A London trust has been auditing mental health referrals in a bid to improve the quality of screening in secondary care older adult mental health patients. Ranjit Mahanta and Seraphim Patel explain the results.

Stigma and awareness in mental health is still a challenge which needs to be overcome in the UK. Mental illness is the biggest single cause of disability in our society and costs the English economy £77bn a year in health and social care.

Over the last decade, there have been several Department of Health initiatives directed at raising awareness and reducing the stigma and discrimination associated with mental health problems. A key aim of this campaign has been to reach isolated groups and communities, hoping to promote discussion and acceptance of mental illness.

Through better identification of individuals with mental illness and the provision of subsequent treatment, the government aims to reduce the associated economic burden. During a time of economic crisis this ambition becomes even more important.

Placed within secondary care, specialist mental health services are in a unique position to manage the needs of patients; providing a range of diagnostic and therapeutic services including inpatient care for mental illness. Over time, several innovations have been attempted and implemented in an effort to improve access to specialist secondary care services. As such, in an effort to promote awareness and enhance access into services, many mental health trusts have adopted an informal ‘open-access’ referral policy for referrers. This strategy allows a liberal and open-minded approach to the identification of patients and at risk individuals; accepting referrals from a wide range of sources including social services, care homes and even the police.

In theory, this appears to be a relatively good approach; it maximises opportunities for intervention and support for the mentally unwell, however in reality there are inherent problems. This system relies on referrers having some knowledge and understanding of mental illness which may not necessarily be in place. If this is the case, referrals made to mental health teams can potentially be poor quality and lack sufficient information.

This problem has already been identified in some trusts with interventions put in place to address the issue. This included educating potential referrers, using uniform referral forms and putting in place a referral screening process. Particularly important is the referral screening process, which places the emphasis and responsibility of providing adequate referrals on the staff members within the accepting team. It is reasonable to assume that the people best placed to generate a referral may be those placed within the mental health teams.

Despite these safeguards, the success of individual referrals into mental health services can be varied and erratic. Many teams have seen an influx of referrals which warrant the intervention of secondary care as a result of this strategy, however, equally, a proportion of the referrals accepted have also been inappropriate, unnecessary and discharged before or after one assessment. This highlights the fundamental flaws in the open access policy which has been adopted by many mental health trusts, creating a system that can be inefficient and counter-productive. Successful implementation of this strategy depends on delivering training to competent individuals within mental health teams in triaging and managing referrals.

Evidence for the effectiveness of an informal referral policy within health services is relatively modest and studies investigating the referral and assessment consequences of an ‘open-access’ policy generally judge this initiative favourably. Earlier studies, demonstrated no significant increase of inappropriate referrals. Indeed, these studies found open-access to be more sensitive in picking up patients that would otherwise have been missed from more traditional referral methods. More recently in the US, one study found that an open-access process reduced waiting times, improved clinical productivity and evaluation of new referrals, and this was sustained for over 4 years.

In the UK, a recent study looking at referrals of older adults with mental illness from social services, found no increase in the total number of annual referrals, with over 90 per cent of referrals suffering from mental illness. The authors suggested that social services play an important role in identifying elderly mental illness and the acceptance of direct referrals is actively encouraged.

However, in 2003 a large study comparing the Netherlands GP ‘gate-keeping’ model and the German ‘open-access’ model for patients with psychological problems found that the GP gate-keeping model is an effective remedy against the over-utilisation of specialist services and the medicalisation of daily hazards within society.

Overall, most of the evidence suggests that an informal open-access referral system does enhance the identification of the mentally unwell however; the success of this system clearly lies within the management of referrals at the point of entry.


The primary objective was to determine the degree to which referrals were screened and examine the consistency of screeners for patient referrals in a typical older adult community mental health team in Central London. By determining the extent of screening at the initial point of entry into the service, we aimed to evaluate clinical practice against the gold standards of care. Although good practice should be commended, our aim was to identify areas of poor practice and evaluate the reasons for this. In highlighting any particular areas where performance was not adequate, we hoped to generate and implement strategies to improve consistency and clinical practice.


An audit of referrals entering the South Kensington & Chelsea Community Mental Health Team (CMHT) over a two month period. Referrals were evaluated by comparing the actions of screening methods undertaken with a set of gold standards. Gold standards were generated from the Central & North West London Mental Health Trust CPA policy on referrals, local CMHT documentation “Roles and Responsibilities of Duty Referrals” and further discussion with team managers and consultant clinical leads.

Referrals were evaluated on the presence or absence of specific variables identified in the Gold Standard by manually scrutinising documentation within the case files. In particular, focusing on the original referral letter/form itself and initial notes made prior to the date of the CMHT initial assessment. Using this data gathering method, there were no missing data characteristics.


Overall, twenty-six files were evaluated for patients referred into the service from April to May 2010. Files were randomly chosen and there were nineteen female and seven male cases assessed. Interestingly, ten of these cases had been closed following one assessment. On the whole, the results were mixed, with good practice demonstrated in some aspects and scope for improvement in other areas.

The first question referred to the detail of the original referral. Reassuringly, each referral appeared to contain sufficient information about the referrer and the patient (88.9 per cent vs. 11.1 per cent). However other aspects of the referral contained insufficient detail, not enough information was available on risk issues (48.1 per cent vs. 51.9 per cent), or sufficient detail to make a decision at that stage (40.7 per cent vs. 59.3 per cent). Many GP referrals did not contain recent pertinent haematological investigations (14.8 per cent vs. 51.9 per cent) with 33.3 per cent of referrals coming from non-GP sources. Additionally, for those capable, few patients had given consent or were aware a referral had been made to the mental health team (25.9 per cent vs. 33.3 per cent).

The second question referred to the initial steps taken by the screener to obtain further information from individuals. For GP referrals, contact was made only rarely (14.8 per cent vs. 74.1 per cent) and only 11.1 per cent of the time was the contact unnecessary. For non-GP referrals, the referrer was only contacted in some instances (29.6 per cent vs. 33.3 per cent) and with 37.0 per cent of the time this being unnecessary. Similarly, relatives and carers were only contacted appropriately in some instances (25.9 per cent vs. 51.9 per cent) with 22.2 per cent unnecessary.

The next question referred to the nature and clarity of the referral post-screening. Encouragingly, following screening, many of the referrals clearly identified the nature/reason for the referral (74.1 per cent vs. 25.9 per cent). Unfortunately, fewer identified the urgency of the referral (55.6 per cent vs. 44.4 per cent), or whether the patient was known to mental health or other services (63.0 per cent vs. 29.6 per cent) with 7.4 per cent non-applicable.

The fourth question referred to the next steps taken by the screener to obtain information from records. Reassuringly, 100 per cent of the patients referred were checked against the trust’s electronic database. Of the patients known to the service, in the majority of cases the medical records were sought (81.5 per cent vs. 7.4 per cent) with 11.1 per cent non-applicable. Although access was available, fewer patients were checked against the social services database (22.2 per cent vs. 37.0 per cent) with 40.7 per cent non-applicable.

The final question referred to the clarity of the final action plan undertaken by the screener before taking the referral to the allocation meeting. Many of the referrals did not explicitly confirm whether the referral had been accepted or declined (42.3 per cent vs. 57.7 per cent). Similarly, many of the referrals did not outline the urgency of the assessment to be undertaken (46.2 per cent vs. 53.8 per cent). In several cases clarity about whether further information should be sought before proceeding was not indicated (46.2 per cent vs. 42.3 per cent) with 11.5 per cent non-applicable.

The majority of GP’s were asked to review patients medically before accepting the referral (19.2 per cent vs. 7.7 per cent) although it was not appropriate in 73.1 per cent of cases. When appropriate, clarity on an agreement for joint assessment with the referrer was not sought often enough (3.8 per cent vs. 11.5 per cent), however this was unnecessary in 84.6 per cent (see figure 1, attached right).


Overall, the results demonstrated were mixed and quite varied. As original referrals enter the service, it is clear that many do not contain sufficient detail to establish their appropriateness for mental health services. More than 50 per cent of referrals did not contain enough information about the nature of the problem or risk to an acceptable standard. GP referrals in particular were consistently lacking in recent baseline haematological test results. Many patients were also not aware that referrals had been made to the service when it was appropriate to have informed them.

This lack of detail made it harder for the team to make a decision regarding the appropriateness or urgency of a significant proportion of the referrals. There could be several explanations for this deficit. Fundamentally, a lack of knowledge and awareness of the service provided by the team is a key issue, and mental health services have been striving to improve this for some years. Understandably, if a referrer is uninformed of the particular services provided, then they will be unable to provide the necessary information required. GPs were generally better in providing the pertinent detail; however, they regularly failed to provide baseline blood tests as part of the referral. Again, lack of understanding of mental health in older adults, in this professional cohort is an important factor.

The standard referral form used by referrers within the trust may additionally contribute to the lack of detail in referrals. The poor structure and lack of content may restrict the referrer in conveying essential issues, and this should be taken into consideration. These points aside, it is clear that many patients are not being included in the decision making with strong evidence that a significant proportion are unaware of referral to the service when it was appropriate to get their consent. In a specialty where engagement and rapport is so crucial, for these unaware patients, a further barrier is possibly created which may hinder the success of a referral.

Given all the above, along with the realistic possibility that a sizeable number of referrals entering the service may not be of a sufficient standard, it is essential that those screening referrals, do so comprehensively and consistently. From this audit, the evidence suggests variable performance with good practice demonstrated in some areas and certainly room for improvement in other aspects.

Credibly, over 80 per cent of the staff who were screening, were consistently checking patients referred against hospital electronic and manual records. However, making contact with the referrer or a relative/carer was more inconsistent and during this audit a successful contact was recorded if an attempt was made regardless of the response. GP’s in particular were not being contacted about referrals when it would have been appropriate to do so. Although it would appear that staff do need to make more contact with crucial individuals during the referral screening process it is important to stress inherent difficulties. These may include inaccuracies in contact details, lack of telephone number and a poor response. It would also be prudent to acknowledge that staff screening come from a variety of professional backgrounds with differing expertise and interests within the team. There may be a tendency for staff to focus detail on their particular expertise unaware of the need for other detail on unfamiliar topics. One can only speculate on these issues and further clarity within the team is necessary.

In finalising an action plan, many screeners did not explicitly clarify important aspects of some referrals including recording an acceptance or refusal, urgency or the need for further information before taking the referral into the allocation process. This could be due to a lack of experience or guidance which may lead staff to allow a referral to ‘drift’ through to allocation. It is important to understand whether staff are being given the necessary support during this crucial process. Additionally it is common practice for staff on triage duty to juggle their routine work with their duty role. It may be the case that they are unable to give sufficient time allowing certain referrals to be overlooked.

On a positive note, staff screening were good at re-directing the referrals to general practice for medical review before acceptance, with nearly three quarters of referrals directed when appropriate. This demonstrates that understanding of potential organic issues in elderly mental health is relatively good within the multidisciplinary team but joint assessments could be considered more often at the point of screening to enhance efficiency of allocation and assessment.

In summary, the evidence from this audit clearly identifies that many of the referrals to the CMHT are of poor quality with insufficient detail. The lack of awareness of the role of mental health services is fundamental to this issue, but is only one of several factors. To ensure that cases are managed appropriately, the staff responsible for screening referrals play a vital role triaging and identifying needs from the core details that are provided from the referral source. In order to do this successfully, there needs to be a comprehensive and consistent approach to the screening of referrals that demonstrates clarity and method. This small audit based in a Central London CMHT demonstrates that current local practice is variable and the success of referrals as a result is erratic. Good practice in some areas is clouded by lack of attention to other aspects. The variable approach has inevitably led to inefficiencies within the screening and allocation process which could be enhanced. It may help to provide a more robust structure whereby the team and its staff have a clear identity of their role and sufficient awareness of clinical need within any given referral, allowing the team to focus on patients in genuine need.


In light of the outcomes described above one can look at positive change in two ways either externally or internally. Externally, it is clear that greater awareness and education of mental health in society is vital to the success of identifying and managing those suffering from mental illness. The open access policy adopted by many mental health trusts is a relatively sensible approach in reducing barriers to access. However, its success is pivotal upon the diligence and aptitude of a well informed community. This is an area of huge focus for the Department of Health with some positive progress in recent years, although much more can be done in the future.

Additionally, it would appear that improvements could be made within mental health services to ensure they are realising the benefits of this widely adopted informal referral process. Further attention and consideration should be focussed on screening referrals which is a fundamental aspect of the service delivery and is too often overlooked. By guiding and educating individuals within the team, they are likely to feel more supported and valued in the critical role they undertake.

Specifically, it would be crucial for the service, to capitalise on the good work occurring already and improving in areas where standards are not being met. The benefit of intervention in this manner will ultimately enhance team practice and service provision, providing clarity and consistency to an important aspect of clinical care. This will reinforce and build on existing staff skills, competency and team efficiency critically leading to an enhanced patient experience whereby mental health teams can focus on those with the greatest need.

We would recommend:

Further training and increased awareness of all members of staff within the team regarding the correct methods of screening a referral based on our agreed gold standards. This will ensure that staff undertaking the screening are aware of their responsibility in this important role.

Implementation of a referral screening tool to aid staff ensuring the clarity of documentation and plan. Introducing a concise form for screening referrals may aid triage, supporting staff in following a robust method in practice.

Re-audit of referrals two months after the introduction of change to assess the success or failure of such interventions within the team. Evaluating the importance of the concepts discussed within this report and ensuring completion of the audit cycle.


  • Feedback of audit report to the Community Mental Health Team and wider audiences both locally and nationally.
  • Suggest implementation of recommendations locally monitoring for a change in practice.


Das, S., & Bouman, W. (2008). Direct referrals from social services to community teams for older adults with mental illness. Psychiatric Bulletin, 32, 164-165

Linden, M., Gothe, H., Ormel, J. (2003). Pathways to care and psychological problems of general practice patients in a “gate-keeper” and an “open-access” health care system. A comparison of Germany and the Netherlands. Soc Psychiatry Epidemiol; 38, 690-697

Macdonald, A., Goddard, C., & Poynton, A., (1994). Impact of ‘open access’ to specialist services – the case of community psychogeriatrics. Int J Ger Psychiatry; 9(9), 709-714

Marriott, S., et al: (1993). The consequences of an open access referral system to a community mental health service. Acta Psychiatrica Scandinavica; 88(2), 93-97

Pomerantz, A., et al. (2008) Improving efficiency and access to mental health care: combining integrated care and advanced access. Gen Hosp Psychiatry; 30(6), 546-551