Freedom of Information legislation has proliferated globally over recent decades, with over 150 countries now maintaining some form of access-to-information law (Ackerman and Sandoval-Ballesteros, 2006; Michener, 2015). These legal frameworks typically aim to enhance democratic accountability by providing citizens and civil society organizations with mechanisms to access government-held information. However, the mere existence of FOI laws does not guarantee meaningful transparency in practice.

Substantial research demonstrates significant gaps between legal provisions and practical implementation across diverse democratic contexts (Grimmelikhuijsen et al., 2019; Hazell and Worthy, 2010; Roberts, 2006). Administrative resistance, resource constraints, and institutional cultures that prioritize secrecy over openness frequently undermine FOI effectiveness (Cuillier, 2019; Walby and Luscombe, 2019). These implementation challenges suggest that formal legal frameworks require supportive administrative cultures and adequate institutional capacity to achieve their democratic objectives.

Journalists represent particularly important FOI users, given their role in facilitating public accountability through investigative reporting. However, research in the United States indicates that journalists submit only a small proportion of total FOI requests in many jurisdictions (Silver, 2016; Hamilton, 2018). Hamilton's analysis of United States FOI usage reveals declining journalist participation, whilst requests from specialized media outlets serving commercial interests have increased substantially. This trend raises concerns about whether FOI systems effectively serve their intended democratic functions or primarily benefit commercial actors with greater resources for navigating complex request processes.

Data journalism has emerged as an increasingly important subset of investigative reporting, utilizing large datasets and analytical tools to reveal patterns hidden within vast information repositories (Appelgren and Salaverría, 2018; Fridman et al., 2023; Tong, 2022). This approach depends critically on access to machine-readable, comprehensive datasets, requirements that many government agencies struggle to meet due to outdated record-keeping systems and challenges to invest in data-sharing infrastructure.

Building on Žuffová's (Žuffová, 2023) empirical investigation of Czech FOI compliance, which revealed how administrative culture and blame-avoidance behaviors constrain transparency despite robust legal provisions, we adopt a similar analytical lens to examine Norwegian municipal practices. Žuffová's study demonstrated that legal frameworks alone prove insufficient for ensuring meaningful transparency; effective FOI implementation requires administrative willingness, adequate resource allocation, and institutional cultures that encourage rather than resist openness (Žuffová, 2023).

Blame-avoidance theory provides crucial insights into why public officials may resist information disclosure even when legally required to comply with FOI requests (Hood, 2011). Officials facing potential criticism for revealed shortcomings may adopt defensive strategies, including partial disclosure, procedural delays, and broad interpretation of legal exemptions (Honig et al., 2023; Hood, 2011; Pasquier and Villeneuve, 2007). These behaviors reflect rational responses to institutional incentives that punish visible failures whilst providing limited rewards for transparency (Honig et al., 2023; Hood, 2011; Pasquier and Villeneuve, 2007).

The "blame game" dynamic emerges when authorities anticipate that releasing data will generate negative coverage or personal culpability (Žuffová, 2023). Common avoidance strategies include: deflection through jurisdictional boundary claims; partial disclosure via unusable formats or excessive redaction; reframing problematic events as isolated incidents rather than systemic issues; and invoking broad legal clauses to justify non-disclosure. These tactics enable technical compliance whilst substantially limiting meaningful access to information (Honig et al., 2023; Hood, 2011; Pasquier and Villeneuve, 2007; Žuffová, 2023).

Administrative capacity constraints compound these cultural barriers. Many local government entities lack sufficient resources for comprehensive data management, making FOI compliance genuinely burdensome rather than merely inconvenient (Roziere and Walby, 2020). Outdated technological systems, insufficient staffing, and limited expertise in data anonymization create practical obstacles that may legitimately prevent full compliance with information requests.

Walby and Yaremko (Walby and Yaremko, 2020) propose systematic FOI audits, coordinated submission of identical requests to multiple public authorities, as an effective method for assessing compliance patterns and identifying institutional variations in transparency practices. This approach enables researchers to distinguish between legitimate capacity constraints and deliberate non-compliance whilst revealing best practices worthy of broader adoption.

FOI audits offer several methodological advantages over traditional survey-based studies of transparency. They provide direct evidence of actual institutional behavior rather than self-reported attitudes or intentions, enable systematic comparison across entities operating under identical legal frameworks, and generate concrete examples of successful and unsuccessful disclosure practices that can inform policy recommendations.

International applications of FOI audit methodology have revealed consistent patterns of uneven compliance, with significant variations in response rates, data quality, and institutional cooperation (Walby and Luscombe, 2019). These studies consistently demonstrate gaps between formal legal commitments to transparency and practical institutional performance, suggesting that implementation challenges represent universal rather than jurisdiction-specific phenomena.

The Norwegian context provides an ideal setting for FOI audit research, given the country's strong democratic institutions, comprehensive legal framework, and municipal system that enables systematic cross-institutional comparison. Moreover, elderly care represents a policy domain where transparency serves clear public interest objectives, making resistance to disclosure particularly problematic from democratic accountability perspectives.

Drawing on these theoretical insights, we analyze Norwegian municipal FOI compliance through three analytical dimensions derived from existing literature: How municipalities interpret and apply FOI legal provisions, including the use of exemptions and procedural requirements (following Žuffová, 2023). Institutional resources, technological infrastructure, and staff expertise are available for processing FOI requests (following Roziere and Walby, 2020). Institutional attitudes toward transparency, blame-avoidance behaviors, and organizational incentives affecting disclosure decisions (following Hood, 2011; Žuffová, 2023).

This framework enables systematic analysis of compliance patterns whilst identifying specific barriers that prevent effective FOI implementation in practice.

Having established these theoretical foundations, we now turn to examine the specific Norwegian institutional context that shapes municipal approaches to information disclosure in the elderly care domain.

Norway has a universal healthcare system that covers all citizens (Norwegian Directorate of Health, 2023). The government provides a range of services to support older people in living independent lives in their own homes, including home assistance on a regular basis called: “home services” («hjemmetjenester»). The provision of home assistance is the responsibility of the local municipalities, which receive funding from the central government to provide these services. A key element of the Norwegian quality assurance system is the Norwegian Code of Practice for Undesirable Patient Events (“Norsk kodeverk for uønskede pasienthendelser”, NOKUP). This document is primarily intended to improve and standardize the reporting of deviant events within the Norwegian healthcare system (Helsedirektoratet, 2021). Commissioned in 2013 by the Ministry of Health and Care Services (HOD, “Helse- og omsorgsdepartementet”) to the The Knowledge Centre for Health Services (“Kunnskapssenteret for helsetjenesten”), and involving representatives from both regional and national administrations, this documents aims to standardize the classification of deviant events so that uniformity is reached at a local, regional, and national level. As stated by Regulation on Management and Quality Improvement in the Health and Care Service, leaders of institutions providing healthcare services have the duty to plan (Helse- og omsorgsdepartementet, 2016a), implement (Helse- og omsorgsdepartementet, 2016b), evaluate (Helse- og omsorgsdepartementet, 2016c), and improve (Helse- og omsorgsdepartementet, 2016d) the quality of the services they offer so that patient and user safety is prioritized. Not to mention the opportunity to avoid the same error from happening twice (Dagens Medisin, 2021).

Norway’s Freedom of Information Act (Act No. 16 of 2006), often called “offentleglova,” seeks to ensure transparency by granting the public broad access to documents held by government agencies and publicly funded organizations. The Act applies to both central and local authorities as well as certain semi-public bodies performing official tasks. Under Section 9, agencies must compile electronically stored data “using simple procedures” when responding to requests, thereby aiming to facilitate straightforward disclosure. Section 13 establishes a duty of confidentiality that prohibits the release of sensitive or identifying information, while Section 14 allows officials to withhold internal documents prepared for case deliberations. Additionally, Section 30 stipulates that once materials are made publicly available, they should be offered in any format already existing within the agency’s records. If a request is refused, applicants can typically appeal through administrative channels or seek judicial review. Together, these provisions form the legal backbone for public oversight in Norway, setting the parameters within which authorities must disclose, or may lawfully withhold, documents and data.

This study employed a systematic FOI audit methodology, following the approach advocated by Walby and Yaremko (Walby and Yaremko, 2020) for assessing transparency compliance across multiple public institutions. The research was conducted through collaboration between the Artificial Intelligence Journalism Research Centre (AIJRC) at Oslo Metropolitan University and investigative journalists from NRK Brennpunkt, Norway's premier investigative journalism program.

The collaborative framework was structured to maintain appropriate boundaries between academic research objectives and journalistic practice. Journalists from NRK Brennpunkt initiated the project with specific investigative goals related to elderly care quality and sought research support for methodological design and data analysis. All FOI requests and direct communication with municipalities were conducted solely by the journalists, ensuring authentic journalistic practice whilst enabling systematic academic analysis of responses and compliance patterns.

FOI requests were submitted between June and December 2021, with follow-up communications extending through the entire first half of 2022. This timeframe enabled systematic comparison across municipalities whilst accommodating varying institutional response capabilities and administrative procedures.

The research design aimed to assess municipal FOI compliance across the three analytical dimensions outlined in our theoretical framework: legal interpretation and procedural adherence; administrative capacity and technological infrastructure; and institutional culture regarding transparency and blame avoidance. This multi-dimensional approach enabled comprehensive analysis of factors influencing municipal transparency practices beyond simple compliance statistics.

Our data collection proceeded through two distinct phases, with methodological refinements in phase 3 based on initial findings and practical constraints identified during the previous phases.

Pilot Study and Initial Methodology: The initial approach targeted 10 municipalities (N=10), over a three-month period, June-August 2021, selected based on journalists’ assessment of their relevance to the investigation, for a focused pilot study examining deviation reports from home care services.

This pilot approach achieved a 50% response rate, with 5 municipalities (N=5) providing some form of response to the initial request. However, only 3 municipalities (N=3) delivered high-quality, analytically useful data, representing a 30% success rate for obtaining usable information. These initial results highlighted significant challenges in municipal FOI compliance whilst demonstrating the feasibility of systematic data collection approaches.

Nationwide Implementation: Building on pilot study insights, the research team expanded the investigation to encompass all 356 Norwegian municipalities and the 15 districts in Oslo (N=370), requesting deviation reports from home care services covering a five-year period (2016-2020). This extended timeframe enabled analysis of temporal trends whilst providing sufficient data volume for meaningful statistical analysis across diverse municipal contexts.

However, the broader scope exposed several methodological challenges not fully anticipated during the pilot phase. Many municipalities proved unable to filter reports by service user age or type of care, preventing a precise focus on elderly home care specifically. The terms "deviation" and "home care services" lacked standardized definitions across municipal reporting systems, leading to widely varying interpretations of request scope. Some municipalities interpreted "deviation" narrowly as serious incidents requiring external reporting, whilst others included minor procedural non-conformities. Similarly, "home care services" encompassed different service categories across municipalities, from basic assistance to comprehensive medical support. Consequently, municipal interpretations varied substantially, with some providing comprehensive incident databases whilst others offered only high-level summary statistics. Of the 265 municipalities (N=265) that provided responses, data arrived in diverse formats ranging from structured Excel files to PDF documents containing unreadable graphs or scanned paper records. Additional delays occurred when municipalities were asked to resend data in Excel format.

The complete dataset from Phase Two encompassed over 120,000 deviation reports (N>120,000), but comparing data across municipalities proved extremely difficult due to definitional inconsistencies and format variations.

Methodological Refinement: To address Phase Two limitations, the methodology was refined for a third data collection round using Microsoft Forms for standardized data submission. This approach enabled municipalities to submit uniform data focusing specifically on total deviation numbers and medication-related incidents, eliminating format inconsistencies whilst reducing municipal administrative burden.

The Microsoft Forms system was emailed, in March 2022, to each municipality requesting specific aggregated statistics: total deviations and medicine-related deviations for the last 6 years, 2016-2021. This standardized approach proved significantly more efficient, with all 265 responses (N=265) providing usable data in identical formats. The resulting dataset contained approximately 160,000 total deviations (N≈160,000), of which 72,000 involved medication errors (N=72,000), representing 45% of all reported incidents.

Although this refined methodology yielded less detailed information than Phase Two's comprehensive incident reports, it facilitated rapid analysis and generated a manageable, consistent dataset. The use of well-defined parameters and user-friendly tools significantly improved both the quality and the reliability of the data in the second round.

The significant improvement in data quality and analytical feasibility demonstrated the importance of balancing comprehensiveness with practical constraints when conducting systematic FOI audits.

The systematic preservation of municipal explanatory comments during the FOI process provided valuable insights into institutional perspectives on transparency obligations and practical compliance challenges. Comments following responses were received from 45 municipalities, enabling thematic analysis of institutional attitudes and concerns.

Technological Transition Challenges: Multiple municipalities emphasized ongoing technological modernization efforts whilst acknowledging historical data limitations:

These responses highlighted the dynamic nature of municipal information systems, with many jurisdictions transitioning between different technological platforms during the study period. Such transitions created legitimate barriers to historical data provision whilst demonstrating institutional commitment to improved record-keeping practices.

Service Integration Complexities: Smaller municipalities particularly emphasized the challenges created by integrated service delivery models:

This structural characteristic of Norwegian elderly care delivery, where administrative efficiency drives service integration, creates genuine complications for targeted data requests. Municipalities expressed concerns that providing integrated data might mislead journalists or create unfair comparisons with jurisdictions maintaining separate administrative systems.

Organizational Learning and Quality Improvement: Several municipalities provided sophisticated explanations of their approach to deviation reporting as a quality improvement tool:

These responses demonstrated awareness of modern quality management principles, emphasizing learning rather than blame. Municipalities implementing systematic quality improvement approaches often expressed greater willingness to share appropriately anonymized data, suggesting that transparent institutional cultures may facilitate both internal learning and external accountability.

Evidence-Based Service Improvement: Advanced municipalities provided specific examples of how deviation data informed service improvements:

This response illustrates sophisticated use of systematic data for identifying problems and evaluating interventions. Municipalities demonstrating evidence-based approaches to service improvement typically provided higher quality responses to FOI requests, suggesting associations between internal analytical capacity and external transparency performance.

Having presented these systematic findings on Norwegian municipal FOI compliance patterns, we now turn to analyze their theoretical implications and practical significance for understanding transparency governance in democratic systems.

Our FOI audit was part of a broader project that aimed to use AI-based methods to analyze incident reports in elderly care to identify patterns of risk and systemic failures. Open data initiatives by governments are widely celebrated for their potential to enhance transparency, foster public trust and improve governance. Yet working with the data obtained through our FOI requests made it clear that releasing and re-using public data carries inherent risks. These risks are heightened when datasets involve sensitive areas such as healthcare and can be combined, modelled, or analyzed with powerful AI tools, especially now that large language models (LLMs) are accessible to non-experts and data ownership/access is increasingly blurred by opaque corporate policies. These concerns surfaced both in our own attempts to prepare the data for AI-supported analysis and in the way municipalities framed the possible misuse, misinterpretation and privacy risks associated with making such data more widely available.

Privacy and Confidentiality Concerns: One of the most prominent risks associated with open data is the potential compromise of individual privacy. In sectors like healthcare, where data often involve sensitive personal information, there is a delicate balance between transparency and the need to protect patient confidentiality. While data on the performance of elderly care institutions are crucial for assessing quality and accountability, such information may inadvertently reveal personal details about patients, healthcare workers, or specific incidents. Even anonymized data can sometimes be re-identified through data triangulation, putting individuals at risk of exposure. For government agencies, the fear of violating privacy laws, such as the General Data Protection Regulation (GDPR) in Europe, may lead to overly cautious behavior, which can hinder the open data movement. This creates a paradox where transparency efforts are stifled by concerns over data protection.

Misinterpretation and Misuse of Data: Open data can be complex and difficult for the general public to interpret correctly, leading to risks of misinterpretation or misuse. In the case of elderly care data, for instance, the non-conformity reports (NCRs) collected at the municipal level are meant to highlight areas needing improvement, such as staff shortages or medication errors. However, without the proper context or understanding, such data could be misinterpreted as indicative of widespread incompetence or failure within the healthcare system. The potential for misinterpretation poses a significant risk for public servants and politicians, who may fear that open data could be weaponized against them. This fear may lead to a defensive stance, as highlighted in the interactions between Norwegian journalists and bureaucrats. Rather than fostering transparency and collaboration, the release of open data might escalate blame avoidance, where officials prioritize minimizing personal or organizational exposure over engaging with systemic issues. In the long run, this undermines the very purpose of open data: to facilitate informed decision-making and public accountability.

Administrative Burden and Resource Strain: The process of preparing, anonymizing, and releasing data for public consumption can impose a significant administrative burden on government agencies, particularly those with limited resources. This risk is especially pronounced in smaller municipalities, where the capacity to handle complex FOI requests may be stretched thin. In the case of Norwegian municipalities, the investigation revealed varying levels of responsiveness to FOI requests, with some municipalities struggling to provide data promptly or in a usable format. This uneven compliance highlights a broader risk associated with open data initiatives: the strain on public sector resources. Government agencies may need to allocate significant time and money to ensure that data is accurate, anonymized, and legally compliant, which can detract from other essential public services. In extreme cases, this resource strain could lead to reduced efficiency in the very sectors that open data is intended to improve, such as elderly care, where the prioritization of data transparency may come at the expense of front-line services.

Erosion of Public Trust: While open data is intended to foster trust between the government and its citizens, poorly managed open data initiatives can have the opposite effect. If the released data are incomplete, difficult to understand, or perceived as misleading, it can erode public confidence in government institutions. Citizens may question the motives behind open data initiatives, suspecting that the government is not fully transparent or that it is hiding critical information. When transparency efforts are perceived as partial or insincere, they can reinforce skepticism about government accountability. This risk is further exacerbated by the "blame game" dynamics, where bureaucrats may seek to deflect responsibility, leaving citizens with the impression that the government is more interested in self-preservation than in addressing systemic issues.

Although the study maintains a clear boundary between academic analysis and policy advocacy, the empirical findings strongly suggest areas for potential reform.

Digital Infrastructure Standardization: First, the research shows that Norway would benefit from national coordination of digital systems, particularly for deviation reporting in sectors such as elderly care. Standardizing these systems could help address disparities in capacity and eliminate format inconsistencies that currently hinder FOI compliance. Furthermore, clearer technical standards for data delivery formats (e.g., requiring machine-readable Excel files) would substantially improve the practical usability of information without imposing excessive administrative burden.

Special consideration should be given to smaller municipalities, which often lack the human and technological resources needed to manage FOI requests. Regional cooperation mechanisms and shared digital platforms could offer economies of scale while preserving local autonomy.

Clarification of Legal Frameworks: The frequent invocation of Section 9 of the FOI Act, which allows officials to decline requests deemed too burdensome, underscores the need for greater legal clarity around what constitutes “reasonable effort.” Clearer guidelines could help public servants distinguish between genuinely excessive requests and those that should be fulfilled in the public interest.

Another legal improvement would be the definition of acceptable file formats for electronic disclosures, helping to eliminate the current patchwork of unusable PDFs, scanned documents, and inconsistent databases. Furthermore, enhancing appeal mechanisms for denied requests would offer a stronger incentive for compliance and bolster the rights of requesters.

Cultural and Organizational Development: The study also underscores the importance of institutional culture in shaping transparency outcomes. Municipalities with learning-oriented leadership and quality improvement systems tended to comply more fully with FOI requests. Therefore, investing in professional development that emphasizes the value of transparency could yield dividends both for service quality and for democratic accountability.

Leadership training programs, particularly for municipal executives and middle managers, could emphasize transparency as a governance competency. Coupled with knowledge-sharing networks across municipalities, such programs could accelerate the diffusion of best practices in FOI compliance and open data management.

However, it is important to highlight that several limitations temper the generalizability of the findings and point toward future research directions.

Firstly, the focus on elderly care as a policy domain is notable. While this is an area of broad public interest and relatively low commercial sensitivity, results might differ in more politically sensitive or commercially protected domains. Thus, cross-domain comparisons would help assess the transferability of the findings.

Secondly, the timing of the study coincided with significant events: municipal restructuring and a national push for digitalization. These contextual factors may have introduced anomalies in FOI compliance. Future longitudinal studies would be useful to assess whether these transparency gaps are transitional or structural.

Third, the research focused on formal compliance, i.e., whether municipalities responded to requests, but could not assess the accuracy, completeness, or relevance of the data provided. Strategic transparency, in which institutions appear compliant while obscuring meaningful information, remains a challenge that requires more nuanced methodologies, potentially integrating qualitative follow-up or content analysis.

Lastly, the Norwegian political and administrative context, characterized by high trust, professionalized public service, and a cooperative political culture, may limit the applicability of findings to more adversarial or resource-constrained settings. Comparative international studies could help define boundary conditions for the theoretical frameworks developed in this paper.

This systematic investigation of Norwegian municipal FOI compliance reveals fundamental tensions between formal transparency commitments and practical accountability implementation in contemporary democratic governance. While Norway's exceptional press freedom record and comprehensive legal framework create strong foundations for transparent governance, our empirical analysis demonstrates that administrative capacity, institutional culture, and blame-avoidance behaviors prove decisive in determining actual transparency outcomes.

The Norwegian experience illustrates that even well-intentioned transparency legislation requires supportive institutional infrastructure, administrative capacity, and cultural commitment to achieve democratic accountability objectives. Legal provisions alone prove insufficient without corresponding organizational development, technological modernization, and cultural transformation that aligns institutional incentives with transparency goals.

Perhaps most importantly, our findings suggest that transparency governance proves more complex than conventional assessments indicate. Our analysis highlights the need for more sophisticated approaches to evaluating and promoting democratic accountability.

As democratic societies worldwide grapple with demands for greater transparency whilst confronting administrative capacity constraints and institutional resistance, the Norwegian experience provides valuable insights into both the potential and limitations of legal approaches to transparency governance. Achieving meaningful democratic accountability requires sustained attention to the institutional, cultural, and technological foundations that enable formal transparency commitments to translate into practical accessibility of government information.

Our investigation reveals institutional practices that satisfy formal legal requirements and international assessments whilst maintaining practical barriers to comprehensive public oversight, a pattern likely characterising many advanced democratic contexts. Understanding and addressing this pattern represents a crucial challenge for contemporary democratic theory and practice, requiring continued empirical investigation, theoretical development, and institutional innovation to bridge the gap between transparency aspirations and accountability reality.