Introduction

There are three types of people who work/exist in healthcare. Medical experts and medical support staff (doctors, nurses, ambulance staff, phlebotomists etc.) are the first type; the people who support the first type (administrators, cleaners, porters, computer experts etc.) are the second type; and patients are the third type. When talking about information systems (ISs) in an organization, a user is normally someone who uses the system on behalf of the organization, a provider, which in healthcare is one of the first two types. In this paper we are going to look upon the user as the patient, the inevitable user of healthcare as a system, and therefore on the receiving end of any IS that the first two types use. The providers tend to determine the ISs used in healthcare because they have the power to do so, controlling budgets and generally being more permanent fixtures in the system.

The purpose of this paper is to suggest, through a user's perspective, some of the important concerns that need to be taken into account in order to successfully use ISs in healthcare. To avoid confusion over the use of the word user, we shall distinguish patients from the normal use of the word user by calling the former ‘patient-users’. This also aptly describes one of the major requirements of any patient-user, namely, patience. This is not the first paper to highlight the need for patients’ concerns to be addressed in developing IS (see the recent paper by Payton et al, 2011, which gives an excellent overview in an interdisciplinary context, and is discussed further below), but it is possibly unique in suggesting what patients’ concerns might be, or at least that the set of concerns put forward are innovatory.

The section ‘The demand for healthcare’ of the paper shows that the wealthier the nation, the more likely that expenditure on healthcare will consume a higher proportion of a nation's wealth. In the section ‘ISs in healthcare’, we set out to show that, while it is generally believed that ISs have a major role to play in healthcare, the achievements they have brought about so far are largely not fit for purpose. The section ‘Why is IS in healthcare difficult: the literature’ draws some tentative conclusions from the literature about the reasons for this. The section ‘A patient-user perspective’ introduces a patient-user perspective of the concerns that must be taken into account for the success of healthcare ISs.

The section ‘Ethnographic studies’ discusses seven stories, or ethnographical studies that the first author (who has Parkinson's disease – PD) presents in support of these concerns. This makes this paper very distinctive in that, in healthcare, the IS professionals might conduct ethnographical studies of patients or patient groups, but there are few where the patient studies the medical providers. In part, this is because the Parkinsonist (sufferer) wanted to write an autobiographical account, which required that some thought be given to what material was to be obtained. The second reason is to do with discussions between patients and the medical providers. As the first author expresses in his views on such discussions ‘It is essential for the patient to think what (s)/he wants to get from a meeting with the medical providers, what they want the providers to know and any queries they may have. I wrote agendas for my meetings, and took copies of other medical reports with me in case the person I was seeing did not have this information in the file. If I was seeing someone other than my PD consultant, I always took a latest copy of my medication as laid out by my consultant. This enabled me to write my autobiographical account (Paul, 2009) relatively easily if not quickly’. In this process, there was no intention to declare a definitive or prescriptive list of reasons for the failure of ISs or lack of user satisfaction with the IS; rather, it was to show that a patient-user perspective can highlight some of the concerns worthy of debate about what the IS should address – in other words, its fitness for purpose.

The last section, ‘Conclusions and a way forward’, proposes a new way of thinking about healthcare. This way of thinking is not intended to replace the concept of healthcare, but it is used as a stalking horse to encourage these considerations to be a part of the debate on healthcare.

As the issues raised in this paper are important as well as interesting, and deserve wide consideration, the authors have attempted a concise clear narrative style. This paper's arguments are not meant to be definitive or prescriptive, merely indicative, and hence we believe that the need for lots of references in support of such a point of view is less relevant here than that the point of view can be successfully argued.

The demand for healthcare

Spending on healthcare is rising faster than increases in wealth as measured by gross domestic product (GDP). Take the United States for example. The following data come from the web pages ‘healthcare spending in the U.S.A’, which can be accessed on the website ‘U.S. Government Spending History from 1900’ (Chantrill, 2012). At the beginning of the 20th century, government spent little on healthcare. Government healthcare did not exceed 1 percent of GDP until the 1960s. In 1965, Congress passed the Great Society legislation at the behest of President Johnson, featuring Medicare, a health subsidy programme for older Americans, and Medicaid, a healthcare provision for the poor. Government health spending breached 2 percent of GDP in 1970, and kept increasing so that by 2009 it was 7 percent of GDP.

Data from the World Health Organization (WHO) show that the trend in spending is for a higher proportion of increasing wealth to be spent on healthcare (WHO, 2012). Table 1 shows percentages of GDP of selected countries that have been committed to spending on healthcare based on data from the WHO's statistics. The expectation is that this trend in spending disproportionately larger proportions of increasing wealth on health will continue. Table 2 gives past and predicted changes in just the expenditure on public health as a percentage of GDP. There is no reason to believe that private expenditure will not either.

Table 1 Total expenditure on health as a percentage of GDP (2000–2005)
Full size table
Table 2 Changes in public expenditure on health as a percentage of GDP
Full size table

The most likely explanation for these data is that it shows that with an increase in the wealth of individuals there is a tendency to spend a disproportionately larger slice of this wealth on health. In other words, individuals one way or another will devote a larger slice of their increasing income to health – presumably in the belief that they will stay healthy. In extreme cases, the objective of the individual will be to delay death. Alternatively, the data could be showing that, as countries get wealthier, the cost of healthcare rises faster. For example, as a country gets wealthier, its medics get disproportionately larger incomes in compensation for doing a job that is harder and more unpleasant than other well-paid jobs. However, whatever the interpretation, it is clear that the demand for healthcare at present cannot be met because there are insufficient resources available to do so, hence the increasing proportion of wealth devoted to health. There may come an age when healthcare can keep someone alive almost indefinitely. However, it is not here yet in 2012, and it seems to be an immeasurably long way away.

ISs in healthcare

Part of the increases in public health expenditure has been directed for the use of supportive technologies, aiming to enhance healthcare provision. One of these is the use of ISs and its underlying technologies in the context of healthcare. The terminology used in the literature is not consistent. This includes, but is not limited to, health information technologies (Furukawa et al, 2006), healthcare ISs (Teixeira et al, 2010; Johnson, 2011), the use of IT in healthcare (Haux et al, 2002), IT applications in healthcare (Chiasson & Davidson, 2004) and so on. In this paper, we use the term healthcare ISs to combine all of these terms, or others that address the use of Information and Communication Technologies (ICT) to support healthcare applications.

The use of healthcare ISs has evolved with the rapid development of ISs and advances in healthcare technology in parallel with current concerns arising over patients’ safety and of course how to cure patients efficiently. Although there are multiple potential benefits of these systems, there are already several high-profile failures (Jha et al, 2009; Teixeira et al, 2010; Johnson, 2011; Pai & Huang, 2011).

Impact of healthcare ISs

It has been postulated that healthcare ISs can offer economic benefits through efficiency savings via appropriate data management to identify potential bottlenecks in the provision and administration of care, which can become more predictable (Johnson, 2011). Perera et al (2011) add that health information sharing between patients and their healthcare providers may improve diagnosis, increase patient education and promote self-care.

Lawler et al (2011) discuss the results of a systematic review of 257 studies related to the impact of health ISs. This shows that 25 percent of studies were associated with four large exemplary academic medical centres, which have mature, custom-developed health ISs. Reliable and general stable evidence demonstrating the positive effects of health ISs on safety and quality remains inconclusive (Berger & Kichak, 2004; Burstin, 2008; Eslami et al, 2008). Studies measuring the positive effects of health ISs are matched by a growing body of research revealing the unintended consequences and potentially error-producing effects of the implementation of health ISs (fitness for purpose).

Electronic patient records (EPRs)

The literature almost unanimously advocates that electronic health records or EPRs have the potential to improve patient care (Teixeira et al, 2010).

The use of EPRs as the primary source of patient medical information is still relatively rare but is increasing from a slow start in Canadian, British and American primary care. In addition to improved legibility, the organized note structure of many EPRs should support high-quality patient summaries desirable for shared clinical care. Similarly, the detailed healthcare information in EPRs should make them important sources of information for clinical studies, research and policy. Therefore, the potential benefits of EPR systems are claimed to be organizational issues, such as improving the exchange of information between healthcare facilities and also the support of standardized procedures that can help to increase consistency between different service providers. Further, EPR should be able to ensure minimum standards across the trajectory of care when patients move between different specializations. Not only have these benefits not been delivered yet, but serious issues have arisen as to how and whether these benefits will be possible.

There are many other problems that are associated with implementing EPR systems. Information has to be retrieved from paper-based records that often lack the aspirations of electronic systems to have everything to do with the patient's health. Individuals who have only used electronic ISs face particular problems in learning how to apply paper-based fallbacks (Johnson, 2011). Aanestad & Jensen (2011) reported that achieving interoperable central EPR systems have proved difficult to realize and challenging on a nation-wide scale (Middleton et al, 2005), both in the United States (Kohn et al, 2000; Ash & Bates, 2005; Berner et al, 2005) and in Europe (Currie & Guah, 2007; Greenhalgh et al, 2008, 2010; Jones, 2004). Although the US Reinvestment Act of 2009 has made the promotion of a national, interoperable health IS a priority, U.S. healthcare providers have been slow to adopt them. Other issues related to EPR systems include privacy concerns where some studies showed mixed views on whether the privacy of electronic medical records could be maintained (Perera et al, 2011). Further, Lawler et al (2011) add that, whether by lack of system interoperability, poorly designed socio-technical systems, scepticism regarding patient confidentiality or other factors, a 2006 survey of 4561 U.S. hospitals showed a wide variation in the adoption of IT. For example, 25 percent had not implemented any health IT system, and of those with IT a majority had implemented two or fewer applications (Furukawa et al, 2006).

More recently, Jha et al (2009) reported that the U.S. healthcare system faces challenges on multiple fronts, including rising costs and inconsistent quality, although there has been considerable political support for EPR systems (Hoffman & Podgurski, 2008). For instance, healthcare informatics was a key element in the campaigns of both Senators McCain and Obama in the 2008 U.S. Presidential election. These commitments have been carried into office with the passage of the American Recovery and Reinvestment Act of 2009, otherwise known as the ‘Stimulus Bill’. This provides for an investment of around US$148 billion in healthcare of which $19 billion is specifically earmarked for healthcare information technology. Part of this investment is intended to support the development of a U.S. national system for the electronic storage and exchange of patient records (Johnson, 2011). However, as Jha et al (2009) discuss, despite a consensus that the use of health information technology should lead to more efficient, safer and higher quality care, there are no reliable estimates of the occurrence of adoption of electronic health records in U.S. hospitals. In a review of high-quality surveys, Jha et al (2006) indicate that, on the basis of responses from 63.1 percent of hospitals surveyed, only 1.5 percent of U.S. hospitals have a comprehensive EPR system (i.e., present in all clinical units) and an additional 7.6 percent have a basic system (i.e., present in at least one clinical unit).

‘First, do less harm’ and ‘process, people, patients and interdisciplinary considerations’

This sub-section refers to two recent publications that offer explanations for the state of ISs in healthcare, as well as discussing that rarely mentioned part of the health system, namely, patients.

The first part of the title of this sub-section is the beginning of the title of a book called ‘First, Do Less Harm: Confronting the Issues of Patient Safety’ by Gordon & Koppel (2012). Chapter 4 by Koppel et al (2012) is called ‘Health Care Information Technology to the Rescue’ and is a very readable presentation of lots of good ideas and the things that have gone wrong with them (Decision Support Systems, Expert Systems, EPR systems etc.). The authors take the reader on a journey of revelatory discovery as system after system is found wanting in a variety of obvious ways – obvious when you know. The chapter ends with a section on ‘What to Do? Specifics’, all of which then are worthy actions to take, but, surprisingly, with no mention of the patient-user!

The second part of the title of this sub-section is part of the paper by Payton et al (2011) addressing Health care IT. The paper is an introduction to a Special Issue on Health care IT, but it is much more than an editorial, being rigorously researched and informed in its own right. The paper covers evidence-based practice, the places and approaches to incorporating the patient view in the healthcare system, web-based patient groups and patient-centred care.

This paper could spend more time extolling and explaining the virtues of the two publications above. However, rather than rehash the authors’ good work in second-class prose, we encourage the reader to read them him/herself, so confident are we of their value in the midst of so much dross.

Why is IS in healthcare difficult: the literature

In this section, the literature is first examined before a patient-user view of the reasons for the difficulty is offered in the section ‘A patient-user perspective’.

Healthcare is a unique and complex domain and healthcare ISs have human safety implications and profound effects on individual patient care (Kohn et al, 2000). So, why do these projects fail? One of the issues discussed by Abraham & Junglas (2011) is that the healthcare industry has recognized the importance of patient centricity, a concept that puts at centre stage the patient and the associated procedural workflow. This idea theoretically moves away from the concept of a fragmented, physician-centric care delivery organization (Porter & Teisberg, 2007). However, implementing IS based on this patient-centric concept and managing the transformation is a formidable challenge for organizations (Day & Norris, 2007; Porter & Teisberg, 2007; Harrison & Kimani, 2009; Carr et al, 2009; Szydlowski & Smith, 2009). In fact, IS implementations are a perturbation in any organization, whether it is a change in processes or in organizational communication and learning (Davenport, 1998). In healthcare, however, there are higher stakes for failure than in traditional businesses, as the slightest disruption caused by the IS can have detrimental consequences (Christensen et al, 2009).

Chiasson & Davidson (2004) add that the effective use and beneficial outcomes of IT applications in healthcare are not guaranteed, however. Healthcare applications are technically complex, and the software and hardware markets are considered to be less mature than the IT markets for other industries and for medical technologies. Leaders in the healthcare field recognize that, in addition to technological challenges, high levels of resource commitment and leadership changes in institutional structures (Schriger et al, 1997), planned and unplanned changes to healthcare practices (Massaro, 1993) and attention to social issues (Kaplan, 2001) must also be addressed if the potential benefits of IT in healthcare are to be realized. Thus, the healthcare industry poses important social challenges and intriguing research possibilities for researchers interested in the development and use of ISs and technologies.

Further, Teixeira et al (2010) believe that there are hazards associated with the introduction of ICTs in this domain, in which a great number of projects have failed because of the lack of a systematic consideration of human and other non-technology issues throughout the design or implementation process, particularly in the requirements engineering process. Some studies such as that by Pai & Huang (2011) have taken some of the human factors that are rarely addressed, such as the acceptance of computers by nurses, and organization and management support. Although their aim is the improvement of health service, the studies are still not patient centric. This is supported by the study by Pai & Huang (2011) in which they confirm that, when users feel more satisfied with the service quality of the healthcare IS, their perceived usefulness and perceived ease of use will be higher. Therefore, medical centres should not only focus on these influential forces during the system introduction but should also continuously improve their service qualities. All of these affect users’ feelings about the IS. By continuously enhancing its service qualities, the system would be able to reach its potential full performance.

However, popular literature on the success and failure of IT projects over the last two decades suggests that, among a number of indicators, one should place emphasis on the measure of organizational change that must take place in order to move from the current to the future state of an organization. Using such criteria shows that wider design-reality gaps occur, with a greater likelihood of failure, in projects in healthcare than in projects in an organization with profit motives because of generic differences in the two key stakeholder groups (electorates and customers). Perhaps, one of the most cited reasons for the U.K.’s state-run National Health Service failure in achieving significant benefits from IT projects in the past few decades is that the financial logic of IT-based modernization efforts are typically based on private sector profit motives (Currie & Guah, 2007). This technique only allows for cost/benefit analysis that dictates calculations in which the cost of new technology must be less than the benefit of savings in labour cost.

A patient-user perspective

The patient-user perspective presented here is based on the experiences of the first author, who has been working as an ISs academic for 25 years and who has been a PD patient for the latter half of this time. What is offered here are some major concerns that do not appear to be well thought-out in the literature properly, if at all, but that, in the view of this patient-user, have an important impact on the development of any healthcare IS. Some of the concerns arising from this patient-user perspective have the potential to negate all the benefits of an IS based on the literature.

An attempt to highlight the relationships between these concerns in a diagram failed when it was realized that all six concerns were related to 4–6 of the other 6. This made a diagram rather superfluous, as there are hardly any pairs of concerns that do not have a relationship. This shows the complexity involved when it comes to trying to develop an IS that is fit for purpose.

Although the first author has a particular complaint, PD, the paper is not directly related to the disease but to the treatment of illnesses in general, as the ethnographic studies will show. However, PD is a serious illness for those who have it. The estimates of the number of worldwide sufferers vary considerably, but a feel for the magnitude of the number is given by a conservative estimate of 4–4.5 million. The number of sufferers in different countries is a function of the demography of the country as can be seen from the following quote from my consultant Peter Bain ‘A recent study, conducted in East Boston (U.S.A.) showed that 15% of the population aged 65–74, 30% aged 75–84 and 53% aged 85 years or more had mild Parkinsonism signs’ (Paul, 2009).

First: the doctor–patient relationship and fear

The relationship between a doctor and a patient is very different from that between a customer and a supplier. Apart from a formal lack of profit motive (of the system at least) if things do not work out, the patient could, in the worst case, die. Therefore, it is not surprising that many such doctor–patient relationships are enmeshed in fear, at least on the patient’s side. The call for a patient-centric system is admirable, but one has to remember that the relationship is a function of the patient’s ability to articulate the symptoms and of the doctor to extract them from the patient. It will always be true that more articulate patients are likely to receive better treatment because of their quality involvement in the process. Most medical diagnoses are largely based on what the patient tells the doctor; hence, a poor ‘bedside manner’ in combination with a patient who cannot express herself/himself well is unlikely to benefit much from any kind of IS (including artificial intelligence). In fact, if the patient-user searches the web, say, and diagnoses herself/himself, the relationship is off to a good start, with the doctor willing to listen usually and the patient already fairly knowledgeable.

Second: patient or physician-centric EPR

Although the literature gives the highest priority to having electronic record systems, and that these should be patient centric, it is hardly surprising that there is a lot of difficulty in developing such a system. Clearly, transferring the contents of a paper record to an EPR system will not suffice in future additions to the system if the system is to be used for patient-centric purposes, because physician-centric systems do not collect all the data needed to examine patients. Worse, it is doubtful whether anyone can answer the question: What data should be collected in a patient-centric system? In a physician-centric system, the data are collected for the physician’s own needs, and a reasonable guess at what this might be can be considered by the physician. However, in a patient-centric system, in order to reap all the potential benefits, the data required by the healthcare community at large would need recording in some meaningful way – and who knows what that is!

Third: there are no healthcare savings

There is no limit to the amount of money that can be spent, especially in a climate of rapid acquisition of new and better cures, medical techniques, operating practice and so on. A healthcare system cannot generate ‘savings’ in its operations. Any attempt to do this is just another way of cutting the budget. ‘Savings’ can always be readily spent on any one of many worthy medical activities. You may be able to improve the healthcare provision within a budget, but how would you measure this? We shall discuss quality control below in number 6. Suffice to say if you could measure the quality of healthcare, this might lead to a league table of quality-of-life medical practices, which could be taken to imply that only the top X practices should be funded (which we are not proposing!).

Fourth: healthcare is naturally subject to change

There are medical advances, fashions in treatment, fringe medicine that comes and goes, much learning from different cultures and so on. In a static age of medical discovery, it may be possible to take a steady-state analysis of the healthcare systems and the demands on them and try and make them more effective using optimization techniques or other system ‘betterment’ approaches. Given the speed of current medical discovery, the search for stable systems that can be made more cost effective is a waste of time. Medical discovery is not at a constant rate either. Hence, even if methods of dealing with systems that exhibit constant change behaviour were to exist, these methods would not work for healthcare either.

We shall see in the next section some examples of medical change that make all the paraphernalia of the management sciences ill-equipped to deal with healthcare systems.

Fifth: longevity of patients

Healthcare is for life, a somewhat unusual length of time for a relationship outside healthcare. How many organizations work with the same human being for as long as 70 years? What will healthcare be like in 70 years for today’s newborn baby? What will anything be like?

Sixth: quality control

Many social, economic and political systems have some form of quality control associated with them (for which the first author has to make the comment that they are largely bunk). What is quality control in healthcare? ‘Take 20 patients with a heart problem, carry out necessary heart bypass surgery, and if they all live 50 years’, that is quality and so on (it would take a long time to find out).

Seventh: death is healthcare failure

Death is guaranteed, so healthcare must eventually ‘fail’. Of course, that is a little unfair as healthcare cannot reverse this – death is inevitable. However, any treatment being received has to be tempered by the patient’s recognition that the treatment being received will at best merely postpone the inevitable. If the treatment received has a track record of variable success, then this will or should be at the forefront of the patient’s mind. Such failures may be what you have to expect. No one is guaranteed any lifespan. Quality of life is a highly personal matter that cannot be captured by formulae and numbers. If this is left out, then the IS may advise letting a patient, who is fighting very hard, go. On the other hand, someone who has had enough may be being kept alive. The spectre of death is always there and must never be forgotten.

Ethnographic studies

Having stated the concerns that need to be addressed when looking at healthcare ISs, this section checks them out against seven reverse-role ethnographic studies based on the patient-users’ experiences as mentioned in the ‘Introduction’ above. It is worth remembering that such studies constituted a major part of the way this particular patient-user managed his PD, keeping himself informed about what each medical provider was doing and thereby keeping them informed of each others’ contributions. The authors can only commend this approach to healthcare by patients. The ethnographic studies of a patient-user are given in the first person as they are so particular and personal. Much of this material has been presented in Paul (2009), an autobiographical account of a PD patient, told with black humour to make it more palatable.

Ethnographic study 1: PD diagnosis

I was officially diagnosed with PD in February 2000, having diagnosed myself in late December 1999. On reflection, my health throughout 1999 was not good, but I put this down to being out of condition because my job was so demanding. There had been warnings. I went to a gym to exercise in Spring 1999, and the person showing me how to use the machines observed that I had a lazy left leg. On holiday that summer, after a scare resulting from what seemed to me to be the possibility of my drowning, I collapsed and found my way with difficulty to a health station. There I was put on a saline drip and told to have my ‘lights’ checked when I returned to the United Kingdom.

After a plethora of checks, my doctor diagnosed a ‘funny turn’! I asked my PD consultant about this sometime later, and he made the point that PD is difficult to diagnose. I had done some reading and checked with him whether it was true that my life expectancy at that point was 8–10 years. He cheered me up by saying that the book I had looked at was out of date. There had been significant advances in medication and that a better estimate would be a reduction in life expectancy of around 2 years. As I left my consultant, my happiness was somewhat reduced when I realized that this estimate meant absolutely nothing to me the individual – it might be of some use to an actuary.

Table 3 shows the outcomes of considering each ethnographic study against the concerns from a patient-user perspective of developing IS for healthcare. The text associated with each ethnographic study gives a fuller picture. Therefore, the seven concerns are as follows for ethnographic study1:

Table 3 Table showing how the ethnographic studies fit against the patient-user concerns
Full size table

The patient–doctor relationship was excellent and the treatment and advice first class. But I was still in fear. The record keeping appeared to be both physician and patient centric; that is, records were kept for both interests (see, e.g., ethnographic study 2 below). There was no indication that my treatment was in any way limited by resource constraints. Change was endemic, medicine came and went. My longevity was apparently reduced by 2 years. There was no observable quality control, other than that I slowly got the symptoms under control with medication – but this was hardly surprising as patients tend to have different experiences one from another, suggesting that PD is a family of diseases rather than just the one common disease. Obviously, the treatment is working at the moment (I am not dead), so for me it has been successful so far. This last result is true for all ethnographic studies, and so it will not be mentioned again in the remaining ethnographic studies.

Ethnographic study 2: parallel illnesses

My consultant behaves as though my PD will be the cause of my death in the end, and anything else that comes along must not get in the way. He quickly gets whatever it is sorted out (with the support of my doctor who happily agrees to what he proposes, but does not have to). So in the early days, for example, I went to a deep vein thrombosis outpatient clinic for a check-up. On arrival, the receptionist asked me ‘Heart or Leg’, and my answer of ‘leg’ assigned me to second-class speed of service; ‘hearts’ were the priority. After a 3-hour wait, I went into an open-plan office space where I met a newly qualified doctor (historically known as a junior doctor in the United Kingdom), who had obviously been working non-stop for days, and a very pleasant student doctor. The worn-out doctor kept coming in and out, mumbling about finding this or that form. After about half an hour, the Registrar joined us, casually ignored the junior doctor, checked my legs, told me they needed exercise and inferred that his time was being wasted.

The seven concerns are as follows for this story:

The patient–doctor relationship was satisfactory, but I was in fear until I was informed that all I needed to do was exercise. Record keeping, as I observed, seemed physician centric but not patient centric. I was not aware of any resource constraints other than the junior doctor's exhaustion. The latter could be interpreted as a way of getting more work from the junior doctor, but the quality would be so poor that it could not be effective. I saw it more as a rite of passage. I was not in this particular part of the system long enough to observe change. What I had experienced had no discernible effect on my longevity. I observed no quality control, especially with the junior doctor.

Ethnographic study 3: as easy as riding a bike

In 2005, I mentioned to my consultant that I sometimes experienced chest pains. I found myself in the Rapid Access Chest Pain Clinic. The activities that occur there bear some extensive telling, as I found them to be a typical example of common happenings. On arrival I was asked to fill in a form. I do not know why, because every question on the form was asked of me verbally later, often more than once. In this process I became Paul Raymond. Nearly all hospital forms ask you for your name in reverse order, family name then given name. Many people thought I had filled in the form the wrong way and it took 30 min for me to get my name back. Then on to the bicycle test, where I was asked, along with many questions from the form, when I felt the chest pains as they wanted to recreate the conditions. I said to do that I would need a large meal, two glasses of red wine and a slight incline/hill to walk up. The nurse was nonplussed and suggested I see the consultant without taking the test first. The consultant told me that he thought I needed an endoscopy or some such, and some other rather evil-sounding things. But he wanted me to do the bike test. Back to the nurse and the bike. She wanted me to fill in a disclaimer form and told me why when I asked her. I put it to her that I was in the clinic because there was a possibility I was heading for a heart attack, and there was a 1 in 10,000 chance (or whatever) that riding the bike would give me one. I was asked if I wanted a copy of the disclaimer form, to which I observed that if I died I could not see that I would have any use for it, and if I did not die I certainly would not need it.

I passed the bike test with flying colours, a probability of a heart attack around the level of all males in the population, far better than my age group. The consultant decided not to do the endoscopy but instead to prescribe some medicines for me. The first was a beta blocker to keep the heart nice and calm. The trouble is, as a Parkinsonist, I need speeding up not slowing down, so they were quickly jettisoned. The second was a spray, which I understood works as follows. If I felt the chest pain, then spray under my tongue. If the pain persists, spray it again. If it still persists, spray once more. And if it still persists – well, you are dead! Comforted by such imagery, I was really cheered up when the pharmacy told me that if I use the spray it would give me a headache.

The seven concerns are as follows for this story:

The patient–doctor relationship scarcely existed, and the doctor was so keen to diagnose without much evidence. Clearly, I was in fear as heart attacks can kill. Record keeping was entirely physician oriented. There were no discernible resource constraints. My longevity was an issue, and I could be dead at any moment! Quality control seemed largely to be a function of the percentage of patients who have heart attacks.

Ethnographic study 4: stress and pain

I was not quite free. A stress echocardiogram was ordered, and eventually (for reasons described in my book, Paul, 2009) I took the test. A normal echocardiogram was first, to benchmark the data, and then a chemical was introduced that made the blood system work harder. When things were quite difficult I was asked on a scale of 1 to 10 where I was. I said 8, but I should have said 11, because more chemical was introduced, and I really felt it. When the test was complete I asked the equipment operators if they had ever tried the test for themselves, and they replied negatively because the test was ‘far too invasive’. I quietly agree. I returned a week later to get the official result, and the clinic head congratulated me on two pieces of good news. Two? I asked what they were and he replied that my heart was essentially sound and that I had been cured of PD! I informed him that there was no cure for PD, and he seemed surprised. So much for total medicine.

The seven concerns are as follows for this ethnographic study:

The doctor–patient relationship was friendly at the human level but professionally poor. Fear worsened, and I was glad to be out of there. Record keeping was entirely for the physician. There were no obvious resource constraints, and this was a relatively new procedure that was changing with use. Again, as in the previous ethnographic study, heart stories can be quite short. The quality control questions are answered by the question I was asked of where on a scale of 1 to 10 I was.

Ethnographic study 5: side effects can kill

In 2008, the PD medication I was taking was found to have caused me a lot of damage. Starting with a blood test at my doctor's, which showed my haemoglobin count approaching half of the recommended levels, I found myself over the next 6 months visiting an outpatient clinic of one sort or the other weekly. I visited the Gastrointestinal Surgery, the National Pulmonary Hypertension Service, Respiratory, Cardiology, Haematology, Clinical Neurophysiology, the Neuromuscular Unit and probably one or two that I have forgotten. There were some delays in gaining entry into the system, which was resolved by an arrangement with Accident and Emergency that I should be a walk-in patient and take it from there. I was visited by the A&E Registrar who queried this irregularity until informed by his colleagues that they had arranged it. The Registrar was not amused, but when the results of some tests came in he became more sympathetic and added to the tests required once he saw the appalling data.

The seven concerns are as follows for this ethnographic study:

The patient–doctor relationship was extremely variable, off-hand at one level to extremely supportive, interested and willing to dirty hands for me. Most encounters generated fear in me. Whether it was heart, or blood, or lungs, none of these seemed good news to me. There was little evidence of collaborating between specialities (only two specialities had a joint meeting to my knowledge). The patient recording systems being used were largely physician centric with little evidence of patient-centric data. There was no evidence of the treatment being restricted by resource issues. Evidence of change was a bit variable, as was the potential effect on longevity. There was not much discernible evidence of quality control.

Ethnographic study 6: thanks but no thanks

As diagnosis and treatment took off apace, the change in PD medication started to have effect, and slowly my body returned to its usual state. One by one the clinics discharged me. In this process, I was booked into hospital for a week with the National Pulmonary Hypertension Service for a series of tests, culminating in microcameras being inserted through sensitive parts of my anatomy to look around. The tests went well, and by the Wednesday it became quite apparent that the symptoms were much improved. A huddle of doctors appeared in the ward, whispering to each other and sneaking looks at me. I guessed what the problem was and when they ventured over I started the conversation with the suggestion that, as the tests had gone so well, did we need the camera work? The sense of relief could be gauged by the worried expressions turning into smiles and after one more straightforward test I went home.

The seven concerns are as follows for this ethnographic study:

The patient–doctor relationship was very good, although fear, because it was the heart, was ever present. Recording data appeared to be both physician and patient centric. Evidence of resource constraints might explain the relief of the doctors at not having to conduct the expensive planned testing. Evidence of change lay in the fact that the whole approach was new. My longevity was a function of my heart with a variable possible number of outcomes. There was no evidence of quality control.

Ethnographic study 7: double deep brain stimulation (DBS)

By 2009, my PD got worse, and I started falling over, so much so that 2 years later I was falling over 50 times a month. Although I could anticipate most falls and ‘roll’ into them to minimize damage, it was obvious that if I did not find a way out I would eventually break something, which, because of PD, would make recovery extremely difficult. My consultant had mentioned an ongoing clinical trial that he was a part of, essentially a double DBS operation. DBS operations are now quite common, but the clinical trial operation implants four electrodes in pairs into different parts of the brain to influence the symptoms of PD and also to affect balance. In late 2010, I agreed to consider the operation. It was made clear to me that I could withdraw at any time for reasons that will become apparent. I went through a large number of tests, including psychological, and it was confirmed that I was a suitable patient. The operation itself costs of the order of $50,000 (there were 20 people in the operating theatre) of which half was to be met by the company supplying the electrodes, batteries and so on (it was in their interests for the clinical trial to be a success) and the other half by the local area health service where I lived.

The operation was conducted in two parts in June 2011, a week apart, the first to drill the four holes into the brain and implant the electrodes and the second to implant the batteries and connect the wiring beneath the skin. I was awake throughout the first operation, as to some extent the surgeon is guided by the patient's responses (hence I could withdraw at any time since the idea of an awake bolted-down reluctant patient being operated upon does not bear thinking about). When I was eventually switched on, the unknown factor was the amount of medicine I now required. The electrodes were set to take up most of the desired benefits, but it is not yet possible to set them accurately enough, and in any case some PD effects are not covered by these electrodes. Hence, some medicine must be taken as well (much less fortunately than before the operation, which can be shown to more than offset the operation costs). However, how much medicine? It took about a week to get to a reasonable level, starting with a dose that, in combination with the trauma of the operation, left me flying on the ceiling!

The seven concerns are as follows for this ethnographic study:

The patient–doctor relationship was excellent, although some members of the research team sometimes forgot that we patients were human beings as well as research tools. This ethnographic study had the same main doctor as ethnographic study 1. But even so, rather than just fear, I was terrified at the prospect of holes being drilled while awake. Data recording appeared to be both physician and patient centric. This is an obvious example of resource limitations (the local area health service could have said no). The operation was part of a clinical trial, and in itself, and in the trial outcome, is a procedure that inherently changes things. Patient longevity is unknown, but no trialist has died yet. I am patient number 6, whatever that says about quality control.

Summary discussion

The table shows that my relationship with doctors is usually good, but fear is ever present. My PD consultant is both physician and patient centric, but most doctors are just physician centric. Two clinics held a joint meeting to discuss my case, which shows some patient centricity. I rarely found resource a problem for my treatment, which I guess says something about the severity of PD. In the Pulmonary Hypertension Unit, relief came from my proposing not to undergo some expensive tests. Change, as far as I could tell, varied a lot, from the creation of new clinics, to PD's fast changing medication, and on to brain operations. The effect of any treatment, and the reasons for the treatment, has variable impact on the length of life. For my DBS operations, there was a 1 in 2000 chance of death occurring during the operation, a statistic I would not have worried about if there were 2000 of me. But there is only 1 of me.

Conclusions and a way forward

The purpose of writing this paper was to take the first author's kaleidoscope of patient experiences and report back to his professional community the observation that ISs are largely undertaken without giving much thought to the patient. In this paper, we have shown that ignoring patient-user concerns is likely to leave IS in healthcare as unsuccessful as it currently is. We have shown patient-user concerns through some real and personal examples of healthcare in action, and these experiences reveal the importance of these concerns in the development of any healthcare IS.

The research in this paper may not be considered science, but it still informs scientific debate. A list of patient-user concerns that influence IS development has been drawn up (to be amended, expanded and improved upon by future research). This paper shows that, when developing IS systems in healthcare, one should remember to consider these concerns. Logically, there is no reason for not doing so. None of the concerns are obviously ridiculous. Consideration of them can at worst be redundant in some instances but more likely beneficial.

Let us now look at a way forward. There is a possible way around some of the healthcare problems, and that is to redefine what healthcare is about in stark terms. It is a feature of modern life that names can often imply opposites – for example, the role of the Ministry of Defence is war. Similarly, we could consider health in terms of death. If we did, it might run something like this.

From the moment one is born, the one thing you can be sure of is that you are going to die. During your life between birth and death, you will undergo various illnesses, some mild and some severe. If you are lucky, there will be intermittent periods of time when you feel you are ‘well’ (how does one define this?). The role of healthcare, or Deathcare, is death avoidance. Death avoidance is automatically a whole person idea, as there is little point in saving a PD patient from many PD symptoms if he is about to die the following week of a brain haemorrhage (to make the point by exaggeration). EPR records in this way of thinking would have to have the complete health record in order to do the job.

No one is suggesting that we call the Ministry of Health the Ministry of Death. But it can be seen that playing mind games like this could improve considerably on what we achieve now.

There is a drawback, a subject gentle folk do not like to discuss. In a resource-limited system (and all healthcare systems are resource limited), decisions are taken as to who is going to die. This is largely taken by doctors in spite of the Hippocratic oath, which probably should have tagged on it ‘while life is worth living’. It is possible that it is convenient that doctors make such decisions, as patients are as likely to give untenable or ‘illegal’ answers as not, and family and friends might not like such decisions coming their way. This is left as a thought for the reader.

In this paper, we have emphasized the difficulties associated with developing ISs for healthcare and hopefully convinced some if not all readers of the paucity of the current literature on the subject, as well as the even worse implementations. This paper has hopefully done its job of using a patient-user perspective to give the problem a more all-round airing.

If the reader were to ask how the problem should be tackled, that would be the subject of another paper. But a clue might be to start with Paul's definition (2010) of what any and every IS is – information systems is Information Technology in Use.

This is the starting point for recognizing the IS user as a part of the IS and that an IS has meaning as an object only at one point in time and a different meaning at another point in time. Then the ideas behind Living ISs (Paul & Deshpande, forthcoming) might give guidance in this area.

Note that IT in use in healthcare would include the patient.