Use of multiple services from childhood to early adulthood among violent juvenile offenders: a document-based descriptive study

Abstract

Background

Young violent offenders frequently face significant developmental adversities, including learning difficulties, mental disorders, and substance use during childhood and adolescence. Many also grow up in circumstances requiring child welfare interventions. However, limited information is available regarding the services provided to them.

Objective

To describe the timing, duration and components of school support services, child welfare services, as well as mental health services and substance abuse services used by young violent offenders.

Participants and methods

Data were derived from the forensic psychiatric examination statements of 42 Finnish violent offenders aged 15–22 years and analyzed using qualitative and quantitative content analyses.

Results

Almost all young people (95%) had used at least one service. Service use typically began between ages 7 and 12, with school support services accessed more frequently than other services during this period. The highest rate of single-service use occurred at ages 8–9. Adolescence was characterized by the most extensive multiservice involvement, with the combination of school support, mental health, and child welfare services peaking at age 15. Interruptions longer than one year were more frequent in mental health services than in other services. Evidence-based parent training, multidimensional family therapy, and specialized placements for youth with substance use and mental health problems were used only seldom.

Conclusions

Most youth who committed violent offences had received multiple services during childhood and adolescence. Our results highlight the need to develop the timing, suitability and acceptability of services for this specific group of young people in their youth.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13034-026-01140-2.

Introduction

The World Health Organization defines interpersonal violence as “The intentional use of physical force or power, threatened or actual, against another person, that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment or deprivation [1]. Young age is a significant risk factor for violent offending, particularly for males [2].

Previous research has shown that the issues faced by young people who have committed violent offences are diverse and often long-standing. Mental health disorders are common, including conduct disorders that often emerge in childhood [3]. A related symptom, bullying others, has shown to be a strong predictor of violent criminality later in life [4, 5]. In addition, learning and school-related difficulties are more prevalent in this group than among their peers, and placement outside the home is also common [4]. Substance use is another risk factor that is found more often among young violent offenders compared to other young people [3, 5]. Adverse childhood experiences cumulatively increase the risk of committing violent crimes [6–9]. It is also well established that adverse experiences constitute a risk factor for learning difficulties as well as mental health and substance-use problems, and they often underlie placements made by child welfare services [10–12]. Taken the numerous risks and disorders, young people who have committed violent offences often require support from multiple services throughout their childhood and adolescence.

However, limited information exists regarding the support and treatments provided to young violent offenders to address their difficulties during childhood and adolescence. Predominantly retrospective studies conducted in England, Finland, and Sweden provide varying estimates on the proportions of users of different services. Reported rates suggest that 23–75% have received special needs education or other educational support, and 19–76% have experienced out-of-home placement [4, 13–15]. However, these studies do not clarify whether some individuals in need remained unreached by services. Evidence further indicates that 40–79% of violent offenders aged 15–25 have used child and adolescent psychiatric services at some point [3, 14, 16]. In the Nordic context, where mental health services are generally accessible, barriers to utilization may reflect not only structural factors but also difficulties in treatment engagement and adherence [17]. Moreover, mental health service use is uncommon in the months immediately preceding a violent offence, and substance abuse treatment within the 12 months prior to the offence appears similarly infrequent [9, 14, 16, 18].

Some studies have reported the services used by young violent offenders during childhood or prior to their violent acts; however, these findings have been presented separately, without analyzing the co-occurring use of multiple services (e.g., 14, 15). Combinations of multiple service use have been studied among adolescents requiring treatment for behavioral symptoms. In two studies 12–29% of youth had a high level of involvement with school services, child welfare services, mental health services, and the juvenile justice system during the 6–12 months preceding treatment [19, 20]. However, the majority (71%) had only used school services intensively, but the other three services moderately or with low intensity [20]. In terms of criminal development, among another at-risk group, adolescents with adverse childhood experiences, about one in ten (9.5%) used intensive mental health services (e.g., inpatient psychiatric treatment) and school-based services (e.g., special education classes) during the 30 days preceding the study [21].

Detailed evidence on the extent young violent offenders have accessed targeted services for their difficulties during childhood and adolescence—prior to committing a serious violent offence—remains limited. This study aimed to increase knowledge about their service use by exploring and describing the timing, duration and components of school support services, child welfare services, mental health services, and substance abuse services used by young violent offenders prior to their violent crimes. The study is part of a series of studies conducted under the Forensic Psychiatry 2020 Initiative by the Finnish Institute for Health and Welfare. In Finland, criminal responsibility begins at age 15. Forensic psychiatric examination is conducted by court order in cases where the court deems it necessary to assess a violent offender’s criminal responsibility against not being guilty by reason of insanity. Based on a multiprofessional inpatient evaluation over a period of approximately two months, a final statement is prepared by a forensic psychiatrist. It includes not only the results of the examination but also condensed data gathered from the subjects and their family of origin, and various original documents from services provided to the young person (e.g., school support services (SCHS), child welfare (CWS), mental health services (MHS), and substance abuse services (SAS)) during childhood, adolescence, and adulthood, prior to the violent offence that led to the examination [22]. A description of SCHS, CWS and SAS in Finland can be found in Additional file 1.

Methods and subjects

Study design

This study was conducted as a retrospective document-based file review using forensic psychiatric examination statements as the primary data source. The aim was to systematically describe the timing, duration, and components of school support services (SCHS), child welfare services (CWS), mental health services (MHS), and substance abuse services (SAS) used by young violent offenders from childhood to early adulthood. The material was analyzed using a deductive qualitative content analysis supplemented by a quantification of service use [23, 24].

Methods

Data collection

Details of the data collection process have been described in a previous publication [16]. To ensure that the services examined were aligned with current practices, statements were selected in order of increasing age, beginning with individuals who were 15 at the time of their forensic psychiatric examination. Ten statements were excluded due to prior residence abroad or the use of services intended for individuals with intellectual disabilities. The final dataset comprised 42 statements about the youngest violent offenders (aged 15–22) written between 2015 and 2019. This textual data included 1,141 pages in PDF format (the length of a single statement ranged from 16 to 37 pages, with a mean of 27 pages).

Deductive content analysis

In this study, we included data on service use. Reliability was improved by EH and RL independently coding deductively three statements, then comparing the coding, and discussing differences until reaching consensus. Then, EH coded 42 statements, as per Table 1.

Table 1.

Categories, subcategories and codes describing the services

Categories	Subcategories	Codes
School support services	Extending time for studies	More time prior to compulsory school More time during school years More time prior to vocational school
	Additional support for learning	More intensive teaching Reduction in the level of requirements
	Adapting the learning environment	Reduction of the size of learning group Increasing support from adults Reduction of length of school day Changing the learning environment
Child welfare services	Open care support services	Family work Support out of home Other
	Residential care	Placement in open care support Custody
Mental health services	Primary mental health services	Examination Treatment Non-defined
	Specialized mental health services	Examination Treatment
Substance abuse services	Out-patient services	Student healthcare, low threshold youth service, community health center Organizations specialized in substance abuse service Type of service
	In-patient services	Substance abuse institution Type of service

Quantification of service use

For the study of service use duration, mentions of the duration were coded in years and months. When the duration of service use was not specified in the data, it was derived from the documented start and end dates, when available. Precise information on the duration of service use in years and months was thus available for 11 users of CWS, 11 users of MHS and seven users of SAS. For 14 subjects, complete information on the support received throughout compulsory education was available, although the timing of SCHS was reported only as the grade level at which the support was provided.

All individuals who had used any service (n = 40) were included in analyses of: (a) age at the initiation of service use, (b) the age when using multiple and co-occurring services and (c) service interruptions lasting longer than one year. If the age (in full years) was not mentioned, it was calculated using the subject’s date of birth and the date of the onset of service use, other descriptions of the timing of the service use, or determined based on the subject’s school grade and school history (for SCHS). Interruptions were determined based on the age when using the service. Multi service use was defined as using more than one service during childhood and adolescence. Co-occurring service use was defined as using more than one service at the same age, not necessarily overlapping. In the analysis of multiple and co-occurring services, any indication of service contact within a given year was coded as the age at the time of use.

Subjects

Descriptive information on the individuals involved in this study has been reported earlier [14] and is presented in Table 2. The offences committed that led to a forensic psychiatric examination, were violent offences: murder or manslaughter (including attempted forms), assault, or severe sexual offences involving physical contact. Additionally, 24 (57%) had committed other types of offences. Eight individuals were minors and 34 were in emerging adulthood (age-range 15.5–22.6 years, mean 19.2 years). At the time of committing the offences the individuals were 15.0-20.6 years old (mean 18.3 years).

Table 2.

Descriptive characteristics of the 42 young violent offenders

	%
Gender
Male	79
Female	21
Education
Ongoing vocational training or high school studies or already graduated or obtained a degree	26
Completed compulsory education only	62
Ongoing compulsory education*	12
Prior convictions (before index offence**)
No	74
Yes	26
Family structure
Both biological parents in the family	28
Single biological parent/other parent deceased/divorced parents	72
Perinatal and developmental difficulties, childhood adversities
Perinatal and developmental (motor, speech and language, learning) difficulties	71
Childhood adversities***	95

*Some due to their age

**Index offence: offence that led to a forensic psychiatric examination. Almost all prior convictions were suspended imprisonment 2–17 months, other: fines, unsuspended imprisonment for two months, or unknown

***Childhood adversities by Kessler et al. 2018 [37], in addition: being a victim of bullying and death of a sibling, friend or supportive adult

The majority of the individuals had completed only compulsory education. For most individuals, the family structure was characterized by the presence of only one biological parent and often by multiple changes. Perinatal and developmental difficulties, as well as childhood adversities were common among the individuals. All except one individual was diagnosed with at least one ICD10 F- diagnosis (International Statistical Classification of Diseases and Related Health Problems, WHO 10th edition). The most common diagnoses were disorders of adult personality and behavior (F60–F69; 69%) and mental and behavioral disorders due to psychoactive substance use (F10–F19; 57%).

Results

Timing, consistency and duration of service use

Nearly all individuals (40, 95%) had used at least one service. The first service use most commonly began between ages 7 and 12. Service use was not always continuous, with gaps observed among some individuals, particularly in mental health services. (Table 3.)

Table 3.

Service use: age at initiation and gaps

	%
Age at service initiation*
< 7 years	25
7–12 years	55
≥ 13 years	20
≥ 1 year gap in service use
SCHS (n = 32)	25
CWS (n = 31)	23
MHS (n = 33)	58
SAS (n = 10)	<50

*All service users, n = 40

The average cumulative use of services, excluding substance abuse services, could extend over several years; this was calculated for the 27 individuals with available data, with service-specific numbers shown in Table 4.

Table 4.

Cumulative duration of service use (years)

Service	Range (median; mean)
School support services (n = 14)	6–10* (9; 8.6)
Child welfare services (n = 11)	3.1–15.9 (8.4; 8.75)
Mental health services (n = 11)	1.25–10.3 (5.4; 5.8)
Substance abuse services (n = 7)	0.25–1.9 (0.7; 0.9)

*Compulsory education only; support reported by grade level rather than duration (e.g. weeks or months)

The proportions of service users by age are illustrated in Fig. 1. SCHS were provided to over one-third (38%) of all individuals during their first school year, at the age of seven. The utilization of SCHS peaked among those aged 14 to 15 (57%). For some individuals, CWS use was initiated during the first year of life. The utilization of CWS increased until age 15 when the highest proportion of CWS users (55%) was observed. The individuals began utilizing SAS at age 13, with the highest use observed at age 17. The utilization of mental health services (MHS) mirrored the trends observed in CWS and SCHS use until age 15. Thereafter, the utilization of all services began to decrease, with the steepest decline observed in SCHS. A final increase in the use of both MHS and SAS was noted at age 20. At this age, half of the individuals were identified as first-time users of MHS.

Fig. 1.

Use of school support services, child welfare services, mental health services and substance abuse services by age among 42 violent offenders aged 15–22 years, %

y= years; 0-15-year-olds: n = 42, 16-17-year-olds: n = 39, 18-year-olds: n = 34, 19-year-olds: n = 24, 20-year-olds: n = 18, 21-year-olds: n = 9, 22-year-olds: n < 5. The curve illustrating mental health service use was previously published in BMJ Open [16].

Use of multiple services and co-occurring service use

Among those who used multiple services, at least one period of co-occurring service use was observed; however, in each group, one or two individuals had consecutive service periods without complete overlap. The use of co-occurring services was most frequent during adolescence, and the most common pattern was the use of three services (Table 5).

Table 5.

Multi service use among 40 service users

Number of services	Users N = 40 %	Peak age (user proportion, %)	Services
One service	17.5	8–9 years (43)	SCHS roughly twice as common as MHS and CWS
Two services	17.5	16 years (31)	All combinations of SCHS, MHS, and CWS roughly equally common.
Three services	47.5	15 years (29)	SCHC + MHS+ CWS most common; SAS as a third service rarely
Four services	17.5	14 years (7)	SCHC + MHS+ CWS + SAS

Providing two services simultaneously within the same service event was uncommon; an example was family work conducted jointly by CWS and MHS. A collaboration between SCHS and CWS was noted in the form of a small class offering tailored education. CWS incorporated specialized expertise from another service in institutions specializing in mental health or substance use disorders, where a few individuals were placed. Specialized expertise from all four services was provided to individuals in reform schools.

Changes in service providers

Many individuals experienced changes in service providers. During the lower grades of compulsory comprehensive school education (grades 1 to 6, ages 7 to 12) more than two in five (43%) attended multiple schools, most commonly two, with a maximum of six schools. During upper grades (grades 7 to 9, ages 13 to 15), nearly one in three (29%) attended two or three schools, and one in six (15%) utilized MHS in three or four healthcare organizations. The most common reason for such changes was arranging additional support for the individual, such as changing school when a special class was located elsewhere. Changes in CWS, such as out-of-home placement or changes in such placements, could also cause discontinuities in schooling and SCHS as well as MHS. Family relocation was reported only once as a reason for discontinuation of services.

Components of services

A significant majority received school support services, child welfare services and mental health services (Table 6).

Table 6.

School support services, child welfare services and mental health services provided to 42 violent offenders (%)*

	%
School support services	76
Extending time for studies (repeating preschool/grade, preparatory education for vocational studies)	45
Additional support for learning (tutoring, special education full/part time, subject adaptations, individualized education plan)	43
Adapting the learning environment (small/special education class, personal assistant/assistant in the class, study carrel, special arrangements for tests, spending recess indoors, shortened school day, temporary placement in home-based education, school in child welfare institution, reform school)	60
No use of services	24
Child welfare services	74
Open care support services (support family, family oriented social work, intensified family social work, visits to social worker, therapeutic or other professional support person, family rehabilitation, economic support, supported housing, out-of-home placement, non-specified)	38
Residential care (foster family, professional family home, children’s home, small group home, reform school)	60
Assessment of need for child welfare services without case initiation	17
No use of services	26
Mental health services	79
Examinations (e.g., psychological assessments, structured interviews)	64
Treatment: Psychosocial interventions (home visits, family visits, individual’s visits to professionals, occupational, music, art and psychotherapy, MDFT**	69
Treatment: Medication (ADHD-medicines, antipsychotics, antidepressants, sleeping aids, sedatives)	67
Urgent examination and treatment	33
No use of services	21
Substance abuse services	24
Outpatient substance abuse withdrawal and rehabilitation (e.g. nicotine withdrawal program, MDFT**)	14
Inpatient substance abuse withdrawal and rehabilitation	12
No service use	76

*Various types of support can be provided to an individual in each service. **Multidimensional Family Therapy

School support services

Additional support for learning and adaptation of the learning environment occurred throughout education (Table 6). Special education services were offered as frequently to students with both learning difficulties and behavioral challenges as to those with only one of these issues. One in two users of special education studied in a small class for at least some time, with some continuing throughout their school years. At ages 13 to 15, the most intensive forms of educational support, such as education in reform schools, were used by some individuals. After compulsory comprehensive education, nearly one in four individuals initially enrolled in preparatory education for vocational qualifications. Despite this, vocational studies were susceptible to early termination. Of those who were old enough to complete a vocational qualification, approximately one in five achieved the qualification, yet many required additional support.

Child welfare services

The use of child welfare services began at the age of 7–12 for nearly half of the individuals, for one in four at the age of 13–15, and was less common for preschool- aged children and those over 16 (Table 6). Families were supported in child and youth open care through family work, intensive family work, family rehabilitation, and parental therapy, as well as financially. Support families, support periods in professional family homes, and individual therapy were arranged for the children. The relationship with a support family could extend over several years. Additionally, support for hobbies and supported vacations were mentioned. In youth services, network work, individual support visits, and support people were mentioned. One in four individuals were placed out-of-home as a support measure in open care. Supported housing was organized starting from the age of 16, and young adults were provided with individualized support from a social worker and financial assistance. More than one in two CWS users experienced being taken into care and placed outside the home. These were most common during school age, occurring equally often between the ages of 7–12 and 13–15. They were often preceded by one or more emergency placements. More than one in three were initially in foster or professional family care but were later transferred to a child welfare institution. Repeated absconding from the child welfare institution frequently resulted in a change of placement. For some families using CWS, a support plan was developed but not implemented. In all but one case, the reason cited was the parents’ opinion of the service offered.

Mental health services

Examinations and treatments were primarily conducted in specialized healthcare settings, although one in three utilized either only primary care MHS or both primary care and specialized MHS (Table 6). Of the MHS users, two in five received inpatient care. Psychosocial treatments and psychotropic medication were roughly equally common, and were reportedly utilized by two out of three individuals. Psychosocial treatments often included scheduled visits to a professional, with more than half of the cases involving a psychiatric nurse. Family visits and parental support visits were reported for one in five individuals, along with various therapeutic interventions that occurred with equal frequency in childhood and adolescence. ADHD-medications were prescribed from the age of six, the antipsychotic medication risperidone from the age seven, quetiapine from the age of 12, and antidepressants during adolescence.

More than one in three individuals visited an emergency care unit for a mental health assessment. In cases involving children or adolescents, planned voluntary crisis inpatient care was primarily provided. If the individuals were placed outside the home, they often returned to a child welfare institution. However, many individuals were involuntarily referred for psychiatric hospital observation from the emergency assessment, although involuntary inpatient treatment was deemed necessary for only a few subjects. Some individuals experienced repeated observations or voluntary short-term hospitalizations over a relatively short period and committed violent offences within days or weeks after discharge.

Substance abuse services

Both outpatient and inpatient services were provided mostly in private foundation-based treatment facilities for adolescents with substance use disorders (Table 6). In some cases, outpatient services were delivered in student healthcare settings, and inpatient services were provided in child welfare institutions or psychiatric units specialized in substance abuse treatment. Descriptions of the content of treatments were sparse.

Discussion

This retrospective study examined the timing, duration and components of school support, child welfare, mental health, and substance abuse services used from childhood to emerging adulthood among individuals who later committed violent offences. The main finding was that nearly all individuals had used one or more services during their childhood and adolescence. Periods involving the use of multiple services were common, and the support or treatment received in one or more services had been intensive for more than two in three individuals. However, individual service paths were often inconsistent, with variable use of services and gaps without any service use. Thus, a key question concerning the use of these services appears to be their suitability and acceptability by the service users.

Multi service use

The use of three or sometimes four services was most common among adolescents. This finding aligns with the fact that the peak age of onset of mental disorders is 14.5 years, and the first symptoms of personality disorders are exhibited at this age and onwards by most individuals who will develop these disorders in the future [25, 26]. Given that a substantial portion of the individuals exhibited conduct-, ADHD- or multiple type symptoms [16], the finding is not unexpected, since youth with behavioral symptoms typically need not only MHS but also SCHS and, in many cases, CWS. This aligns with previous reports on subgroups with high involvement in multiple service use among children and adolescents with behavioral symptoms, conduct disorders and adverse childhood experiences [19–21, 25]. While obtaining the necessary services is essential, our findings support the view that the quality of these services plays a central role in mitigating risks [27, 28].

A prominent feature of service use was that in all services, the organization or the professionals providing the services may change during the service need. Partly, these changes are related to the structure of the service system, such as the division of labor based on the severity of issues, as seen in each service. For instance, in child welfare, behavioral and mental health problems are linked to changes in the placement of children and adolescents [29]. Additionally, both school support and mental health services involve age-based transitions to organizations serving the next age group, as well as reaching adulthood in child welfare. Consequently, when the individual’s needs were complex and long-lasting, the likelihood of changes in service providers over time across all services increased. The accumulation of numerous professionals providing support or care can reduce confidence in receiving effective help [30, 31]. This situation may have been particularly harmful to children and young people in the sample, as many experienced multiple changes in their family structure during their upbringing. The transition from adolescent psychiatry to adult psychiatry is notoriously prone to treatment discontinuation [32, 33]. It would be appropriate to conduct more research into the experiences of also users of other than psychiatric services regarding how transitions and changes in staff and organizational structures affect service utilization.

The limited collaboration between services may have partly contributed to the decrease in service usage after age 15 and in emerging adulthood. Young people often wish for comprehensive support for their situation rather than help that focuses solely on one problem [34]. Additionally, the variation in the intensity of services may also affect young people. Many young people engaged in behaviors that endangered their health and development, and in such situations, urgent measures by CWS and MHS were needed. However, following strong support during a crisis, a young person in need of long-term support may feel abandoned as the service continues with less intensity post-crisis [27].

Important questions arise: can the current service system promote the mental health of such vulnerable children and adolescents and treat their mental health disorders, as well as protect them from further traumatization?

Strengths and limitations

A major strength of this study is the unique dataset on multiple service use from childhood to emerging adulthood among young violent offenders. The data consists of statements based on Finnish forensic psychiatric examinations, which are comprehensive, structured evaluations and are considered reliable [35]. The data coverage is robust, including all 42 written statements in the national archive of forensic psychiatric examinations from 2015 to 2019 concerning the youngest offenders. Through a qualitative approach, a comprehensive understanding of the quality and quantity of service use was achieved. A significant limitation, however, is that due to the strict confidentiality of the data only one author was responsible for the hands-on coding of most data, albeit under the guidance of two senior researchers. Additionally, the data was originally created for another purpose, specifically for the court to assess criminal responsibility against not being guilty by reason of insanity. As with all document-based studies, a significant limitation of the data is that the absence of a particular mention may indicate either that the issue was not present, or it may indicate missing data. There were some deficiencies in the content of the statements; for example, the duration of service usage could only be determined for a subset of young people. The description of the concurrent use of services is age-based and does not indicate the duration or intensity of service use. Additionally, the absence of active contact does not necessarily indicate the termination of a service relationship. Information about the relocation of families was not comprehensive. The data also did not allow for determining whether arrests related to youth offenses hindered their use of services. However, in prisons and detention facilities there are healthcare and psychiatric services available. Moreover, the reporting of the contents of the services was inconsistent. The data contained no information on possible mental health-promoting or supporting actions implemented in early childhood education services or by student welfare staff during school age. The sample size was small but sufficient to yield reliable results with the methods used [36]. Most of the young people were male, and due to the small proportion of females, potential gender differences in the use of services were not analyzed. As the data was based on young violent offenders referred to a forensic psychiatric examination, the results do not represent the general juvenile offender population. We also do not know how these four services have been used by young people with similar backgrounds and problems, but who have not committed a violent crime. Future studies should therefore include a comparison group. Despite these limitations, valuable information is provided on the support offered to future young violent offenders during childhood, adolescence and young adulthood prior to committing a violent act.

Conclusions

Youth who commit violent offenses receive multiple and often co-occurring services for years during childhood and adolescence. However, their service paths are frequently inconsistent, with variable use of services and gaps without any service use. Services for families and children with multiple needs should be better integrated. They should be developed by utilizing the experiences of children, adolescents, and families who use the services, thereby increasing the suitability and acceptability of the services. Future research should investigate in more detail how services respond to identified risks and the recognized need for multiple services.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

EH and AA-R contributed to the planning and data management of the study. EH and RL collaborated on creating the categories, with EH focusing on coding and analyzing the data, supervised by RL and MM. EH drafted the original manuscript, and all authors participated in its review and editing.AI assisted in refining the English, all authors reviewed and edited the text, and a professional language editor provided stylistic suggestions incorporated by the authors.

Funding

Open Access funding provided by Finnish Institute for Health and Welfare. No funds, grants or other support were received.

Data availability

The datasets analysed during the current study are not publicly available due to strict confidentiality orders concerning the data. Additional information about the availability of the data can be obtained from Findata. [Data permits - Findata](https:/findata.fi/en/permits/data-permits).

Declarations

Ethics approval and consent to participate

The study was accepted by the research ethical committee of the Finnish Institute for Health and Welfare (January 14th, 2020).

Consent for publication

Not applicable. By law, informed consent is not required in Finland for register-based studies if the registered people are not contacted.

Competing interests

The authors declare no competing interests.