Article Text

Download PDFPDF
Out of sight, out of mind: a prospective observational study to estimate the duration of the Hawthorne effect on hand hygiene events
  1. Alon Vaisman1,
  2. Grace Bannerman2,
  3. John Matelski3,
  4. Kathryn Tinckam2,
  5. Susy S Hota1
  1. 1 Infection Prevention and Control, University Health Network, Toronto, Ontario, Canada
  2. 2 Multi-Organ Transplant Program, University Health Network, Toronto, Ontario, Canada
  3. 3 Biostatistics Research Unit, University Health Network, Toronto, Ontario, Canada
  1. Correspondence to Dr Alon Vaisman, Infection Prevention and Control, University Health Network, Toronto, ON M5G 2C4, Canada; alon.vaisman{at}


Background Human auditing has been the gold standard for evaluating hand hygiene (HH) compliance but is subject to the Hawthorne effect (HE), the change in subjects’ behaviour due to their awareness of being observed. For the first time, we used electronic HH monitoring to characterise the duration of the HE on HH events after human auditors have left the ward.

Methods Observations were prospectively conducted on two transplant wards at a tertiary centre between May 2018 and January 2019. HH events were measured using the electronic GOJO Smartlink Activity Monitoring System located throughout the ward. Non-covert human auditing was conducted in 1-hour intervals at random locations on both wards on varying days of the week. Two adjusted negative binomial regression models were fit in order to estimate an overall auditor effect and a graded auditor effect.

Results In total, 365 674 HH dispensing events were observed out of a possible 911 791 opportunities. In the adjusted model, the presence of an auditor increased electronic HH events by approximately 2.5-fold in the room closest to where the auditor was standing (9.86 events per hour/3.98 events per hour; p<0.01), an effect sustained across only the partial hour before and after the auditor was present but not beyond the first hour after the auditor left. This effect persisted but was attenuated in areas distal from the auditor (total ward events of 6.91*6.32–7.55, p<0.01). Additionally, there was significant variability in the magnitude of the HE based on temporal and geographic distribution of audits.

Conclusion The HE on HH events appears to last for a limited time on inpatient wards and is highly dynamic across time and auditor location. These findings further challenge the validity and value of human auditing and support the need for alternative and complementary monitoring methods.

  • infection control
  • social sciences
  • nosocomial infections
  • hospital medicine
  • compliance

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Twitter @AlonVaisman, @hotasusy

  • Contributors Concept and design: AV, SSH, GB, KT. Acquisition, analysis or interpretation of data: GB, AV, JM. Drafting of the manuscript: AV, SSH. Critical revision of the manuscript for important intellectual content: All authors. Statistical analysis: JM. Supervision: SSH.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement No data are available for request for this publication.