The Chief Public Health Officer's Report on the State of Public Health in Canada 2014 – Digital technology as a tool for public health

Digital Technology as a Tool for Public Health

As technology has evolved, so has its use and application to public health. The telephone has gone from being a tool to transmit electrocardiograph data early in the 20^th century to a way for patients to receive health information and advice from health professionals remotely, via telehealth systems.^{Footnote 345,Footnote 346} Canadians have become accustomed to the technology that routinely surrounds them when they access health services or programs, from computerized health information systems and electronic medical records to diagnostic tools and treatment equipment. Personal computers and mobile device apps are used to connect with health-related information and tools. Behind the scenes, scientific researchers make use of technology to develop new methods and tools for public health and also to create new technology itself.

Technology can provide public health professionals, communities and individuals with a wide range of tools to address issues of public health more efficiently and quickly in a more connected way. As technology continues to advance, its role in public health will also change. The vast range of available technologies will continue to be used in educating, informing, training and communicating with both the public and public health professionals; in surveillance and data collection for detecting infectious disease outbreaks; in monitoring of chronic disease and injuries; in monitoring and evaluating programs; in making transparent and evidence-based decisions; to improve the speed and accuracy of diagnoses; and in providing new and more effective treatments. New and innovative uses of technology for public health will also emerge as the field continues to grow and evolve.

This section presents just some of the ways in which certain digital technologies can be and are being used to address several key functions of public health.

Health promotion and protection

Health promotion is a core function of public health. Public health practitioners and professionals work with communities, agencies and individuals to develop and implement programs and interventions aimed at positively influencing health behaviours.^{Footnote 347}

EHealth interventions

Once a program or intervention is developed it must then be delivered. Technology can help by providing the means through which such programs and interventions are offered. In particular, eHealth interventions allow public health to benefit from current technologies.^{Footnote 348}

EHealth interventions can take numerous forms and address a vast range of issues.^{Footnote 348} To provide a manageable and focused assessment, the review excluded eHealth interventions involving telehealth, telemedicine, television or radio, use of electronic health records, gaming or videogames, and personal wearable devices, and interventions were narrowed to include traditional public health functions.^{Footnote 348} The literature identified studies in the United States, the Netherlands, the United Kingdom, Australia and Canada that used websites and web portals; email and text messaging; goal setting; assessment and monitoring/tracking; risk assessments; online training, counselling/motivational interviewing; tailored feedback and peer and expert advice; and social network sites, live chats and discussion boards as intervention tools.^{Footnote 348} The public health issues addressed through the interventions included diet, nutrition, healthy weights, increasing physical activity, smoking cessation, sexual health promotion, immunization uptake, substance use reduction and general and/or multiple lifestyle issues.^{Footnote 348} The intervention approaches were classified into three categories: web-based; mHealth (mobile technologies, such as cell phones, tablets etc.); and computer-based (stand-alone computers without Internet, such as a computer kiosks or CD-ROMs).^{Footnote 348}

The studies included in the review mostly targeted adults from the general population as well as adolescents, college/university students and adults in workplace settings.^{Footnote 348} A large number of the studies targeted youth in relation to improving healthy eating and physical activity.^{Footnote 348} Young adults were also most often involved in studies related to sexual health such as increasing screening for sexually transmitted infections (STIs) and adopting healthier sexual behaviours.^{Footnote 348} Employee interventions targeted multiple lifestyle behaviours, mostly related to healthy eating and increasing physical activity.^{Footnote 348} Only a few studies analyzed interventions with seniors; one examined seniors' use of mobile phones and another focused on a web intervention, with both aiming to increase older adults' level of physical activity.^{Footnote 348} In light of the aging demographic in Canada, the increased use of the Internet among those over 55 years of age, the lack of interventions for this population and the promise shown in the above studies, this is a potential area for growth.^{Footnote 348}

Canadian studies focused mostly on web-based or online (enhanced and/or interactive websites or web portals) programs and on one texting intervention.^{Footnote 348} The studies addressed a variety of health topics: uptake of vitamin C, reduction in alcohol use, reduction in smokeless tobacco use, increase in physical activity, general health-seeking behaviours, reduction in drug use and increased access to information on sexual risk behaviours and testing for STIs (see the textbox "Using technology to increase testing for sexually transmitted infections"). Interestingly, over half of these studies targeted adolescents and young adults.^{Footnote 348}

Overall, study results indicated that eHealth interventions are feasible in controlled settings and that individuals are generally open to using technology to monitor and improve health behaviours, attitudes and beliefs.^{Footnote 348} In some cases, certain features of the technology, for example, brief, relevant and positive text messages, were better at enhancing engagement.^{Footnote 351}

The effectiveness of interventions varied based on the health topic. Alcohol use reduction/prevention studies and sexual health studies showed mostly positive and significant outcomes, whereas smoking cessation and multiple risk factor studies showed both positive and negative outcomes, depending on the type of intervention.^{Footnote 348} The eHealth intervention studies associated with the most positive and significant outcomes were those that mixed components (e.g. website plus face-to-face support), tailored components to individual needs (personalized feedback or advice) and integrated behaviour change theory, irrespective of health topic.^{Footnote 348} In addition, many studies showed effects over the short term (less than 6 months), but longer-term results were lacking.^{Footnote 348} In assessing the level of effectiveness of such interventions, reported study limitations such as self-reported outcome measures, limited ability to generalize of results due to small sample sizes, high attrition rates and loss to follow up must be kept in mind.^{Footnote 348}

As with other uses of technology as a public health tool, accessibility should be taken into account to ensure equitable program and service delivery. For example, in all but one of the mHealth intervention studies, participants had to have a cell phone, and in some cases, a data plan.Footnote 348,Footnote 352 Using text messaging can potentially marginalize populations with low literacy levels or without access to a mobile phone.Footnote 348 Interventions need to match the target population and be integrated into the pool of public health interventions.Footnote 348 EHealth strategies need to deliver health information through culturally and socioeconomically acceptable formats; increase skills in delivering accessible health information; and engage online social network members who assume leadership roles as well as trusted people (e.g. elders).Footnote 348,Footnote 353 To address health equity concerns, factors that need to be considered include lack of physical access to technology; lack of meaningful access (information needs to be designed to reach and appeal to diverse populations); language abilities and eHealth literacy skills; age; disability; and cultural relevance of tools.Footnote 348,Footnote 354 It is critically important to engage end users in co-designing interventions to ensure relevance, uptake and sustained use of eHealth interventions.Footnote 348

There is definitely room for careful application of eHealth interventions in public health in Canada, with attention needed to address challenges to health equity. There is also a need to support further research of eHealth applications in the Canadian context.^{Footnote 348} New interventions, including some that have already been implemented, continue to be developed (see the textbox "ImmunizeCA").

Looking at the way in which technology is being used to encourage health-promoting and protecting behaviours, it is clear that the potential exists to improve use of social networks.^{Footnote 348} Some individual health behaviours and resulting chronic conditions, such as smoking and obesity, have been shown to be "contagious" to some degree, that is, they tend to "spread" within social networks.^{Footnote 360-362} If an individual has social connections to others who smoke or are obese, they are more likely to smoke or be obese themselves.^{Footnote 360-362} Knowing this, social media in particular could be used to identify public health programs and interventions and target individuals, rather than an entire network, with the intention that individual change will spread to others.^{Footnote 360}

The Internet is an extremely common source of health information in North America.^{Footnote 366,Footnote 368} A 2012 survey found that three-quarters of adults in the United States searched for health information online in the previous year.^{Footnote 369} In Canada, 70% of home Internet users searched for medical or health information online in 2009, up from 59% only two years earlier.^{Footnote 368} With more than 200 million monthly active users worldwide on Twitter and more than one billion on Facebook, social media is positioned to connect millions of Canadians with important public health information.^{Footnote 370,Footnote 371}

Although social media is popularly believed to be the domain of younger people, research in the United States found that almost two-thirds of Internet users aged 50 to 64 years and almost one-half of those aged 65 years and older were using social networking sites such as Facebook and Twitter in 2013.^{Footnote 372} Consequently, it only makes sense that health organizations reach out to this ready-made audience.^{Footnote 366}

Aside from its built-in user base, social media has a number of features or aspects that make it an effective means of sharing health information. For example, minimal effort is necessary to share information in real time.^{Footnote 373} Organizations can take advantage of the existing infrastructure, which may make using it inexpensive.^{Footnote 373} The multidirectional nature of the communications allows for quicker and more extensive distribution of the information than traditional methods.^{Footnote 373}

Simply having access to the information, however, is just one aspect of health literacy; the user must also be able to understand and evaluate the information to make appropriate health decisions.^{Footnote 267} The information and approach must therefore be adjusted to users’ levels of reading and comprehension.^{Footnote 366} In addition, knowing how family, social context, culture and education play a role is key.^{Footnote 374},^{Footnote 375} These influences can affect how individuals receive health information, perceive health problems, express their symptoms and their views on what, and by whom, treatment should be provided.^{Footnote 375Footnote 376}

Exclusive reliance on text in health messages can be a barrier to health literacy.^{Footnote 377} Online health information providers should also consider alternative and complementary media formats—photographs, illustrations, animation, video, live seminars and interactive games—and learning environments that allow Canadians with literacy or language issues to more easily understand the information (see the textbox "T2X: Getting the message to teens").^{Footnote 374,Footnote 377}

The strength of social media is the ease with which users can exchange information, rather than merely passively receive one-way statements.^{Footnote 382} Dynamic back-and-forth interchanges can take place with peers or with healthcare professionals. People find it helpful to discuss symptoms, treatments and concerns with their peers with similar health issues or similar questions.^{Footnote 377,Footnote 383} Social media allows Canadians to do so from the comfort and privacy of their own homes. This allows them to ask questions they might be too embarrassed to ask their doctors.^{Footnote 377} Communicating with people with whom they perceive they share a problem can help patients open up.^{Footnote 377} In addition, it may be easier to understand information shared by peers than the jargon used by some healthcare professionals.^{Footnote 377} People can connect with each other where otherwise they may never have due to barriers in geography, distance or the rarity of their condition.^{Footnote 383} This access to a broader range of information can increase their knowledge and help them make better, more informed decisions about their health.^{Footnote 377,Footnote 383} Even between peers the information exchange can go beyond text to include photos or video. For example, a fertility specialist in the United States asked one of his patients to produce a video showing her self-administering her daily in vitro fertilization (IVF) injections. The video was then posted on YouTube for other patients to watch to help them overcome their fears and gain confidence in their own abilities.^{Footnote 377}

The aspects of social media that make it well suited as a positive tool for health literacy are also its main drawbacks. The speed and range of messages that pass through social media, along with its largely un-moderated format, create potential risks.^{Footnote 374,Footnote 377} The risk of spreading inaccurate, biased or incomplete information or misinformation is considerable given that basically anyone is allowed to post anything they want.^{Footnote 366,Footnote 374,Footnote 377} The source of the information may be missing, thus preventing assessment of the credibility of the information.^{Footnote 374} The information may be correct, but written in such a way that it could be misunderstood and have negative health consequences.^{Footnote 366,Footnote 374} The viral nature of social media then allows this information or misinformation to spread very quickly, especially during times of public anxiety or fear.^{Footnote 366,Footnote 374}

Health organizations may also consider privacy issues and the value of employee time as reasons against adopting social media as a tool for public communication.^{Footnote 384} They must also be careful not to cross the fine line between providing general information and giving specific medical advice in a public venue.^{Footnote 384}

Some of the risks associated with social media can be lessened, however, by building trustworthy websites and by monitoring and moderating content. Official websites such as those of the Public Health Agency of Canada, the United States Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO) can be promoted to Canadians as trustworthy and credible. The content of social media websites should be monitored to prevent spam, malicious content or privacy violations without eliminating the individual user's ability to freely post content.^{Footnote 374}

Social media tools are only of use if people are able to access and use them, which not all Canadians can.^{Footnote 377,Footnote 385,Footnote 386} Barriers to access include cost, location, low literacy levels, disability and factors that relate to people's capacity to use these technologies appropriately and effectively.^{Footnote 377,Footnote 386} These are the barriers faced most often by some of the most vulnerable Canadians who are also some of those most in need of improved health literacy. They include older Canadians, people living in low-income households and those living in remote and northern communities, which may include First Nations, Inuit and Métis people.^{Footnote 387}

Moving past the barriers requires improving access to technology and, when necessary, tailoring the information and messages to those with limited health literacy. In terms of access, one solution is to make the technology publically available in schools, libraries and health clinics or doctors' offices.^{Footnote 377} In fact, most schools and libraries already offer public computer use; some may also have staff on hand who can assist or instruct users.^{Footnote 377} Placing accessible computers in clinics or other places used for health-related reasons would allow people to access information that may be pre-screened or recommended by knowledgeable staff.^{Footnote 377} For those with low general literacy or physical limitations that may make traditional browsing difficult, touchscreens with visual cues and icons could help eliminate those barriers.^{Footnote 377}

Surveillance

Surveillance—the collection, analysis and reporting of data in order to track and forecast health events and determinants—is central to any public health system.^{Footnote 325} Through surveillance public health officials can identify and respond to public health threats, create practical, evidence-based policies and programs, and meet Canada's international public health obligations.^{Footnote 325}

The Internet and social media's value as surveillance tools lies in the fact that changes in information and communication patterns on the Internet can act as an early-warning system for epidemics and outbreaks or other changes in population health.^{Footnote 389-391} Conversely, public health officials can use data about the information being shared by the public to target public health campaigns, educate, correct misinformation and alleviate fears.^{Footnote 390,Footnote 392} This is what WHO did during the days following the Japanese tsunami in 2011.^{Footnote 393} Social media was used as a communication tool to allow WHO to manage critical health information being shared with the public after the incident (see the textbox "Fukushima radiation threat: Informing the public through social media").^{Footnote 393}

In addition to informal web and social media posts of members of the public, public health surveillance can make use of online secondary data such as news reports, expert newsletters and aggregate information that has already been synthesized, analyzed and/or reported.^{Footnote 390,Footnote 394} A number of Internet surveillance systems that use a more selective approach and choose high quality, expert-curated secondary data sources are in use. These include the Global Public Health Intelligence Network (GPHIN), developed in Canada in 1997, and HealthMap, developed in the United States in 2006.^{Footnote 389,Footnote 390,Footnote 394} These systems use an automated process to monitor and analyze online sources, facilitating early detection of global public health threats (see the textbox "Global Public Health Intelligence Network").^{Footnote 389,Footnote 390,Footnote 394-396}

Syndromic surveillance

Syndromic surveillance, just as its name suggests, aims to track symptoms associated with a defined syndrome, such as influenza-like illness or acute respiratory illness, rather than depending on laboratory-confirmed disease data.^{Footnote 398-400} Traditionally, this surveillance is carried out through reports from official sources such as emergency departments or primary care providers.^{Footnote 398,Footnote 401} Nowadays, however, members of the general public can also contribute directly to these data through social media.^{Footnote 399,Footnote 402} While people discuss their symptoms, illnesses and state of health with others in open forums, they are providing additional information for these surveillance systems.^{Footnote 399,Footnote 400}

One area that has benefitted from the use of social media-based syndromic surveillance is influenza monitoring. Although diagnostic tests can be used to identify individual cases of influenza, the prevalence of influenza in the population at any given time can only be estimated.^{Footnote 402} Those estimates often rely on syndromic surveillance to capture symptoms related to influenza-like illness (ILI). However, not all people experiencing symptoms will visit an emergency department or physician, so traditional methods do not capture these data.^{Footnote 401} In addition, traditional systems rely on a limited number of sentinel sites for reporting and delays are frequent.^{Footnote 401} Capturing user-generated ILI-related information on the Internet may provide a more complete real-time picture of influenza trends and cases.^{Footnote 401-403}

Google has taken that concept and created Google Flu Trends (GFT).^{Footnote 403,Footnote 404} The idea behind GFT is that people with ILI perform search queries on their symptoms to find information that helps them self-diagnose. GFT tracks the search queries looking for keywords and phrases related to ILI. In fact, the estimates of influenza activity from the system have been shown to be highly correlated with data reported by official public health organizations around the world (see Figure 5).^{Footnote 403,Footnote 404}

Figure 5 Canadian flu activity and Google Flu Trends, the week of September 28, 2003, through the week of March 30, 2014^{Footnote 405-408}

* Influenza-like illness per 1,000 patient visits.
† Estimates based on Google search data.

As good as these web-based systems are, however, they are not as accurate as traditional systems and can lead to incorrect estimates.^{Footnote 409} In 2009, GFT greatly underestimated the prevalence of ILI at the beginning of the H1N1 pandemic in the United States.^{Footnote 410} Despite altering its algorithm after Pandemic H1N1, GFT severely overestimated peak flu levels for early in 2013 in the United States.^{Footnote 409} Yet even with their limitations, systems such as GFT can still provide useful information. However, they are not meant as a replacement for traditional surveillance and reporting systems, but rather, they should be seen as complimentary to them. In fact, research has shown that a combination of data from both GFT and the CDC in the United States produced better estimates than either system on its own.^{Footnote 409}

As people's Internet browsing behaviours only record proxy measures of influenza, they cannot measure other influenza-related data such as healthcare usage or the clinical spectrum of cases.^{Footnote 401} This type of data can be collected, however, by having individuals use an Internet-based questionnaire to record their specific symptoms and healthcare usage over time.^{Footnote 401} These Internet-based cohort systems have been implemented in several European countries, namely Belgium and the Netherlands (under the name "Der Grote Griepmeting"—the Great Influenza Survey), Portugal ("Gripenet") and Italy ("Influweb").^{Footnote 401} These systems have been shown to provide results in line with those from traditional surveillance methods and may be able to detect increased influenza activity more rapidly.^{Footnote 401} Such systems can work alongside traditional systems to provide complementary additional information.^{Footnote 401}

Infectious disease outbreaks and trends are not the only things that can be captured through the mining of Internet and social media activity. People's online searches, discussions and postings can also provide information useful in identifying issues surrounding chronic diseases, including mental illness.^{Footnote 411,Footnote 412} For example, social media mining can show pockets of poor mental health or mental illness in communities, thereby allowing the relevant health service to develop early intervention strategies.^{Footnote 412} Researchers in Taiwan found suicide to be correlated with lottery sales as an indirect indicator of feelings of hopelessness at the social level.^{Footnote 413} A South Korean study tested the hypothesis that social media data such as weblog contents are more promising sources to gauge the public mood than activity such as lottery sales. The study found a significant association between social media data and the national suicide rate and resulted in a predictive model that could potentially be used to develop new models for use in forecasting and prevention of suicide.^{Footnote 414}

Continuing efforts

This section has only touched on a very small piece of the picture in terms of the countless possibilities for the relationship between public health and technology. For instance, the idea of technology as a public health issue itself, such as the dangers of texting and driving, the impact of social media on mental health or the change in youth interaction via social media as it relates to bullying, also needs to be addressed as technology continues to permeate more and more of Canadians' everyday lives. In addition, this report has not explored the very important ethical, legal and privacy concerns surrounding the use of social media as a source of data. These are legitimate concerns that must be brought to the forefront of any planning and discussions around this topic.

There are endless ways in which public health and technology do, can and will come together. Although the role of technology in public health in the future is as yet unknown, Canadians can take a cue from today and try to plan for and guide the relationship.