A nationwide analysis of 350 million patient encounters reveals a high volume of mental-health conditions in primary care

Primary-care physicians (PCPs) are central providers managing acute and chronic conditions, prevention, and referrals. Despite many PCPs feeling ill-equipped for specialized psychiatric care, they are often the first contact for mental-health concerns. With global efforts to strengthen primary care and integrate mental-health services, accurate accounting of what PCPs manage is needed to inform training, service delivery, and system planning. The study asks: How much of PCPs’ work is devoted to mental-health conditions? What types and volumes of mental-health conditions do PCPs address across the lifespan? And how does this volume compare with encounters for other medical conditions across body systems? Leveraging comprehensive nationwide primary-care data from Norway (with universal assignment to PCPs and minimal cost barriers), the study aims to provide population-level estimates of mental-health workload in primary care and contextualize it relative to other conditions.

Prior reports vary widely regarding the share of primary-care visits involving mental health, with claims such as up to 70% (APA) and 40% (UK Mind). More systematic U.S. data from the National Ambulatory Medical Care Surveys (NAMCS) indicate increases in visits with a primary mental-health diagnosis from 3.4% (2006–2007) to 6.3% (2016–2018), but these surveys have limitations: variable response rates (<50% in 2018), a 1-week reporting window, and restriction to adults ≥18 years. The study positions itself as providing more comprehensive, population-wide evidence from a national system with complete primary-care coverage, addressing gaps in prior survey-based and adult-only data.

Design and data source: Population-based analysis of all primary-care encounters recorded in Norway from January 2006 through December 2019. All Norwegian residents are assigned a PCP; access to specialist care typically requires PCP referral. Visits are highly subsidized (free for juveniles), and PCPs submit at least one primary diagnosis or reason for visit for reimbursement, minimizing underreporting. Information is recorded using ICPC-2 (International Classification of Primary Care), organized into chapters by body systems. Population: 4,875,222 individuals (2,433,975 males; 2,441,247 females) born in Norway between February 1905 and December 2017, resident during the observation period or until death. Encounters for patients aged 0–100 years were included. Encounters: Included direct PCP–patient contacts (office, telephone, home visits). On average, each visit yielded 1.18 (sd = 0.47) codes; 84.5% had one code, 13.9% two, 1.6% three or more. Condition classification: Mental-health conditions were identified using ICPC-2 Chapter P (Psychological) and grouped into 24 categories (Supplementary Table 2). For comparisons with physical health, encounters were grouped by ICPC-2 chapters representing body systems: A (General/Unspecified; excluding A97 no disease and A98 health maintenance), B (Blood/Immune), D (Digestive), F (Eye), H (Ear), K (Cardiovascular), L (Musculoskeletal), N (Neurological), R (Respiratory), S (Skin), T (Endocrine/Metabolic/Nutritional), U (Urological), W (Pregnancy/Childbearing/Family Planning), X/Y (Female/Male Genital). ICPC-2 Z (Social Problems) was excluded (<0.5% of codes). Mental-health volumes were also compared with infections, pain symptoms/complaints, and injuries across the body. Units of analysis and measures: Two units were used—patients and encounters. Patient-level prevalence: number of patients with any mental-health code divided by total patients visiting PCPs over 14 years, stratified by age and sex. Encounter-level share: number of encounters with a mental-health condition divided by all encounters over 14 years, stratified by age and sex. Comparative analyses assessed the proportion of mental-health encounters versus other ICPC-2 chapters and versus infections, pain, and injuries. Statistical approach: Descriptive statistics and data visualizations only; no inferential testing due to the very large sample size (n > 250 million encounters). Analyses performed in SAS v9.4 and SYSTAT v14.0; graphs in R (tidyverse v4.0.2). Ethics approvals obtained from REK South East Norway and Duke University IRB.

• Population and volume: 4,875,722 patients generated 354,516,219–354,516,291 primary-care encounters (2006–2019).
• Patient-level prevalence: 47% (N = 2,309,787) of patients presented at least once with a mental-health condition. Age patterns: children often for sleep disturbances, continence issues, ADHD; young/middle-aged adults for depression; older adults for memory difficulties and sleep disturbances.
• Encounter-level share: 11.7% of all encounters involved a mental-health condition (n = 41,616,704). Mental-health encounters occurred across the lifespan.
• Composition of mental-health encounters (of 41,616,704 encounters): depression 23.8%; anxiety 14.1%; sleep disturbances 12.1%; substance abuse 8.3%; acute stress reaction 1.7%; psychosis 1.6%; dementia/memory problems 1.5%; ADHD 1.3%; phobia/compulsive disorders 1.7%; developmental delays/learning problems 1.5%; PTSD 1.3%; personality disorder 1.2%; <1% each for child/adolescent behavior problems, sexual concern, neurasthenia/chronic fatigue, eating disorder, somatization, continence issues, suicide/suicide attempt, adult phase-of-life problem, stammering/stuttering/tic, fear of mental disorder, irritability/anger. Unspecified/psychological symptoms or other disorders: 9.1% of mental-health encounters.
• Comparison to other body systems: Mental health accounted for 11.7% of all encounters, second only to musculoskeletal conditions (17.4%). Mental health was comparable to cardiovascular (12.1%) and respiratory (11.0%) and exceeded skin, endocrine/metabolic, digestive, neurological, urological, male/female genital, pregnancy/childbearing/family planning, eye, ear, and blood/immune chapters. PCPs had roughly twofold or greater volume of mental-health encounters relative to each of these other body systems.
• Temporal trend: Mental-health share rose from 11.0% in 2006 to 12.8% in 2019, preceding the COVID-19 pandemic.
• Additional comparisons: The volume of mental-health encounters was similar to infections and exceeded pain and injuries when examined across the body (per Supplementary Tables/figures).

The findings directly answer the research question: mental-health conditions constitute a substantial share of primary care workload—about one in nine encounters—and nearly half the population seeks primary care for mental-health concerns over a 14-year period. Mental-health encounters are diverse, spanning mild to severe conditions and all ages, not limited to depression and anxiety. Relative to other medical domains, only musculoskeletal conditions account for more encounters, underscoring the central role of PCPs in mental-health care. These results have implications for clinical training and system design: PCPs should be prepared as mental-health generalists; primary care should be supported to integrate mental-health services (for example, collaborative care and warm handoffs); and workforce planning should consider the high and diverse mental-health workload to mitigate burnout and turnover. The observed pre-pandemic increase suggests growing demand, highlighting the importance of strengthening primary care’s capacity to prevent escalation of mental-health problems and improve access to effective treatments across the lifespan.

Using comprehensive nationwide primary-care data from Norway, the study shows that mental-health conditions constitute 11.7% of all primary-care encounters (second only to musculoskeletal) and that 47% of patients present at least once for mental-health concerns across 14 years. Mental-health encounters are diverse and span all ages, indicating PCPs’ pivotal role in addressing mental health. These findings support enhancing mental-health training for PCPs, integrating mental-health services into primary care, and planning workforce capacity accordingly. Future research should assess care processes and outcomes within encounters (treatments provided and their effectiveness), incorporate secondary diagnoses and reasons for visit to capture the full mental-health burden, evaluate trends post-2019 (including pandemic effects), refine classification and reduce misclassification in primary care coding, and examine generalizability across different health systems and socioeconomic contexts.

• Generalizability may be limited outside Norway due to differences in healthcare and socioeconomic structures.
• ICPC-2 psychological codes reflect reasons for seeking care, not prevalence of mental disorders in the population; analyses estimate service contacts, not disorder prevalence.
• Only primary diagnoses/reasons for visit were analyzed; secondary diagnoses were unavailable, likely underestimating mental-health involvement in visits (physician-survey data suggest 2–3× higher when considering additional reasons/diagnoses).
• Classification challenges in primary care: potential misclassification (for example, between depression and anxiety), use of placeholder Chapter P codes when uncertain, and inter-doctor variation; ICPC-2, like other diagnostic systems, is imperfect.
• Inclusion of sleep disturbances and dementia as psychological codes may overestimate mental-health volume; conversely, exclusion of pain codes and fear-of-illness codes from mental health may underestimate it.
• No information on treatments delivered during encounters; cannot evaluate management quality or outcomes.
• Observation period ends in 2019; estimates predate COVID-19 and may underestimate current mental-health care volume.