Article review
article review on the article: Cataract visual impairment and quality of life in a Kenyan population.

This is the assignment

1 A4 max

Aim of study

•What questions is this study trying to answer?

Method(s):Give a description of the population(s), measurements and method(s) that have been used?

Why do you think the author(s) used this method(s)?

Answer the following questions:

• Why this population?

• Why this sample size?

• Why these type of measurements?

• this type of data analysis?

Discussion (half A-4):

Answer the following questions:

• What are the advantages of the chosen method(s)?

• What are the possible drawbacks of the chosen method(s)?

• What do you think the reproducibility of the chosen method(s) is? Discuss bias and generalization.

• Do you think the chosen method(s) is valid?

• How can you increase the reliability of the chosen method(s)?

EXTENDED REPORT
Cataract visual impairment and quality of life in a Kenyan
population
Sarah Polack, Hannah Kuper, Wanjiku Mathenge, Astrid Fletcher, Allen Foster
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
See end of article for
authors’ affiliations
. . . . . . . . . . . . . . . . . . . . . . . .
Correspondence to:
S Polack, London School of
Hygiene & Tropical
Medicine, Keppel Street,
London WC1E 7HT, UK;
[email protected]
Accepted 19 January 2007
Published Online First
31 January 2007
. . . . . . . . . . . . . . . . . . . . . . . .
Br J Ophthalmol 2007;91:927–932. doi: 10.1136/bjo.2006.110973
Aims: To evaluate the World Health Organization Prevention of Blindness and Deafness 20-item Visual
Functioning Questionnaire (WHO/PBD VF20), a vision-related quality of life scale, and to describe the
relationship between cataract visual impairment and vision- and generic health-related quality of life, in
people >50 years of age in Nakuru district, Kenya.
Methods: The WHO/PBD VF20 was pilot tested and modified. 196 patients with visual impairment from
cataract and 128 population-based controls without visual impairment from cataract were identified through
a district-wide survey. Additional cases were identified through case finding. Vision- and health-related
quality of life were assessed using the WHO/PBD VF20 scale and EuroQol generic health index (European
Quality of Life Questionnaire (EQ-5D)), respectively. WHO/PBD VF20 was evaluated using standard
psychometric tests, including factor analysis to determine item grouping for summary scores.
Results: The modified WHO/PBD VF20 demonstrated good psychometric properties. Two subscales (general
functioning and psychosocial) and one overall eyesight-rating item were appropriate for these data.
Increased severity of visual impairment in cases was associated with worsening general functioning,
psychosocial and overall eyesight scores (p for trend ,0.001). Cases were more likely to report problems
with EQ-5D descriptive dimensions than controls (p,0.001), and, among cases, increased severity of visual
impairment was associated with worsening self-rated health score.
Conclusion: The modified WHO/PBD VF20 is a valid and reliable scale to assess vision-related quality of life
associated with cataract visual impairment in this Kenyan population. The association between health-related
quality of life and visual impairment reflects the wider implications of cataract for health and well-being,
beyond visual acuity alone.
Cataract is the leading cause of blindness and low vision
worldwide, estimated to be responsible for at least 17.7 of
the 37 million cases of blindness in the world.1 Cataract
extraction is one of the most cost-effective medical interventions,
2 and yet coverage of cataract surgery in low-income
countries remains low.3
Visual impairment and outcomes from sight-restoring
surgery have traditionally been assessed using objective clinical
measures, such as visual acuity (VA). In recent years, however,
there has been increasing recognition of the importance of
assessing patients’ views regarding the impact of medical
conditions and interventions, and quality of life assessment has
gained increasing interest and acceptance.4
Vision-related (disease specific) quality of life (VRQOL)
scales assess patients’ experiences of visual acuity. By contrast,
generic health-related scales are designed to be applicable to a
range of conditions, interventions and populations. The
majority of studies assessing the association between cataract
and quality of life are from high-income countries.5–9 Many
well-validated scales exist for evaluating vision function (VF)
and VRQOL in these settings.10 Fewer scales and studies
exist for low-income settings, and in particular for Africa.11 12
No studies exploring the impact of cataract on wider healthrelated
quality of life in African countries were identified.
Recently, the World Health Organization recommended that
more attention be given to the assessment of VF and VRQOL
in people with visual impairment, and highlighted the
need for cross-cultural methods.13 Based on this, the Indian
VF33 questionnaire (INDVFQ33),14 which was developed
through focus group discussion and psychometric evaluation,
was reviewed to produce a 20-item visual functioning
questionnaire (WHO/PBD VFQ-20). It was recommended that
this scale be validated by field testing,13 but this has not
occurred to date.
This study aimed to evaluate the WHO/PBD VFQ20, a new
VRQOL instrument, and describe the relationship between
cataract visual impairment and vision- and health-related
quality of life, in people >50 years of age in Nakuru district,
Kenya.
METHODS
Study population
This study was conducted in Nakuru district between February
and June 2005 as part of a wider case–control study to evaluate
the impact of cataract surgery on quality of life and poverty. To
estimate the required number of cases (visually impaired from
cataract) and controls (with no visual impairment), sample size
calculations were based on previous findings of a difference of
at least one-third in mean VRQOL.14 15 The power to detect this
difference required a sample of 133 cases and 133 controls, with
an a of 0.01 and 80% power.
Cases were recruited via three methods: a population-based
survey of 3500 adults aged >50 years, using systematic cluster
sampling with probability proportionate to size (82 cases)16;
community-based case detection (65 cases) using the same
cluster sampling procedure; and the first 50 patients attending
the Rift Valley Hospital, Nakuru, Nakuru district, Kenya, who
Abbreviations: EQ-5D, European Quality of Life Questionnaire; EuroQol,
European quality of life; HRQOL, health-related quality of life; INDVFQ33,
Indian VF33 questionnaire; PCA, principal components analysis; QOL,
quality of life; SES, socioeconomic status; VA, visual acuity; VAS, visual
analogue scale; VF, vision function; VRQOL, vision-related quality of life;
WHO/PBD VFQ20, World Health Organization Prevention of Blindness
and Deafness 20-item Visual Functioning Questionnaire
927
www.bjophthalmol.com
met the case definition. This hospital is the main centre for
cataract surgery in Nakuru district and serves people from
across the district. Three different methods were employed
because of logistical and time constraints. Procedures for
ophthalmic examination, case selection criteria and consent
were the same in each.
Ophthalmic examination and case definition
The case definition was people aged >50 years with best
corrected VA ,6/24 in the better eye due to cataract, living in
Nakuru district. All clinical examinations and diagnoses were
made by ophthalmologists. VA was measured with available
correction using a tumbling ‘‘E’’ chart. For each case in the
survey, one or two age- and sex-matched control subjects (VA
>6/18) were randomly selected from the same cluster.
Vision-related quality of life
The WHO/PBD VFQ2013 was translated into Swahili and two
local languages (Kikuyu and Kalenjin), and back-translated by
independent translators, who were asked to comment on the
appropriateness of language used for the target population. A
review was held to discuss differences in the translations and to
modify them accordingly. The scale was pilot tested on 20
patients in the eye unit, Nakuru District Hospital, and small
modifications to the wording of some items were made to
ensure local understanding. One question, ‘‘how much difficulty
do you have in seeing because of glare from bright
lights?’’, caused difficulties for respondents from rural areas
where there was a lack of electricity or car lights. Following
consultation with an ophthalmologist, this question was
removed. For test–retest reliability assessment, the questionnaire
was administered to 20 patients at the eye unit, Nakuru
Hospital in the afternoon, and again the next morning by the
same interviewer.
Health-related quality of life
To assess health-related quality of life, items from the European
Quality of Life Questionnaire (EQ-5D) were used. This scale
was designed by the European quality of life (EuroQol) group to
be brief, simple and practical for use in surveys alongside
disease-specific measures.17 Evidence of validity and reliability
in high- and low-income settings has been shown.18–20 The EQ-
5D includes two components. The first consists of five
descriptive dimensions: mobility, self-care, usual activity,
pain/discomfort and anxiety/depression, each with three
response options: no problem, some problem or extreme
problem. The second is a visual analogue scale (VAS), with
scores ranging from 0 (‘‘worst imaginable health state’’) to 100
(‘‘best imaginable health state’’). Respondents are asked to
indicate on the scale where they rate their ‘‘own health state
today’’. For all study members this scale was described verbally,
enabling those members unable to see the scale to respond. The
same translation procedure described above was used to
translate the EQ-5D. However, due to time constraints, this
was carried out independently from the EuroQol group, and the
versions used in this study have therefore not been approved by
the EuroQol group.
Interviews
Six interviewers were trained for 1 week, and interviews were
observed periodically throughout the study.
Ethical considerations
Informed signed or thumb-printed consent was obtained from
all study subjects. All cases were offered free cataract surgery at
the district hospital. People with visual impairment, but not
eligible to be study cases, were examined and referred to the
district hospital accordingly. Ethical approval for this study was
obtained from the ethics committees of the London School of
Hygiene & Tropical Medicine, London, UK, and the Kenya
Medical Research Institute, Nairobi, Kenya.
Statistical analysis
Visual acuity
For analysis, presenting VA in the better eye with available
correction was grouped into the following categories: normal
vision (>6/18, controls only), moderate visual impairment (,6/
24, >6/60), severe visual impairment (,6/60, >3/60), blind
(,3/60, .PL) and perception of light (PL).
Vision-related quality of life
Validity and reliability of the WHO/PBD VF20 (minus one item)
were evaluated by standard psychometric methods, including
item acceptability, internal consistency, test–retest reliability,
within-scale analyses and analyses against external criteria,
and using thresholds specified by Lamping et al.21 Analyses were
conducted on data from cases only, except for testing the ability
Table 1 Characteristics of cases and controls
Cases,
n (%)
Controls,
n (%)
Age- and sexadjusted
OR (95% CI)
Age (years)
50–59 9 (4.6) 11 (8.6) Baseline
60–69 34 (17.4) 30 (23.4) 1.3 (0.5 to 3.8)
70–79 66 (33.7) 51 (39.8) 1.6 (0.6 to 4.1)
>80 87 (44.4) 36 (28.1) 2.9 (1.1 to 7.8)
Sex
Male 79 (40.3) 51 (39.8) Baseline
Female 117 (59.7) 77 (60.2) 1.0 (0.6 to 1.6)
Education
None 148 (76.7) 74 (58.7) Baseline
Some 45 (23.3) 52 (41.3) 0.4 (0.2 to 0.7)
Literacy
Cannot read 148 (75.9) 65 (50.8) Baseline
Can read 47 (24.1) 63 (49.2) 0.3 (0.1 to 0.5)
Marital status
Single/widowed 110 (56.7) 59 (46.8) Baseline
Married 84 (43.3) 67 (53.2) 0.7 (0.4 to 1.2)
Socioeconomic status
1 (poorest) 57 (29.8) 22 (17.9) Baseline
2 52 (27.2) 27 (22.0) 0.8 (0.4 to 1.6)
3 46 (24.1) 35 (28.5) 0.5 (0.3 to 1.0)
4 (least poor) 36 (18.9) 39 (31.7) 0.4 (0.2 to 0.8)
Visual acuity
>6/18 0 (0.0) 128 (100.0) N/A
,6/24, >6/60 78 (39.8) 0 (0.0)
,6/60, >3/60 41 (20.9) 0 (0.0)
,3/60, .PL 36 (18.4) 0 (0.0)
PL 41 (20.9) 0 (0.0)
Mean (95% CI) Mean (95% CI) p Value
Vision-related quality
of life*
Overall eyesight 3.9 (3.9 to 4.1) 2.1 (2.0–2.3) ,0.001
General
functioning
43.6 (41.5 to 45.8) 17.8 (16.6–19.1) ,0.001
Psychosocial 12.2 (11.4 to 12.9) 5.5 (5.0–6.0) ,0.001
Self-rated health 47.6 (45.1 to 50.1) 59.4 (56.3–62.5) ,0.001
PL, perception of light.
Some data were missing.
*Higher score denotes poorer quality of life.
Higher score denotes better self-rated health.
928 Polack, Kuper, Mathenge, et al
www.bjophthalmol.com
of the scale to differentiate between groups known to be
different, which compared cases and controls. Three subscales
were originally proposed: visual symptoms (3 items), general
functioning (12 items) and psychosocial (4 items), with one
overall eyesight-rating item. As modifications were made, a
rotated exploratory factor analysis was conducted to determine
how items should be grouped for summary scores. Maximum
likelihood estimation was used, and the number of distinct
factors in the scale were taken as those with eigenvalues .1.22
Covariates
Standard sociodemographic data and indicators of socioeconomic
status were collected, as they have been shown to
influence QOL.14 A socioeconomic status (SES) index was
calculated for each household using principal components
analysis (PCA) to determine weights for a list of assets and
wealth indicators using the method of Filmer and Pritchett.23
Variables entered into the PCA included building materials of
the house, ownership of household assets, animal ownership
and education of the head of the household. Assets and wealth
indicators included in the PCA were selected on the basis of
published literature and discussion with local key informants.
The index was created using STATA V.9 and was divided into
quartiles from poorest (lowest SES index) to least poor (highest
SES index).
The associations between QOL measures, VA and socioeconomic
variables were assessed initially using analysis of
variance (WHO/PBD VF20 subscales and EQ5D VAS score) and
x2 (EQ5D dimensions). Multivariate linear or logistic (as
appropriate) regression analyses were conducted using forward
selection of variables, forcing age and gender into the models.
All analyses were conducted using STATA V.9.
Table 2 Internal consistency and skewness values for World Health Organization Prevention
of Blindness and Deafness 20-item Visual Functioning Questionnaire summary scores
Interitem range
(mean)
Item-total range
(mean) Cronbach a Skewness
Overall eyesight – – – 20.59
General functioning 0.33–0.88 (0.70) 0.61–0.93 (0.85) 0.96 20.15
Psychosocial 0.62–0.80 (0.72) 0.82–0.92 (0.89) 0.91 20.28
Pain/discomfort in eye – – – 0.21
Table 3 Forward selection multivariate adjusted linear regression analyses of World Health
Organization Prevention of Blindness and Deafness 20-item Visual Functioning Questionnaire
scores, presenting visual acuity and sociodemographic variables in cases visually impaired
from cataract
Model 1:
overall eyesight
Model 2:
general functioning
Model 3:
psychosocial
Mean (95% CI) Mean (95% CI) Mean (95% CI)
Presenting visual acuity
,6/24, >6/60 3.7 (3.6 to 3.9) 36.5 (33.7 to 39.3) 10.7 (9.4 to 11.8)
,6/60, >3/60 3.7 (3.7 to 4.1) 39.2 (33.6 to 39.3) 10.9 (9.3 to 12.6)
,3/60, .PL 4.0 (3.8 to 4.2) 47.2 (43.1 to 51.3) 14.2 (12.4 to 16.1)
PL 4.7 (4.5 to 4.9) 59.2 (55.3 to 63.1) 14.5 (12.9 to 16.1)
p for trend ,0.001 ,0.001 ,0.001
Age (years)
50–59 4.2 (3.9 to 4.1) 41 (32.2 to 49.8) 11.5 (7.8 to 15.3)
60–69 3.9 (3.7 to 4.1) 40.5 (36.3 to 44.7) 11.1 (9.3 to 12.9)
70–79 4.0 (3.8 to 4.1) 44.9 (41.8 to 47.9) 13.0 (11.7 to 14.3)
>80 4.0 (3.9 to 4.2) 44.4 (41.8 to 47.0) 12.0 (10.9 to 13.2)
p for trend 0.54 0.14 0.47
Sex
Male 3.9 (3.8 to 4.1) 43.8 (40.9–46.6) 12.3 (11.0 to 13.6)
Female 4.0 (3.9 to 4.1) 43.7 (41.4 to 46.0) 12.1 (11.0 to13.1)
SES
1 (poorest) 4.2 (4.0 to 4.4) 49.9 (46.6 to53.1)
2 4.0 (3.8 to 4.2) 43.3 (40.0 to 46.7) –
3 3.9 (3.8 to 4.2) 39.3 (35.7 to 42.9) –
4 (least poor) 3.7 (3.5 to 3.9) 40.1 (36.0 to 44.3) –
p for trend 0.001 ,0.001
Marital status
Married – – 10.6 (9.4 to 11.9)
Widowed/single – – 13.3 (12.2 to 14.4)
Amount of variance explained
by VA (full model)
23.5% (25.6%) 30.0% (37.6%) 7.8% (12.1%)
PL, perception of light; SES, socioeconomic status; VA, visual acuity.
Age and sex were always kept in the model and adjusted means are presented.
–Did not contribute significantly to the model.
Higher score denotes poorer vision-related quality of life..
Cataract visual impairment in a Kenyan population 929
www.bjophthalmol.com
RESULTS
In all, 196 cases and 128 controls were included. Controls were
younger, more likely to have had some education and were in
higher socioeconomic groups than cases (table 1).
Vision-related quality of life
The WHO/PBD VF20 fulfilled most standard psychometric
criteria. The proportion of missing data for each item was ,1%.
Floor and ceiling effects were ,80% for each item and for
summary scores (ie, ,80% people endorsed response categories
at the top and bottom of the scale for each item and for the
summary scores; table 2). Ten items in the general functioning
subscale had inter-item correlations (ie, correlations with other
items in the same subscale) above the maximum criteria of
0.75, which suggests some item redundancy. Two distinct
factors were identified from the factor analysis. All originally
proposed general functioning items, with one visual symptom
item loaded onto factor 1 and all proposed psychosocial items
loaded onto factor 2, suggesting that two subscales, general
functioning and psychosocial, are appropriate for these data.
One visual symptom item, ‘‘because of your eyesight how much
pain and discomfort do you have in your eyes’’, did not load
well onto either factor. As the pain/discomfort item did not
clearly belong to either of the subscales, it was analysed
separately. Good subscale internal consistency was demonstrated
by high Cronbach a and item-total correlations (table 2).
Test–retest correlations were above the acceptable level of 0.80.
Cases had significantly poorer general functioning, psychosocial,
overall eyesight and pain/discomfort scores than
controls, satisfying the known-group differences criteria
(table 1). The scale showed good convergent validity; poorer
VA was associated with poorer mean general functioning,
psychosocial and overall eyesight rating scores (test for trend
p,0.001). The exception was the pain/discomfort item, which
was not significantly associated with visual acuity.
Discriminant validity findings were mixed. Age was not
associated with mean VRQOL scores among cases. However,
cases in the lower SES group had poorer general functioning (p
for trend ,0.001) and self-rated eyesight scores (p=0.04).
Widowed/single cases had poorer psychosocial scores
(p=0.004). Women, widowed/single cases and cases with no
formal education reported poorer pain/discomfort scores. In
Table 4 Response distribution to European Quality of Life Questionnaire domains and
adjusted odds ratios in cases and controls
EQ-5D domain Cases n (%) Controls n (%) Adjusted OR* (95% CI) p Value
Mobility
No problem 65 (33.2) 93 (72.7) Baseline ,0.001
Some problem 118 (60.2) 35 (27.3) 5 (3.0 to 8.2)1
Confined to bed 13 (6.6) 0 (0.0) – –
Self-care
No problems 106 (54.1) 114 (89.1) Baseline ,0.001
Some problems 74 (37.8) 14 (10.9) 6.2 (3.3 to 11.7)1
Unable 16 (8.2) 0 (0.0) – –
Usual activities
No problems 44 (22.5) 93 (72.7) Baseline ,0.001
Some problems 103 (52.6) 33 (25.8) 8.9 (5.2 to 15.1)1
Unable 49 (25.0) 2 (1.6) – –
Pain/discomfort
None 30 (15.3) 47 (36.7) Baseline` ,0.001
Moderate 131 (66.8) 75 (58.6) 2.7 (1.5 to 4.8)
Extreme 35 (17.9) 6 (4.7) 6.8 (2.5 to 18.9)
Anxiety/depression
None 40 (20.4) 64 (50.0) Baseline` ,0.001
Moderate 96 (49.0) 57 (44.5) 2.8 (1.6–4.8)
Extreme 60 (30.6) 7 (5.5) 13.9 (5.4 to 35.9)
EQ-5D, European Quality of Life Questionnaire.
*Odds ratios (ORs) from forward selection logistic regression analysis, with age and sex always included in the model.
Adjusted for age, sex and location.
`Adjusted for age, sex, literacy and location.
1Owing to small cell sizes ‘‘some problem’’ and ‘‘severe problem’’ were combined to calculate ORs for ‘‘any problem’’.
Table 5 A forward selection multivariate linear regression
analysis of self-rated health (from visual analogue scale),
visual acuity and sociodemographic variables
Self-rated health score
Coefficient (95% CI)
Visual acuity
,6/24, >6/60 Baseline
,6/60, >3/60 0.5 (26.2 to 7.2)
,3/60, .PL 26.2 (213.2 to 0.8)
PL 29.6 (216.3 to 3.0)
p for trend 0.001
Age (years)
50–59 Baseline
60–69 4.5 (29.6 to 18.7)
70–79 12.9 (20.8 to 26.6)
>80 7.2 (26.4 to 20.8)
p for trend 0.66
Sex
Male Baseline
Female 1.4 (24.7 to 7.5)
Socioeconomic status
1 (poorest) Baseline
2 8.3 (1.7 to 14.8)
3 8.3 (1.7 to 14.9)
4 (least poor) 10.7 (3.3 to 18.1)
p for trend 0.009
Marital status
Single/widowed Baseline
Married 6.7 (0.4 to 12.8)
930 Polack, Kuper, Mathenge, et al
www.bjophthalmol.com
multivariate analyses (table 3) controlling for visual acuity,
overall eyesight and general functioning scores were worse in
poorer cases, and widowed/single cases were more likely to
have poorer psychosocial scores. There were no significant
multivariate predictors of pain/discomfort in the eyes of the
cases.
Health-related quality of life
All five EQ-5D dimensions discriminated between cases and
controls (table 4). Among cases, adjusting for covariates, poorer
VA was associated with higher odds of reporting any problem
with mobility (p for trend =0.003), self-care (p for trend
,0.001), usual activities (p for trend=0.008) and pain/
discomfort (p for trend =0.01). There was no significant
association with depression.
Self-rated health score was worse in cases than in controls
(p,0.001); (table 1), and mean health score worsened with
increasing severity of visual impairment in cases (table 5).
Widowed/single and cases in the lower SES groups had lower
mean health scores.
DISCUSSION
This study evaluated the WHO/PBD VF20, an instrument
recommended for measuring VRQOL in low-income settings,
in Nakuru district, Kenya. To our knowledge no other studies
have explored the association between visual impairment from
cataract and quality of life in Kenya.
The results show that the modified WHO/PBD VFQ20, with
one overall rating item and two subscales, (general functioning
and psychosocial) is a valid and reliable scale in this Kenyan
setting. One item about glare from glare from bright lights was
removed, but this might have more relevance in other settings.
The item about pain in the eye did not correlate well with other
items in the scale. Cases reported worse pain scores than
controls; however, among the cases, there was no significant
variation in pain score with VA. Pain and discomfort are not
typical symptoms of cataract, but this item might be more
relevant for other ocular morbidities. However, it was not
included in the original INDVFQ33 which was developed on the
basis of 46 focus group discussions exploring patient’s perceptions
about their eye conditions and associated impact on daily
living.24 There was some redundancy in items, suggesting that it
might be possible to shorten this questionnaire further,
although the full questionnaire should be evaluated in other
settings to confirm this.
Cases were more likely to report problems with the EQ-5D
dimensions than controls. Increasing severity of visual impairment
was associated with higher odds of reporting problems
with mobility, self-care, usual activities, and pain/discomfort
and with mean self-rated health score. These findings
correspond to studies from high-income settings,5 and highlight
an impact of visual impairment on wider well-being that is not
necessarily reflected by vision related scales.
Sociodemographic and economic variables influenced
response independently of VA. This is in accordance with other
studies,5 14 and suggests that experiences of visual impairment
may vary according to individual circumstances. In a study in
Hong Kong, Lau et al25 comment that, despite comparable VA
and using the same scale, mean VRQOL scores were better than
those in China and Nepal, and suggest that this may be due to
differences in modern household utilities which facilitate selfcare
activities. A similar reason may explain why cases in the
lower SES groups in Nakuru had worse general functioning
scores. Promotion of surgical services at early stages of cataract
in poor communities should remain a priority. Being widowed/
single increased the social and emotional burden of cataract
visual impairment, as reflected by poorer psychosocial scores
compared with married people. In contrast, but in accordance
with findings from the INDVFQ 33 in India,14 there was no
association between VRQOL scores and age or sex.
This study has its limitations. Three different case recruitment
methods were used. However, all cases were from the
same district and met the same case definition. According to
power calculations, 133 controls were required for the study,
but only 128 were identified. However, the power calculations
were based on very conservative estimates, so the effect of this
is probably minimal. The WHO/PBD VF20 was recommended as
an instrument to assess all ocular morbidities and we focused
only on cataract. Further, we focused only on people aged
>50 years, and the scale might perform differently in other age
groups. However, the original INDVFQ33 was also developed in
people aged >50 years, and, although other eye conditions
were included, was largely dominated by cataract, reflecting the
relative importance of this condition in the Indian setting. Our
results indicated that some items in the Kenyan setting were
not relevant or were redundant. Future studies should look at
the performance of the full WHO/PBD VF20 scale in other
populations or other disease groups. The translation of the EQ-
5D questionnaire was not validated by the EuroQol group,
although standard translation procedures were followed.
Multiple tests of statistical significance for correlated measures
were made using these data. However, analyses were repeated
using the Bonferroni correction and the multivariate analyses
were essentially unchanged.
In this study, evidence of the validity and reliability of a new
scale were shown, and the data suggest that this scale would be
suitable for assessing the outcome of cataract surgery. The
findings add weight to the evidence of disability and poorer
self-perception of own health associated with cataract visual
impairment among people in an African country.
ACKNOWLEDGEMENTS
We thank all the people from Nakuru district who participated in the
survey. We also thank the office manager (Redempta Muibu), the
ophthalmologists (Oscar Onyango, Godfrey Nyaga, Tina Eusebio), the
ophthalmic clinical officers (Maurice Oduoo, Flora Kosgey, Devina
Kisorio and Maina James), the interviewers (Philip Lumula, Nimmo
Gicheru, Marie Anne Cege, Emma Kamau, Mike Kepkembai, Agnes
Maingi and Rose Kagwe), the information officer (Irene Chelagat), the
drivers (Daniel Mutai, John and Sammy) and data entry clerks (Faith
Lumula and Terry). The Rift Valley Provincial Hospital kindly offered us
office space on their premises. The assistance of the Medical Officer of
Health, Nakuru, and the Electoral Commission Office in Nakuru is
acknowledged.
Authors’ affiliations
. . . . . . . . . . . . . . . . . . . . . . .
Sarah Polack, Hannah Kuper, Astrid Fletcher, Allen Foster, London School
of Hygiene & Tropical Medicine, London, UK
Wanjiku Mathenge, Rift Valley Provincial General Hospital, Nakuru Town,
Nakuru, Kenya
Funding: This study was funded by grants from Sight Savers International,
Christian Blind Mission and ORBIS International.
Competing interests: None.
REFERENCES
1 Resnikoff S, Pascolini D, Etya’ale D, et al. Global data on visual impairment in
the year 2002. Bull World Health Organ 2004;82:844–51.
2 Javitt J, Venkataswamy G, Sommer A. The economic and social aspect of
restoring sight. In: Henkind P, eds. ACTA:24th International Congress of
Ophthalmology. New York: Lippincott, 1983:1308–12.
3 Foster A. Vision 2020: the cataract challenge. J Community Eye Health
2000;13:17–19.
4 Ellwein LB, Fletcher A, Negrel AD, et al. Quality of life assessment in blindness
prevention interventions. Int Ophthalmol 1994;18:263–8.
5 Wang JJ, Mitchell P, Smith W. Vision and low self-rated health: the Blue
Mountains Eye Study. Invest Ophthalmol Vis Sci 2000;41:49–54.
Cataract visual impairment in a Kenyan population 931
www.bjophthalmol.com
6 Crabtree HL, Hildreth AJ, O’Connell JE, et al. Measuring visual symptoms in
British cataract patients: the cataract symptom scale. Br J Ophthalmol
1999;83:519–23.
7 Chia EM, Mitchell P, Rochtchina E, et al. Unilateral visual impairment and health
related quality of life: the Blue Mountains Eye Study. Br J Ophthalmol
2003;87:392–5.
8 Chan CWN, Wong JCC, Chan KSK, et al. Evaluation of quality of life in patients
with cataract in Hong Kong. J Cataract Refract Surg 2003;29:1753–60.
9 Lee JE, Fos PJ, Zuniga MA, et al. Assessing quality of life in cataract patients: the
relationship between utility and health-related quality of life measurement. Qual
Life Res 2000;9:1127–35.
10 Massof RW, Rubin GS. Visual function assessment questionnaires. Surv
Ophthalmol 2001;45:531–48.
11 Schemann JF, Leplege A, Keita T, et al. From visual function deficiency to
handicap: measuring visual handicap in Mali. Ophthalmic Epidemiol
2002;9:133–48.
12 van Dijk K, Lewallen S, Chirambo M, et al. Creation and testing of a practical
visual function assessment for use in Africa: correlation with visual acuity,
contrast sensitivity, and near vision in Malawian adults. Br J Ophthalmol
1999;83:792–5.
13 WHO. Consultation on development of standards for characterization of vision
loss and visual functioning. Geneva: World Health Organization, 2003.
14 Gupta SK, Viswanath K, Thulasiraj RD, et al. The Development of the Indian
Vision Function Questionnaire (IND-VFQ): field testing and psychometric
evaluation. Br J Ophthalmol 2005;89:621–7.
15 Fletcher A, Vijaykumar V, Selvaraj S, et al. The Madurai Intraocular Lens Study.
III: Visual functioning and quality of life outcomes, Am J Ophthalmol
1998;125:26–35.
16 Mathenge W, Kuper H, Polack S, et al. Rapid Assessment of Avoidable Blindness
in Nakuru District, Kenya. Ophthalmology 2007;114:599–605.
17 Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol
Group. Ann Med 2001;33:337–43.
18 Brazier J, Jones N, Kind P. Testing the validity of the Euroqol and
comparing it with the SF-36 health survey questionnaire. Qual Life Res
1993;2:169–80.
19 Hurst NP, Kind P, Ruta D, et al. Measuring health-related quality of life in
rheumatoid arthritis: validity, responsiveness and reliability of EuroQol (EQ-5D).
Br J Rheumatol 1997;36:551–9.
20 Tidermark J, Bergstrom G, Svensson O, et al. Responsiveness of the EuroQol (EQ
5-D) and the SF-36 in elderly patients with displaced femoral neck fractures.
Qual Life Res 2003;12:1069–79.
21 Lamping DL, Schroter S, Marquis P, et al. The community-acquired pneumonia
symptom questionnaire: a new, patient-based outcome measure to evaluate
symptoms in patients with community-acquired pneumonia. Chest
2002;122:920–9.
22 Hays RD, Fayers PM. Evaluating multi-item scales. In: Fayers PM, Hays RD, eds.
Assessing quality of life in clinical trials methods and practice. New York: Oxford
University Press, 2005:41–53.
23 Filmer D, Pritchett L. Estimating wealth effects without expenditure data—or tears.
An application to educational enrollment in states of India. Demography
2001;38:115–32.
24 Murthy GVS, Gupta SK, Thulasiraj RD, et al. The development of the Indian
Vision Function Questionnaire (IND-VFQ): questionnaire content. Br J Ophthalmol
2005;89:498–503.
25 Lau J, Michon JJ, Chan WS, et al. Visual acuity and quality of life outcomes in
cataract surgery patients in Hong Kong. Br J Ophthalmol 2002;86:12–17.
Save your favourite articles and useful searches
Use the ‘‘My folders’’ feature to save and organise articles you want to return to quickly—saving
space on your hard drive. You can also save searches, which will save you time. You will only
need to register once for this service, which can be used for this journal or all BMJ Journals,
including the BMJ.
932 Polack, Kuper, Mathenge, et al
www.bjophthalmol.com