Skip to main content
Log in

Individual validation of model predictions of sleepiness and sleep hours

Validierung eines Modells zur individuellen Vorhersage von Schläfrigkeit und Schlafenszeiten

  • ORIGINAL ARTICLE
  • Published:
Somnologie - Schlafforschung und Schlafmedizin Aims and scope Submit manuscript

Zusammenfassung

Es wurden bereits mehrere mathematische Modelle zur Vorhersage der Schläfrigkeit entwickelt. Jedoch wurden wenige Validierungen an einzelnen Individuen durchgeführt.

Fragestellung:

In dieser Studie wird eine Validierung des Drei-Prozeß-Modells zur Vorhersage der Aufmerksamkeit an Schichtarbeitern durchgeführt. Modellvorhersagen zu den Schlafzeiten wurden ebenfalls überprüft.

Methode:

16 Schichtarbeiter nahmen teil. Die Schläfrigkeit wurde alle 2 Stunden über drei Schichten hinweg per Fragebogen angegeben. Das Modell wurde zur Vorhersage dieser Angaben eingesetzt. Das Modell benutzte nur Angaben zum Beginn und Ende der Arbeitsschichten sowie einer zusätzlichen Analyse von Aktigraphie-Daten.

Ergebnisse:

Die Vorhersagen basierend nur auf den Angaben zu den Arbeitsschichten korrelierten mit r = 0,55 mit der Schläfrigkeit. Die Vorhersagen blieben im Konfidenzintervall ± 1 des Fragebogens. Die zusätzliche Information der Aktigraphen verbesserte die Vorhersage nur geringfügig. Die Vorhersagen zu Schlafbeginn und Schlafende waren recht gut.

Schlussfolgerung:

Die Modellvorhersagen weisen eine recht hohe Zuverlässigkeit bezogen auf Schläfrigkeit und Schlafenszeiten auf. Vermutlich kann zusätzliche Information über individuelle Unterschiede die Genauigkeit noch weiter verbessern.

Summary

Several mathematical models for prediction of sleepiness have been developed. Few validations on individual levels are available.

Background

The present study was designed to provide validation on the individual level of predictions using the Three Process-Model of alertness regulation. Model predictions of sleep timing were also tested.

Method

Sixteen shift workers participated in the study. Ratings of sleepiness were made every 2 h across three shifts. The model was used to predict empirical ratings using as input only information of beginning and end of work shifts, as well as using information on sleep from actigraphs (in a separate analysis).

Results

The prediction using only information on work shifts correlated r = 0.55 (p < 0.001) with empirical ratings. Predictions were generally within ± 1 confidence interval of the ratings. Adding actigraphy sleep data improved predictions marginally. The model predictions of onset and offset of sleep were generally close to the target.

Conclusions

It was concluded that model predictions have a rather high validity both with respect to sleepiness and to sleep timing. It is probable that other information on individual differences will further improve predictability.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

References

  1. Borbély AA (1982) A two-process model of sleep regulation. Hum Neurobiol 1:195–204

    PubMed  Google Scholar 

  2. Mallis MM, Mejdal S, Nguyen TT, Dinges DF (2004) Summary of the key features of seven biomathematical models of human fatigue and performance. Aviat Space Environ Med 75:A4–A14

    PubMed  Google Scholar 

  3. Folkard S, Åkerstedt T (1987) Towards a model for the prediction of alertness and/or fatigue on different sleep/wake schedules. In: Oginski A, Pokorski J, Rutenfranz J (eds) Contemporary Advances in Shiftwork Research. Krakow: Medical Academy, pp 231–240

  4. Åkerstedt T, Folkard S, Portin C (2004) Predictions from the three-process model of alertness. Aviat Space Environ Med 75:A75–A83

    PubMed  Google Scholar 

  5. Folkard S, Åkerstedt T (1991) A three process model of the regulation of alertness and sleepiness. In: Ogilvie R, Broughton R, eds. Sleep, Arousal and Performance: Problems and Promises. Boston: Birkhäuser, pp 11–26

  6. Åkerstedt T, Folkard S (1995) Validation of the S and C components of the Three-process model of alertness regulation. Sleep 18:1–6

    PubMed  Google Scholar 

  7. Åkerstedt T, Folkard S, Kecklund G (1993) The prediction of intentional and unintentional sleep onset. In: Ogilvie RD, Harsh JR (eds) Sleep Onset – Normal and Abnormal Processes; Washington, DC; American Psychological Association, 1994, 73–88

  8. Åkerstedt T, Folkard S (1996) Predicting duration of sleep from the threeprocess model of alertness regulation. Occup Environ Med 53:136–141

    Article  PubMed  Google Scholar 

  9. Axelsson J, Åkerstedt T, Kecklund G, Lowden A (2004) Tolerance to shift work – how does it relate to sleep and wakefullness? Int Arch Occup Environ Health 77:121–129

    Article  PubMed  Google Scholar 

  10. Åkerstedt T, Gillberg M (1990) Subjective and objective sleepiness in the active individual. Int J Neurosci 52:29–37

    Article  PubMed  Google Scholar 

  11. Torsvall L, Åkerstedt T (1980) A diurnal type scale. Scand J Work Environ Health 6:283–290

    PubMed  CAS  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to T. Åkerstedt.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Åkerstedt, T., Axelsson, J. & Kecklund, G. Individual validation of model predictions of sleepiness and sleep hours. Somnologie 11, 169–174 (2007). https://doi.org/10.1007/s11818-007-0315-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11818-007-0315-7

Schlüsselwörter

Key words

Navigation