Cataract blindness in Paraguay - results of a national survey

Original article
Cataract blindness in Paraguay – results
of a national survey
Ophthalmic Epidemiology
0928-6586/03/US$ 16.00
Rainald Duerksen, MD1
Ophthalmic Epidemiology
Hans Limburg, MD, PhD3
– 2003, Vol. 10, No. 5,
pp. 349–357
Prof. Jorge E. Carron2
© Swets & Zeitlinger 2003
Prof. Allen Foster, FRCS, FRCOphth4
Accepted 29 April 2003
1Programa Vision and 2Facultad de Medicina,
Asuncion, Paraguay, 3Prevention of Blindness and Deafness,
World Health Organization, Geneva, Switzerland, and
4International Centre for Eye Health, Department of Infectious
& Tropical Diseases, London School of Hygiene & Tropical
Medicine, London, U.K.
Abstract
Correspondence and reprint
requests to:
purpose
To estimate the burden of visual loss and blindness due to
Hans Limburg, MD PhD
cataract in people aged 50 years and over in Paraguay.
Nijenburg 32
1613 LC Grootebroek
methods
Forty clusters of 60 persons each who were 50 years and
The Netherlands
older (2400 eligible persons) were selected by systematic random sam-
E-mail: hlimburg@quicknet.nl
pling from the entire population of Paraguay. A total of 2136 persons
were examined (89% coverage).
Acknowledgements:
We are grateful to the World Health
results
For the population 50 years and over, the age- and gender-
Organisation, Prevention of
adjusted prevalence of bilateral blindness (VA < 3/60 with available
Blindness Programme in Washington
correction) was 3.14% (95% CI: 2.2–4.4). The adjusted prevalence of
and Geneva for technical expertise,
bilateral cataract blindness (VA
CBM International for ?nancial
< 3/60) was 2.01% (95% CI: 1.3–3.0),
support, and all the ophthalmologists
making cataract the major cause of bilateral blindness in this age group
and eye care workers who took part
(64%). The adjusted prevalence of bilateral severe visual impairment
in the survey. It is their work and
(VA < 6/60 with available correction) was 5.17% (95% CI: 3.9–6.7)
their data.
and the adjusted prevalence of severe visual impairment due to bilat-
eral cataract (VA < 6/60) was 3.09% (95% CI: 2.2–4.3).
The cataract surgical coverage (persons) was 44% for bilaterally
blind persons with VA < 3/60; 36% for persons with bilateral VA < 6/60;
and 28% for any eye with VA < 6/60 due to cataract.
With IOL implantation, 77% of the operated eyes could see 6/18,
against 46% of the non-IOLs (p < 0.005), a signi?cant better outcome.
conclusion
There is a need to increase the cataract surgical cover-
age in Paraguay. The number of eye surgeons is adequate but the
Cataract blindness in Paraguay
349

---

accessibility of cataract surgical services in rural areas and the afford-
ability of surgery to large sections of society are major constraints.
Key words
Cataract prevalence; cataract surgery; blindness preva-
lence; population survey; Paraguay
Introduction
Age-related cataract remains the major cause of
visual impairment and blindness in most countries. A declining birth
rate and increased life expectancy is resulting in a sharp increase in the
number of people of 50 years and older. In many countries this has
caused an increase in the prevalence of cataract blindness and a greater
demand for cataract surgical services.
Paraguay lies in the centre of South America, sharing borders
with Bolivia (north), Brazil (east) and Argentina (south and west). The
1992 national census of Paraguay reported a total population of
4.153 million people, of which 507,486 people were 50 years or older
(12.22%).1 The population in 1999 is estimated at 5.4 million, giving a
population density of 14 persons per square kilometre. However, most
cities and towns are concentrated in the southern third of the country
while the northern and western two-thirds form the arid wilderness of
the Chaco; 95% of the population is mestizo (mixed Spanish and
Guarani Indians). The population growth is 2.6% annually and life
expectancy at birth 71.4 years for males and 76.5 for females. The lit-
eracy rate, de?ned as all people of 15 years of age or older that com-
pleted at least the second year of primary education, is 90.6% (1995
estimate).1 The per capita GDP in the year 2000 is estimated at US$
4,750. Approximately 36% of the population have an income not
enough to provide for basic healthy food, health services, housing and
education.2
There are an estimated 140 ophthalmologists in Paraguay, an average
of 1 per 40,000 people. Approximately 90% are located in or close to
Asuncion. Most of the ophthalmologists work exclusively in the private
sector. A few ophthalmologists give some of their time to work in the
government or university sectors. There is no training programme for
ophthalmic nurses or optometrists. Primary eye care activities are
implemented through Health Promoters and general medical staff.
Approximately 3500 cataract operations are performed each year in
Paraguay giving an estimated cataract surgical rate of 650 cataract surg-
eries per million population. Assuming that half of the 140 ophthal-
mologists are surgically active, this would mean an average of one
cataract operation per surgeon per week. The proportion of IOL
implants is not known.
Methods and materials
No population-based survey on blind-
ness has been conducted before in Paraguay. Local ophthalmologists
estimate the prevalence of blindness (VA < 3/60 with available cor-
rection) to be around 0.6% in the total population, of which 60% is
estimated to be due to cataract. The prevalence of bilateral cataract
blindness is thus estimated at 0.36%. Assuming that senile cataract only
occurs in people of 50 years and older, the prevalence of cataract blind-
ness in persons 50+ in Paraguay is estimated to be 3.0%.
350
R. Duerksen et al.

---

Because of the scattered rural population and the limited human
resources for ?eldwork, it was decided to examine 40 clusters of 60
persons of 50 years and older. Earlier studies from similar surveys indi-
cated a design effect of 1.7 for a cluster size of 60.3 Allowing for a pre-
cision of ±30%, with a con?dence of 95%, a design effect of 1.7 and
using a systematic cluster sampling methodology, the required sample
size was calculated as 2,340.
Computerised census data from 1992 were used to create a list of set-
tlements and their respective populations, adding a column with the
cumulative population. The total population was then divided by the
number of clusters to get the sampling interval. The ?rst cluster was
selected by multiplying the sampling interval with a random number
between 0 and 1. The resulting number was traced in the cumulative
column and the corresponding settlement was the settlement from
which the ?rst cluster was taken. Each of the following 39 clusters was
found by adding the sampling interval to the previous number. This sys-
tematic random sampling procedure selects clusters with a probability
proportional to the size of the population. This distribution of the clus-
ters is shown on the map (Figure 1).
In each cluster, a two-stage sampling procedure is used. Houses were
selected at random and within each house only men and women aged
50 years and older, residing in that house, were eligible for examina-
tion. The purpose of the study and the examination procedure were
explained to the subjects and verbal consent was obtained before
examination.
For the rapid assessment, a standardised protocol4 was used. For each
eligible person a standardised Survey Record was ?lled with six dif-
ferent sections: General information; Vision and pinhole examination;
Lens examination; History, if not examined; Why cataract operation has
not been done; and Details about cataract operation.
Visual acuity is measured with an ‘E’ chart with 6/18 (20/60) on one
side and 6/60 (20/200) on the other side. The 6/60 is used at 6 or 3 metres
distance. All measurements are taken with available correction in full
daylight. If the VA is less than 6/18 in either eye, pinhole vision is also
taken for each eye.
The WHO de?nes blindness as visual acuity (VA) less than 3/60 in
the better eye with the best possible correction. A VA less than 6/60,
Fig. 1. Cataract eyes with VA < 6/60
per million persons in Paraguay.
Cataract blindness in Paraguay
351

---

but equal to or better than 3/60 in the better eye, is classi?ed as severe
visual impairment and a VA less than 6/18, but equal to or better than
6/60 in the better eye, is classi?ed as visual impairment. Some patients
may have more than one eye disorder causing visual impairment. We
followed the WHO convention to assign the major cause to the dis-
order that is easiest to treat.
After measuring visual acuity, the examinee is taken inside the house,
into a shaded or dark area. There, the lens status is assessed by torch
and binocular loupe and by distant direct ophthalmoscopy at 20–30 cm
distance in semi-dark condition, without dilatation of the pupil. The
lens in each eye is examined and graded as ‘normal lens’, ‘obvious lens
opacity present’, ‘lens absent (aphakia)’, ‘IOL implanted without pos-
terior capsule opaci?cation’ or ‘IOL implanted and posterior capsule
opaci?cation present’. If the lens cannot be seen because of corneal
scarring, phthisis bulbi or other causes, ‘No view of lens’ is marked.
One or two teams of one ophthalmologist, one ophthalmic nurse and
one enumerator can examine one cluster (60 persons) in one day. The
population in each cluster was noti?ed prior to the survey. However, if
after repeated visits an eligible person was still not available, informa-
tion about the subject’s visual status was collected from relatives or
neighbours. Simple ailments were treated by the team. Patients with
cataract, or those who required specialist treatment, were offered free
services and given appointments.
A software programme for data entry and standardised data analy-
sis has been developed in Epi-Info version 6.04. Data was double-
entered to identify any errors. After data entry has been completed, the
user selects the required level of vision (VA < 3/60, VA < 6/60 or VA <
6/18) and then the required analysis report using the menu system. This
report can be printed. The following standard reports can be produced:
prevalence of all cause blindness and visual impairment;
prevalence of cataract blindness and visual impairment;
prevalence of aphakia and pseudophakia;
cataract surgical coverage;
visual outcome of cataract surgery;
barriers to cataract surgery;
age at surgery, place of surgery, type of surgery;
use of glasses, reasons for not using glasses.
Further customised analysis is also possible using the analysis facilities
of Epi-Info.
Results
Of the total sample of 2400 persons aged 50 years and
older, 2136 persons (89%) were physically examined: 987 males and
1149 females. Two-hundred-and-four eligible persons were not at
home, even after repeat visits. Information about their visual status was
obtained from relatives or neighbours, but this anecdotal data was not
included in the analysis. All data from one cluster (60) was lost in trans-
fer to the central data entry facility.
Out of the total 2136 examined persons, 76 (3.56%) were bilaterally
blind (VA < 3/60 with available correction) due to all causes. Of these,
352
R. Duerksen et al.

---

Level of visual acuity
Males
Females
Total
(n = 987)
(n = 1149)
(n = 2136)
Cases
Prev.
Cases
Prev.
Cases
prev. (95% CI)
Blindness (VA < 3/60 with available correction)
All bilateral cases
32
3.24
44
3.83
76
3.56 (2.6–4.8)
Bilateral cataract cases
18
1.82
27
2.35
45
2.11 (1.4–3.1)
Eyes with cataract
62
3.14
94
4.09
156
3.65 (3.0–4.5)
Blindness + Severe visual impairment (VA < 6/60
with available correction)
All bilateral cases
51
5.17
75
6.53
126
5.90 (4.7–7.4)
Bilateral cataract cases
26
2.63
42
3.66
68
3.18 (2.3–4.4)
Eyes with cataract
91
4.61
143
6.22
234
5.48 (4.6–6.5)
Blindness + severe visual impairment + low vision
(VA < 6/18 with available correction)
All bilateral cases
139
14.08
192
16.71
331
15.50 (13.5–17.7)
Bilateral cataract cases
40
4.05
62
5.40
102
4.78 (3.7–6.2)
Eyes with cataract
143
7.24
195
8.49
338
7.91 (6.9–9.1)
Bilateral aphakia
11
1.11
12
1.04
23
1.08 (0.6–1.9)
Unilateral aphakia
20
2.03
27
2.35
47
2.20 (1.5–3.3)
Pseudo(aphakic) eyes
42
2.13
51
2.22
93
2.18 (1.6–2.9)
45 persons (59%) were bilaterally blind due to cataract, giving a preva-
ta b l e 1. Sample prevalence of
blindness and cataract, by gender, in
lence of 2.11%. Detailed results are given in Table 1.
2136 people aged 50 years and over.
The prevalence of blindness (due to cataract as well as to other
causes) increases with age and is usually higher in females. When the
age and gender composition of the sample differed from the actual
population in the survey area, the prevalence rates from the sample
were adjusted to re?ect the true prevalence. The age and gender
adjusted prevalence of all cause bilateral blindness (VA < 3/60 with
available correction) in people of 50 years and older was 3.14% (95%
CI: 2.2–4.4%), an estimated total of 15,958 people (see Table 2). Assum-
ing that blindness (VA < 3/60) under the age of 50 is nil, the prevalence
of all bilateral blindness (VA < 3/60 with available correction) in the
entire population of 5.4 million (estimate 1999, at time of survey) would
be 0.3%, 3000 cases per million population. As other total population-
based surveys suggest that approximately 1 in 10 blind people are
under 50 years of age,5 the prevalence of blindness in the general pop-
ulation (all ages) was likely to have been around 0.33% in 1999.
The adjusted prevalence of bilateral cataract blindness in people of
50 years and older was 2.01% (95% CI: 1.3–3.0%), giving an estimated
total of 10,213 patients or 1900 cases per million. Cataract was respon-
sible for 64% of bilateral blindness. At the level of less than 6/60 vision,
54,816 eyes had cataract, around 10,150 cataract eyes per million
population.
Patients who were blind or severely visually impaired due to cataract
(42) were asked why they had not been operated so far. The major bar-
Cataract blindness in Paraguay
353

---

ta b l e 2. Age- and gender-adjusted results for the 50 years and over population of Paraguay.
Males
Females
Total
(age adjusted)
(age adjusted)
(age- and gender-adjusted)
(n = 245,184)
(n = 262,305)
(n = 507,489)
Cases
Prev.
Cases
Prev.
Cases
Prev. (95% CI)
Blindness (VA < 3/60 with available correction)
All bilateral cases
6,733
2.75
9,225
3.52
15,958
3.14 (2.2–4.4)
Bilateral cataract cases
4,366
1.78
5,847
2.23
10,213
2.01 (1.3–3.0)
Eyes with cataract
15,735
3.21
20,282
3.87
36,016
3.55 (2.9–4.4)
Blindness + Severe visual impairment (VA < 6/60
with available correction)
All bilateral cases
10,514
4.29
15,701
5.99
26,216
5.17 (3.9–6.7)
Bilateral cataract cases
6,012
2.45
9,656
3.68
15,669
3.09 (2.2–4.3)
Eyes with cataract
22,465
4.58
32,351
6.17
54,816
5.40 (4.5–6.4)
Blindness + severe visual impairment + low vision
(VA < 6/18 with available correction)
All bilateral cases
29,231
11.92
41,206
15.71
70,437
13.88 (11.9–16.1)
Bilateral cataract cases
8,907
3.63
14,060
5.36
22,967
4.53 (3.4–5.9)
Eyes with cataract
32,980
6.73
43,276
8.25
76,256
7.51 (6.5–8.7)
Bilateral aphakia
2,362
0.96
2,414
0.92
4,775
0.94 (0.5–1.8)
Unilateral aphakia
4,588
1.87
5,815
2.22
10,403
2.05 (1.3–3.1)
Pseudo(aphakic) eyes
9,311
1.90
10,643
2.03
19,954
1.97 (1.43–2.66)
ta b l e 3. Cataract surgical coverage
CSC – persons (95% CI)
CSC – eyes (95% CI)
(CSC) by persons and eyes in people
aged 50+.
VA < 3/60
44 (29–59)
37 (29–46)
VA < 6/60
36 (24–49)
28 (22–36)
VA < 6/18
32 (22–43)
22 (17–27)
riers reported were “Don’t know where to go” (48%), “can see well,
need not felt” (24%) and “cannot afford” (10%).
By comparing the number of (pseudo)aphakic persons, or eyes, with
the number of cataract blind persons, or eyes, the Cataract Surgical
Coverage could be calculated, the proportion of the cataract blind
people, or eyes, that were provided surgical services.6 This was calcu-
lated by gender and for various levels of visual loss, and gives an idea
of the uptake of cataract surgical services by the population during the
past. Six in ten bilaterally blind persons due to cataract (VA < 3/60)
were not operated in either eye and seven in ten eyes with an acuity
less than 6/60 due to cataract were not operated. The coverage rates
for males were slightly higher, though not signi?cant. The results are
summarised in Table 3.
Visual acuity was measured in all aphakic or pseudophakic eyes in
the sample (Table 4). It is important to realise that these cases include
patients operated recently as well as many years earlier, by skilled as
well as less skilled surgeons under optimal as well as less optimal con-
354
R. Duerksen et al.

---

ta b l e 4. Post-operative visual
Category of
IOL’s
Non-IOL’s
All eyes
acuity in 93 eyes following cataract
visual acuity
surgery, with available correction,
eyes
%
eyes
%
Eyes
%
and by IOL status.
Can see 6/18
43
77
17
46
60
64
<6/18 - 6/60
7
12
1
3
8
9
<6/60
6
11
19
51
25
27
Totals
56
100
37
100
93
100
ditions. Good results from recent surgeries could be outbalanced by
less good results of operations conducted in the past.
With IOL implantation, 77% of the operated eyes could see 6/18,
against 46% of the non-IOLs (p < 0.005), a signi?cantly better outcome.
Results from operations during the previous 5 years were slightly better
than from operations more than 5 years earlier. Unfortunately, there
were no details on possible causes of poor outcome in this study. The
non-IOL’s may also have included uncorrected aphakics, failed IOL’s
and patients who lost vision due to other eye diseases.
Of all patients, 45% were operated on in government hospitals, 25%
in private hospitals and 22% in non-governmental institutions. Treat-
ment was provided free of cost to 20% of the cases, 40% paid part of
the costs and another 40% paid everything.
Discussion
The results represent averages for entire Paraguay.
With a highly unequal distribution of the population as well as cataract
surgical services, and a large proportion of the population with less
access to eye care services, a strati?ed survey would have been more
appropriate. However, this would have increased the sample size con-
siderably, while human resources for this survey were limited.
Rapid Assessment of Cataract Surgical Services (RACSS) is not
intended to replace a detailed blindness survey. It is a simple method
to collect baseline data to facilitate adequate planning of cataract inter-
vention programmes and to evaluate such programmes over time. With
trained staff and adequate resources, the lens can be examined with
portable slit lamp and dilated pupils. This may enhance the accuracy of
the diagnosis.
The RACSS procedure has been used extensively in India7–9 and a
number of other countries like Vietnam, Turkmenistan,10 Cambodia,
Myanmar, Pakistan,11 Mexico, Peru, and Mauretania.
The age- and gender-adjusted prevalence of bilateral cataract blind-
ness in persons of 50 years or older was 3.14% (95% CI: 2.2–4.4%).
Extrapolated to the total population this would give an age- and
gender-adjusted prevalence for all bilateral blindness (VA < 3/60) of
0.30%, assuming all blind people are 50 years or above. As some blind
people are under the age of 50 years, the true prevalence is likely to be
closer to 0.33%.6
For every million people in Paraguay, at least 3000 are bilaterally
blind and nearly 5000 people have a VA < 6/60 in the better eye. Of
Cataract blindness in Paraguay
355

---

these, approximately 3000 people are due to bilateral cataract, and over
10,000 eyes have a VA < 6/60 due to cataract per million population.
The Cataract Surgical Coverage for eyes varies from 37% for VA <
3/60 to 22% for VA < 6/18. This means that for every operated eye,
three to four cataract eyes are in need of surgery. The small difference
in coverage rates suggests that there is no particular threshold applied
for cataract surgery in the country and patients with early cataract are
operated while there is still a considerable backlog of more severe
cataract. It may also suggest that availability, accessibility and afford-
ability of cataract surgery are not yet optimal throughout Paraguay.
Since this survey was not strati?ed, variation in prevalence by region
or socio-economic background could not be determined.
Besides patient-related constraints there are also provider-related
barriers to surgery. It is not clear how much the cost of surgery is a con-
straint. Paraguay is a large country with a low density of population.
Few ophthalmologists work outside Asuncion, so distance and access
to an ophthalmologist for examination and advice are likely to be sig-
ni?cant barriers. With one ophthalmologist per 40,000 people, there
should be suf?cient capacity to increase the output.
Although most eye surgeons are private practitioners, nearly half of
all operations were conducted in government hospitals, where most
surgeries were free or against partial payment only.
The visual outcome after cataract surgery with available correction
is in the same range as in other studies. Experience suggests that these
outcome ?gures could be improved considerably with the provision of
adequate optical correction.
Conclusions
For each million people in Paraguay:
– there are an estimated 10,000 eyes (<6/60) needing cataract surgery
(backlog);
– assuming the incidence of cataract blindness to be 20% of the preva-
lence, the incidence is conservatively estimated at 2000 eyes per year;
– there are 26 ophthalmologists;
– at present, 700 cataract operations are performed per year, of which
approximately 1 in 10 has a poor outcome (<6/60) (see Fig. 2).
It is planned to use these ?ndings in Paraguay to develop a National
VISION 2020 programme, emphasising good quality cataract surgery
with an IOL implant, made accessible to rural communities and afford-
able to all sectors of society.
Fig. 2. Paraguay: 40 clusters of 60
persons each, 50 years of age or
older.
356
R. Duerksen et al.

---

References
intervention programmes. Comm
1 Direccion General de Estadisticas,
Eye Health. 1998;25:3–6.
Encuestas y Censos. Censo Nacional
7 Limburg H, Kumar R. Follow-up
del Paraguay 1992. Fernando de la
study of blindness attributed to
Mora, Paraguay.
cataract in Karnataka state, India.
2 Social Affairs of?ce of the
Ophthalmic Epidemiol. 1998;5:
President’s Of?ce. Draft for
211–223.
consultation and discussion. National
8 Rapid assessment of cataract
Strategy for Reducing Poverty and
blindness in districts of Gujarat.
Inequality. Asuncion, Paraguay,
State Ophthalmic Cell, Government
November 2002.
of Gujarat, India, 1998.
3 Limburg H, Kumar R, Indrayan A,
9 Murthy GVS, Gupta S, Ellwein LB,
Sundaram KR. Rapid assessment of
et al. A population-based eye survey
prevalence of cataract blindness at
of older adults in a rural district
district level. Int J Epidemiol. 1997;
of Rajasthan. I. Central vision
26:1049–1054.
impairment, blindness and cataract
4 World Health Organization. Rapid
surgery. Ophthalmology. 2001;108:
Assessment of Cataract Surgical
679–685.
Services. A package for data entry
10 Amansakhatov A, Volokhovskaya
and analysis from population based
ZP, Afanasyeva AN, Limburg H.
rapid assessments. Geneva: WHO/
Cataract blindness in
PBL/01.84.
Turkmenistan: results of a national
5 Thylefors B, Negrel A-D,
survey. Br J Ophthalmol. 2002;86:
Pararajasegaram R, Dadzie KY.
1207–1210.
Global data on blindness. Geneva:
11 Haider S, Hussain A, Limburg H.
WHO, 1994 [WHO/PBL/94.40].
Cataract blindness in Chakwal
6 Limburg H, Foster A. Cataract
District, Pakistan – results of a
Surgical Coverage: an indicator to
survey. Ophthalmic Epidemiol.
measure impact of cataract
2003;4:249–258.
Cataract blindness in Paraguay
357

---

---