2.2 Prevalence sources

Data sources for the prevalence of VI in the ROI include published studies and the NCBI register of people registered as blind. As demonstrated in Section 2.2.1, there have been few past studies on VI and blindness in the ROI. The best data source for estimating the current prevalence rates and causes of blindness in the ROI is the NCBI register. However, as noted by the Vision Impaired Service Providers Alliance (VISPA) (Jackson et al, 2008), these data should not be used in isolation, in particular due to under-registration of blind people in the ROI (Kelliher et al, 2006).

The ROI does not have a national register of people with partial vision (mild or moderate VI). However, a range of published studies have estimated the prevalence of partial vision loss within communities in the ROI and similar countries using survey data and modelling approaches.

A variety of these sources have been used in this study to model prevalence rates for VI in the ROI.

2.2.1 Past prevalence studies for Ireland

There have been few past studies on VI and blindness in the ROI. The VISPA found only six of 138 studies referencing blindness/VI and Ireland referred to overall prevalence of VI in the ROI or Northern Ireland (Jackson et al, 2008).

A literature search was undertaken for this report to identify studies reporting the prevalence of blindness and VI in the ROI, with key search terms based on VA cut-off scores (e.g. 6/12, 6/18, 6/60, 3/60), and a range of terms for VI, blindness and causes of vision loss. Medical databases searched included Medline, EMBASE and PubMed, and general internet search engines. Due to a paucity of studies identified for the ROI in this search (five studies identified), the search was expanded to include Northern Ireland studies.

In total, only nine studies were identified that reported the prevalence of blindness and/or low vision in communities within the ROI or Northern Ireland. These included the six studies identified by VISPA (Jackson et al, 2008), the VISPA report itself, and two additional studies (O’Donoghue et al, 2010; Donnelly et al, 2005). Five of the studies cover the ROI, and four studies were undertaken in Northern Ireland.

A summary of the nine published studies reporting the prevalence of blindness and/or low vision in communities within the ROI or Northern Ireland is presented in Table 2.2. Please refer to the audio files to listen to the table or to the PDF to view the table.

A comprehensive picture of VI in the ROI cannot be gained from any one study, since each study focuses on a particular age group or other population bracket. For example, four studies (O’Donoghue et al, 2010; Donnelly et al, 2005 Flanagan et al, 2003; Khan et al, 2007) look specifically at VI and blindness in children only. Additionally, comparability of the results between studies is restricted due to different VI and blindness definitions, and differing focus on specific causes and eye conditions.

Of those papers that do reference the overall prevalence of VI and blindness in adults, nearly all draw on data from the blind registers in the ROI and Northern Ireland (Canavan et al, 1997; Kelliher et al, 2006; Munier et al, 1998). The exception is a study by Coffey et al (1993) which studied the prevalence of glaucoma in the west of Ireland. This study recruited 2,186 people aged over 50 years from County Roscommon and derived estimates for the overall prevalence of VI and blindness in the sample, using ophthalmologist-measured BCVA measurements. However, these prevalence rates were not disaggregated by age and gender and their application is therefore limited.

Due to limitations with these past studies, the prevalence of blindness in the ROI was estimated directly from the NCBI’s latest register data as described in Section 2.2.2. The methods for adjusting these data for under-reporting and modelling the prevalence of mild and moderate VI using these data and other sources are described in Section 2.4.

2.2.2 Republic of Ireland blind register

The ROI maintains a national, centralised database of registered blind people in the country through the NCBI. The criteria for registration is legal blindness, defined as an ophthalmologist-measured visual acuity of 6/60 or less in the better eye, or a visual field restricted to 20 degrees or less.

Eligible blind patients are registered with the NCBI by the assessing ophthalmologist or optometrist. Registration is voluntary but entails the incentive of practical and monetary benefits including the Blind Welfare Allowance.

The register is valuable for assessing the prevalence of blindness in the country. However, it is noted that data obtained specifically from national registers of blindness can significantly underestimate the true national prevalence of VI and blindness (Bunce et al, 1998; Kelliher et al, 2006; Robinson et al, 1994).

Previous UK studies have found 45% to 60% non-registration rates amongst those who are eligible for blind or partial sight registration (Barry and Murray, 2005; Bunce et al, 1998; Charles 2007; Robinson et al, 1994). For example, a recent study, Barry and Murray (2005) found 45% of eligible patients were not registered; 28% for blind registration, and 72% for partial sight registration.

Kelliher et al (2006) examined under-registration in the ROI context and found that 21% of eligible blind patients at an outpatient clinic were not appropriately registered. While this is lower than estimates from the English studies, it nonetheless suggests that under-registration is a concern with applying register data to directly estimate blindness prevalence in the ROI.
As with previous English studies, Kelliher et al (2006) also found that people with temporary causes of blindness (i.e. cataract) were more likely to be non-registered. Robinson et al (1994) and Bunce et al (1998) found that patients with permanent diseases undergoing active treatment were more likely to be non-registered. This was not found in the Irish study (Kelliher et al, 2006).

Analysis of current blind register data

Current blind register data (i.e. for 2010) was requested from the NCBI by age group and gender disaggregated by the following registered health conditions (primary causes of blindness):

• AMD;
• cataracts;
• DR;
• glaucoma; and
• other causes.

There were 10,223 registered blind people meeting the blind criteria (VA of <6/60 in the better eye, or visual field of 20 degrees or less) in the ROI in 2010. Approximately 57% of these were female.

An additional 1,909 people were registered blind by an ophthalmologist but had no further details of their VA recorded. Discussions with the NCBI revealed that, historically, people may have been registered without a formal assessment of their VA. Therefore, to accurately estimate the prevalence of blindness according to the definitions used in this study, people without a VA record in the register were excluded from the estimates of blindness prevalence. This produces a conservative estimate of the number of people who are registered blind in the ROI and meet the blindness criteria for this study.

The 10,223 people with recorded visual acuity and visual field data represent approximately 0.22% of the estimated population of the ROI in 2010 (CSO, 2008). This is similar to the estimate of blindness prevalence for adults aged 16 years and older (0.23%) obtained from 2003 blind register data by Kelliher et al (2006).

The age breakdown of registered blind people is presented in Chart 2.1 and is described below. A rising age distribution is noted, with the 90 years plus age group representing the largest proportion (15%) of total registered blind people.

Blindness by cause

There were 10,203 registered blind people having a recorded ‘primary cause of vision loss’ as assessed by an ophthalmologist. The cause breakdown for people on the blind register is presented in Chart 2.2, focusing on those conditions related to the interventions evaluated in Chapter 7 of this report – AMD, cataracts, DR and glaucoma. People whose primary cause of blindness was not coded as any of these four causes were coded as having blindness due to ‘other causes’.

The majority of people on the register (62%) fell into the ‘other causes’ category. As shown below, this proportion is higher for younger age groups, and is consistent with blindness being caused by a wider range of pathologies than just AMD, cataracts, DR and glaucoma (particularly in younger people). Kelliher et al (2006) assessed the three most common ‘other causes’ in 2003 to be retinitis pigmentosa (7% of all blindness), myopia (5% of all blindness), and optic atrophy (4% of all blindness). In total, Kelliher distinguished 15 primary causes of blindness, and included an ‘other causes’ category.

Other reasons for the high proportion of blindness due to ‘other causes’ are: (1) some older people registered many years ago may not have had their cause of blindness recorded at that time; (2) potential miscoding of primary cause in some NCBI register data.

The largest specific cause category was AMD, comprising 24% of registered people with a recorded cause of blindness. Glaucoma and DR were identified as the primary causes for 8% and 4% of registered blindness, respectively. Cataracts comprised the smallest specific cause category (2% of people with recorded cause). These figures for 2010 are similar to Kelliher et al’s (2006) analysis of the NCBI register in 2003, which reported the proportions by primary cause to be: 25% AMD; 12% glaucoma; 5% DR; 4% cataracts; and, hence, 54% other causes. Kelliher noted that glaucoma and cataracts decreased as a proportion of registered blindness between 1996 and 2003. The data presented here suggest these trends have continued.

The figures reported below are less consistent with UK estimates. Access Economics (2009) estimated the proportion of all blindness due to AMD, glaucoma, DR and cataracts to be 51%, 17%, 9% and 13%, respectively. The differences may be partly due to the modelling assumptions in the UK study, where data were taken from a variety of sources (including UK data for the causes of blindness in people aged 75 years and older only) compared with ROI data derived directly from the NCBI register for all ages.

It is useful to analyse cause distributions by age groups to see how causes differ, and how certain causes (i.e. AMD) are primarily age-related or progressive (i.e. DR).

Approximately 97% of blindness for the age group 0-19 years can be attributed to ‘other causes’. Glaucoma and cataracts represent 1.5% and 1.3% of blindness, respectively. For 0.2% of this group their cause of blindness is coded as AMD. This may be due to miscoding including cases of juvenile macular degeneration. There were no people registered with DR as the primary cause of blindness in this age group.

For people aged 20 to 44 years, ‘other causes’ again account for the largest portion of registered blindness at 92% (see Chart 2.3, which is described below). Cataracts comprise 3% of total blindness. Glaucoma and DR form roughly similar shares of total blindness (around 2% each). In this age group, AMD constitutes only 1% of total registered blindness.

For people aged 45 to 59 years, there are slight increases in the proportions of blindness due to AMD, glaucoma and DR relative to the younger age groups (see Chart 2.4, which is described below). ‘Other causes’ continue to account for the largest proportion of blindness (85%), with DR being the largest single cause (5%).

For people aged 60 years or older, there is a notable jump in the proportion of blindness caused by AMD relative to younger age groups (see Chart 2.5, described below and Chart 2.4, described above). AMD is the primary cause of 35% of registered blindness cases in people aged 60 years or older. Glaucoma comprises 11% of total blindness (also a large increase from the younger age groups), with DR and cataracts comprising 4% and 2% of cases, respectively.

These figures for people aged 60 years and older are consistent with UK data that show the proportions of partial sight and blindness (VA < 6/18) by cause to be 36% AMD, 8% glaucoma, and 2% diabetic eye diseases in people aged 75 years and older (Access Economics, 2009). However, the UK data reported 25% of partial sight and blindness being due to cataracts in people aged 75 years and older. This substantially higher proportion may reflect a greater incidence of cataracts in people aged 75 years and older than in people aged 60-74 years.

The results above show that, as expected, the proportions of registered cases of blindness due to AMD, cataracts and glaucoma increase with age. These three eye conditions are all related with aging. The proportion of blind cases due to DR shows less marked increases with age and is 4% for the total population registered blind in the ROI.