Lecture T1 Digital Health Communication Persuasive Technology
Learning goals in this lecture:
• Understand the main factors increasing and decreasing the impact of digital health applications;
• Understand and explain the importance of theory use in the development of digital health
applications.
Learning goal 1: Understand the main factors increasing and decreasing the impact of digital health
applications
More and more organizations develop digital health applications, which suggest they will be (cost-) effective.
But are they actually effective? Is there evidence? Are they just gadgets or do they help us? To know this, we
look at several meta-analyses (= a summary of several studies and their outcomes).
Physical activity
A systematic review and meta-analysis aimed to determine the effectiveness of smartphone apps for increasing
objectively measured physical activity in adults.
Cross-sectional studies were excluded because we know that correlations between app-use and physical
activity could be confounded/there is risk for self-selection (for example: people that already move a lot tend
to think they do not need such an application).
Therefore, only randomized controlled trials are included: people are randomly assigned to either an
experimental or control group.
Studies were also excluded when self-reports were included (they can be biased, for example with socially
desirable answers or people may just forget what they actually did for physical activities) or when smartphone
apps were not the primary component to intervention (since this was the end goal to measure: what is the
effect of the apps on physical activity?).
Overall, we see that the general level of the studies is not that good: there are not a lot of randomized control
trials (and thus that measure the level of physical activity objectively).
Which theories were used in these studies and the development of these apps?
- In 4 of the 7 included apps, theory was actually used (behaviour change theory);
- The most employed theory was the social cognitive theory.
When you look at the apps, all of them had a visible display of steps. Some of them had physical activity of
performance summary, goal setting, visual display of goal achievement and/or motivation prompts.
What did this very robust meta-analysis found (only including the best studies that only included RCT studies
and thus objective studies)?
→ overall, the studies did not show significant effects of app use on physical activity
→ however, the interventions were effective (significantly) when the intervention duration was 3
months or less (compared with longer interventions)
How come that a long usage of an application does not have an effect?
→ engagement in such an intervention can decrease over time. Think about yourself: do you ever use an app
more than 3 months?
→ physical activity apps that targeted physical activity in isolation/alone were more effective than
apps that targeted physical activity in combination with diet or alcohol intake
,This was also a finding in the paper of Camerol et al.!! A possible explanation could be that focusing on more
health behaviours could delute (= verdunnnen) the effects, as participants may choose to focus on a single
behaviour. Therefore for some the intake of fruit and vegetables grows, for others their smoking behaviour
decreases, but overall they find small effects.
To summarize:
- No effect overall of smartphone applications on physical activity;
- But effects of shorter interventions and those only focusing on 1 behavior seem to be stronger.
Eating behaviour
A systematic review and meta-analysis aimed to determine the effectiveness of smartphone apps for changing
nutrition behaviors and nutrition-related health outcomes.
- A little bit less strict in this meta-analysis, also non-randomized control trials were included for
example.
- Also self-reports included.
The majority of the apps used techniques like feedback & monitoring, but also social support and goals &
planning. Most apps included 4 different Behavior Change Techniques.
Results of the meta-analysis:
- Overall, a small significant effect of app-based mobile interventions on nutrition behaviors and
nutrition-related outcomes;
- Only studies targeting short term (less than 3 months) and/or intermediate-term (3-6 months) follow-
up intervals yielded significant (small) effect sizes; effects of long term (more than 6 months) follow up
not significantly.
So, we cannot really say that in general digital health applications are super effective. You can say that the
results are mixed. There is still a lot to be done for designers and developers of these apps!
We saw that the impact of interventions is very mixed: there is no clear evidence for the impact of digital
health applications yet.
But, which factors are important to consider for maximizing this impact?
1. You should have an effective ingredient: the effective components of the intervention should be
evidence-based.
This means: The application employs features that target determinants from health behavior theories
, / features that are based on established behavioral change techniques which are shown to actually
have an effect in earlier research.
2. You should provide this in an effective dose: the uptake of an application should be sufficient, people
should really use it (not only once). An effective app that is not used will not have an effect.
Thus: The app should be designed based on scientific theories of technology acceptance and
engagement. The app should be human-centered.
These ideas are also the idea behind the CEHRES Roadmap = a holistic framework to improve the uptake and
impact of eHealth technologies. It is all about effective ingredients and sufficient use. It is a manual to develop
impactful interventions.
Learning goal 2: Understand and explain the importance of theory use in the development of digital health
applications.
Many apps include many different features, so how do we know which features are actually causing the
effects?
Again, another meta-analysis: the extent of theory use in digital health apps.
In non-digital health interventions: effectiveness of intervention is higher
when theory-based and when it employs one or more behavioral change
technique.
Aims of the meta-analysis:
• Provide overview of extent of (= mate waarin) theory use in digital health applications;
• Determine importance of the extent of theory use.
So, are digital health applications that are developed based on theory more effective than those that are not?
It is not simply black and white that applications are based on theory or not. There are different levels of theory
use:
Level 1: Mentioning/referencing a theory
- Just mentioning or referencing a theory (superficially);
- Explicitly describing the relationship between the theoretical constructs and the target behavior of
interest;
Level 2: applying the theory in the intervention development
- For example: using theory to tailor intervention techniques to recipients.
Level 3: Measurement and testing of theory
- For example: measuring theory-relevant constructs before and after the intervention (so for instance
an intervention that is meant to increase self-efficacy and thus measure the self-efficacy measuring
prior and after use of the application to see if the application indeed impacts self-efficacy).
, Level 4: The study actually contributed to building and/or refining the theory
- Discussing findings in relation to the theory → get back to the theory in the discussion and explain the
effect of the application/intervention on the theory/theoretical constructs
- Measuring the theoretical construct (prior and after the intervention) and statistically tested if these
construct could explain the effect of the interventional target behaviour.
- Giving suggestions for future refinements of the theory.
Summary:
- The extent of theory use in studies about digital health applications could be higher;
- This is really important! The meta-analysis showed that a significant positive effect of the digital health
intervention on weight-related outcomes when extent of theory use was higher, namely:
o at least one or more theoretical constructs are explicitly linked to an intervention
technique
o when theoretical constructs are included in evaluations (measured pre and post
intervention)
Why is this the case? Why are theories so important?
- Theories are the key to effectively organizing the information in a meaningful way, such that it
accounts what is known about moderators and mechanisms of the target behaviour and therefore
interventions are more likely to include effective components (since the chance is bigger that the
authors used thorough literature use);
→ as digital health applications can provide large amounts of data, they have a great potential to provide
scientific data on the effectiveness and these factors causing the effects. This can help to further refine
theories.
Many of the applications that you will find in the Appstore are not evidence based. How can users know that an
application is evidence or theory based? → The GGD AppStore only offers evidence based applications!
Preparation assignment U@Uni digital intervention → papers of Epton and Cameron
- Multi-device digital health intervention to stimulate health behavior (e.g., fruit and vegetable intake,
physical activity, alcohol consumption, smoking) among young people during the transition from
school to university
- Freshman 15 = the weight gain of first year college students (which is shown to be quite high because
you move out of your parental house, drink more alcohol, etc.)
A lot of studies focus on this period, because:
- Highly relevant for public health perspective (since weight gain is not healthy);
- It offers a good case to study the prevention of behaviour change/unhealthy behaviour because most
people are quite weight stable, but this is a sensitive period that we know people are likely to gain
weight and therefore also the effect of preventive measures and interventions can be established.
How did the intervention look like?
- A personal homepage with all kind of information about health behaviours with links to these health
behaviours that are bad for your health and why there were bad for your health;
- Possible to make plans on the website;
- Possible to link to your social media like Twitter;
- Several videos, nice to know facts, etc.
- Facebook page with several information;
- Android application with all kinds of information that targeted different determinants of the theories.
Think about the following questions:
1. What was the extent of theory use* in the digital application/intervention described in
the papers of Epton et al. (2013) and Cameron et al. (2015)?
• Mentioning/referencing a theory: yes/no
Which theories: self-affirmation theory; theory of planned behaviour; implementation
intentions.
