Evaluation of science advice during the COVID-19 pandemic in Sweden

The paper situates Sweden’s COVID-19 response in a long history (>280 years) of cooperation among government, authorities, and science that produced major public health successes (e.g., midwifery, maternal care, injury prevention, HIV response). When COVID-19 reached Europe, Sweden adopted a comparatively less restrictive, mitigation-focused strategy emphasizing individual responsibility, keeping most schools open, and avoiding lockdowns. The authors note claims of “Swedish exceptionalism” and the influence of Sweden’s approach internationally. The study’s purpose is to evaluate how scientific evidence and integrity informed Sweden’s COVID-19 policy up to December 31, 2020, assess whether a pre-existing pandemic strategy existed and how it was implemented, and examine the roles of scientists, policymakers, and politicians in decision-making. The work uses a narrative approach and does not aim to advance policy theory.

There is no standalone literature review section. The authors base their analysis on a comprehensive systematic literature search in PubMed/Medline and Web of Science on Sweden’s and Nordic countries’ pandemic handling; official documents; governmental reports; Freedom of Information (FOI) requests (emails, agendas, notes, press releases); and mainstream/digital media. They reference extensive international guidance (WHO, ECDC) and peer-reviewed studies on transmission, mitigation strategies, face masks, school transmission, health system capacity, and social impacts, using these to contrast Sweden’s approach with international scientific consensus.

Design: Non-experimental narrative case study applying the Narrative Policy Framework (meso- and macro-levels) to examine how policy actors construct and communicate narratives influencing policy processes and how narratives permeate institutions and culture. The key actor is the Public Health Agency of Sweden (PHA). Timeframe focuses on 2020. Data sources and procedures: Content analysis informed by multidisciplinary author discussions and input from national/international experts; systematic searches of peer-reviewed literature; FOI requests for emails, meeting agendas/notes, and press releases (government and PHA); public communications (open letters, debate articles, petitions); mainstream and digital media. Analytic framework: Policy evaluation structured around four tasks (Lasswell/Nachmias): (a) identify policy goals; (b) metrics to assess progress; (c) evidence/data related to metrics (official statistics, healthcare reports); and (d) judgments of responsibility to inform future policy learning. Scientific integrity is defined per Douglas and Bour as proper reasoning processes and respect for empirical evidence. The authors assess scientific integrity in four contexts: (1) preparedness; (2) actor evaluations; (3) errors/inconsistencies in recommendations/communications; (4) consequences for healthcare and society. Ethics: Not applicable (no human subjects). All authors resided in Sweden during part of 2020–2021.

• Preparedness and strategy: Sweden’s pandemic plans were influenza-focused (2015, 2019) and insufficiently adapted to SARS-CoV-2. The response emphasized mitigation, personal responsibility, and keeping society open, with no lockdowns or legal enforcement during 2020.
• Governance: Government crisis management largely ceded operative responsibility to the Public Health Agency (PHA), with minimal cross-agency coordination. The government stated it had “no strategy,” relying on PHA advice. Transparency was limited; meeting records and modelling assumptions were often unavailable or redacted.
• Deviation from international evidence: PHA frequently diverged from WHO/ECDC guidance, disputing or downplaying airborne transmission, asymptomatic/pre-symptomatic spread, the effectiveness of masks, and children’s role in transmission. The precautionary principle was not applied.
• Testing, tracing, isolation: Testing and contact tracing were delayed and never fully implemented in 2020; quarantine/isolation policies were weakly enforced.
• Healthcare and elderly care: Reports of triage limiting ICU access (especially older age/comorbidities); many patients were not admitted or examined. IVO (Nov 24, 2020) found systemic failures: about one-fifth of nursing home residents were denied an individual medical assessment; fewer than one-tenth of COVID-19 patients were examined by a physician; some received end-of-life treatment without testing. Morphine was reportedly administered instead of oxygen to many elderly despite available supplies.
• Face masks and PPE: PHA discouraged public mask use, citing ineffectiveness and potential harm; healthcare/care-home mask policies were delayed (recommended June 25, 2020, after >5,000 deaths) and often limited to suspected/confirmed cases; face shields were common despite limited efficacy against airborne spread. Some institutions disallowed masks, leading to staff discipline or public denial of access.
• Schools and children: Compulsory attendance remained for <16-year-olds, with limited mitigation; parents could be fined for keeping children home. Testing for children was restricted. Authorities downplayed children’s transmission and disease burden; internal communications indicated consideration of using children to build herd immunity.
• Secrecy and data issues: Limited transparency, alleged data manipulation concerns (e.g., child deaths), delayed and inconsistent reporting between agencies, and redaction/erasure of emails.
• Advisory processes: PHA’s advisory group (April 17, 2020) lacked diverse independent experts and did not discuss overarching issues; universities’ independent input had limited impact; critics were publicly discredited. The Royal Swedish Academy of Sciences recommended masks (June 2020), but advice was largely not adopted. Informal groups (e.g., Science Forum COVID-19) attempted to provide evidence-based advice.
• Outcomes and comparisons: Sweden’s COVID-19 mortality in 2020 far exceeded neighboring Nordics (about 10× Norway’s death rate in 2020). Peaks in 7-day rolling average deaths reached 99 (Apr 16) and 90 (Dec 31, 2020). Sweden scored lowest on ICU bed accessibility in a 14-country European study, with rural shortfalls. OECD/EU (Nov 2020) ranked Sweden lowest among 35 countries on multiple management metrics; Standard Ethics downgraded Sweden’s ethical rating (May 21, 2020) for noncompliance with WHO recommendations.
• Inequities: Excess mortality disproportionately affected elderly, care-home residents, migrants, and socioeconomically disadvantaged groups; narratives blaming language or immigrant status emerged, with limited remedial action. The approach appeared to benefit middle/upper classes able to work from home.
• International influence: Sweden’s model influenced arguments for looser restrictions abroad and underpinned the Great Barrington Declaration, viewed internationally as unscientific and unethical.
• Accountability: No clear accountability was assumed by government, regions, or municipalities; responsibilities remained opaque.

The analysis indicates Sweden’s response was dominated by the Public Health Agency, lacked a multidisciplinary scientific advisory process, and diverged from international evidence and guidance. This undermined the stated goals of minimizing mortality/morbidity and societal impacts. The findings address the research question by showing scientific advice was selectively used or ignored, and scientific integrity compromised through narrow expertise, secrecy, inconsistent communication, and a failure to apply the precautionary principle. Consequences included avoidable excess mortality (especially among elderly and disadvantaged groups), strained healthcare capacity, delayed non-COVID care, and erosion of trust for critics. Internationally, Sweden’s stance contributed to polarization and potentially undermined suppression efforts elsewhere. The authors argue structural societal and governance factors—decentralization, weak crisis coordination, lack of transparency and accountability, and an absence of an independent infection control institute—shaped these outcomes. They emphasize the need for transparent, ethical, and evidence-based decision-making, inclusive of multidisciplinary expertise and open public communication, to restore democratic processes and improve future pandemic responses.

Sweden’s COVID-19 response was characterized by a laissez-faire approach that prioritized maintaining the national image and individual responsibility over an evidence-based, precautionary strategy. A small group of officials with narrow expertise held disproportionate influence, international guidance was disregarded, and transparency and accountability were lacking. The strategy failed to protect vulnerable groups, especially the elderly, and yielded worse outcomes than Nordic neighbors without apparent compensatory benefits. The authors recommend re-establishing the scientific method within authorities, recreating an independent Institute for Infectious Disease Control, instituting open and democratic multidisciplinary advisory processes, ensuring accurate public communication, and initiating a self-critical review of political culture and accountability to prevent repetition of failures in future pandemics.

• Scope limited to Sweden’s 2020 response (through Dec 31, 2020) and national-level policy, with regional/municipal details addressed selectively.
• Narrative case-study design; no experimental methods.
• Reliance on available documents, FOI materials, and media sources; many records were redacted or unavailable due to secrecy practices, limiting verification and completeness.
• No primary data collection from human participants; some analyses depend on secondary reporting and official statistics with changing case definitions and reporting delays.
• Not intended to advance policy theory; focuses on evaluating the use/misuse of science within policy and governance contexts.