South London and Maudsley NHS Foundation Trust (SLaM)

Transcription

1 Personal health record (PHR) case study South London and Maudsley NHS Foundation Trust (SLaM)

2 Acknowledgements This case study was informed by discussions with: > Michael Holland, chief clinical information officer, South London and Maudsley NHS Foundation Trust (SLaM) > Stephen Docherty, chief information officer, SLaM > Kumar Jacob, former non-executive director, SLaM It is also based on a review of published SLaM reports (see references). Many thanks to all for their help. Overview South London and Maudsley NHS Foundation Trust (SLaM) is a mental health trust providing the widest range of NHS mental health services in the UK. In 2010, SLaM developed an in-house personal health record (PHR), called myhealthlocker (MHL), to allow their patients to have more access to their health information 1. MHL version 1.0 allows users to access their SLaM care plans, complete surveys to keep track of how they are feeling, and access resources and tips on health and wellbeing. The PHR also provides information about medications, emergency contacts and opportunities to get involved in activities and research. The initial drivers for the development of MHL v1.0 were a shift in health and care policy for personalised records, the desire to empower patients and to put SLaM users right at the heart of their healthcare experience. SLaM also wanted to make sure that patients outside of the hospital environment were not disconnected from the trust and were more directly able to contribute to the care planning process. MHL v1.0 now has approximately 1,000 users, of which approximately log on at least once a week. In 2015, SLaM has been working on MHL version 2.0 which they hope will be more attuned to service user needs. Key recommendations > PHRs can help engage mental health service users who may otherwise be quite isolated. > Clinicians may be concerned that PHRs will increase their workload. This has not been shown to be the case, but new ways of working may need to be adopted. > It is a challenge when features of PHRs need to be changed when intermediary supplier organisations are involved. > Developers need to be skilled in user interface design and should include all end users (patients, carers and healthcare professionals) in the design of PHRs. > PHRs need to be sufficiently flexible to meet the needs of different users. > Important research can be conducted on PHR data but it is important that the primary focus of the PHR is related to patient care. Methods and lessons learnt Design and development Prior to 2010, SLaM s patient files had become electronic using a system called electronic Patient Journey System (epjs). A few patients and clinicians were recruited to be involved in the design of the first version of myhealthlocker (MHL v1.0). The interface was completely directed at patients, as clinicians were not expected to use it. The architecture of MHL v1.0 was based on Microsoft HealthVault. There were many issues with the design of MHL v1.0. The multiple drop-down menus made it difficult to navigate and it was also not optimised for viewing on mobile devices. The layout of care plans provided in MHL v1.0 was also problematic with important information, such as crisis planning, towards the end of the document and less useful information, such as demographics, at the top. There were also issues with patient-completed surveys, as the data visualisation was often hard to interpret, for example the Warwick-Edinburgh Mental Wellbeing Scale was displayed as a number of coloured points on a graph without any legend to explain what the data showed or how it should be interpreted. It is important that people involved in the development of PHRs are skilled in user interface design and involve all end users during the process. The research department in SLaM received some of the original funding for MHL v1.0 and wanted to focus on research opportunities which drove the development in a different direction. There is a risk of competing agendas where research opportunities become the focus rather than the main objective of improving patient care and self-management.

3 A key lesson learnt from MHL v1.0 was the importance of communicating with as many clinicians as possible and engaging them in the design of the system so that it fits with their working practices. It is essential when designing a PHR to be aware of how clinicians work. For example, some clinicians did not create care plans in the EPR system and so these were not auto-populated in MHL v1.0, which created confusion for patients. Furthermore, all stakeholders need to be involved in the design of the tool 2. Carers were not involved in the design of MHL v1.0, which meant their needs were not fully addressed. This was particularly problematic for specific conditions, such as dementia, where carers were often using the PHR on behalf of the service user. Carers are now being involved in the design of MHL v2.0 and have provided valuable contributions. The PHR has to do more than simply enable patients to view their record. A tool is required which patients can use to understand their condition and monitor it and that clinicians can use with the patient to help them with their care. PHRs need to be technically agnostic, so that the majority of patients can use them. For example, it will discourage people from using the tool if it is only available on a specific operating system or does not integrate with a wide variety of mobile applications (apps). SLaM is currently working on v2.0 of MHL. There was feeling in the trust that although the principles of v1.0 were good, there was a need for an updated version which better meets patient requirements. In February 2015, the trust developed a prototype of MHL v2.0 within a 10-week period. The executive board were happy with the prototype and SLaM is now working on the architecture required for an operational system. The design of the prototype involved 127 patients and the trust is now talking to different patient groups for further feedback. MHL v1.0 is continuing in parallel during this development and a phased approach based on usage and interest will be adopted. The governance model in place involves the trust medical director as the sponsor and agreement of features by the chief clinical information officer and chief information officer. An expert in user led design has been employed to find out what end users require. Clinician stakeholders involved in the new design included occupational therapists, clinical psychologists, psychiatrists, nurses and nursing assistants but not GPs. User recruitment and registration Once the PHR was live, it was initially offered to patients who had psychotic illness as the hardest to reach group. It was advertised to the patients and clinicians but the clinicians were not really involved in it being offered. There were also occupational therapist led drop-ins in the community and on the wards where patients were informed about MHL v1.0 and were able to sign up if they wished to. Patients are also able to sign up via the myhealthlocker website. Once a patient decides they would like an account, it takes an administrator minutes to sign the patient up and they are then sent a letter in the post with their login details. A two person team is responsible for the content management of the PHR. In order to log in to MHL v1.0 users needed to have a Microsoft Live account which had to be linked to Microsoft HealthVault, which in turn was linked to the SLaM record. This required users to have multiple sign in details, which was cumbersome. It was also difficult when things needed to be changed in the system as the trust had to go via the PHR supplier, Microsoft HealthVault. Patient perspectives Although it was not possible to interview any patients for this case study, Dr Holland provided the following information about how patients use MHL v1.0: > Most users currently only use one or two features of MHL v1.0 on a regular basis; the most popular use is to view care plans. > The patient-completed surveys are not particularly useful in isolation; it is much more meaningful to see trends which develop over time. > Virtual consultations between patients and their clinicians have been tested with individuals with attention deficit hyperactivity disorder (ADHD); however patient uptake is quite low as many patients like to see a doctor face-to-face. > Patients have found it helpful to have access to their crisis pathway and planning available in MHL v1.0. This provides them with easy access to this information to support them when they are unwell.

4 A study conducted by Robotham et al showed that patients with severe mental illness found MHL v1.0 beneficial in managing their health and wellbeing 3. Fifty-eight service users took part in the study, of which 32 used MHL v1.0 (defined as logging in at least twice and completing a patient-reported outcome measure). One hundred per cent of the service users who used MHL v1.0 said they found it useful and 84% said they would continue to use it. There were no apparent differences in usage across gender, diagnosis, and length of service use history. Mental health patients can often feel isolated, especially those who suffer from paranoid ideation. One SLaM service user liked the fact that MHL v1.0 helped to keep him engaged with other people (the following quotations are provided on publically available videos uploaded on YouTube): When I first heard of myhealthlocker it looked very interesting, it seemed like it was going to be very useful because of instant access, because of the way there is input from the professionals and it seemed very easy to use. With myhealthlocker I check my messages every day so even if it is something very simple that I do, it has a significant effect in terms of bringing me back in with other people. Patients have expressed that the PHR helps them selfmanage their condition and become more aware of the options available to them. The parent of a child service user said: The greatest benefit will be for him to be aware that he does have a disability, how his disability can affect him, knowing to look back on the days that he has recorded on how he is feeling as to where he is now. As he gets to use it and gain the access he will become more aware of what his choices are as he gets older and give him a greater confidence in supporting himself. Clinicians were initially worried that the use of the PHR would increase their workload. However, this was not found to be the case, but rather it involves a different way of working. For example, many clinicians in mental health use their notes to formulate their diagnoses (eg query schizophrenia); it may be quite distressing and cause unnecessary anxiety for patients if they had full access to these clinician formulations. Shared clinical records therefore need to be sensitively written and clinician work patterns have to change given this more dynamic way of working, ie building time into daily routines to deal with electronic/phr queries. Dr Martin Baggaley, SLaM medical director said: Rather than the clinicians holding onto the records, it is about the service user having the records, it s about being able to see them, it s about being able to understand what s written about them and also more importantly to actually write about themselves, it s about bringing their experience and sharing it with clinicians rather than vice-versa, and I think that is going to fundamentally alter the relationship between clinicians and service users. Currently, 20 GP practices have signed up to access MHL v1.0. Dr Adrian McLachlan, a GP based in Lambeth and chair of Lambeth clinical commissioning group felt MHL v1.0 provides a valuable supplement to existing electronic patient records (EPRs): Whilst I do not see this being intended to replace the electronic record in primary care or elsewhere, it provides a valuable supplement to it, in terms of access and potentially use in circumstances where that record is not available. I look forward to this being the normal and expected way of working, with patient access to a comprehensive record universally available, and in future this will probably be much more with mobile devices. Health and care professional perspectives One of the benefits of MHL v1.0 is that clinicians do not need to wait until the next appointment with the service user to be updated on their progress, for example they can see how a patient is dealing with their current medication prior to the next medication review meeting. This requires engaged patients to input data and allows a continuing dialogue between conventional scheduled meetings.

5 More engaged patients are likely to be more willing to take part in clinical research and MHL v1.0 allows patients to sign up to research studies. In future, systems such as MHL might allow service users to choose which parts of their EPR they wish to share on a research register. Dr Louise Johns, consultant clinical psychologist said: The service user research enterprise based at the institute of psychiatry is conducting a big study collecting feedback from service users on the system and evaluating that. It is very important to know what service users are thinking about the system and how they are finding it and also practical aspects, are they able to use it and make the most of it, have they got any reservations, anything that may get in the way of them using it. Future plans MHL v2.0 will include a clinician view which will give a summarised view of patients on their caseload who use the PHR. Clinicians will be able to see when their patients last logged on, and what information patients have added to the PHR. The trust sees MHL v2.0 as a platform which other things can plug into, for example health and wellbeing mobile apps. References Ennis L, Robotham D, Denis M, Pandit N, Newton D, Rose D and Wykes T. Collaborative development of an electronic Personal Health Record for people with severe and enduring mental health problems. BMC Psychiatry. 2014;14: Robotham D, Riches S, Perdue I, Callard F, Craig T, Rose D, Wykes T. Consenting for contact? Linking electronic health records to a research register within psychosis services, a mixed method study. BMC Health Services Research. 2015;15:199.

6 Health Informatics Unit Royal College of Physicians 11 St Andrews Place Regent s Park London NW1 4LE Tel: +44 (0) informatics@rcplondon.ac.uk