ABSTRACT
%2{%Obp' Purpose: To quantify the prevalence of cataract, the outcomes
7 *`h/ of cataract surgery and the factors related to
(Hb:?( unoperated cataract in Australia.
FW G6uKv Methods: Participants were recruited from the Visual
!u[eaLxV Impairment Project: a cluster, stratified sample of more than
r\-uJ~8N 5000 Victorians aged 40 years and over. At examination
|8l<$J sites interviews, clinical examinations and lens photography
m4hg'<<V were performed. Cataract was defined in participants who
b)qoh^ had: had previous cataract surgery, cortical cataract greater
':3pq2{ than 4/16, nuclear greater than Wilmer standard 2, or
jN;@=COi posterior subcapsular greater than 1 mm2.
DpvI[r//'* Results: The participant group comprised 3271 Melbourne
c]n1':FT" residents, 403 Melbourne nursing home residents and 1473
2q=AEv/ rural residents.The weighted rate of any cataract in Victoria
bB1UZ O was 21.5%. The overall weighted rate of prior cataract
NflD/q/ L surgery was 3.79%. Two hundred and forty-nine eyes had
x7!L{(E3 had prior cataract surgery. Of these 249 procedures, 49
( u\._Gwsx (20%) were aphakic, 6 (2.4%) had anterior chamber
;Q OBBF3HG intraocular lenses and 194 (78%) had posterior chamber
-$cmG4 intraocular lenses.Two hundred and eleven of these operated
!Sh&3uy_qN eyes (85%) had best-corrected visual acuity of 6/12 or
Bn#?zI better, the legal requirement for a driver’s license.Twentyseven
ORHp$Un~) (11%) had visual acuity of less than 6/18 (moderate
"4+&-ms vision impairment). Complications of cataract surgery
@s ? caused reduced vision in four of the 27 eyes (15%), or 1.9%
hKLCJ#T of operated eyes. Three of these four eyes had undergone
2Mc3|T4)U intracapsular cataract extraction and the fourth eye had an
Zw
5Ni Xj opaque posterior capsule. No one had bilateral vision
Z<1FSk,[ impairment as a result of cataract surgery. Surprisingly, no
v&Yi particular demographic factors (such as age, gender, rural
A+ZK4]xb residence, occupation, employment status, health insurance
([T>.s status, ethnicity) were related to the presence of unoperated
m? J0i>H
cataract.
u~7hWiY<2 Conclusions: Although the overall prevalence of cataract is
I7 |Pi[e quite high, no particular subgroup is systematically underserviced
ZkRx1S"m in terms of cataract surgery. Overall, the results of
J%v=yBC2 cataract surgery are very good, with the majority of eyes
qT{U( achieving driving vision following cataract extraction.
n$xc];j Key words: cataract extraction, health planning, health
ov`h services accessibility, prevalence
&=Ar INTRODUCTION
bH7X'%r Cataract is the leading cause of blindness worldwide and, in
bf ]f=
;.+ Australia, cataract extractions account for the majority of all
QUq_:t+Dv ophthalmic procedures.1 Over the period 1985–94, the rate
Zd^rNHhA of cataract surgery in Australia was twice as high as would be
H*RC@O_hv expected from the growth in the elderly population.1
zT=Ho
Although there have been a number of studies reporting
]fx"4qKM the prevalence of cataract in various populations,2–6 there is
>BIMi^ little information about determinants of cataract surgery in
[-65PC4aN the population. A previous survey of Australian ophthalmologists
s:>VaGC showed that patient concern and lifestyle, rather
YIn',]p: than visual acuity itself, are the primary factors for referral
yzbx . for cataract surgery.7 This supports prior research which has
+X#vVD3" shown that visual acuity is not a strong predictor of need for
_X^1IaL cataract surgery.8,9 Elsewhere, socioeconomic status has
63q^ $I been shown to be related to cataract surgery rates.10
9TgIB To appropriately plan health care services, information is
CY
4gSe? needed about the prevalence of age-related cataract in the
wj*,U~syB community as well as the factors associated with cataract
<{dVKf,e surgery. The purpose of this study is to quantify the prevalence
4,bv)Im+ ` of any cataract in Australia, to describe the factors
p#ol*m5wE related to unoperated cataract in the community and to
mC4zactv describe the visual outcomes of cataract surgery.
S@jQX METHODS
^!<U_;+ Study population
s[M?as Details about the study methodology for the Visual
fV &KM*W*@ Impairment Project have been published previously.11
#zG&|<hc Briefly, cluster sampling within three strata was employed to
\kp8S'qVo recruit subjects aged 40 years and over to participate.
c&
r70L, Within the Melbourne Statistical Division, nine pairs of
(n*^4@"2 census collector districts were randomly selected. Fourteen
zcel|oz) nursing homes within a 5 km radius of these nine test sites
q$ZHd were randomly chosen to recruit nursing home residents.
3p39`"~ Clinical and Experimental Ophthalmology (2000) 28, 77–82
q4R5<LW" Original Article
whmdcVh. Operated and unoperated cataract in Australia
b=g8eMm Catherine A McCarty PhD, MPH, Mukesh B Nanjan PhD, Hugh R Taylor MD
8qY79)vD4E Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, Melbourne, Victoria, Australia
vL|SY_:4 n Correspondence: Dr Cathy McCarty MPH, Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, 32 Gisborne Street, East Melbourne,
)T/0S$@ Victoria 3002, Australia. Email:
cathy@cera.unimelb.edu.au R{WE\T ' 78 McCarty et al.
C\3y {s Finally, four pairs of census collector districts in four rural
c2h{6;bfY Victorian communities were randomly selected to recruit rural
M2HomO/X) residents. A household census was conducted to identify
$h5xH9x
; eligible residents aged 40 years and over who had been a
a;rdQ> resident at that address for at least 6 months. At the time of
W0y '5` the household census, basic information about age, sex,
I, -hf=- country of birth, language spoken at home, education, use of
;Yx )tWQI corrective spectacles and use of eye care services was collected.
&%8'8,. Eligible residents were then invited to attend a local
:H7D
~ n examination site for a more detailed interview and examination.
ks3`3q 7 The study protocol was approved by the Royal Victorian
&+a9+y
Eye and Ear Hospital Human Research Ethics Committee.
k'gh Assessment of cataract
Ao&\E cIOT A standardized ophthalmic examination was performed after
3;)>Fs; pupil dilatation with one drop of 10% phenylephrine
@AK
n@T5 hydrochloride. Lens opacities were graded clinically at the
NS9B[*"Jl time of the examination and subsequently from photos using
3Vsc 9B"w the Wilmer cataract photo-grading system.12 Cortical and
sSOOXdnGG posterior subcapsular (PSC) opacities were assessed on
l/BLUl~z retroillumination and measured as the proportion (in 1/16)
b^ L
\>3 of pupil circumference occupied by opacity. For this analysis,
$.C=H[QC cortical cataract was defined as 4/16 or greater opacity,
BS /G("oZ[ PSC cataract was defined as opacity equal to or greater than
\[BK1J
P 1 mm2 and nuclear cataract was defined as opacity equal to
h"Xg;(K or greater than Wilmer standard 2,12 independent of visual
_6_IP0; acuity. Examples of the minimum opacities defined as cortical,
p%iGc<vHX nuclear and PSC cataract are presented in Figure 1.
v$~QU{& Bilateral congenital cataracts or cataracts secondary to
2]I4M[|&z intraocular inflammation or trauma were excluded from the
8j Mk
)- analysis. Two cases of bilateral secondary cataract and eight
P6E3-?4j cases of bilateral congenital cataract were excluded from the
<!L>Exh&r analyses.
k xP-,MD A Topcon® SL5 photo slit-lamp (Topcon America Corp.,
(H)2s Y Paramus, NJ, USA) with a 0.1 mm slit beam of 9.0 mm in
wkP#Z"A0~ height set to an incident angle of 30° was used for examinations.
>QA
uEM Ektachrome® 200 ASA colour slide film (Eastman
Ae
mDJ8Y Kodak Company, Rochester, NY, USA) was used to photograph
"nZ*{uv the nuclear opacities. The cortical opacities were
)u3 Zm photographed with an Oxford® retroillumination camera
aJYgzr, (Marcher Enterprises Ltd, Hereford, UK) and T-MAX® 400
n_$
:7J film (Eastman Kodak). Photographs were graded separately
umD!2
w by two research assistants and discrepancies were adjudicated
:/
y1yM by an independent reviewer. Any discrepancies
9U{a{~b between the clinical grades and the photograph grades were
Fkvl%n resolved. Except in cases where photographs were missing,
LAVAFlK5 the photograph grades were used in the analyses. Photograph
6;C3RU] grades were available for 4301 (84%) for cortical
1v,Us5s<"6 cataract, 4147 (81%) for nuclear cataract and 4303 (84%)
7#@cz5Su for PSC cataract. Cataract status was classified according to
cj+ FRG~u the severity of the opacity in the worse eye.
EG0WoUX| Assessment of risk factors
$SM#< @ A standardized questionnaire was used to obtain information
)_{dWf1 about education, employment and ethnic background.11
A6GE,FhsG Specific information was elicited on the occurrence, duration
M(jgd and treatment of a number of medical conditions,
Wi[ ~fI8^! including ocular trauma, arthritis, diabetes, gout, hypertension
~2<7ZtV= and mental illness. Information about the use, dose and
SxdE?uCUS duration of tobacco, alcohol, analgesics and steriods were
lrnyk(M}Q. collected, and a food frequency questionnaire was used to
U"q/rcA determine current consumption of dietary sources of antioxidants
b`)){LR and use of vitamin supplements.
an4GSL Data management and statistical analysis
=:D aS`~V Data were collected either by direct computer entry with a
{LX.iH
9}l questionnaire programmed in Paradox© (Carel Corporation,
]?3un!o3o Ottawa, Canada) with internal consistency checks, or
e_s&L,ze on self-coding forms. Open-ended responses were coded at
Vnx,5E& a later time. Data that were entered on the self-coded forms
RQ'exc2x0 were entered into a computer with double data entry and
`kbSu} reconciliation of any inconsistencies. Data range and consistency
NG checks were performed on the entire data set.
a;QMAd! SAS© version 6.1 (SAS Institute, Cary, North Carolina) was
(4'$y`Z employed for statistical analyses.
A;/-u<f Ninety-five per cent confidence limits around the agespecific
zunV<2~(2} rates were calculated according to Cochran13 to
x%+aKZ(m) account for the effect of the cluster sampling. Ninety-five
2WtRJi?b| per cent confidence limits around age-standardized rates
"*LD 3 were calculated according to Breslow and Day.14 The strataspecific
8>X d2X data were weighted according to the 1996
2
Yp7 Australian Bureau of Statistics census data15 to reflect the
h3*Zfl<] cataract prevalence in the entire Victorian population.
@;<ht c Univariate analyses with Student’s t-tests and chi-squared
DA2}{ tests were first employed to evaluate risk factors for unoperated
@O@GRq&V cataract. Any factors with P < 0.10 were then fitted
+n<k)E@>J into a backwards stepwise logistic regression model. For the
pGf@z:^{*- Figure 1. Minimum levels of cortical (a), nuclear (b), and posterior subcapsular (c) cataract.
" R-!(9k^` final multivariate models, P < 0.05 was considered statistically
]UH`Pdlt significant. Design effect was assessed through the use
P(XaTU&- of cluster-specific models and multivariate models. The
(9<guv design effect was assumed to be additive and an adjustment
K4]g[z made in the variance by adding the variance associated with
_U{zMVr the design effect prior to constructing the 95% confidence
N\HQN0d9 limits.
4O;OjUI0a RESULTS
~mO62(8m Study population
U5-@2YcH A total of 3271 (83%) of the Melbourne residents, 403
J|X
6j&- (90%) Melbourne nursing home residents, and 1473 (92%)
~v'3"k6 rural residents participated. In general, non-participants did
}|5VRJA not differ from participants.16 The study population was
jS+AGE?5e representative of the Victorian population and Australia as
{X[ HCfJd a whole.
Qt,M!i, The Melbourne residents ranged in age from 40 to
C_4)=#@GU 98 years (mean = 59) and 1511 (46%) were male. The
3/b;7\M Melbourne nursing home residents ranged in age from 46 to
WdZ:K,
101 years (mean = 82) and 85 (21%) were men. The rural
\mw(cM#: residents ranged in age from 40 to 103 years (mean = 60)
59zENUYl and 701 (47.5%) were men.
@#P,d5^G
Prevalence of cataract and prior cataract surgery
{5d9$v7k4 As would be expected, the rate of any cataract increases
hUD7_arKF
dramatically with age (Table 1). The weighted rate of any
b:R-mg.VT{ cataract in Victoria was 21.5% (95% CL 18.1, 24.9).
f)g7
3= Although the rates varied somewhat between the three
0vVV%,v strata, they were not significantly different as the 95% confidence
f8SL3+v limits overlapped. The per cent of cataractous eyes
xB@|LtdO9; with best-corrected visual acuity of less than 6/12 was 12.5%
uP<0WCN (65/520) for cortical cataract, 18% for nuclear cataract
HLBkR>e (97/534) and 14.4% (27/187) for PSC cataract. Cataract
f>s?4 surgery also rose dramatically with age. The overall
$t5V=}m> weighted rate of prior cataract surgery in Victoria was
.AYj'Y 3.79% (95% CL 2.97, 4.60) (Table 2).
F q!fWl Risk factors for unoperated cataract
?6nF~9Z' Cases of cataract that had not been removed were classified
$5v0m#[^ as unoperated cataract. Risk factor analyses for unoperated
Qj3a_p$)P cataract were not performed with the nursing home residents
<BO)E( as information about risk factor exposure was not
1Nu1BLPm available for this cohort. The following factors were assessed
gORJWQv in relation to unoperated cataract: age, sex, residence
]r3Kg12Mi (urban/rural), language spoken at home (a measure of ethnic
V97,1` integration), country of birth, parents’ country of birth (a
mKT>,M measure of ethnicity), years since migration, education, use
2kv7UU#q2 of ophthalmic services, use of optometric services, private
PlYm& health insurance status, duration of distance glasses use,
lc_E!"1 glaucoma, age-related maculopathy and employment status.
lpRR& In this cross sectional study it was not possible to assess the
6YuY|JD level of visual acuity that would predict a patient’s having
T~fmk
f$ cataract surgery, as visual acuity data prior to cataract
[]0mX70N surgery were not available.
}|B=h The significant risk factors for unoperated cataract in univariate
+fx8mu
z:y analyses were related to: whether a participant had
Fkf97O
i ever seen an optometrist, seen an ophthalmologist or been
N)S!7%ne diagnosed with glaucoma; and participants’ employment
0wFH!s/B status (currently employed) and age. These significant
k z"3ZDR factors were placed in a backwards stepwise logistic regression
oo=#XZkk model. The factors that remained significantly related
Gn)y>
AN to unoperated cataract were whether participants had ever
'!P"xBVAu seen an ophthalmologist, seen an optometrist and been
mR8W]'gl.L diagnosed with glaucoma. None of the demographic factors
BZb]SoAL were associated with unoperated cataract in the multivariate
!GW,\y model.
r-kMLw/)
The per cent of participants with unoperated cataract
CC;! <km who said that they were dissatisfied or very dissatisfied with
F?m?UQS'u Operated and unoperated cataract in Australia 79
VD4C::J Table 1. Prevalence of any cataracta (excluding previous cataract surgery) by age, gender and cohort
y<XlRTy[} Age group Sex Urban Rural Nursing home Weighted total
Y${l!+q (years) (%) (%) (%)
}j1!j&& 40–49 Male 2.56 4.04 0.00 3.00 (1.91, 4.08)
=2y8CgLj Female 2.61 1.70 0.00 2.36 (1.61, 3.10)
F8e<}v&7R 50–59 Male 7.69 6.94 0.00 7.47 (5.80, 9.15)
-ng=l; Female 6.67 7.56 0.00 6.92 (5.60, 8.24)
>x3ug]Bu 60–69 Male 20.7 25.0 10.0 22.0 (18.1, 26.0)
7&O`p(j Female 27.9 35.7 37.5 30.3 (26.0, 34.7)
%~j2 ('Y 70–79 Male 43.1 58.7 39.1 48.1 (41.2, 55.1)
Ufo>|A
6;$ Female 58.6 66.2 55.6 61.0 (56.0, 65.9)
1\dn1Hh 80–89 Male 74.0 89.3 92.0 79.3 (72.3, 86.3)
yBLUNIr Female 91.9 97.0 80.2 92.6 (86.4, 98.8)
+:6Ii9GN 90 + Male 100.0 100.0 75.0 98.8 (96.2, 100.0)
rZ_>`}O2 Female 100.0 100.0 93.8 98.6 (97.0, 100.0)
5)zn :$cz Age-standardized
+VEU:1Gt (95% CL) Combined 19.7 (16.3, 23.1) 23.2 (16.1, 30.2) 16.5 (2.06, 30.9) 21.5 (18.1, 25.0)
R.j1?\ aCortical > 4/16, nuclear > standard 2, posterior subcapsular > 1 mm2
?p[O%_Xf their current vision was 30% (290/683), compared with 27%
[O-sVYB (26/95) of participants with prior cataract surgery (chisquared,
,]Zp+>{
1 d.f. = 0.25, P = 0.62).
|0bc$ZY: Outcomes of cataract surgery
|!d"*.Q@F Two hundred and forty-nine eyes had undergone prior
tPHS98y cataract surgery. Of these 249 operated eyes, 49 (20%) were
*! :QdWLq left aphakic, 6 (2.4%) had anterior chamber intraocular
7Tf]:4Y" lenses and 194 (78%) had posterior chamber intraocular
TTI81:fku lenses. The rate of capsulotomy in the eyes with intact
i}TwOy<4s posterior capsules was 36% (73/202). Fifteen per cent of
)_jSG5k eyes (17/114) with a clear posterior capsule had bestcorrected
=lr) gj visual acuity of less than 6/12 compared with 43%
-ewQp9)G of eyes (6/14) with opaque capsules, and 15% of eyes
S+x_c4 T (11/73) with prior capsulotomies (chi-squared, 2 d.f. = 7.21,
[JTto!Ih$ P = 0.027).
Uhh
l3%p The percentage of eyes with best-corrected visual acuity
^9"KTZc-* of 6/12 or better was 96% (302/314) for eyes without
Mxz,wfaH> cataract, 88% (1417/1609) for eyes with prevalent cataract
nl/UdgI and 85% (211/249) for eyes with operated cataract (chisquared,
xc'vS>& 2 d.f. = 22.3), P < 0.001). Twenty-seven of the
y@
vj;3: operated eyes (11%) had visual acuities of less than 6/18
v%{0 Tyk (moderate vision impairment) (Fig. 2). A cause of this
+p:Y=>bTj moderate visual impairment (but not the only cause) in four
LhJ a)jFQ (15%) eyes was secondary to cataract surgery. Three of these
utO.WfWP four eyes had undergone intracapsular cataract extraction
gb-{2p>} and the fourth eye had an opaque posterior capsule. No one
XYbyOM VI had bilateral vision impairment as a result of their cataract
:.IN?X surgery.
hCc0sRp DISCUSSION
/NBTvTI To our knowledge, this is the first paper to systematically
4L:>4X[T assess the prevalence of current cataract, previous cataract
z?.(3oLT surgery, predictors of unoperated cataract and the outcomes
)&1!xF of cataract surgery in a population-based sample. The Visual
]EL\)xCr Impairment Project is unique in that the sampling frame and
-Wjh* * high response rate have ensured that the study population is
sB7" 0M representative of Australians aged 40 years and over. Therefore,
y$oW! these data can be used to plan age-related cataract
6M><(1fT services throughout Australia.
Lar r}o= We found the rate of any cataract in those over the age
>BiRk%x of 40 years to be 22%. Although relatively high, this rate is
>A jCl significantly less than was reported in a number of previous
y lL8+7W studies,2,4,6 with the exception of the Casteldaccia Eye
pC^[ [5A Study.5 However, it is difficult to compare rates of cataract
v}`1)BUeF between studies because of different methodologies and
O~'FR[J cataract definitions employed in the various studies, as well
!t-K<' as the different age structures of the study populations.
:EB,{|m Other studies have used less conservative definitions of
%,? vyY cataract, thus leading to higher rates of cataract as defined.
ZgF/;8!~V- In most large epidemiologic studies of cataract, visual acuity
V*qY"[ has not been included in the definition of cataract.
_Y#Bm/* Therefore, the prevalence of cataract may not reflect the
;X7i/DQ actual need for cataract surgery in the community.
HDj$"pS 80 McCarty et al.
on50+)uN Table 2. Prevalence of previous cataract by age, gender and cohort
zPBfiK_hV Age group Gender Urban Rural Nursing home Weighted total
DB8s (years) (%) (%) (%)
UPO^V:.R4 40–49 Male 1.14 0.00 0.00 0.80 (0.00, 1.81)
5B|,S1b Female 0.00 0.00 0.00 0.00 (
u\5g3BH 50–59 Male 0.68 0.58 0.00 0.65 (0.13, 1.17)
{G.jB/ Female 0.57 0.00 0.00 0.41 (0.00, 1.00)
^srs$
w] 60–69 Male 2.12 7.59 0.00 3.80 (1.41, 6.19)
'>]&r