Clinical Validity Needed for Accurately Reporting Dry Eye Symptoms

Numerous patient-reported outcome dry eye questionnaires (PRO-DEQs) are used to identify dry eye disease, but many have not been clinically validated, according to a review published in the American Journal of Ophthalmology. Exclusively using clinically validated tests will ensure quality, comparability, and replicability among studies, according to the report.

Investigators performed a scoping literature review in PUBMED to identify PRO-DEQs published up to July 2018. A total of 973 records were included in the analysis which consisted of reports providing information into the design and validation of PRO-DEQs (n=56), clinical studies using PRO-DEQs (n=885) and other studies mentioning PRO-DEQs (n=32). 

The researchers identified 49 PRO-DEQs. The number of survey items ranged from 1 to 57. Among these questionnaires, 22 had an established validity or reliability status. Administration methods included self-administration (n=16), interviewer guided (n=1), and mixed administration (n=1). The administration method was not specified in 4 of the PRO-DEQs. 

The ocular surface disease index (OSDI) was the most widely used questionnaire, appearing in 600 records. A total of 15 modified OSDI versions encompassing between 5 and 29 items were discovered. 

Dry eye is categorized into different degrees of severity in 4 of the validated questionnaires, 9 categorize the presence or absence of dry eye, and 9 lack a cutoff point for dry eye. 

“Among the advantages of [patient reported outcome] questionnaires are that they avoid information bias on the part of the assessor, are simple to use and can be used at a low cost,” according to the researchers. “The use of these can help make decisions in clinical practice adapted to the needs of the patient.”

The study authors stress, however, “Without validation of the instrument, the results or conclusions drawn may be meaningless or even inappropriate.”

Study limitations include the use of a single database (PUBMED) and failure to include articles published after July 2018.