Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Streamlining a specialist A&E service to enhance patient care

  • Comment

(2008) Streamlining a specialist A&E service to enhance patient care. This is an extended version of the article published in Nursing Times; 104: 26, 29-30.
This article describes an initiative to improve services in an ophthalmic A&E department. The multidisciplinary team streamlined the service by introducing a formal triage tool suitable for a unit open during normal working hours, an informal appointments system and ophthalmic nurse practitioners. An internal audit has shown that patients experience appropriate, effective and efficient care. In addition, staff are now using an evidence-based tool with manageable workloads, and four-hour waiting targets are consistently met.


Ring, L., Linnell, A.

Lynn Ring, BSc, RN1, RN5
, is clinical nurse specialist; Anne Linnell, MBBS, FRCSEd, FRSOPHTH, is consultant ophthalmologist; both at the Roy Harfitt Eye Unit, Sutton Hospital, Surrey.


The Epsom and St. Helier University Hospitals NHS Trust multidisciplinary ophthalmic team began to review the delivery of its ophthalmic A&E service in late 2005. This service is considered to be a type II A&E service defined as a ‘consultant-led single specialty with designated accommodation for reception of patients’ (Department of Health, 2003). The trust crosses the boundaries of both an outer-London PCT and a provincial PCT with a population of 2.9million people.

The ophthalmic A&E service provides care for the local community from the ‘hub site’ outpatient department. The overall ophthalmic service is provided from several ‘spoke site’ outpatient departments and two surgical bases, one with inpatient facilities. The hub site is situated in a peripheral hospital without an adjacent general A&E department.

In 2005, a traditional walk-in A&E service was provided at the hub site, whereby all patients (regardless of priority or appropriateness) were treated and/or referred to the ophthalmic outpatient department for further investigation and treatment. This often led to a perception of inappropriate referrals, packed waiting rooms, disgruntled patients and, at times, doctors being overwhelmed by sheer numbers. The team decided to formalise triage, to support the service with a triage tool and to review the overall A&E experience for patients.

Literature review

We carried out a literature search using Cinahl and Medline, with ‘ophthalmic triage’, ‘triage tools’ and ‘nurse practitioner’ as keywords. The search was limited to the last 10 years to ensure all information was up to date.

Ezra et al (2005) discussed the accuracy of diagnosis and referrals from general A&E senior house officers (SHOs) and emergency nurse practitioners (ENPs). Overall nurse practitioners’ referrals appeared to be more accurate than those of SHOs. Many articles discussed the use of telephone triage in ophthalmic emergency departments (for example, Marsden, 2000; Foat, 1999; Marsden, 1998), which would appear to be an effective way of managing patient flow into the department.

A constraint of existing tools was that our service is limited to working hours only, with out-of-hours ophthalmic care provided via a referral-only service based at a large central teaching hospital. We also carried out a manual search of journals and books due to the limited literature found.

Stollery et al (2005) established prioritisation of ophthalmic emergency care based on conditions, that is, urgent care for conditions such as central retinal vein thrombosis, acute glaucoma and chemical burn. However, Marsden (2006) discussed triage and stated that the triage consultation is not long enough to make a diagnosis. In addition, clinical conditions are not always an indication of priority, as they do not include subjective problems such as pain assessment. Marsden (2006) highlighted the use of a reductive method of triage – that is, practitioners should assume the worst, then ask questions to prove or disprove the theory using the Manchester triage tool (Mackway-Jones, 2005).

However, the Manchester triage tool did not allow for our service’s limitations. For example, if a patient arrived just before the department closed with a red eye and no reduced vision, it was not clear whether the nurse should refer the patient to the out-of-hours service or whether, clinically, the patient could wait until the following morning. No suitable triage tools have been published to support the implementation of appropriate triage within the confines of this service. Therefore, the team needed to develop a local triage tool that was appropriate for the unit’s opening hours.

Developing the tool

This part of the process involved many meetings of ophthalmic consultants, ophthalmic training-grade doctors and ophthalmic nurses. A complete review of current documentation, skill mix and informal triage processes took place.

A triage tool based on the presenting conditions of each patient was developed. However, the team realised that nurses carrying out triage would have to be extremely knowledgeable to make clinical diagnoses during the triage process. The service had found experienced ophthalmic nurses were in short supply during recruitment drives, so there was reluctance to develop a tool with inherent limitations. A two-week pilot found the triage process became too lengthy, often delaying treatment rather than prioritising it, as identified in Marsden (2006).

Further refinement was carried out and, finally, a symptom-based triage tool was implemented. Both nursing and medical staff suggested the categories and symptoms. Consultants, as a group, endorsed the tool’s use.

Five categories were originally defined (Table 1) and used in the triage process, along with colour-coordinated trays in the examination room. The doctors then saw patients according to clear priorities. The team also discussed other ways to streamline attendances within the unit, alongside the introduction of a formal triage tool.

Table 1. Symptom based triage tool for adults and children

Within 2 hoursWithin 24 hoursWithin 24–48 hoursClinic referralNurse-led next available
  • Major trauma

  • Chemical

  • Cut to eye

  • Major trauma

  • Cut to lid

  • No trauma
  • Minor trauma
  • Sudden loss of vision less than 6 hours
  • Sudden loss of vision more than 12 hours

  • New flashes

  • Curtain-like loss of vision

  • Sudden loss of vision

  • more than 24 hours but less than 4 weeks

  • Distortion

  • Haloes

  • Floaters/increased floaters

  • Gradual loss of vision more than 4 weeks

  • No sudden loss of vision

  • No sudden loss of vision
  • Painful eye

  • Worst ever

  • Painful eye

  • Keeping awake at night

  • Painful eye but tolerable

  • Red eye

  • No eye pain
  • Gritty

  • FB sensation

  • Red eye, painless

  • Headache

  • Worst ever

  • Sudden onset

  • Headache

  • Tender temples

  • Photophobia
  • Discharge
  • Droopy eyelid/ptosis, new
  • Double vision, new
  • Child under 12
  • Child 12–16
  • Any problem/complaint by contact lens wearer
  • Broken contact lens

Optometrist reference:

  • Hypopyon

  • Hyphaema

  • IOP >40mmHg


Optometrist reference:

  • Abnormal pupil and visual disturbance

  • IOP 39–30mmHg

Optometrist reference:

  • IOP <30mmHg

  • Abnormal pupil - symptomless


Appointment system

Royal Bournemouth and Christchurch Hospitals NHS Foundation Trust presented an audit on the introduction of an appointments system, including nurse-led clinics in ophthalmic A&E, at a national conference in 2006. Our team discussed how this experience could be adopted to improve patient satisfaction, reduce waiting times and manage the flow throughout the department. We arranged a visit to the eye unit at this other trust for both medical and nursing staff. The multidisciplinary group found it invaluable to see how change could be implemented. Four action points were agreed:

  • All new attendees were to be informed about the new system and to undergo formal triage into the categories shown in Table 1;

  • A&E follow-ups were to be reduced and patients requiring review were to be directed to a dedicated primary care clinic;

  • Nurse-led sessions were to be introduced within existing resources, as a trial, using existing ophthalmic nurses’ skills to prove effectiveness and to support a business case for nurse practitioners;

  • Patients were to be encouraged to telephone the department before arriving and the concept of telephone triage was to be introduced alongside the use of a formal tool.

Various activities had to take place to aid implementation. Documentation needed to be modified to include triage categories. We also had to arrange training sessions for the tool’s use and installation of a dedicated telephone line as first point of contact. New clinic templates needed to be agreed and set up. An adapted competency assessment document was used to support the development of the ‘see and treat’ method. The team recognised that having to completely design and approve a competency programme would be a lengthy task. As a result, the author and organisation of the original competency programme (Agnes Lee, lecturer practitioner, Manchester Royal Eye Hospital) were acknowledged within the adapted document to recognise their contribution and support.


Development of new documentation and training in the triage tool’s use took place during the six months before implementation. At the same time, ophthalmic nurses were trained to undertake a ‘see and treat’ method of dealing with minor injuries. Our trust’s senior nurses approved the new roles for ophthalmic nurses including the training programme, and lines of governance were set up. A medical supervisor was established for advice, guidance and support if required.

The triage tool and new documentation were introduced in January 2007. The service’s manager set up new primary care clinics so that A&E follow-up patients were redirected into a new primary care slot. The review of patients in A&E was changed to a dedicated clinic, rather than recorded as A&E attendance. This improved data accuracy for the purposes of Payment by Results, which expanded to include emergency care in April 2006. In hindsight, the department should have indicated its intention to change this practice so that discussions could take place with the local PCT regarding the increase in outpatient activity.

Patients were given the number of the dedicated telephone line and were encouraged to call the unit for advice before attending. This informal approach to telephone triage was shared with local general A&E units. Nevertheless, a formal proposal on changing to a referral-only service has not been made to the PCTs as it is felt that patients should still be able to self-refer, if appropriate, without any delay in either main A&E units or GP surgeries.

An audit was carried out six months after implementation – 50 triage decisions and 50 patients were examined and overall 92% accuracy in triage was achieved. Inaccuracies were checked and it was found that two patients were seen immediately due to inactivity, but two patients were classed in a higher category than diagnosis warranted.

Patients were asked whether the new informal appointments system improved the service provided. Overall 80% were in favour but 20% felt it did not improve the service and would have preferred to wait and be seen, regardless of waiting time. While patients were happy to wait even if this breached the government’s four-hour A&E waiting time target (DH, 2000), the trust did not want to breach this. As a result we have persisted in our new approach. We used posters in the waiting area to explain to patients that they would be seen in order of clinical priority rather than arrival time. Since the new system started, the department has seen a reduction in four-hour waiting time breaches and, currently, 98–100% of patients classed as type II A&E are seen within the four-hour timeframe. Clinical emergencies such as acute glaucoma, which may need more than four hours’ treatment, are the only exception.

Following an internal review of the audit, questions were asked about staff perception of the change and whether patients sought help from other health professionals before coming to the unit. In line with this, a further audit was carried out in December 2007.

The team also reviewed the categories, as a 48-hour limit did not lend itself to adequate service provision. Friday afternoon was a particular problem for patients with injuries that, according to the 24–48-hour (green) category, could wait. However, Monday morning was beyond the 48-hour limit according to our triage tool.

Other issues had been highlighted over the year, including difficulties encountered due to the position of the ophthalmic A&E, which is away from the main hospital site and general A&E. Consequently, the team reviewed and further refined the triage tool (Table 2).

Table 2. Refined symptom-based triage tool

<4/24 (same session)
<12/24 (same day)
<72/24 (within 3/7)
Nurse-led, next availableNot appropriate/needs clinic referral
  • Any child <8 years or symptoms <24 hours
  • 8–16 years or symptoms <48 hours
  • 12–16 years or symptoms <1 month
  • Any child >1 month dependent on symptoms
  • Chemical injury (alkaline)

  • Laceration to eye

  • Velocity

  • FB

  • Laceration to lid

  • Blunt force trauma

  • Minor injury

  • Scratch

  • Poke

  • FB

  • Sudden complete loss of vision <4 hours
  • Sudden loss of vision <12 hours

  • New flashes of light

  • Portion or curtain-like loss of vision

  • Sudden loss of vision >12 hours but <1 week

  • Visual distortion <1 week

  • NB

  • Detailed/central

  • Haloes

  • Increased floaters

  • Double vision (new/sudden or worse)

  • Sudden change to vision <4 weeks

  • Single floater with no flashing lights

  • Blurred

  • Watery

  • Gradual loss of vision >4 weeks

  • No sudden changes in vision

Eye pain:

1 (none)–10 (worst ever)

  • 8–10 scoring

  • No relief from oral analgesia

  • 6–8 scoring

  • No relief: analgesia

  • Keeping awake at night

  • Waking during night

  • 4–8 scoring

  • Relieved by analgesia

  • Photophobia

  • <4 scoring

  • Gritty

  • FB sensation

  • º pain


1 (none)–10 (worst ever)

  • 10+ (worst ever)

  • Sudden onset

  • Painful scalp

  • Brow pain

  • Painful temples

  • Tender temples
  • New droopy eyelid/ptosis

  • Acute swollen lids (vision distorted)

  • Swollen lids (vision normal)
  • Puffy lids

  • Redness

  • Epiphora

  • Discharge

Cornea/ conjunctiva
  • Cloudy

  • Red +++

  • Hazy

  • Red ++

  • Clear cornea

  • Red around limbus

  • Red mild to moderate

  • Contact lens lost/damaged

Optom referral 
  • Hypopyon

  • Hyphaema

  • IOP >40mmHg

  • Papilloedema

  • Abnormal pupil with visual symptoms

  • IOP 39–30mmHg

  • IOP 39–30mmHg
  • Unequal pupil size

  • No ptosis

  • No visual loss

  • IOP<29mmHg

  • Acutely ‘unwell’ adult
  • Very hot adult ‘feverish’
  • Audit results

Out of 100 patients, 87 responded to a questionnaire distributed on arrival. The majority experienced their eye problem for less than one week, although 12 registered complaints that lasted for more than four weeks. Most attempted to seek advice from their GP or optometrist before attending the department, with three using NHS Direct. The majority came to the unit on the advice of their GP or optometrist, but three arrived following advice from GP receptionists.

It was surprising that less than 50% of patients knew the unit was only open during working hours, considering it had been established for 10 years. They also failed to realise that the A&E unit was delivering care based on clinical need, in spite of posters and advice from other health professionals. Anecdotally, it was often referred to as the ‘little walk-in clinic’. However, those who had attended the unit before, knew relatives or friends who had attended before, or had been referred from the main A&E department, were aware of clinical priorities.

Patients’ comments were positive, such as: ‘I have no knowledge of the previous systems but have found this good’ and ‘Very good system, saves queuing for long periods not knowing how long’. However, some commented that triage seemed to be a long process, as a second nurse completes a history-taking interview following quick triage.

The triage accuracy was consistent with previous audit data using inter-reliability and criterion validity as described by Twomey et al (2007), and 92 forms were completed in a randomly chosen week. Some 91% (84) were accurate. Further exploration showed clinical examination and more detailed history-taking would have moved patients from the same-session triage category to same-day.

Staff views

The medical and nursing staff were asked their opinions about the new system. Five doctors usually involved in the A&E service (two consultants, one associate specialist and two senior house officers) responded. Two had previous experience using a traffic-light system and felt patients were seen appropriately some of the time, with three stating most of the time. Four said the workload was more manageable most of the time and that the system’s introduction had reduced work-related stress. All the doctors felt the system had improved services for patients, staff and the trust.

Comments made by doctors – such as, ‘All have benefited, even when staffing levels [are] low’, ‘Less stressful, very well organised, patients and GPs are more aware of eye conditions’ and ‘Patients often say they like the A&E service and they think it is efficient’ – were very encouraging.

Nurses were positive about the change. Nine who had over one year’s experience in ophthalmology (not necessarily in A&E) responded and included both registered and unregistered nurses. All felt the tool was fairly easy or very easy to use and all had used it in clinical practice. They had all used the tool during a telephone consultation and eight had used it to explain to patients why they were not suitable for A&E, which helped reduce potential confrontations. Nurses also felt the system had improved services for patients, with comments such as: ‘It’s a tool providing a uniform service and patients see equality and understand clinical priority’, ‘The system still allows for urgent cases to come in immediately but has the facility to reassure patients that they will be seen appropriately’, ‘Number of patients is adequate so we can manage. No more moaning/complaints’, ‘Helps staff improve their knowledge’ and ‘Get time to be with patient now’. The team was particularly pleased with this last comment. It seems that patients have now become the centre of care and nurses are able to give attention where it is needed.

Reflecting on developments

The audit has shown that further public relations work needs to be carried out with GPs and local optometrists to ensure accurate information is given to all patients about services provided in the unit. The following action points have been highlighted:

  • Promoting the direct-line advice service;

  • Dissemination of the triage tool to other healthcare providers, both locally and across the sector.

The nurse-led service was more difficult to introduce. Currently one nurse has completed the necessary training as part of the nurse specialist role, achieving all the compulsory competencies. The supply of medication for patients with minor injuries was not fully explored initially, and the trust required additional supporting guidance. ENP treatment protocols have since been developed to support the role. The nurse specialist has undergone training as an independent prescriber and achieved registration with the NMC but to facilitate the supply of medications by nurses without independent prescribing rights, patient group directions have been designed. These were approved by the trust.

The current environment has also hampered the full implementation of a nurse-led service. Space limitations within the department meant that nurse-led clinics could not run concurrently with medical clinics and appropriate patients could not be triaged to be seen by appropriate health professionals. Nevertheless, the nurse specialist’s training and expertise has enabled two nurse-led sessions per week. This practitioner sees all patients, regardless of presenting symptoms. The nurse specialist has access to a named consultant for advice and support for conditions outside of their professional competence.

Since the introduction of our triage tool, researchers have published articles discussing triage and ophthalmic A&E departments. Rossi et al (2007) carried out a two-phase approach to triage in an emergency room using a specific scoring system. Phase one analysed over 160,000 electronic patient records for symptom-based problems, such as redness, pain and loss of vision, and allocated an appropriate score (white, green or yellow, with yellow being the highest score). Phase two reviewed 1,000 patients and found a high correlation between Rome Eye System for Scoring Urgency and Emergency (RESCUE) codes assigned prospectively and the post-diagnosis coding. The authors felt this system may have value in the wider ophthalmic community but required further validation in multiple centres to be a truly robust, effective tool. It is based on care delivery outside the NHS and may not be transferable to a typical stand-alone ophthalmic unit.

Implications for practice

The audit has shown that experienced nurses can use a triage tool effectively and that the new informal system of appointments provides appropriate care for patients in a timely and efficient manner. Capacity and demand still fluctuate, with the busiest times on Monday and Friday. It is thought that ensuring patients telephone the department before arrival should guarantee a match between capacity and demand, thus reducing the ‘peak and trough’ effect.

To be truly useful, however, the triage tool must also be effective when used by an inexperienced ophthalmic nurse or a nurse in a non-ophthalmic setting such as a GP practice or general A&E department. A research proposal is being developed to investigate this hypothesis.

Formal discussions with local PCTs must take place to reach agreement about changes to provision of urgent ophthalmic care. The team feels the audit shows a positive response to the informal appointments system and that a move to a formal appointment-only service now needs to take place. Safeguards must be in place, however, to continue to support patient self-referral. In addition, most organisations providing an ophthalmic A&E service need to support the expansion of the nurse practitioner role.


These changes to the ophthalmic A&E service have streamlined it and improved standards for both patients and staff, as shown by the audit results. Effectiveness and efficiency have increased, benefiting the trust and achieving government targets. Overall, the changes have taken place as a result of joint working by the multidisciplinary team aimed at achieving improvements.


Department of Health (2003) SITREPS 2003–2004 Final Version. Definitions and Guidance.

Department of Health (2000) The NHS Plan: A Plan for Investment, A Plan for Reform.

Ezra, D. et al (2005) Reliability of ophthalmic accident and emergency referrals: a new role for emergency nurse practitioner? Emergency Medicine Journal; 22: 696–699.

Foat, P. (1999) Setting up a direct line referral system: eye triage by telephone. Ophthalmic Nursing; 3: 3, 8–12.

Marsden, J. (ed) (2006) The care of patients presenting with acute problems. In: Ophthalmic Care. Chichester: Wiley Publishers.

Marsden, J. (2000) An evaluation of the safety and effectiveness of telephone triage as a method of patient prioritisation in an ophthalmic accident and emergency service. Journal of Advanced Nursing; 31: 2, 401–409.

Marsden, J. (1998) Decision-making in A&E by expert nurses. Nursing Times; 94: 41, 62–65.

Mackway-Jones, K. (ed) (2005) Emergency Triage. London: BMJ Books.

Rossi, T. et al (2007) Triaging and coding ophthalmic emergency: the Rome Eye Scoring System for Urgency and Emergency (RESCUE): a pilot study of 1,000 eye-dedicated emergency room patients. European Journal of Ophthalmology; 17: 3, 413–417.

Stollery, R. et al (2005) Ophthalmic Nursing. Oxford: Blackwell Publishing.

Twomey, M. et al (2007) Limitations in validating emergency department triage scales. Emergency Medicine Journal; 24: 477–479.

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.