March 2019


Research highlight
Study: Common dry eye questionnaires show modest diagnostic accuracy for detecting clinical signs

by Vanessa Caceres EyeWorld Contributing Writer

Dr. Nolan examines a patient.
Source: Mark Nolan, OD


Eyecare practitioners need to evaluate both symptoms and signs when diagnosing dry eye disease

Does your practice use dry eye questionnaires to help assess the severity of dry eye? How helpful have those questionnaires been to guide diagnosis or treatment?
A study published in JAMA Ophthalmology found that the five most commonly used questionnaires had modest diagnostic accuracy values for detecting clinical dry eye signs, although two of the questionnaires were slightly better than the others.1 These findings would suggest that both symptoms and signs need to be assessed when diagnosing dry eye disease, and that neither symptoms nor signs should be evaluated in isolation.
Study researchers from the New Zealand National Eye Centre, University of Auckland, Auckland, New Zealand, addressed the recent recommendation from the Tear Film and Ocular Surface Society Dry Eye Workshop II (TFOS DEWS II) to use the Ocular Surface Disease Index (OSDI) and the 5-item Dry Eye Questionnaire (DEQ-5) as part of the dry eye diagnostic battery. The researchers’ prospective, investigator-masked study included OSDI, DEQ-5, McMonnies Dry Eye Questionnaire, Symptom Assessment in Dry Eye (SANDE), and Standard Patient Evaluation of Eye Dryness (SPEED) to evaluate the ability to detect objective signs of dry eye disease.
The study’s 211 participants were all age 18 or older and did not have any surgical procedures within 3 months of study participation. Nearly 60% of the study participants were female, with a mean age of 41 years. Participants were of European (60.7%), East Asian (25.6%), South Asian (11.4%), and other (2.4%) ethnicity.
Instructions for use of the questionnaires was given, and an independent observer evaluated right eye ocular surface parameters using keratography. Researchers used the presence of clinical signs of dry eye disease as assessed according to the homeostasis markers arm of the TFOS DEWS II dry eye diagnostic criteria, which includes a noninvasive tear film breakup time of less than 10 seconds, tear film osmolarity of 308 mOsm/L or greater, interocular difference in osmolarity of 8 mOsm/L or higher, more than five corneal staining spots, more than nine conjunctival staining spots, or eyelid margin staining of 2 mm or greater in length and 25% or higher in width.

Study results

Nearly 70% of participants met the TFOS DEWS II criteria for clinical dry eye signs, with the following parameters:
• median noninvasive tear film breakup time of 8.6 seconds
• mean tear osmolarity of 310 mOsm/L
• mean interocular difference in osmolarity of 8 mOsm/L
• 33.6% of participants had more than five corneal spots, 50.7% had nine or more conjunctival spots, and 46% had eyelid margin staining 2 mm or greater in length and 25% or greater in width
Researchers found that the OSDI and SANDE questionnaires had a diagnostic accuracy significantly greater than chance (C statistic, 0.65; 95% CI, 0.56–0.73, P<.001 for the OSDI; C statistic, 0.63; 95% CI, 0.55–0.72, P=.002 for the SANDE), but the discriminative abilities of the SPEED, DEQ-5, and McMonnies did not have that same level of accuracy. Although the OSDI and SANDE had better discriminative ability to detect clinical dry eye signs, other statistical values for all five of the questionnaires were “relatively modest,” the researchers observed. “Sensitivity and specificity values of all five questionnaires were less than 80%,” said study co-researcher Michael Wang, MBChB, honorary research fellow, Department of Ophthalmology, New Zealand National Eye Centre.
The challenges with interpreting the findings reported by previous diagnostic accuracy studies assessing dry eye questionnaires may be related to the heterogeneity of study populations, methodologies, and reference standards used, the researchers said. “This heterogeneity can introduce significant challenges when interpreting the relative performance of these screening instruments,” they wrote.
Researchers think this is the first study that directly compares the five questionnaires within the same study population and uses the TFOS DEWS II diagnostic criteria for clinical dry eye signs as the reference standard.

From research to practice

The study results do not surprise Ming Wang, MD, PhD, Wang Vision Cataract & LASIK Center, Nashville, Tennessee, who found after a few years that dry eye questionnaires did not help much with patients’ dry eye management decisions; that could be related to the fairly low sensitivity and specificity of the questionnaires, he observed. “The low specificity [in the study] indicates that all of these questionnaires may be likely to miss patients who may have clinical signs, which could result in an underdiagnosis of dry eye if relied on too heavily,” he said.
Dr. Ming Wang’s office also stopped using the questionnaires because he thinks patients are already overwhelmed by paperwork they complete for the office.
After reviewing the study, Mark Nolan, OD, ReVision LASIK & Cataract Surgery, Columbus, Ohio, said he is now more likely to use the OSDI questionnaire than the DEQ-5 already in use at his office because the former appears to be more effective. However, he will continue to heavily rely on exam findings with the use of diagnostic tools such as the slit lamp. “A patient could present with more or less severe dryness based on the examination findings,” he said.
Doctors treating dry eye who use these questionnaires may want to consider which one is most appropriate contextually, according to Dr. Michael Wang. “The OSDI, for instance, is useful as a global assessment of dry eye symptomology and their overall impact on the quality of life and visual function of a patient. As a relatively shorter and less time-consuming questionnaire, the SANDE visual analogue scales might be more suited in resource- and time-constrained settings, such as repeated measurements conducted over a period of time to monitor the efficacy of dry eye management therapies,” he said.
The Ocular Surface Laboratory at the University of Auckland has other research underway related to management therapies, diagnostic methodology, and epidemiology of dry eye disease, Dr. Michael Wang said.


1. Wang MTM, et al. Comparative evaluation of 5 validated symptom questionnaires as screening instruments for dry eye disease. JAMA Ophthalmol. 2018 Nov 8. Epub ahead of print.
2. Wolffsohn JS, et al. TFOS DEWS II Diagnostic Methodology report. Ocul Surf. 2017;15:539–574.

Editors’ note: The physicians have no related financial interests related to their comments.

Contact information

Michael Wang:
Ming Wang:

Study: Common dry eye questionnaires show modest diagnostic accuracy for detecting clinical signs Study: Common dry eye questionnaires show modest diagnostic accuracy for detecting clinical signs
Ophthalmology News - EyeWorld Magazine
283 110
220 125
True, 3