Proposed supplements and amendments to ‘A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects’, the Rechtschaffen & Kales (1968) standard

Abstract

In 1967, the Association for the Psychophysiological Study of Sleep (APSS) chartered a committee of sleep researchers to establish a standard system for visually scoring stages of sleep. The Committee’s terminology and scoring system (edited by Alan Rechtschaffen and Anthony Kales, co-Chairpersons of the Committee) were quickly adopted after its 1968 publication under the auspices of the UCLA Brain Information Service as ‘A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects’. The manual and its recommendations have been well accepted and the system has spread across the world. The proposal to standardize recording techniques and scoring criteria was intended to increase the comparability of results reported by different investigators. Researchers who have applied the system correctly have increased the reliability of their sleep stage scoring. In the several decades since its publication, however, a number of serious areas of unreliability in the 1968 standard scoring system have been identified. In particular, researchers developing computer-based automatic sleep staging systems have encountered numerous vague and ambiguous areas in the current standard. At the 15th Annual Meeting of the Japanese Society of Sleep Research (JSSR) in 1990, Terashima’s report from the Japanese Conference for Sleep Analysis emphasized the need to re-examine the definitions of sleep stages when developing computer staging systems based on the standard scoring system. The results of multiple comparisons among institutes and laboratories using methods for automated sleep staging based on the standard scoring system were unexpectedly poor. The lowest level of congruence between automation and records staged visually using the 1968 standard system was found for Stage 1 sleep, with the second worst agreement (still less than 70%) for sleep Stages 3 and 4. To increase both the within- and between-researcher agreement in sleep stage identification and to foster the development of computer algorithms for automatic analyses of sleep, a need for additional definitions was recognized. In 1991, the Subcommittee for Automatic Sleep Staging (SASS) was formed by the JSSR. The Subcommittee comprised 53 investigators and seven project leaders selected for their skill in scoring sleep records: S. Sugita (Chair), M. Okawa (co-Chair), T. Kobayashi, T. Hori, A. Miyasita, S. Shirakawa and Y. Atsumi. In 1995, based on their 5 year discussions, the Subcommittee proposed supplementary definitions and amendments for the Standard Scoring System to the JSSR. These proposals were reported in the JSSR Newsletter (1996; No. 13, February 1, pages 5–13). In 1997, at its 22nd Annual Meeting, the JSSR restructured the SASS to form the Sleep Computing Committee (SCC). The SCC working group has now formalized Supplements and Amendments to the Standard Scoring System based upon the 1995 SASS proposals. The Committee has met several times for discussion and to test their new definitions using carefully selected representative polysomnographic recordings. The final JSSR Sleep Computing Committee proposals have been published in Japanese as the ‘Learning Manual of PSG Chart: Polysomnogram, Sleep Stage Scoring, Interpretation’. The Committee has now prepared the English language version of their proposals for evaluation and empirical testing by sleep researchers all over the world. The JSSR’s proposals to supplement and amend the currently used criteria for sleep staging (A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects, Rechtschaffen A, Kales A (eds). BIS/BRI University of California, Los Angeles, 1968; hereafter, the ‘standard criteria (SC)’) are detailed below. When research purposes are not met by the SC, different movement time criteria may be used. However, in such cases, the modified criteria should be clearly stated and the degrees of modification should not exceed those in the following examples. MT is judged following the SC, with the addition of scoring parts of the recording with higher muscle tones and degrees of artifact as ‘gross movement’ or ‘movement with duration more than 50% of an epoch’, and the number of such segments is reported. Even when an epoch fulfils the SC definition of MT, the epoch may still be scored as a stage of sleep using additional criteria (below). In this case, an SC-defined epoch of MT can simultaneously be defined as an epoch of sleep and a body movement epoch. Examples of criteria: (i) MT epoch is scored as the same stage as the preceding epoch and (ii) MT epoch is scored as the same stage as the following epoch. The following supplementary definitions are used for scoring a transition from stage wake to Stage 1 sleep. Stage 1 is not necessarily defined as a sleep onset (sleep onset is defined below). Alpha waves are electroencephalographic waves with a frequency of 8 Hz or more but less than 13 Hz. A vertex sharp wave is a