Medication and rehabilitation of persons applying for disability pension due to depression: a register-based retrospective study from Finland

Abstract Purpose To investigate how drug therapies and rehabilitation options have been utilised before applying for a disability pension due to depression. Methods A retrospective register-based study of the 3604 persons who applied for a disability pension from the Social Insurance Institution of Finland (Kela) in 2019. In Finland, disability pension is usually preceded by an incapacity for work lasting for 1 year, during which time therapeutic procedures, which were analysed here, are applied. Results Approximately half (56.0%) of the applicants had reimbursed purchases of two or more antidepressants during the 12 months preceding the disability pension application. Psychotherapy was received by 13.8% and 19.2% of the applicants 1 and 5 years before application, respectively. The share of applicants receiving some form of rehabilitation 1 year before application was 24.8% and 39.0% in the 5 years preceding application. During the 4 months before application, 19.6% of the applicants had no antidepressant purchases. In total, 12.2% of the applicants had both antidepressant treatment and psychotherapy in the year preceding the application, and 9.9% had neither psychotherapy nor antidepressant treatment. Conclusion Before applying for disability pension, only a minority of the applicants had received effective treatment for depression in the form of psychotherapy and antidepressants. However, most of the applicants had received some form of treatment, but it appears to have been insufficient.


Introduction
Measured in years lived with disability, depression is the leading cause of disability worldwide [1].Furthermore, it causes a significant economic burden to societies worldwide due to treatment expenses and lost productivity [2][3][4].The highest costs are related to workplace absenteeism and presenteism [3,4].In the US, work-related costs of depression had the highest increase, while individual direct costs declined during 2010-2018 [4].According to recent meta-analyses, depression is associated with a 1.52-fold risk for sick leave [5] and a 1.68-fold risk for disability pension [6].
There are many effective treatment and rehabilitation options for people with depression.Depression can be treated with antidepressants, antipsychotics or lithium as augmentation, psychotherapies, or neuromodulative treatments.Combining antidepressants and psychotherapies is more effective than either treatment alone [7].However, earlier studies indicate that the available treatment methods are not utilised efficiently enough, and, in Finland, a minority of persons applying for disability pension due to depression received either two different antidepressants or psychotherapy [8][9][10].No significant improvement in the matter was observed in the samples of those applying for disability pension due to depression in 1993-1994 and 2003-2004 [9].
Since the earlier studies, many actions have been carried out to improve the treatment and rehabilitation of people with depression in Finland.Access to longer psychotherapies became easier after the 2011 change from discretionary to subjective right to rehabilitative psychotherapy, which tripled the use of the service.Furthermore, the 2012 reform of the Occupational Health Care Act advanced the assessment of return to work and the need for rehabilitation.After the reform, a statement has been required from an occupational health physician if sick leave lasts longer than 90 d.Clinical practice guidelines have also been updated and local treatment regimens have been developed [7,11,12].
Despite the efforts, there is an upwards trend in disability pensions and sickness allowances due to the major depression in Finland, and depression is the most common single reason for disability [13,14].A recent register-based study indicates that treatment-resistant depression is still insufficiently managed in Finland [15].According to the nationwide study, monotherapy was the most common treatment method even in the fifth line of medication [15].Nevertheless, among individuals applying for a disability pension due to depression, purchases of psychotropic drugs appear to increase significantly before applying for a disability pension and decrease after the decision [16].
Currently, there is not sufficient knowledge on how the treatment and rehabilitation of persons with depression is implemented prior to applying for a disability pension.Earlier studies have included small samples and been restricted to medical statements concerning drug therapies.Therefore, we set out to analyse the realised treatment and rehabilitation of persons applying for disability pension due to depression from the Social Insurance Institution of Finland (Kela).

Methods
The disability pension scheme in Finland covers all permanent residents.The earnings-related pension scheme covers disability pension in employee pension insurance companies for those with sufficient work history.If a person has insufficient work history, a disability pension can also be paid by Kela as part of the national pension.Disability pension can be granted only after the applicant's incapacity for work is medically assessed to last longer than 1 year.In shorter periods of incapacity for work, loss of income is compensated with sickness allowance, which is also paid by Kela.Therefore, a sickness allowance period of 1 year usually precedes applying for a disability pension.Applications for both sickness allowance and disability pension are required an enclosed statement from the treating physician.Disability pension can be granted until further notice or as a rehabilitation subsidy for a fixed period.A partial disability pension is not included in Kela's pension benefits.
In this study, we used Kela's register data.We collected data on all new disability pension applications due to major depressive disorder or recurrent depression (ICD-10 main diagnoses F32-F33) with an arrival date between 1.1.2019and 31.12.2019.Applications for both permanent and fixed-term disability pension were included.However, we excluded applications for an extension to a fixed-term disability pension as well as applicants who had received a disability pension from Kela during the previous 12 months.In the case of multiple applications, only the first application with depression as the main diagnosis was included in the data.In association with disability pension applications, Kela's registers hold information on the main diagnosis plus a maximum of two side diagnoses.We excluded applications with recorded side diagnoses of psychoses (ICD-10 diagnoses F20-F29), manic or bipolar disorders (F30-F31), organic mental disorder (F00-F09), and intellectual disabilities (F70-F79).In addition, applicants living outside Finland between 2013 and 2019 were excluded because we examined earlier treatment during the 5 years preceding application, and applied register data contain medication and rehabilitation information only on individuals living in Finland.
In total, 7663 individuals applied for disability pension due to depression in 2019.Of those, 3,882 individuals were excluded due to having received a disability pension from Kela during the previous 12 months, and 54 individuals were excluded due to their side diagnoses (F20-29, F30-31, F00-F09, F70-F79).In addition, the 123 individuals who were living outside Finland between 2013 and 2019 were excluded.Eventually, 604 individuals were included in the analyses.
The characteristics of the study population, such as age and sex, were collected from Kela's population register.All information on treatment and previous sickness allowance days were calculated in relation to the day Kela received the disability pension application.Decisions on the applications were collected from Kela's register on national pension decisions.Data on disability days on sickness allowance 12 months prior to the disability pension application were also gathered from Kela's registers.Sickness allowance is paid due to temporary incapacity for work after a waiting period of 10 working days.We calculated calendar days including the waiting periods.This, however, excludes disability periods of less than 10 working days.
All reimbursed purchases of psychotropic drugs (ATC codes N05-N06) were collected from the year preceding the disability pension application.Drug purchases were calculated in terms of defined daily doses (DDD).DDD is the assumed average maintenance dose per day for a drug used for its main indication in adults [17].
In register-based studies, there are different methods for the construction of drug exposure from drug purchases [18].We selected to estimate medication use by utilising drug purchases in DDDs with the assumption of one DDD per day.Drug usage starts from the date of purchase, and some limitations were applied in order to tune the assumption closer to a more realistic number of days of drug usage.The purpose was to evaluate the number of days of drug usage within the follow-up intervals prior to the disability pension application.In Finland, Kela can reimburse drug purchases equivalent to no more than 3 months of treatment at any one time.Therefore, the duration from each purchase was restricted to 100 d.Since a grace period enhances the performance of the applied DDD method [18], we filled the gaps between subsequent drug purchases of the same drug up to 90 d with the assumption of the drug being in use.Reimbursement from Kela for a new batch is available 3 weeks before the previous batch runs out at the earliest.Therefore, stock piling was allowed for 21 d.
The implementation of drug treatments was described by alternating reimbursed drug purchases and estimated days of drug usage.The prevalence of individuals using any antidepressants (ATC code group N06A), antipsychotics (ATC code group N05A, includes both neuroleptics and lithium), or psychotropic drugs (ATC code groups N05 and N06) was calculated for the previous 12 and 4 months before disability pension application.A period of 12 months was selected because a disability pension is usually preceded by 1 year of incapacity for work, allowing time for therapeutic procedures.Drug purchases from the previous 4 months are a good indication of current medication because reimbursement from Kela is available for 3 months of treatment at a time.We calculated the mean amount of purchases in DDDs.To assess the used dosages, we also calculated the purchases in DDDs in relation to the estimated days of usage among individuals with at least one estimated day of drug usage.Furthermore, we calculated the prevalence of individuals using different numbers of distinct antidepressants during the previous 12 months.We also analysed how many different antidepressant subgroups were used and whether the antidepressants were from the same subgroup in four categories: non-selective monoamine reuptake inhibitors (ATC code group N06AA), selective serotonin reuptake inhibitors (N06AB), monoamine oxidase inhibitors (N06AF + N06AG), and other antidepressants (N06AX).Further, we calculated the prevalence of combined use of antidepressants (N06A) and antipsychotics (N05A), anxiolytics (N05B), or hypnotics (N05C) during the preceding 4 and 12 months before application as well as the mean and interquartile range of days they were in simultaneous use.
Kela arranges both medical and vocational rehabilitation.Vocational rehabilitation is intended for persons whose ability to work or study and earning capacity have significantly declined due to an illness or injury.Psychotherapy is one form of medical rehabilitation, and it can be arranged as rehabilitative psychotherapy or intensive medical rehabilitation.
We analysed the implemented Kela-provided rehabilitation in three categories: vocational rehabilitation, psychotherapy (including both rehabilitative psychotherapy and psychotherapy as intensive medical rehabilitation), and medical rehabilitation other than psychotherapy.We calculated the prevalence of individuals exploiting different rehabilitation services during the 1 and 5 years preceding the application.In addition, we analysed the combined use of antidepressants and Kela's psychotherapy by calculating the share of disability pension applicants receiving either or both of the treatments in the previous year.
Disability days on sickness allowance during the year before disability pension application were categorised into two groups: under 6 months and over 6 months.Receiving specific treatment for depression (antidepressants or psychotherapy) was cross-tabulated with disability days on sickness allowance.A chi-square test was used to calculate the p-value for statistical significance.

Results
There were more female than male applicants for disability pension due to depression; 2336 (65%) were female and 1268 (35%) were male.The study population covered different age groups, the largest of which was the 50-59-year-olds comprising 30% of the study population, 22% were 20-29-year-olds, 19% were 30-39-year-olds and 21% were 40-49-year-olds.Over 60-year-olds comprised 7% and under 20-year-olds 1% of the study population.The mean, median and interquartile range of the age was 42, 43 and 30-54 years, respectively.Nearly half (49%) of the applications were rejected, while 43% of the applicants were granted a fixed-term rehabilitation subsidy, and 8% received a disability pension until further notice.
Table 1 shows the prevalence of individuals using antidepressants, antipsychotics, and psychotropic drugs in the 4 and 12 months before applying for disability pension, mean purchased drugs in DDDs, and mean purchased drugs per estimated drug exposure day.A majority of the applicants had some form of drug treatment for depression before applying for disability pension, as 88.5% had purchased antidepressants and 92.5% had purchased any psychotropic drugs during the 12 months before application.During the 4 months preceding the disability pension application, 80.4% had purchased antidepressants and 85.8% had purchased psychotropic drugs.For the 12 months preceding the application, the average drug purchases in DDDs per estimated use day were 1.6 DDDs/d and 1.8 DDDs/d for antidepressants and psychotropic drugs, respectively.
Table 2 describes the prevalence of individuals using a different number of distinct antidepressants 1 year before applying for a disability pension.Of those, 56.0% had used two or more antidepressants and 24.7% had used three or more distinct antidepressants.Among those who had used two distinct antidepressants, 42.8% used antidepressants from the same subgroup, and 27.3% of those who had used three distinct antidepressants, had all drugs from the same subgroup of antidepressants.
Table 3 presents the prevalence of individuals using antipsychotics, anxiolytics, or hypnotics in addition to antidepressants 4 and 12 months before applying for disability pension.In addition to antidepressants, antipsychotics, anxiolytics, and hypnotics were used in the preceding year by 36.6%, 26.9% and 12.7% of the study population, respectively.

Rehabilitation
Table 4 shows the prevalence of individuals who were granted different rehabilitation services from Kela 1 and 5 years before application.A majority of the applicants had not received any type of rehabilitation, as 75.2% had no rehabilitation during the preceding year and 60.7% had no rehabilitation during the preceding 5 years.The most common form of rehabilitation was psychotherapy; 19.2% and 13.8% of the applicants had received psychotherapy alone or combined with another form of rehabilitation 5 and 1 years before application, respectively.The share of applicants receiving vocational rehabilitation alone or combined with another form of rehabilitation was 16.7% and 8.7% in the preceding 5 and 1 years, respectively.

Combination of depression treatments
Table 5 presents the combined use of antidepressants and Kela's psychotherapy in the treatment of depression during the year before applying for a disability pension.Of the applicants, 12.2% had received both antidepressants and Kela's psychotherapy, 76.3% had received only antidepressants, and 1.6% had received only psychotherapy.Furthermore, 9.9% had received neither antidepressants nor Kela's psychotherapy.There were statistically significant difference in treatment between applicants with rejected application and granted fixed-term rehabilitation subsidy or disability pension until further notice.The share of applicants receiving both psychotherapy and antidepressants were similar regardless of the decision, but there were more persons without any specific treatment and less antidepressant treatment among applicants whose application was rejected.

Sickness allowance
Disability pension applicants had a mean of 228.6 disability days on sickness allowance during the year before disability pension application.The share of individuals without specific drug therapy or psychotherapy was statistically significantly (Chi 2 -test p-value <.0001) higher among those with under half-year disability days on sickness allowance (15.5%) compared to those with over half-year disability days on sickness allowance (7.7%).

Statement of principal findings
Approximately half (56.0%) of the persons who applied for disability pension due to depression had purchased two or more antidepressants during the year preceding the application.Psychotherapy was received by 13.8% and 19.2% of the applicants in the 1 and 5 years preceding the application, respectively.Around a quarter of the applicants (24.8%) had received some form of rehabilitation during the year before application, and 39.3% had received rehabilitation during the past 5 years before application.In the 4 months before application, 19.6% of the applicants had no antidepressant purchases.However, a majority of the applicants had some drug therapies during the year before application; 88.5% had used antidepressants and 92.5% had used psychotropic drugs.Furthermore, the mean purchased amounts of drugs per estimated drug use days 1 year before application were appropriate in relation to assumed average maintenance doses with 1.6 DDDs/d and 1.8 DDDs/d for antidepressants and all psychotropic drugs, respectively.During the year preceding the application, 12.2% of the applicants received both antidepressants and Kela's psychotherapy, whereas 9.9% received neither of the two treatment types.

Strengths and weaknesses of the study
A large register-based data covering all individuals who applied for disability pension due to depression from Kela in 2019 are a significant strength of this study.In addition, the data include information on all drug purchases and rehabilitations reimbursed by Kela.However, the data are restricted to the information included in Kela's registers and, therefore, this study does not contain information on other treatments than medications and Kela's rehabilitation.Other treatments dispensed in healthcare, such as electrotherapies, transcranial magnetic stimulation, transcranial direct current stimulation, invasive neuromodulative treatments, or psychotherapies and other interaction-based treatments, are not covered in this study.Also, our analyses are focused on recent year concerning drug therapies and recent 5 years concerning rehabilitation.Some patients may have earlier treatment efforts.Furthermore, this study is restricted to information on disability pension applications sent to Kela and does not contain information from pension insurance companies.Therefore, our results cannot be directly generalised to the earnings-related pension scheme and individuals with longer working careers.This study concentrates on new disability pension applications and does not include applications for an extension to earlier disability pension.However, earlier disability pensions from pension insurance companies cannot be excluded.
When compared to earlier studies, our research reveals that the treatment and rehabilitation of individuals applying for disability pension due to depression is still insufficient.An earlier study conducted in the context of pension insurance companies in 2012 [10] reported that 34% of the applicants had received two or more antidepressants, while a slightly higher result of 56.0% was found in this study.However, earlier studies have been based on medical statements describing prescribed medications in varying accuracy, whereas this study is based on register data containing all reimbursed drug purchases indicating realised treatment.
One constant appears to be the share of people applying for disability pension due to depression without any antidepressant treatment.Earlier samples from the Finnish national disability pension registers show that 13% of the applicants in 1993-1994 [8] and 14% of the applicants in 2003-2004 [9] had no antidepressant treatment, while Suominen et al. reported a share of 17% without antidepressants.In this study, 11.5% and 19.6% had no antidepressant purchases during the preceding 12 and 4 months, respectively, before applying for disability pension.
Psychotherapy appears to have been applied similarly in earlier studies.According to Suominen et al., rehabilitative psychotherapy was suggested to 11% of the applicants.In this study, the corresponding percentages are slightly higher at 13.8% and 19.2% during the 1 and 5 years before application, respectively.Our data contains information on realised rehabilitation.
Our results of insufficient treatment of depression among disability pension applicants is in line with the findings reported in Lähteenvuo et al. regarding treatment-resistant depression.Lähteenvuo et al. studied the treatment of treatment-resistant depression and found that the most common treatment, even in the fifth treatment line, was antidepressant monotherapy (33.4%), followed by antidepressant combinations (27.5%) and augmentation with mood stabilizers or antipsychotics (24.2%).Although not directly comparable due to methodological differences, our finding of 36.6% of the applicants using both antidepressants and antipsychotic drugs during the year before application and 26.4% using them during the 4 months before application suggests a somewhat higher use of augmentation treatment.However, the data do not reveal whether the applicants used antipsychotics in dosages recommended in augmentation treatment, and some low-dosage use may also occur for sleep disorders.

Meaning of the study
Despite the many efforts to enhance the treatment and rehabilitation of depression, drug therapies and rehabilitation are still realised insufficiently.In total, 19.6% of those applying for disability pension due to depression did not use antidepressants at the time of application, and only one-fourth of the applicants received any form of rehabilitation.The rejection rate of disability pension applications was 49%, suggesting there are likely several individuals whose condition is not severe enough to prevent participating in rehabilitation.Apparently, rehabilitation and drug therapies do not efficiently target those in risk for disability.

Conclusion and implications for further research
A majority of the disability pension applicants received insufficient drug therapies, psychotherapy, or other rehabilitation in accordance with current care guidelines [7].Only a minority received both antidepressants and psychotherapy.This is an unexpected result, considering the burden depression poses on individuals and society and the efforts to improve treatment and rehabilitation.Future research should address the factors and possible reasons associated with the realisation of treatment.

Disclosure statement
Tero Kujanpää has received a grant from the Jalmari and Rauha Ahokas Foundation and works as research physician at Kela and occupational health physician at Terveystalo Ltd.Sauli Jäppinen and Matti Rantanen work as data scientists at Kela.Kirsi Suominen works as a medical advisor at Kela and Ilmarinen Mutual Pension Insurance Company and has worked as chief physician at the city of Helsinki Health and Substance Abuse Services.Veera Pohjolainen works as a medical advisor at Ilmarinen Mutual Pension Insurance Company and as a part-time consultant psychiatrist at Terveystalo Ltd.She has also worked as chief physician at the city of Helsinki Health and Substance Abuse Services.During the past 3 years she has received speaking fees from Janssen and H. Lundbeck.Riitta Luoto works as medical director in Research at Kela.

Table 1 .
Prevalence of persons using antidepressants, antipsychotics, and psychotropic drugs, the average purchases, and estimated use during 4 and 12 months before applying for disability pension due to depression.
a Mean purchased drugs measured in defined daily doses.bEstimatedmean days the drugs were in use.cMean purchased drugs as defined daily doses divided by estimated use days among those with at least one exposure day.

Table 2 .
Prevalence of Persons using a different number of distinct antidepressants during the 12 months before applying for disability pension due to depression.

Table 3 .
Prevalence of persons using antipsychotics, anxiolytics, or hypnotics in addition to antidepressants and estimated mean days and interquartile range of simultaneous use during 4 and 12 months before applying for disability pension.
a Estimated mean days of simultaneous use with antidepressants among those with at least one use day.b interquartile range of estimated days in simultaneous use with antidepressants.

Table 4 .
Prevalence Vocational rehabilitation and psychotherapy and other medical rehabilitation.
a Medical rehabilitation other than psychotherapy.b Vocational rehabilitation and psychotherapy.c Psychotherapy and other medical rehabilitation.d Vocational rehabilitation and medical rehabilitation other than psychotherapy.e

Table 5 .
Prevalence of Persons using antidepressants and Kela's psychotherapy during the year before disability pension application among all applicants and separately for applicants with rejection and granting.
a all disability pension applicants due to depression.b disability pension applicants who were granted fixed-term rehabilitation subsidy or disability pension until further notice.c disability pension applicants, whose application was rejected.d chi 2 -test p-value for difference between granting and rejection groups.e No psychotherapy or antidepressant use during the past year.f antidepressant drug use but no psychotherapy during the past year.g Psychotherapy but no antidepressant use during the past year.h Both antidepressant use and psychotherapy during the past year.