select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
A study published in the Journal of Epidemiology & Community Health looks at primary care utilisation for mental health problems.
Prof Stella Chan, Charlie Waller Chair in Evidence-Based Psychological Treatment, University of Reading, said:
“This study examined the trends of mental health prevalence in Norway across fourteen years. Using routinely collected administrative data from primary care, the analyses were robustly conducted without being compromised by sampling biases that typically confounded research. While the methodological details were appropriate, the dataset was insufficient in identifying the causes underlying the observed changes. In other words, the study told us about what had been recorded in primary care services in Norway, but it offered limited insights into whether, and how, mental health has changed over time.
“Key findings suggested a significant increase in symptom-level presentations of depression and anxiety, but not disorder-level. While this trend is interesting, readers should be aware that within the field of mental health science and practice, there has been a longstanding debate and controversies around the way we define mental health problems. While the findings could reflect changes in help-seeking and practitioners’ coding over time, there is no denying that any increases in symptoms should be taken seriously as an indicator of deteriorating mental health. On a positive note, improvement in help-seeking behaviours is an encouraging news. The key implication here is to implement more effective preventative intervention to support these individuals at an early stage before symptoms develop into a more severe and recurrent pattern.”
Professor Celso Arango, Chair of the Child and Adolescent Department of Psychiatry at Hospital La Paz, CIBERSAM, said:
“The main limitation is that the study cannot disentangle whether the findings reflect lower help-seeking thresholds, changes in GP referral practices, or a true increase in mental disorders. It is also possible that GPs increasingly refer individuals with milder symptoms and that a broader cultural and clinical shift towards using less pathologising terminology (e.g., mental health problems, psychological distress, or emotional difficulties rather than formal psychiatric disorders) has influenced both coding and referral practices. The authors themselves note that GPs may preferentially use symptom codes to avoid stigmatising patients, suggesting that secular changes in diagnostic language may partly explain the observed trends.
“Finally, these findings should not be generalised to other psychiatric conditions, particularly neurodevelopmental disorders, whose incidence and prevalence have also increased substantially over recent decades and are likely driven by different mechanisms.
Comments from our friends at SMC Spain:
Javier José Pérez Flores, a lecturer in the Department of Psychobiology at the University of La Laguna, says:
“The study has one clear strength: it draws on a very large sample and covers a sufficiently long time period to detect relevant epidemiological trends. This type of design allows us to observe changes in the way mental health problems are referred to primary care and recorded in healthcare systems. However, precisely because of the breadth of the data, we should be cautious in our interpretation. What the study shows is not necessarily a direct change in the actual prevalence of mental disorders, but rather a change in the way certain psychological problems are recorded, coded and possibly addressed within the Norwegian healthcare system.
“The first important point is that these findings are not new in this field of study. The discrepancy between the prevalence of symptoms and the prevalence of diagnoses had already been highlighted over the past decade. A clear example is the work by Archer et al., published in 2022, which described a very similar trend in the United Kingdom in relation to anxiety symptoms. There, too, an increase in records relating to symptoms was observed, without this automatically translating into an equivalent increase in formal diagnoses of mental disorders.
“This distinction is fundamental. In the case of the Norwegian study, what the data appear to show is that, between 2010 and 2024, the rise in mental health contacts in primary care is concentrated particularly in records of symptoms of anxiety and depression. By contrast, the codes corresponding to disorders have increased much less or remained relatively stable. That is, strictly speaking, the conclusion that the data support. Any further interpretation (for example, that there is more suffering but not more disorders, or that diagnoses are on the decline) requires much greater caution.
“In fact, the study by Archer and colleagues itself helps to explain why. That study included interviews with doctors, and some professionals noted that they preferred to code symptoms rather than disorders in order to reduce the stigma associated with psychiatric diagnosis. This raises the possibility that the increase in the coding of symptoms does not necessarily imply that there are fewer diagnosable cases, but rather that professionals are choosing to label these problems differently. In other words, part of the phenomenon could lie in the coding practice itself.
“Something similar occurs when the study is interpreted from a public policy perspective. At first glance, it might seem that these results call for a fundamental rethinking of the organisation of mental health care. However, they can also be interpreted as the consequence of health policies already in place in Norway. In 2012, the country introduced a reform that gave local authorities greater autonomy in the provision of health services, reinforcing a model more heavily reliant on primary care and community resources.
“This approach ties in with the Escalation Plan for Mental Health 2023–2033, which explicitly sets out the need to lower the threshold for receiving help and to prioritise municipal services over a model that is overly focused on specialist care. The plan itself acknowledges that many people with mental health problems are already in contact with their GP and envisages interventions such as supportive conversations, stress management courses or brief treatments for mild forms of anxiety and depression. In other words, the Norwegian system appears to recognise a middle ground between psychological distress and a formally diagnosed mental disorder.
“The Prompt Mental Health Careprogramme, launched as a pilot in 2012, should also be viewed in this context; it is specifically designed to provide psychological treatment within primary care for mild or moderate symptoms of anxiety and depression. Seen in this light, the rise in reported symptoms need not be interpreted solely as an epidemiological warning sign, but also as a reflection of an organisational decision: to identify distress earlier, address it through more local services, and not necessarily wait for it to take the form of a fully codified disorder.
“Therefore, I do not believe it can be concluded that there are fewer diagnoses in Norway despite more symptoms being reported. What we are seeing, rather, is a combination of changes in professional coding, expanded access to primary care and the choice of specific approaches to addressing psychological distress. Whether this is positive or negative is another matter. To assess this, we would need to analyse whether these people receive adequate care, whether serious cases are referred appropriately, whether unnecessary medicalisation is avoided, or whether, on the contrary, problems that would require more specialised intervention are trivialised.
“One final piece of data sheds light on the issue from another angle. Anmella and colleagues analysed the prescribing of antidepressants in primary care in Catalonia between 2010 and 2019 and found a very marked increase in antidepressant prescriptions, far exceeding the rise in diagnoses of depression. This type of finding raises an uncomfortable question: to what extent does it matter that the system codes symptoms or disorders if, in practice, the approach ends up being the same?
”That is, probably, the underlying issue. The debate should not be limited to whether symptoms are on the rise or diagnoses are increasing, but rather to what is done for people who present at primary care with psychological distress. If symptoms are coded to reduce stigma, facilitate access and provide appropriate brief interventions, this may be a reasonable strategy. If, on the other hand, symptoms are coded but the response is always the same pharmacological approach or inadequate care, the change in terminology adds little.”
José César Perales, professor in the Department of Experimental Psychology at the University of Granada, said:
Is it of good quality? Are the conclusions backed up by solid data?
“This study examines consultations for mental health reasons in a cohort of 3.7 million users of the primary care system in Norway over several years (2010–2024). It considers different age groups, from pre-adolescents and adolescents to young adults, and includes a comparison group of adults aged between 31 and 46. The analysis distinguishes between consultations coded as diagnoses of depressive or anxiety disorders and those recorded as the presence of symptoms of depression or anxiety that do not meet the threshold for a clinical diagnosis.
“The descriptive data show that the increase in symptom codings over the period analysed clearly outstrips that observed in diagnostic codings; in other words, there is a clear decoupling between the two trends, with a marked shift towards symptom codings.”
“The size and representativeness of the sample (virtually the entire population of Norway in the age groups under consideration), together with the stability of the trends, make it unlikely that this divergence is due to chance. Everything points to this being a consistent pattern in the population studied.
“One plausible interpretation is that the recent increase in mental health-related consultations – and consequently their greater visibility both amongst professionals and in the public sphere – is largely due to changes in people’s help-seeking behaviour and to changes in clinical coding practices. From this perspective, the recorded increase would reflect not so much a change in the underlying incidence of disorders as a shift in how they are detected, named and recorded.
However, this interpretation should be qualified. Whilst it is consistent with the data, it does not rule out other possible explanations. It is important not to confuse the quality and robustness of the data with the strength of a particular interpretation of them”.
How does this work fit in with the existing evidence?
“The study is based on a well-known and indisputable fact: the rise in diagnoses of depression and anxiety disorders, as well as in consultations for associated symptoms—particularly amongst adolescents and young adults—can be partly explained by a greater propensity to seek help or by improved detection. The available evidence supports this possibility.”
What are its limitations?
“The existence of this phenomenon of diagnostic inflation does not entirely rule out the possibility that there has also been an increase in underlying mental health problems. It would, however, explain why these underlying trends appear to be less clear-cut and universal than is claimed by certain viewpoints, which are not always strictly scientific.
“Assessing these underlying trends presents an inescapable difficulty. We do not have objective indicators that allow us to measure them directly. As this study shows, diagnoses alone do not provide information on undetected cases, whilst studies based on self-reports have their own inherent limitations.
“Among these limitations, a possible shift in the meaning and use of expressions such as ‘being depressed’ or ‘suffering from anxiety’ stands out; this may lead different cohorts to interpret and respond to the same questions in non-equivalent ways.
“In this context, any attempt to infer the true trend in mental health problems involves a high degree of interpretation. It is therefore advisable to adopt a cumulative perspective and consider the body of evidence as a whole, drawn from diverse methodologies and employing different analytical strategies. Focusing the discussion on a single study narrows the scope of the conclusions and increases the risk of overreacting to results which, on their own, cannot settle the debate.”
What are the implications for public policy?
“As regards the measures to be adopted, it is advisable to maintain a cautious stance without succumbing to alarmism. The available evidence does not support the idea of a disproportionate increase in mental health problems among adolescents in recent years. An excessive reaction to a perceived threat can be just as harmful as inaction.
“Beyond the possible over-interpretation of trends, the priority is to identify the factors that have a significant impact on adolescent mental health and to design coordinated interventions targeting them. This requires tailoring the intensity of interventions according to the weight each factor carries within the general population or specific groups. It also involves avoiding single-cause explanations and resisting the temptation to focus attention on the most visible determinants at the expense of others that are less apparent but potentially more influential.”
‘Secular trends in primary care utilisation for mental health problems: a Norwegian register based population-wide study’ by Kathryn Christine Beck et al. was published in Journal of Epidemiology & Community Health at 23:30 UK time Tuesday 30th June 2026.
DOI: 10.1136/jech-2026-226059
Declared interests
Prof Stella Chan: “No CoI.”
Professor Celso Arango: “Dr. Arango has been a consultant to or has received honoraria or grants from Abbott, Acadia, Ambrosetti, Angelini, Biogen, BMS, Boehringer, Carnot, Gedeon Richter, Janssen Cilag, Lundbeck, Medscape, Menarini, Minerva, Otsuka, Pfizer, Roche, Rovi, Sage, Servier, Shire, Schering Plough, Sumitomo Dainippon Pharma, Sunovion, Takeda and Teva.”
Javier José Pérez Flores: He declares that he has no conflicts of interest.
José César Perales: He declares that he has no conflicts of interest