Do debts lead to disability pension? Evidence from a 15-year follow-up of 54,000 Finnish men and women

Abstract

While over-indebtedness has emerged as a new social risk among Europeans as a consequence of economic recessions, its associations with health and disability are poorly understood. This study utilises longitudinal, register-based data to assess the associations of over-indebtedness with disability retirement. Severely over-indebted people were identified from the Finnish credit information register. For each over-indebted person, one matched control was retrieved from the population register (total N: 54,494). Register data on socio-demographics, health-related factors and pensions were used to analyse the incidence of disability pensions due to different diagnoses among over-indebted people and their controls during the period 1995–2009 using Cox regression analysis. The analyses showed that over-indebtedness was strongly associated with the risk of disability retirement in all diagnostic groups. The associations were stronger among women than among men. The adverse effects of over-indebtedness on health and disability should be acknowledged. In addition to standard socio-demographic indicators, debtor status may add to our knowledge of the predictors of poor health. Policymakers should adopt measures to prevent over-indebtedness and to overcome its harmful effects.

Keywords

Disability pension disability retirement Finland mental health morbidity over-indebtedness

Introduction

Exiting the labour force because of an inability to work is a problem that practically all Organisation for Economic Co-operation and Development (OECD) countries are currently having to deal with (OECD, 2010). A high proportion of disability pensioners burdens the financing of the welfare state due to increased costs of income transfers and the lack of worker input. On average, approximately 6 percent of the working-age population in the OECD (2010) received a disability benefit in 2008, with a variation rate of between 2 and 10 percent depending on the country. In Finland, approximately 7 percent of the population aged 16–64 were on disability pension at the end of 2013. The most common causes for new disability pensions in Finland are mental and behavioural disorders (33%) and musculoskeletal diseases (32%) (Finnish Centre for Pensions (FCP) and Social Insurance Institution of Finland (SIIF), 2014).

The latest financial crisis in Europe has brought about major threats to the health of the general population in many countries (Karanikolos et al., 2013). The crisis has most probably multiplied the number of over-indebted Europeans, even though no comprehensive statistics are currently available. Already before the deepest phase of the crisis, there were concerns about over-indebtedness (European Commission, 2008). These concerns are certainly warranted given the fact that throughout the 2000s, household debt relative to disposable income has risen in most OECD countries (OECD, 2014). Accordingly, concerns have risen regarding the effects of increasing debt on people’s health and wellbeing (McCloud and Dwyer, 2011; Sweet et al., 2013).

Earlier studies have demonstrated that the probability of entering a disability pension is related to social background, most often measured by education, occupational class and income (e.g. Leinonen et al., 2012; Polvinen et al., 2013). The result is evident given the vast amount of existing knowledge on the associations between health and individuals’ socio-demographic characteristics: for example, low income may affect health through poor material conditions and by complicating social participation (Marmot, 2002). However, it has been suggested that the traditional indicators of socio-economic status do not cover the whole spectrum of socio-economic variation. The authors of several studies have proposed that debtor status and over-indebtedness, a new social risk, may add to our understanding of the links between socio-economic status and health (Drentea and Reynolds, 2012; Münster et al., 2009; Richardson et al., 2013; Sweet et al., 2013). It has even been found that a debtor status is more consistently associated with mental health than any of the traditional indicators of socio-economic status (Drentea and Reynolds, 2012). Another recent study (Lallukka et al., 2015) found that frequent economic difficulties increased the risk of disability retirement. However, we are not aware of any previous register-based, individual-level follow-up studies on the associations between severe over-indebtedness and disability.

The aim of this study was to analyse the associations of severe over-indebtedness with the incidence of disability retirement in a 15-year follow-up assessment among a sample of Finnish over-indebted people compared to their matched controls.

Associations of over-indebtedness with health

Over-indebtedness refers to a household’s long-term inability to meet the expenses related to its contracted financial commitments without reducing its minimum standard of living (European Commission, 2008). The administrative definition of over-indebtedness – the definition used in this study – uses the criteria of an official registration of non-payment of debts by the authorities or a court declaration concerning non-payment (Betti et al., 2001). In Finland, a payment default entry is the result of extended payment difficulties and is recorded when neglect of payment has been confirmed via a court decision or an executive authority (Raijas et al., 2010). If over-indebted people have any eligible income or assets, they may be subject to enforcement until the debt is eventually paid in full.

Over-indebtedness may be a consequence of reckless consumption and risk-taking, which characterises the so-called dispositional route to excessive debt, claimed to be common among people with low self-control and with an orientation towards present time instead of long-term planning (Webley and Nyhus, 2001). Also, and perhaps more often, debt problems may be the result of unforeseen financial difficulties in the household, which in turn may be related to, for example, unemployment, sickness, divorce, decreases of property values or increases in interest rates because of macro-economic cycles (Muttilainen, 2002; Raijas et al., 2010). Whatever the ultimate cause, over-indebtedness is accompanied by the known downsides of living under severe financial strain, such as not being able to consume health-enhancing food and services, not being able to meet current health needs, not having access to health care and not being able to fully participate in social life; it is also associated with psychological distress, shame and anxiety, which may have adverse effects on both physical and mental health (Drentea and Lavrakas, 2000; Drentea and Reynolds, 2012; Hayes, 2000; Jacoby, 2002; Münster et al., 2009; Ochsmann et al., 2009). Furthermore, a payment default entry may rule out chances of being granted a credit card or other lines of credit, renting an apartment or finding employment (European Commission, 2008).

Given the distress related to over-indebtedness, it is not surprising that numerous studies have called attention to the detrimental effects of over-indebtedness on the health and wellbeing of individuals. In particular, studies have demonstrated an association between over-indebtedness and a range of mental disorders (Alley et al., 2011; Bridges and Disney, 2010; Cannuscio et al., 2012; Drentea and Reynolds, 2012; Jenkins et al., 2008; McLaughlin et al., 2012; Meltzer et al., 2013; Nettleton and Burrows, 1998; Sweet et al., 2013). Also, studies have highlighted the associations between over-indebtedness and physical health (Cannuscio et al., 2012; Drentea and Lavrakas, 2000; Münster et al., 2009; Ochsmann et al., 2009; Sweet et al., 2013).

Two recent systematic reviews of studies published between the years 1994 and 2013 found that over-indebtedness has been associated especially with mental health – in particular, depression and suicidal ideation – as well as with self-rated health and health behaviour (Richardson et al., 2013; Turunen and Hiilamo, 2014). Several studies have concluded that debt-related financial stress, worry, the inability to make health-maximising choices and harmful health behaviour serve as key intermediating factors between debt and its impacts on health (Drentea and Lavrakas, 2000; Drentea and Reynolds, 2012; Turunen and Hiilamo, 2014). However, causality is difficult to establish because of a lack of longitudinal studies (Richardson et al., 2013).

Over-indebtedness in Finland during and after the 1990s

The economic history of Finland offers a sound setting for studying the associations of over-indebtedness with the health of individuals. In the early 1990s, Finland entered a deep recession. During the years 1991–1993, the gross domestic product (GDP) dropped by 12 percent and the unemployment rate rose from 3.5 to 18 percent (Kalela et al., 2001). Many people lost their jobs, their companies, their property and/or their credit rating. Those who were most hard hit lost their assets, as the value of property plummeted at the same time as interest rates skyrocketed. Consequently, the recession led to a steep growth in the number of over-indebted people. The number of people with a payment default entry (often a result of unsuccessful enforcement activities) reached its peak in 1997 with 368,000 people defaulting (Suomen Asiakastieto, 2014). In the 1990s, the most common reasons for over-indebtedness included unemployment, social crisis, difficulties in running a business, guarantee liabilities and divorce (Ministry of Social Affairs and Health (MSAH), 2002).

This study analyses the extent to which severe over-indebtedness has been associated with disability retirement in a 15-year follow-up assessment of a sample of Finnish over-indebted men and women compared to their matched controls. Our sample, defined below, included people who became severely over-indebted during or before the recession of the 1990s. The problem of reverse causality has been acknowledged in several previous studies: debt problems may have adverse effects on health and wellbeing, but those people with suboptimal health are also likely to face financial problems (Gathergood, 2012; Meltzer et al., 2013). We in part tackle this problem by using longitudinal data.

Data and methods

Sample

Our data set was constructed through linking yearly observations from several national registers covering the years 1989–2010. We were able to use a retrospective method to single out people who had been over-indebted for at least 15 years, after which their debts were erased due to an exceptional legislative amendment. By utilising the database of Suomen Asiakastieto Ltd, a credit rating company that maintains the leading database of consumer and business credit history in Finland, we were able to acquire personal identification numbers of those people whose debt records were erased from the credit information register in April 2010 due to outdated payment liability. At that time, the payment default records of all expired payment liabilities had to be erased retrospectively from the credit information register following an amendment to the Credit Information Act that took effect on 1 April 2010 (L 527/2007; L 933/2009).

According to the updated Finnish Enforcement Code that took effect in 2008 (L 705/2007), payment liability can expire, that is, debts can be forgiven, 15 years after the judgment by a court, or after 20 years in some special cases. Until that time, the expected liability for debts had been lifelong. Thus, the debt records erased in 2010 concerned unpaid debts that had been judged no later than in the year 1995. Our sample of over-indebted people thus consisted of people whose debts originated during the recession years of the early 1990s or even during years prior to the recession, and who had lived as over-indebted people until the first decade of the 2000s, when their debts were erased. The data included approximately 37,000 individuals.

For each over-indebted person, one control matched for age, gender and place of residence was selected by means of random sampling from the population data file of the SIIF, with the restriction that the controls did not have any removal of payment default entries in the credit information register in April 2010. However, since comprehensive information on people’s debt and credit histories is lacking, some of the controls may also have been over-indebted at some previous point in time but had been able to pay off their debts, or they may have had debts that had not yet reached the 15-year limit required for pardoning.

The sample of over-indebted people and their controls was restricted to working-age people (aged 25–62) at the end of 1994, which was chosen as the starting point for the follow-up assessment. Since over-indebtedness could only be concluded retrospectively, the sample included only people who were still alive in 2010. For the statistical analyses of the incidence of disability retirement, the sample was further restricted to people who were not receiving pensions at the end of 1994. After these restrictions, the data included 54,494 individuals: 36,828 men and 17,666 women (Table 1).

Table 1.

Distributions of the study sample (%): Finnish men and women aged 25–62 who did not receive any pensions at the end of 1994.

	Men (%)	Women (%)
Over-indebted people or controls
Over-indebted	50.0	50.0
Control	50.0	50.0
Age
25–34	40.1	48.8
35–44	38.7	34.7
45–62	21.2	16.5
Marital status
Married	44.1	47.3
Never married	37.7	30.9
Divorced or separated	17.8	20.3
Widowed	0.3	1.4
Region
Helsinki and surroundings (Uusimaa)	31.3	32.6
Other parts of Southern Finland	23.0	23.4
Western Finland	22.5	22.2
Northern/Eastern Finland	23.2	21.8
Sickness allowance days in 1989–1994
No benefit days	92.8	89.6
1–100 days	4.9	7.3
More than 100 days	2.3	3.0
Number of chronic diseases treated with reimbursable medicines in 1989–1994
None	92.5	90.4
At least one	7.5	9.6
Chronic diseases in 1989–1994
Hypertension	2.9	2.6
Chronic bronchial asthma and COPD	1.4	2.5
Diabetes	0.8	0.5
Psychoses	0.4	0.6
Chronic coronary heart disease	0.3	0.1
Other diseases	2.2	4.3
Total	100	100
N	36,828	17,666

Source: Own calculations based on the data set used in this study (see ‘Data and methods’ section).

COPD: chronic obstructive pulmonary disease.

Independent variables

The analyses were performed separately by gender. The status of being over-indebted in contrast to being in the control group was the main independent variable. Demographic covariates, which are known to be associated with the incidence of disability retirement, were retrieved from the population register and included age, marital status and geographical region.

Covariates used to control for previous health status were retrieved from the registers of the SIIF and included the number of sickness allowance days during the 6-year period (1989–1994) prior to the follow-up, as well as chronic diseases during the same 6-year period, measured through entitlements to special medicine reimbursement. These entitlements, granted by the SIIF in conjunction with a relevant doctor’s certificate, guarantee that the patients can obtain the medicine needed for treatment of certain chronic diseases for free or for only a small out-of-pocket share of the cost (SIIF, 2012). These entitlements, which are part of the Finnish National Health Insurance programme, may be granted universally to anyone based solely on his or her health status, and are not dependent on income or other characteristics of the patient. Thus, they are often used as a proxy for chronic disease (Saastamoinen et al., 2012). The distributions of the covariates are shown in Table 1.

Measurement of disability retirement

Data on disability retirement were available as year-end information from the pension registers of the SIIF (responsible for the payment of basic level national pensions) and from the FCP (earnings-related pensions). Disability pensions may be granted based on long-term reduced work capacity, which must be demonstrated through a physician’s detailed assessment of the claimant’s health status and ability to work (MSAH, 2013). As the sample was restricted to those not receiving any pension at the start of the follow-up, the first year-end occasion of being on disability pension was recorded as the point at which a patient entered into a new disability pension. Most common diagnostic groups, based on the primary diagnosis of the pension, were analysed separately (see Table 2). Only the first entry into a disability pension based on each diagnosis was taken into account, but in some cases, a particular person may be included in the incidence rates of several diagnostic groups. This is possible because disability pensions may be granted either fixed-term or permanent status. If a person first enters into a fixed-term pension period, but later enters into another fixed-term (or permanent) pension period based on some other primary diagnosis, he or she is then counted in both diagnostic groups in analyses that focus on pensions that start due to a specific diagnosis.

Table 2.

Proportion (%) and number (N) of people entering into disability pensions based on the primary diagnoses during 1995–2009^a: Finnish men and women aged 25–62 who did not receive any pensions at the end of 1994.

	Men				Women
	Over-indebted		Controls		Over-indebted		Controls
	%	N	%	N	%	N	%	N
All disability pensions	19.3	3546	5.6	1036	20.6	1817	5.0	439
By diagnosis:
Mental and behavioural disorders	7.5	1382	2.1	391	10.6	935	2.3	202
Depression	2.2	408	0.9	159	4.9	431	1.3	117
Other mental and behavioural disorders	5.7	1053	1.4	258	6.5	578	1.2	104
Diseases of the musculoskeletal system	3.9	709	1.5	282	4.9	429	1.7	154
Diseases of the nervous system	3.3	603	0.6	117	2.7	236	0.4	32
Diseases of the circulatory system	2.0	375	0.7	127	1.1	94	0.3	23
Other diseases	4.2	778	1.1	197	3.5	311	0.7	65

Source: Own calculations based on the data set used in this study (see ‘Data and methods’ section).

The groups partially overlap since people may have been granted pensions based on different diagnoses during the follow-up time.

Statistical analyses

The risk of entering into new disability pensions during the period 1994–2009 was examined using descriptive methods and survival analysis. Cox proportional hazards regression models, suitable for analysing event history or survival data (Cox, 1972), were utilised to model the time in years from the start of the follow-up (end of 1994) until the first year-end record of having received a disability pension. If the person did not enter into a disability pension during the follow-up, the length of the follow-up was set at 15 years. Since older people may receive old-age pensions (currently flexibly between ages 63 and 68), all observations were censored at the time the person turned 63 if they did not enter disability pension before that. The results of the regression models are presented as hazard ratios (HRs), which can be interpreted as the risk ratios that the event in question would occur at any given point in time in the group of interest (here: the over-indebted) compared to the reference group (the controls) (Sedgwick, 2015). Results are shown for two types of models: unadjusted models including only over-indebtedness versus being in the control group (Model 1), and models adjusted for all covariates (Model 2).

Ethical considerations

The study only uses secondary data retrieved from registers. Conventions of good scientific practice, data protection and information security were applied when analysing the data and presenting the results. According to Finnish law, an ethical review is not required if no human subject is contacted and if the research is based on secondary data, such as public documents, registers or archived data (Finnish Advisory Board on Research Integrity, 2014).

Results: the associations of over-indebtedness with disability pensions

During the follow-up period, 1995–2009, 19 percent of over-indebted men and 21 percent of over-indebted women were granted a new disability pension, while the corresponding proportions were only 6 and 5 percent among men and women in the control group, respectively (Table 2). The most typical diagnostic group was mental and behavioural disorders: 8 percent of over-indebted men and 11 percent of women began receiving a disability pension due to these diagnoses. With respect to mental disorders, the diagnosis was most often one of the other mental or behavioural disorders than depression. Four percent of over-indebted men and 5 percent of over-indebted women, respectively, began receiving a disability pension because of diseases of the musculoskeletal system and connective tissue, while the proportions were less than 2 percent among the controls. The proportions (shown in Table 2) for different diagnoses partly overlap since some individuals had been granted successive fixed-term disability pensions based on different primary diagnoses during the follow-up time.

The HRs for entering into a disability pension during the follow-up time according to independent variables are presented in Table 3 (men) and Table 4 (women). Compared to the control group, the adjusted risk of disability retirement was 3.23-fold among over-indebted men and 4.15-fold among over-indebted women, respectively. For the different diagnostic groups, the HRs for the over-indebted varied from 2.34 to 4.25 among men and from 2.80 to 6.40 among women. We observed the highest HRs for diseases of the nervous system among both men and women, while we observed the lowest HRs for depression among men and for diseases of the musculoskeletal system among women. When looking at mental disorders in more detail, our estimates for depression were lower than our estimates for other mental and behavioural disorders. In all diagnostic groups, our estimates were higher for women than for men.

Table 3.

Hazard ratios (HRs) for the risk of new disability pensions in 1995–2009: Finnish over-indebted men and their controls who did not receive any pensions at the end of 1994.

	1. All disability pensions	2. Mental and behavioural disorders	2a. Depression	2b. Other mental and behavioural disorders	3. Diseases of the musculoskeletal system	4. Diseases of the nervous system	5. Diseases of the circulatory system	6. Other diseases
	HR	HR	HR	HR	HR	HR	HR	HR
Model 1 (unadjusted)
Over-indebted (ref. control)	3.66***	3.62***	2.58***	4.16***	2.54***	5.22***	2.97***	4.00***
Model 2 (adjusted)
Over-indebted (ref. control)	3.23***	3.07***	2.34***	3.42***	2.45***	4.25***	2.66***	3.64***
Age (ref. 25–34)
35–44	1.87***	1.24***	1.45***	1.11	2.39***	3.25***	4.24***	1.71***
45–62	3.20***	1.38***	1.60***	1.19	5.34***	3.99***	11.4***	3.13***
Marital status (ref. married)
Never married	1.75***	2.26***	1.02	3.14***	0.85	2.81***	1.44**	1.38***
Divorced/separated	1.56***	1.73***	1.34**	2.07***	1.13	2.29***	1.46***	1.37***
Widowed	1.41	0.99	0.49	1.33	1.07	2.75*	1.60	1.33
Region (ref. Helsinki and surroundings)
Other parts of Southern Finland	1.25***	1.17*	1.21	1.12	1.40***	1.49***	0.96	1.11
Western Finland	1.32***	1.28***	1.26*	1.26**	1.39***	1.67***	1.26	1.12
Northern / Eastern Finland	1.47***	1.31***	1.19	1.35***	1.77***	1.68***	1.64***	1.27**
Sickness absence benefit days in 1989–1994 (ref. 0)
1–100 days	1.58***	1.64***	1.82***	1.54***	1.96***	1.56**	0.91	1.27
More than 100 days	3.05***	3.20***	3.69***	2.95***	3.87***	1.30	1.94***	2.70***
Chronic diseases in 1989–1994
Hypertension	1.27**	1.00	1.14	0.93	1.48**	0.62	2.14***	1.04
Diabetes	2.35***	1.01	1.49	0.77	0.80	1.68	3.37***	4.21***
Chronic bronchial asthma and COPD	1.52***	1.20	1.49	1.11	1.42	1.14	1.04	2.43***
Chronic coronary heart disease	1.69**	0.82	1.07	0.57	0.48	1.63	8.34***	2.10*
Psychoses	3.47***	8.84***	4.66***	9.36***	0.23*	0.21	0.53	0.13*
Other diseases	1.74***	1.48**	1.32	1.57**	1.33	2.29***	1.37	2.16***

Source: Own calculations based on the data set used in this study (see ‘Data and methods’ section).

COPD: chronic obstructive pulmonary disease.

Significance levels: *p<0.05; **p<0.01; ***p<0.001.

Table 4.

Hazard ratios (HRs) for the risk of new disability pensions in 1995–2009: Finnish over-indebted women and their controls who did not receive any pensions at the end of 1994.

	1. All disability pensions	2. Mental and behavioural disorders	2a. Depression	2b. Other mental and behavioural disorders	3. Diseases of the musculoskeletal system	4. Diseases of the nervous system	5. Diseases of the circulatory system	6. Other diseases
	HR	HR	HR	HR	HR	HR	HR	HR
Model 1 (unadjusted)
Over-indebted (ref. control)	4.50***	4.83***	3.75***	5.70***	2.83***	7.46***	4.10***	4.85***
Model 2 (adjusted)
Over-indebted (ref. control)	4.15***	4.21***	3.40***	4.93***	2.80***	6.40***	3.72***	4.81***
Age (ref. 25–34)
35–44	1.84***	1.38***	1.58***	1.11	2.21***	2.60***	2.80***	1.83***
45–62	2.63***	1.47***	1.87***	1.05	4.02***	2.26***	6.62***	3.02***
Marital status (ref. married)
Never married	1.06	1.23**	1.03	1.38**	0.66***	1.70**	1.10	0.95
Divorced/separated	1.26***	1.46***	1.23*	1.60***	0.91	1.94***	1.35	0.93
Widowed	0.85	0.91	0.31*	1.58	0.81	2.04	1.00	0.46
Region (ref. Helsinki and surroundings)
Other parts of Southern Finland	1.14*	0.98	1.11	0.88	1.56***	0.88	1.03	1.39*
Western Finland	1.09	0.95	0.09	1.02	1.56***	0.74	0.86	1.29
Northern/Eastern Finland	1.26***	1.12	1.03	1.22	1.72***	1.08	1.08	1.53**
Sickness absence benefit days in 1989–1994 (ref. 0)
1–100 days	1.58***	1.34**	1.40*	1.34*	2.29***	1.36	1.29	1.64**
More than 100 days	2.47***	2.61***	2.92***	2.44***	3.04***	1.24	1.62	2.09***
Chronic diseases in 1989–1994
Hypertension	1.35**	1.11	1.21	1.06	1.06	0.69	3.62***	1.62*
Diabetes	1.70*	0.92	0.84	0.79	1.45	0.71	0.75	3.68***
Chronic bronchial asthma and COPD	1.48***	1.40*	1.48	1.29	1.56*	1.23	1.58	2.25***
Chronic coronary heart disease	2.90**	5.74***	4.40*	3.98	2.23	^a	3.82	1.58
Psychoses	3.64***	7.79***	2.84***	11.7***	0.43	0.96	^a	0.27
Other diseases	1.68***	1.23	1.14	1.40*	1.95***	1.65*	2.22**	1.82**

Source: Own calculations based on the data set used in this study (see ‘Data and methods’ section).

COPD: chronic obstructive pulmonary disease.

Not enough observations for an estimation.

Significance levels: *p < 0.05; **p < 0.01; ***p < 0.001.

As expected, we found that the control variables were also associated with the risk of disability retirement (Tables 3 and 4). When adjusted for other covariates, advanced age was generally associated with a greater risk of entering into a disability pension, with the exception of other mental and behavioural disorders. Being never married or divorced/separated was associated with men’s higher risk of disability pension. In contrast, only being divorced/separated was rather consistently associated with a greater risk of receiving a disability pension across different diagnostic categories among women. Region was also associated with risk of disability pension – men living in any other region compared to the Helsinki capital region were at a higher risk of disability pension, with some variation across diagnostic groups. Among women, we found fewer and weaker associations between region and the risk of disability retirement; this partly had to do with the smaller sample size. The number of previous sickness allowance days predicted entering disability pension in all of our analyses, except for diseases of the nervous system or diseases of the circulatory system among women. Chronic diseases were also associated with an increased risk of disability retirement; however, our estimates and their statistical significance vary between the diagnostic groups. Due to space limitations, the results concerning the covariates cannot be further elaborated.

We also conducted sensitivity analyses to assess the association with varying lengths of follow-up time: we ran similarly adjusted models with a 5-year and a 10-year follow-up time. The HRs for over-indebtedness increased with an increasing follow-up time, but the estimates were always statistically significant. Among men, the adjusted HR for disability retirement was 1.8 with a 5-year follow-up time and 2.4 with a 10-year follow-up time. Among women, the adjusted HRs were 2.3 and 3.4, respectively. A plausible explanation has to do with the cumulative burden of long-term financial difficulties and its consequences with respect to health.

Discussion

Over-indebtedness has emerged as a new social risk in the European agenda. This study used longitudinal register-based data on over-indebted people and their matched controls to study whether over-indebtedness was associated with the onset of disability retirement during a 15-year follow-up in a Finnish cohort. We are not aware of any previous studies that have aimed to quantify the associations of over-indebtedness with disability in a longitudinal setting, even though several previous studies have examined the association of over-indebtedness with other health measures. Furthermore, register-based studies on the effects of over-indebtedness on health are practically non-existent.

Our analyses showed that over-indebtedness was strongly and statistically significantly associated with the incidence of disability pension in all of the diagnostic groups examined. These results are in line with previous studies on the adverse health consequences of over-indebtedness, both in terms of mental and physical health (e.g. Bridges and Disney, 2010; Cannuscio et al., 2012; Drentea and Lavrakas, 2000; Drentea and Reynolds, 2012; Jenkins et al., 2008; McLaughlin et al., 2012; Meltzer et al., 2013; Münster et al., 2009; Ochsmann et al., 2009; Sweet et al., 2013). Our recent paper (Blomgren et al., 2016) also demonstrates an association between over-indebtedness and both mental and physical chronic disease.

Among the most common diagnostic groups of disability pensions, we found the strongest associations with diseases of the nervous system and the weakest associations – albeit also clear and statistically significant – with depression and diseases of the musculoskeletal system. The comparatively low HRs of musculoskeletal diseases may be partly explained by the fact that these diseases are among the most common reasons for disability retirement in the general population (FCP and SIIF, 2014), and they accounted for a considerably larger percentage of all new disability pension in the data for our control group than for the over-indebted group. Thus, musculoskeletal diseases may not be as strongly connected to the social background of people as the other diagnoses in this data set.

According to a recent meta-analysis that combined the results of previously published studies, the unadjusted odds ratio of having mental disorders was 3.2 among those with debt problems compared to those without debt problems (Richardson et al., 2013). The unadjusted HRs that we obtained for disability retirement due to mental disorders were relatively close: 3.6 for men and 4.8 for women. The association between depression and debt as well as other types of financial distress has been widely reported in previous studies (Bridges and Disney, 2010; Drentea and Reynolds, 2012; McLaughlin et al., 2012). Likewise in our study, over-indebtedness was clearly associated with depression-based disability retirement, but the result is somewhat overshadowed by the larger HRs concerning mental and behavioural disorders other than depression and those concerning diseases of the nervous system. The results are in concordance with those obtained in the above-mentioned meta-analysis of Richardson et al. (2013), who obtained a smaller estimate for the association between debt and depression than for the association between debt and mental disorders in general.

In the diagnostic group of other mental and behavioural disorders than depression, schizophrenia and bipolar disorders are common diagnoses. These diseases may be related to a weak labour market position, unemployment and, for example, substance use (Nesvag et al., 2015; Zimmerman et al., 2010). Furthermore, the strong association of over-indebtedness with disability pensions due to diseases of the nervous system may partly reflect the fact that the degeneration of the nervous system due to alcohol belongs to this disease category (World Health Organization (WHO), 2015). Previous studies have shown that over-indebtedness is often linked to health through health behaviour (Drentea and Lavrakas, 2000; Drentea and Reynolds, 2012); our study indirectly supports such findings. Stress related to over-indebtedness as well as long-term unemployment, which has been a common cause of over-indebtedness in Finland since the 1990s, may be related to increased alcohol use. It is known that a large share of those who retire due to mental disorders and diseases of the nervous system in Finland have a substantial history of unemployment (Blomgren et al., 2011). Work disability and unemployment are thus intertwined, and it is difficult to determine the order of causality. In addition, both are related to over-indebtedness, which is known to be associated with problems of finding employment and remaining employed (Turunen and Hiilamo, 2014).

Overall, the associations between over-indebtedness and disability retirement were greater among women than men, which is in line with previous results obtained by Nettleton and Burrows (1998). Research has demonstrated that women are more risk averse than men (Jianakoplos and Bernasek, 1998), and they may thus experience more stress over over-indebtedness than men. However, selection into being over-indebted may also have played a role: our data set consisted predominantly of over-indebted men rather than women, and women may thus have been more strongly selected into the over-indebted group in terms of their social background, for example.

Compared to previous studies that have used more traditional measures of socio-economic status, such as education or occupation-based socio-economic status, we obtained somewhat different results concerning the strength of the association for different diagnoses. For example, previous studies have found larger differences with respect to musculoskeletal diseases rather than for mental diseases when comparing educational groups (Bruusgaard et al., 2010) or occupational status groups (Polvinen et al., 2014). On the other hand, previous studies have found an especially strong social gradient in disability retirement based on the use of psychoactive substances, such as alcohol (Polvinen et al., 2014), which is in line with the results of our study.

In conclusion, the differences in our results compared to studies using traditional socio-economic indicators corroborate the proposition that debtor status is an independent indicator of socio-economic position, and it may considerably add to our knowledge of predictors of poor health (Drentea and Reynolds, 2012; Münster et al., 2009; Richardson et al., 2013; Sweet et al., 2013). Over-indebtedness may be considered a special indicator of severe financial scarcity and economic difficulty, which in turn have been associated with health and disability (e.g. Lallukka et al., 2015; Marmot, 2002). In part, the association is related to the lack of financial ability to consume health-enhancing food and services, to meet current health needs, to have access to quality health care and to fully participate in social life. Additionally, indebtedness entails living under constant stress, shame and anxiety, all of which have detrimental effects on health (Drentea and Lavrakas, 2000; Jacoby, 2002). Financial stress and worry have also been shown to undermine the capability for long-term planning and making wise decisions (Mani et al., 2013). Over-indebtedness may thus affect disability at least through its effects on the financial ability to take care of one’s health and through the stress effects.

On the other hand, over-indebtedness and living under the threat of foreclosure may act as an incentive to leave the workforce since all efforts to increase income through work are punished by withholding a part of a person’s income that exceeds the protected portion, which is currently (in 2016) €678 per month for a person not supporting anyone else with his or her income (National Administrative Office for Enforcement (NAOE), 2016). Depending on the net income, an increasing share of the additional income is garnished by the enforcement authority – it is generally one-third, but it can be as much as one half (NAOE, 2016). Through the disincentive to work and increase one’s income, especially at lower wage levels, debtors may be more prone to remain unemployed or to aim at leaving the workforce via pensions. Even though entering into a disability pension always requires decreased working capacity certified by a doctor, it also requires the person’s own initiative to apply for such a pension, and the motivation for applying may be higher among those with poorer incentives to remain in the workforce. Thus, entering into a disability pension may also be seen as a survival strategy for the over-indebted.

Strengths and limitations

The study was based on unique register-based data on severely over-indebted individuals and their matched controls. We identified the over-indebted cohort retrospectively based on people who were alive in 2010 but who were known to have been over-indebted for at least 15 years. We were not able to focus on a sample of all Finnish people who were over-indebted at baseline in 1994. People who had been able to pay off their debts or who were released from them through debt adjustment were not included in our cohort. Thus, the sample represents the most severely over-indebted people in Finland during this particular time period: those people who had lived under lifelong liability for their debts until their debts were forgiven due to legislative changes in the 2000s. This may have caused an upward bias in our HRs for over-indebtedness, as these people have, most probably, been in a weaker position than those over-indebted people who ultimately succeeded in managing their debts.

In contrast, the control group may include people who had been over-indebted at baseline or at some other point in time but who did not have their record expunged from the credit information register in 2010 – or who had not yet reached the 15-year limitation of debt liability. This may cause some downward bias with respect to the difference between the over-indebted group and the control group. Approximately 300,000 Finns (7% of the adult population) had a payment default entry in 2009 (Suomen Asiakastieto, 2014), but a large part of those entries are likely to be based on more recent debts. Also, a large proportion of such people were already included in our cohort of over-indebted people. We may rather safely assume that the share of over-indebted people waiting for their debts to expire among the controls is likely to be small. In any case, the estimates are on the safe side: if all those people with some debt problems could be cleaned from the controls, the observed differences would be even larger than those observed here.

Our results showed strong associations between over-indebtedness and disability, but they do not demonstrate a causal effect. It may be assumed that some of the observed differences between over-indebted people and their controls may be attributed to unmeasured socio-economic factors, such as education, their earlier position in the labour market and income. Unfortunately, these variables were not available in the data set. Also, the amount of original debt or the date of debt was not available.

Despite the fact that we examined the incidence of disability pensions among people who did not receive any pensions at baseline and had control variables for their previous health status, it is possible that some people may have already been developing a disease and/or disability before the event of over-indebtedness occurred. Also, the direction of causation between over-indebtedness and ill health is likely to run in both ways: worse health may also be one of the causes of over-indebtedness (Meltzer et al., 2013). Reverse causality cannot be overruled as an explanation for the observed associations in our study. Even though we are not able to show proof for causality, we find a strong and consistent association between over-indebtedness and work disability.

Our results are indicative of a general connection between over-indebtedness and work disability. However, we need to take into account the special context of the recession in Finland during the early 1990s. A long history of restricted access to business and private credit was, prior to legislative amendments in the 2000s, reflected in penalising legislation against debtors with no chance to declare personal bankruptcy and no chance for debt pardoning. Over-indebtedness was and perhaps still is a more stigmatising event than in more liberal credit markets.

Further research is needed on the health effects of the most recent economic downturn and especially on the interactions between different socio-economic factors and over-indebtedness. Mechanisms through which debt is related to health are also poorly understood. More longitudinal prospective cohort studies are needed in order to establish the order of causation between debt and health.

Conclusion

The study found strong associations between over-indebtedness and work disability. Policymakers should strive to prevent over-indebtedness by enhancing financial education, especially for young people, and by controlling access to credit. Also, debt advice and social credit should be offered to heavily indebted people. If debt problems cannot be avoided, their negative consequences should be tackled. Health and social care professionals should be aware of the potential harmful health effects of over-indebtedness, such as financial stress and its consequences. Furthermore, the current regulations of enforcement should also be reconsidered, as currently there may not be enough incentives for over-indebted, unemployed people to take up low-paid employment. As negative health effects may further weaken an individual’s ability to manage his or her debts, it is also in the creditors’ interests to maintain the working ability of the over-indebted so that those people can ultimately obtain financial resources and eventually repay their debts.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was supported by the Academy of Finland, project Social Consequences of Economic Depressions, grant number 259216.

References

Alley

D.E.

Lloyd

Pagan

J.A.

Pollack

C.E.

Shardell

Cannuscio

(2011) ‘Mortgage Delinquency and Changes in Access to Health Resources and Depressive Symptoms in a Nationally Representative Cohort of Americans Older than 50 Years’, American Journal of Public Health 101(12): 2293–8.

Betti

Dourmashkin

Rossi

M.C.

Verma

Yin

(2001) ‘Study of the Problem of Consumer Indebtedness: Statistical Aspects’, Final Report, Commission of the European Communities, Directorate-General for Health & Consumer Protection, Belgium, October. London: ORC Macro.

Blomgren

Hytti

Gould

(2011) Työkyvyttömyyseläkkeelle siirtyneiden työttömyys- ja sairaustausta eri eläkejärjestelmissä (Nettityöpapereita 26). Helsinki: Kela.

Blomgren

Maunula

Hiilamo

(2016) ‘Over-Indebtedness and Chronic Disease: A Linked Register-Based Study of Finnish Men and Women during 1995–2010’, International Journal of Public Health 61(5): 535–44.

Bridges

Disney

(2010) ‘Debt and Depression’, Journal of Health Economics 29(3): 388–403.

Bruusgaard

Smeby

Claussen

(2010) ‘Education and Disability Pension: A Stronger Association than Previously Found’, Scandinavian Journal of Public Health 38(7): 686–90.

Cannuscio

C.C.

Alley

D.E.

Pagán

J.A.

Soldo

Krasny

Shardell

. (2012) ‘Housing Strain, Mortgage Foreclosure, and Health’, Nursing Outlook 60(3): 134–42.

Cox

D.R.

(1972) ‘Regression Models and Life-Tables’, Journal of the Royal Statistical Society Series B 34(2): 187–220.

Drentea

Lavrakas

P.J.

(2000) ‘Over the Limit: The Association among Health, Race and Debt’, Social Science & Medicine 50(4): 517–29.

10.

Drentea

Reynolds

J.R.

(2012) ‘Neither a Borrower nor a Lender Be: The Relative Importance of Debt and SES for Mental Health among Older Adults’, Journal of Aging and Health 24(4): 673–95.

11.

European Commission (2008) Towards a Common Operational European Definition of Over-Indebtedness. Brussels: European Commission.

12.

Finnish Advisory Board on Research Integrity (2014) ‘Ethical Review’, available at www.tenk.fi/en/ethical-review-human-sciences/ethical-review.

13.

Finnish Centre for Pensions (FCP) and Social Insurance Institution of Finland (SIIF) (2014) Statistical Yearbook of Pensioners in Finland 2013. Helsinki: FCP and SIIF.

14.

Gathergood

(2012) ‘Debt and Depression: Causal Links and Social Norm Effects’, Economic Journal 122(563): 1094–114.

15.

Hayes

T.A.

(2000) ‘Stigmatizing Indebtedness: Implications for Labeling Theory’, Symbolic Interaction 23(1): 29–46.

16.

Jacoby

M.B.

(2002) ‘Does Indebtedness Influence Health? A Preliminary Inquiry’, Journal of Law, Medicine & Ethics 30(4): 560–71.

17.

Jenkins

Bhugra

Bebbington

Brugha

Farrell

Coid

. (2008) ‘Debt, Income and Mental Disorder in the General Population’, Psychological Medicine 38(10): 1485–93.

18.

Jianakoplos

N.A.

Bernasek

(1998) ‘Are Women More Risk Averse?’, Economic Inquiry 36(4): 620–30.

19.

Kalela

Kiander

Kivikuru

Loikkanen

H.A.

(2001) Down from the Heavens, up from the Ashes. The Finnish Economic Crisis of the 1990s in the Light of Economic and Social Research. Helsinki: Government Institute for Economic Research.

20.

Karanikolos

Mladovsky

Cylus

Thomson

Basu

Stuckler

. (2013) ‘Financial Crisis, Austerity, and Health in Europe’, Lancet 381(9874): 1323–31.

21.

L 527/2007. ‘The Credit Information Act 11.5.2007/527’. Finnish legislation.

22.

L 705/2007. ‘The Enforcement Code 15.6.2007/705’. Finnish legislation.

23.

L 933/2009. ‘The Amendment to the Credit Information Act 20.11.2009/933’. Finnish legislation.

24.

Lallukka

Mauramo

Lahelma

Rahkonen

(2015) ‘Economic Difficulties and Subsequent Disability Retirement’, Scandinavian Journal of Public Health 43(2): 169–75.

25.

Leinonen

Martikainen

Lahelma

(2012) ‘Interrelationships between Education, Occupational Social Class, and Income as Determinants of Disability Retirement’, Scandinavian Journal of Public Health 40(2): 157–66.

26.

McCloud

Dwyer

R.E.

(2011) ‘The Fragile American: Hardship and Financial Troubles in the 21st Century’, Sociological Quarterly 52(1): 13–35.

27.

McLaughlin

K.A.

Nandi

Keyes

K.M.

Uddin

Aiello (2012) ‘Home Foreclosure and Risk of Psychiatric Morbidity during the Recent Financial Crisis’, Psychological Medicine 42(7): 1441–8.

28.

Mani

Mullainathan

Shafir

Zhao

(2013) ‘Poverty Impedes Cognitive Function’, Science 341(6149): 976–80.

29.

Marmot

(2002) ‘The Influence of Income on Health: Views of an Epidemiologist’, Health Affairs 21(2): 31–46.

30.

Meltzer

Bebbington

Brugha

Farrell

Jenkins

(2013) ‘The Relationship between Personal Debt and Specific Common Mental Disorders’, European Journal of Public Health 23(1): 108–13.

31.

Ministry of Social Affairs and Health (MSAH) (2002) Report on Social and Health Issues 2002 (Publication in Finnish: Sosiaali- ja terveyskertomus 2002). Helsinki: Ministry of Social Affairs and Health.

32.

Ministry of Social Affairs and Health (MSAH) (2013) Characteristics of the Social Security System in Finland. Tampere: Ministry of Social Affairs and Health.

33.

Münster

Rüger

Ochsmann

Letzel

Toschke

A.M.

(2009) ‘Over-Indebtedness as a Marker of Socioeconomic Status and Its Association with Obesity: A Cross-Sectional Study’, BMC Public Health 9: 286.

34.

Muttilainen

(2002) Luottoyhteiskunta. Kotitalouksien velkaongelmat ja niiden hallinnan muodonmuutos luottojen säännöstelystä velkojen järjestelyyn 1980- ja 1990-luvun Suomessa. Helsinki: National Research Institute of Legal Policy.

35.

National Administrative Office for Enforcement (NAOE) (2016) ‘Attachment of Salary’, available at www.oikeus.fi/ulosotto/en/index/velallisenaulosotossa/palkanulosmittaus.html.

36.

Nesvag

Knudsen

G.P.

Bakken

I.J.

Hoye

Ystrom

Suren

. (2015) ‘Substance Use Disorders in Schizophrenia, Bipolar Disorder, and Depressive Illness: A Registry-Based Study’, Social Psychiatry and Psychiatric Epidemiology 50(8): 1267–76.

37.

Nettleton

Burrows

(1998) ‘Mortgage Debt, Insecure Home Ownership and Health: An Exploratory Analysis’, Sociology of Health & Illness 20(5): 731–53.

38.

Ochsmann

Rueger

Letzel

Drexler

Münster

(2009) ‘Over-Indebtedness and Its Association with the Prevalence of Back Pain’, BMC Public Health 9: 451.

39.

Organisation for Economic Co-operation and Development (OECD) (2010) Sickness, Disability and Work: Breaking the Barriers. A Synthesis of Findings across OECD Countries. Paris: OECD Publishing.

40.

Organisation for Economic Co-operation and Development (OECD) (2014) OECD Factbook 2014: Economic, Environmental and Social Statistics. Paris: OECD publishing. Available at https://dx-doi-org.web.bisu.edu.cn/10.1787/factbook-2014-en.

41.

Polvinen

Gould

Lahelma

Martikainen

(2013) ‘Socioeconomic Differences in Disability Retirement in Finland: The Contribution of Ill-Health, Health Behaviours and Working Conditions’, Scandinavian Journal of Public Health 41(5): 470–8.

42.

Polvinen

Laaksonen

Gould

Lahelma

Martikainen

(2014) ‘The Contribution of Major Diagnostic Causes to Socioeconomic Differences in Disability Retirement’, Scandinavian Journal of Work, Environment & Health 40(4): 353–60.

43.

Raijas

Lehtinen

Leskinen

(2010) ‘Over-Indebtedness in the Finnish Consumer Society’, Journal of Consumer Policy 33(3): 209–23.

44.

Richardson

Elliott

Roberts

(2013) ‘The Relationship between Personal Unsecured Debt and Mental and Physical Health: A Systematic Review and Meta-Analysis’, Clinical Psychology Review 33(8): 1148–62.

45.

Saastamoinen

Aaltonen

Maljanen

Tuominen

Martikainen

(2012) ‘Health Registers as a Source of Data for Research and Policy Making’, Dosis 28(3): 199–205.

46.

Sedgwick

(2015) ‘Interpreting Hazard Ratios’, BMJ 351: h4631.

47.

Social Insurance Institution of Finland (SIIF) (2012) ‘Special Reimbursement of Medicines’, available at www.kela.fi/web/en/reimbursements-for-medicine-expences_special-reimbursement.

48.

Suomen Asiakastieto (2014) ‘Statistics on Payment Default Entries’, available at www.asiakastieto.fi/web/fi/asiakastieto/ajankohtaista/-/asset_publisher/gxzzwXCm45vn/content/maksuhairioisten-kuluttajien-maara-lahenee-laman-jalkeista-ennatysta.

49.

Sweet

Nandi

Adam

E.K.

McDade

T.W.

(2013) ‘The High Price of Debt: Household Financial Debt and its Impact on Mental and Physical Health’, Social Science & Medicine 91: 94–100.

50.

Turunen

Hiilamo

(2014) ‘Health Effects of Indebtedness: A Systematic Review’, BMC Public Health 14: 489.

51.

Webley

Nyhus

E.K.

(2001) ‘Life-Cycle and Dispositional Routes into Problem Debt’, British Journal of Psychology 92(3): 423–46.

52.

World Health Organization (WHO) (2015) ‘ICD-10 Version: 2015. International Statistical Classification of Diseases and Related Health Problems 10th Revision’, available at http://apps.who.int/classifications/icd10/browse/2015/en.

53.

Zimmerman

Galione

J.N.

Chelminski

Young

Dalrymple

Ruggero

C.J.

(2010) ‘Sustained Unemployment in Psychiatric Outpatients with Bipolar Disorder: Frequency and Association with Demographic Variables and Comorbid Disorders’, Bipolar Disorders 12(7): 720–6.