TABLE 1.
Characteristics of included studies
| First author (year) | Study focus | Study aim | Study type (subtype) a | Number of participants | Country | Age | Females (%) | Study findings in relation to research question | Quality rating – Total |
|---|---|---|---|---|---|---|---|---|---|
| French et al. (2012) | ARMS identification tool (RQ1) | To assess the ability of the Primary Care Checklist (PCCL) to accurately identify individuals with an ARMS. | Quantitative (descriptive) | 176 (83% met the diagnostic criteria for ARMS); 37 (21%) screened with PCCL by their GP | UK | 14–34 (M = 20.78, SD = 4.16) | 37.5% b |
Instrument used: PCCL checklist (French & Morrison, 2004) Findings: PCCL checklist has high sensitivity and low specificity in identifying ARMS adolescents. Better sensitivity/specificity ratio for short 6‐item version or the original version with different weighting |
4 (medium) |
| Quijada et al. (2010) | ARMS identification tool (RQ1) | To describe and evaluate an ARMS screening programme and the demographic and clinical characteristics of people presenting to the service. | Quantitative (descriptive) | 20 b individuals with an ARMS | Spain | 14.7–16.8 b | 40% b |
Instrument used: Spanish version of ERIraos checklist (Maurer et al., 2006) Findings: ERIraos checklist could help identifying individuals with an ARMS in primary care. |
3 (medium) |
| Chen et al. (2019) | Strategy to improve identification of ARMS in primary care (RQ2) | To identify common symptoms and patterns of symptoms presented to the GPs prior to the diagnosis of first psychotic episode. | Quantitative (non‐RCT – Case–control study) | 3045 patients with recorded FEP and 12 180 controls | UK | 16–45 (Me = 30) | 37.1% |
Strategy: Examination of patients' medical records Findings: Patterns of consultations: Higher number of GP consultations in patients who later develop psychosis Symptoms: Mood disorders, ‘neurotic’ symptoms, behavioural change in volition, substance misuse, physical symptoms, perceptual changes (relatively rarely, but significantly more common than in healthy controls). Three distinct prodromal patterns: (1) no/minimal symptoms cluster (if symptoms, then mood or physical health), (2) Mood cluster (most commonly 2 symptoms, e.g., low mood and ‘neurotic’ symptoms), (3) multiple symptom cluster (e.g., mood, physical health, behavioural problems). The first two clusters were more common. Cluster one likely youngest and male; cluster three likely older and more likely female and long DUP. Time consultation‐diagnosis: 2–2.5 years (shorter for perceptual changes). |
5 (high) |
| Falloon et al. (1996) | Strategy to improve identification of ARMS in primary care (RQ2) | Evaluation of the ‘Buckingham project’ – collaboration between GPs and specialist mental health services. (pilot study) | Quantitative (descriptive) | 18 GPs | UK | n/a | n/a |
Strategy: Different service set‐up Findings: Formal screening for ARMS in GP setting, combined with family and specialized mental health support resulted in reduced incidence of schizophrenia in the area. |
Not assessed |
| Perez et al. (2015) | Strategy to improve identification of ARMS in primary care (RQ2) | Establishing if 1) low intensity (postal information campaign) or 2) high intensity (postal information + theory‐based educational intervention) lead to different outcomes in terms of the number of ARMS referrals from primary care. | Quantitative (RCT) | 30 GP practices included in high‐intensity intervention and 34 in low intensity intervention (from Peterborough and Cambridgeshire) | UK | n/a | n/a |
Strategy: ARMS educational intervention Findings: High intensity intervention was more effective than low intensity intervention in increasing the number of referrals to first episode psychosis and ARMS services. High number of true and false positives referred via the high intensity group. Intervention was costly but has a potential to lead to long‐term savings due to earlier detection/intervention. Low intensity intervention was no more efficient than no intervention. |
5 (high) |
| Platz et al. (2006) | Strategy to improve identification of ARMS in primary care (RQ2) | To assess help‐seeking behaviours and main presenting symptoms of individuals with an ARMS presenting to the GPs. | Quantitative (non‐RCT – Cohort study) | 50 individuals with an ARMS | Switzerland | 21 | 38% |
Strategy: Examination of patients' self‐reported symptoms and help‐seeking behaviour Findings: Symptoms: Depression, social decline, social withdrawal. ‘Typical’ psychosis symptoms (e.g., hallucinations) were less common/rare compared to the first‐episode psychosis group. Patterns of consultations: GPs were most consulted for negative/non‐psychosis‐specific symptoms (e.g., depression). |
3 (medium) |
| Reynolds et al. (2015) | Strategy to improve identification of ARMS in primary care (RQ2) | Evaluation of GP training (1 session) on ARMS recognition and referrals to appropriate service. | Quantitative (non‐RCT – Cohort study) | 29 GP practices; 54 individuals referred/identified as ARMS by the GPs | UK | M = 21.85 (SD = 4.16) | 41% |
Strategy: ARMS educational intervention Findings: 50% of referrals by the GPs were correctly identified as ARMS. 1 h GP training increased the number of ARMS (but not EIP) direct referrals. Increased number of false and true positives. |
3 (medium) |
| Simon et al. (2010) | Strategy to improve identification of ARMS in primary care (RQ2) | To see if a repeated exposure (sensitisation) to the clinical vignette can improve diagnostic knowledge of ARMS in GPs. | Quantitative (non‐RCT – Cohort study) | 750 GPs b | Switzerland | n/a | n/a |
Strategy: ARMS educational intervention Findings: GPs exposed to the intervention showed a significant improvement in diagnostic knowledge of ARMS at 6‐ and 12‐month follow‐up (the effect persisted after sensitisation). This was not observed for non‐sensitized GPs. Sensitized GPs also improved their knowledge of symptoms of ARMS that are often under‐identified (e.g., social withdrawal and functional decline). |
3 (medium) |
| Sullivan et al. (2018) | Strategy to improve identification of ARMS in primary care (RQ2) | To see if a primary care consultation pattern for ARMS can be used to identify patients who later develop psychosis. | Quantitative (non‐RCT – Case–control study) | 530 primary care practices; 11 690 patients with psychosis and 81 793 matched controls | UK | M = 51.34 (SD = 21.75) | 57.4% b |
Strategy: Examination of patients' medical records Findings: Symptoms: Bizarre behaviour, suicidal behaviour (strongest predictor), cannabis‐associated problems, depressive symptoms, blunted affect, ADHD‐like symptoms, OCD‐like symptoms, social isolation, role functioning problems, mania symptoms, sleep disturbance, smoking‐associated problems. Positive predictive value of symptoms greater for men than women. Patterns of consultations: More common GP consultations; increasing number of consultations over time. |
5 (high) |
| Jacobs et al. (2011) | Barriers/facilitators to identifying ARMS in primary care (RQ3) | Understanding GPs appraisals of ARMS. | Quantitative (descriptive) | 72 GPs | US | M = 52.7 b (Me = 53.0) | 47.6% b | Barriers: Lack of knowledge about ARMS (i.e., thinking about it as schizophrenia spectrum); lack of diagnostic category to capture the symptoms of ARMS. | 3 (medium) |
| Jacobs et al. (2012) | Barriers/facilitators to identifying ARMS in primary care (RQ3) | Exploring how different practitioners (GPs, clinical psychologists and psychiatrists) treat individuals with an ARMS. | Quantitative (descriptive) | 68 primary care practitioners | US | M = 52.6 (SD = 10.9) | 48% b | Barriers: Lack of knowledge about ARMS, its identification and treatment. | 3 (medium) |
| Russo et al. (2012) | Barriers/facilitators to identifying ARMS in primary care (RQ3) | To identify factors that influence the identification of individuals with an ARMS in primary care using theory of planned behaviour. | Quantitative (descriptive) | 82 GPs | UK | M = 45.6 (SD = 9.4) | 47% |
Barriers: Thinking that their peers (e.g., psychiatrist) would not approve them diagnosing individuals with an ARMS (systemic barriers); low level of confidence and perceived control over identification of ARMS, lack of skills. Facilitators: Positive attitudes and intentions towards identifying individuals with an ARMS, high personal motivation/interest in ARMS and mental health, knowledge of patient and their background. |
3 (medium) |
| Simon et al. (2009) | Barriers/facilitators to identifying ARMS in primary care (RQ3) | The international GP study on early psychosis ‐ to assess the diagnostic knowledge, treatment practices, attitudes, and preferences for support of GPs in different countries. | Quantitative (descriptive) | 2784 GPs | International ‐ Switzerland, Austria, UK, New Zealand, Czech Republic, Canada, Australia, Norway | M = 46.4 (SD = 9.44) | 45.30% |
Barriers: Lack of knowledge about ARMS (about symptoms of ‘functional decline’), lack of low‐threshold services to refer individuals with an ARMS to. Facilitators: Good knowledge of ‘positive symptoms’ of psychosis; being a ‘gatekeeping GP’ (have better diagnostic knowledge than non‐gatekeeping GPs), practicing in ‘Commonwealth’ countries (have better diagnostic knowledge than continental European GPs). |
3 (medium) |
| Smith et al. (2021) | Barriers/facilitators to identifying ARMS in primary care (RQ3) | To understand GPs' comfort and understanding of ARMS; to understand GPs' interest in specialized training. | Quantitative (Descriptive) | 75 GPs | Australia | n/a | n/a |
Barriers: Lack of knowledge about ARMS (31% of GPs not aware of the concept of ARMS) Facilitators: Motivation for further training (almost all (95%) of GPs interested in further training in YP mental health). |
3 (medium) |
| Strelchuk et al. (2021) | Barriers/facilitators to identifying ARMS in primary care (RQ3) | To investigate GPs views about identifying individuals with an ARMS in primary care. To identify barriers and facilitators related to the identification. | Qualitative | 20 GPs | UK | 32–63 (M = 46.0, SD = 8.6) | 40% |
Barriers: Lack of knowledge about ARMS, lack of mental health training, diagnostic similarities between ARMS and other mental health problems, lack of diagnostic categories (e.g., ARMS), lack of time (short GP consultations), difficulties making appointment with the GP, high threshold for accessing secondary care, difficulties about getting an appointment in secondary care, fears about labelling patients, patients not seeking help due to lack of motivation, depression and stigma, Facilitators: Increasing knowledge about specialist referrals and ARMS treatment. |
5 (high) |
| Tor and Lee (2009) | Barriers/facilitators to identifying ARMS in primary care (RQ3) | To compare attitudes of Singapore psychiatrists vs. GPs about ARMS. | Quantitative (descriptive) | 107 primary care practitioners | Singapore | 57.9% aged between 30 and 40 | 51.40% | Barriers: Lack of knowledge about ARMS (GPs more likely to diagnose patients with psychosis), lack of confidence in identifying ARMS (less than a third of GPs advocate for screening for ARMS in high‐risk groups), low confidence in treating individuals with an ARMS (almost all GPs not wanting to treat ARMS), low tolerance of psychosis‐like symptoms. | 2 (low) |
Abbreviations: ADHD, attention‐deficit hyperactivity disorder; EIP, early intervention in psychosis; ERIraos, Interview for the Retrospective Assessment of the Onset and Course of Schizophrenia and Other Psychoses (German version); n/a, the study did not report participants' gender or age; OCD, obsessive–compulsive disorder; RQ1‐3, Research Question 1–3, PCCL, Primary Care Checklist; YP, young people.
Study type as defined by MMAT (quality appraisal tool).
Study characteristics reported in relation to the whole sample (i.e., sub‐group statistics were not reported).