Patients’ uses of information as researchable domains of social practice

Abstract

In this article we argue that research into information for patients has to extend beyond an evaluation of particular information resources to studies of how those resources are engaged with, made sense of and used in practice. We draw on empirical data collected in the course of a study of a patient information resource designed for breast cancer patients in Liverpool and Newcastle in order to demonstrate the limitations of a restricted focus on information resources alone – namely, that it does not take into account the specific ways in which information is incorporated within what patients do as the grounds of ‘further inference and action’. Our interest is less in discussing the strengths and weaknesses of this particular resource than in explicating some neglected aspects of the commonplace ways in which patients ‘work’ with information. We conclude by sketching some broad features of those ‘reading’ and ‘linking’ practices, the study of which, we believe, would help us as researchers to explicate the ‘problem of information’ as it is actually encountered and resolved by patients in realworld settings for their own practical purposes. Taking our lead from ethnomethodological studies and related research in various fields, we argue patients’ uses of information are social practices that can and should be treated as researchable phenomena.

Introduction

As Bowker and Star have put it:

During this [last] century the information sciences … grappled with new ways of configuring, storing, and retrieving information. The rise of networked computing, and the extravagant advances in processing capacity have increased the pace and pressure of this struggle. We are clearly at a point today where we are witnessing the birth of an information technology as fundamentally new as was the printing press in its day ¹

The ‘technological revolution’ Bowker and Star talk of has had, is having and will have major implications for medicine and healthcare provision, and a range of authoritative commentators in these fields (including the US Government ² and the UK’s Royal Society ³ ) have sought to map the rapid proliferation of advanced information and communication technologies across all aspects of medical and healthcare work.^{4 –9} Although the scope, scale and implications of the ‘revolution’ are far from fully understood (or, indeed, universally welcomed or accepted), policymakers have been quick to seize on its potential, particularly in promoting patient involvement in healthcare by using the provision of information to support the ‘logic of choice’^{10 –15}. Patients’ associations have reacted with similar enthusiasm. According to advocacy groups, the growing openness and availability of medical information could lead to a new, more equitable relationship between the different parties with a stake in the health of patients, amounting to no less than a ‘reformation in healthcare’, a socio-cultural revolution following in the wake of a socio-technical one. ¹⁶

Unsurprisingly, given this, there has been a rapid growth of studies of the web-based information resources that patients now routinely seek out and make use of. ¹⁷ However, while advanced information and communication technologies are increasingly prominent features of the patient information landscape, it would be a mistake to think that they have displaced more ‘primitive’ technologies. Patients operate in environments which are information and, perhaps more importantly, inferentially rich, ¹⁸ and, in the course of the different paths they follow through healthcare systems, they continue to rely upon communicative interaction with others (including face-to-face and telephone and indirectly through over-hearing and noticing), written information in the form of files, letters, appointment cards, leaflets, posters and signs, as well as newspapers, magazines, journals, reports and policy documents, and older technologies such as calendars, diaries and pillboxes to keep track of, for example, where they are meant to be when or what medication they should be taking at which times.

The diversity of the information resources available to patients is reflected in the literature on this subject. In a field which often comes across as a ‘confusion of tongues’, ¹⁹ we find loose conglomerations of frequently one-off studies based in different research traditions, conducted in different registers and proceeding in different directions. Different research teams have investigated different kinds of information given to patients with different conditions receiving different types of treatment grounded in different areas and levels of medical practice at different points in the treatment process, communicated using different media and received at different sites from different providers in different national healthcare contexts. They have employed different theories, different methods and have arrived at different (and often competing) conclusions as a result.^{20 –24} The diversity of work in this area testifies to the position information for patients has come to occupy at the shifting intersections and points of contact between governments, large-scale health bureaucracies, healthcare professionals, health informatics and a diverse range of patient publics and ‘counter-publics’. Patient information is not the sole province of any single constituency but of many at once, and its changing forms are part-and-parcel of wider, ongoing reconfigurations in their relationships. It would be a mistake to under-estimate just how hotly those reconfigurations have been, and are being, contested, with doctors in particular quick to defend against perceived encroachments on their professional autonomy via information, communication and other ‘accountability’ technologies by politicians, health service managers and patient bodies (for a particularly trenchant restatement of the ‘doctor knows best’ view of patient information see Sokol ²⁵ ).

Our focus in this article is not, however, ‘patient information’ in the broad. Nor do we have any wish to take sides. Instead, we will discuss, from an ethnomethodologically-oriented perspective, an aspect of patient information that is often taken-for-granted within these debates, namely the practical character of patients’ interactions with information resources. As a starting point, we want to distinguish information for patients (i.e. available to and used by patients) from information on or about patients, and we do so for two main reasons. Firstly, there have been a number of what we regard as exemplary studies of information on or about patients, information typically generated in the course of different kinds of medical and healthcare work.^{4 –9, 20} Research of this kind is grounded in work in science and technology and workplace studies, with particular emphasis on collaboration and coordination, human–computer interaction, computer-supported cooperative working, information systems design and ubiquitous computing in a wide variety of settings, including telecommunication, informatics and healthcare.^{26 –34} While this body of research includes studies of ordinary users of information technology,^{20 –32} how patients, specifically, access and manage information beyond the immediacies of particular kinds of medical encounters has not been the subject of sustained examination in the same way. Secondly, and as a consequence, we believe extending the practice-oriented approach taken in studies of this kind to the ways in which patients access, manage and use information more broadly can bring much needed conceptual and methodological clarity to debates in this area. This is not to say there have not been many sophisticated studies in this area: there have.^{5, 10, 11, 35 –38} We do, however, think work in this area has a tendency to introduce too sharp a contrast between patient and professional informational ‘work’, treating them as separate domains to be studied in different ways. There is a danger in this that categorical distinctions are simply assumed from the outset and so read into practices rather than being shown to be relevant within them. Extending the alternate, ethnomethodologically-informed approach that has informed studies of (a) technologies-in-use and (b) professional-patient interaction to (c) patients’ information work (an area much studied but not by ethnomethodologists) is particularly helpful here because it provides a set of resources that enable us to navigate such dangers safely.

We do, however, recognise that technologies such as patient records and, more recently, electronic patient records (EPRs), complicate matters in that they are accessed by professionals and patients alike.^39,40 Moreover, a great deal of the information available to and used by patients has been put together by or derives from the work of professionals of many kinds working within healthcare systems, much of it in the course of their interaction with patients.^{4, 8, 9, 20} While that is undeniably the case, what we are interested in is providing an account of some features of the patient information ‘landscape’ from the patient’s point-of-view. For that reason, in what follows we will present and discuss examples which deal with the simpler case of an information resource designed specifically for use by patients.

The position we set out is grounded in two separate evaluative studies: the first in 2006 (involving 28 participants) and the second in 2008 (involving 22 participants). These studies examined how patients with breast cancer in Liverpool and Newcastle, respectively, engaged with and made use of a multi-platform information resource that had been designed to support them at different stages in their treatment, from diagnosis through to remission, based on the organisation of health and social care services in those particular locales. Although these studies began with patients’ uses of a particular resource, the aim was to examine, as far as possible, their engagement with that system in terms of their methods for making use of and orienting to information resources more broadly. Given our interest in the latter, we will not say very much about the strengths or weaknesses of the resource itself here. Rather, we have chosen to highlight commonplace activities that proved relevant to the use of that system: navigating forwards and backwards from a website menu and keeping a personal diary. This is entirely in keeping with investigations of mundane ‘informational’ work within ethnomethodology, a tradition which encompasses studies of such ordinary activities (and their associated competencies and fields of social production) as following a series of signs through an unfamiliar building, ⁴¹ referring to instructions in the process of assembling a chair, ⁴² puzzling out the meaning of a passage of prose,^{43, 44} finding a book on a library shelf ⁴⁵ or connecting to a wireless internet server for the first time. ⁴⁶ While focussed on seemingly trivial matters, analyses of such activities raise important considerations about ‘seen-but-unnoticed’ features of information work. ⁴⁷

Based on our examples, we will argue that the investigation of patient information cannot be unproblematically limited to the study of the technical properties (typically content and organisation) of particular information resources. Proceeding as if it can neglects the fact that information acquires its relevance for patients in the course of their wider activities. Our goal in what follows is to explain why we think neglecting patients’ practices represents a problem. We also set out what we think can be done about it: namely, a shift away from a view of patient information as a disparate collection of stand-alone technologies to a view of patient information that investigates patients’ interactions with those technologies as forms of social practice in their own right. In the conclusion, we suggest some areas where this reorientation is likely to produce particularly interesting results.

Assessing patient information resources

As a search of any database of scholarly articles demonstrates, the place where the vast majority of investigations of patient information resources begin and end is with the resource itself, particularly content, the information made available through the resource as well as how that information is organised for access. In examining a given resource, the broad question is, does it do what it is designed to do? Does it provide access to ‘item[s] of information in a form that helps each user … to find and interpret [them]’ ⁴⁸ as and when they are needed? Researchers therefore ask: Is it comprehensive? Is it up-to-date? Is it clear? Is it readable? Is it well structured? Does it mislead? Is it free from errors, bugs or glitches? Does it communicate what it is meant to? Research on these questions tends to be prescriptive, with an emphasis on the creation of standardised evaluative metrics in order to determine whether particular technologies meet the mark. ⁴⁹

This approach to the investigation of patient information resources is very often underpinned by a philosophy that sees the provision of information as an extension of healthcare. ⁵⁰ Researchers proceed on the basis that information ‘needs’ can be objectively defined and measured, problems identified and appropriate interventions prescribed, as with medical conditions (typically by those with the requisite expertise: medical professionals). For proponents, it becomes important to determine the health-related properties of information—the properties that enable information to discretely ‘act on’ patients—through techniques like randomised controlled trials, ²⁴ and subject that information to strict quality controls through kite-marking or other accreditation schemes. ⁴⁹ If information is to be a ‘valuable … part of a multifaceted approach to managing specific conditions’, ⁵² it must be subject to the same set of regulations that govern the prescription of drugs and other forms of medical intervention—something we see in the case of ‘information prescriptions’.^{52, 53}

It is possible to critique this view of patient information on a number of different levels. We might, for example, question the sense in which medical or healthcare professionals, or information scientists, are best placed to determine what information patients should or should not access, how they should or should not interpret it and what they should or should not do with it once they have it. Alternatively, we might question what is a highly decontextualised and asymmetrical approach: if it makes little sense to treat the kinds of information medical professionals work with in this way, it is unclear why it makes more sense when it comes to the kinds of information patients are said to ‘need’. ⁵⁴ Finally, we might question the reliance on functional explanations (as in e.g., Clerehan and Buchbinder ⁵⁵ ): stating that a resource is readable because it can be read adds as little to our understanding as stating that opium induces sleep because it has sleep-inducing properties. ⁵⁶

However, to dismiss work of this kind out of hand would be to ignore the point that questions of content and organisation are far from trivial and ought not to be overlooked. A badly designed resource that is difficult to access and read, and is full of errors, is unlikely to be of much use to anyone (if not potentially dangerous in a healthcare context^{6, 7, 35}). However, equally clearly, questions of content cannot be exhaustive and for a range of different reasons. What we want to stress here is that the ‘same’ resources can be taken up in quite different ways depending on the situation.

We would like, first, to invite readers to imagine a medical textbook. Depending on the circumstances, that textbook may be used: (a) as a teaching aid; (b) as a learning aid; (c) as a diagnostic tool; (d) as a description of medical knowledge and practice; (e) as a standard against which to assess medical knowledge and practice; (f) to resolve differences of opinion; (g) to look up definitions, names, references or dates; (h) to highlight gaps in an evidence-base; (i) as a status claim (when displayed on a desk); (j) as a way of raising problems (when produced by a patient in the course of consultation); or, among many other possibilities, (k) as a paperweight, should it find no other practical uses. If we were to focus on the content and organisation of the textbook alone, we could not account for the different ways in which it might potentially be used and so the courses of action and interaction it might come to be embedded within. Those forms of use and courses of action are not unrelated to content but nor are they determined by it.

The difficulties associated with the prioritisation of content over use become more acute when we turn to examples like the following, taken from our own work in this area.

Example one:

N:

Was it [the PDA] easy to use then?

S:

I think in general … because I was briefed how to use it [the PDA] beforehand so … it was double-clicking on each title and it goes on … and there’s a back button as well so … going through the previous buttons and going back to the original page

N:

How useful was it?

S:

I would use something like that … but I think just one thing might be practical to make it a bit more user friendly. So whatever buttons you need, have them around like, you know, the screen, forward-back navigation button, you know? Like on the internet you generally have forward-backwards, yeah? Something like that … I’m an IT user but you might not find a lot of people who are so I think it might just help.

Example two:

M:

Did you have any problems?

P:

No, no.

M:

You’ve spoken about the kinds of routes that you take when you’re looking for information. Do you think somebody who was not as IT literate, would have trouble?

P:

I don’t know. All I did was use the back button … at the top, ha, ha. I mean, I’m not that good on websites. All I did was that, you know, if I’d gone into it, into stage four, and found out what I was looking for I’d just press the back button till I got onto the main menu that I wanted and … go on to the next one. So I mean … I’m not very sort of … I’m not very expert at getting round … websites.

Example three:

A:

How often do you use a computer?

R:

Never, ha. No, never. Well I watch the children using it and, you know, that’s my [limit] … They do all sorts on it and I feel that I … I want to keep pace

A:

Did you find the information you wanted?

R:

Yeah … I did because I was able to, to you know

A:

Ok. Did it involve . . .

R:

A lot of effort, well no it didn’t really

A:

Did you have any problems with it?

R:

No, I didn’t, no … I was able to get the … you know, the ehm … what is it? … the typeface … and the ehm … like the mammograms and then the different treatments, you know. So I was able to go back into that and it explained what they were. Even the eh needles, using the needles … which I’ve had done. So …

A:

So I see, you knew what was going on

R:

So I knew what to expect in that sense, you know.

Example one is interesting because S, in explaining what he thought about accessing the resource through a personal digital assistant (PDA), justifies his comments with reference to his status as an ‘IT user’. His reference to navigation using forward and back buttons is tied into a claim about expertise as someone who is familiar with information technology (IT). S proposes that because he already knew how to use related systems, he could navigate quite easily, whereas those who lacked such familiarity might have difficulties. In example two, P also talks about navigation, specifically using the back button to navigate to the main menu from a particular point (‘stage four’). However, despite demonstrating the same set of competencies employed by S, in this case she explicitly states that she’s ‘not very expert’. In other words, P’s response is based partially on the idea that there must be much more to ‘getting around … websites’ than simply using the backwards and forwards buttons to navigate toward and away from the main menu, something which S with his familiarity knows is not the case. Finally, we have the case of R, a retiree who actually laughs a little at the suggestion she might be a regular computer user. Here, again, navigation presents few difficulties, not because R claims any experience with IT at all, but because she was able to use her knowledge of the treatment she had already undergone as a way of orienting within the system. In each case, the users accessed different information for different reasons. However, in each case, they improvised, adapting navigational strategies familiar to them from other domains to locate information within the framework of an unfamiliar database. These accounts are thus as much about what these individuals were able to bring to bear on the resource to exploit an affordance ‘within’ it as they are about features the resource might be said to independently possess.

The outcome is the same in all three cases—they found their way to what they were interested in—but how they got there is not. The three interviewees were not engaged in the same kinds of activities. ⁴² The difference is particularly striking when we compare the first two and the third examples: R has an entirely different orientation to the technology and draws upon a completely different set of resources in order to make it work for her. It is the ‘human texture’ ³¹ of these interactions with information technology that drop from view when we concentrate on the ‘functional integrity’ of the resource alone. The picture produced when this happens inevitably leaves much of interest unexamined. As even these short examples show, when we start to ask how individuals go about such apparently simple tasks as navigating forwards and backwards from a menu, we are quickly led to a much wider set of considerations. Nor does this involve an analytical leap on our part: S, P and R themselves guide us into more complex terrain, instructing us as to the relevant background against which their practical engagement with this resource ought to be properly seen and understood (for a particularly robust statement of this position see Crabtree et al. ⁴⁵ ).

If we concentrate narrowly on the technical question of how a given resource extends our access to information, we also fail to capture the moral significance of our interactions with technology. ³¹ In our study, the designers built an electronic treatment and mood diary into the information resource. This facility was designed to help patients keep track of the treatment process and its effects on them. One patient, D, stated that she had not and would not use this particular facility (although she would continue to use others). This was not because she did not know how to use it, or because she thought it was useless. Almost the reverse was true in fact: she avoided it because she thought it would do its job too well. It was because D wanted to ensure that no documentary record of this particular part of her life was left behind and so she shunned diaries of any kind.

Example four:

M:

I know there are so many appointments and places to go to, do you keep quite a detailed diary for yourself?

D:

No, no. No. I ehm, I really … I really don’t want that … I really do not want it because it’s something I will never be able to go back to. I really do not want to have to read about it. I mean I’d say the majority of things are up here anyway {points to her head}. How I felt … at individual times … it’s all locked away there.

M:

Yeah.

D:

When I was diagnosed, a girl from work, she sent me the most gorgeous bunch of flowers. Now, I love flowers. But … I can still remember the smell from them, and I’m not too happy about flowers any more. I was glad to see them go in the bin. And even the get well cards … they’ve all gone as well. So I don’t know if I’m being a bit, you know, ha, ha, but it was like me trying to clear this all away. And I wished there was a door I could go through, close it, and it’d all be behind me, it’d all be gone.

What such examples show is that the status of something as ‘information’ cannot be separated from particular courses of practical action and reasoning. ⁴⁷ Information is ‘tied to the things that people do; to the worlds to which they belong’ ¹ and acquires its relevance in terms of how it is ‘exploited in activities by which … tasks [are] accomplished’. ⁵⁷ Almost anything, in this view, may count as an ‘information resource’. The question of interest is what patients themselves treat (or indeed avoid ¹⁰ ) as sources of ‘further inference and action’. ⁵⁸ Answering such questions is a descriptive not evaluative exercise: these are empirical questions the dimensions of which cannot be stipulated in advance using standardised metrics.

Reading and linking practices

The mundane examples we have discussed help to remind us what, as ordinary members of society, we already know: that the problem of information is as much a social, cultural and indeed moral issue as it is a technical one. By concentrating on technical matters first and patients’ interactions with and orientations to these technologies second, we are thus in danger of oversimplifying this area of human affairs.

But how can we evaluate information resources if we adopt the second approach? As we have already conceded, such studies are descriptive not evaluative. Moreover, given their particularistic focus, they seem to provide no easy route to generalisable conclusions or straightforward implications for design. ⁵⁹ We do not think, however, this represents a weakness. As research in the areas discussed earlier has demonstrated, rather than generating detached schematics, researchers should be examining the ways in which information resources are made to fit (or come to be seen not to fit) within the lives of those who use them. In this sense, rather than deriving a set of criteria for assessing information resources, we should look to see what methods patients themselves employ to differentiate the good from the bad, the useful from the useless, the situationally relevant from the situationally irrelevant. We believe that refocussing on the work that patients do on, and with, information will help to guard against any tendency to oversimplification.

There is already a body of research examining questions of this kind in the area of patient information, including work in the fields of library and information science, information-seeking in context, everyday life-seeking, health education/communication and health literacies.^36–38 There have been, for example, studies of how patients look for different kinds of answers based on the sources of information available to them at different stages of the treatment process.^{22, 60} However, we believe the scope of such investigations should be extended much further. At the same time, more has to be done to avoid the reification of distinctions between the informational work of professionals and patients by treating those distinctions as embedded in and elaborated through practice rather than as belonging to ontologically separable ‘worlds’ (for critiques of this reifying tendency see Davenport ³⁸ and Crabtree et al. ⁴⁵ ). Based on insights into vernacular sense-making practices within philosophy, phenomenology, ethnomethodology, conversation analysis and actor–network theory, and building on the examples above, we would point initially to two kinds of practices which have, as yet, received scant attention within the literature: what we are terming, hopefully uncontroversially, reading and linking practices. Given the commonplace character of the kinds of activities these practices are bound up with, we believe the points we raise have a much wider applicability.

Reading practices

Ethnomethodologists (along with others like Derrida ⁶¹ ), have long been interested in explicating the plurality of ordinary reading practices and the vernacular competencies exhibited in and through them.^{43, 44, 62 –65} In terms of patient readers, one particularly interesting aspect of these practices is their motivated character: patients’ readings are often ‘directed readings’. That is, a given text will be read, not cover to cover and for its own sake, but just for the (often as yet unidentified) things that may turn out (upon reading) to be relevant to the person reading it.

Patient readers employ a variety of techniques that enable them to work through different ‘orders of organisation’ within the texts they read in directed ways.^63–65 These include, for example, methods of discriminating between different aspects of their topical organisation, where this might involve following transitions from one topic to the next or reading more deeply into particular topics within the resource as a whole. Alternatively, they may draw on extra-textual resources like their knowledge of similar resources or of the processes being discussed (as in examples one to three) in order to make the resource readable. Among other things, such practices enable patients to focus on specifics, make out the structure of the resource as a whole, trace (and make) progressions from simple to more complex information within it, and so on, in order to work out ‘what is connected in the display’. ⁶² The techniques employed include those normally grouped within studies of information practices under the headings of searching and scanning but also under such things as skimming, skipping, bypassing, combing, plumbing, surveying, mining and even ‘foraging’ for relevant materials. ³⁶ It is important to bear in mind, however, following Garfinkel, that readings are accomplishments that can be studied in their details and are not blind or automatic. Readings proceed on the basis of a provisional, wait-and-see proviso (‘I’ll know it when I see it’) and thus have a developing (prospective-retrospective) sense. ⁴⁷ Readings are also oriented to ‘seriality’ in that they look forwards or back from particular points within a text (or, indeed, stages of treatment as in example three), from ‘firsts’ to discernible ‘nexts’ and, from there, ‘lasts’. ⁶⁶ It is by orienting to such orders of organisation that patients are able to determine whether what they are reading has a local or more global relevance, whether something holds for specific times and places or has a more general significance. Understanding how patients read for relevant information and not just what they are reading, is, we suggest, one area where a great deal remains to be learned.

Linking practices

As our fourth example shows in miniature, information resources are rarely treated as self-contained but as pointing beyond themselves to a universe of people and objects, relationships, sites and settings, bodies of knowledge and, indeed, other texts and resources. In engaging with particular resources, then, patients are engaged in making outward connections, relating what they glean from one resource to the many others it can (upon reading) be seen to stand in relation to. There are important differences between computer-based information, visually-based information, paper-based information and face-to-face, interaction-based information, but the patients in our research rarely treated these categories as entirely discrete or distinct. Instead, they routinely transformed information from one format to another. Discussions with doctors and nurses were written down or recorded. Online materials were printed off and used in consultations. Hospital signs, appointment cards, instructions before surgery, etc., were turned into questions—particularly where there was an element of doubt. Answers received were fed back into searches for information, in libraries and online, which led to further appointments, meetings and discussions which, in turn, were coordinated using paper calendars and reminder notes stuck to refrigerators. Information that was not easily transformed, and so brought to bear in the widest set of situations, was regarded as effectively useless. Information gleaned in the course of consultations, for example, is not easily remembered so, if it cannot be recorded in another format, its relevance is often ‘lost’ to the patient. Similarly, if information acquired outside the medical setting could not be translated into questions for healthcare personnel, it was also seen as having limited use. Finally, this did not just apply to individual interactions with information resources. Those in our study explained that they would field questions from anxious relatives, friends and colleagues by directing them to particular information resources, thereby delegating, ⁶⁷ as it were, the reassurance work to the technology rather than having to undertake it themselves. Access to resources was thus threaded into a whole series of relationships and interactions. What we require is a much greater understanding of the range of techniques that enable patients to do this inter-textual linking (and out-sourcing) work and so exploit information resources in a range of ways by adapting them to their own practical purposes.^{26, 67–69} At the moment, we have little understanding of the ways in which these ‘occasioned’, purpose-built ‘informational assemblages’ are put together and put to use by patients. We think a study of these kinds of practices, along with a study of the multiplicity of reading practices interwoven with them, would, again, deliver fresh insights into patient information as a real-world concern.

Conclusion

As Marshall Sahlins has noted in relation to a very different form of hunting, our investigations of ‘hunting’ for information too often involve the ‘exaggeration of the importance of tool over skill’. ⁷⁰ Dazzled by the rapid rate of technological innovation and diffusion, we are in danger of becoming fixated on information technologies at the expense of the skill of the users who find a use and place for it within their lives. We would argue that the functional integrity of a given resource can only really be gauged after we have a sense of how it is or might be used and to what practical ends, i.e. we need to ask ‘Is it good for that?’ rather than just ‘Is it good?’.

We do not believe that information resources are the answer to all of healthcare’s problems. Nor, by the same token, do we believe they are necessarily either pernicious or useless. Information resources are sources of inference and action, and formal resources will be more or less useful depending on how they are designed, the content they organise and make available, and what people find they afford them. Our argument is that it is time to move beyond studies that focus solely on the what (i.e. categories of information resource/content delivered) to an investigation of the how (i.e. information using activities). Despite the surge of interest at the academic, practitioner and patient levels, we still understand relatively little about how patients read and link different types of information together for practical purposes in everyday situations. Talk of ‘what patients do/want/need to know’ is still largely based on disengaged accounts of action, not systematic empirical investigation. This can change and it is an area where ethnomethodological studies can make a direct contribution to understandings of this complex field by treating patients’ uses of information as a researchable domain of social practices. In this, we feel investigations of the kinds of commonplace engagements with information resources we have begun to broach in this article represent a promising place to begin.