Hearing aid clinical study software

Pure-tone audiometry records the quietest tone a person hears at each tested frequency, in dB HL, separately for each ear and for air and bone conduction. Typical protocols test standard octave frequencies from low to high, and may add interoctave points; take the exact list from your protocol. Build one section per ear and one numeric field per frequency, with units in the field and a range check so impossible values raise a query at entry. Masking, test equipment and the test room are worth a field each, because they explain unexplained differences between sites.

Capture's built-in calculated fields cover a list of clinical formulas (BMI, BSA, MAP, eGFR, QTc, age from date of birth, time between two date-times), and a pure-tone average is not on it. Enter the average from the audiometer, or compute it in your statistics package from the exported frequency fields, and say which in the data management plan. Likewise, Capture does not claim an integration with audiometers or fitting software: results are entered by the audiologist, or transcribed from a printout, and every change keeps its audit trail with old value, new value and reason.

Speech tests need the same care. Record the test name, the listening condition (quiet, in noise, aided or unaided, sound field position), the list used and the score or signal-to-noise ratio. A condition field per row lets one repeating table hold aided, unaided and different-settings results for a participant, each row with its own audit trail. Pair it with a visit schedule so the exports are tidy: baseline, fitting, a few weeks of use and final outcome visits. For the generic issues of device studies, see EDC for medical device trials and how to run a medical device clinical study.

Self-reported benefit is often as important as the audiogram. Hearing questionnaires can be sent to the participant's phone through a secure link with reminders, completed at home between visits, and compared with the clinic result. Several widely used hearing questionnaires are copyrighted or carry conditions; Capture does not supply the instruments or verify licences, so the licence stays with your study documents and you should check the rights holder's terms before use. See ePRO for patient-reported outcome studies.

Blinding a hearing aid trial is rarely complete. The wearer knows whether the device is on. What can be blinded is the tester (who does not know which setting or device was used for a given test), the data analyst and sometimes the participant, in a crossover comparing two programmes that sound different but are not labelled. Capture's blinded roles never receive treatment-arm values, enforced by masked database views and row-level security, and a crossover can be built as epochs with randomization; see the blinding guide.

Many participants are older, and some have cognitive impairment or limited dexterity with phones. Offer a clinic-completed version of every questionnaire, and where a participant cannot consent for themselves, use the legally authorised representative step in eConsent. Large print and a person nearby are practical measures, not features the platform claims. Hearing loss also affects consent itself: allow time, written information and a quiet room, and record how consent was explained.