Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
Title: Intra- and Inter- Examiner Repeatability of Cycloplegic Retinoscopy Among
Young Children
Running Head: Repeatability of Cycloplegic Retinoscopy
Authors: Sara J McCullough1, Lesley A Doyle1*, Kathryn J Saunders1
1 Biomedical Sciences Research Institute, School of Biomedical Sciences, University
of Ulster, Cromore Road, Coleraine, N. Ireland.
* Corresponding Author
Email: [email protected]
Keywords: vision science, optometry, retinoscopy, cycloplegia
Disclosure: The authors report no conflicts of interest and have no proprietary
interest in any of the materials mentioned in this article.
Word Count: 2102
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Abstract
Purpose To evaluate the intra- and inter-examiner repeatability of cycloplegic retinoscopy in
young children aged 4-5 years old.
Methods Examiner 1 refracted all children in the first sample (n=108); firstly with masked loose
lenses, then using unmasked loose lenses (intra-examiner repeatability). Examiners
1 and 2 refracted all children in the second sample (n=97) using unmasked loose
lenses, blind to the child’s refractive error, presence/magnitude of habitual spectacle
correction and to each other’s findings (inter-examiner repeatability). Refractions
were performed on one eye chosen at random. Mean differences and 95% limits of
agreement (LOAs) and confidence intervals were calculated for intra- and inter-
examiner repeatability of sphere, cylinder and spherical equivalent refraction (SER).
Results Participants had a wide range of refractive errors (-1.50DS to +7.25DS; ≥4.50DC).
Mean differences (95% LOAs) were small for both intra- and inter-examiner
repeatability [Intra:Sphere 0.00D (-0.85, +0.85D), Cyl -0.03D (-0.68, +0.62D), SER -
0.06D (-0.90, +0.78D); Inter:Sphere -0.08D (-0.92, +0.76D), Cyl -0.08D (-0.75,
+0.59D), SER -0.13D (-0.95, +0.69D). A statistically significant proportional bias was
present for intra-examiner repeatability of cylinder (ρ=0.20, p=0.04) and SER
measurement (ρ=0.19, p=0.049). Proportional bias was not present for any other
measure (p>0.12). Examiners agreed on cylinder axis within ±20o in 71% of
refractions where astigmatism of -0.75D or higher was present. 80% of intra- and
inter-examiner measures fell within ±0.50D for spherical and cylindrical components.
Conclusions Differences of ±1.00D and ±0.75D or more for spherical and cylindrical measures
respectively can be considered significant when performing cycloplegic retinoscopy
on young children.
2
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Introduction
Full cycloplegia has been recommended to obtain an accurate refraction in children
under 12 years of age1 particularly in the presence of suspected strabismus
(esotropia), latent hyperopia or where there is poor fixation or cooperation from the
child. Repeatability and validity of the measurement of refractive error using
cycloplegic autorefraction in children is well established2-6 and the agreement
between cycloplegic retinoscopy and non-cycloplegic refraction techniques has been
investigated.7-9 Zadnik et al.4 and Walline et al.10 measured intra-examiner
repeatability of cycloplegic retinoscopy in a small sample of healthy adults (n=40, 20
to 43 years old) but to date no published reports exist detailing intra- and inter-
examiner repeatability of cycloplegic retinoscopy in the population in which its use is
regarded as gold-standard. An appreciation of the repeatability of cycloplegic
retinoscopy is necessary in clinical practice and for epidemiological studies of
refractive error in order to determine a ‘real’ change in refractive error between
clinicians or over time. The purpose of the present study is to evaluate the intra- and
inter-examiner repeatability of refraction by cycloplegic retinoscopy in a group of
young children aged 4-5 years old.
Methods
Participants
A total of 198 children in their first year of formal education within mainstream
schools in Northern Ireland (Kindergarten equivalent) were recruited for the study.
Participants were recruited from state primary schools that were non-selective in
academic ability and drew children from a range of socioeconomic backgrounds and
rural/urban environments. Intra-examiner repeatability was conducted in one sample
of children (n=108) and inter-examiner repeatability conducted in a second sample
(n=97). Seven children were included in both samples. Written informed consent
was obtained from the parents/guardians of the participants and verbal assent was
given by the participant on the day data collection took place. Data collection took
place on school premises during school time.
3
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
Ethical Approval
The study was approved by University of Ulster Research Ethics Committee and the
conduct of the study adhered to the tenets of the Declaration of Helsinki.
Procedures
The magnitude of spherical and cylindrical refractive error was assessed using
streak retinoscopy (Keeler Professional) by two experienced optometrists (Examiner
1-SJM, Examiner 2-KJS) at least 30 minutes after the instillation of one drop of 0.5%
proxymetacaine hydrochloride and one drop of 1.0% cyclopentolate hydrochloride in
each eye. Retinoscopy was performed on one eye chosen in a pseudorandom
fashion. Right eyes were measured using the right eye of the examiner holding the
retinoscope with their right hand and left eyes measured with the left eye of the
examiner holding the retinoscope in the left hand.11 Neutrality of the retinoscopy
reflex was achieved using a combination of spherical and minus cylindrical lenses
placed in a trial frame (Keeler Oculus Universal). Retinoscopy was carried out in a
darkened room while the child fixated on the retinoscope reflex. A ‘working’ distance
of 67cm from the retinoscope to the trial frame was maintained using a fixed string
attached to the retinoscope. The ‘working’ distance was checked using the string at
the start of the procedure, intermittently throughout and at the end of the procedure
when the examiner achieved neutralisation.
Intra-examiner repeatability
In the assessment of intra-examiner repeatability, Examiner 1 (SJM) refracted all
children within the first sample (n=108) firstly using masked loose lenses in
0.25DS/0.25DC steps held within a turntable and subsequently with unmasked loose
lenses in 0.25DS/0.25DC steps held within a trial lens case. The masked lenses
were marked with a number identifier and were decoded after data collection was
complete.
4
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
Inter-examiner repeatability
In the assessment of inter-examiner repeatability Examiners 1 (SJM) and 2 (KJS)
refracted all children within the second sample (n=97) using unmasked loose lenses
in 0.25DS/0.25DC steps from a trial lens case and were blind to each other’s
findings.
Statistical Analysis
Statistical analyses were performed using Stata 13.0 (StataCorp, College Station,
TX, USA). Spherical equivalent refraction (SER) was calculated using [Sphere+
(Cylinder/2)]. Mean differences, 95% limits of agreement (LOA’s) and confidence
intervals (CI’s) were calculated for the SER and the spherical and cylindrical
components. LOA’s were calculated as ‘mean difference ± (1.96 X standard
deviation)’ and have been used to allow for Bland Altman analysis and direct
comparison to previous studies.4,10 CI’s were calculated as ‘limit of agreement ± (1.66
X standard error)‘ and have also been included as recommended by McAlinden et
al.19 when reporting on the study of agreement and precision. While the spherical
and cylindrical components and cylindrical axis are most relevant to clinical practice,
SER and vector components (J0 and J45) have also been included in analysis. SER is
widely reported in epidemiological studies of refractive error12-14 and the use of vector
analysis allows for the consideration of both the magnitude and direction of
astigmatism. Bland and Altman15 plots were also used to inspect the intra- and inter-
examiner repeatability. Wilcoxon matched-pairs signed rank tests were used to
determine statistically significant intra- and inter-examiner differences. Spearman’s
correlations were used to identify proportional bias. A p value of less than 0.05 was
considered statistically significant.
Results
Intra-Examiner Group
5
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
Participants were 50 males (46%) and 58 females (54%), with a mean age of
5.1±0.37 years (range 4.0 to 5.8 years). The majority of participants were white
(n=100, 93%) consistent with the demographics of the Northern Irish population.16 A
total of 56 right (52%) and 52 left eyes (48%) were examined for intra-examiner
repeatability.
The spherical and cylindrical components of refractive error, SER and vector
components determined by both Examiner 1 and Examiner 2 were not normally
distributed. Table 1. details the median, inter-quartile range and range for the
spherical, cylinder and vector components as determined by Examiner 1 for intra-
examiner repeatability.
Inter-Examiner Group
Participants were 44 males (45%) and 53 females (55%), with a mean age of
5.2±0.37 years (range 4.4 to 5.9 years). The majority of participants were white
(n=93, 96%). A total of 50 right (52%) and 47 left eyes (48%) were examined for
inter-examiner repeatability.
Table 2. details the median, inter-quartile range and range for the spherical, cylinder
and vector components as determined by Examiners 1 and 2 for inter-examiner
repeatability.
Intra-Examiner and Inter-Examiner Repeatability
Mean differences, standard deviations (SD), 95% limits of agreement (LOA’s) and
confidence intervals (CI’s) for intra-examiner and inter-examiner repeatability are
detailed in Table 3 and 4. The percentage of measures falling within ±0.25D and
±0.50D are also reported.
6
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
Figures 1 (A, B & C) & 2 (A, B & C) show the Bland and Altman plots for intra-
examiner and inter-examiner repeatability for the spherical and cylindrical
components and SER respectively. There was no statistically significant
proportional bias for the intra-examiner repeatability for the spherical component
(ρ=0.11, p=0.26). A statistically significant proportional bias was found for the intra-
examiner repeatability of the cylindrical component (ρ=0.20, p=0.04) and the SER
(ρ=0.19, p=0.049). No statistically significant proportional bias was found for inter-
examiner repeatability for the spherical and cylindrical components and the SER (all
Spearman correlations, p>0.12).
Inter-examiner Repeatability of Cylinder Axis
For those participants judged to have an astigmatic error of -0.75DC or higher by
either examiner [i.e. those cylinders likely to have a significant impact on visual
acuity17] (n=18), agreement of cylinder axis was within ±10o for 53% and within ±20o
for 71%.
Discussion
This is the first study to report the intra-examiner and inter-examiner repeatability of
cycloplegic retinoscopy in young children over a wide range of refractive errors
representative of the population at this age.18 Our results show that mean differences
between intra-examiner repeat measures for sphere, cylinder and spherical
equivalent refraction (SER) were small (<0.07D), representing less than one step in
the clinical procedure (0.25D) and showed no statistically significant differences.
There was no statistically significant difference between two examiners’ spherical
component measures and the mean inter-examiner difference was again small (-
0.08D) and represented less than one clinical step (0.25D). Statistically significant
differences were found between the two examiners when cylindrical component and
the SER were considered. On average, Examiner 1 over-estimated the magnitude of
the cylindrical component (mean difference= -0.08, 95% limits of agreement= -0.75
to +0.59) when compared to Examiner 2, which in turn resulted in a less positive
SER (mean difference= -0.13, 95% limits of agreement= -0.95 to +0.69). The limits
7
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
of agreements for both intra- and inter-examiner repeatability of cycloplegic
retinoscopy indicate that differences between repeat measures and between
examiners of ±1.00D or more for sphere and ±0.75D or more for cylinder can be
considered significant and denote a ‘real’ change in refractive error when using
cycloplegic retinoscopy in children. These findings are comparable to data available
for the repeatability of cycloplegic retinoscopy conducted in adults. Zadnik et al. 4
and Walline et al.10 reported intra-examiner mean differences (95% limits of
agreement) for the measurement of the spherical and cylindrical components as
0.075D (-0.87 to 1.02D) and 0.06D (-0.69 to 0.80D) respectively in a group of
healthy, young adults using cycloplegic retinoscopy. Our results show approximately
60% of both intra- and inter-examiner measures for sphere and spherical equivalent
refraction were within ±0.25D, increasing to over 80% within ±0.50D. Almost 80% of
intra-examiner measures and 70% of inter-examiner measures were within ±0.25D
for the cylindrical component increasing to almost 95% within ±0.50D for both intra-
and inter-examiner measures.
When investigating agreement, there are two potential sources of disagreement
between methods; fixed and proportional bias. Fixed bias occurs when one method
produces measures that are consistently higher or lower than the other by a constant
or fixed amount. Whereas proportional bias occurs when one method produces
values that are higher or lower than the other by an amount that is proportional to the
magnitude of the measured variable.18,19 A statistically significant proportional bias
was found for the cylindrical component and spherical equivalent refraction when
intra-examiner measures were compared. This bias was not evident in the spherical
component measures. Examiner 1 under-estimated the cylindrical component and
the level of hyperopic spherical equivalent refraction when using unmasked lenses
compared to when using masked lenses. The examiner may have had pre-
conceptions about the refractive error of the child when not masked to the
measurement which resulted in under-estimation when compared to the masked
results. There were few children within the study who were current spectacle
wearers (intra-examiner repeatability group n=9, inter-examiner repeatability group
n=8) and neither examiner was aware whether children habitually wore spectacles
when conducting retinoscopy. Few studies assessing repeatability of cycloplegic
8
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
refraction have reported on proportional bias; however inspection of Bland and
Altman plots from studies of the repeatability or reproducibility of cycloplegic
subjective refraction or autorefraction show increasing variability with increasing
ametropia.5,6 The typical procedure of cycloplegic retinoscopy in clinical practice is
better represented by the inter-examiner methodology of the present study, where
masked lenses were not used and where no proportional bias was found. These
data may be a more accurate reflection of the repeatability of cycloplegic retinoscopy
in a typical clinical setting.
Our findings show relatively good agreement between examiners for cylindrical axes
with 71% of recorded axes agreeing to within ±20o where astigmatism of -0.75DC or
more was present. The inter-examiner agreement in cylinder axis (53% within ±10o)
within the present study are similar to those reported by Walline et al.10 for repeat
measures by the same examiner using cycloplegic retinoscopy on adults; their
repeat measures agreed to within ±10o in 64% of astigmatic errors of -0.75DC or
higher.10
Strengths and Limitations
The data contain few myopic refractive errors which is reflective of the refractive
error distribution of young children of white ethnicity,14,20 however a wide range of
hyperopic and astigmatic errors were sampled.
Repeatability of cylindrical axis was assessed only between examiners and not
between repeat measures by the same examiner as it was not possible to
satisfactorily mask this component of the retinoscopy procedure in the latter
condition.
Both examiners were paediatric specialists, experienced in carrying out cycloplegic
retinoscopy in children. Whilst the findings of the present study may not be
generalizable to all clinicians, the limits of agreement reported for both intra- and
9
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
inter-examiner repeatability may be considered the ‘minimum’ difference required
before a true change in refractive error using cycloplegic retinoscopy in children can
be determined.
Conclusions
The 95% limits of agreement reported within the present study are useful for
clinicians and researchers wishing to identify ‘real’ refractive change between
measures made by the same examiner over time and/or between measures made
by different examiners. Differences of ±1.00D and ±0.75D or more for spherical and
cylindrical measures respectively can be considered significant when performing
cycloplegic retinoscopy on children aged 4-5 years. Over 80% of intra- and inter-
examiner measures of cycloplegic retinoscopy within the present study fell within
±0.50D for both spherical and cylindrical components.
Acknowledgements
This work was supported by a research grant from the College of Optometrists
(London, UK).
References
1. The Royal College of Ophthalmologists. Guidelines for the Management of
Strabismus in Childhood, 2012. Available from: https://www.rcophth.ac.uk/wp-
content/uploads/2014/12/2012-SCI-250-Guidelines-for-Management-of-Strabismus-
in-Childhood-2012.pdf
2. Harvey EM, Miller JM, Wagner LK, Dobson V. Reproducibility and accuracy of
measurements with a hand held autorefractor in children. Br J Ophthalmol.
1997;81:941-8.
10
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
3. Chat SW, Edwards MH. Clinical evaluation of the Shin-Nippon SRW-5000
autorefractor in children. Ophthalmic Physiol Opt. 2001;21:87-100.
4. Zadnik K, Mutti DO, Adams AJ. The repeatability of measurement of the ocular
components. Invest Ophthalmol Vis Sci. 1992;33:2325-33.
5. Steele G, Ireland D, Block S. Cycloplegic autorefraction results in pre-school
children using the Nikon Retinomax Plus and the Welch Allyn SureSight. Optom Vis
Sci. 2003;80:573-7.
6. Choong YF, Chen AH, Goh PP. A comparison of autorefraction and subjective
refraction with and without cycloplegia in primary school children. Am J Ophthalmol.
2006;142:68-74.
7. Saunders KJ, Westall CA. Comparison between near retinoscopy and cycloplegic
retinoscopy in the refraction of infants and children. Optom Vis Sci 1992;69(8):615-
22.
8. Chan OY, Edwards M. Comparison of cycloplegic and noncycloplegic retinoscopy
in Chinese pre-school children. Optom Vis Sci 1994;71(5):312-8.
9. Ozdemir O, Özen Tunay Z, Petriҫli IS, Ergintürk Acar D, Erol MK. Comparison of
non-cycloplegic photorefraction, cycloplegic photorefraction and cycloplegic
retinoscopy in children. Int J Ophthalmol. 2015;8:128-31.
10. Walline J, Kinney K, Zadnik K, Mutti DO. Repeatability and validity of
astigmatism measurements. J Refract Surg. 1999;15:23-31.
11. Safir A, Hyams L, Philpot J, Jagerman LS. Studies in refraction. I. The precision
of retinoscopy. Arch Ophthalmol. 1970;84: 49-61.
12. Negrel AD, Maul E, Pokharel GP, Zhao J, Ellwein LB. Refractive error study in
children, sampling and measurement methods for a multi-country survey, Am J
Ophthalmol. 2000;129:421-426.
11
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
13. Ojaimi E, Rose KA, Smith W, Morgan IG, Martin FJ, Mitchell P. Methods for a
population-based study of myopia and other eye conditions in school children: the
Sydney Myopia Study. Ophthalmic Epidemiol. 2005;12:59-69.
14. O’Donoghue L, McClelland JF, Logan NS, Rudnicka AR, Owen CG, Saunders
KJ. Refractive error and visual impairment in school children in Northern Ireland. Br J
Ophthalmol. 2010;94:1155-9.
15. Bland JM, Altman DG. Statistical methods for assessing agreement between two
methods of clinical measurement. Lancet. 1986;1:307-310.
16. Northern Ireland Statistics and Research Agency [Internet], Census 2011, Key
Statistics for Northern Ireland [cited 20th August 15] Available from
http://www.nisra.gov.uk/Census/key_stats_bulletin_2011.pdf
17. Sandhu RK, Munoz BE, Swenor BK, West SK. Refractive error and visual
function difficulty in a Latino population. Ophthalmology. 2012;119:1731-1736.
18. Ludbrook J. Comparing methods of measurement. Clin Exp Pharmacol Physiol.
1997;24:193-203.
19. McAlinden C, Khadka J, Pseudovs K. Statistical methods for conducting
agreement (comparison of clinical tests) and precision (repeatability or
reproducibility) studies in optometry and ophthalmology. Ophthalmic Physiol Opt.
2011;31:330-8.
20. Giordano L, Friedman DS, Repka MX, et al. Prevalence of refractive error among
preschool children in an urban population: The Baltimore Pediatric Eye Disease
Study. Ophthalmology. 2009;116:739-746.
12
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
Tables Table 1. Median, inter-quartile range (IQR) and range for the spherical, cylindrical
and vector components of refractive error for intra-examiner repeatability masked
and unmasked measurements.
Refractive Error Intra-Examiner GroupMasked Measurement
Intra-Examiner GroupUnmasked Measurement
Median Sphere(IQR)
[Range]
+1.50D(+1.25 to +2.00D)[-1.00 to +7.00D]
+1.50D(+1.13 to +2.12D)[-1.50 to +6.00D]
Median Cylinder(IQR)
[Range]
-0.25DC(-0.75 to 0.00DC)[-4.50 to 0.00DC]
-0.50DC(-0.50 to 0.00DC)[-3.50 to 0.00DC]
Median SER(IQR)
[Range]
+1.38D(+0.88 to +1.81D)[-1.50 to +6.63D)
+1.25D(+1.00 to +1.81D)[-1.75 to +6.00D]
Median J0
(IQR)[Range]
0.10(-0.13 to 0.29)[-0.63 to 2.11]
0.13(-0.17 to 0.25)[-0.47 to 1.53]
Median J45
(IQR)[Range]
-0.01(-0.01 to 0.01)[-0.77 to 1.29]
-0.01(-0.06 to 0.01)[-0.56 to 1.64]
13
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
Table 2. Medians, inter-quartile ranges (IQR) and ranges for the spherical,
cylindrical and vector components of refractive error for inter-examiner repeatability
Examiner 1 and 2 measurements.
Refractive Error Inter-Examiner GroupExaminer 1 Measurement
Inter-Examiner Group Examiner 2 Measurement
Median Sphere(IQR)
[Range]
+1.50D(+1.00 to +2.50D)[-1.50 to +7.25D]
+1.50D(+1.00 to +2.50D)[-1.00 to +8.00D]
Median Cylinder(IQR)
[Range]
-0.25DC(-0.50 to 0.00DC)[-3.50 to 0.00DC]
-0.25DC(-0.50 to 0.00DC)[-3.50 to 0.00DC]
Median SER(IQR)
[Range]
+1.50D(+0.75 to +2.25D)[-1.75 to +6.75D)
+1.50D(+1.00 to +2.25D)[-1.38 to +7.25D]
Median J0
(IQR)[Range]
0.13(-0.13 to 0.25)[-0.75 to 1.41]
0.13(-0.13 to 0.25)[-0.70 to 1.64]
Median J45
(IQR)[Range]
-0.01(-0.01 to 0.01)[-0.49 to 1.64]
-0.01(-0.01 to 0.01)[-0.60 to 1.15]
14
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
Table 3. Mean differences, standard deviations (SD), 95% limits of agreement
(LOA’s) and 95% confidence intervals (CI’s) for the spherical, cylindrical and vector
components of refractive error for intra-examiner repeatability and the percentage of
repeat measures falling within ±0.25D and ±0.50D.
INTRA-EXAMINER REPEATABILITYRefractive
ErrorMean
Difference (SD)(D)
95% LOA’s
(D)
95% CI’s(D)
Statistically significant difference?
Within ±0.25D
Within ±0.50D
Sphere 0.00(0.43)
-0.85 to +0.85
-0.99 to 0.99
Noz=-0.194, p=0.846
66% 86%
Cylinder -0.03(0.33)
-0.68 to +0.62
-0.78 to +0.73
Noz=-1.240, p=0.215
79% 94%
SER -0.06(0.43)
-0.90 to +0.78
-1.04 to +0.92
Noz=-0.672, p=0.501
59% 81%
J0 0.04(0.21)
-0.37 to +0.45
-0.41 to 0.49
Noz=-0.462,p=0.644
J45 -0.02(0.11)
-0.24 to +0.20
-0.26 to 0.22
Noz=-0.077,p=0.939
15
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
Table 4. Mean differences, standard deviations (SD), 95% limits of agreement
(LOA’s) and 95% confidence intervals (CI’s) for the spherical, cylindrical and vector
components of refractive error for inter-examiner repeatability and the percentage of
between examiner measures falling within ±0.25D and ±0.50D.
INTER-EXAMINER REPEATABILITYRefractive
ErrorMean
Difference(SD)(D)
95% LOA’s
(D)
95% CI’s(D)
Statistically significant difference?
Within ±0.25D
Within ±0.50D
Sphere -0.08(0.43)
-0.92 to +0.76
-1.07 to +0.91
Noz=-1.724, p=0.085
64% 89%
Cylinder -0.08(0.34)
-0.75 to +0.59
-0.87 to +0.71
Yesz=-2.011, p=0.044
67% 94%
SER -0.13(0.42)
-0.95 to +0.69
-1.10 to +0.84
Yesz=-3.257, p=0.001
59% 89%
J0 -0.02(0.25)
-0.51 to +0.47
-0.58 to 0.54
Noz=0.282,p=0.778
J45 0.10(0.33)
-0.55 to +0.75
-0.64 to 0.84
Noz=0.858p=0.391
16
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
Figures
17
436
437
438
439
440
Figure 1. Bland and Altman plots for intra-examiner repeatability of (A) spherical
component, (B) cylindrical component and (C) spherical equivalent refraction. The
18
441
442443
444
solid black line represents the mean difference and the dashed black lines represent
the 95% limits of agreement. The shaded grey areas encapsulate intra-examiner
differences of less than or equal to ±0.50D.
19
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
Figure 2. Bland and Altman plots for inter-examiner repeatability of (A) spherical
component, (B) cylindrical component and (C) spherical equivalent refraction. The
20
461
462463
464
solid black line represents the mean difference and the dashed black lines represent
the 95% limits of agreement. The shaded grey areas encapsulate inter-examiner
differences of less than or equal to ±0.50D.
21
465
466
467
468
469