The sight test fee: effect on ophthalmology referrals and rate of glaucoma detectionBMJ 1994; 309 doi: https://doi.org/10.1136/bmj.309.6955.634 (Published 10 September 1994) Cite this as: BMJ 1994;309:634
- Eye Hospital, Bristol BS1 2LX Department of Epidemiology and Public Health Medicine, University of Bristol, Bristol BS8 2PR Department of Ophthalmology, University of Bristol, Bristol BS1 2LX
- a Bristol
- Correspondence to: Mr Laidlaw.
- Accepted 24 June 1994
Objective: To assess changes, if any, in the numbers of referrals and outcome of glaucoma referrals to the hospital eye service since the introduction of the sight test fee on 1 April 1989.
Design: Review of referral records and clinical notes.
Setting: Referrals to the Bristol Eye Hospital.
Subjects: 51 919 patients referred to the Bristol Eye Hospital between 1984 and 1992. 9438 case notes of patients referred between 1987 and 1991 were examined in detail.
Main outcome measures: Numbers of referrals; rate of adult true positive glaucoma referrals.
Results: Referrals to the Bristol Eye Hospital were between 13.7% and 19.0% fewer than expected20after the introduction of the sight test fee. True positive glaucoma referrals were reduced by the same proportion. Conclusions - The numbers of patients being identified as requiring treatment or follow up for potentially blinding glaucoma have declined by nearly one fifth since the introduction of the sight test fee. An increased prevalence of preventable blindness may result.
The sight test is the only existing means of screening adults in Britain for treatable but potentially blinding ocular diseases such as glaucoma
Since 1 April 1989, 60% of adults have had to pay for sight tests
During 1989-92 up to 19% fewer than expected new patient referrals were received at the Bristol Eye Hospital
True positive glaucoma referrals were also reduced by 19%
The rate of identification of glaucoma has declined by nearly one fifth in Bristol since the introduction of the sight test fee
An increased prevalence of preventable blindness may be expected
The sight test is the only existing means of screening adults in Britain for sight threatening but treatable ocular diseases such as glaucoma. Over 80% of glaucoma referrals to ophthalmologists are initiated as a result of screening during routine sight tests.1 Unconditionally free sight testing was withdrawn in the United Kingdom on 1 April 1989.
Concern was expressed that the introduction of a sight test fee would deter people, particularly elderly people, from having sight tests. It was suggested that as a consequence the rate of detection of blinding but potentially treatable ocular diseases such as glaucoma would fall.2 The effect of the sight test fee on the rate of detection of eye disease has not been investigated. We therefore tested the hypothesis that there had been no change in the number of referrals to, or rate of identification of glaucoma at, the Bristol Eye Hospital since the introduction of the sight test fee.
Subjects and methods
Records of referrals to the Bristol Eye Hospital were examined for the number of referrals received each year during 1984 to 1992. Referral data from before 1984 were not available. The case records of adult referrals received between 1 July and 31 December each year during 1987 to 1991 were targeted for examination. A sample of these records was examined to determine the outcome of referrals for suspected glaucoma. Adult referrals were targeted because sight tests on children are still funded by the NHS.
There is a latency of up to six weeks between the identification of suspected disease by an optometrist and receipt at the hospital of a referral from the patient's general practitioner. To ensure that we were comparing referrals initiated after the sight test fee was introduced with those initiated before we targeted referrals received between 1 July and 31 December of each year during 1987 to 1991. The case records of patients referred with suspected glaucoma were examined to determine whether the referral was a true positive referral. Suspected glaucoma was defined as mention in the referral letter of glaucoma, abnormal intraocular pressure, optic disc changes compatible with glaucoma, family history of glaucoma, or field loss compatible with glaucoma. True positive cases were those in which the patient was noted at the first clinic visit to require follow up for confirmed or suspected glaucoma, ocular hypertension, or low tension glaucoma.
True positive glaucoma referrals represent about 6% of all new ophthalmic referrals.1 From this we calculated that a minimum sample size of 1500 referrals from each of the targeted six month periods was required to detect with 95% confidence a 10% reduction in the number of true positive glaucoma referrals being received in any of the targeted periods. A sampling protocol was devised to allow identification of up to 1800 referrals per six month period. Altogether 9438 of the 14 657 referrals received in the five six month periods were examined. An estimate of the total number of adult true positive glaucoma referrals received in each six months was obtained by multiplying the number identified from the sample by the sampling fraction.
Simple linear regression with confidence intervals and the X2 test were used for analysis.
The table gives the number of referrals received per year between 1984 and 1992 and the numbers of referrals with 95% confidence intervals predicted from linear regression of those received between 1984 and 1988. Numbers of referrals from 1989 onwards fell below the number received in 1988 and also below or very close to the lower 95% confidence intervals for the predictions. With the exception of the slight upturn in 1991 the numbers of referrals in each year were 18-19% fewer than predicted.
The 6466 referrals in 1988 exceeded the number (6123) predicted from linear regression of trends between 1984 and 1987. This higher than expected total, however, was well within the 95% confidence interval of the prediction (5091 to 7145 referrals).
From the sampled notes the proportion of referrals which were adult true positive glaucoma cases averaged 5.6%. There was no significant variation in this proportion over the five years (table) (X2=5.47; df=4; P=0.34).
Before 1 April 1989 routine sight tests performed by either optometrists or ophthalmic medical practitioners in the United Kingdom were funded by the NHS. Unconditional funding was withdrawn after that date. Subsequently around 60% of tests have been performed privately at an average cost of pounds sterling 13.20 each.3 Free tests have remained available to people aged under 16 (or under 19 if in full time education), those receiving income support, patients with diabetes mellitus, glaucoma patients and their first degree relatives aged over 40, patients requiring complex lenses, and those registered either partially sighted or blind.
The sight test is the only existing means of screening adults in Britain for ocular disease. Glaucoma is a prevalent chronic and incurable disorder, occurring in 7% of people aged over 75.4 Owing to predicted demographic changes the incidence of this condition is expected to increase. Glaucoma is an important cause of blindness, accounting for up to 15% of new blind registrations.*RF 5-7* Unlike many other causes of blindness, glaucoma is treatable if identified in its early, usually asymptomatic stages.*RF 8-10* Up to 90% of confirmed cases of glaucoma are first suspected as a result of screening during a routine sight test.1,11,12 True positive screening referrals are therefore the source of the vast majority of cases of glaucoma.
Concern was expressed that introducing a fee would deter people, especially elderly people, from having sight tests.2 It was hypothesised that fewer tests would result in less screening, that fewer patients would be referred to the hospital eye service, and that important sight threatening diseases such a glaucoma would remain untreated.
The effect of introducing the sight test fee on either the rate of referral of patients with suspected eye disease or the rate of identification of serious eye disease such as glaucoma had not been determined. We measured the total number of referrals and the number of adult true positive glaucoma referrals being received as a means of detecting any such shifts. Estimates of the effect on the number of sight tests being performed cannot be verified because no official records are kept of private tests.3,13,14
A total of 979 (15%) fewer referrals were received at the Bristol Eye Hospital in 1989 than in 1988. Some 1.8 million extra sight tests were performed in the 15 months leading up to the introduction of the fee.14 Despite this, however, only 343 (5.6%) more referrals than predicted were received in 1988. The reduced totals from 1989 onwards should not therefore be attributed to early referral.
The proportion of all referrals which were adult true positive glaucoma cases did not vary significantly between 1987 and 1991 and averaged 5.6%. Between 13.7% and 19.0% fewer than expected referrals of all kinds were received after the introduction of the sight test fee. The rate of identification of glaucoma was therefore correspondingly reduced after the introduction of the fee. From these figures, in 1992 a total of 356 true positive glaucoma referrals would have been received at the Bristol Eye Hospital; 438 would have been predicted from trends observed before the introduction of the fee. This shortfall represents one case a year per 10 000 of the populations.15 If this local reduction in the rate of glaucoma detection reflects a sustained nationwide phenomenon an increased incidence of preventable blindness can be anticipated.
We thank Miss L Porter, Miss S J Torrington, Mrs M Mathews, and Mrs J Hooper, of the Bristol Eye Hospital's medical records department, who collected most of the data. We also thank Dr R Midwinter, of the Bristol University Department of Public Health, Mr R P L Wormald, of St Mary's Hospital Medical School, and Professor N Butler, of the City University Social Statistics Unit, for helpful advice. The study was funded by the International Glaucoma Association.