemic 2020 compared to the two preceding years, and then fell sharply to 31 percent lower than pre-pandemic levels in march 2021 and thereafter. Finally, there were substantial differences in rejection rates between residential regions. Compared to the mean rejection rate level, the rates ranged from 52 percent lower in the county of Östergötland, to 82 percent higher in the county of Jämtland. Lower rejection rates were found in the more densely populated south of Sweden in counties such as Södermanland, Kalmar, Skåne, Västra Götaland, Stockholm and Västmanland. Higher rejections rates were present in the more sparsely populated north and west part of Sweden in counties such as Norrbotten, Västernorrland, Dalarna, Gävleborg, Värmland and Örebro.

The gender difference in sick leave is substantial [30], still the difference in rejected claims between genders is marginal as reflected in official statistics [14], and the present study only report a 2 percent lower rate for women when a number of other factors are accounted for. Those in ages 20 − 24 have a higher rejection rate and those in ages 60 − 64 a lower rejection rate, which could reflect presupposed work ability on the broader labour market [3]. The somewhat higher rejection rates among divorced persons may indicate a socially disadvantageous situation and economic stress [3], potentially hampering the ability to claim ones rights [27–29].

There are substantial differences in rejection rates between different diagnosis categories in line with expectations. Lower rejection rates were present for diagnoses with distinct objective findings such as neoplasms, pregnancy related diagnoses, genitourinary diseases and circulatory diseases. Higher rejection rates were found for musculoskeletal diseases, injuries, symptoms, diseases of ear and eye, skin diseases, endocrine diseases and mental disorders. For mental disorders and symptoms diagnoses this could probably be due to the presupposed lack of objective medical findings and/or a bigger challenge for physicians to adequately describe how such ailments reduces work ability in the medical certificates presented to the SSIA [7, 11, 35]. For the other physical diagnoses with high rejection rates there may also be an assumption that there exists some residual work ability, e.g. in physically lighter occupations for those with musculoskeletal diseases (e.g. back-pain) or mobility impairments due to injury [7, 10, 11].

Furthermore, the socioeconomic pattern is evident with lower rejection rates among white collar occupations and higher rejection rates among blue collar occupations [10, 13, 27–29]. A similar pattern was found between branches with high rejection rates for those working within branches with high sick leave rates such as construction, transportation, hotel-restaurant-entertainment, education and social services (i.e. healthcare, childcare, care of elderly and disabled persons etc.), with more adverse work environments, compared to the lower rejection rates among those working within land management, manufacturing, trade and business services [3, 26], Those working in private companies have higher rejection rates than those working in the public sector. This could potentially be due to lower degree of trade union organisation among employees in the private than in the public sector [36], and consequently less organised legal support in claiming sickness insurance benefits.

In line with expectations persons having solely primary education have higher rejection rates than those with secondary education or more [10, 27]. Another significant difference is the high rejection rates for those born abroad compared to native swedes, which have been previously reported [10]. The fact that those who lack employment have lower rejection rates for prolonged compensation could be considered illogical since the work ability of unemployed is assessed towards jobs at the entire labour market from the first day of sick leave. However, the employment status for those on extensive sick leave probably reflect their health status and the fact that they have lost their previous employment due to longstanding illness and inability to work [37].

The higher rejection rates initiated from 2015, when the SSIA started to use the discretionary scope within contemporary sickness insurance legislation to reduce sick leave aiming at the specified goal set for the sickness rate [7, 10, 11, 13], culminated during the pandemic, and then fell sharply after changes in regulations were launched in march 2021 [6, 14, 15]. Finally, there are substantial differences in rejection rates between residential regions partly reflecting the regional organisation of sickness insurance within the SSIA during the study period [15]. Lower rejection rates are found in the more densely populated south of Sweden where sick leave rates are comparably lower [30]. Higher rejections rates are present in the more sparsely populated north and west part of Sweden where sick leave rates are comparably higher [30]. An interpretation of the regional differences in rejection rates is that the discretionary scope within the contemporary sickness insurance legislation have been used to reduce prevailing regional differences in sick leave rates [7, 15, 30].

Strengths and limitations

This study has several major advantages in terms of generalisation. Firstly, the data are in general very reliable, though registration errors cannot be entirely ruled out. Secondly, administrative data are not hampered with low response rates as is often the case with standard survey techniques and relatively few cases had to be omitted due to missing values. Third, since all claims during 3.5 years for prolonged sick leave compensations after 180 days of sick leave made in the compulsory Swedish sickness insurance scheme were included in the study, the validity for the Swedish setting is evident, as well as in comparable societies with similar sickness insurance schemes. A further strength in the study is that a number of relevant confounders were accounted for in the regression analysis, e.g. age, diagnosis and occupation. Nevertheless, the study has limitations. Comorbidity is in general reported to further reduce work ability and increase sick leave duration [38, 39] and the inability to account for this is a shortcoming in the study. As in all observational studies the possible impact of residual confounding from other unmeasured or poorly measured covariates cannot be excluded. However, there is no single factor that have not been included in the analyses that is a likely candidate to explain the main findings by confounding. Still, the observational nature of the study inherently opens up for the possibility that other potential predictors influence the outcomes.

Practical Implications

This study show that rejection rates are substantially higher for medically multifaceted diagnoses where the connection between illness and the reduction in work ability is more complex given the presumption, during most of the study period, that this assessment is directed towards any job at the entire labour market. Furthermore, rejection rates are higher among disadvantaged groups at the labour market such as immigrants, less educated and blue-collar workers. Since several of the factors associated with higher rejection rates coincide, for instance for less educated, foreign born, privately employed blue-collar workers, in branches with disadvantageous work environments and a musculoskeletal disease, there is reason to believe that the actual rejection rates within the Swedish sickness insurance scheme could have been considerable for some groups during the study period. The results also indicate that the stricter assessment of work ability employed by the SSIA could have had negative welfare effects for vulnerable groups at the labour market, as indicated in earlier studies [4, 6, 7, 13, 18–21, 23, 24]. In addition, systematic unexplained differences in rejection rates were identified between geographic regions. Further studies are warranted to conclude if differences between groups and regions found in the present study are justified and can be attributed to other factors foreseen, for instance substantial differences in the reduction in work ability beyond what is reflected in medical diagnoses. Nevertheless, rejection rates have been varying over time and place within the Swedish sickness insurance scheme in a way that is incompatible with social justice [7, 11–13].

Acknowledgements

The author is grateful to Huiqiao Liang and Peter Abrahamsson who assisted in the preparation of the data.

Funding

Open access funding provided by Karolinska Institutet. No other funding has been provided.

Declaration of Interest: The author report no conflict of interest.

Ethics approval

Since only register data were used no informed consent was required from participants. The current register data is stored and processed under strict secrecy in accordance with Swedish law on official statistics (2001:99) and the data can only be used for production of social insurance statistics or research within the SSIA. The Swedish Social Insurance Agency (SSIA) is the Government authority in Sweden responsible for social insurance statistics and the author is a senior research officer at the Statistics Department at the SSIA.

Data availability

The data in this study are pseudonymized individual data from Swedish official statistics registers and are not publicly available according to Swedish legislation. They can be obtained from the Swedish Social Insurance Agency upon request from researchers fulfilling relevant legislation for research and data protection.

OECD, Sickness, Disability and Work. Breaking the barriers. A Synthesis of Findings across OECD Countries. 2010, OECD Publishing: Paris.
OECD, Mental Health and Work: Sweden. 2013, OECD Publishing: Paris.
Lidwall, U., et al., Mental disorder sick leave in Sweden: A population study. Work, 2018. 59(2): p. 259-272.
Vaez, M., et al., A follow-up of the introduction of a maximum entitlement period for receiving sickness benefits in Sweden: A nationwide register-based study. Scand J Public Health, 2020. 48(2): p. 144-154.
Lidwall, U., Termination of sickness benefits or transition to disability pension after changes in sickness insurance: a Swedish register study. Disabil Rehabil, 2013. 35(2): p. 118-24.
ISF, Utan inkomst efter avslag [Without income after rejected claim]. Vol. Rapport 2023:4. 2023, Gothenburg: Inspektionen för socialförsäkringen [Swedish Social Insurance Inspectorate].
ISF, Konsekvenser av förändringar i sjukförsäkringen [Consequences of changes in sickness insurance]. Vol. Rapport 2023:10. 2023, Gothenburg: Inspektionen för socialförsäkringen [Swedish Social Insurance Inspectorate].
Seing, I., et al., Policy and practice of work ability: a negotiation of responsibility in organizing return to work. J Occup Rehabil, 2012. 22(4): p. 553-64.
Seing, I., et al., Return to work or job transition? Employer dilemmas in taking social responsibility for return to work in local workplace practice. Disabil Rehabil, 2015. 37(19): p. 1760-9.
ISF, Ökning av antalet personer som får beslut om indragen sjukpenning [Increased number of rejected sickness cash benefit claims]. Vol. Rapport 2018:12. 2018, Stockholm: Inspektionen för socialförsäkringen [Swedish Social Insurance Inspectorate].
ISF, Försäkringskassans utredningsskyldighet [The investigation obligation of the Swedisch Social Insurance Agency]. Vol. Rapport 2021:3. 2021, Gothenburg: Inspektionen för socialförsäkringen [Swedish Social Insurance Inspectorate].
Karlsson, E.A., et al., Exploring Interactions in the Sickness Insurance System in Terms of Power and Trust. J Occup Rehabil, 2022. 32(3): p. 483-493.
Altermark, N. and A. Plesner, Austerity and identity formation: How welfare cutbacks condition narratives of sickness. Sociology of Health & Illness, 2022. 44(8): p. 1270-1286.
SSIA, Antal och andel avslag efter dag 180 i rehabiliteringskedjan [Number and proportion of rejected claims after day 180 in the rehabilitation chain], in Official Statistics of Sweden, SwedishSocialInsuranceAgency, Editor. 2023: Stockholm.
SSIA, Annual report. 2022, Stockholm: SwedishSocialInsuranceAgency.
Laaksonen, M., R. Gould, and J. Liukko, Labor market position after a rejection of a disability pension application: a register-based cohort study. Disabil Rehabil, 2018. 40(25): p. 3022-3029.
Barr, B., et al., To what extent have relaxed eligibility requirements and increased generosity of disability benefits acted as disincentives for employment? A systematic review of evidence from countries with well-developed welfare systems. J Epidemiol Community Health, 2010. 64(12): p. 1106-14.
Jensen, N.K., et al., Too sick to work, too healthy to qualify: a cross-country analysis of the effect of changes to disability benefits. J Epidemiol Community Health, 2019. 73(8): p. 717-722.
Reeves, A., et al., Does capping social security harm health? A natural experiment in the UK. Social Policy & Administration, 2021. 56(3): p. 345-359.
Dwyer, P., et al., Work, welfare, and wellbeing: The impacts of welfare conditionality on people with mental health impairments in the UK. Social Policy & Administration, 2019. 54(2): p. 311-326.
Simpson, J., et al., Effects of social security policy reforms on mental health and inequalities: A systematic review of observational studies in high-income countries. Soc Sci Med, 2021. 272: p. 113717.
McAllister, A., et al., How do macro-level structural determinants affect inequalities in mental health? - a systematic review of the literature. Int J Equity Health, 2018. 17(1): p. 180.
Barr, B., et al., 'First, do no harm': are disability assessments associated with adverse trends in mental health? A longitudinal ecological study. J Epidemiol Community Health, 2016. 70(4): p. 339-45.
Orchard, C., N. Carnide, and P. Smith, How Does Perceived Fairness in the Workers' Compensation Claims Process Affect Mental Health Following a Workplace Injury? J Occup Rehabil, 2020. 30(1): p. 40-48.
Lundberg, O., The next step towards more equity in health in Sweden: how can we close the gap in a generation? Scand J Public Health, 2018. 46(22_suppl): p. 19-27.
SSIA, Sjukfrånvaron på svensk arbetsmarknad [Sick leave at the Swedish labour market]. Vol. Socialförsäkringsrapport 2018:2. 2018, Stockholm: SwedishSocialInsuranceAgency.
Svendsen, M.T., et al., Associations of health literacy with socioeconomic position, health risk behavior, and health status: a large national population-based survey among Danish adults. BMC Public Health, 2020. 20(1): p. 565.
Stahl, C., et al., Social insurance literacy: a scoping review on how to define and measure it. Disabil Rehabil, 2021. 43(12): p. 1776-1785.
Karlsson, E.A., et al., Communication characteristics between clients and stakeholders within the Swedish sickness insurance system - a document analysis of granted and withdrawn sickness benefit claims. Disabil Rehabil, 2020. 42(23): p. 3316-3326.
Lidwall, U., ed. Social Insurance in Figures 2023. 2023, Försäkringskassan [Swedish Social Insurance Agency]: Stockholm.
Ludvigsson, J.F., et al., The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. European journal of epidemiology, 2009. 24(11): p. 659-667.
StatisticsSweden, Swedish Standard Classification of Occupations 2012. MIS 2012:1. 2012: Örebro.
StatisticsSweden, Swedish Standard Industrial Classification 2007. MIS 2007:2. . 2007: Örebro.
NationalBoardofHealthandWelfare, Swedish version of International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10). 2011: Stockholm.
Starzmann, K., P. Hjerpe, and K.B. Bostrom, The quality of the sickness certificate. A case control study of patients with symptom and disease specific diagnoses in primary health care in Sweden. Scand J Prim Health Care, 2019. 37(3): p. 319-326.
StatisticsSweden, Labour force surveys AKU. 2022, StatisticsSweden: Örebro.
Wagenaar, A.F., et al., Employment contracts and health selection: unhealthy employees out and healthy employees in? J Occup Environ Med, 2012. 54(10): p. 1192-200.
Ubalde-Lopez, M., et al., Influence of new secondary diagnoses on the duration of non-work-related sickness absence episodes. J Occup Environ Med, 2013. 55(4): p. 460-4.
Ubalde-Lopez, M., et al., Beyond Return to Work: The Effect of Multimorbidity on Work Functioning Trajectories After Sick Leave due to Common Mental Disorders. J Occup Rehabil, 2017. 27(2): p. 210-217.

No competing interests reported.

Excluded from sickness insurance: rejected sickness cash benefit claims after 180 days of sick leave in the Swedish rehabilitation chain

Status:

Version 1

Abstract

Figures

Introduction