Many features of social media platforms can influence safety in relation to self-harm/suicide-related content. In 2019, platforms including Meta updated their policies regarding self-harm/suicide content. This included community guidelines, content restrictions, and signposting. The impact of these safety measures has yet to be fully evaluated. This study aimed to better understand the perspectives of social media users on safety measures related to self-harm/suicide content, including perspectives of those with a history of self-harm. This was a cross-sectional mixed methods study. Participants were recruited through convenience sampling. Quantitative data were collected using a fixed-response cross-sectional survey (5294 respondents aged 16-84 years, largely women/girls or nonbinary). Questions related to the experience of self-harm and suicide content, the change policy in 2019, and specific features of platforms. Qualitative data were collected via semistructured interviews (17 participants aged 16-56 years) with interview guides broadly following the topic areas of the survey. Interviews were analyzed using thematic analysis. Half of the survey respondents (49.8%, 95% CI 47.8%-51.8%; 2476/4970) stated the current restrictions made social media feel safer, 34.9% (95% CI 32.6%-37.2%; 1736/4981) reported that this changed what they saw on social media, and just 9.9% (95% CI 7.5%-13.0%; 492/4479) stated that this changed what they posted. Furthermore, 64.3% (95% CI 62.6%-65.9%; 3212/4997) stated that they would be "very likely" to click on a post marked with a generic content warning, with 15.1% (95% CI 12.6%-17.9%; 753/4998) being "very likely" to click on a post with a self-harm specific content warning. Of the respondents who had content removed due to mental health or self-harm content, ~90% indicated that this was harmful or very harmful to them at the time, particularly where content was removed because of visible scars (extremely harmful 58.3%, 95% CI 54.3%-62.3%; 599/1027; harmful 31.3%, 95% CI 26.3%-36.7%; 321/1027). Thematic analysis of in-depth interviews revealed 13 subthemes organized into 7 overarching thematic areas. Participants described the experience of social media platforms before and after the policy changes, with opinions divided on whether existing safety features were effective. Participants described the distress caused by inappropriate censoring of images, for example, where healed self-harm scars were visible. Suggestions were made for how platforms can continue to improve, including age verification procedures for young people, tailored signposting, and increased control over content. To our knowledge, this is the first study to explore the 2019 change in platform policies with respect to self-harm/suicide-related content from the perspectives of those with a history of self-harm. This research adds to the evidence of the risks and unintended consequences of imposing untested blanket restrictions in online settings. Full evaluation of these restrictions is essential to allow platforms to continue to improve, encouraging a safe space for supportive communities while mitigating potential harm.
This study examined whether direct expert elicitation of gambling harm severity, applied to blinded first-hand narrative accounts, produces ordered and consistent estimates with respect to quantitative self-reports. As such, this represented independent convergent validation evidence of the Gambling Harms Scales (GHS) instruments for people who gamble and affected others (AOs). Six experts with significant research or clinical expertise in the gambling field evaluated 60 transcripts from people who gamble and AOs experiencing varying levels of gambling-related harm (no-harm, low-harm, moderate-harm, and high-harm). Each transcript was evaluated using four protocols (visual analogue scale, rank ordering, time trade-off, and standard gamble). Narratives from participants with higher levels of harm (as assessed by the GHS) were associated with significantly higher expert-elicited health disutility. Disutility increased monotonically with GHS harm bands, and this pattern was consistent across all four elicitation protocols and all six expert raters. Pooled disutility estimates were 0.084 for low-harm, 0.187 for moderate-harm, and 0.221 for high-harm. Harm severity elicited by interviews and expert assessment was consistent with harm severity as assessed by the brief GHS measures. Expert evaluation of blinded narrative accounts provides independent external evidence supporting the GHS suite of instruments as measures of gambling-related harm.
Alcohol consumption often leads to harm to individuals other than the drinker. Yet, few studies have quantified an exposure-response gradient using "nominated" drinkers' alcohol volume and their heavy episodic drinking (HED) frequency to predict harm-type burden and overall severity among people harmed by someone else's alcohol use. We examined these dose-response relationships across sociodemographic groups. Data were drawn from the cross-sectional Australian 2021 Alcohol's Harm to Others (AHTO) Survey. A total of 2574 adults (18+ years) participated, of whom 554 reported harm from another person's drinking in the past 12 months. Respondents identified the person who harmed them the most (the "nominated drinker") and reported on the person's weekly alcohol volume and frequency of HED. Harms experienced included an overall harm score (0-10 scale) and a count of 14 specific harms. There was a significant positive dose-response relationship between the nominated drinker's alcohol consumption (both weekly volume and frequency of HED) and the degree of harm experienced by respondents. Although subgroup analyses suggested stronger effects among women, younger respondents, and when the drinker was a (former) partner, these differences were not statistically significant. Greater alcohol consumption and more frequent HED are associated with more severe harms to others, regardless of respondents' sociodemographic background. A clear dose-response relationship exists between alcohol consumption by a known drinker and harm to others. The dose-response gradient was stronger at higher exposure levels, suggesting that reducing very frequent heavy drinking may yield disproportionately larger reductions in AHTO burden.
Patient safety, defined as the absence of avoidable harm, in orthodontics is still in its early stages of development. Knowledge on the subject is limited. (1) To conduct an incident analysis, (2) to measure avoidability, (3) to investigate the relationship between contributory factors and perceived avoidability of harm from orthodontic care, and (4) to suggest methods to mitigate potential patient harm incidents. Data were retrospectively collected from consecutively filed orthodontic-related claims from the Danish Dental Compensation Association (DDCA) and included all approved-for-compensation claims from September 2019 to August 2024. Reason for filing the claim, harm category, and harm-causing events were recorded. Incident analysis was conducted for each claim based on the Eindhoven criteria and AI-assisted qualitative analysis. A large language model was used to identify and interpret the interplay between the contributory factors. Perceived avoidability of harm was based on a 6-point rating system. A total of 125 claims of approved patient harm were included and reviewed. Based on the Eindhoven incident analysis, the most frequent root causes were human related factors (monitoring, verification and clinical knowledge among others) followed by organizational factors (management priorities, protocols and workplace culture), illustrating complex interrelationships. The majority of claims (70.1%, n = 82) were avoidable harm, being rated as either probably avoidable in more than 50% of cases, high certainty of avoidability, or completely avoidable. While low-avoidability harm events were characterized by few contributory factors, high-avoidability cases were characterized by the involvement of numerous human and organizational failures. The analysis of 125 compensated orthodontic harm cases showed that 70.1% of the cases were deemed avoidable and most harms were multifactorial in nature, with human and organizational contributors playing the primary role. This multifactorial pattern provides new insights and forms a foundation for future patient-safety initiatives in orthodontic practice and research.
To assess the effectiveness and safety of active pharmacotherapy for patent ductus arteriosus (PDA) initiated based on echocardiography-criteria versus expectant management in extremely preterm infants <29 w gestation in the first three weeks of life. Systematic review with meta-analysis and trial sequential analysis (TSA). We searched PubMed, Embase, CENTRAL and Emcare for randomised trials published between 2010 and May 2026. We synthesised results using random-effects meta-analysis and conducted a TSA. Extremely preterm infants with PDA considered as significant on echocardiography. Pharmacotherapy versus expectant management. Mortality by 36 weeks postconceptional age or before discharge and chronic lung disease (CLD). We included 12 trials involving 2344 extremely preterm infants. Meta-analysis found increased risk of mortality with active pharmacotherapy (RR 1.34, 95% CI 1.08 to 1.67; p=0.008, I2=0%; 12 randomised controlled trials (RCTs)). Number-needed-to-treat for an additional harmful outcome (i.e. mortality) was 25 (95% CI 17 to 100). On TSA conducted with the stringent alpha error 3.3% and beta 10%, the cumulative Z-score touched the O'Brien-Fleming monitoring boundary at the current sample size, supporting the meta-analysis findings under the specified TSA assumptions.For CLD, meta-analysis showed little or no difference (RR 0.99, 95% CI 0.92 to 1.06, p=0.74, I2=14%, 12 RCTs). The certainty of evidence was high for mortality and moderate for CLD. In extremely preterm infants, active pharmacotherapy for PDA based on current echocardiographic criteria in the first three weeks of life was associated with increased mortality with a clinically important number needed to harm. These findings should inform treatment decisions and discussions with parents while ongoing research seeks to identify infants who may benefit from treatment. CRD420261303992.
Universal school-based mental health programs for adolescents intend to promote wellbeing, but some evidence suggest that they may also cause harm. This scoping review aimed to map the state of knowledge on potential harms associated with these programs from the recent literature. Following the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews guidelines, four databases were searched for peer-reviewed studies published between 2019 and 2024. Articles were included if they evaluated a universal school-based mental health program for adolescents (13-18 years) and reported at least one quantitative outcome measure. Potential harms identified were classified into five harm types based on the observed data. Results showed that, of the 58 included studies, 29% (17 of 58) reported at least one type of potential harm. Statistically significant change in the overall sample was examined in all studies, with 16% (nine of 58) showing deterioration in at least one outcome. Statistically significant change by subgroup (e.g., age, gender, and baseline mental health) was examined in 21 studies, with 29% (six of 21) of these showing deterioration in at least one subgroup. Additional harms were identified through clinically meaningful deterioration, qualitative participant feedback, and safety protocols. Our findings confirm that universal school-based mental health programs can be associated with potential harms. Implications for research and practice are discussed.
Social scientists have developed dozens of measures for assessing partisan bias in redistricting. However, these measures are not easily adapted to other groups, including groups defined by race, class or geography, nor are they applicable to single- or no-party contexts, such as local redistricting. Here we propose a unified framework of harm for evaluating the impacts of a districting plan on individual voters and the groups to which they belong, to overcome these limitations. We consider a voter harmed if their chosen candidate is not elected under the current plan but would be under a different plan. Harm improves on existing measures by both focusing on the choices of individual voters and directly incorporating counterfactual plans. We discuss strategies for estimating harm using redistricting simulations, and demonstrate the utility of our framework through three applications to US redistricting. Overall, harm provides a flexible way to precisely quantify redistricting's individual impacts.
Gastrointestinal foreign body ingestion or insertion can be an act of intentional self-harm. We aimed to provide a large-scale assessment of this clinical scenario. This is a retrospective cross-sectional analysis of adult patients who presented with gastrointestinal foreign bodies and intent to self-harm. Hospitalizations were identified using the National Inpatient Sample from 2016 to 2023. Each hospitalization was characterized by the management methods used including no intervention, endoscopy, and surgery. Evaluated outcomes included ICU-level care, length of stay, and cost. Additionally, we assessed factors associated with the need for surgery after endoscopy. A total of 16,150 hospitalizations were identified. The mean age was 33.9 years and most patients were male (65.7%). Over one-third (35.2%) of patients had a cluster B personality disorder while 39.3% had schizophrenia and 24.2% were incarcerated. The majority (61.2%) of patients were managed with endoscopy alone while 26.4% required no intervention and 12.3% required surgery. ICU-level care was increased in those who required surgery, but the difference was not statistically significant (4.5% vs. 3.0%; OR 1.54, 95% CI: 0.91-2.62, p = 0.111). Small bowel location was associated with the need for surgery after endoscopy (aOR 2.50, 95% CI 1.91-3.28; p < 0.001). Over 12,000 hospital days and $25,000,000 were attributable to self-harm-associated gastrointestinal foreign bodies annually. Self-harm-associated gastrointestinal foreign bodies occur predominantly in patients with significant psychiatric comorbidity. Although mortality and severe morbidity are uncommon, healthcare utilization is substantial. Small bowel location is a key predictor of the need for surgical intervention.
The study aims to assess the mental health burden among Chinese adolescents aged 10 to 19 years via the 2021 Global Burden of Diseases Study (GBD). Data on the disease burden owing to mental health conditions of Chinese adolescents aged 10 to 19 years from 1990 to 2021 were obtained from the GBD 2021. Disability-adjusted life of years, years of life lost (YLLs), years lived with disability (YLDs), and the prevalence rates of mental disorders (MDs), substance use disorders (SUDs) and self-harm were plotted against sociodemographic index from 1990 to 2021. Temporal trends in the disease burden were evaluated using the Joinpoint Regression. The Nordpred model was applied to project the disease burden over the next 25 years. The burdens of MDs, SUDs, and self-harm accounted for 31.3% of total YLDs and 7.1% of total YLLs among Chinese adolescents aged 10 to 19 years in 2021. These conditions contributed 1439 YLDs (average annual percentage change [AAPC] = -0.2, 95% CI: -0.3 to -0.1) and 139 YLLs (AAPC = -5.2, 95% CI: -5.6 to -4.8). Significant increases in YLDs were observed for eating disorders (AAPC = 1.3, 95% CI: 1.2 to 1.4), whereas intellectual disabilities and developmental disorders significantly decreased (AAPC = -1.4, 95% CI: -1.5 to -1.3). Differences were evident across age and sex groups. The burden of SUDs reaches to the lowest point with an sociodemographic index value of 0.68. Among GBD-quantified risk factors, bullying victimization was a major contributor to disability. The burden of MDs and SUDs is projected to continue increasing over the next 25 years. The mental health burden among Chinese adolescents is a leading contributor to disability. Model-based projections suggest increasing future burdens of MDs and SUDs, underscoring the need for regularly updated, age- and sex-specific adolescent mental health policies.
Gambling consumers show low uptake and engagement with tools designed to support safer gambling practices and reduce the risk of experiencing gambling-related harms. We used co-design principles to design and develop a digital tool for safer gambling (BetWell) to overcome known barriers to tool uptake, including a focus on gambling problems. BetWell aims to increase awareness of personal gambling expenditure and knowledge of how gambling products function to support informed decision-making about gambling. It was designed based on behavioral change theories and presents a personalized amalgamation of gambling expenditure relative to alternative spend options, and psychoeducational information via a quiz. This exploratory study assessed the perceived acceptability and usefulness of the newly designed digital tool and gathered end-user feedback to improve future prototypes. The study explored whether usability, gambling severity, financial well-being, gambling frequency, spend tracking, activity statement use, and the number of accounts held impacted acceptability and perceived usefulness, and the extent to which these factors independently predicted acceptability and perceived usefulness when controlling for life satisfaction, gambling satisfaction, and demographic variables. This cross-sectional study recruited 140 gambling consumers (mean 41.3, SD 10.9 years) via the market research panel CRNRSTONE. Participants accessed and engaged with BetWell and completed an online survey to share their perceptions towards the tool. The overall acceptability (mean 32/40, SD 5.2) and perceived usefulness (mean 20/25, SD 3.9) of the tool were considered "good" and "useful," respectively by participants. Individuals with higher gambling severity scores were more likely to perceive the tool as useful than those of the lower risk categories. Individuals with higher financial well-being were more likely to perceive the tool as useful and acceptable compared to those in lower financial well-being categories. Higher usability scores corresponded with higher tool acceptability. Previous efforts to monitor and track gambling spend was associated with acceptability and perceived usefulness. Gambling frequency and the number of gambling accounts held were not related to acceptability and perceived usefulness. Qualitative feedback included participant identified suggestions for improvements such as a need for interactive elements, a more detailed view of gambling expenditure, automation of activity statement upload functionality, more challenging and positively framed quiz content, and an emphasis on data security. The study provides preliminary support for the acceptability and perceived usefulness of a newly developed tool to increase awareness of gambling expenditure and knowledge on how gambling products function, providing directions for future improvements to the tool.
Recent Australian and New Zealand clinical practice guidelines for schizophrenia demonstrate important methodological advances, including use of the Grading of Recommendations, Assessment, Development, and Evaluations framework and early involvement of people with lived experience. However, these guidelines remain largely silent on the role of coercion in psychiatric care, despite its prevalence and potential for harm. People diagnosed with schizophrenia experience disproportionately high rates of compulsory interventions-including community treatment orders, involuntary admission, seclusion, and restraint-many of which lack robust evidence of effectiveness. Community treatment orders, the most common form of coercion, show wide and unexplained variation in use across jurisdictions and are disproportionately applied to First Nations people, culturally and linguistically diverse populations, and individuals from socioeconomically disadvantaged areas. Evidence of effectiveness is limited, particularly in more robust controlled designs that have used matching, randomisation, or multivariate analyses to adjust for confounding, while adverse outcomes such as coercion, disempowerment, and disengagement remain under-researched. Similar concerns apply to involuntary admission and other restrictive practices, where evidence of therapeutic benefit is mixed and harms-including trauma and fear-are well documented. Despite this, the latest clinical practice guidelines devote minimal attention to these issues, with potential harms mentioned only briefly and without substantive guidance on minimising coercion. Future iterations of schizophrenia clinical practice guidelines would be strengthened by explicit appraisal of the evidence on coercive practices, systematic consideration of their harms, and incorporation of lived-experience perspectives on the use of coercion, thereby better supporting clinicians and policymakers in delivering care that balances effectiveness, safety, equity, and respect for autonomy.
Nucleic acid synthesis is a dual-use technology that can benefit fields such as biology, medicine, and information storage. However, synthetic nucleic acids could also potentially be used negligently and ultimately cause harm, or be used with malicious intent to cause harm. Thus, this technology needs to be appropriately safeguarded. Sequence screening is one component of a biosecurity protocol for preventing such harm and consists of identifying Sequences of Concern (SOCs). There exist many fit-for-purpose tools that have been developed for nucleic acid synthesis sequence screening. However, questions remain regarding their performance with respect to the consistency of screening. To aid in determining if screening tools are harmonized in regard to baseline sequence screening (which represents a minimum acceptable level of performance), the National Institute of Standards and Technology (NIST) constructed a test dataset based on current screening recommendations. NIST then sent blinded datasets to sequence screening tool developers for testing. Overall, there was a general agreement between the tools and NIST labels given to the sequences, and all tools had a baseline performance of >95% sensitivity and >97% accuracy. Disagreement on specific sequences largely arose from single tools and could be traced to differences in defining a SOC and/or methodological differences in screening algorithms.
This commentary responds to Duggan, Hardy, and Waller's systematic review and meta-analysis of cognitive-behavioural therapy (CBT) for outpatients with anorexia nervosa (AN). Although the review synthesises evidence of clinical benefit, it does not define, extract, or discuss iatrogenic harm or reasons for dropout. This omission limits what can be inferred from the review's conclusions. Effectiveness and safety are distinct evidentiary domains: improvement in weight or eating disorder (ED) symptoms does not demonstrate absence of iatrogenic harm, especially when clinical trials have not systematically defined, monitored, or reported treatment-related risks. Evidence-based care requires not only evidence of effectiveness, but also evidence about safety. This commentary argues that evidence syntheses may produce an unduly favourable account of treatment impact when they quantify and emphasise benefit without equivalent attention to iatrogenesis, weakening informed consent and obscuring patient-reported experiences that may not be captured by symptom or weight outcomes alone. The paper considers methodological, clinical, ethical, epistemic, and intersectional implications of this asymmetry. It recommends that future ED clinical trials and systematic reviews evaluate benefit and harm together, undertake balanced assessments of benefits versus risks, use clearer harm definitions and patient-reported measures, report reasons for dropout, assess longer-term outcomes, and involve lived experience contributors substantively in study design, interpretation, and translation.
Artificial intelligence (AI) in healthcare is assumed to introduce risks that are not easily addressed by dominant philosophical models for thinking about responsibility. When an AI tool makes an error that results in patient harm, the question of who is responsible is rarely straightforward. Dominant models of responsibility work when harm can be traced to a single actor, but they fail in socio-technical systems where decisions and actions are distributed across multiple human and technological agents. Iris Marion Young's social connection model of responsibility offers a systematic way to rethink responsibility in these contexts. Rather than locating responsibility solely after harms occur, Young proposes a forward-looking account that attaches responsibility to those who participate in and benefit from the structural processes that produce harm. Young's framework organised around parameters of reasoning-power, privilege, interest, collective ability, and personal connection- provides an alternative way for understanding differentiated responsibilities across core actors in AI-driven healthcare.
Conversational AI, powered by artificial intelligence, is becoming a common tool for accessing health information, educating patients, and obtaining general medical advice. These advanced systems, known as large language models, can produce responses that sound remarkably human. Nevertheless, these systems are prone to "AI confabulations," whereby they confidently generate incorrect information that could harm patients. This highlights the need to inform healthcare workers and individuals who may be prone to trusting these devices. New evidence suggests that AI may also exacerbate mental health conditions, particularly psychosis, paranoia, and related vulnerable states, especially among susceptible individuals. We conducted a targeted literature review and case-based analysis of 35 reported instances in which interactions with generative AI systems were temporally associated with the onset or worsening of psychotic symptoms. Across cases, recurrent patterns included reinforcement of delusional beliefs, amplification of pre-existing psychiatric vulnerabilities, promotion of harmful behaviors, and dissemination of unsafe medical guidance. Common contributing factors included prior psychiatric history, substance use, sleep disturbance, and prolonged AI engagement. We propose a conceptual hypothesis termed the delusional feedback loop, in which AI-generated responses iteratively validate distorted beliefs, contributing to their persistence and escalation. This process can be conceptualized as involving four components: underlying vulnerability, exposure to conversational AI, validation of distorted beliefs, and reinforcement through repeated interactions. Despite the rapid integration of conversational AI into health information seeking, there is currently no framework in the neuropsychiatric literature describing how AI interactions may relate to psychosis vulnerability. Existing reports are limited to isolated case descriptions without a common mechanism. This review addresses this gap.
Understanding the experiences of Intimate Partner Violence (IPV) in adolescent dating relationships is critical to addressing its physical and psycho-social impacts. Intervention research on IPV in dating relationships often focuses on high-income contexts. To address this gap, we conducted a systematic review of qualitative studies on dating violence among adolescents and young people aged 10 to 25 years in Low-and-Middle-Income Countries (LMICs). Twelve databases were systematically searched last in March 2023. Twenty-three publications describing 22 qualitative studies across 13 countries met the inclusion criteria. Thematic synthesis of the included study findings led to five themes: experiences of violence, causal attributions, impact, help-seeking and expectations of support. Findings revealed cross-regional experiences of physical violence, sexual coercion, control, and emotional violence both offline and online, associated with physical harm and adverse psycho-social effects. Traditional gender norms were key drivers of dating violence, maintained and normalised by myths of romantic love. Help-seeking was limited to informal sources, unless medical attention was necessary. Our findings highlight the severity of harms associated with dating violence and the limited resources available for adolescents in low-resource settings. We make a case for developing prevention interventions that address risk factors at multiple levels - individual, interpersonal, social and cultural - and that overcome the barriers identified by adolescents and young people from LMIC settings. The review protocol was registered prospectively on PROSPERO (ID: CRD42021267236).
The pathogenesis of critical illness drives a dynamic cascade of clinical, metabolic, and immunologic abnormalities through the course of disease. Recent randomized trials show that early full-dose feeding for critically ill patients is ineffective and potentially harmful compared to trophic feeding. The aim of this review is to propose a response-contingent phase-specific strategy of nutrition therapy appropriate for this patient population. Barriers to delivering effective nutritional therapy exist, ranging from bioenergetic failure related to mitochondrial dysfunction and gastrointestinal dysmotility with feeding intolerance, to microbial dysbiosis and impaired substrate utilization characterized by inefficient energy production and futile substrate cycling. Emerging concepts are helping to design more appropriate regimens, including estimation of potential benefit from nutritional therapy based on disease severity, risk of net harm, and determination of the stage of enteral nutrition responsiveness, which can be conceptualized as the capacity for achieving phase-specific goals within the constraints of physiologic state, metabolic capacity, safety, and feasibility. Calibrating goals, dosage, monitoring strategies, and expectations from nutritional therapy is required to overcome barriers imposed by the underlying pathophysiology and to optimize support, as these complex patients transition through the acute phase of critical illness towards recovery and rehabilitation.
Since 2016, foster care entries declined by over 35%. This amounts to a net difference of 444,000 fewer children who entered foster care over this period. Despite the profound ramifications of this decline in terms of the health and well-being of children and families, this topic remains largely unexplored. In this discussion article, we investigate two questions foundational to understanding the decline: 1) Is the decline concentrated among certain populations, and 2) can it be explained? In 1), we examine if and how states have differentially driven the decline and how the national composition of entries has changed by age, race/ethnicity, urbanicity, and prior maltreatment. In 2), we consider multiple potential explanations for declining foster care entries, including the role of the Family First Prevention Services Act, the use of kinship diversion, and reductions in child harm or risk of harm. We present available data that does or does not support each explanation and discuss each in some detail. In short, we do not find the decline is substantially concentrated within any one population (e.g. age, race/ethnicity, state politics), and is instead experienced broadly. At this time, we find no strong evidence for any single explanation.
In the United States and worldwide, there is an increasing body of reliable evidence suggesting that ultra processed food intake is linked to many adverse outcomes. Prostate is one of the most common cancers and has dietary influences. Data are sparse about ultra processed foods and prostate cancer. We used nationally representative data from the United States National Health and Nutrition Examination Survey from 2003 to 2023. Ultra processed food intake was estimated from two 24-hour dietary recalls using the validated and widely used NOVA classification and categorized into quartiles based on the percentage of total energy intake. Prostate cancer was defined by self-report. We used relative risks to estimate effects. We used 90% confidence intervals to test for significance as harm was prespecified. Crude and multivariable logistic regression models were adjusted for age, smoking status, race/ethnicity, and poverty status. Among 17,024 male adults, compared to those in the lowest, participants in the combined second, third, and fourth quartiles had a significant 29% higher risk of prostate cancer (p = 0.031), while those in the combined third and fourth quartiles had a significant 30% higher risk (p=0.035). Perhaps due to smaller sample sizes, those in the fourth quartile had a possible but non-significant increase of 31% (p=0.079). In adjusted analyses, corresponding values were 27% (p=0.045), 31% (p=0.037), and 34% (p=0.072). In these data, people with higher ultra-processed food intake had higher risks of prostate cancer. Analytic studies are necessary to test this hypothesis.
Nordic countries have traditionally governed drug use and related policies either as criminal or health-related issues - an approach described as a dual-track model. Far less attention has been paid on human rights perspectives in the Nordic drug policy context. Internationally, human rights challenges in drug policies have been increasingly addressed by civil society organizations representing people who use drugs, as well as harm reduction actors. In 2026, the human rights perspective was brought on the Nordic political agenda more prominently than before as, for the first time, the Nordic civil society organizations representing people who use drugs held a joint international conference on the human rights theme. This report provides notes from the conference and discusses current international and Nordic debates on drug policy as a human rights issue.