Digital Health

Question 1

What is digital health?

Answer 1

A term used to encompass wide range of technologies used for health care, health informatics, health education, health promotion and public health purposes.

eHealth, mHealth, connected health, pervasive health, health 2.0

Digital health has been driving a revolution in healthcare.

Question 2

Why is digital health driving a revolution in health care?

Answer 2

Digital health is driving a revolution in healthcare for it’s potential to improve ability to accurately diagnose and treat disease and enhance the delivery of health care for an individual.

Question 3

What are some digital healthcare milestones?

Answer 3

1897: first long distance diagnosis
1950-1999: growth, with technology development
2000 - 2015: maturation, with the desire to digitize
Future: investment, digital health units and expansion into general healthcare

Question 4

What are the 3 shifts in culture that digital health transformations have facilitated/pushed?

Answer 4

1. Health –> Wellbeing: beyond illness and including psycho/social/physical health, both on community and individual levels.
2. Paternalistic Care –> Collaborative/Patient Centered Care: data accessible to both patient and doctor making for more shared decision making
3. Conventional Medicine –> Lifestyle Medicine

Question 5

What are some theories that talk about why and how new technologies emerge?

Answer 5

Diffusions of Innovations Theory (Rogers 1962), Telemedicine Community Readiness Model /TCRM (Care4Saxony, 2020), Digital Hype Theory

Question 6

What should we remember about the Diffusion of Innovations Theory?

Answer 6

• Any innovation follows a certain parabolic pattern: it requires time for innovations to be picked up by the population and diffuse. It is not linear and does not mean the innovation will be there forever (could be competing and complementary innovations that follow) and most innovations fail to diffuse.
• Factors that affect diffusion include: cost, effectiveness, simplicity, compatibility, the characteristics of the adopters, and the context of the innovation (for example; the social/political climate, salience for need).

Question 7

What should we remember about the Telemedicine Community Readiness Model (TCRM)?

Answer 7

We can use the model beyond telemedicine, it is the assessment of community status/readiness that then allows for improvement measures to be proposed to reach a higher level of readiness.

Levels of readiness are determined by 3 dimensions:
1. Status of telemedicine initiatives
2. Community involvement
3. Evidence for telemedicine in the community

Question 8

What are the digital inequalities that challenge the digital health transformation?

Answer 8

We need an informed and empowered “patient” for the digital health transformation: there are two inequalities that make this hard:

1. Digital Access
2. Digital Capabilities

Question 9

What is the knowledge gap theory?

Answer 9

There is an information gap (those who can access/understand information) which leads to a knowledge gap, and then a participation gap.

Those with higher education will know more about a topic that those with less, for topics that are highly publicized.

Question 10

What are the 5 reasons for the knowledge gap?

Answer 10

1. Numeracy/literacy skills
2. Stored information (ppl w/ higher formal education have built up info/ready to integrate w/ new info)
3. Relevant social contacts/social capital (those w/ more higher formal education have more ppl to discuss topics with)
4. Selective exposure to info (ppl w/ higher education expose themselves more broadly and open to more contrast)
5. Media system (public affairs topics tend to be delivered via print, more easily accessible to more educated ppl)

Question 11

What is first order and second order digital divide?

Answer 11

First order digital divide = differences in access

Second order digital divide = differences in motivation and competencies

Question 12

Is there still a digital divide?

Answer 12

Yes, there is still a digital divide (south/north west/east). We don’t have data on some countries, that likely have less access. Averages within countries could also be misleading. Within countries with access, there is a gap between those who are educated and with higher education having access to internet as well.

Question 13

What is online health information?

Answer 13

Online health information definition is very general, it covers “anything regarding the symptoms, diagnoses, and treatments of different diseases or simply general information about weight loss, healthy diets or wellness tips”.

The questions we should ask are: who searches for online health information and what sources are used.

Examples include Federal Office of Public Health, PubMed, Forums, Blogs, Apps, personalized or tailored info.

Question 14

How is online health information classified?

Answer 14

Online health information is classified by source (user generated content, health professionals, journalists, researchers, etc), message type and language (news, factual information, comments, testimonials, sponsored content, etc), and online format (websites, social media, messages, etc).

Question 15

What are some models/theories related to motivations to seek health information online?

Answer 15

1 - Comprehensive Model of Information Seeking (where information utility is an important mediator)

2 - Planned Risk Information Seeking Model

3 - Situational Theory of Problem Solving (in which personal relevance, efficacy and looking for information is dependent on the situation).

Question 16

What is communicative action, within the Situational Theory of Problem Solving?

Answer 16

We looked at the Situational Theory of Problem Solving in regards to online health information seeking behavior. Communicative action are the many ways in which we interact. It could be how we select information (forefending = blocking, vs permitting), information transmission (sharing/forwarding) or information acquisition (how we process and integrate the information).

Question 17

What is digital or eHealth literacy?

Answer 17

eHealth literacy combines analytical skills (traditional + numeracy, information + media literacy) with context specific skills (health, computer, science literacy).

There are dynamic and contextual factors to eHealth literacy.

Question 18

What is the Transactional Model of eHealth Literacy?

Answer 18

This model outlines a hierarchy of health literacy skills, if the individual has these skills they are more likely to be informed and empowered. This also shows the importance of the context of the information (what info is given, what sources do I have access to? In what language?)

Overall, we also consider noise: could be from a very specific situation, like a patient in extreme pain.

Once the individual is empowered, the virtous circle makes the eHealth contextual factors easier for me.

Question 19

What is the difference between misinformation and disinformation? Fake news and Deepfakes?

Answer 19

Misinformation is false information that is shared without the intent to cause harm.

Disinformation is false information that is knowingly created and shared, to cause harm.

Fake news is fabricated information that mimics the news, that is deliberately meant to confuse people.

Deepfakes is disinformation that is manipulated with the aim to rapidly spread fake information.

Infodemic as a health issue.

Question 20

What are some ways to combat misinformation and disinformation?

Answer 20

Fact checking, (post) debunking, and prebunking/warning/nudging/inoculation.

Fact checking and labelling quality markers could be like the use of the HON Foundation, a foundation that is internationally known for health information ethics (and establishment of HONcode, ethical conduct) for health information online.

Question 21

What is the Digital Hype Model?

Answer 21

The Digital Hype Model looks at how and why new technologies emerge. It shows the cycles.

Disillusionment = the obstales, ethic issues, security, costs and other things we can’t ignore thinking about new technologies. Then, with the slope of enlightenment, we readjust our expectations.

Question 22

What are some characteristics of digital healthcare innovations, per the Digital Hype Model?

Answer 22

Question 23

How are things financed and funded in the digital health industry? Where does the majority of the money go?

Answer 23

Financing is needed not only to develop new technologies, but also to maintain.

Funding comes from investors/venture capital, time/work for shares in company, taxes/reinvestment (public section), pay out of pocket (customers), advertisements (personal data) and corporate foundations.

The majority of the money goes to disease management (like telemedicine). Majority of money goes here because disease management has an established reimbursement system (investor has a secure profit).

Prevention/promotion software development has also grown.

Any detection money goes to AI.

Question 24

What are some of the health care systems/ health insurance models?

Answer 24

Question 25

What is big data?

Answer 25

Extremely large data sets that may be analyszed computationally to reveal patterns, trends and associations, especially relating to human behaviros and interactions.

Question 26

What are the 6 V’s in big data?

Answer 26

The 6 V’s are characteristics of big data. These characteristics give us an idea of the potential, but also the challenges of big data.

Question 27

Where do we use big data in health?

Answer 27

We use big data in health for data driven models (rather than theory driven models). It is used at both an individual level and a population level in a few ways.

Question 28

What is digital phenotyping?

Answer 28

Digital phenotyping is big data at an individual level: it is computerized technology that allows us to look at hundreds of variables and individual health. “The patient in context”

Question 29

What are the data sources for “the patient in context?”

Answer 29

Genotype: genetic code of individual

Phenotype: Expression of the genotype, observable things outside the human (including behavior)

Digital phenotyping: continous quantification of human phenotype, using digital devices (like the smart phone, watches, etc).

Question 30

Where does big data come from in health?

Answer 30

Question 31

What are the uses of digital phenotyping?

Answer 31

Precision medicine:

1. surveillance/self tracking
   (motivation self entertainment, self design, self discipline –> positive outcomes = better informed, reward, motivation, increased consciousness. challenges = privacy concerns, wrong usage/data interpretation, development of obsessive compulsive health thoughts, acceptance of patient generated data by healthcare providers).
2. multimodal profiling for early disease/detection/diagnosis
   (sensors tell us about behavior, which tell us about predictors of disease. The question is always validity and reliability of digital phenotypes.

Digital health is the quantification of our behavior.

Question 32

What are some considerations for digital phenotyping in regards to passive data and active data?

Answer 32

Active data = surveys, diaries
Systematic errors are the issue: for example, desirability bias, estimation bias, recall bias, heuristics, reporting of retrospective situation, etc.

If we move to ecological momentary assessments, (like surveys 1-2x a day, we overcome some of these bias.

Passive data = sensors.
These digital traces overcome these bias.

We’re still research the combination of active and passive data on an individual basis.

Some ppl say this is the way to go, some ppl say be careful.

Question 33

What are some of the challenges of digital phenotyping?

Answer 33

1 - scientific evidence: validity and reliability of digital collected biomarkers, clinical effectiveness

2 - availability: digital divide

3 - acceptance, trust: adoption by patients, researchers and healthcare providers

4 - personal vs public good: data ownership, data sharing, anonymization of data

Question 34

What is machine learning?

Answer 34

Machine learning comes in when we think of what to do with the health data we collect: it is the study of tools and methods for identifying patterns in data. These patterns can be used to either increase our understanding of the current world (risk factors for infection) or make predictions about the future (predict who will become infected).

Decisions/thinking that are data driven, rather than theory driven.

Question 35

What are some of the types of machine learning methods?

Answer 35

Machine learning is an approach that involves various methods of enabling an algorithm to learn:

• supervised machine learning (we create a catalogue/dictionary to help the machine learn to find groups, similarities, categories.)
• unsupervised learning (we ask the computer to find groups, similarities, categories).
• deep learning (artificial brain, w/o human guidance). Deep learning was able to better diagnose disease than healthcare providers in one study, with high sensitivity and high specificity. We can only rely on this tech in some cases, like diagnostics.

-natural language processing (subfield of AI that involves text - semantic understanding, info exchange, translation, ex: Chat GBT).

Question 36

What are some things to keep in mind when looking at big data in health and diagnostics, regarding the impacts and potential bias?

Answer 36

Veracity: where is the data coming from?

We need governance to understand better the impacts, bias.

Fairness: a lot of data comes from WEIRD population.

Accountability: who is liable for mistakes? The doctor, the vendor, the government?

Question 37

What are the 4 pillars/principles of ethics?

Answer 37

1 - Respect Autonomy
2 - Beneficence (benefits)
3 - Non-Maleficence (prevent harm)
4 - Justice (accessibility, equality, equity, fairness)

Question 38

What are the moral issues with health research?

Answer 38

Health research, which aims to improve the health/life of individuals and draws special attention to the most vulnerable populations: entails moral issues because it is about deciding what is right or wrong and is about asking someone to give up something/ accept some risks.

1- Consequentialism/Utilitarianism
2 - Norms/Principles
3 - Virtues

The main point is that ethics is not black/white, everything (study, person, etc). is unique.

Public health ethics can be seen as both the application of principles and norms to guide the practice of public health and as a process for identifying, analyzing and resolving ethical issues inherent in the practice of public health.

Question 39

What are some of the opportunities of big geo-social data?

Answer 39

It is easily accessible, steadily growing, mirrors real world offline trends. It also helps us to understand the relationship between the place based social determinants of health and the given health indicators. It helps us to identify where communicable and non-communicable disorders cluster to inform timely and targeted prevention and intervention strategies. (For example based on google searches or flu related tweets).

Question 40

What is ethical deliberation?

Answer 40

1 - identifying and clarifying the ethical dilemma posed
2 - anaylzing it in terms of alternative courses of action and their consequences
3 - resolving the dilemma by deciding which course of action best incorporates and balances the guiding principles and values

Question 41

What is trust and why is it important? (Thinking about ethics)

Answer 41

Trust is believing someone will/will not do something in a competent/honest way. It is reliability.

Trust is important because it gives us legitimacy, and participation. (If the company/study/research is trusted, they ppl will participate and engage in the study/program/platform).

You can build trust in an app/platform, by looking through the mHAT checklist, which includes brand familiarity, accuracy, understandability, privacy features, user control, recommendations from friends.

Question 42

What are the 3 things that we learned in the ethics conversation?

Answer 42

1 - Trust in digital health is key

2 - Ethics in this space is a grey area

3 - It is important to bring on many ppl/ideas, to encourage conversation re: ethics for studies (especially in using data online). Is data online a public place?

Question 43

What are some features of social media platforms? What are things we should consider when researching them?

Answer 43

Are they:
text vs visual
private vs (semi) public
ephemeral vs forever
active use vs passive use

It is difficult to research social media, the platforms change quickly, it is hard to get data and many ppl use multiple platforms (also based on age, generation, etc). The public concerns are for the adolescents: because they use social media heavily and are vulnerable.

There are two perspectives looking at social media + digital health:
social media FOR digital health
social media AND digital health

Question 44

In class, when we discussed the positives and negatives of social media in regards to health, what came up?

Answer 44

Typically in research and popular discussion there is a negative bias.

Positive features of social media were more linked to social health (community building, etc).

Negative features of social media were more linked to its impact on physical health (lack of movement, screen time, eye sight, etc).

Same behaviors could be negative or positive, it depends on the platform, type of use and the individual.

While studying, we don’t differentiate the platform and the context. It is also hard to know the cause and the consequence (for example, is SM to blame for more anxiety or is it more political unrest, pandemics, etc that is to blame and ppl turn to SM?).

Question 45

What are some types of online social support?

Answer 45

Informational (info/advice)
Emotional (empathy/affection)
Instrumental/tangible (offer services)
Appraisal support (supplying practical resources for self evaluation

These are used for measuring quality of care (in a caregiver for example).

Question 46

What are some pros/ cons of social media health information?

Answer 46

Pros: allows ppl suffering from mental health issues the opportunity to read, watch, listen to and understand health experiences - relating them back to their own reality. Provides increased interaction with others, increased accessibility.

Cons: quality concerns and lack of reliability, confidentiality, privacy and information overload.

Question 47

What are some of the opportunities and concerns with health influencers on social media?

Answer 47

Opportunities: intimacy and trust with hard to reach audiences with relational and affective content, wide audience, timely communication.

Concerns: “deprofessionalization” of health expertise (dr. in white coat vs someone more approachable/less professional), trivialisation of health problems/pandemic, blurring boundaries between experts and laypeople, issues around patient privacy, concerns about time demands.

Question 48

What are some mediators of the impact of social media on health?

Answer 48

1 - Social anxiety: adolescents are more likely to be embarrassed, leading to social anxiety. Social media can cause anxiety, fo example it can lead to social comparison (inspiration which is useful for health behavior change, or despair).

Social anxiety can also cause an individual to feel safer on social media, or social media use. It is bi-directional/ a loop.

2 - FOMO: social media positivity bias/ best version of us. Also circular, SM fosters FOMO, and then further increases it. Lower mood/lower life satisfaction.

3 - Body Image: appearance focused social media, individual treated as a body: media -ideal internalization and comparisons, leading to unhappiness with body, shame, restriction, etc.

4 - Depression: constant comparison with others, compensate for and escape from real life, but also can get information they need and support when they have a lack of resources in real life.

5 - Loneliness: does it maintain relationships or change them?

Question 49

What are the two contradictory frameworks for the mediator of loneliness, when looking at SM and health?

Answer 49

1 - Internet enhanced self disclosure hypothesis: SM allows us to maintain relationships, it is an extended space for socialization, gives a sense of belonging, self disclosure

2 - Evolutionary Mismatch Model: We are social beings, internet related behaviors take us away from in person relationships. Strong vs weak ties.

Question 50

What is digital public health, for the individual?

Answer 50

Digital public health for the individual is digital phenotyping: it is the quantification of our behavior. We have passive data and active data, that then can lead to the digital health interventions. Allows for the ability to tailor digital health interventions. Can be used to describe, diagnose, so on…

Question 51

What is a digital health intervention?

Answer 51

Intervention delivered via technology. It is an added value on an intervention that helps one to achieve health objectives, and can help overcome obstacles like access, or equity + be more efficient. From a patients prospective, they can have access to information and care, support in health promotion and disease prevention and support in disease self-management and treatment.

Question 52

What is important to consider when appraising digital health interventions?

Answer 52

It is important to appraise digital health interventions (DHI) in order to find which solutions add something/ are valuable and which are just part of the hype. We look at the need (the problem), the benefits, accessibility, acceptability (will ppl adopt it? is it perceived as useful and easy to use?), what components of the intervention are helpful, the effectiveness & efficiency, the potential harms and feasibility and utility.

Question 53

What do the guidelines proposed that help us evaluate digital health interventions include?

Answer 53

The some of the guidelines from the App Evaluation Model: American Psychiatric Association, mHealth Assessment include:

Choice of comparator: what is the DHI an alternative to? What is the unmet medical need?

Multi-stakeholders’ perspective: simultaneous assessment of DIH’s benefits from view of patients, clinicians, payers, and health care managers.

Organization impact: the health care system’s preparedness/ readiness for the intervention

Multidimensional outcome: quantification of the performance of the digital health solution. Does it make an impact/work?

Interoperability: does it connect to other data sources?

Question 54

What are the two ways we discussed to evaluate a digital health intervention?

Answer 54

One way discussed were some guidelines from the App Evaluation Model, which includes an assessment of questions like what is the comparison to/unmet medical need, is the healthcare system ready for the intervention, does it make an impact, does it connect to other data sources, etc.

A second way we discussed was looking at this pyramid, in order to determine the effectiveness and efficiency compared to alternatives, starting with the stability and usability of the technology, moving to the user experience, the implementation and finally the summative evaluation to determine if clinically effective.

Question 55

What is the issue of the black box?

Answer 55

In digital interventions, the issue of the black box is the inability to distinguish the effectiveness of certain features included in a digital intervention, you don’t know what particular feature worked, for example: chat bots, reminder system, communication/support, decision aids, motivation/action plans, gamification, etc.

Question 56

What is digital public health, at the population level?

Answer 56

Digital public health, at the population level, is a reimagination of public health, with digital concepts. Digital becomes an asset to overcome gaps, be more efficient and work towards equity. Examples would be surveillance (i.e. how does disease spread and what are risk factors) , also evidence collection and research (like the access to systematic reviews and summarized evidence on PubMed, WHO, etc). and also could be regulation, quality control (checking standards) and support ( services like clinician decision support tools, education of healthcare workers, empowerment).

Question 57

What are some pros and cons for the use of chatbots in a digital health intervention (DHI)?

Answer 57