After data are cleaned, analysis is recommended to be conducted by either a biostatistician or a researcher with statistical analysis experience.
What data should be excluded?
If more than 3 baseline refractions were conducted on a USP, exclude the refraction(s) that are the most different to the others.
If USPs were identified by certain optical services, the number and percentage of visits should be reported, however their data should be removed from all other analysis.
Q.REC indicators
The Q.REC indicator(s) are used to compare the dispensed spectacles from optical services to the USP’s baseline prescription.
Q.REC indicator 1: Optimally prescribed spectacles
Q.REC Indicator 1 is the key primary outcome for this study. It has been demonstrated that when glasses pass this indicator that USPs have significantly better vision and comfort compared to glasses that do not pass this indicator.5 This shows us the quality of refractive error care and information to plan and monitor optical services.
Table 1: Q.REC Indicator 1 criteria
| Spectacle component | Tolerance limits compared to baseline prescription |
|---|---|
| Spherical power | ± 0.50 dioptre |
| Cylindrical power | ± 0.50 dioptre |
| Cylindrical axis (if baseline |cylindrical power| ≤ 0.50 DC) | ± 7 degrees |
| Cylindrical axis (if baseline |cylindrical power| > 0.50 DC – ≤ 1.50 DC) | ± 5 degrees |
| Cylindrical axis (if baseline |cylindrical power| > 1.50 DC) | ± 2 degrees |
| Horizontal prism (total) | < 1 prism dioptre (in/out direction) |
| Vertical prism (total) | < 0.50 prism dioptre (up/down direction) |
Q.REC indicator 2: Adequately prescribed spectacles
For settings where sphero-cylindrical lenses are rarely prescribed or available, Q.REC indicator 2 can be used as the primary outcome that shows the quality of refractive error care. Similarly, it can help with planning and monitoring optical services, however is not as accurate as Q.REC indicator 1. For projects that use Q.REC indicator 1 as the primary outcome, Q.REC indicator 2 can be a secondary outcome or not used at all.
Data transformation
Baseline refractions
For each eye, all components of the refraction should be averaged, this includes the distance spherical power, near spherical power (particularly if the USP has a near addition with their prescription) cylindrical power, cylindrical axes, distance and near pupillary distances, and distance and near visual acuities.
Categorising into refractive error types
Each USP can be grouped into at least one refractive error type based on the baseline refraction. The following definitions for each type are recommended:
- Myopia: spherical equivalent < -0.50 DS in at least one eye
- Hyperopia: spherical equivalent > +0.50 DS in at least one eye
- Astigmatism: > 0.50 DC in at least one eye
- Emmetropia: spherical equivalent ≥ -0.50 DS and ≤ +0.50 DS in both eyes
- Presbyopia (objective definition): ≥ 1.00 D added to the best optical distance correction
To calculate spherical equivalent, the following formula can be used: Spherical equivalent power = Spherical power + (Cylindrical power)/2
Visual acuities
Distance and near visual acuity for the right, left and both eyes should be converted into logMAR decimal format. As each USP is likely to have unique near working/viewing distances, near visual acuity should be recalculated to account for their working distance.
Categorical variables of ‘good vision’ can be created for the right and left eyes separately and with both eyes open. USP while wearing dispensed spectacles is considered to achieve ‘good vision’ if the corrected visual acuity is <1.5 lines worse than baseline best-corrected visual acuity at distance or near.
Suitability of spectacle lens types
A USP is considered suitable for a spectacle lens type (either single vision distance / single vision near / bifocal / multifocal) if at least one of the study optometrists ticks that they are suitable for that lens type.
Analysing and presenting the data
The following section is a recommendation of data that can be analysed and presented. This initial analysis might produce emerging outcomes that are of interest to your context and warrant further analysis.
Q.REC Flowchart
A flowchart of the optical services sampling frame, the ones excluded, recruited, USPs recruited, services visited and the number of spectacles dispensed are the minimum details to be included.
USP characteristics
Describing USP characteristics can be presented in a table. Examples of descriptors include:
- Age – mean and standard deviation or categorised into groups
- Gender
- Refractive error type
- Presence of presbyopia
- Pupil distances – mean and standard deviation
- Types of spectacle lenses the USPs are suitable for
The table can be presented as the total group or disaggregated. If USPs are disaggregated, Fisher’s exact tests or Chi-square tests for categorical variables or ANOVA for continuous variables can be used to test USP differences by their disaggregated groups.