Behavioural determinants of testing behaviour during a hypothetical avian influenza outbreak: an interview study

This interview study identifies key behavioural determinants of testing during a hypothetical avian influenza outbreak, highlighting that individual autonomy, personal and familial health benefits, external constraints, and trust in authorities significantly influence testing decisions, while past pandemic experiences shape current attitudes.

The Gist

Imagine a future where a new kind of bird flu starts jumping from chickens to people, and then, scary as it sounds, from person to person. To stop it from becoming a global disaster, doctors need to test everyone who feels sick. But here's the catch: you can't force people to get tested if they don't want to.

This paper is like a "pre-game interview" with 17 regular people. The researchers asked them: "If this bird flu outbreak happened tomorrow, would you get tested? Why or why not?" They showed the participants a short cartoon animation to help them visualize the scenario, since no one has actually lived through a human-to-human bird flu pandemic yet.

Here is the breakdown of what they found, translated into everyday language:

1. The "Do-Not-Tell-Me-What-To-Do" Button (Autonomy)

The biggest thing the researchers found is that people hate being bossed around.

• The Analogy: Imagine you are at a buffet. If someone walks up and says, "You must eat this salad, or you can't leave," you might suddenly lose your appetite and refuse to eat it, even if you were hungry. But if they say, "Hey, this salad is great for you, feel free to grab some," you're much more likely to take a plate.
• The Finding: When people felt like the government was forcing them to get tested, they got angry and said, "No way, I'm not doing it." They wanted to feel like they were making their own choice. If testing felt like a punishment or a rule, they resisted. If it felt like a personal choice to stay safe, they were more willing.

2. The "Why Bother?" Question (Usefulness)

People are practical. They want to know what's in it for them.

• The Analogy: If you go to a mechanic and they say, "We need to check your engine," but you don't know if they can fix it or if it even matters, you might just drive away.
• The Finding: Many people asked, "If I get tested and it's positive, but there's no vaccine and no cure, what's the point?" They felt testing was useless unless it helped them personally. They cared more about protecting their own health or their grandma's health than "saving the country."

3. The "Ghost of Pandemics Past" (COVID-19 Memories)

Everyone is still carrying the baggage from the last few years.

• The Analogy: It's like going to a restaurant where you once had a bad experience. Even if the chef has changed, you might still be suspicious of the menu because you remember the last time you got sick.
• The Finding: People kept bringing up COVID-19. If they felt the government lied to them or was too controlling during the pandemic, they didn't trust them now. However, some people trusted local health workers (like the nurses at the testing sites) more than the politicians in the capital.

4. The "Social Shield" (Protecting Loved Ones)

While people didn't care much about "society" in the abstract, they cared deeply about their immediate circle.

• The Analogy: You might not wear a helmet just to be a "good citizen," but you will definitely wear one if it means you won't hurt your kids or your spouse.
• The Finding: The strongest reason people said they would get tested was to protect their family, their "soulmates," or vulnerable neighbors. They didn't want to be the one who accidentally gave the flu to their sick uncle.

5. The "Too Far to Walk" Problem (Logistics)

Sometimes, the barrier is just simple inconvenience.

• The Analogy: If the gym is a 20-minute drive away and you have to park in a different lot, you might just stay on the couch.
• The Finding: If the testing site was too far away, or if getting there took too much time, people were less likely to go. They also worried about the "sanction" of testing: if they tested positive, they'd have to stay home and couldn't go to work or see friends. That fear kept some people away.

The Big Takeaway

The researchers concluded that if the government wants people to get tested during a bird flu outbreak, they can't just shout orders from a megaphone.

• Don't be a dictator: Give people the feeling that they are choosing to help, not being forced.
• Be honest: Explain clearly why testing matters (e.g., "This helps us protect your grandma," not just "It's the law").
• Build trust: People need to believe the people asking them to test are on their side, not just trying to control them.
• Make it easy: If the testing site is a hassle to get to, people won't go.

In short: To stop a virus, you need to win people's hearts and minds, not just their compliance. If you make people feel trapped, they will run away. If you make them feel empowered and informed, they might just step up to the plate.

Technical Summary

1. Problem Statement

Avian Influenza (AI) poses a significant pandemic threat, particularly if the virus mutates to allow efficient human-to-human transmission. Effective outbreak management relies heavily on widespread public participation in testing for timely detection, contact tracing, and isolation. However, historical data from the COVID-19 pandemic in the Netherlands revealed fluctuating test rates and low engagement in certain populations, driven by factors such as distrust in authorities, unclear policies, and perceived lack of utility.

Current literature on testing behavior is largely derived from the COVID-19 context. There is a critical gap in understanding the specific behavioral determinants that would influence testing compliance during a hypothetical human-to-human AI outbreak, which may present different severity levels, transmissibility, and societal contexts. Without early identification of these determinants, public health strategies may fail to secure the necessary population engagement to mitigate transmission.

2. Methodology

The study employed a qualitative research design utilizing semi-structured, in-depth interviews based on the Theoretical Domains Framework (TDF).

• Participants: A heterogeneous sample of 17 participants was recruited via purposive and convenience sampling in the South Holland province of the Netherlands. The group included diverse ages (median 44, range 20–81), genders (71% female), education levels, and countries of birth. Recruitment specifically sought variation in general willingness to test for infectious diseases.
• Data Collection: Interviews occurred between May 2024 and February 2025. A key feature of the data collection was the use of a one-minute animation to simulate a hypothetical AI outbreak scenario. The animation depicted two stages:
  1. Initial outbreak in poultry with sporadic human cases (no confirmed human-to-human transmission).
  2. Escalation with confirmed human-to-human transmission, increased hospitalizations, and no available vaccine.
• Analysis: Verbatim transcripts were analyzed using a deductive thematic analysis approach, organized within the COM-B model (Capability, Opportunity, Motivation) of behavior. Coding was performed by one researcher with 25% double-coded by a second researcher to ensure reliability. Discrepancies were resolved by a third researcher.

3. Key Contributions

• Context-Specific Insights: This is one of the first studies to specifically investigate testing determinants for a hypothetical AI outbreak, moving beyond the general "pandemic" data of COVID-19.
• Autonomy as a Central Theme: The study identifies "Need for Autonomy" not just as a sub-domain of motivation, but as an overarching, critical determinant that transcends the standard COM-B categories.
• Differentiation of Trust: The research highlights a nuanced distinction in public trust, where participants often differentiated between the "Government" (political) and "Public Health Services" (operational), finding the latter more trustworthy.
• Individual vs. Collective Motivation: The study provides empirical evidence that individuals primarily frame testing decisions through an individual benefit lens (protecting self/family) rather than a public health lens (protecting society/hospitals), challenging assumptions that collective good is a primary driver.

4. Key Results

The analysis identified four primary thematic clusters influencing testing behavior:

A. Need for Autonomy and Freedom of Choice

• Finding: A strong desire for autonomy was the most significant driver. Participants expressed that testing would be more acceptable if it was a voluntary, informed choice.
• Impact: Perceived coercion or mandatory testing (e.g., government mandates) significantly reduced willingness to test. Participants stated they would rather isolate voluntarily than submit to mandatory testing regimes.
• Quote: "It is mainly the fact that people feel that things are required... At that point, I no longer want to (get tested)."

B. Motivation (Drivers and Barriers)

• Disease Severity & Usefulness: Willingness to test was contingent on perceived severity. Mild symptoms were often dismissed as "just a flu." Furthermore, the perceived lack of treatment or vaccine for AI led some to question the utility of testing ("What do I gain?").
• Protective Motives: The primary motivation for willing participants was protecting their immediate social circle (family, vulnerable partners) and themselves.
• Psychological Reassurance: A negative test result provided peace of mind, serving as a positive motivator.
• Sanctions: The prospect of mandatory quarantine or isolation following a positive test acted as a deterrent (a "sanction") for some.

C. Opportunity (External Conditions)

• External Restrictions: Mandatory testing requirements by employers or for event access generally increased testing rates but simultaneously generated resistance and resentment.
• Accessibility: Physical distance and travel time to testing sites were practical barriers.
• Requests vs. Orders: "Invitations" or clear advice (e.g., via letter or media) were more effective than coercive orders.

D. Capability (Knowledge)

• Knowledge Gaps: Participants often lacked understanding of the broader public health utility of testing (e.g., contact tracing, halting transmission). Knowledge deficits correlated with lower perceived disease severity and lower testing willingness.

E. The Role of Past Experiences (COVID-19)

• Participants heavily referenced their experiences with the COVID-19 pandemic. Negative experiences with shifting policies, perceived government overreach, and vaccine mandates shaped their skepticism toward AI testing protocols. Trust in authorities was fragile and heavily influenced by these prior interactions.

5. Significance and Implications

The study concludes that testing behavior is not disease-specific but is deeply rooted in psychological and social dynamics. The findings offer critical guidance for future pandemic preparedness:

• Balancing Autonomy and Control: Public health strategies must balance the necessity of external measures (to stop transmission) with the psychological need for autonomy. Over-reliance on coercion may backfire, reducing compliance.
• Communication Strategy: Messaging should leverage existing individual motivations (protecting loved ones, personal peace of mind) rather than relying solely on abstract public health arguments, which participants found less compelling.
• Trust Building: Rebuilding trust in institutions is paramount. Communication must be transparent, and the distinction between political government and operational health services should be leveraged, as the latter often retains higher trust.
• Policy Design: To maximize uptake, testing strategies should offer clear, accessible information and, where possible, alternative preventive measures that allow individuals to feel they are making an informed, autonomous choice rather than being forced.

Limitations: The study relies on hypothetical scenarios, meaning actual behavior during a real outbreak may differ. Additionally, the qualitative nature limits generalizability to the entire population, though the heterogeneous sample provides robust thematic insights.

Conclusion: Effective AI outbreak management requires a shift from purely restrictive policies to autonomy-supportive communication that rebuilds trust and aligns public health goals with individual motivations.