Project design and planning

| Home | | Hospital pharmacy |

Chapter: Hospital pharmacy : Research and development

Collaborating with others is important at every stage in R&D, finding people who are interested in the topic you want to study and also who are experts in research methods.


Project design and planning

 

Who needs to be involved?

 

Collaborating with others is important at every stage in R&D, finding people who are interested in the topic you want to study and also who are experts in research methods. Talking with colleagues and with other healthcare profes-sionals can help to develop a broader perspective from initial ideas and allows input from those with a track record in R&D. Collaborators might include academics at a school of pharmacy or other health or social sciences discip-line. Advice can also be sought from an organisation’s local NHS R&D office that can also provide contact with a local forum for input from patients and the public. Depending on whether the idea is research or service development, a local clinical governance office can also be helpful.

 

As others become involved, it is useful to establish early in the planning of the project the extent of collaboration and the contribution they will provide; this is particularly true for academic colleagues in terms of expectations around co-authorship of published output.

 

What question am I trying to answer?

 

Designing a robust study is dependent on having a clear question to answer. The starting point may be from interest in a particular area of research; this will need to be channelled into a general, and then more specific, research question. The aim is to arrive at a single question that is specific and as unambiguous as possible. Then work to break down the question further and to develop objectives for the study is required. For example, a team of researchers aiming to reduce medication problems after discharge from hospital asked: ‘What is the impact of a pharmacist-facilitated hospital discharge programme?’ and defined their objectives as: ‘to characterise medication discrepancies at hospital discharge and test the effects of a pharmacist intervention on healthcare utilisation following discharge’.4 This example does not explicitly state the outcomes that will be used other than the broad term ‘healthcare utilisation’. Pharmacy practice research is now strengthening its use of outcomes and there are still areas where exist-ing research is particularly sparse, for example for clinical and economic outcomes.

 

Two issues are raised for pharmacy because many of its interventions are as part of care provided by a team: firstly, the need for multidisciplinary research to take this into account and, secondly, not choosing outcomes where it would be unlikely or even impossible for a pharmacy intervention alone to lead to improved health outcomes. The Medical Research Council has produced guidance for evaluating complex interventions that include conducting feasibility studies to define interventions and outcomes.5 This framework could be followed if the research asks complex questions com-paring pharmacist-led services with usual care.

 

It is perhaps helpful to consider an example. If the service were a hyper-tension clinic, a feasibility study would define the processes in care and facilitate power calculations to be undertaken to determine an appropriate sample size for a randomised controlled trial that would demonstrate any likely impact on the health outcome, in this case, blood pressure control. Process outcomes such as patient satisfaction and acceptability of the service by other healthcare professionals may provide positive outcomes but there may be no difference in blood pressure control. Economic evaluation may be appropriate in such a study to demonstrate that a change in service delivery is more efficient with no detriment to patient care.

 

If the question is more about why and how, then the outcomes will be more descriptive and qualitative methods may provide the answers. If the question is what, then a questionnaire may provide the answer. The challenge is to generate a question that is not too ambitious and can clearly be answered. Once the question is developed, before proceeding any further, there is a need to know if the work has been done before.

 

Has someone already answered my question?

 

Searching and reviewing relevant literature will prevent reinventing the wheel and is a key element in study design. Summarising what is known and not known from previous work will help give a better understanding of both the context and issues in the chosen topic. It is then possible to build upon existing published evidence and make a useful and relevant contribution. Depending on the strength of the researcher’s skills in con-structing an electronic literature search, assistance from the medicines information team or from a medical librarian can be key at this stage. As for clinical evidence, the principles of information mastery apply, drilling down through a hierarchy of sources beginning with systematic reviews and moving through to individual studies.6 A useful starting point might be the Cochrane Collaboration on effective professional practice. Learning resources available through the UK Medicines Information web-site (www.ukmi.nhs.uk) may help with your search strategy and provide examples of research in medicines information. It is important to remember that there are many published abstracts from pharmacy conferences, such as the British Pharmaceutical Conference, Health Services Research and Pharmacy Practice Research Conference and UK Clinical Pharmacy Association (UKCPA) conference. These can usually found at the relevant website and are a valuable source because not all of these make their way into the wider literature if they are not subse-quently written up as full papers. Local NHS R&D and clinical gover-nance offices may also hold local and national databases of R&D projects. An advanced Google search can also be useful. Accessing full copies of published papers may be possible through the organisation’s clinical library or medicines information team.

 

The literature review must go beyond being descriptive to being critical, showing an understanding of the strengths and limitations of the studies being appraised. Results and conclusions need to be assessed in the light of the research design used and the appropriateness of the methods to answer the question posed.

 

Finding out whether key authors in the relevant field have ongoing studies that are due to be published is another important step. Most researchers are happy to be contacted and there may even be the possibility for collaboration. Papers in peer-review journals almost always have address and e-mail contact details for the corresponding author; otherwise ‘googling’ by name and organ-isation is usually fruitful.

 

Which methods?

 

Depending on the question and the outcomes to be measured, an experimental or descriptive approach or a mixture of the two may be selected; quantitative or qualitative methods, or both, may be appropriate. Experimental studies are used to test the effects of an intervention and they aim to investigate causes and associations. Experimental research designs include ‘true’ experiments (randomised trials) and quasiexperimental studies. Their aim can be ‘proof of concept’ or ‘proof of generalisability’. Choosing the appropriate method/s is a balance between selecting the ideal design (highest validity) with the most practical (highest feasibility).

 

Both quantitative and qualitative methods are used in pharmacy practice research. In the former the data are numbers and measurements, and in the latter narrative descriptions and observations. Qualitative methods are valu-able in stand-alone studies and also have an important role in setting content for quantitative methods, including surveys. Survey studies are commonly used; they collect information in a standardised form from groups of people and are useful providing the questions have been framed. They usually employ questionnaires or highly structured interviews to obtain a small amount of data from a large population. Self-completed questionnaires are efficient in terms of researchers’ time and effort. Closed questions will provide more concise replies, but wording needs to be carefully developed and tested. Rating scales are often used to collect data in questionnaires, for example, strength of agreement or disagreement about statements. Open questions generate qual-itative data that can be subjected to a content analysis. Coding of content can be used to produce a quantitative profile of responses that can be supplemented by illustrative quotations; this is a time-consuming process requiring consid-erable skill. Semistructured interviews are a key tool in qualitative research and enable rich and in-depth information to be collected. Focus groups use a topic guide and facilitated discussion to cover the topics. The Delphi and nominal group techniques are used in specialist areas of practice to obtain consensus, but again can be time-consuming. To choose the best methodology or tool, other research in the topic can be examined as a guide; there may even be a validated tool that would be ideal for the planned research.

 

Consideration must be given to the study population, how many subjects are required and how they will be recruited. The inclusion and exclusion criteria will need to be established to help arrive at participants who will help answer the question posed. For a quantitative study a sample size or power calculation will be necessary, checking to see the study has sufficient partici-pants to give an answer that is not merely down to chance. If there is no statistical expertise in the team, the NHS R&D office should be able to help. It is also a good idea to consider how the data will be analysed and to obtain statistical advice at an early stage. This may also help design the data collec-tion tools. A decision on who will measure the outcomes will need to be made. Depending on the intervention, it may be more appropriate to engage an independent researcher such as a research nurse to administer questionnaires or take clinical measurements. The local R&D office or clinical research facility may be able to help.

 

It is a good idea at this stage to develop a study protocol outlining the project as this provides a focus for collaborators to provide feedback and it will be required for approval processes. It will also help to clarify whether or not they will be part of the project team. A more detailed protocol will need to be developed as plans progress. There is no set outline for a research protocol and funding bodies may specify their own template, but the following head-ings would usually be expected

 

·           title of the study

 

·           study team and contribution of individual members

 

·           summary of aims, objectives, methods and outcomes to be measured background: this section justifies the need to undertake the research based on the literature review and any pilot work done

 

o    aims and objectives

 

o    study design and methods:

 

o    setting where the research will take place

 

o    subjects/patients, including how they will be identified and recruited, and inclusion and exclusion criteria. The sample size must be justified and, for a quantitative study, say whether a sample size calculation has been used

 

o    methods of assessment or measurement: explain what data will be collected, data collection instruments to be used and why they were chosen. If equipment needs to be used, describe it, and how its measures will be used

 

o    outcome measures need to be stated for both quantitative and qualitative studies

 

o    interventions (if applicable): if the study involves an intervention, it should be described. If giving a treatment or investigation, the dose, timing, method of providing, administering and receiving the treatment should be detailed. All necessary safeguards and potential risks should be made clear, including the methods by which intervention will be monitored

 

·           data collection, management and analysis, including how data will be collected, method of data entry, plan of analysis, and any data analysis packages

 

·           ethical considerations

 

·           benefits of the research to patients and the NHS

 

·           costs, including staff salaries, equipment, running costs (for example stationery, telephone, postage, travel) and any overheads (such as costs for office space, lighting, heating)

 

·           timescales

 

·           references

 

·           appendices (draft or actual data collection forms, questionnaires, interview schedules)

 

·           which approvals are needed for my study?

 

The main aim of research governance is to ensure all NHS researchers comply with a framework of activities and minimum standards and to be able to demonstrate their research quality through defined audit trails. There are two main types of NHS approval for research studies: (1) research ethics; and (2) R&D management approval. If the study is going to involve patients or staff working within the NHS, advice must be sought on whether ethical and local R&D approval is needed. This hinges on whether the project is considered to be ‘research’, hence the importance of the earlier definitions. The local R&D office will be able to advise on this as well as on local procedures that may be required for audit or service evaluation, which are sometimes overseen by clinical effectiveness teams. The electronic Integrated Research Application System ethics form (www. myresearchproject.org.uk/) gives an idea of the issues for consideration. Projects which are not ‘research’ will probably require approval through clinical governance structures. Irrespective of whether the study is deemed ‘research’ or ‘audit’, Caldicott principles, patient consent and data protection should be considered, though the project design will determine if particular actions are required. Training programmes may be available through the local NHS R&D office or through partner academic institution.

 

What resources will I need?

 

Regardless of whether the study is self-funded or an application for funding is to be made, it is important to understand the resources that will be needed, including time, equipment, consumables, travel and dissemination costs. These need to be clear on all approval forms and the local NHS R&D office may be able to help with templates for estimating and costing resources. The National Institute for Health Research is a major funder of health research (www.rdinfo.org.uk; http//rdfunding.org.uk) and has several funding streams, some of which are commissioned on specific topics and some are responsive and intended to meet needs identified within the service. There are many other possible sources of funding, such as www. cso.scot.nhs.uk (in NHS Scotland) and many clinical specialist charities. Any bid needs to be targeted appropriately to the relevant body. Some pharmacy organisations have awards to support pilot or development work and small projects, and these are valuable to test out ideas if you are thinking of applying for funding for a larger study. It is important that the proposal has been well reviewed by all members of the project team before it is submitted for funding or approval.

 

Bringing it all together

 

An example of how a research programme was developed from the starting point of an audit of care is shown in Text box 16.1.

 

Box 16.1 Developing a research programme in pharmaceutical care in secondary prevention of stroke

 

Starting from audits of care in patients who had had a stroke, an application was made for a research grant to:

• explore beliefs and concerns about medicines

• identify the difficulties experienced taking medicines

• design documentation incorporating the evidence base and pharmaceutical needs as identified by patients and carers.

The application was successful and the study was conducted with 30 patients who were purposively sampled and had been discharged after stroke or transient ischaemic attack within the past 12 months. Semistructured interviews covered self-reported adherence with prescribed medication and patients were asked to complete the Beliefs about Medicines questionnaire. The medicine-related problems identified were then addressed. The draft document produced as a result was reviewed by patients, general practitioners, community pharmacists and the local stroke managed clinical network.

Using the results from the study a further grant application was made to conduct a randomised exploratory trial of a pharmacist-led home-based clinical medication review in people after stroke. The purpose of this study was to conduct an exploratory trial to define the intervention, outcome measures and sample size. Pharmacist-led home-based clinical medication review at 1 and 3 months after discharge (n = 20) is being compared with ‘usual care’ (n ¼ 20), and 6- month follow-up which ended in May 2010. Further grant application has been made to test the feasibility of recruitment from primary care and to include telephone interview.

The research is a collaboration between pharmacy, geriatric medicine, community health sciences, psychology and the Stroke Research Group.

 

How should I disseminate my findings?

 

Much greater emphasis is now placed on dissemination than used to be the case. Effective dissemination of results to those who need to take action is now widely recognised as an essential part of the research process. In the past there has been more emphasis on academic methods of dissemination so that infor-mation is publicly available and open to peer review. This has changed to incorporate dissemination to key stakeholders because ultimately R&D activ-ities are done with a view to change and improvement. This can only happen if the findings are acted upon. Research can be disseminated in many ways, including full original peer-reviewed journal publications, articles in profes-sional journals, oral and poster conference presentations, newsletters, reports, briefings and entries in research databases. A dissemination plan should be agreed by the project team at the outset of the study, as there may be differ-ences in opinion as to where research should be published. Guidance on preparing and presenting research is available from various pharmacy organ-isations. For example the UKCPA, UK Medicines Information and European Society of Clinical Pharmacy include guidance on preparation of abstracts and posters on their websites (http://www.ukcpa.org.uk; www.ukmi.nhs.uk; www.escpweb.org). Abstracts from posters and presentations at previous pharmacy conferences (for example, British Pharmaceutical Conference, Health Services Research and Pharmacy Practice Research Conference and UKCPA conference) can usually be found on their websites. It may be possible to present findings at both pharmacy conferences and at multidisciplinary specialist conferences such as Diabetes UK.

 

Publications are rewarding for researchers and raise not only their indi-vidual profile but also the profile of the profession in a multidisciplinary environment. Proactive dissemination to stakeholders is an important part of the overall strategy and requires creative thinking. A strategy might include targeted presentations to relevant boards or committees, written briefings sent to a targeted audience and multi-stakeholder workshops at which find-ings can be discussed and actions needed prioritised. Again collaboration is helpful, whether it is academic support for writing for peer-review journals and conferences, or review of briefings and articles by some of the intended audience.

 

Contact Us, Privacy Policy, Terms and Compliant, DMCA Policy and Compliant

TH 2019 - 2024 pharmacy180.com; Developed by Therithal info.