# The Prompt
my opthalmic consultant friend is producing a statistical analysis.  bullet points below share the information she has given me about it. comments from myself are in ()

* in a statistical analysis we have included both eyes of the same patient, at different times over 2 years. 
(it is not quite clear what this means specifically with respect to data collection)
* One of the reviewers of the paper said that we should have done the statistics to include inter- patient variability. 
(by inter-patient variability i am inclined to believe there is a concern regarding using data points for each eye of a patient as 2 sources of independent data....  but this is a guess on my part.)
* although eyes behave similarly, the disease is usually asymmetrical. 
(my interpretation of the information so far is she has a measurement for left eye taken periodically over 2 years. Ditto for right eye. These values say something about the disease. The disease is asymmetrical so only one eye will show changes in data if disease present inpatient.) 
* they have analysed 126 patients.  26 patients they used both eyes to contribute to the dataset.  in total that is 152 data points.
* Continuous variables are reported as mean ± SD or median (IQR) as appropriate; categorical variables as counts (percentages). 
* Analysis of variance (ANOVA) followed by the Tukey test for multiple comparisons was performed for continuous data. 
* Student's t-test was used in the analysis of paired means. 
* Binomial data were analysed using McNemar's test. 
* Two-sided P < 0.05 was considered statistically significant. No formal sample size or power calculation was performed. 
(whilst it is good to know the tests performed i think im lacking information about the data and hypotheses tested to make any judgements at this stage about whether the analysis has been conducted correctly.)


* apparently the reviewer said the analytical approach could be more clearly explained. 
* apparently the reviewer also said: since outcomes were assessed repeatedly at multiple timepoints (baseline, post-loading, year 1, and year 2), a repeated-measures analytical framework such as mixed-effects modelling or repeated-measures ANOVA may be more appropriate to account for within-eye correlations across time. 
(im not entirely sure what he is referring to here and i guess would need more info about how the analysis was performed to understand what he is referring to )
* apparently the reviewer also said: if both eyes from the same patient were included in the analysis, this may violate the assumption of independent observations. The authors should clarify whether any statistical adjustment for intra-patient correlation was performed. 
* No statistical adjustment for clustering at the patient level (e.g., mixed-effects modelling or generalised estimating equations) was performed, and this should be considered when interpreting the results

    



my friend wants to know if has to redo the analysis in light of the reviewers comments.