The role of the Electronic Patient Record in clinical research

Transcription

1 The role of the Electronic Patient Record in clinical research Rune Pedersen & Gunnar Ellingsen University of Tromso Abstract. Electronic Patient Records have many purposes in hospitals. They are expected to be a foundation for the clinical treatment and care processes, reimbursement issues, as well as a foundation for research purposes. Accordingly, there is a gap between the information found in the EPR and the information required for clinical research. We explore what this gap consists of, how it is managed and who is responsible for closing it in order to make the information usable for research purposes. We wish to highlight the work that goes into this process conducted by both the research staff as well as the ordinary health workers. With this we want to move away from the perception that research data is given as long as a hospital-wide EPR is in use. We have asked, what is the nature of integrated systems in clinical research built on the notion of quality, and secondary use of documentation? Introduction Electronic Patient Records (EPRs) have many purposes in hospitals. They are expected to be a foundation for the clinical treatment and care processes (Ellingsen. 2004), reimbursement procedures through the diagnosis-related group (DRG) coding systems which is a reimbursement issue, as well as being a foundation for research purposes. One research area is clinical research. Clinical research or clinical trials are studies that test how new medical approaches or procedures may improve the treatment of patients. Each study answers scientific questions and tries to find better ways to prevent, screen for, diagnose or treat a disease. Clinical trials may also compare a new treatment to a treatment that is already available. The general idea is that by using a large-scale EPR, you may extract whatever information you need for research purposes, especially if the EPR consists of lot of structured data elements. From a purely technical perspective it is assumed that this data may be extracted, compared, accumulated and summarized, see for instance (Berg and Goorman. 1999), however this view is problematic as the

2 research studies often requires slightly different data as well as data of a higher quality than what is needed in daily clinical work. Sometimes, even different information sources are used to supplement the clinical information found in the EPR. Clinical research documentation is also attached to different types of standardization that should make data transmittable over country barriers. Like Common Terminology Criteria for Adverse Events v3.0 (CTCAE) which includes Adverse Events applicable to all oncology trials regardless of chronicity or modality. Accordingly, there is a gap between the information found in the EPR and the information required for clinical research. In this paper we explore what this gap consists of, how it is managed and who is responsible for closing it in order to make the information in the EPR usable for research purposes. Our contribution is to highlight the work that goes into this process conducted by both the research staff as well as the ordinary health workers. With this we want to move away from the perception that research data is given as long as a hospital-wide EPR is in use. Theoretically, we draw the notion on information infrastructures (Hanseth and Monteiro. 1998) for two important reasons. Firstly, this notion reflects very well the EPR concept as a hospital-wide infrastructure. Secondly, information infrastructure pay attention to the social aspects of technology, and this is particularly relevant for our case. Empirically, we have studied the process of collecting data for clinical research in the oncology department at the University hospital of Northern Norway (UNN HF). They have a historically long tradition in collaborating clinical trials, and have been organized with extended personnel and physicians. The physicians have coordinated both trials initiated by the medical industry and by local researchers. Another important fact to add is that cancer research as a field is broad and the possibilities for attending trials are many. The reminder of this paper is organized as followed. First, we elaborate on the theoretical foundation for the paper thru secondary use, quality and information infrastructure than we describe the settings for the empirical investigation and describe the method used, followed by the case analysis. In the conclusion and implications, we reflect on how the case fits with the notion of information infrastructure and secondary use of information. Theory Secondary use of personal information in research The discussion around secondary use of EPR data is found world wide. (Powell, J et.al. 2005) (Safran et.al. 2007) (Eike-Henner. 2004).

3 On a world basis secondary use of health data applies personal health information (PHI) for users outside of direct health care delivery. It includes such activities as analysis, clinical research, quality and safety measurements, public health, and commercial activities. (Safran et.al. 2007). Other areas are, international research particularly in the U.S. has pay-forperformance models where commercial enterprises pay for health care data to derive products and services to sell. (Safran et.al. 2007). The American medical informatics association (AMIA) has initialized a national discourse on secondary use of data for research. Points to mark are benefits and risk, particularly concerning patient rights, anonymization, audit the use of the data, and how to give feedback on results from the research. (Safran et.al. 2007). The Norwegian social- and health department is working under the same presumption in their report samspill 2.0 to get a seamless ICT system where information can travel freely country wide (Helse- og Omsorgsdeartementet Samspill ). The modern health care institutions spawn the need for a solution making it possible to guarantee data exchange across different types of health information applications, being multi-site organizations that have implemented heterogeneous information management systems interacting with distributed databases. The implementation process is depending on a 3-tier structure including reusability, flexibility, and cost reduction. (Chu. Cesnic. 2000). In much of the literature on EPRs, this secondary utilization is only seen to depend on the question whether the IT connections are in place. It is then simply a matter of selecting which information to transport and to where. Berg and Goorman says that, the further information has to be able to circulate (i.e. the more diverse contexts it has to be usable in), the more work is required to disentangle the information from the context of its production. Who has to do this work, and who reaps the benefit? (Berg. and Goorman. 1999). This thus underpinning the belief that data is easy extracted and used for research purposes as long as the information is easy extractable from the EPR, and the number of additional documentation sources are limited. Information Infrastructures (II): A more complex view on the EPR is to conceptualize it as an information infrastructure. The changes towards a more integrated and efficient health care thru the introduction of EPRs in hospitals increases the need to coordinate work activities beyond organizational boundaries. This includes the flow of patient information between departments, institutions, and between the EPR and other information sources like the CRF, and hand-written forms. Health care work activities exhibit strong network characteristics but current technologies and use patterns of EPRs do not support this kind of cooperation. (Nilsson. et.al. 2002).

4 The information infrastructure at UNN HF today exist of the EPR, which includes DIPS, other information sources that we will discuss, and human actors. To understand the design or the change of IIs the understanding of installed base, what already exist and is implemented, is crucial (Hanseth. 1996). The IIs are never built from scratch, rather in a process of building upon what already exists. As the installed base grows, it becomes increasingly more different to change it. It is apparent that in the modern healthcare, with constantly increased integration and collaboration, an EPR is no longer a tool for single users only. It becomes by default part of a large and complex II, which is used for various purposes (Nilsson. et.al. 2002). Infrastructural arrangements denote the various entities that support integrative initiatives, such as the electronic patient record. Together with standards, procedures, and classification schemes the EPR is intended to improve planning of the patient case, higher quality of care and better cost containment (Ellingsen. Munkvold. 2006) all which are seen as important for growth and continuance of an II community. (Hanseth and Monteiro. 1998). IIs are more than pure technology they are rather socio- technical networks. Infrastructures are heterogeneous concerning the qualities of their constituencies. They encompass technological components, humans, organizations, and institutions. This is true for information technologies in general, as they will not work without the support of people. An information system does not work either if not the users are using it properly. (Hanseth and Monteiro. 1998). As many layers of technology accrue and expand over space and time, systems of classification (and standardization) form a juncture of social organizations, moral order, and layers of technical integration. (Bowker and Star. 1999). Due to the independent development of component systems, heterogeneity occurs at various levels and for various reasons. (Hasselbring. 2000). Fundamental to the ongoing efforts of overcoming institutional and interdisciplinary boundaries are infrastructural arrangements such as electronic patient records (EPRs), standards, procedures and classification schemes. The advantages of the electronic health care record over its paper-based counterpart are clear, it is always available, information can be transferred, and it can support different views of the record for nurses, doctors and other users. (Grimson, J. Grimson. Hasselbring. 2000). Thru give methodological attention to the architecture and use of different systems like EPRs and other health record systems that are deeply attached to clinical research, we can achieve a deeper understanding of how it is that individuals and communities meet infrastructure. One of the main tasks of this study is to look at the documentation-flow between patient information and the case report form (CRF), this will be done by looking closer into the way which

5 the EPR is used, the collaboration over professional boundaries and handling of other important actors human, and non-human. Identifying the key aspects of Information Infrastructures, and in particularly what makes them different from information systems. Infrastructure is defined as; An substructure or underlying foundation; esp. the basic installations and facilities on which the continuance and growth of a community, state etc. depends as roads, schools, power plants, transportation and communication systems, etc. (Hanseth and Monteiro. 1998). Infrastructures are enabling, shared and open. Infrastructures have a supporting or enabling function, meaning that it is designed to support a wide range of activities, not especially tailored to one. It is enabling in the sense that it is technology intended to open up a field of new activities, not just improving or automating something existing. This is practically a notion that supports the integration of clinical research in a hospital department, the flexibility towards supporting a wide range of activities. An infrastructure is shared by a larger community, and by its members in the sense that it is the one and the same object used by all of them. Infrastructure cannot be split into separate parts being used by different groups independently. The different elements of an infrastructure are integrated through standardized interfaces. In the case of IIs the role of infrastructure is believed to be important as its enabling character point to what may be kept as a stable basis in an increasingly more complex and dynamic world. Infrastructures are open in the sense that there are no limits for number of user, stakeholders, vendors involved, nodes in the network and other technological components, application areas or network operators. (Hanseth and Monteiro. 1998) A relevant example of this which is mentioned in the book but also a part of this particular task is the exchanging of information among medical institutions even in other countries. Information gathered at the hospital is in a way recoded thru standards and the patients anonymity are maintained coding their participation. Accordingly the hospital is sharing information with virtually any other sector in society, and the information exchanged among different partners is overlapping. More exactly different hospitals are gathering information which is sent to the main investigator of the study, all information from every participant is united during this process. An unlimited number of users, developers, stakeholders, components and use areas imply several activities with varying relations over time, varying constellations and alliances, changing and unstable conditions for development and changing requirements. In sum all this implies heterogeneity.

6 Quality Traditionally uniform and standardized IT-solutions in health care are considered mechanisms for increased control, efficiency and quality. (Ellingsen, G 2004.). The key question is how to balance local use and heterogeneity and take into consideration how the infrastructures influence each other. In a book about extremely routine service-work in McDonalds franchising companies (Leidner. 1993) describes how management maximizes its control of the work process in order to predetermine how to conduct those tasks and thus promote increased efficiency and simplicity. Leidner also elaborate the relationship between standardization of work and quality, this is a key tool for offering a proper quality to the customers. A precondition, however, is that the customers accept standardization of themselves as customers and play their role defined by McDonalds (Leidner p.25). If the customers are parting from the predefined script wanting more flexible services than the company will allow, consequently when both efficiency and quality are major goals, quality becomes subordinated. Clinical research procedures are standardized and shaped towards integration with the patient documentation of the hospital. Good quality is assured thru the guidelines of Good clinical practice, monitoring of the elaboration of data, and standards used to globalize the grade of e.g. adverse events. Good Clinical Practice is an international ethical and scientific quality standard for designing, recording and reporting trials that involve the participation of human subjects. Compliance with this standard provides public assurance that the rights, safety and well being of trial subjects are protected consistent with the principles that have their origin in the Declaration of Helsinki, and that the clinical trial data are credible ( d ) Another assumption is that the use of EPRs could give a better information flow among actors in the socio-technical network surrounding it. In this paper, we define integration of documentation over professional boundaries as importent for succeeding with increased quality of data entered into the case report form of clinical research. Our assumptions are made thru elaboration on three perspectives which are focused on physicians, nurses, and clinical research. Both physicians and nurses documentation is varying valuable integrated with clinical research, and the extended standardization of research in general should be used as a tool for quality assurance of those processes in the same time as it is demanding extended documentation. Standardization efforts should balance the management level s need for increased coordination and control, and the local levels need for flexibility. (Ellingsen ).

7 To make sure that unsuitable circumstances (e.g. proliferation of incomplete solutions) are not allowed to take root [so] standardization must be started as soon as possible in order to set the development in right track (De Moor p.4) Method Research settings The research was conducted at the University Hospital of Northern Norway (UNN HF), having about 5000 employees, including 450 physicians and 1000 nurses. The hospital has a total of 600 beds. The actual study took place at the Oncology department. The Oncology department is a regional center for medical oncology and radiation therapy in Northern Norway. The department offers exposition and treatment to grown-up patients at the age of 16 years and up. The department was established in 1985, having 120 employees in four units, the Ward, the Radiation department, the Outpatient clinic, the Clinical research unit (EKF). The department has an extended activity and treats patients from the Health region North which includes inhabitants distributed over the range of square kilometers as shown in (Figure I). The ward has 30 beds distributed around singular, double and four-bed rooms. The approximated bed time is 6,2 days. The ward has 43 nurses included the management and 3 assistant nurses. Of the nurses there are 25 with complementary education. There are also 9 secretaries and 17 physicians who are working around the department including the Radiotherapy unit and the Outpatient clinic.

8 Figure I: the web page at the Oncology department. Shows us the priority in medical oncology in the region of Northern Norway. The patients treated at the ward are typically in palliative treatment, some of those are just treated waiting for supportive care from their municipality or their local nursing home. Another group of patients are those treated with high- dose chemotherapy or other regimes that is depending on a 24 hour follow-up over several days or weeks. Patients treated with chemotherapy in general are often getting their first treatment at the ward. They are further transferred to the polyclinic or their local hospital for further treatment. Patients receiving radiotherapy treatment are both patients at the ward and polyclinic patients, depending on which physical condition they are in. Research method In practice the flow of health care work activities is often much less linear than it is in other arenas, with flexibly defined roles and overlapping. Because of this complexity, standard quantitative research methods such as surveys fail to expose the subtle problems. Qualitative research techniques, on the other hand, can provide deep insight and can both identify problems, and answer the why, and the how questions that quantitative studies cannot answer. (Ash. Et.al. 2003). Interpretive methods of research in IS are aimed at producing the information system, and the process whereby the information system influence and is influenced by the context. Interpretive research can help the IS researcher to understand human thought and action in social and organizational context, it has the potential to produce deep insight into information systems phenomena including the management etc. (Klein. and Myers. 1999). As of this, the study

9 adheres to an interpretive research tradition. The principles are derived primarily from hermeneutics, meaning that observations, document analysis and interviews are all considered, bringing meaning to the complex Whole of the data. The hermeneutic circle, a complex whole of shared meanings emerges. This interpretation should be used in applying the principle of interaction between the researcher and the subjects (Klein and Myers p. 71). Philosophically this is an ethnographic study with an interpretive research approach with focus on interviews, document analysis and observations as methods. Totally the first author conducted 12 hours of observations, interdisciplinary meetings, nursing handover and writing reports and nursing plans. Handwritten field notes were transcribed shortly after gathering information. We did no special considerations due to electing type of actors, only interactions like described thru handovers and meetings. The main reason is the fact that the first author almost ten years of work practice from this particular ward and consider this practice very valuable counting as field-work. The first authors extended knowledge of the practice of the department makes us suitable to highlight potentially different interpretations thru experience.there were conducted five interviews. The interviews lasted an average of 0,5 to 1,0 hours, (Table I). I also used some time doing document analysis after the interviews in order to validate on statements given during the interviews. Interview Duration of Method Timepoint object interview Nurse 1 1 hour, over two One phone interview, and one interview Feb 2008 interviews taking notes Nurse 2 1 hour, over two Two interviews taking notes Feb 2008 interviews Nurse 3 30 minutes Interview using Feb 2008 Nurse 4 30 minutes Interview using Feb 2008 Physician 20 minutes Phone interview taking notes Feb 2008 Table I: Overview of all interviews included in this article All interviews wore transcribed shortly after the interview. All the transcription has been done by the first author, according to (Malterud. 2003) this is crucial in case of clarifying uncertainties and the meaning of not clarified sentences. The three interviews that where done taking notes using countersigns to get all the crucial information was transcribed immediately after the interview. The interview objects have also read thru the transcriptions and approved the content. The last two interviews were done using a tape recorder, the questions was the same, few and open, and the interviews were transcribed shortly after.

10 All the nurses and the physicians were selected because of their extended work-practice at this particular ward, making it possible to get historical information about clinical research. This was crucial because of alternating activity. In practice, the problem of access turns out to be a cluster of, for the most part, manageable problems which includes gaining entry to the work setting, gaining acceptability, being able to hang around and more. Even if the entry is successful, some areas might be regarded as sacred and of limits to the observer. (Randal. et.al. 2008). The fact that this particular department has been the first authors employer thru almost ten years, and that clinical research documentation was the focus of the first authors work practice erased most of the possibility to get a slow start or no access to the department. Insiders, of course, can lat claim too much of the expertise associated with the domain and already know about important features of the job in hand (Randal, Harper. chapter 6, p ). A result of this is that the learning curve is less steep. In contrast, some have argued that there is an advantage to outsider accounts, notably that a more dispassionate account will be provided and there is less risk of going native. The issue is what activities an ethnographer will need to undertake for his views to be treated as from the inside rather than from the outside of the work. The amount of people in the work practice and the considerable work-pressure could make it difficult to communicate why an ethnographer is her? Personally the first author entered a new role as a researcher, the employees know me as a study nurse. Getting accepted in this new role could be the same as getting accepted at all. The first authors approach as an ethnographer towards the mangers of the department has been open about the purpose of the study. My presence at the department has been announced locally and I had the possibility to make individual contacts planning interviews continuously. Case The EPR at the University hospital North Norway In 2004 the University hospital North Norway implemented a new large-scale EPR named DIPS EPR delivered by the vendor DIPS ASA.

11 This EPR covers somatic, psychiatry, nursing, physicians and laboratory. All together this makes a good starting point to extract information for research purposes. The expectation towards clinical research at the initial implementation time of the EPR was high. Researchers saw the possibility to extract information e.g. about blood pressure directly from the EPR to research databases in some kind of a time sawing procedure. Today the physician opinion is that the EPRs are based upon the same system as old-fashioned "paperjournals". The only difference is that information today is in electronic form. The potential benefits of storing data digitally are thus far from being exploited. Clinical information should be categorized if possible. Categorization of clinical data should be according to well known and internationally establishes clinical instruments and grading systems. Registration rates increases and the data quality is improved by categorization. Today, clinical departments do registration as a separate activity, which is time consuming and demands extra resources. (Physician) Data could easily be stored within the EPR or easily extracted from the EPR, in such a way which allows for analysis and development of reports within EPR or potentially outside EPR system for more complex data analysis. (Physician) After implementation there are multiple sources of documentation that need adjustment towards integration with DIPS, clinical research is one of those sources, containing a large grade of integration possibilities. The research environment at the Oncology ward Being a University hospital demands focusing on clinical research, the Oncology department have a long tradition of participating in clinical trials. Everyone working at the department knows that the patients from time to time are getting their treatment thru clinical studies, and that special effort has to be offered to get the patient thru the treatment. The environment is always ready for research patients, but the challenges can be difficult to handle in a shaped environment because of the day to day work that can be demanding. The process of gathering information which is used in research has become easier, earlier we had to run between departments to get copies, no we find most of it in the EPR (Nurse 1). The study nurse s work is organized in the Clinical research center (KFS), but the focus of the work is against patients at the Oncology department, and thereby protocols initiated in cooperation with researchers at the department, mainly

12 medical oncologists. The employment fraction is under the Oncology department, but the location of her office is in the Clinical research center. The study nurse is responsible to accomplish the clinical research work under guidance of the physicians. The physicians are responsible for research thru representation in national interest groups, and are the once responsible have extended work practice and research competence within that particular field of research. The main task is integration and use of study specific procedures as patient treatment, x-ray, collaborate source data, laboratory tests and shipment of specimens as well as quality assurance in general Selecting the right patient to the study The study nurse and the physician responsible are selecting patients thru the outpatient clinic and the wards incoming list. Each protocol has inclusion and exclusion criteria describing which patients to include in the particular study. This criterion is typical, as followed and in order, type of cancer, stage of cancer, other specifications like pathology and immunology, gender, age, blood samples, and radiology. The criterion is described in the study protocol the study nurse is also responsible for preparing short versions of each protocol which is published on the internal web pages as illustrated in (figure II). Figure II: This print screen illustrates which clinical studies is active at the department, and which are in follow-up. Each study are also linked up to other pages with general information about the study. The study nurse and secretaries are informed about the possibility of the inclusion in a clinical trial thru the physicians incoming list, from this point on they use the internal documentation on the web to get information about the

13 particular study and which coding to use while ordering blood samples, x-ray and other relevant procedures The arrival of the patient The patients includable are informed about the possibility of enrolment in a study in advance, more specific before entering the department or first thing after arrival. Normally patients take blood samples and CT-scan before entering the department, in case of inclusion the samples has to be taken after inclusion, this is because of the possibility that the procedures diverge from the standard procedures. The study nurse inform the patient about study specific procedures at his/her arrival, this includes the signature of the informed consent that gives the physicians and the study nurse the opportunity to carry out with the study procedures including samples, the informed consent is also an insurance document for the patient. It allows personnel to transfer information from the source data to the clinical research form (CRF) and further to anonymous data sources. A picture of a typically CRF page is indicated in (Figure III) on the next page. The next phase is general quality assurance thru the patient trajectory included the follow-up which ends according to protocol procedures. The trajectory consists of e.g. weekly treatment visits demanding laboratory tests, blood pressure and evaluation every second or third time. Most studies are quality assured thru sequent monitoring from the responsible group or medical company. The monitoring is when the Clinical Research Associate (CRA) typically working for the medical industry is gathering information due to the collection and registration of patient documentation from the patient journal to the Case Report Form (CRF) 1. 1 A Case Report Form is a paper or electronic questionnaire specifically used in clinical trial research. The Case Report Form is the tool used by the sponsor of the clinical trial to collect data from each participating site. All data on each patient participating in a clinical trial are held and/or documented in the CRF, including adverse events. The sponsor of the clinical trial develops the CRF to collect the specific data they need in order to test their hypotheses or answer their research questions. The size of a CRF can range from a handwritten one-time 'snapshot' of a patient's physical condition to hundreds of pages of electronically captured data obtained over a period of weeks or months. (It can also include required check-up visits months after the patient's treatment has stopped.). Case report forms contain data obtained during the patient's participation in the clinical trial. When the study administrators or automated mechanisms process the CRFs that were sent to the sponsor by local researchers, they make a note of queries. Queries are non-sensible or questionable data that must be explained. To ensure quality control, these queries are usually addressed and resolved before the CRF data is included by the sponsor in the final clinical study report.

14 Figure III: This illustrates how an CRF page typically looks like. The top row has a sign on field for the physician responsible. Including patient number, or in this case birth date, since this particularly study also is used as a national register. Other information is blood samples, treatment, radiology and pathology Establishing temporary information sources Not everything can be found in the existing information sources. Additional information that only is stored in the CRF is paper forms from local hospitals or General practitioners where the patients often take laboratory tests or CT-scans during the study trajectory. This doesn t include information that is scanned into the EPR. This is handwritten information which is attached to the CRF, a ring binder, and stored with it for as long as the guidelines demands, normally 15 years. The information is stored because of the fact that all information entered in the CRF need a document source which in this case is a paper form from the local hospital etc Reminders The study nurse makes extended efforts to get all information applicable for each study. The Yellow note is a possibility in the DIPS system to admonish the physicians to take specific laboratory tests or other study specific procedures, what is important to document in this particular protocol, and how to document it? It is also used to get answers to questions about the study or the patient. The study nurse establish the yellow note after opening the actual patient screen in the EPR, the next step is to decide which physician that shall receive the note. After filling out the suitable information or questions the note is sent to the physician. The study nurse will receive information on whether the note has been received,

15 or a message in return answering questions added. The yellow note appears when the physician opens the particular patient EPR. It appears in an automatic list of work tasks together with other tasks concerning this particular patient. The same yellow notes are used on the outpatient clinic, the nurse use it to give messages to the physicians in front of the patient visit. The type of messages is varying but is normally related to earlier visits or phone calls to the nurse close in time to the visit. This could be adverse reactions due to treatment, patient requirements like suspension of treatment or the need of a medical certificate due to treatment. We use the yellow notes in the communication with the physicians, sometimes this is the only way we communicate. (Nurse 3) In case of clinical research there is specific study or clinical protocol codes that are to be added on the requisition ordering laboratory tests. The study specific tests are often other tests then which is taken in general for that particular disease. They have standard tests e.g. to adjuvant treated breast cancer. When the requisition is attached with the right code the laboratory staff knows which tests to take. There are similar challenges ordering x-ray or CT-scan from the radiology department. The measurement of disease using e.g. CT-scan has to be done by standardized procedures. It is therefore important to give this information to the radiologist as a part of the radiology requisition. The study nurse uses the yellow note to inform the physicians that secondly inform the radiologist Repair work Documents as the chemo-therapy work-chart which are filled out both in research and in general treatment are done in the ward. The work-chart is illustrated in (Figure V) on the next page. I ll think we are very accurate with the work-sheet for administrating the chemotherapy, none of us dare to be sloppy with handling chemotherapy. We are afraid to do mistakes, we write everything, start, and stop, adverse events and blood pressure (nurse 4).

16 Figure V: The chemo-therapy work-chart Problems occur in relation to clinical research, specified information disappear when the outpatient clinic nurses don t fill out the forms. Information as start and stop time for chemotherapy administration is information that only is used documenting clinical research. The study nurse comment; I have, after gathering experience seen that the work-charts for chemotherapy is essential for documentation related to patients in clinical research. I don t have the time to monitor every patient visit personally. The routines at the polyclinic have to be changed so that we don t loose more information (nurse 1). The work chart and the work chart description is partly a quality assurance for securing that procedures are done in straight order, dependent time limits etc. It also contains crucial information concerning adverse events, which are used to inform the patient during the treatment. This is a good way to ensure the quality of information. Additionally to this the study nurse have the opportunity to adjust the work chart individually for each clinical study, if the study require start and stop date for the chemotherapy treatment this can be added to the chart for this particular study. All information concerning treatment is potentially added to this form. The physician documentation is mostly used and has the greatest significance, but the nursing documentation is used, but more confined. Documentation made by nurses is mostly used in case that the physician notes lack information. The study nurse and the physician will typically and

17 retrospective screen the nursing reports for valuable information that could clarify the situation. We are a bit sloppy on where we put information like blood pressure and temperature, it is supposed to be written in the curve but sometimes it is impossible to find. Instead of attaching the information during the day we put it in the nursing report, and as you understand this don t satisfy the physicians. (Nurse 3) When procedures are not followed this creates problems in the documentation routines. In this specific example the physicians are used to find e.g. blood pressure in the paper chart, the lack of this specific information results in documentation failure. The physicians miss the opportunity to document the blood pressure in their e.g. discharge note which leads to that this information that should have been documented both in the chart and the discharge note is gone. Secondly, the study nurse is able to find it in the nursing documentation, having knowledge about routine failure. Data for giving ECOG status is often found in the nursing reports and the nursing plan. The meaning of the ECOG status is described in (Table II) below. If this concrete information is lacking from the physician incoming journal or discharge note the most reasonable place to look for this type of soft data is in the nursing plan/report. Soft data, typically describe the patients physically, social and psychically condition. Example of ECOG status; ECOG PERFORMANCE STATUS* Grade ECOG 0 Fully active, able to carry on all pre-disease performance without restriction 1 Restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature, e.g., light house work, office work 2 Ambulatory and capable of all self-care but unable to carry out any work activities. Up and about more than 50% of waking hours 3 Capable of only limited self-care, confined to bed or chair more than 50% of waking hours 4 Completely disabled. Cannot carry on any self.-care. Totally confined to bed or chair 5 Dead Table II: As published in Am. J. Clin. Once.: Oken, M.M., Creech, R.H., Tormey, D.C., Horton, J., Davis, T.E., McFadden, E.T., Carbone, P.P.: Toxicity And Response Criteria Of The Eastern Cooperative Oncology Group. Am J Clin Oncol 5: , Another valuable way to integrate the nursing report/plan with clinical research is to follow up on this enunciation;

18 If the 12 definitions of function areas in DIPS could have been sorted out individually, it would have been easier to use the info in clinical research. You could e.g. look at elimination which is task number 5 and get only information concerning this matter thru the patient trajectory (Nurse 2) There have been a limited use of work-sheets in clinical research the study nurse has made retrospective approaches to gather information concerning study patient. It seems like it has been some kind of communication failure between us, the study nurse and the physicians. We have e.g. had no knowledge about cancer prostate patients receiving chemotherapy before getting questions from the study nurse about gathering source-data. There has in forehand been no information about research activity, and we have done procedures according to the roles of ordinary treatment, which means no work-sheets and no prolonged reports (Nurse 3). The study nurse has commented on this issue after being asked to deepen on the statement. This particular study have no specific demands towards the nurses at the Outpatient clinic, the study nurse and the physicians deal with the recommendations of the protocol at the patient visit. Another important issue to this is the fact that one of the surgery departments is responsible for this study, and thereby the collecting of data. The treatment at the Oncology department is a smaller part of this study Transferring, copying data from the information sources to the CRF The study nurse at the ward is copying patient documentation from the genuine patient journal to the case report form (CRF). The journal is illustrated thru the EPR DIPS in (Figure IV) on the next page. This process is manually drifted, after gathering together all information sources available the process of transferring data begins. The transferring consists of reproducing information like blood pressure, weight, and temperature. The standardization of adverse events demands other transferring procedure where several inputs of information are answering to the adverse event standard. There are two ways of copying data. The historically mostly used method is to screen the EPR from the computer with the case report form open, filling in the requested information by hand. The other method require more time than usually available. Gathering information during the patient stay, doing the procedures her selves followed by doing the documentation in the EPR. This is a backwards

19 procedure requiring extended time use, which have to be operated by the study nurse. It is also only possible for treatment given locally. Patients given study specific treatment at the University hospital could theoretically be monitored during the treatment by the study nurse her selves. The study nurse try to look up all the patients during the visit, but mostly for letting them know that she is there and for answering questions related to the research. Otherwise, there are studies that demand surveillance thru the treatment period, and which also have complicated logistics attached to it. In those cases it is easy for the study nurse to plan her presence thru this trajectory. Figure IV: This picture illustrates the screening of the EPR, reports and nursing plan. As illustrated this patient which is a patient at the ward have a nursing plan as described in the interview. All patient have nursing plans as of what is informed. The formula to succeed is strictly coordination of every part involved. The case report form (CRF) has historically been handwritten, but electronic record systems have been more common over the last years. In case of electronic record systems, the sponsor or the research group behind the protocol provides the research site with a computer containing the appropriate software. The CRF is unique, this is secured thru the law of patient treatment and ethical guidelines thru Good Clinical Practice (GCP) 2 2 Good Clinical Practice is an international ethical and scientific quality standard for designing, recording and reporting trials that involve the participation of human subjects. Compliance with this standard provides public assurance that the rights, safety and well being of trial subjects are protected consistent with the principles that have their origin in the Declaration of Helsinki, and that the clinical trial data are credible to use patient data in research, there are guidelines both to protect the patient s anonymity and their security thru the protocol treatment. All patients participating in the protocol treatment has to sign a consent allowing the physician to use information from the patient trajectory thru the study. ( d

20 Information of interest is varying from protocol to protocol and between cancer diagnosis and whether the intention of treatment is curative or palliative. Most treatments are evaluated with CT-scan others with disease specific markers (tumor-markers) like pain, ECOG -status (These scales and criteria are used by doctors and researchers to assess how a patient's disease is progressing, assess how the disease affects the daily living abilities of the patient, and determine appropriate treatment and prognosis) or Prostate-specific antigen (PSA) 3 In addition to physicians and nurses documentation, x-ray, pathology, clinical chemistry, micro-biology and immunology are of interest. Other parts to mention is equipment like scanners and fax machines for transferring data, the source-data itself, applications like the CRF used to organize the information, the use of standards that facilitates interconnection and interoperation between networks and the people, nurses, secretaries, patients and physicians that create the information. The document trajectory implicates use of different standards witch are common for clinical research thru guidelines. These standards are not in common use outside the protocol treatment, but adverse events as (e.g. nausea) is connected, by name to this standard. To find the right grade of e.g. nausea the case report form asks questions that are to be filled out. How often during 24 hours, does the nausea require use of medication, the dose of medication, type of medication, weight los due to reduced food ingestion, etc. The questions are resulting in grading nausea as an adverse event due to treatment. Different study groups are using the same standards on a global level. In day to day work this information could e.g. come from different sources of documentation like related blood samples, the physician journal and the nursing plan. Discussion and analysis An infrastructure is shared by a larger community, and by its members in the sense that it is the one and the same object used by all of them (Hanseth and Monteiro. 1998). Infrastructure cannot be split into separate parts being used by different groups independently. The different systems covered by the EPR is used individually in the different work practices, nurses and physicians have different needs, and the study nurse has needs that partly depend on nurses, physicians, laboratory, pathology and radiology. This is transferable to information infrastructure the role of infrastructure is believed to be important as its enabling character point to what may be kept as a stable basis in an increasingly more complex and dynamic world. 3 Is a protein produced by the cells of the prostate gland. The PSA test measures the level of PSA in the blood. The doctor takes a blood sample, and the amount of PSA is measured in a laboratory. Because PSA is produced by the body and can be used to detect disease, it is sometimes called a biological marker or tumor marker.)

21 1.0. The quality of information is not given, but dependent on the use. The customers at McDonalds in the Leidner example accept standardization of themselves as customers and play their role defined by McDonalds. (Leidner p.25). A precondition, however, is that the customers accept standardization of themselves as customers and play their role defined by McDonalds (Leidner p.25). If the customers are parting from the predefined script wanting more flexible services than the company will allow is problematic, consequently when both efficiency and quality are major goals, quality becomes subordinated. The context of Leidners work is that the quality of information is not given, but dependent on the use. The physicians having great interest in research purposes sees the great potential of the EPR, but it is held back of ethical and organizational issues as restrictions due to the transferring of information, and lack of categorization. The term of quality is assessed different whether it is discussed on a contextual level between human and non-human actors or in relation to work practice. The quality of documentation in the EPR could be good for clinical purposes, but be lacking information for research purposes at the same time. The systems described in this article, the EPR and the CRF are standardized. The EPR is standardized to improve the quality of the documentation, efficiency, and research. The CRF is standardized to make data transmittable over country barriers, and easy integration with the EPR thru standardized descriptions of adverse events. When we describe the gap between these two systems, it seems to occur because of information failure between different professionals, problems coordinating the information for the study nurse having too few resources, and technical issues concerning the EPR which e.g. affects the multidisciplinary and boundary work Making information transportable We have focused on the utilization of primary health care data for secondary use purposes in research. In much of the literature on EPRs, this secondary utilization is only seen to depend on the question whether the IT connections are in place. It is then simply a matter of selecting which information to transport and to where. (Berg. and Goorman. 1999). Information should be conceptualized as always entangled with the context of its production. The disentangling of information from its production context is possible, but that entails work (Berg and Goorman p.51). Berg and Gorman propose a law of medical information. The further information has to be able to circulate (i.e. the more diverse contexts it has to be