Patient Headache Questionnaire Leads to Greater Diagnosis Rates, Less Non-Specific Diagnoses in Children

Christina Szperka, MD, MSCE

In a recently published study, the use of an electronic health record (EHR)-integrated patient headache questionnaire resulted in improved specificity of migraine diagnosis and in turn, increased treatment prescription rates. The tool remains available for other institutions to utilize and provides a source of data for observational research on headache disorders.¹

Between July 2011 and December 2022, the time of the analysis, 31,484 new patients seen in neurology were given a headache-related diagnosis. Led by Christina Szperka, MD, MSCE, director of the Pediatric Headache Program at the Children’s Hospital of Philadelphia, the use of the patient headache questionnaire form nearly doubled the odds of making a specific diagnosis (OR, 1.90; 95% CI, 1.65-2.19) after adjusting for variables. Throughout that time, the proportion of new patients with headache who were given a diagnosis of migraine increased 9.7% and non-specific headache diagnoses decreased 21.0%.

The patient headache questionnaire (HAQ) was designed initially on paper in 2013-2014, then in REDCap (July 2014-March 2016), then fully integrated into EpicCare in May 2017 along with a paired provider Smartform. In its development, questions were designed to elicit concerning features that might suggest a dangerous underling cause for headache, as well as features of primary headache according to the International Classification of Headache Disorders, 3rd edition. In addition, it was developed in collaboration with other academic centers and incorporated the National Institutes of Health Common Data Elements, the Pediatric Migraine Disability Scale (PedMIDAS), or MIDAS for adults, and the elements captured by the American Registry for Migraine Research.

At each stage, patients completed the forms and clinicians performed a semi-structured interview. From 2011 to 2022, the diagnosis of migraine increased from 52.3% to 62.1% while non-specific headache diagnoses decreased from 39.7% to 18.7%. After the launch of the EHR-integrated questionnaire in May 2017, more than half (57.9%; 8759 of 15,122) of new patients who were given a headache-related diagnosis completed the questionnaire. Completion rates increased over the first few years and then plateaued around 60% in 2020-2022.

READ MORE: Phase 3 PRODROME Study Highlights Ubrogepant’s Impact on Migraine Prior to Attacks

At the conclusion of the analysis, findings showed that encounters in which the provider reviewed the patient answers also increased the odds of a specific diagnosis compared with encounters of no patient HAQ (OR, 1.20; 95% CI, 1.10-1.30). Notably, encounters from providers in the headache programs and practicing at satellite locations increased odds of making a specific diagnosis (headache program: OR, 1.80 [95% CI, 1.58-2.04]; satellite: OR, 1.47; 95% CI, 1.35-1.60); however, there was no interaction between form use and provider being in the headache program.

A sensitivity analysis to further confirm migraine diagnosis specificity was conducted by determining the agreement across ICD coding and ICHD-3-based algorithm utilizing patient questionnaire answers. All told, the diagnoses were concordant for 56.5% of encounters overall, 48.8% when there was no patient HAQ, 56.7% when clinician reviewed answers, and 58.3% when clinician used the form. The diagnostic agreement was increased by both provider use of the form (OR, 1.39; 95% CI, 1.24-1.56) and reviewing of the patient’s answer (OR, 1.39; 95% CI, 1.21-1.61).

In total, 70.6% of new patients seen for headache were prescribed a new medication or continued a medication that had been started previously. Acute medications, accounting for 33.4% of the cohort, were the most common prescribed, while preventives were used much less (11.4%) and the combination of both used slightly less (25.7%). After adjusting for other covariates, making a specific diagnosis resulted in nearly twice the odds of prescribing a therapy for migraine (OR, 2.39; 95% CI, 2.20-2.60). Visits in satellite locations were associated with decreased odds of prescribing treatment (OR, 0.59; 95% CI, 0.55-0.64).

Earlier this year, at the 2023 American Headache Society (AHS) annual meeting, held June 15-18, in Austin, Texas, Szperka presented a lecture overviewing the use of such questionnaire and its impact on improving the diagnosis of migraine. Following her presentation, Szperka sat down with NeurologyLive^® to discuss how the EHR-integrated questionnaire can be utilized in real-world settings. In the video below, she discussed the importance of patients receiving a more accurate diagnosis and how this questionnaire helps healthcare institutions and patient care.

REFERENCE
1. Szperka CL, Witzman S, Ostapenko S, et al. Patient headache questionnaires can improve headache diagnosis and treatment in children. Headache. 2023;00:1-13. doi:10.1111/head.14647