Developments in consumer health informatics in the next decade

Making predictions about use of technology is dangerous, as Professor D. Hartree would testify ‘We have a computer here in Cambridge, there is one in Manchester and another at the National Physical Laboratory. I suppose there ought to be one in Scotland, but that’s all’ (Prof. D. Hartree 1951 1). Nevertheless, this article discusses how the provision of health information for the public and patients may change over the next decade. To take full advantage of technological advances, alternatives need to be tested for acceptability, efficiency and cost-effectiveness. We also need to identify learning needs, or organizational changes, that will enable best use to be made of these new systems. This article considers these issues under four headings: (a) methods and places in which patients and the public may obtain information; (b) developments in the provision of information; (c) tailoring information to the individual patient’s needs; (d) integration of consumer health informatics with clinical practice. Health professionals have always been a source of health information. Family members and friends are very often the primary source, with television, newspapers and magazines playing an increasing role. Use of specialized telephone services, books and computers is likely to expand over the next 10 years. People are becoming accustomed to using telecommunications and computing technology. In 1994, 7–9%2,3 of households were without a telephone, in 1996 5%, and currently we could assume that this is now less. There has been a dramatic rise in the use of mobiles, with press claims of 20 million users. Use of telephone banking and insurance has also grown in recent years. Telephone health information services have existed since the 1980s. 4 Political backing for a one-stop telephone service, NHS Direct, will ensure an expansion of availability and use. 5 Use by patients and professionals is increasing, 6–9 but evaluation of NHS Direct showed that older people are less likely to use it. 10 Issues of cost, physical barriers (such as hearing difficulties) and readiness to adopt new technologies, may leave certain sections of society more disadvantaged. While, the World Wide Web (WWW) is a good source of health information and support, and there is evidence of benefit for patients and their families, 11 three barriers to use may mean that the Internet perpetuates inequalities of information access. 12,13 Technical problems. The interface, including both hardware and navigation, are important and may be different for naïve rather than regular computer users. Prevalence of previous computer use amongst older people in Scotland is still not high; for example in 1996, 14 only 20% of gastroenterology outpatients had used a computer more than a few times. In an American study 15 about 20% were unable to use an Internet connected computer on their own. However, an American study has reported older women as having benefited from home use of the Internet. 16 Information problems. Much information on the WWW is of dubious quality or relevance. 17,18 Finding relevant information is not always easy. In a review of information for cancer patients, 49 707 indexable Web pages were identified by search engines. Investigation of the 292 most relevant revealed 126 independent sites of which only 10 sites provided cancer information for patients. Two local community general health resources were excluded, leaving a total of eight oncology specific patient information sites; three were produced and maintained in Britain, and five in America. To carry out this process took many days (M. Gillis, personal communication). Cost. North American predictions about Internet use may not apply here unless the charging mechanism for local telephone calls is changed. There are increasing demands for the method of charging to be changed, 19,20 but in the meantime, the cost in the UK of having health information systems permanently connected to the Internet during daytime is considerable. Nevertheless, most would argue that home access is more convenient than access in health centres where patients may feel stressed and under pressure of time. 14,21 Access problems may be overcome by the development of easy-to-use Websites developed specifically for WebTV. For example, a Website on breast cancer awareness in English and Spanish was designed for WebTV and is being accessed by publicly available WebTVs throughout Alaska. 22 There are some Government sponsored initiatives to bring Internet access to communities in the UK. 23 E-mail on the Internet has allowed patients with similar conditions to exchange information and support each other. 24 In the USA it is increasingly being used by patients to contact their physicians 25 and although there are many concerns over the rise of ‘Webdoctors’ these too are likely to increase. 26 Personalized e-mail can also be used by health services for continuing health promotion and follow-up of susceptible groups, for example with reminders about ‘contracts’ for healthy eating or exercise (J. Marsden, personal communication). In the next decade, given the barriers to home use, public (non-health service) sites may perhaps offer already ‘filtered and validated’ information. If the interface is easy enough and the computer is readily available, most people will use it as was illustrated by a series of evaluation studies in Scotland and Spain between 1989 and 1998, 27–31 of a touchscreen health kiosk called Healthpoint (in Spanish Infosalud). The objectives of Healthpoint are to meet consumer demand for information, to raise awareness of health issues, to provide a basis for health promotion campaigns, and to measure interest in various topics by place and time. Evaluation methods included use of automatic monitoring by the computer, interviews with users and non-users, postal questionnaires, a random telephone survey and observational studies. Healthpoints were sited in supermarkets, shopping centres, pharmacies, health centres, hospitals, bars, sports centres, post offices, town hall, housing centres, libraries and staff canteens. Results showed that 17% of a town had used one of 10 Healthpoints sited there over a 5-month period. Although more people (20%) under the age of 50 had used it than those (13%) aged over 50, this ‘gradient’ of use was not steep. In studies in general practice, there were a good number of Healthpoint users aged over 70. Evidence from different sites suggested that Healthpoint received a more positive reaction from ‘less educated’ populations. Therefore, public access computers may be one way of creating more equal access to health information. There is little private space in health service settings for longer use (30–60 min) of a computer in a ‘booth’. 21,32 Public libraries are recognized sources of information in the community and have accommodated Healthpoints. Patients recruited to a randomized trial of computer-based treatment for anxiety in Glasgow, and referred to use one of 10 touch-screen computers sited in libraries seemed satisfied with these sites, 33 as did clinicians. In other centres, high street shops have been set up providing health information including a variety of computer-based information resources (e.g. Grampian Health Board high street health information shop, 181 Union Street, Aberdeen AB11 6BB, UK, Tel: 01244 589901). Nevertheless, older people and those unfamiliar with computers may be reluctant to use such a facility and experiments in which, for example, schoolchildren were recruited to act as ‘guides’ to the facility at weekends, would be worth trying. In health service sites, the practicalities of linking information given to patients to their own medical record are easier. In the longer term when space can be found, we can envisage the development and use of patient workstations. 34 Outpatient clinics might have a ‘bank’ of multimedia touch-screens at which patients complete a computer consultation in which they undertake a structured interview 35 and receive background information before seeing the clinician. The Government is aiming to get all schools linked and using the Internet. One of the features of access should be health information. Another way in which schools or the community can be ‘brought into the loop’ is to also become providers of health information and advice. 36,37 Computer-produced patient-held paper records have been used for many patients for some time. 38 Most GPs already have the ability to issue patients with a paper copy of their own medical record and we may expect to see this practice increase over the next decade. The benefits for patients include the opportunity to audit and correct the information, their increased involvement and satisfaction, and a means of conveying information to other health professionals who may need it. Information exchange could be achieved by comprehensive networking, but there are concerns about security and confidentiality. In the meantime, a paper record from general medical practice provides patient-controlled and useful information. 39 Much health information whether on the Internet 17,18 or in leaflets, is of dubious quality. 40,41 Various systems have been proposed 42–44 to evaluate health information. Efforts by the Cochrane Consumers and Communication Review Group ( rio@hna. ffh.vic.gov.au) and others, 45 aim to ensure that recognizable, good quality evidence-based information is available to consumers. ‘Kitemarking’ good-quality information has been suggested but, as in the sale of consumer products, this could be misused. The development of quality-controlled gateways on the Web and the new National Electronic Library for Health may offer solutions 46 (but see also the following section). In the next decade, we can expect to see more tailoring of information to patients’ needs. Patients need information to make decisions about health treatment, to adjust themselves psychologically, and to feel more in control of their situation. Computers can tailor information in various ways. 47 For example, the information offered can be ‘indexed’ using the patient’s own medical record. A study amongst 525 cancer patients suggests that such an approach is worthwhile. 48 Patients starting radiotherapy were randomized to one of three groups: personal, general or booklets. The personalized and general systems had the same content and style, both using a mix of hypertext and menus, like a simplified Web browser. The only difference was that the personal system started with the patient’s own medical summary. As patients looked at information in more depth, they moved ‘away’ from personal towards the general. More patients offered personalized information said they had learned something new, thought the information relevant, used the computer again, and showed their computer print-out to others. This study in cancer originally used Artificial Intelligence and Natural Language techniques to generate the personalized text and other studies have used similar approaches. 49,50 For example, an information system for patients with migraine 50 generated personalized hypertext pages of information based on an initial short computer interview. Other studies have used simpler methods, for instance a computer-system for patient education in schizophrenia included ‘cut and pasted’ information from the patient’s own medical record. 51 Whether such techniques become widespread in the next decade depends on progress in the development of the Electronic Patient Record. 52 Computers might also tailor the information by context. A public access Web browsing kiosk might adapt information from the Web, presenting for example in larger font, with no scroll bars for use with touch-screen, and only display information for that location. Such developments, however, will require more collaboration over Web-based health information than we see at present (see discussion). 53 Another way of filtering the information is to encourage patients to use gateways, and as GPs develop Websites the natural starting point and filter of valid sources may be through the GP’s Website. 54 Lastly, information will be presented according to the situation. Patients facing a health treatment decision will be offered information as a decision aid. A recent review of 17 decision aids found that they improved patients’ knowledge, reduced decisional conflicts and stimulated patients to be more active in decision-making without increasing their anxiety. 55 There is a danger that consumer health informatics will develop separately from clinical practice. Patients’ use of sources such as the Internet may lead them into conflict with, or mistrust of, their doctor and some clinicians may feel threatened by well-informed patients. On the other hand, clinicians could find that referral of patients to computer-based systems could help reduce their workload. At present, we do not know the best ways of integrating consumer health informatics into clinical practice, but the growing work on ‘Patient Partnership’ is addressing this. 56 More research is needed into the practical issues of healthcare organization and into clinical education. For example, a recent UK study 57 identified a list of 46 learning outcomes in eight areas to be addressed in planning curricula, and 30 action points for health service and educational managers. These should be incorporated into clinical curricula over the next decade. Learning materials might include, for example, the use of scenarios and simulated patients. Effective clinician–patient consultations require good communication skills, but clinicians also need to learn how best to help patients gather and assimilate information, and how they themselves can best learn from the patient and from this process. We can expect to see this field develop over the next decade. There will be many opportunities for patients and their families to take a more active role in seeking information and using various forms of information technology. Consumer health informatics could have far reaching effects on health and health services. However, as in all other aspects of health care delivery there are many issues to be investigated concerning equity, effectiveness and efficiency.