Clinical and Experimental Ophthalmology 2011; 39: 760–765 doi: 10.1111/j.1442-9071.2011.02556.x
Original Article Telemedicine model to prevent blindness from familial glaucoma Sandra E Staffieri BAppSci(Orth),1 Jonathan B Ruddle FRANZCO,1 Lisa S Kearns BOrth,1 Julie M Barbour DOBA,2 Thomas L Edwards MBBS PhD,3 Padma Paul MS1 and David A Mackey MD FRANZCO1,3,4 1
Centre for Eye Research Australia, University of Melbourne, Department of Ophthalmology, Royal Victorian Eye and Ear Hospital, Melbourne, Victoria, 2Launceston Eye Institute, Launceston, 3Royal Hobart Hospital/University of Tasmania, Hobart, Tasmania, and 4 Centre for Ophthalmology and Visual Science, Lions Eye Institute, University of Western Australia, Perth, Western Australia, Australia
ABSTRACT Background: To develop, implement and evaluate a telemedicine model to reduce glaucoma blindness through the early detection of undiagnosed glaucoma in high-risk individuals. Design: Prospective study, private ophthalmology practice and public outpatient clinics in Tasmania. Participants: One hundred and thirty-three individuals with primary open-angle glaucoma were invited to enrol their first-degree relatives (FDRs) to undergo an eye examination. Within the study period, 211 FDRs were available for examination. Methods: A registered nurse was trained to perform the required assessments. Clinical data were entered into a purpose-built database, converted to a portable document format and graded offsite by an ophthalmologist to determine the presence, absence or risk of developing glaucoma. Participants were notified of the grading result and recommendations for review. Main Outcome Measures: Incidence of undiagnosed glaucoma in a high-risk population. Results: Previously undiagnosed glaucoma was identified in 5% of those examined. For every 19 participants screened, one new case of previously undiagnosed case of glaucoma was identified.
Additionally 15% of participants showed suspicious signs of glaucoma, and 6% had ocular hypertension. Conclusions: A telemedicine model is an efficient method for screening, grading and notifying participants of examination results. Nurses can be adequately trained to undertake the initial screening examinations, with grading of the results performed offsite by a suitably qualified ophthalmologist. Targeted screening for glaucoma increases the yield of identifying individuals with undiagnosed glaucoma or those at greatest risk. Cost efficiencies for this model of glaucoma screening should be further explored and implemented to prevent blindness from familial glaucoma. Key words: glaucoma, screening, telemedicine.
INTRODUCTION Glaucoma is the leading cause of untreatable blindness in the world. Primary open-angle glaucoma (POAG) is the most common subtype, and half of people with POAG are unaware that they have lost vision as a result of the condition. Early diagnosis and treatment can prevent the progression of visual loss and eventual blindness. Glaucoma has a strong genetic predisposition, with first-degree relatives (FDRs) at 10 times the population risk of developing the disease.1 Current international guidelines suggest that glaucoma screening for the entire population is not cost-effective, lacking sensitivity and specificity, except in high-risk groups: patients with diabetes, those with myopia and relatives of people
䊏 Correspondence: Professor David A Mackey, Centre for Ophthalmology and Visual Science, Lions Eye Institute, University of Western Australia, 2 Verdun St, Nedlands, WA 6009, Australia. Email:
[email protected] Received 28 July 2010; accepted 11 February 2011. © 2011 The Authors Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
Telemedicine model for glaucoma screening with glaucoma.2 Most guidelines now recommend opportunistic case detection or screening of highrisk groups, such as those with a family history of glaucoma. Despite studies demonstrating the feasibility of using telemedicine as an efficient method for diagnosing glaucoma,3,4 there is currently no universally accepted protocol for glaucoma screening in the community using this technology nor have targeted familial glaucoma screening programmes been evaluated. The ‘Telemedicine Model to Prevent Blindness from Familial Glaucoma’ was designed to determine the feasibility of screening high-risk individuals for undiagnosed glaucoma using a telemedicine approach. In addition, participation rates and barriers to participation were reviewed so that public awareness programmes could be better targeted in an effort to reduce the incidence of familial glaucoma blindness.
METHODS Recruitment Ethics approval was obtained from the Human Research and Ethics Committees at the University of Tasmania. Individuals with advanced POAG were identified from the records of the Eye Clinic at the Royal Hobart Hospital and private eye clinics in Southern and North-Western Tasmania. Once the diagnosis was confirmed, these individuals were invited either in person, by letter or telephone to involve their FDRs (parents, siblings and children over 40 years of age) in the project by having them undergo a free, comprehensive eye examination. Index cases were provided with a brochure highlighting the increased risk of FDRs developing glaucoma to pass on to their family members.
Training A registered nurse was trained to perform the eye assessments that were required including: • •
•
• •
Visual acuity – standard logMAR Bailey-Lovie visual acuity chart Refractive status – Nidek Autorefractor/ keratometer. Model ARK-30 Type R (Nidek Co, Gamagori, Japan) Computerized visual field testing – Humphrey 24-2 SITA Zeiss Humphrey Visual Field Analyser Intraocular pressure (IOP) measurement – Tonopen XL (Reichert, Inc., New York, USA) Central Corneal thickness (CCT) measurement – Pachmate DGH 55 (DGH Technology, Inc., Exton, PA, USA)
761 •
Stereoscopic optic disc photographs – Nidek 3-Dx Stereodisc camera (Nidek Co. Ltd, Gamagori, Japan)
Tropicamide (Mydriacyl) 1% was instilled in each eye to adequately dilate the pupils prior to taking the optic disc photographs. The total examination time for each participant was approximately 1 hour, including the 20-min waiting period for dilating eye drops to take effect. Following the resignation of the trained nurse, the examinations were performed by a previously trained technician, orthoptists and ophthalmology registrar trainees.
Data collection and assessment A database that could store clinical data and then produce the required reports for grading and feedback to the participant was developed using MS Access 2007 (Microsoft Corp., Redmond, WA, USA). Data collation and imputation into the specifically designed database were completed by the study coordinator. A report for each participant was then converted to a summary portable document format (pdf, Adobe Systems Inc. San Jose, CA, USA). This included all the clinical parameters, visual field results and stereoscopic optic disc photographs. These documents were then sent via a secure server for viewing and grading by an ophthalmologist with subspecialty training in glaucoma (JBR). Subjects were classified as having definite glaucoma on the basis of characteristic optic nerve head changes (cup:disc ratio [CDR] outside the 97.5 percentile for the normal population or rim width less than 0.1 CDR at the superior and inferior poles of the disc) and definite visual field defect consistent with glaucoma.5 Individuals with stereoscopic disc photos consistent with structural damage but in whom field testing was unreliable or unobtainable were classified as glaucoma suspect. In both glaucoma cases and suspects, no alternative explanation for the disc structural change (such as high myopia or previous vascular disease) or the field structural change (such as macular degeneration or cerebrovascular disease) could be present. Ocular hypertension (OHT) was diagnosed on the basis of an IOP greater than 21 mmHg. As gonioscopy was not performed by the nurses and orthoptists performing the survey, there was no way of determining whether patients had primary openangle or angle-closure forms of glaucoma. When an individual was known to have glaucoma and was already receiving treatment, this was confirmed on the clinical data obtained. The outcomes for the other participants were graded as:
© 2011 The Authors Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
762
Staffieri et al. Declined
Unwell (n = 3) 9%
(n = 34) 19%
Deceased (n = 12) 7%
No FDRs (n = 11) 32% Not Interested (n = 20) 59%
Index available (n = 133) 74%
Figure 2. Reason for index case declining participation. Figure 1. Index case recruitment. Not yet booked,
• • •
Glaucoma – for urgent referral Suspicious signs of glaucoma – for routine referral No glaucoma – recommended review in 1, 2 or 5 years time depending on the clinical findings and the participant’s age.
(n = 173) 29%
FDRs Examined (n = 211) 36%
Declined
Participants were then sent a letter, with copies for the general practitioner and usual eye health-care provider, advising them of the grading outcome along with a recommendation for review.
(n = 21) 4%
Deceased (n = 182) 31%
RESULTS
Figure 3. First-degree relatives (FDRs) Recruitment.
One hundred and seventy-nine individuals with glaucoma were identified and invited to participate. Of these, 133 were examined to confirm the diagnosis and invite participation of their FDRs (Fig. 1). Thirty-four index cases declined involvement, 59% of whom were not interested in participating or involving their family members (Fig. 2). Only 76 of these index cases had FDRs who were available or willing to participate. A total of 587 FDRs were identified for possible recruitment; however, given the age (65 years or more) of the index cohort, FDRs included parents, the majority of whom were deceased. Of these, 211 could be examined and included during the study period (Fig. 3). Male FDRs declined participation in the study more commonly than females (67% vs. 33%), citing ‘not interested’ or ‘too busy’ as their reasons (Fig. 4). Of the 29% of FDRs who had not yet been booked, 45% lived away from the examination sites, were unwell or were unable to attend the examination sessions on the scheduled dates (Fig. 5). The as yet unexamined FDRs will be contacted and invited to be examined as part of the ongoing Glaucoma Inheritance Study in Tasmania.
Female (n = 7) 33%
Male (n = 14) 67%
Figure 4. Male : female first-degree relatives (FDRs) declining participation.
Although 6% of FDRs had already been diagnosed with glaucoma, an additional 5% were identified with signs of glaucoma requiring prompt referral for treatment. Another 15% were found to have suspicious signs of glaucoma. Six per cent of participants were found to have OHT (Fig. 6). Participants labelled as glaucoma suspects or OHT were advised to be reviewed again in 6 months or a year.
© 2011 The Authors Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
Telemedicine model for glaucoma screening other (n = 9) 5%
763
unable to attend (n = 29) 17%
unable to contact (n = 9) 5% not yet booked (n = 66) 38%
unwell (n = 10) 6%
Interstate (n = 30) 17% estranged (n = 12) 7%
Overseas (n = 8) 5%
Figure 5. Reasons first-degree relatives (FDRs) not yet examined.
Previously diagnosed glaucoma (n = 12) 6%
Newly Diagnosed Glaucoma (n = 11) 5% Glaucoma Suspect (n = 31)15%
Ocular Hypertension Nil Glaucoma
(n = 13) 6%
(n = 144) 68%
Figure 6. Grading result of pdf for first-degree relatives (FDRs) examined.
For every 19 FDRs screened, one new case of glaucoma was diagnosed (total FDRs ⫼ new glaucoma identified = number needed to screen).
DISCUSSION In their briefing paper, the European Health Telematics Association recognized telemedicine as a rapidly developing area in the delivery of health care via technology. The potential applications of the telemedicine model of health care are wide and varied, enabling service providers to provide medical care in a cost-effective manner utilizing appropriate resources, particularly in remote or rural areas.6 The use of telemedicine in screening for diabetic eye dis-
ease,7,8 and retinopathy of prematurity9 has been explored and successfully implemented. This is not yet the case for glaucoma. Burr et al. (2007)2 recommended that glaucoma detection could be improved with increased participation or utilizing technology as an initial case-identifying strategy, translating to improved cost-effectiveness. Additionally, a sharedcare model for glaucoma screening has been reported utilizing telemedicine technology, with good results for detecting undiagnosed glaucoma as part of a routine eye examination.10 However, a telemedicine model for large-scale screening for glaucoma in an at-risk population is yet to be implemented. The concept of targeted screening for a preventable blinding eye disease increases the likelihood of identifying disease. Glaucoma, with its reported hereditary nature, fits suitably into a targeted screening approach.11,12 Although 20% of index cases in our cohort were not interested in participating, a relatively high participation rate was observed for the FDRs. This would suggest that informed individuals are keen to take preventative measures for their health care if readily accessible. It is well documented that a positive family history for glaucoma increases an individual’s risk of developing the disease.11,12 Thus, for targeted screening to be effective, it is vital that both individuals and primary care physicians understand the implications of a known family history of glaucoma. Notably twice as many FDR men as women chose not to participate. Given the lower rate of access to medical testing in males,13 the target audience for developing the information brochure for FDRs was primarily middle-aged men. Increased resources for promoting health awareness and strategies for prevention, particularly to this target audience and primary care physicians, are required. Barriers to participation for the FDRs included either being interstate or overseas or unable to attend on the respective examination dates. With sufficient resources, more FDRs could be examined and participation rates enhanced by providing opportunities for examination on weekdays, weeknights and weekends. For some, the length of time required for the examination and subsequent inconvenience of the dilating drops, which can last for up to two or more hours, hindered participation. With the release of the new generation non-mydriatic stereoscopic fundus cameras (e.g. Kowa nonmyd WX3D, Kowa Company Ltd, Tokyo, Japan), the required optic disc photographs could be reliably obtained in most patients without the use of dilating eye drops. Not only would this reduce the total examination time to approximately 40 minutes, but the short-term effects of pupil dilatation for the participant would be eliminated, allowing immediate return to their daily activities.
© 2011 The Authors Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
764
Staffieri et al.
Nurse-led diabetic8,14 and retinopathy of prematurity9 screening programmes have shown favourable results, enhancing eye health-care delivery. In this present study we demonstrated that a nurse could be trained in the examination techniques required for glaucoma screening. The orthoptist-trained registered nurse resigned after identifying the index cases and examining only 20 FDRs. In the interest of completing the project in the time frame required, the project continued with a research orthoptist as coordinator. A technician, with no medical qualifications, had been previously trained by the research team for earlier projects to perform the visual acuity, visual fields, tonometry and CCTs. The technician completed over half of the clinical examinations as well as approximately a third of the photography. Given the original nurse was trained to administer the necessary examinations and the technician performed examinations on more than half of the remainder of participants, the premise that non-eye health-care practitioners can be trained to perform the necessary clinical assessments was adequately demonstrated, and the conclusions regarding training not compromised. Training to examination for shallow anterior chamber, an indication not to dilate the pupils because of the risk of causing acute angle-closure glaucoma, was not undertaken; however, studies have shown that this is also possible.15 As this pilot study is targeting familial open-angle glaucoma, mostly in younger individuals, this was not routinely conducted in our screening protocol. This telemedicine model demonstrated that offsite grading of clinical data and images is feasible. Central to its success was the ease and speed with which the data could be entered and converted to a useful and readily accessible document for grading. Additionally, prompt feedback of the grading result to the participant for information and appropriate action was a vital component of the success of the screening programme. Using MS Access 2007 (Microsoft) the module we designed enabled the collection and efficient retrieval of relevant clinical data in a report format for grading. For future use, a secure web-based SQL format would enhance accessibility of this database. Utilizing the systems developed, a shared care model could be envisaged involving local eye care providers including orthoptists, optometrists or ophthalmic-care nurses who feedback screening results in a standardized manner to a reporting centre with glaucoma subspecialty expertise. Compared with an unselected population screening programme that detects 3–5% of the population over 40 to have glaucoma,16 our study newly diagnosed 5% of the examined population as well as 6% of cases who were already diagnosed, giving a cumu-
lative incidence of 11% with glaucoma. In the recruitment phase of FDRs, a diagnosis of glaucoma was not always evident as appointments were made by a third party or patients were unaware of their diagnosis of glaucoma. This is an issue that could be addressed in improving patient awareness and education. Future recruitment protocols could include questions regarding current ocular treatment in an effort to better identify individuals already diagnosed with glaucoma. ‘Glaucoma suspect’ is the diagnostic label given to patients who have features consistent with, but not diagnostic for, glaucoma. This can include OHT, patients with IOP above the population standard deviation of >21 mmHg but no visual field or optic disc changes. Glaucoma can also be suspected on the basis of an increased CDR, thin corneas or visual field tests with a decreased mean deviation. Previous studies have also shown that individuals with untreated OHT will go on to develop glaucoma at twice the rate of those who receive IOP-lowering treatment.17 Although treatment decisions depend on individual patient factors such as age and perception of risk, it is significant that 6% of our cohort were diagnosed with OHT and thus will need either treatment or closer monitoring. It is highly likely that a large proportion of our glaucoma suspects (15%) will go on to develop the disease. In addition to the major risk factor of an affected FDR, these patients also exhibited an enlarged CDR or performed poorly on their visual field test. A follow-up study is recommended to corroborate the results of the patients referred with suspected glaucoma and OHT. To further evaluate the effectiveness of such a targeted screening approach, we considered the number needed to screen (NNS) in order to identify a new, previously undiagnosed case of glaucoma. Examining 3654 individuals from an unselected population, the Blue Mountains Eye Study, Mitchell et al.16 identified 108 cases of glaucoma, 49% of whom were newly diagnosed. This yielded an NNS of 66 (3564 ⫼ 54 = 66). Compared with our NNS of 19, targeted screening of high-risk family members would yield a higher number of otherwise undiagnosed glaucoma cases, allowing for earlier treatment and prevention of glaucoma blindness. This project developed a telemedicine model for glaucoma screening in high-risk individuals. The feasibility of training non-eye health-care professionals to undertake the required screening examinations and successful grading of clinical summary documents remotely by a suitably qualified ophthalmologist was demonstrated. With fully portable equipment, the telemedicine model could successfully bring improved medical care to remote and rural communities. Screening for glaucoma in individuals with a family history of this preventable,
© 2011 The Authors Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
Telemedicine model for glaucoma screening blinding eye disease would be cost-effective and resource efficient, utilizing current and future technology as well as medical expertise.
ACKNOWLEDGEMENTS We would like to thank the Tasmanian Ophthalmologists for their assistance of this project and the GIST over the last 17 years; the participating ophthalmologists for allowing access to their private records and use of their private consulting suite facilities for field work (Dr Paul McCartney, Hobart Eye Surgeons, Hobart; Dr Michael Haybittel, NorthWest Medical Centre, Burnie; Dr Brendan Vote, Launceston Eye Institute, Launceston; Dr Ian Murrell, Launceston Eye Institute, Launceston; and Dr Nick Downie, Launceston Eye Hospital, Launceston); Ross Dunn, Centre for Eye Research Australia, Melbourne, for assistance with database development; Dr Evelyn O’Neill, Centre for Eye Research Australia, Melbourne; Dr Paul O’Connell, Centre for Eye Research Australia, Melbourne; Nicole Van Bergen, Centre for Eye Research Australia, Melbourne, and Rachel Adams, Launceston Eye Institute, Launceston, for assistance with field work; Patricia Coleman, Royal Hobart Hospital, Hobart, for laboratory assistance; Maree Ring, Royal Hobart Hospital, Hobart, for genealogy assistance and Anneliese Wicks RN, Royal Hobart Hospital, Hobart, for initial recruitment and examination work on the project. This project was federally funded by the National Eye Health Initiative, Department of Health and Ageing. Support from ORIA and the RANZCO Eye Foundation is acknowledged. Professor David Mackey is a recipient of the Pfizer Australia Senior Research Fellowship. In addition, the Centre for Eye Research Australia receives Operational Infrastructure Support from the Victorian Government.
REFERENCES 1. Wolfs RC, Klaver CC, Ramrattan RS, van Duijn CM, Hofman A, de Jong PT. Genetic risk of primary openangle glaucoma. Population-based familial aggregation study. Arch Ophthalmol 1998; 116: 1640–5. 2. Burr JM, Mowatt G, Hernandez R et al. The clinical effectiveness and cost-effectiveness of screening for open angle glaucoma: a systematic review and economic evaluation. Health Technol Assess 2007; 11: iii–iv, ix–x, 1–190. 3. Bergua A, Mardin CY, Horn FK. Tele-transmission of stereoscopic images of the optic nerve head in glaucoma via internet. Telemed J E Health 2009; 15: 439–44.
765 4. Kumar S, Giubilato A, Morgan W et al. Glaucoma screening: analysis of conventional and telemedicinefriendly devices. Clin Experiment Ophthalmol 2007; 35: 237–43. 5. Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The definition and classification of glaucoma in prevalence surveys. Br J Ophthalmol 2002; 86: 238–42. 6. European Health Telematics Association. Sustainable telemedicine: paradigms for future-proof healthcare. A briefing paper. 2008. Available from: http://www. ehtel.org/forum/tasks-sources/task-force-sustainabletelemedicine-and-chronic-disease-management/ehtelbriefing-paper-sustainable-telemedicine 7. Cuadros J, Bresnick G. EyePACS: an adaptable telemedicine system for diabetic retinopathy screening. J Diabetes Sci Technol 2009; 3: 509–16. 8. Mayes PA, Silvers A, Prendergast JJ. New direction for enhancing quality in diabetes care: utilizing telecommunications and paraprofessional outreach workers backed by an expert medical team. Telemed J E Health 2010; 16: 358–63. 9. Murakami Y, Silva RA, Jain A, Lad EM, Gandhi J, Moshfeghi DM. Stanford University Network for Diagnosis of Retinopathy of Prematurity (SUNDROP): 24-month experience with telemedicine screening. Acta Ophthalmol 2010; 88: 317–22. 10. de Mul M, de Bont AA, Reus NJ, Lemij HG, Berg M. Improving the quality of eye care with teleophthalmology: shared-care glaucoma screening. J Telemed Telecare 2004; 10: 331–6. 11. Green CM, Kearns LS, Wu J et al. How significant is a family history of glaucoma? Experience from the Glaucoma Inheritance Study in Tasmania. Clin Experiment Ophthalmol 2007; 35: 793–9. 12. Guedes RA, Guedes VM, Chaoubah A. Focusing on patients at high-risk for glaucoma in Brazil: a pilot study. J Fr Ophtalmol 2009; 32: 640–5. 13. Harris MF, McKenzie S. Men’s health: what’s a GP to do? Med J Aust 2006; 185: 440–4. 14. Kirkwood BJ, Coster DJ, Essex RW. Ophthalmic nurse practitioner led diabetic retinopathy screening. Results of a 3-month trial. Eye (Lond) 2006; 20: 173–7. 15. Nuriyah Y, Ren XT, Jiang L, Liu XP, Zou YH. Comparison between ophthalmologists and community health workers in screening of shallow anterior chamber with oblique flashlight test. Chin Med Sci J 2010; 25: 50–2. 16. Mitchell P, Smith W, Attebo K, Healey PR. Prevalence of open-angle glaucoma in Australia. The Blue Mountains Eye Study. Ophthalmology 1996; 103: 1661–9. 17. Kass MA, Heuer DK, Higginbotham EJ et al. The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary openangle glaucoma. Arch Ophthalmol 2002; 120: 701–13; discussion 829–30.
© 2011 The Authors Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
