Feigning daytime sleepiness: potential effects on the psychomotor vigilance test Free

Excessive daytime sleepiness (EDS) is a multi-faceted construct that can be quantified with a variety of self-report questionnaires and objective tests. The conventional approach to validate a patient’s subjective report of EDS involves an objective measure such as the Multiple Sleep Latency Test (MSLT) or Maintenance of Wakefulness Test (MWT). While these tests often provide corroborating evidence of sleepiness, they may be expensive, time-consuming, difficult to perform, and have poor test–retest reliability [1, 2]. Additionally, a shared limitation of nearly all tools that aim to quantify EDS is that their validity is reliant on earnest cooperation on the part of the patient. Participants motivated to shorten or lengthen their sleep latency on MSLT or MWT may be able to do so [3, 4].

In view of these issues with objective assessment of daytime sleepiness, there has been an upsurge in the clinical application of the Psychomotor Vigilance Test (PVT) [5], a relatively brief procedure well-validated in experimental protocols [6, 7]. Studies to date suggest that PVT has utility in characterizing impaired cognitive performance in hypersomnolent patients [7–9]. Specifically, hypersomnolent patients perform significantly worse than non-sleepy controls on measures of mean and median reaction time, reciprocal of reaction time (RRT), lapses of at least 500 milliseconds, and RRT of the fastest and slowest 10% of responses [7]. Additionally, the slope of RRT over time (i.e. linear regression slope of RRT on trial minute) shows significantly more time-on-task worsening for the hypersomnolent group than for controls [7]. In this brief report, we explored to what extent a PVT performance profile might serve to differentiate hypersomnolent patients from normal subjects instructed to feign drowsiness.

In order to mirror the age distribution of the clinical population of CDH patients presenting at our center, we recruited healthy young adults (n = 16) from the community to serve as controls. They were instructed on the use of the 10-minute, handheld PVT device, and performed a 1-minute practice demonstration prior to testing. They were then told to try to make their PVT results appear as though they were sleepy, without any further instructions on how to achieve this. All participants performed the PVT alone in a quiet room, to minimize environmental distractions. Potential participants were excluded from the control cohort if they endorsed an Epworth Sleepiness Scale score >10 or self-reported habitual sleep durations >9 hours. Participants were not screened for the use of sleep- or wake-promoting medications, but rather for the presence or absence of self-reported sleepiness.

Standard PVT metrics were computed via React software (version 1.1.05, Ambulatory Monitoring, Inc). PVT metrics from these controls attempting to simulate sleepiness (CON-SS) were then compared to previously published data [7] from 33 controls instructed to achieve their best possible performance (CON) and patients with documented CDH (76 with idiopathic hypersomnia, 22 with narcolepsy type 2, and 19 with narcolepsy type 1) who were unmedicated for sleepiness at the time of testing. Differences in demographics, clinical characteristics, and PVT measures were compared across groups using ANOVA. In the case of significant group differences, pairwise comparisons were performed via Tukey test, controlling for multiple comparisons. Statistical analyses were performed in SAS (version 9.4). This study was approved by the Emory University Institutional Review Board and all participants gave informed consent.

Mean age for all participants was 33.9 years (+/− SD 14.2), not significantly different across groups (34.7 +/− 13.6 CDH, 32.8 +/− 15.9 CON, 30.3 +/− 15.8 CON-SS). One hundred twenty-one (72.9%) were women, with somewhat more women in the hypersomnolent group (78.6% CDH, 57.6% CON, 62.5% CON-SS, p = 0.03). By definition, Epworth scores were ≤10 in both CON and CON-SS groups, without significant differences (6.2 +/− 2.8 in CON-SS and 6.3 +/− 2.5 in CON, p = 0.85). CDH patients had higher Epworth scores (15.4 +/− 4.8), as expected.

PVT metrics by group are presented in Table 1. For most measures (RRT, lapses, fastest 10%, and slowest 10%), the three groups demonstrated significant differences in performance, with worst performance by the CON-SS group and best performance by the CON group. For mean and median reaction times, a similar pattern held true, but with only a trend toward significant differences between CON-SS and CDH groups (p < 0.06 for both pairwise comparisons). However, RRT slope demonstrated a very different pattern, such that the CON and CON-SS had similar RRT slopes, showing minimal accrual of deficits over time, compared with the CDH group, whose performance worsened over time.

The CDH group tended to complete the PVT at an earlier clock time than the other two groups (10:23 am +/− 1.5 hours for CDH, 12:49 pm +/− 2.7 hours for CON, and 12:57 pm +/− 2.9 hours for CON-SS, p < 0.001, CDH < CON = CON-SS). However, controlling for time of PVT administration did not meaningfully change results, with the same pattern of significant group differences still being observed for RRT, lapses, fastest 10%, slowest 10%, and slope in the multivariate analysis.

These results suggest a distinct profile of PVT performance for non-sleepy participants who are attempting to simulate sleepiness. This profile has two main features: (1) dramatically worse performance on most measures than that which is seen in hypersomnolence patients with impaired psychomotor vigilance, (2) normal RRT slope of performance over time, similar to that seen in healthy controls performing their best and superior to hypersomnolent patients. In essence, these results could be interpreted as suggesting that healthy participants, attempting to simulate sleepiness, will do so by consistently responding implausibly slowly, without considering any need to adjust their reaction time over the 10-minute test duration to mimic fatigue.

This pilot study has clear limitations. We were unable to phenotype controls simulating sleepiness using MSLT or MWT, thus their objective sleepiness was unknown. However, it has previously been shown that, among people with OSA, poorer PVT performance (more lapses, lower RRT, and slowest 10% of reaction times) is correlated with more subjective sleepiness on the Epworth but unrelated to measured sleepiness on the MSLT [10]. If such a pattern were true for controls and CDH participants, it would tend to mitigate the effect of this limitation on our results. The educational level of participants is not known and objectively measured sleep the night prior to PVT was not collected for those feigning sleepiness; both educational level and sleep deprivation may impact performance. Additionally, our sample size was small.

These results require replication, perhaps even in CDH patients asked to exaggerate their baseline performance deficits. However, if confirmed, our results suggest that clinical use of the PVT could be informed by assessing the profile of response, with the combination of marked outliers in most measures of reaction time but normal slope over time suggesting the possibility of embellishment or failure to attempt best performance. Our results also serve to highlight previous observations that the PVT may be affected by a variety of factors, including now participant instruction and intention.

Funding

Research reported in this publication was supported by the National Institutes of Neurological Disorders and Stroke of the National Institutes of Health under award number R01NS111280. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Disclosure statement

Financial disclosure: Dr. Rye and Mr. Saini report consulting fees from NextSense, outside the submitted work. Dr. Bliwise reports consulting fees from CliniLabs, Eisai, Ferring, Huxley, Idorsia, and Merck, outside the submitted work. The remaining authors report no financial disclosures. Nonfinancial disclosure: Dr. Trotti is a member of the Board of Directors of the American Academy of Sleep Medicine; Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors, and do not necessarily reflect the views of the American Academy of Sleep Medicine. The remaining authors report nothing to disclose.

Data Availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References

Author notes