JMIR Publications

ABSTRACT

Background Worldwide, the prevalence of mental disorders is very high and the guideline-oriented care of patients depends greatly on an early diagnosis as well as a regular and valid evaluation of the course, to be able to intervene quickly in case of imminent recurrence or deterioration in therapeutic terms. Experienced physicians and psychotherapists are neccessary for diagnostics and treatment but not available in sufficient numbers everywhere, especially in rural areas or in less well-developed countries. The human language is capable of revealing the psychic situation of the speaker by altering the non-contentual aspects (speech melody, intonations, speech rate, etc.). The time and experience to learn the speechpatterns of a patient in healthy and ill moments is often unavailable, leaving the opportunities inherent in capturing linguistic, non-contentual information unused. In order to improve the care of patients with mental disorders just under these aspects, we have developed a concept for assessing the most important mental parameters through a non-contentual analysis of the active speech. Using speech analysis for assessment and tracking of mental health patients, opens also the invaluable possibilities of remote, automatic and ongoing evaluation, when used with patients‘ smartphones, as part of the strong digital and mobile health trends. Methods/Design In this paper, we describe a two-arm, randomized controlled trial. The participants are all recruited in one outpatient neuropsychiatric treatment center. Inclusion criteria are e.g. a psychiatric diagnosis made by a specialist, no terminal or life-threatening illness, fluent use of the German language, exclusion criteria are e.g. psychosis, dementia, speech or language disorders, addiction history or suicide attempt currently or in the last 12 months. The measuring instruments are the "VoiceSense" voice analysis tool, which enables the analysis of 200 specific speech parameters and assessment of the findings through the use of psychometric questionnaires. Discussion The importance of content-free speech patterns should not be overestimated. This is particularly evident in the interpretation of the psychological findings. Applying a software analysis tool can increase the accuracy of finding assignments and improve patient care. Trial Registration This study is registered at „clinicaltrials.gov“, Number was NCT03700008, registration date 09 October 2018, http://www.clinicaltrials.gov.