For many years, voice professionals have bemoaned the lack of standard, scientifically vetted protocols for voice assessment. Facilities and clinicians have largely been left to develop their own protocols, which have not been based on an established set of procedures with a firm scientific basis.
And that situation has made it virtually impossible to make valid comparisons of voice assessment measures across patients and facilities, limiting the extent to which such measures can be used to assess the effectiveness of treatments.
So, in 2001, members of ASHA Special Interest Group 3 (then Division 3), took action, launching a quest to develop such guidelines. The effort has spawned a standard method for auditory-perceptual assessment of voice quality and is now moving toward consensus on basic protocols for the instruments commonly used in voice assessment. Establishing these protocols will help us better evaluate our clinical effectiveness in response to a pressing need for outcomes research in the new era of health care reimbursement.
Birth of a measure
A complete clinical voice assessment typically includes auditory-perceptual evaluation, endoscopic imaging, and acoustic and aerodynamic assessment of voice functioning. In 2001, our SIG decided to start with developing a standard method for auditory-perceptual evaluation, considered the voice assessment procedure that speech-language pathologists use most. Many of us felt that auditory-perceptual assessment of voice was fraught with difficulties, including lack of standard terminology or scaling procedures for describing disordered voice quality, and inherent poor reliability of auditory-perceptual judgments of voice quality.
Thus, in 2002, we convened and sponsored the Consensus Conference on Auditory-Perceptual Evaluation of Voice, hosted by Katherine Verdolini Abbot at the University of Pittsburgh. Out of this gathering of international voice scientists, experts in human perception and SLPs came the Consensus Auditory-Perceptual Evaluation of Voice rating form and standard procedures. We distributed this clinical and research tool via the ASHA website, presentations at national and international meetings, and ultimately in the American Journal of Speech-Language Pathology in a 2009 article whose lead author was Gail B. Kempster.
Since then, a number of researchers have attempted to assess the validity and reliability of the CAPE-V. Our SIG funded the work of one of these groups, led by Richard I. Zraick. His team found that CAPE-V ratings corresponded well with ratings of the same voices by the same judges on the already existing GRBAS (grade, roughness, breathiness, asthenia, strain) scales. This finding established the empirical (concurrent) validity of the CAPE-V. GRBAS is widely used internationally but has several shortcomings—such as lack of standard instructions for administration and analysis—that CAPE-V seeks to fix. Zraick's team found slightly improved rater reliability for the CAPE-V over GRBAS ratings. But this reliability still varies greatly across raters and perceptual parameters.
It is not surprising that the CAPE-V does not completely resolve reliability issues. These issues continue to plague all clinical approaches for rating voice quality, and are the subject of ongoing basic research to better understand and control the sources of such variation. Future studies also need to better establish the true construct validity of the CAPE-V.
Overall, however, the CAPE-V achieves the important goals of establishing a standard (minimal) set of terms to describe voice quality, as well as standard procedures for administering and scoring that have some scientific underpinning and can be performed expediently in a clinical setting. These achievements, in turn, should facilitate more valid comparisons of voice quality estimates between facilities and patients, and in repeated assessments of the same patient.
Toward comprehensive standard guidelines
Next, SIG 3 turned its attention to developing comprehensive guidelines for clinical voice assessment. ASHA's National Center for Evidence-Based Practice helped with a systematic review of the voice assessment literature. The review—lead author Nelson Roy, published online in 2012 in the American Journal of Speech-Language Pathology—identified some evidence for use of selected voice measures in effective clinical evaluation, but the evidence was not sufficient to support developing guidelines for a standard clinical voice assessment protocol.
A chronic issue is the lack of consistency in performance of voice measures, which severely limits the extent to which we can combine data to build an evidence base. We therefore concluded that, to improve the evidence base, we need first to establish basic minimal specifications and protocols for the most common instruments used in voice assessment: endoscopic imaging and acoustic and aerodynamic assessments. ASHA's Board of Directors approved our request to form an eight-person ad hoc committee of experts to develop these specifications and protocols, and the committee began work in November 2011.
The group aims to create standard protocols that will include minimum technical specifications for instrumentation, analysis methods, and environmental factors that should be controlled to ensure valid measurement. The protocols also will delineate a minimum set of appropriate voice and speech tasks, and provide guidance on how results should be interpreted, integrated and reported. We are working to structure the protocols so they can be used across the range of voice-assessment settings, from a practice with limited access to advanced instrumentation to the high-end voice center.
Using expert consensus to establish clinical guidelines and protocols when the evidence base is lacking has become a widely accepted approach in health care—and is a key step toward assessing and improving our clinical effectiveness. Our expert committee on voice assessment plans to present its standard protocols to the ASHA membership for testing, peer review and modification over the next two years. Once finalized, the protocols are expected to improve the state of the art in clinical voice assessment and to facilitate collection of the clinical data needed to build the evidence base for treating voice disorders.