First-episode psychosis : primary care experience and implications for service development

RESULTS Out of 284 GPs, 123 (43%) responded. General practitioners are unlikely to start treatment before referring to a specialist service: 63 GPs (51% of responders) start treatment in10% or less of individuals with first-episode psychosis and 19 (15.5%) GPs start treatment in 75% or more before referring them to psychiatric service; 42 GPs (34%) refer those who request/accept a referral and 66 GPs (53%) refer all even if they refuse. Overall, 92 GPs (74%) agreed that an early intervention in psychosis service is needed and 77 (63%) GPs welcome having a mental health clinic in their surgery.

First-episode psychosis: primary care experience and implications for service development

AIM AND METHODS
To capture the local primary care experience of first-episode psychosis before developing a local early intervention in psychosis service.A survey of Northamptonshire general practitioners (GPs) using a confidential questionnaire was carried out.

RESULTS
Out of 284 GPs, 123 (43%) responded.General practitioners are unlikely to start treatment before referring to a specialist service: 63 GPs (51% of responders) start treatment in10% or less of individuals with first-episode psychosis and 19 (15.5%)GPs start treatment in 75% or more before referring them to psychiatric service; 42 GPs (34%) refer those who request/accept a referral and 66 GPs (53%) refer all even if they refuse.Overall, 92 GPs (74%) agreed that an early intervention in psychosis service is needed and 77 (63%) GPs welcome having a mental health clinic in their surgery.

CLINICAL IMPLICATIONS
Individuals are more likely to accept referral to a psychiatric service if offered than to ask for it.People disengaging, stigma, the service being difficult to access/inappropriate and carers' lack of knowledge about mental illness are the likely causes for delayed referral.
The attitude to treating psychosis has recently changed from simply focusing on severe enduring mental illness to also including early intervention. 1Treating psychotic disorders in their earliest stages has become a focus for research and clinical care. 2 New terms appeared including: first-episode psychosis; duration of untreated psychosis (DUP; from onset of positive psychotic symptoms until starting treatment); and duration of untreated illness (from onset of prodrome until starting treatment). 3First-episode psychosis studies show that the average time between onset of symptoms and first effective treatment is often 1 year or more. 4Long DUP is undesirable for various reasons.
1. Early treatment helps minimise the risk of serious consequences as a result of changes in mental state and behaviour. 5,62. Shorter DUP is associated with better clinical response. 7. Early treatment can reduce suffering. 8. Early results suggested that an early intervention in psychosis service is more cost-effective than generic services. 9ans have been put in place to develop a network of early intervention in psychosis services across the UK. 10 Most general practitioners (GPs) see one to two new people with first-episode psychosis a year. 11This small figure may lead to underestimating the importance of the primary care role of early intervention in psychosis.However, Shiers & Lester believe that early intervention in psychosis needs greater involvement of primary care for its success. 11Our study examined the local primary care experience of first-episode psychosis before developing the local early intervention in psychosis service.

Method
We undertook a cross-sectional survey of GPs using a colour-coded confidential questionnaire of eight questions requiring ticks to multiple-choice options and free text comments.The questionnaire was developed by M.E.-A., reviewed and approved by a panel of ten clinicians including five psychiatrists, three community psychiatric nurses, one mental health social worker and one GP with an interest in mental health.The questionnaire focused on GP's experiences with individuals with first-episode psychosis regarding referral trends, prescribing medication, engaging in treatment, preferred model of service and causes of delayed referral to the psychiatric service.The clinical governance support team sent the questionnaires to all Northamptonshire GPs in 2004, with a covering letter highlighting the importance of their contribution, as it would influence the local early intervention in psychosis model, and requested their response within 3 weeks.Non-responders were not followed-up.Data collection and analysis was carried out by the clinical governance support team.
Northamptonshire is in the eastern part of central England with an area of 2364 km 2 and a population of 646 600 (mid-2003 estimate) including: 95.1% White, 2% South Asian, 1.2% Black African and Caribbean, and 1.8% from other ethnic minority groups.Population density is 272/km 2 .Northamptonshire is largely rural and Northampton is the largest town with a population of 194122.

Results
Out of 284 GPs, 123 (43%) responded.The response rate varied between 27% and 53% across the eleven catchments areas.Fifteen (7%) GPs requested a copy of the final report.

Starting treatment
Sixty-three GPs (51% of responders) start treatment in 10% or less of individuals with first-episode psychosis before referring them to the psychiatric service and eleven GPs (9% of responders) start treatment in 25% of individuals.Nineteen GPs (15.5% of responders) start treatment in 50% of people with first-episode psychosis and nineteen GPs (15.5%) start treatment in 75% or more of people before referring to the psychiatric service.Eight GPs (7%) were unsure and three GPs (2%) did not answer.

GP referral trends
Five GPs (4% of responders) do not refer individuals with first-episode psychosis to the psychiatric service until the clinical picture is clear.Two GPs (2%) only refer those who request a referral, whereas nine GPs (7%) refer those who accept referral.Forty-two GPs (34%) refer those who request or accept being referred.Sixty-six GPs (53%) refer all people with first-episode psychosis even if they refuse being referred.

Individuals'attitude towards referral
Twenty-one GPs (17%) did not know the proportion of people with first-episode psychosis that request or accept referral to a psychiatric service (Fig. 1).Five GPs (4%) believed that 75% of individuals with first-episode psychosis ask for referral, 11 GPs (10%) believed that 50% ask and 51 GPs (42%) believed that only 10% or less ask.If people with first-episode psychosis are offered a referral, 58 GPs (48%) believed that 75% or more would accept, 23 GPs (19%) believed that 50% would accept and 11 GPs (9%) believed that only 25% would accept.

Engaging with treatment
In total, 116 GPs (95%) answered this question and 7 GPs (5%) did not (Fig. 2).Twenty-one GPs (17%) responded with 'don't know' about first-episode psychosis engagement before or after referral.Eighteen GPs (15%) believed that 75% engage before referral, whereas 77 GPs (63%) believed that 75% engage after referral.Twenty-seven GPs (22%) believed that 50% are likely to engage before referral and twelve GPs (10%) believed that 50% are likely to engage after referral.Nineteen GPs (16%) believed that 25% are likely to engage before referral and five GPs (4%) believed that 25% engage after referral.Thirty-one GPs (25%) believed that 10% or less are likely to engage before referral and one GP (1%) believed that 10% or less engage after referral.

Number of individuals with first-episode psychosis
We asked the GPs how many individuals with firstepisode psychosis they had seen in the past 12 months.Overall, 121 GPs (98%) answered and 2 GPs (2%) did not; 84 GPs (68%), saw one to two people with first-episode psychosis, 16 GPs (13%) saw three to five, 3 GPs (2%) saw five or more and 18 GPs (15%) did not see any.
Is there a need for an early intervention in psychosis service?
Overall, 5 GPs (4%) did not feel the need for an early intervention in psychosis service, 26 GPs (21%) did not know and 92 GPs (74%) agreed that an early intervention in psychosis service is needed (Fig. 3).If there was an early intervention in psychosis service, 3 GPs (2%) would not use it, 13 GPs (11%) did not know and 107 GPs (87%) would use it (Fig. 4).

Preferred service model
A total of 119 GPs (97%) answered and 4 GPs (3%) did not answer.Seventy-seven GPs (63%) welcome having a mental health clinic in their surgeries with 37 GPs (30%) preferring a mental health clinic run by a clinician from the mental health service, 12 GPs (10%) preferring a clinic run jointly by the GP and a mental health clinician and 28 GPs (23%) accepting either.However, 21 GPs (17%) refused having a mental health clinic in their surgery and 21 GPs (17%) had other views including: no need to slice an already overstretched service to give it a new name with another waiting list; GPs wish to have an open access to psychiatric out-patient slots; and urgent home visits at GP request are more important.

Discussion
This is the first UK primary care survey that studied GPs' experiences of first-episode psychosis with a relatively good overall response rate compared with similar studies (e.g. a postal survey in 1995-6 of GPs in the Midlands produced an initial response rate of 32%). 12This may reflect the growing importance of mental health in primary care.

GPs'prescribing and referral trends
Study results show wide variation in GPs' experiences with first-episode psychosis.General practitioners are likely not to start treatment in first-episode psychosis before referring to a specialist service with variations in prescribing trends that may reflect differences in GPs' training, experience and interest.
General practitioners' tendency to only refer individuals with first-episode psychosis when the diagnosis is clear or those who request and/or accept referral may lead to delayed access to specialist care.Later, many of those individuals will settle into a relapsing pattern, present in emergency situations to be dealt with by psychiatric services in a fire-fighting manner exhausting healthcare resources. 13Referring all individuals with firstepisode psychosis even if they refuse is likely to be associated with a high rate of 'did not attend'.Thus, early intervention in psychosis services need to facilitate access to care and GPs need to be more proactive with a higher index of suspicion and a lower referral threshold.This would agree with the findings of Edwards & McGorry who stated that GPs have a potentially important role in early recognition and referral of first-episode psychosis. 14in causes for delayed referral Interestingly, GPs identified individuals disengaging, services being difficult to access or inappropriate, stigma, carers' lack of knowledge about mental illness and unclear diagnosis as the most important reasons for delayed referral of people with first-episode psychosis.Lincoln & McGorry identified similar factors in tackling delayed access to services in first-episode psychosis including stigma, clinical features, gender and inaccessible services. 15ost GPs see one to two new people with firstepisode psychosis a year which agrees with that indicated by Shiers & Lester. 11Some GPs reported figures of two or more new cases a year probably because of higher levels of morbidity and high turnover in their catchment area (e.g.Maple Centre for homeless individuals).In most industrial countries the incidence of new cases of schizophrenia is 15/100 000 a year. 16Northamptonshire's population is 650 000 (www.northampton.gov.uk/site/scripts/documents___info.php?documentID=171&pageNumber=2), i.e. about 100 new cases of schizophrenia are expected a year locally.Some may argue whether this number is big enough to require a new specialist service.However, early intervention in psychosis services will deal with all forms of early psychosis (e.g.affective, drug induced) as well as those that may prove later not to be psychosis.Early intervention in psychosis services may have to see and follow-up a much larger number of individuals to identify cases of psychosis.Some GPs may be inhibited from seeking specialist help for various reasons, for example a lack of knowledge about resources and how to access them, disillusionment with mental health services and a belief that individuals and families may be offended by recommending psychiatric help. 15It is vital that early intervention in psychosis service developers consider this and address the underlying issues relevant to GPs and professionals in other services (e.g.teachers).
Novel approach and a partnership are needed A major GP and public concern is waiting lists and times.It is important that early intervention in psychosis services do not develop long waiting lists/times and use a novel and less stigmatising approach, otherwise the whole concept will lose its meaning (i.e.'early intervention').
Early intervention in psychosis requires a partnership of primary and secondary care and wider involvement of the community to make a difference. 11Early results from Early Treatment and Identification of Psychosis (TIPS) in Norway and Denmark suggest that public education and specific education for teachers, youth workers and GPs about first-episode psychosis can assist early detection and reduce DUP. 11,17An early intervention in psychosis service model that suits an inner-city area of a big city such as London may not suit a small town or its surrounding rural areas.Therefore the local model has to be flexible and needs based. 18uestionnaire surveys of clinical practice do not necessarily provide an accurate picture of clinical process and activity.Care professionals completing the questionnaires might feel that their services are being scrutinised and provide 'acceptable answers'. 18This study questionnaire was confidential and focused on GPs experience, views and expectations.This is likely to help avoid or at least minimise such a tendency.

Conclusion
General practitioners' experience with first-episode psychosis indicates that people are more likely to accept referral to a psychiatric service if offered it than to ask for it.The likely causes for delayed referral of people with first-episode psychosis to a psychiatric service are: individuals disengaging, stigma, service being difficult to access or inappropriate, carers' lacking knowledge about mental illness and GPs waiting to make a diagnosis.General practitioners need to be adequately informed about early intervention in psychosis and the importance of their role for its success.

Declaration of interest
None.

Fig. 1 .
Fig. 1.General practitioner experience of first-episode psychosis: attitudes towards referral to psychiatric services.