Author Affiliations: Singapore Eye Research Institute and Singapore National Eye Centre, Singapore (Drs Quek, Perera, Lamoureux, and Aung and Ms Ong); Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Ear Hospital, Victoria (Dr Lamoureux); and the Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore (Dr Aung).
Glaucoma, the leading cause of irreversible blindness worldwide,1 frequently requires long-term medical treatment to lower intraocular pressure (IOP) and prevent progressive visual field loss.2,3 Patients, however, often do not use medications as prescribed4- 10 and are prone to poor adherence and persistence,11 leading to suboptimal treatment outcomes and increased health care costs.12,13 Persistence measures the duration of time that a patient fills a prescription and continues a prescribed therapy,14 which is a prerequisite for adherence. Persistence rates of patients who started taking initial glaucoma medications in the United States have been reported to be as low as 25% at 1 year,8,9,15,16 with prostaglandin analogues reportedly having higher persistence rates than beta-blockers and other classes of ocular hypotensive therapies.6,8- 10,16- 22
It has been estimated that by 2020,1 47% of the estimated 79.6 million people with glaucoma will be Asian, but to date, there are no data on persistence rates of IOP-lowering medications in Asian persons with glaucoma. The aim of this study was to determine the persistence rates during 3 years after Asian patients started taking topical IOP-lowering monotherapy in a Singapore hospital.
The study had the approval of the hospital's institutional review board. We performed a retrospective review of the pharmacy database of patients at the Singapore National Eye Centre (SNEC) during a 4-year period from October 1, 2005, to September 30, 2009. Patients who started taking a single IOP-lowering medication between October 1, 2005, and September 30, 2006 (and who were never dispensed any IOP-lowering therapy in the year before) were identified to measure persistence. Pharmacy dispensing records of this cohort were traced from the date of the first prescription to 3 years thereafter. The SNEC pharmacy database was created on October 1, 2004, and includes information on patients' demographics, billing status (those receiving government subsidies vs those without), and the type, quantity, and laterality of medication(s) prescribed. At the SNEC, almost all patients fill their prescriptions at a single pharmacy located within the center. Patients who were not receiving government subvention paid approximately 7% more for all medications (except for Travoprost, which was made available to patients with government-subsidized health care at half the price paid by patients with nonsubsidized care). Those who chose to have their prescriptions filled at other pharmacies did not receive the subsidies.
A patient was defined as persistent if he or she was prescribed the same medication before or within 90 days after the previous prescription had lapsed. These patients were further subdivided into those who required medications in addition to the one originally prescribed and those who did not. Patients who were not persistent were classified as having terminated therapy (never received another prescription for any IOP lowering medication), changed medication (received a prescription for another IOP-lowering medication without a lapse), or restarted therapy (received a prescription for the same IOP-lowering medication after a lapse). Patients who restarted therapy were subdivided into those who received (1) the same medication alone, (2) other medication(s) in addition to the original one, or (3) different medication(s) from the original.
Adjustments for the effect of bottle size were not made. This was because expected refill patterns predicted by bottle size and the mean number of drops have been shown not to reflect observed refill patterns, with patients who were dispensed larger bottle sizes returning for refills sooner than expected, regardless of the class of therapy.23 At the SNEC, patients were instructed to discard each bottle of medication 30 days after opening, and sufficient quantities of each medication were dispensed, according to the interval to the next scheduled review.
Persistence rates for each medication were calculated at 1 and 3 years. Patient demographics (age, sex, race, resident status), billing status, and laterality of eye drop prescription (unilateral vs bilateral) were compared between those who were and were not persistent. As a secondary analysis, we computed the corresponding persistence rates using varying intervals (14, 28, and 60 days) between successive prescriptions in the definition of persistency.
A total of 2781 patients started receiving IOP-lowering monotherapy between October 1, 2005, and September 30, 2006. There were similar numbers of men and women (50.3% men), with a mean (SD) age of 61.1 (15.7) years. Of these patients, 81% (n = 2254) were Chinese, 85% (n = 2374) were Singaporean residents, and 51% (n = 1423) were receiving government subsidies for their medications (Table 1). Timolol maleate, 0.5%, was the most frequently prescribed medication (62.2%), followed by Travoprost, 0.004% (10.4%), latanoprost, 0.005% (10.4%), brimonidine, 0.15% (9.5%), dorzolamide, 2% (4.1%), pilocarpine, 4% (1.9%), bimatoprost, 0.03% (0.8%), and brinzolamide, 1% (0.7%). A total of 62.5% of patients (n = 1738) were prescribed unilateral therapy, and the remaining 37.5% (n = 1043) started receiving bilateral treatment.
At the end of 1 year, only 626 of 2781 patients (22.5%) persistently took the same therapy. Of these, 62.1% (n = 389) continued taking the same medication, while 37.9% (n = 237) were prescribed additional IOP-lowering medication(s). Table 2 shows the persistence rates of various medications at 1 year. Of the 2155 patients (77.5%) who were not persistent at the end of 1 year, 66.3% (n = 1428) terminated therapy and were never subsequently prescribed any IOP-lowering medication, 17.6% (n = 380) started taking a different medication (without a lapse), 12.7% (n = 273) restarted therapy with the same medication, 3.1% (n = 67) restarted therapy a different medication, and 0.3% (n = 7) restarted therapy with other medication(s) in addition to the initial one (Figure 1).
Persistence rates at the end of 1 year.
Of the 1428 subjects who terminated therapy at the end of 1 year and were never subsequently prescribed any IOP-lowering medication, only 30.4% (434 of 1428) defaulted follow-up and never subsequently attended SNEC. The remaining 69.9% were still attending SNEC after 1 year.
Compared with persistent patients, the nonpersistent group tended to be younger (mean age, 60.2 vs 64.1 years; P < .001) and more likely to not be Singaporean (16.3% vs 12.0%; P = .008), not be receiving government subsidies (51.7% vs 42.3%; P < .001), and be receiving unilateral therapy (68.6% vs 41.5%; P < .001). There were no significant differences in sex or race between the 2 groups (P = .51 and P = .25, respectively). When compared with all other medications combined, prostaglandin analogues as a group (travoprost, latanoprost, bimatoprost) had a better persistence rate at 1 year (29.6% vs 20.6%; P < .001). Similarly, the persistence rate of prostaglandin analogues as a group was better than that of timolol (29.6% vs 23.7%; P = .004).
After 3 years, only 320 of 2781 patients (11.5%) persistently received the same therapy. Of these, 40.9% (n = 131) continued taking only the same medication, while 59.1% (n = 189) required additional IOP-lowering medication(s). Of the 2461 subjects (88.5%) who were not persistent at the end of 3 years, 56.3% (n = 1386) terminated therapy and were never subsequently prescribed any IOP-lowering medication during the duration of the study period, 17.7% (n = 436) began taking a different medication (without a lapse), 20.1% (n = 494) restarted treatment with the same medication, 5.6% (n = 137) restarted treatment with a different medication, and 0.3% (n = 8) restarted treatment with other medication(s) in addition to the initial one (Figure 2). Timolol had a better persistence rate at 3 years (12.5% vs 9.8%; P = .03) than all other medications combined. There was no significant difference in persistence rates (at 3 years) of prostaglandin analogues as a group (12.9%) compared with all other medications combined (11.1%) or prostaglandin analogues compared with timolol (12.5%; P > .05).
Persistence rates at the end of 3 years.
We further computed the corresponding persistence rates using varying intervals between successive prescriptions as the definition of persistence. The overall persistence rates at 1 year using 14-, 28-, and 60-day maximal intervals in the definition of persistence were 8.6%, 12.8%, 19.0%, respectively (Table 3 and Table 4).
As secondary analysis, we excluded patients who began taking a single IOP-lowering medication for a period of less than 30 days and never subsequently prescribed either the same or any other IOP-lowering medication thereafter. Such patients were presumably prescribed short-term IOP-lowering therapy for transient spikes in IOP. After excluding these patients, the persistence rates at the end of 1 and 3 years were 34.9% and 18.0%, respectively.
We performed a further analysis of the patients who began taking unilateral monotherapy for a period of more than 30 days. Of the 887 patients who belonged to this group, persistence rates were 29.8% (n = 264) and 15.6% (n = 138) at the end of 1 and 3 years, respectively. These rates remained statistically lower than those of patients who began receiving bilateral monotherapy for a period of greater than 30 days (40.2% and 20.5% at the end of 1 and 3 years, respectively; P < .001 and P = .008, respectively).
Using a 90-day interval between successive prescriptions as the definition of persistence, the persistence rates among Singaporean individuals at the end of 1 year was 23.4%. This is not significantly different from the overall persistence rate of 22.5% (P = .44). Hence, although persons in the nonpersistent group were more likely to not be Singaporean (P = .008), the persistence rate of Singaporean individuals was similar to that of Singaporean individuals and those who are not Singaporean combined.
Patients fail to be persistent with glaucoma therapy for various reasons including cost, tolerability, difficulty with administration, denial, lack of education, forgetfulness, and physician's satisfaction with IOP control. To our knowledge, this is the first study undertaken to assess the persistence rates of patients who began taking glaucoma medication in an Asian country, namely Singapore. We found persistence rates to be only 22.5% at 1 year and 11.5% at 3 years, figures that are notably lower than have been reported in white populations.9,17,21,24- 26 The increased likelihood of angle-closure glaucoma diagnoses in this Asian population might skew the persistence data compared with the greater number of patients with open-angle glaucoma in the previously published Western studies. The findings are of concern and have implications for glaucoma care in Singapore and possibly other countries in Asia. The poor persistence shows that patients frequently change or discontinue their treatment regimen, with potential effects on therapeutic effectiveness. These results indicate that strategies are urgently needed to elucidate reasons for poor persistence and to improve persistence in this population.
Previous studies have reported prostaglandin analogues to be associated with better persistence than any other drug class. For example, latanoprost-treated patients have been found to have higher persistence and lower rates of therapy failure and discontinuation compared with patients treated with other IOP-lowering medications, including beta-blockers.6- 10,16- 19,21,22 Our study concurs that prostaglandin analogues, as a group, have better persistence rates at 1 year compared with all other medications combined. In a retrospective survey of pharmaceutical records of American patients recruited during a 6-month period and followed up for 12 months, 1-year persistence rates of 69.4% for latanoprost, 70.6% for travoprost, and 68.1% for bimatoprost were reported.24 This is in contrast to our 1-year persistence rates of 36.7% for travoprost, 23.6% for latanoprost, and 14.3% for bimatoprost. Possible reasons for the differences between the 2 studies could be differing inclusion criteria, socioeconomic backgrounds, and health care systems. However, at 3 years, we found the persistence rate of timolol to be higher than for all other medications combined. Various factors including the cheaper cost of timolol, low frequency of ocular adverse effects, and physician and/or patient preferences could account for this observation.
Previous studies have used various time intervals between successive prescriptions to define persistence.7,9,15,16,22,25,27 Admittedly, the low persistence rates in our study could be a result of using a maximum interval of 90 days between successive prescriptions compared with intervals as long as 120 to 180 days in other studies.6,16,17,21 We also determined persistence for all patients who began receiving IOP-lowering monotherapy; in contrast, some studies only included patients who were persistent for at least the first 90 days of therapy.24 Excluding patients who are not persistent during the first 90 days of therapy has the advantage of leaving out patients who may have had a short hypertensive episode that did not require long-term treatment. On the other hand, imposing this arbitrary 90-day exclusion criterion may also lead to an overestimation of persistence, as it excludes patients who default within the first 90 days of commencement of therapy. Nonetheless, a reason for the low persistence rates in our study could be the inclusion of patients requiring only short-term therapy (for example, for postoperative IOP spikes, after trauma, or uveitis), as supported by the high proportion of unilateral (68.6%) therapy in the nonpersistent group and possibly the high proportion of beta-blockers compared with prostaglandin analogues. After having excluded patients who were presumably prescribed short-term IOP-lowering therapy for transient spikes in IOP, the persistence rates at the end of 1 and 3 years were 35% and 18% respectively (compared with 22.5% and 11.5%). However, these persistence rates were still low compared with previous figures in the literature.24,25
We found that younger age was associated with poor persistence. This finding is consistent with a recent article indicating that younger subjects (35-64 years) were more likely to not be persistent compared with older patients.25 It is not surprising that residents who were not Singaporean were less likely to be persistent, as they may elect to continue treatment back in their country of origin. Patients who began taking unilateral therapy were also found to be less persistent than those who received bilateral therapy. A plausible explanation could be that those who began taking bilateral therapy were more aware (and perhaps fearful) of the risk of losing sight in both eyes and hence more persistently receiving therapy. We found that patients who did not receive government subsidies for their medications were less likely to be persistent. Travoprost, which was made available to patients with government-subsidized care at a much lower price than other prostaglandin analogues had the highest persistence rate at the end of 1 and 3 years. These observations highlight the importance of affordability of treatment as a factor influencing persistence (and adherence) in chronic diseases like glaucoma that require long-term therapy.
We found that, of the patients who were persistent at 1 year, 62.1% took only the same medication, while 37.9% required additional IOP-lowering medication(s). This suggests that about one-third of patients who begin taking and persistently receive glaucoma monotherapy either have inadequate IOP control and/or disease progression after 1 year, requiring additional IOP-lowering medication(s). This is consistent with the Ocular Hypertension Study and Collaborative Initial Glaucoma Treatment Study, which reported that 49% and 75% of subjects, respectively, required adjunctive therapy.28,29 Of those who were not persistent at the end of 1 year, two-thirds (66.3%) stopped receiving therapy and were never subsequently prescribed any IOP-lowering medication. Possible reasons for patients discontinuing therapy include seeking treatment at other institutions, opting for other forms of treatment such as laser or surgery, death, or noncompliance. As this was a review of pharmaceutical dispensing records and not patient records, we were unable to elucidate the exact reasons for termination of therapy in these patients. Interestingly, about 18% of those who were not persistent at 1 year began taking a different medication, without a lapse in prescriptions, for possible reasons such as inadequate IOP control, adverse effects, cost, or poor adherence.
Our study has a few limitations. Being a retrospective review of pharmacy dispensing records, we were unable to determine the reasons for discontinuation of therapy or poor persistence such as adverse effects, cost, or poor efficacy of medications. Our database also did not include the underlying diagnoses of patients; hence, this study was not able to elucidate differences in persistence between patients with different types or severity of glaucoma.
In conclusion, we found that persistence rates of patients in a Singapore hospital who began taking IOP-lowering monotherapy were very low after 1 and 3 years. We believe that these results are reflective of persistence of patients with glaucoma in this region and fear that persistence may be even worse in less developed countries in Asia. Prospective cohort studies investigating reasons for lack of persistence, including the effect of the type and severity of glaucoma on persistence, should be carried out to help characterize patients who tend to be nonpersistent. In the long term, there is a need to implement appropriate strategies to improve persistence in patients with glaucoma.
Correspondence: Tin Aung, FRCOphth, PhD, Singapore National Eye Centre, 11 Third Hospital Ave, Singapore 168751 (email@example.com).
Submitted for Publication: August 3, 2010; final revision received October 10, 2010; accepted November 9, 2010.
Published Online: January 10, 2011. doi:10.1001/archophthalmol.2010.345
Financial Disclosure: Dr Aung reports receiving research support from Alcon and Allergan and travel support/honoraria from Alcon, Pfizer, Allergan, Santen, and Merck.
Funding/Support: This study was supported by a grant from the National Medical Research Council, Singapore.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Ophthalmology editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 8
Customize your page view by dragging & repositioning the boxes below.
The Rational Clinical Examination: Evidence-Based Clinical Diagnosis
The Rational Clinical Examination: Evidence-Based Clinical Diagnosis
All results at
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.