Statistical methods for conducting agreement (comparison of clinical tests) and precision (repeatability or reproducibility) studies in optometry and ophthalmology

McAlinden, Colm; Khadka, Jyoti; Pesudovs, Konrad

doi:10.1111/j.1475-1313.2011.00851.x

Cited by 361 publications

(325 citation statements)

References 48 publications

Supporting

Mentioning

320

Contrasting

Order By: Relevance

“…cant difference in the distribution of test results was found between eyes, only data from the left eye were used for subsequent analyses because averaging the data from both eyes would add little information and could potentially confound correlation and repeatability analyses. 28 For reference values, normality of the data was assessed with the Shapiro-Wilk test. Nonnormally distributed data are presented as median and 95% CR (2.5th to 97.5th percentile).…”

Section: Methodsmentioning

confidence: 99%

Reference values, intertest correlations, and test-retest repeatability of selected tear film tests in healthy cats

Sebbag

Kass

Maggs

2015

javma

View full text Add to dashboard Cite

Objective—To determine reference values, intertest correlations, and test-retest repeatability of Schirmer tear test 1 (STT-1), phenol red thread test (PRTT), tear film breakup time (TFBUT), tear osmolarity, and meibometry in healthy cats. Design—Evaluation study. Animals—135 healthy domestic cats aged 0.5 to 12.8 years. Procedures—Each test was performed once in 120 cats and repeated in 40. Pearson correlation was used to assess correlation among tests. Intraclass correlation coefficients (ICCs) and 95% limits of agreement (LOA) were used to evaluate test-retest repeatability. Results—Median (95% central range) values were 18 mm/min (9 to 34 mm/min) for STT-1, 29 mm/15 s (15 to 37 mm/15 s) for PRTT, 12.4 seconds (9.1 to 17.7 seconds) for TFBUT, 322 mOsm/L (297 to 364 mOsm/L) for osmolarity, and 32 meibometry units (MU; 11 to 114 MU) for peak meibometry value. The STT-1 and PRTT values were positively correlated. Age was weakly associated with TFBUT and osmolarity. Meibometry measurements were higher for strips that contacted the tear film (285 MU) than for those that touched the eyelid margin only (32 MU). All ICCs were < 0.75, and 95% LOA were wide. Conclusions and Clinical Relevance—Tear deficiency should be suspected in cats with STT-1 < 9 mm/min, PRTT < 15 mm/15 s, or TFBUT < 9 to 10 seconds. Generally poor correlation among tests suggested that thorough tear film analysis requires performance of multiple tests in concert. Relatively poor test-retest repeatability should be considered when repeated tests are used to monitor tear film dysfunction and response to treatment.

show abstract

Section: Methodsmentioning

confidence: 99%

Reference values, intertest correlations, and test-retest repeatability of selected tear film tests in healthy cats

Sebbag

Kass

Maggs

2015

javma

View full text Add to dashboard Cite

show abstract

“…A further consideration is bias when two eyes of one patient are used as there is a greater intra-correlation between the two eyes of one patient compared with the inter-correlation between the two eyes of two different patients. 18 Specific bias issues including small sample sizes (power calculation) and confounding factors are discussed below in the review of each trial.…”

Section: Sources Of Biasmentioning

confidence: 99%

Selective laser trabeculoplasty (SLT) vs other treatment modalities for glaucoma: systematic review

McAlinden

2013

Eye

View full text Add to dashboard Cite

Purpose Systemic review to compare selective laser trabeculoplasty (SLT) to other glaucoma treatment options in terms of their intraocular pressure (IOP)-lowering effect. Methods Searches of the following databases were performed: PubMed, Cochrane Central Register of Controlled Trials, Ovid, EMBASE, metaRegister of Controlled Trials, and ClinicalTrials.gov. Only randomised controlled trials (RCTs) published in peer-reviewed journals comparing SLT to other glaucoma treatment options were considered. The main outcome measure was the change in IOP from baseline. Results An initial search of PubMed identified 23 RCTs with 17 meeting the inclusion criteria. Nine RCTs compared 1801 SLT to 1801 argon laser trabeculoplasty (ALT) and one trial compared 3601 SLT to 3601 ALT, all reporting no difference in terms of IOP reduction from baseline. One RCT reported better outcomes with SLT at 1 year but this effect regressed at 2 years. Three trials compared 3601 SLT to medical therapy and found no difference between the two treatment options. One trial found greater IOP reduction with latanoprost vs 901 and 1801 SLT, and greater IOP reduction with 1801 and 3601 SLT versus 901 SLT, however no differences were found between 3601 SLT versus latanoprost or 3601 vs 1801 SLT. Two trials compared 1801 SLT to 3601 SLT finding no difference in IOP reduction. Two trials compared 1801 SLT to 901 SLT, one finding no significant difference and one finding greater IOP reduction with 1801 SLT over 901 SLT. One trial compared excimer laser trabeculotomy (ELT) to 1801 SLT, finding no differences in IOP reduction up to 3 months follow-up but greater IOP reduction with ELT at time intervals between 9 and 24 months. There were no RCTs identified that compared SLT to surgery. Conclusion In terms of the IOP lowering effect, there is no difference between SLT and ALT. Three trials indicate no difference between 3601 SLT and medical therapy, with one of the trials indicating greater IOP reduction with latanoprost over 901 and 1801 SLT. Three trials indicate no difference between 1801 SLT and 3601 SLT. It is inconclusive whether 901 is less efficacious than 1801 SLT. One trial reports greater IOP reduction with ELT over 1801 SLT in the long term.

show abstract

“…To assess the repeatability of repeated measurements, intraclass correlation coefficients (ICC) were calculated 18 . The ICC was defined as the ratio of the betweensubject variance to the sum of the pooled within-subject (S w ) and between-subject variances.…”

Section: Accepted Manuscriptmentioning

confidence: 99%

Choroidal Thickness and Volume in Healthy Young White Adults and the Relationships between them and Axial Length, Ammetropy and Sex

Sánchez‐Cano

Orduna

Segura

et al. 2014

American Journal of Ophthalmology

View full text Add to dashboard Cite

PURPOSE:To evaluate choroidal thickness in young subjects using Enhanced Depth Imaging Spectral Domain Optical Coherence Tomography (EDI SD-OCT) describing volume differences between all the defined areas of the Early Treatment Diabetic Retinopathy Study (ETDRS). DESIGN:Prospective, clinical study. METHODS:Seventy-nine eyes of 95 healthy, young (23.8±3.2years), adult volunteers were prospectively enrolled. Manual choroidal segmentation on a 25-raster horizontal scan protocol was performed. The measurements of the nine subfields defined by the ETDRS were evaluated. RESULTS:Mean subfoveal choroidal thickness was 345.67±81.80µm and mean total choroidal volume was 8.99±1.88mm 3 . Choroidal thickness and volume were higher at the superior and temporal areas compared to inferior and nasal sectors of the same diameter respectively. Strong correlations between subfoveal choroidal thickness and axial length (AL) and myopic refractive error were obtained, r = -0.649, p<0.001 and r = 0.473, p<0.001 respectively. Emmetropic eyes tended to have thicker subfoveal choroidal thickness (381.94±79.88µm versus 307.04±64.91µm) and higher total choroidal volume than myopic eyes (9.80± 1.87mm3 versus 8.14±1.48mm3). The estimation of the variation of the subfoveal choroidal thickness with the AL was -43.84µm/mm. In the myopic group, the variation of the subfoveal choroidal thickness with the myopic refractive error was -10.45µm/D. CONCLUSIONS:This study establishes for the first time a normal database for choroidal thickness and volume in young adults. Axial length, and myopic ammetropy are highly associated with choroidal parameters in healthy subjects. EDI SD-OCT exhibited a high degree of intraobserver and interobserver repeatability. M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT INTRODUCTIONThe development of optical coherence tomography (OCT) technology has revolutionized the diagnostic, monitoring and therapeutic approaches to many retinal diseases. Spectral domain OCT (SD-OCT) offers improved axial resolution (3 µm); by providing 19,000 A-scans per second, it shortens examination times, reducing the eye exposure as well as artifacts 1 . The latest development in OCT technology, sweptsource longer-wavelength OCT (SS-OCT), has a longer-band light source than does the conventional instrumentation (1 µm band light source), providing higher penetration through the retinal pigment epithelium (RPE) and allowing for better visualization of the choroid; however, at the present time, SS-OCT use is limited to research.The role of the choroid in a number of diseases, including central serous chorioretinopathy, high myopia, age-related macular degeneration, choroidal melanoma, and polypoidal choroidal vasculopathy, emphasizes the importance of understanding choroidal structure in ocular disease [2][3][4][5] . Indocyanine green has been the best tool for studying choroidal vasculature; however, it does not provide a quantitative evaluation of the layer; other imaging methods such as echography aid in evaluating the layer, and MRI ...

show abstract

Statistical methods for conducting agreement (comparison of clinical tests) and precision (repeatability or reproducibility) studies in optometry and ophthalmology

Cited by 361 publications

References 48 publications

Reference values, intertest correlations, and test-retest repeatability of selected tear film tests in healthy cats

Reference values, intertest correlations, and test-retest repeatability of selected tear film tests in healthy cats

Selective laser trabeculoplasty (SLT) vs other treatment modalities for glaucoma: systematic review

Choroidal Thickness and Volume in Healthy Young White Adults and the Relationships between them and Axial Length, Ammetropy and Sex

Contact Info

Product

Resources

About