A Glaucoma Case Finding Pilot with the African Caribbean Community

A glaucoma case finding pilot with the African Caribbean community

Author:Shaun Leamon and Helen Lee;Publisher:RNIB;Year of publication:2015

Note on versions

The results of this study have been published as a peer-reviewed article,'A Mixed-Methods Evaluation of a Community-Based Glaucoma Check Service in Hackney, London, UK', in Ophthalmic Epidemiology. Slightly revised figures for the results of the glaucoma case finding pilot are available in the peer reviewed article.

Key findings from the research

Approximately 1 in 6 people invited to attend the free glaucoma check based in a local GP surgery attended the appointment.
The service was universally well received, with 96.7% of respondents rating the experience as positive.
Overall, 2.6% of patients who underwent the free glaucoma check were either diagnosed as having glaucoma or identified as glaucoma suspects.
The cost per patient identified was £9,013

Background

Glaucoma is a chronic eye condition that, if left untreated, can lead to blindness. Each year, more than 3,000 people over the age of 40 are certified blind or partially sighted because of glaucoma; making it the second most common cause of certifiable sight loss in the UK (1).

The early detection of glaucoma is important in order to initiate appropriate monitoring and treatment and to minimize the risk of irreversible visual field loss. At present, the detection of glaucoma is based on opportunistic identification, with optometrists responsible for nearly all referrals to hospital eye service for suspected cases (2,3) .

Population screening for glaucoma is currently not recommended by NICE (4) but research suggests it may be a cost effective strategy for high risk groups such as people of Black Caribbean and Black African descent (5,6). There are varying opinions regarding how such a service could be configured: what checks should be used, who should deliver the checks, where should the checks take place and who should be the target group? (7).

People of Black African and Caribbean descent are up to eight times more likely to develop open angle glaucoma and 10 to 15 years earlier than people of other ethnicities. In the UK they are also more likely to present to eye care services when the eye condition is more advanced (8).

This research briefing describes a pilot glaucoma case finding service based in the London borough of Hackney. The project was a collaborative initiative between RNIB, City and Hackney Primary Care Trust, GP practices in Hackney, London School of Hygiene and Tropical Medicine, Moorfields Eye Hospital, the Local Optical Committee and the community in Hackney. The project sought to address some of the knowledge gaps in relation to glaucoma screening in the UK, providing evidence and insight for a future large-scale screening trail.

Methods

The aim of the project was to develop, promote and measure the up-take of a pilot glaucoma check service, based in general practice. The target population for the pilot service was people of Black Caribbean and Black African descent, aged between 40 and 65 years. The location of the pilot study was the London borough of Hackney.

Four GP practices took part in the pilot, one of which also ‘hosted’ the service and was responsible for appointment booking, processing referrals and notifying patients’ originating practice of appointment outcomes. The other three practices (practice A, B and C) were all situated within a 1.5 miles radius of the ‘host’ practice. To be eligible for the service, patients had to be registered with one of the four practices taking part in the pilot. The service ran for six months, from October 2012 to March 2013.

Eligible patients were sent a letter and information leaflet by their GP practice, inviting them to make an appointment to attend for a glaucoma check at the ‘host’ practice. Appointments were made via a choose-and-book system, based on availability. These patients formed the ‘standard’ intervention. In two of the practices (the ‘host’ practice and practice A), a further ‘reminder’ letter was sent to those who did not respond to the initial invitation. Patients at practices B and C did not receive any further reminder to attend the check.

A random half of the patients from the ‘host’ practice who did not respond to the initial invitation and reminder letter were allocated to receive an ‘enhanced’ intervention which consisted of a further invitation by phone to attend a check.

The glaucoma check was performed by a sessional Optometrist based in the ‘host’ practice. Several sessions a week were offered in order to provide a mix of appointment times (including Saturdays). The check consisted of:

standard automated perimetry using a suprathreshold test;
measurement of intraocular pressure using Goldmann applanation tonometry;
Optic nerve head imaging, assessing the clinical features.

Patients identified as potential glaucoma cases were referred to secondary care for ongoing investigation. Clinical protocols were developed for governance and quality assurance purposes. Regular audits of care were conducted by an RNIB optometrist and the optometric advisor from Hackney and City Primary Care Trust.

The pilot was evaluated by London School of Hygiene and Tropical Medicine. The evaluation comprised an outcome, process and economic evaluation of the service. Outcome measures included uptake and did not attend (DNA) rates for the service; referral rates to secondary care; the rate of attrition between referral and attendance at secondary care; the false positive rate at secondary care; and the number of patients diagnosed with glaucoma and stage of disease at presentation. A patient survey was also administered to ascertain people’s reasons for attending the check and satisfaction with the service. The process evaluation comprised qualitative interviews with project staff directly involved in developing and operationalising the programme: City & Hackney PCT staff, the sessional optometrists, GP surgery staff and hospital staff. The economic evaluation took the form of a cost consequence analysis.

The evaluation was granted ethical approval from Bromley REC (reference number 11/LO/1264), and local R&D approval from NHS East London & City (reference number RD/022).

Findings

3.1 Summary of the findings

A total of 3,041 patients were invited to attend the free glaucoma check. Overall:

581 patients (19.1% of those invited) made an appointment for a check.
459 attended (15.1% of those invited) and 122 did not attend (DNA) their appointment (21% of those who made an appointment).
Attendance at the service was higher among women (17.1%) than men (13.3%).
Attendance was also higher among patients registered at the ‘host’ practice.
More than 1 in 10 (11.4%) people attending the free glaucoma check had never had an eye examination.
Amongst people reporting a family history of glaucoma, fewer than half (48.9%) had been for an eye test in the last year; 1 in 6 (13.3%) had never had an eye examination.
A key motivation to attend the check was the information leaflet supplied with the invitation letter, which featured a local community member encouraging people to attend the service.
The service was universally well received, with 96.7% of people rating the experience as positive. The vast majority of people (89.0%) also found the location of the service convenient.
Of the 459 people who attended the check, 29 patients (6.3%) were referred to the hospital glaucoma clinic.
Overall, 2.6% (12/459) of patients who underwent the glaucoma check were either diagnosed as having glaucoma, or identified as glaucoma suspects.
The cost per patient identified was £9,013. This includes all start up costs for the study.

3.2 Uptake of the offer of a free glaucoma check

Table 1 shows the proportion of people booking an appointment for the free glaucoma check, according to patient practice and invitation type.

The proportion of people booking an appointment was higher among patients registered at the ‘host’ practice and also higher among patients who received a reminder letter.

At the ‘host’ practice, 15.5% of eligible patients (71/457) had booked an appointment after receipt of the initial invitation letter, compared with 14% (228/1630) at practice C.

After receipt of the invitation and reminder letter, the proportion of patients at the ‘host’ practice who had booked an appointment was 31.3% (143/457), compared with 17.9% (171/954) at practice A.
27.9% (39/140) of people in the enhanced intervention group booked an appointment. Although the booking rate among those actually contactable by phone was much higher at 50.6% (39/77).

Table 1. Proportion of people booking an appointment for the free glaucoma check, according to patient practice and invitation type

Practice / Practice list size / Booked an appointment (% (n/N))
Letter / Letter + reminder / Letter + reminder + phone call
Host / 457 / 15.5
(71/457) / 31.3
(143/457) / 27.9
(39/140)†
A / 954 / * / 17.9
(171/954) / n/a
B & C / 1630 / 14.0
(228/1630) / n/a / n/a

†of the 140 people approached by telephone, only 77 were actually contactable, giving a booking rate among those contacted of 50.6% (39/77). *it was not possible to separate the up-take and DNA among patients who booked an appointment following the initial invitation letter and those that did so following the reminder in practice A, as the respective letters were not sent out in single batches given the size of the cohort in the practice.

3.3 Attendance and did not attend rates

Attendance and DNA rates according to patient practice are shown in Table 2, and by the type of invitation in Table 3.

Overall attendance was markedly higher among patients registered at the ‘host’ practice (39.8%) compared with Practice B (17.9%) and Practices C & D (14.0%) the other three practices. The proportion of patients attending an appointment was also higher in the ‘host’ practice for each intervention time-point.
The DNA rate varied by practice. Overall the DNA rate at the ‘host’ practice was 17%, at Practice B it was 18.1% and at Practices C and D it was 26.3%.

Table 2. Overall attendance and did not attend (DNA) rates at the free glaucoma check, according to patient practice

Practice list size / Total up-take
(% (n/N)) / Total
DNA rate
(% (n/N))
Host Practice / 457 / 39.8
(182/457) / 17.0
(31/182)
Practice A / 954 / 17.9
(172/954) / 18.1
(31/171)
Practice B & C / 1630 / 14.0
(228/1630) / 26.3
(60/228)

Table 3 shows the DNA rate at the ‘host’ practice according to different intervention types. The proportion of patients who DNA increased with increasing investment in terms of patient engagement: DNA rate was 11.3% amongst people booking an appointment after the first invitation letter; 19.4% amongst people booking after receipt of the reminder letter; and 23.1% amongst people who booked following the phone call (Table 3).

Table 3. Did not attend (DNA) rates at the ‘host’ practice, according to intervention type

Intervention / Invitation letter only
(% (n/N)) / Invitation letter & reminder letter
(% (n/N)) / Invitation letter & reminder letter & phone call
(% (n/N))
DNA / 11.3 (8/71) / 19.4 (14/72) / 23.1 (9/39)

3.4 Referrals to secondary care

Thirty-one patients were referred to secondary care; two of these were referred to the cataract clinic and the remaining 29 to the glaucoma clinic. Patient data were returned for 28 of the 29 patients.

Of the 28 patients referred to the glaucoma clinic, 22 attended their appointment; 13 were women and 9 were men.
Twelve of the 22 patients referred to the clinic were either diagnosed as having glaucoma (n=4) or identified as glaucoma suspects (n=8).
Despite fewer men than women being referred to the clinic, more men were diagnosed with glaucoma (1.6% vs 0.4%) or identified as glaucoma suspects (3.2% vs 2.2%; ns).
Overall, 2.6% (12/459) of patients who underwent the free glaucoma check were either diagnosed as having glaucoma or identified as glaucoma suspects.

3.5 Follow-up survey with service users

3.5.1 Participant characteristics

243 patients completed a post glaucoma check questionnaire, giving a response rate of 52.9%. More women (57.9%) than men (42.1%) completed the questionnaire. The mean age of respondents was 49.9 years.
A high proportion (71.1%) of respondents (n = 240) reported that they wore spectacles or contact lenses, with significantly more women than men (116 vs 56) wearing spectacles or contact lenses (p < 0.001).
Nearly one in five people (18.9%) reported a family history of glaucoma (n = 233), and more women than men reported a family history of glaucoma; a difference that approached significance (23.1% vs 13.1%; p = 0.090).

3.5.2 Experience of eye care services reasons for attendance

More than 1 in 10 (11.4%) respondents had never had an eye examination (n = 234), with significantly more men than women having never engaged with optometry services (16.0% vs 7.5%; p = 0.020).
Amongst people reporting a family history of glaucoma, fewer than half (48.9%) had been for an eye test in the last year; 1 in 6 (13.3%) had never had an eye examination.
The most commonly reported motivation for attendance was the information leaflet sent with the invitation letter (61.2%). Wanting to find out if they had glaucoma (51.8%) and wanting to look after eyesight now and in the future (51.8%) were also popular motivations (n = 224).
Encouragement by family and friends was a stronger motivation to attend in man than women (17.6% vs. 5.6%; p = 0.003), as were concerns about being at risk of glaucoma (25.0% vs 12.9%; p = 0.020).

3.5.3 Acceptance with the service

The service was universally well received, with 96.7% of respondents rating the experience as positive (n = 209).
The system of chose-and-book was also well received, with 90.0% of respondents stating that booking an appointment was easy and 78.1% stating that the appointment times were convenient (n = 219).
The vast majority of people (89.0%) said the location of the service convenient (n = 218) and the majority either walked to the clinic (51.8%) or travelled by public transport (30.8%) (n = 224). For most people (57.5%) it took less than 15 minutes to reach the clinic (n = 237).
Approximately 1 in 7 (13.8%) people had to take time off work to attend the glaucoma check (n = 232), although this did not appear to affect people’s positive rating of the service.

3.6 Cost of the service

The total cost of the project was estimated to be £108,161. This figure includes the set-up and management costs for the project. Examining the costs more closely reveals:

Staff costs accounted for 62% of the total costs, of which clinical staff costs were 42% of all staffing costs.
Equipment and room hire made up 22% of the total costs and consumables a further 8.4%.
A large proportion of the staffing budget (45%) was spent on project set up costs, including the development of materials.

This equates to a cost of £9,013 per patient identified at Moorfields with suspected or definite glaucoma.

An alternative staff skill mix would affect the total costs and cost per person diagnosed. Assuming the detection rate was unchanged, a 15% reduction in the cost of clinical staff would result in a cost per case identified of £8,658.

Similarly, once the programme was established, by reducing the non-clinical staff costs by 50%, the cost per person identified would reduce to £7,380.

Conclusions

The project succeeded in its aim to develop, promote and measure the up-take of a glaucoma check service based in general practice. The findings of the evaluation suggest general practice is an acceptable setting for such a service; staff were supportive of the project and satisfaction amongst patients who attended was high.

The results show that up-take of the service varied by a practice and invitation type. Overall attendance was considerably higher in patients registered at the “host” practice suggesting that there was additional “buy-in” amongst those patients. Attendance was also higher among women, though ultimately a higher proportion of men were identified as having glaucoma (or being a “glaucoma suspect”).

The findings also indicate that the pilot reached people who had not previously engaged with eye services; more than one in ten (11.1%) had never had an eye examination (with men more likely to report this was the case than women). The service also engaged people with a known family history of glaucoma who were not attending an annual eye examination.

There are, however, questions that remain, about the model itself and in particular who should administer the checks. The cost was £9,013 per patient identified, this includes initial set up costs.

Ultimately, the goal of such a glaucoma check service, were it to be rolled out, would be to prevent sight loss; a goal which was beyond the scope of this pilot. A larger trial with longer follow up exploring disease progression would be needed. This pilot provides the parameters for such a trial and gives insights into the practicalities of delivery.

Priorities for further research

Based on the findings from the study it is recommended that an expert working is convened to assess the feasibility of pursuing a large-scale trial of glaucoma testing in the light of the findings from this pilot.

The working group would comprise members of the Glaucoma Screening Platform Group, the UK National Screening Committee, the Royal College of Ophthalmologists, the College of Optometrists and other interested parties (such as the Federation of (Ophthalmic and Dispensing) Opticians, the Association for Independent Optometrists and Dispensing Opticians, and the British and Irish Orthoptic Society).

Any future study would need to provide evidence to confirm the following:

Who should deliver the service; should it be technician- or optometrist-led?
What tests should be performed; it is appropriate to be using ‘gold standard’ methods which are more diagnostic as opposed to shorter ‘screening’ methods?
Where should services be located and how should they be administered; would a centrally administered service be preferable to the ‘hub and spoke’ model administered within general practice?
How do you address issues of up-take; specifically among men?
Is screening for glaucoma in the UK cost effective, and which delivery models represent the best use of resources?

References

Public Health England. Public Health Outcomes Framework. Public Health England, 2014. (accessed 21 December 2014).
Sheldrick JH, Ng C, Austin DJ et al., 1994. An analysis of referral routes and diagnostic accuracy in cases of suspected glaucoma. Ophthalmic Epidemiol, 1(1), 31-9.
Bowling B, Chen, SDM and Salmon JF, 2005. Outcomes of referrals by community optometrists to a hospital glaucoma service. Br J Ophthalmol, 89(9), 1102-1104.
Burr JM, Mowatt G, Hernández R et al., 2007. The clinical effectiveness and cost-effectiveness of screening for open angle glaucoma: a systematic review and economic evaluation. Health Technology Assessment, 11(41).
Hernández RA, Burr JM and Vale LD, 2008. Economic evaluation of screening for open-angle glaucoma. Int J Technol Assess Health Care, 24(2), 203-11.
Campbell SE, Azuara-Blanco A, Campbell MK, et al., 2012. Developing the specifications of an Open Angle Glaucoma screening intervention in the United Kingdom: a Delphi approach. BMC Health Serv Res, 12(447), 10.1186/1472-6963-12-447
Cross V, Shah P, Bativala R and Spurgeon, 2007. ReGAE 2: glaucoma awareness and the primary eye-care service: some perceptions among African Caribbeans in Birmingham UK. Eye, 21, 912-920.

rnib.org.uk