Planning your research and audits

Article Outline

• Summary
• Continuing professional education
• Introduction
  • Literature review
  • Aims and objectives
  • Data collection
  • Data collection tools
  • Sample size
  • Ethical considerations
  • Timescales
  • Data analysis
  • Dissemination of results
  • Funds
• Conclusion
• References
• Copyright

Summary 

Research and audit projects are a common feature of today’s health care in the quest for evidence based practice. There may be a project you, or your team, want to undertake. You might be undertaking a course, be it first degree or at a higher level in which completion of a project is a requirement. This article outlines some of the considerations necessary to translate your ideas for a project into a written proposal. Written proposals are necessary for a variety of reasons including approval to undertake the project from managers, for access to data, ethics committee approval, or those who might be providing grants towards the cost of the project.

Back to Article Outline

Continuing professional education 

As orthopaedic nurses and healthcare professionals we provide care to individuals with neuromuscular and skeletal disorders and injury that requires a specific body of knowledge to ensure high quality. Such knowledge is acquired, maintained and advanced through lifelong learning in practice and research. Articles included in this section aim to provide you with a sound knowledge base and help meet some of the professional, organisational and individual demands of being a safe and competent orthopaedic practitioner.

Examples of how this may be achieved and possible evidence for inclusion in your professional profile are given throughout the article. Other ways to demonstrate your continuing professional development (CPD) may be to:

Back to Article Outline

Introduction 

“Preparation, Preparation, Preparation” Three words of advice from a decorator friend of mine. We were talking about the quality of the finish I was aiming for. These words also hold true when you want to do some research. Taking the time to prepare and plan what it is you want to research, and how you are going to collect and analyse your data will pay dividends. This article will help you to prepare a clear plan which will be necessary to guide you through the tasks of data collection and data analysis. Importantly, you will need a plan if you need to convince colleagues, managers, Ethical Committees (Institutional Review Boards), or grant awarding bodies about the need for the research. There are several basic elements which make up a plan Box 1. This paper considers the research planning process from the identification of aims and objectives onwards.

Research plans, often called research proposals, are a bit like a good cookbook recipe. A recipe tells you the ingredients you need, and how much of each. It tells the order in which to put them together and details cooking methods and so on. With a good recipe you expect to get a product which looks like the one in the photograph in the book. As many a reader will know, recipes do not always have all the details you would like, and the results you get are based on a bit of good fortune if you are lucky – or you opt for the takeaway if you are not. The same can be said about a research plan. A good plan is one which another could take and know exactly what you intended and be able to repeat the process.

The research plan begins with a reason for undertaking the research. This may come from clinical experience, a chance observation in practice, or casual conversation at shift changeover. This is explored more by Justham (2006). Having an idea for research is usually followed through with a review of the current literature. If you are asking for a grant to be able to do the research, or if the research needs be approved from an ethical perspective, you will need to be able to demonstrate that there is a lack of knowledge about the topic you want to explore, and that the method you plan to use to collect data is appropriate.

Literature review 

Justham (2007) provides a framework for this, though most textbooks on doing research will offer guidance. The essential requirement of the review is to be able to demonstrate why the research you are proposing needs to be done. This can be achieved through a critical review of the previous research to demonstrate what has been achieved already and how this differs from what you propose. The review may highlight weaknesses and gaps in the existing literature. The review needs to leave the reader appreciating that there is a deficit in knowledge or understanding which your proposal seeks to address (in whole or in part). If you are simply repeating someone else’s study the rationale needs to justify why it is important to repeat the study – this may be due to the passage of time.

Aims and objectives 

These are fundamental to any research proposal. Having a clearly stated aim sets the agenda for the project. The aim will suggest the choice of method. It will also identify the topic for the research, and the population about which the research is concerned. For example, if the aim is to “test different approaches to cleaning and dressing pin sites to minimise the risk of infection” then a clinical trial would be required, whereas an aim about “exploring patient experiences of living with an external fixator following a fractured pelvis” suggests a more qualitative interview based approach. It may also be useful to provide more detail by developing specific objectives. Objectives in this sense help to interpret the aim. Thus, our intent to look at pin sites might generate specific objectives which could include “clean with betadine only if there is evidence of infection compared to cleaning every day with betadine”. Some studies require a hypothesis – a testable statement. This is a typical requirement of a clinical trial. Thus our hypothesis for the trial about pin site care might be “Pin sites left dry unless otherwise indicated have a lower infection rate than those dressed daily with an antiseptic agent”. Hypotheses tend to be precise. See Polit and Beck (2006) for a fuller discussion. The research then seeks to support or disprove the hypothesis.

Data collection 

The research plan cannot cover all eventualities. Nevertheless, a good plan will give a reasoned account of how you intend to collect data. You might be planning a study requiring interviews of patients. If so it would be helpful to explain how you propose to undertake the interviews, mentioning, where the interviews would take place, how the interview would be recorded, how long you anticipate interviews to last, how you propose to transcribe tape recordings (if collected). Evidence of how you would standardise the interview processes to eliminate concerns about interviewer bias might be needed. If the interviews involve “vulnerable” patients, would you allow a friend or relative to be present, and if so would they be allowed to contribute within the interview.

Data collection tools 

It is probable that your research may require you to develop a data collection tool, such as questionnaire, interview schedule or observation checklist. Your plan should include a discussion of the development of these, or if not fully developed, what steps would be taken to develop and pilot test the tools. Pilot tests are important to help answer concerns about reliability and validity. Reliability is to do with reproducibility and consistency. Thus, within an observational study, can different data collectors, observing the same event, make similar recordings of the event? Validity has something to say about the relevance of the data collection tool to the research question. Thus in a questionnaire to plaster technicians about casting techniques questions about hobbies such as fishing would be inappropriate. The quality of data collected via qualitative research methods needs to be assessed. Many researchers use the framework advocated by Lincoln and Guba (1985), which addresses trustworthiness of data though four dimensions of credibility, dependability, confirmability and transferability.

Sample size 

This is a concern in any research or audit proposal. In quantitative research, the simple rule that the larger the better should not be overlooked. Wherever possible the sample size should be determined “a priori”, that is before the study begins (Justham, 2005). But this is not always possible or appropriate. Clearly, there are constraints to any study because of time and resources (whether financial or personnel). In studies such as clinical trials, in which the data will be analysed using inferential statistical tests, it is usual to calculate the minimum sample size needed to ensure that possible errors in the sample are below pre-determined limits for Type 1 and 2 errors. For a fuller discussion of this type of sample size calculation see Polit and Beck (2003). Nevertheless, this should not become an excuse for limiting the sample size. Sample sizes are usually smaller in qualitative research but this should not be a reason for choosing a qualitative approach. Let us assume that in a qualitative research which is interview based, the plan requires thematic content analysis, the sample size might be reached only when no new categories of data emerge from the analysis – this might be after only 5 or 6 interviews, though it could be after 30 or more. For a fuller discussion of this approach see Glaser and Strauss’s (1967) Discovery of Grounded Theory.

Ethical considerations 

Research proposals are a requirement as part of any application to a Research Ethics Committee (sometimes known as Institution Review Board). The important consideration of such committees is the ethical implications that stem from undertaking the research. It is appropriate therefore that the research plan recognises the possible ethical issues which might arise. The stringency of ethical requirements may vary form country to country. The primary ethical issue for any plan is that it clearly describes what that plan is. It would be unethical to approve a plan which is unclear, contains uncertainties about possible actions, recruitments of subjects etc. Beyond this basic consideration, ethical committees must consider the risks to participants within any study. The two overriding principles are that the research should do the participants no harm, and that the research is aimed at benevolence (achieving good). A fuller description can be found in Evans and Evans (1996).

Timescales 

Whether your study is a routine workplace audit, a first degree dissertation or a doctoral thesis the plan should identify relevant timescales. When do you propose to start the study, how long will recruitment and data collection take, what timescale will be allowed for data analysis, discussion of the findings and preparing the final report. Many plans will include a time line, or Gantt chart, to illustrate the anticipated start and completion dates for various activities within the study. An agreed plan before the start of any study helps to maintain the focus within the study, and ensures that the project achieves completion within the stated timeframe.

Data analysis 

This is a part which is often addressed weakly within proposals, yet it is crucial to the outcome of any study. Thinking about the data that will be collected in relation to the aims of the study will naturally suggest how the data will be analysed. In studies such as clinical trials, in which comparisons are going to be made between the control group and the trial group to establish if differences had developed between the groups, then decision making (or inferential) statistical tests will be needed. But the plan would expect more than this to be stated. If it is intended to explore differences in nominal data then a Chi square test might be named, or if the difference was being explored between 2 sets of interval data a t-test would be undertaken, or if more than two sets then the ANOVA test is the test of choice. There are various resources to help with the understanding of statistics (never easy for nurses). Some resources other than texts books are Reusable Learning Objects. These can be web based. For example see www.nottingham.ac.uk/nursing/sonet/rlos/statistics for some examples.

When analysing qualitative data, the rigour should not be any less. The plan should outline the method to be used for analysing the data. There are various approaches to the analysis of narrative data. The default setting is usually thematic content analysis of which there are different frameworks, and Chapter 16 in Parahoo (2006) provides a useful introduction. However, qualitative data analysis includes techniques for analysing observational data, and other approaches to narrative data (Polit and Beck, 2003).

Dissemination of results 

Grant awarding bodies like to know how the results of a completed study will be disseminated. Who are you going to tell about what you found? Plans for dissemination of audit results may be very local, within ward or team meetings. In studies involving major funding support from charitable, governmental or industrial sponsors it would be politic to include plans for publication, mentioning specific journals, or conferences. Thus you might plan to prepare a paper or two for publication in the Journal of Orthopaedic Nursing, or seek to tell others at a conference on orthopaedic nursing. Do not worry about this if you have never done it before. Even seasoned authors had to start somewhere. The purpose of publication is at least twofold. Firstly, it is to make known the results of your study. Publication is an easier process than many presume. There is a lot of good work undertaken which fails to get accessed and read because potential authors are fearful of the publication process. Reputable journals have a peer review process which seeks to help authors by advising them, if necessary, how changes to a paper can improve the presentation and content. Secondly, publication aims to advise others who might also be undertaking similar studies what your findings are so that similarities or differences in results can be explored and explained.

Funds 

Grantsmanship is a wonderful term used to describe the skills needed to attract funds to finance a study. One colleague estimated that only one in ten research proposals submitted for funding were successful. However, even well described plans, do not always get funded. But the chances of getting the necessary monies are increased if as many of the “i”s are dotted and “t”s crossed as possible. A good plan, which not only outlines why the study is needed, and how you propose to do it, but also realistically identifies the costs associated with the various activities is more likely to be granted funds than one which gives a broad estimate for the overall project. Even seeking the time from your manager to undertake the study needs to recognise the costs to the manger’s budget of releasing you to do the study. Such costs do not necessarily stop at your time, but the costs involved in buying someone in to replace you. When thinking about the costs, these might be divided into capital costs, and revenue costs – which may include salary costs, costs for disposables like paper, printer cartridges, photocopying, but also overheads. Overheads are sometimes less easy to cost, but include things like heating, lighting, contribution Human Resource support when advertising and recruiting for data collectors, etc. Academic institutions often advise grant applicants to add a fixed percentage for overheads. Such percentages will vary but may be as much as 50%. Rarely is it less than 10%. In health services research, grant awarding bodies may limit the amount, which can be applied for as overhead costs. Health service providers may absorb some research costs, seeing these as a necessary part of practice improvements and developments, or consequential to a course that a nurse may be undertaking.

Back to Article Outline

Conclusion 

This article has commented on a number of issues which combine to make a good proposal. Proposals are written for a target audience, be it a manger to enable a study to take place, an academic institution because you want to pursue a research degree, an ethics committee because you need permission to approach patients or staff to participate, or for grant awarding bodies, because you need money to undertake the study. Each target audience may hold different expectations of what they want to see in a proposal, or they may limit the amount of words you can use to state your case. Whatever constraints you may be placed under will be handled more easily if the time is taken initially to consider and write up all the aspects outlined. A well written plan can then be tailored as necessary to meet the needs of the various interested parties.

Back to Article Outline

References 

1. Evans D, Evans M. A Decent Proposal: Ethical Review of Clinical Research. Chichester: John Wiley; 1996;
   • View In Article
2. Glaser B, Strauss A. The Discovery of Grounded Theory: Strategies for Qulitative Research. New York: Aldine; 1967;
   • View In Article
3. Justham D. Measuring performance through audit projects in orthopaedics – how do you know your performance is up to standard?. Journal of Orthopaedic Nursing. 2005;9(1):43–47
   • View In Article
4. Justham D. What is research?. Journal of Orthopaedic Nursing. 2006;10(4):228–231
   • View In Article
5. Justham D. Finding and evaluating an article. Journal of Orthopaedic Nursing. 2007;11(1):60–65
   • View In Article
6. Lincoln YS, Guba EG. Naturalistic Inquiry. Newbury Park: Sage; 1985;
   • View In Article
7. Parahoo K. Nursing Research: Principles, Process and Issues. second ed.. Basingstoke: Palgrave Macmillan; 2006;
   • View In Article
8. Polit DF, Beck CT. Nursing Research: Principles and Methods. seventh ed.. Philadelpia: Lippincott Williams and Wilkins; 2003;
   • View In Article
9. Polit DF, Beck CT. Essentials of Nursing Research: Methods, Appraisal and Utilization. sixth ed.. Philadelphia: Lippincott Williams and Wilkins; 2006;
   • View In Article