Umeå University's logo

Mot särskild bakgrund av den ökande förekomsten av psykisk ohälsa i sjukfrånvarostatistiken, samt de svårigheter som detta medför för arbetsgivare vid bedömning av arbetsförmåga och behovet av arbetsanpassning och rehabilitering, analyseras i artikeln arbetsgivarens rättsliga ansvar att bedöma arbetstagares arbetsförmåga i enskilda fall. Särskilt fokus läggs på i vilken utsträckning detta ansvar kan aktualisera en skyldighet för arbetsgivaren att anlita företagshälsovård för att bistå med bedömningen. Analysen visar att kravet på individuella och proportionella bedömningar, anpassade till arbetsgivarens ekonomiska och organisatoriska kapacitet, utgör en medveten kompromiss mellan flexibilitet och tydlighet, liksom speglar en uppfattning att arbetsförmågebedömningar utgör ett område som lämpar sig dåligt för precisa gränsdragningar i regleringen. Regleringsmodellens utformning skapar emellertid en onödigt stor brist på tydlighet i hur företagshälsovården böreller skaintegreras i arbetsgivarens arbetsförmågebedömningar.

Sickness absence due to psychiatric diagnoses, particularly stress-related mental health conditions, has become the most common cause of sick leave in Sweden since 2014. These diagnoses are linked to the longest durations of absence and the highest risk of not returning to work, with women being overrepresented. Psychiatric disorders now account for a third of new and half of all ongoing sick leave cases. To receive sickness benefits, a person’s work capacity must be reduced by at least 25% due to illness. However, assessing the extent of this reduction, particularly for psychiatric conditions, is challenging due to their multifactorial nature, including biological, psychological, and social factors. These complexities often lead to inconsistent evaluations of work capacity. This article analyses the legal framework for assessing work capacity in cases of psychiatric illness, combining traditional legal sources with research and guidelines from relevant authorities to explore the specific challenges involved in these assessments.

In 2025, the Court of Justice of the European Union (CJEU) ended the extensive debate on whether the General Data Protection Regulation (GDPR) includes a right to an explanation when a data subject is digitally profiled and a decision is made based on the profile. The Court also addressed the level of detail required in the explanation and several other issues. However, while the Court has settled the existence of an explanation duty, it has not exhausted its scope. Post the recent case law, this article argues that the unresolved centre of gravity in the GDPR’s “right to explanation” is no longer what must be explained about the data processing and profiling. However, the rulings leave several research questions unanswered, including whether – and if so – how far the explanation must extend to external norms, which in the public sector are usually statutory provisions and their interpretation and operationalisation in automated decision rules applied to a set of personal data, including profiles, thereby forming a decision regarding the data subject.

Occupational health services are fundamentally mandated by the employer, who both chooses and pays for the services. This creates a potential conflict between acting in the interests of the employer and meeting the support and intervention needs of employees. According to Chapter 3, Section 2 c of the Swedish Work Environment Act, occupational health services are to act as an independent expert resource in matters of work environment and rehabilitation. The article shows that despite this regulation, there are no direct mechanisms to ensure compliance, which makes the principle of inde-pendence weak from a legal standpoint. It is, therefore, argued that lawma-kers should consider introducing explicit legal requirements for occupational health services to maintain their independence and for employers to be obli-ged to respect this independence, regardless of organisational structure. A regulation does not automatically guarantee an independent operation, but it represents a symbolically and legally significant measure to promote and protect such independence. The legitimacy of occupational health services fundamentally depends on their ability to maintain independence.

This thematic section explores oversight not as a discrete instrument, but as a governance architecture embedded in law, institutions, and technical systems – and enacted in practice. Whether technology is the object of oversight or the means by which it is carried out, accountability ultimately remains anchored in public authority and attaches to identifiable institutions and persons, even where execution is automated. Reasons must be provided, decisions explained and reviewable, and mistakes corrected. The contributions gathered here map that responsibility across legal bases (such as the GDPR, the AI Act, administrative law, freedom of information etc.), across organisational roles (such as providers, deployers, regulators, auditors), and across jurisdictions and regulatory sectors. They also show that effective oversight must operate both preventively and reactively. It begins in design and continues in review, binding what happens in a single file to the standards and interfaces that govern the system as a whole, and connecting duties in the individual case with architectural choices, interoperability arrangements, and impact-assessment practice.

In Sweden, mental illness has become the leading cause of sick leave, most often involving so-called common mental disorders such as depression, anxiety, and stress-related conditions.[1] Despite being widespread, these disorders pose particular challenges within the sickness insurance system, especially where medical and legal uncertainties intersect. One reason is that eligibility criteria for sickness benefits (sjukpenning) are built on the assumption that medical, social, and other factors affecting work capacity can be separated and clearly defined. In cases of mental illness, however, such distinctions are rarely straightforward, since causal links and boundaries are often blurred.

In Human Needs and the Welfare State, Bent Greve undertakes the ambitious task of addressing, in only just over 140 pages, some of the most intricate and profound questions underpinning the architectures of welfare states – namely, who are the needy and do we know what their needs are? This approach is reasoned on the basis of a perceived gap in welfare state discussions: although welfare state provision has expanded significantly over time and has come to encompass a varying but wide range of responsibilities, there has been little reflection on whether we truly understand who the needy are and whether their needs are genuinely being met.

Whereas hackathons are widespread within and outside academia and have been argued to be a valid pedagogical method for teaching interdisciplinarity, no detailed frameworks or methods are available for conceptualizing and organizing educational hackathons, i.e., hackathons dedicated to best achieving pedagogic objectives. This paper is dedicated to introducing EDUCational Hackathons for learning how to solve Interdisciplinary Challenges (EDUCHIC) through: (1) defining the fundamental principles for framing an activity as an EDUCHIC, integrating principles from pedagogical methods, hackathon organization, and interdisciplinarity processes; (2) describing general properties that EDUCHIC possess as a consequence of the interaction of the fundamental principles; (3) developing operational guidelines for streamlining the practical organization of EDUCHIC, including an exhaustive end-to-end process covering all the steps for organizing EDUCHIC and practical frames for carrying the key decisions to be made in this process; and (4) a demonstration of these guidelines through illustrating their application for organizing a concrete EDUCHIC.

This article examines the integration of data analytics and automated fraud-detection systems within welfare sectors, focusing on the evolving legal landscape in Europe. Throughout Europe, the use of advanced technologies to monitor welfare payments has intensified. However, large-scale data profiling raises significant legal challenges, exemplified by the judicial scrutiny of the Dutch SyRI system, which was found to violate the European Convention on Human Rights and the General Data Protection Regulation (GDPR). This study addresses a pivotal question: whether the use of such systems under EU law constitutes law-enforcement activities or routine administrative operations. This classification impacts the legal framework—GDPR or the Law Enforcement Directive (LED)—governing data processing practices, and influences obligations under the forthcoming EU Artificial Intelligence Act. Through an analysis of European law, including GDPR, LED, the EU Charter of Fundamental Rights, and relevant case law, this article highlights the complexities of this legal demarcation and the potential for conflicting regulatory incentives. Swedish examples are used to illustrate how different legal and institutional setups affect these issues, offering insights into the broader implications for the legality and governance of fraud detection systems.

his chapter explores aspects of the relationship between technology, transparency, and accountability in public decision-making. It addresses how technological advancements have increased accessibility and automation while complicating decision process reviewability. It explores transparency as a relational concept and focuses on legal obligations on documentation and records-keeping, such as in the EU General Data Protection Regulation and the upcoming EU Artificial Intelligence Act, as a means to bolster transparency and improve reviewability. In particular it also discusses the feasibility of gathering and analysing ‘paradata’—data pertaining to data processes—as a means to safeguard legality and transparency in automated decision-making, notably within the public sphere.