How data-driven feedback improved hospital staff hygiene by up to 54%

Intervention · Health and Safety

Abstract

The COVID-19 pandemic has given us a first-hand look into the importance of hand-washing. This is especially true for front-line healthcare workers, who are at constant risk of exposure to bacteria and viruses. 

Unfortunately, healthcare workers can forget about the importance of hand-washing and skip this vital step. This leaves them and their patients vulnerable to infections. While many methods have been tried to ensure that hospital staff comply with these health and safety measures, most have failed. 

From February 2018 to April 2019, Danish university hospitals explored whether they could change this behavior by nudging healthcare workers to wash their hands using sensory technology and digital progress updates. These methods proved effective: the gentle change in messaging and data driven messaging increased hygiene compliance by 54%. 

WANT TO WORK TOGETHER ON A RELATED PROBLEM?

Effective interventions start with a nuanced understanding of how decisions are made. Our mission is to help large organizations be better and do better, using behavioral science.

Learn about what we do

Rating = 4/5 (Significant effects; direct positive impact on participants; small sample size)

How feedback and social norms increased hygiene compliance by 54%
Condition
Results 
Positive feedback via electronic badge each time hygienic procedures were completed.

Bi-weekly meetings to review performance updates and set new target hygiene goals.
On average, doctors’ hand-washing nearly tripled, while nurses saw an average 20% increase. 
Personalized feedback to compare hygenic performance with peers. Nurses’ hygienic performances doubled.

Key Concepts

Nudges:  Nudges are gentle suggestions towards a desired outcome. They subtly alter the context in which we make our decisions. They often occur below our conscious awareness. 

Availability Heuristic: We typically use information that comes to mind easier as a shortcut when making decisions. If the consequences of our actions aren’t cognitively available, we take them less seriously. 

The Problem

The role of handwashing in infection prevention

Healthcare workers sometimes lapse in their adherence to hygiene guidelines. This results in hospital-acquired infections (HAI’s), which increase morbidity, mortality, absenteeism, and drive up treatment costs. It is estimated that approximately 15% to 30% of these HAIs are caused by inadequate hand-washing, making it an important key in fighting this healthcare issue. 

Previous attempts

Many methods have been attempted to remedy HAI’s, yet have only been moderately successful or did not contribute to lasting change. Some of these methods included education sessions, performance feedback, increasing administrative support, constant reminders, and combinations of these techniques. Unfortunately, these methods only scratch the behavioral surface of why health care workers weren’t complying.

The potential for behavioral science

As hospital-acquired infections occur multiple days after a healthcare worker exposes the patient, it is difficult to make the connection between the initial behavior and the subsequent infections. In a busy, stressful environment it is unsurprising that many healthcare workers make decisions based on energy-saving heuristics. As the importance of hand-washing isn’t cognitively prioritized, it is unsurprising that workers can skip this step. Therefore, the issue can be remedied by two behavioral steps: 

  1. Increasing the availability of information on the importance of hand-washing. 
  2. Changing the environment into a more hygienic one.

Design

The set-up

To do this, these scholars performed an observational test in a large Danish university surgical department, which had been previously plagued by infectious disease outbreaks. The department’s unfavourable history with diseases implied that hand-washing compliance was subpar, and the size of the ward (29 beds) provided a decent sample size for this behavioral intervention. 

The study was divided into three phases. For the first two months, researchers simply observed the current level of hand hygiene compliance in the ward. This first phase acted as a control period, providing baseline levels for the next phases. 

The researchers activated their interventions in the second phase, which was approximately 12 months long. They applied both data-driven performance feedback and light nudges.

The (Sani) nudge: a handwashing data-collector

The nudges were supplied by the Sani Nudge, an automated hand hygiene monitoring system. 

The Sani Nudge works like this: healthcare personnel are given a badge with an electronic ID. The badge can be picked up by a variety of sensors placed throughout the hospital. When a healthcare worker uses a soap dispenser, the name of the worker is recorded. To nudge towards improved sanitation, the sensor shows a green smiley face after the worker washes their hands. When the healthcare worker approaches a patient’s bed, sensors will also record their presence. Therefore, if the worker doesn’t wash their hands before approaching the patient, there will be data showing that they interacted with the patient, but did not use the soap dispenser.

Data-driven feedback

Taking this data, the researchers began the second facet of the experiment: data driven performance updates. The head doctor and nurses would gather the unit’s healthcare workers for a brief biweekly meeting, in which they would go over data for the past week, discuss the barriers that prevented people from reaching their goals, and selected a new goal for the healthcare workers to perform in the following week to improve handwashing compliance. 

Personalized data-driven feedback 

The final phase overlapped with the second phase, occurring 8 months into Phase 2. In this phase, nine volunteer nurses began to receive personalized weekly compliance data. They were able to see their data on the job, and compare their data to the rest of their coworkers.  While an important phase for discerning individual effects, there remains a potential selection bias present for these stage 3 participants. As these nurses volunteered to join this aspect of the experiment, some of their success may be attributed to their confidence in their own hand-washing prowess.

Results and Application

The increase in regular handwashing

Overall, the implementation was highly successful. Doctors in the surgical unit were notoriously bad at washing their hands, washing 16% of the time in the patient rooms, and 24% in all other rooms before the intervention. After the intervention, doctors were washing their hands 42% of the time in patient rooms and 78% of the time in the remaining rooms. Nurses were also highly successful, jumping from 27% to 43%, and 39% to 64% in the patient and non-patient rooms, respectively. 

The power of self-selection and personalized feedback

The most successful subsection of the group were the nine nurses who self-selected to engage in the third phase. Compared to the other nurses, who rose to 42% and 78%, these nurses rose to 55% and 80% from baseline rates, implying that individualized feedback was highly predictive of further success with the nudge program. This study shows the value of behavioral science in forwarding compliance and hygiene in hospitals, as well as reducing HAI’s. 

Industry
Application
Health & Wellbeing Positive messaging and weekly performance monitors could be used to incentivize smoking cessation amongst smokers who want to quit.
Education Providing students with individualized performance reviews concerning their reading rates can be a great way to increase reading by leveraging social norms.
Climate & Energy Nudging households to increase their recycling and composting rates through individualized feedback can be an easy and effective way to achieve sustainable outcomes.

Ethics

  • Study made use of volunteers for the third phase of its intervention, therefore fully respecting the participant’s choice and will.
  • The intervention managed to nudge towards desired outcomes without altering the participant’s environments drastically.
  • No information on the diversity of the healthcare workers participating in the study.
Dimension
Verdict
Comments
Welfare

Does the intervention demonstrably improve the lives of those affected by it?
Positive
The intervention improved the lives of physicians and patients by limiting the spread of germs and diseases.
Does the intervention respect the privacy (including the privacy of identity) of those it affects?
Positive
Data was anonymized to both the researchers and the participants until the third stage, where nurses received their own personal information. Only personal data given is a profession.
Does the intervention have a plan to monitor the safety, effectiveness, and validity of the intervention?
Room for Improvement
In this study, participants were debriefed after the experiment. The study abided by an overall transparent process, but does not give information on long-term monitoring.
Autonomy

Does the intervention abide by a reasonable degree of consent?
Positive
Participation in the study was voluntary. All participants were briefed on the studies goals and fully understood how the Sani Nudge system functioned.
Does the intervention respect the ability of those it affects to make their own decisions?
Positive
The intervention allowed participants to change their own behavior by implementing feedback and performance updates. It gave healthcare workers the chance to reflect and gave them full ability to make their own decisions.
Does the intervention increase the number of choices available to those it affects?
Not Applicable
The study does not necessarily offer new methods of hygiene to the healthcare workers, it simply encourages them to abide by current best practices.
Equity

Does the intervention acknowledge the perspectives, interests, and preferences of everyone it affects, including traditionally marginalized groups?
Insufficient Information
No information was given on the acknowledgement of traditionally marginalized groups, which in this case could be the many lives lost to poor hospital hygiene.
Are the participants diverse?
Insufficient Information
No information was given concerning the diversity of the healthcare workers involved.
Does the intervention help ensure a just, equitable distribution of welfare?
Room for Improvement
The intervention helps ensure a just treatment of patients in hospitals, but does not necessarily touch on the access to healthcare in the first place.

Related TDL Content

Encouraging Social Compliance During COVID-19: During a global health crisis, compliance to things like handwashing and mask wearing are essential for protecting ourselves and others. In this piece, TDL’s Siddharth Ramalingam examines how we can use similar techniques to the ones used in this intervention to help maintain public health during COVID-19. 

Dr. Mitesh Patel on Nudging, Tech, and Health Care: Healthcare is a profession full of important choices, ranging from washing your hands to deciding where to perform surgery. If you are further interested in the ways decision science can improve healthcare outcomes, The Decision Lab recently interviewed Dr. Mitesh Patel, the founder of the first behavioral design team within a health system. This enlightening interview focuses on his innovative views on nudging, technology, and medical decision-making.  

Sources

Iverson, A. M., Stangerup, M., From-Hansen, M., Hansen, R., Paladin Sode, L., Kostadinov,K, Hansen, M. B., Calum, H., Ellermann-Erikson, S. & Dahl Knudsen, J. (2021).Light-guided nudging and data-driven performance feedback improve handhygiene compliance among nurses and doctors.American Journal of InfectionControl, Vol. 49, No. 6, 733-739.

Notes illustration

Eager to learn about how behavioral science can help your organization?