Subjective sleepiness

Subjective measures of sleepiness are used to assess the perceptions of participants of their own sleepiness levels. These measures can be divided into two categories:

1) Trait sleepiness, how sleepy a person feels in general

2) State sleepiness, how sleepy a person feels at a specific time point

In the context of driver sleepiness trait sleepiness is important because if a person is consistently sleepy they may have greater difficulty maintaining alertness whilst driving than someone who is not. State sleepiness is important because people continue to drive based on their own perceptions of how sleepy they are and use this information to decide when they should stop.

3.5.3.1 Trait sleepiness

Trait sleepiness is most commonly quantified using the Epworth Sleepiness Scale (ESS). The ESS measures average sleep propensity, an individual’s general level of sleepiness independent of the current situation (Shen et al. 2006).

The ESS is a widely used 8 item self rating scale. Participants indicate the probability of them falling asleep under different set circumstances. In each circumstance participants can choose

0 = no chance of dozing 1 = slight chance of dozing 2 = moderate chance of dozing 3 = high chance of dozing

Figure 3.5-1 Epworth Sleepiness Scale

The resulting scores are totalled ranging from 0 to 24 with scores over 12 suggesting excessive daytime sleepiness (Johns 1991).

This scale is quick and easy to use so it popular in a clinical setting to assess for EDS in OSA patients. It has been shown to correlate with RDI (Johns 1993) and improves in OSA patients once started on CPAP treatment (Engleman et al. 1996). It is generally accepted as a good tool to identify OSA. However, some studies have not found significant correlation between ESS and OSA severity as measured by AHI and MSLT

The Epworth Sleepiness Scale

How likely are you to fall asleep in the following situations? Please indicate, using the following scale, which is most appropriate given the situation.

0 = Would never doze

1 = Slight chance of dozing

2 = Moderate chance of dozing

3 = High chance of dozing

Situation Chance of Dozing

Sitting and Reading ………..

Watching TV ………..

Sitting inactive in a public place (e.g. theatre/meeting) ………..

As a passenger in a car for an hour without a break ………..

Lying down in the afternoon when circumstances permit ………..

Sitting and talking to someone ………..

Sitting quietly after lunch without alcohol ………..

(Chervin et al. 1997, Olson et al. 1998). Despite criticism; some studies have found correlations between ESS and MSLT (Johns 1991) though Johns does argue that the MSLT should not be considered the ‘gold standard’ measure of sleepiness (Johns 2000a, Johns 2000b, Johns 1991). ESS has been found to correlate to inappropriate line crossings on a dual carriageway in OSA drivers and controls (Philip et al. 2008).

The ESS relies on honest self reporting of symptoms; as such it may be subject to misinterpretation and untruthful responses (Shen et al. 2006). An example of these problems can be demonstrated by asking people to complete an ESS about their spouse, as one study has found bed partners give a higher ESS than the patients themselves (Walter et al. 2002b). However, another study has found close agreement between scores of spouses (Olson et al. 1998).

In comparison to the equivalent objective measures, MSLT and MWT, the ESS is easier to use and much more cost effective. During the screening day, participants completed the ESS (appendix 2).

3.5.3.2 State sleepiness

The Karolinska Sleepiness Scale (KSS) is a widely used Likert scale to record sleepiness at a set time point (Akerstedt et al. 1990). It is quick and simple to complete allowing a participant to give an instant uniformed response to the question ‘How sleepy do you feel?’ Response is given as a number on a scale from 1 to 9.

The KSS has been validated to EEG activity showing that EEG activity changes when a KSS of 7 is given by healthy individuals; this is predominantly linked to an increase in theta and alpha power (Akerstedt et al. 1990, Baulk et al. 2001, Kaida et al. 2006). KSS has also been found to correlate to actual inappropriate line crossings on a dual carriageway with both OSA drivers and controls (Philip et al. 2008).

KSS is a subjective measurement so it is possible that participants may not answer honestly, to try to eliminate this, the investigator built up a good relationship with participants during the approximately three hours spent with each participant before they completed a study drive. The KSS has been validated against EEG, which has shown participants scoring 7 or higher on the KSS coinciding with EEG activity suggests sleepiness (Akerstedt et al. 1990).

An alternative to the KSS is a Visual Analogue Scale (VAS). VAS requires participants to indicate on a line where their subjective sleepiness falls. It is suggested that VAS is more sensitive than KSS (Shen et al. 2006) but for the purpose of a driving simulator it is more practical to use a scale with a verbal response.

Validation of the KSS to EEG shows that healthy people are able to tell when they are sleepy; in the case of driver sleepiness it is important that people recognise this and take action. They will only do this if they feel the sleepiness may impact on their

The Karolinska Sleepiness Scale

1. Extremely alert 2. Very alert 3. Alert 4. Rather alert

5. Neither alert nor sleepy 6. Some signs of sleepiness

7. Sleepy, but no effort to keep awake 8. Sleepy, some effort to keep awake

9. Very sleepy, great effort to keep awake, fighting sleep

chance of actually falling asleep. In order to assess this, a Likelihood of Falling Asleep (LHoFA) scale can be used in conjunction with the KSS (Reyner and Horne 1998b). Whereas the KSS can show if someone has insight into their level of sleepiness the LHoFA demonstrates if they perceive this to be a problem. For example if someone rates 9A they suggest they know they are very sleepy (very sleepy great effort to keep awake) but they believe they are very unlikely to fall asleep.

Figure 3.5-3 The Likelihood of Falling Asleep scale

On all test days state sleepiness was recorded throughout the drives using the KSS and rating the Likelihood of falling asleep in the next 5 minutes (LHoFA). The scales were explained to participants at the screening day and they practiced using it during the test drive. Every 200 seconds of the drive is marked as a sleep check because a recorded voice asked “sleep check?” at these times, to which participants responded with a number from the KSS and a letter from the LHoFA scales. Responses to both these scales are reflective of how the participant is feeling at that particular point in time. 36 responses were collected for each participant in each 2 h drive. The scales were constantly visible, presented on the dashboard of the car.

Whereas the KSS can show if someone has insight into their level of sleepiness the LHoFA demonstrates if they perceive this to be a problem. For example if someone rates 9A they suggest they know they are very sleepy but they believe they are very unlikely to fall asleep.

Likelihood of falling asleep in the next 5 minutes

A Very unlikely

B Unlikely

C Neither

D Likely

For analysis subjective sleepiness scores are taken to represent the 100 seconds prior and post a “sleep check?”.