Research In Clinical Practise
Research In Clinical Practise
Introduction to Portfolio
The research articles to be critiqued relate to the author's area of practice; community nursing and in particular; the cost-effectiveness of community leg ulcer clinics. The author currently manages a leg ulcer clinic and an insight into the research underpinning their cost-effectiveness would be of benefit in her quest to deliver evidence-based practice in line with the principles underpinning clinical governance.
Management of venous leg ulcers had advanced considerably over the last decade. This is due to various factors from greater knowledge of the aetiology of leg ulceration to more recent developments, such as dedicated leg ulcer clinics (Dowsett, 2004). However, it is a concern that leg ulcer clinics are being justified by nurses and organizations based on uncontrolled studies and narrative accounts (Bosanquet et al, 1993; Fletcher, 1995; Stevens et al, 1997; Thorne, 1998; Audit Commission, 1999). It has been widely suggested that the establishment of community leg ulcer clinics could increase healing rates for leg ulcers and standardise care (Moffatt et al, 1992; Simon et al, 1996; Stevens et al, 1997). However, research surrounding their cost effectiveness is extremely limited.
As interest in healthcare research grows, the number of definitions and explorations of the topic has increased. Most definitions of research follow a similar theme with the search for knowledge being emphasised as a fundamental reason for undertaking research studies (Clifford, 1997). Cormack (2000) defined research as an attempt to increase the sum of what is known by the discovery of new facts or relationships through a process of systematic scientific enquiry - the research process.
Nursing has been striving to become a research based profession since the Briggs Report (1972) and the lack of apparent integration of research findings into nursing practice has been lamented consistently ever since (Walsh and Ford, 1989). As long ago as 1989 the Department of Health (DoH), in their Strategy for Nursing stated; 'All clinical practice should be founded on up-to-date information and research findings.
Practitioners should be encouraged to identify the needs and opportunities for research presented by their work. Research is fundamental to achieving evidence-based practice in nursing. Evidence-based health care aims to promote clinical and cost-effective care/treatment through the explicit, conscientious, and judicious use of the currently available best evidence from research to guide decisions (Sackett et al, 1996).
Recently, there has been an increasing emphasis on evidence-based practice. The NHS information strategy, the development of the NHS net and the National Electronic Library for Health all testify to the NHS commitment of bringing research evidence closer to clinical decision makers (Thompson et al 2001). Journal based initiatives are also available such as Evidence Based Nursing (Cullum et al. 1997) and Clinical Effectiveness in Nursing (Newell 1997). Furthermore, guidance can be found in publications such as NICE (National Institute of Clinical Excellence) guidelines.
Evidence-based nursing and clinical governance presuppose research, without it evidence will remain elusive and no credence will be given to clinicians who base their work on routine and tradition alone (Lawton et al, 2000).
Often, consumers mistakenly believe that if a research report was accepted for publication then the study must be sound. Unfortunately, this is not necessarily the case. Indeed most research will have limitations and weaknesses and, for this reason, no single study can provide unchallengeable answers to research questions (Polit and Hungler, 1997). Rees (1997) adds that we must remember that there is rarely such a thing as the perfect research project.
Research findings should be challenged before practitioners change their practise and this can be achieved through critiquing research articles. The purpose of critical evaluation is to ensure that if implementation of findings is proposed, patients will derive genuine benefits. Therefore, a critical review must identify the strengths and weaknesses within a piece of research and should be carried out in a systematic manner (Eachus, 2003). Polit and Hungler (1997) assert that a good critique objectively identifies areas of adequacy and inadequacy, virtues as well as faults. They go on to say that, a critique of research should reflect a thoughtful, objective and balanced consideration of the study's validity and significance. They concur with Eachus (2003) that the process of scanning an article should be thorough and systematic. Cormack (2000) concludes that by being aware of the form and function of the various parts of a research article, readers are in a position to evaluate the worth of the material presented. The form and function of the various parts of a research article can be evaluated by following a research critique framework, such as the Cormack framework (2000) (Appendix I), in order to minimise the potential for bias. This includes evaluation and comparison of research methods, sample selection processes, data collection methods and methods of data analysis, ethical considerations and future research and learning. Rees (1997) believes in critiquing it is important to do two things; describe the content under the headings within the critique framework and state how well the author has accomplished that particular element.
Literature Search.
The author undertook a literature search using electronic databases, during November 2004 (Appendix II) posing the question;
'Is treatment in a community leg ulcer clinic cost effective compared to treatment in the patient's home?'.
From this, only one relevant randomised controlled trial (RCT) was found;
* Morrell C, Walters S, Dixon S, Collins K, Brereton L, Peters J, Brooker C, (1998) Cost effectiveness of community leg ulcer clinics: randomised controlled trial. BMJ, 316(7143) 1487-91.
Three other studies answering the question posed were found to be suitable for the purpose of research appraisal:
* Bosanquet N, Franks P, Moffatt C, et al (1993) Community leg ulcer clinics: cost-effectiveness. Health Trends, 25(4)146-8. ISSN: 0017-9132.
* Moffatt CJ, Franks PJ, Oldroyd M, et al (1992) Community clinics for leg ulcers and impact on healing. BMJ, 305(5)1389-92.
* Simon DA, Freak L, Kinsella A, Walsh J, Lane C, Groarke L, McColum C, (1996) Community leg ulcer clinics: a comparative study in two health authorities. BMJ, 312: 1658-61.
A mini-review entitled; 'Community leg ulcer clinics vs. home visits: which is more effective?' was undertaken in 2002 by Thurlby and Griffiths. This review looked at three aspects of the management of leg ulcers: healing rates, re-occurrence rates and cost-effectiveness. Following a literature search, Thurlby and Griffiths, (2002) rejected several studies from their mini-review concluding that the control group within the study by Moffatt et al (1992) was historical, the study by Simon et al (1996) was non-random and Bosanquet et al (1993) compared community clinics to hospital care (Thurlby and Griffith, 2002). Thurlby and Griffith (2002) included only one study - Morrell et al, 1998.
As it is a RCT, the first of the two articles selected by the author is;
Morrell C, Walters S, Dixon S, Collins K, Brereton L, Peters J, Brooker C, (1998) Cost effectiveness of community leg ulcer clinics: randomised controlled trial. BMJ, 316(7143) 1487-91.
The article is from the British Medical Journal (BMJ). The BMJ aims to publish rigorous, accessible and entertaining material that will help doctors and medical students in their daily practice, lifelong learning and career development. In addition, it seeks to be at the forefront of the international debate on health. To achieve these aims, the BMJ publishes original scientific studies, review and educational articles, and papers commenting on the clinical, scientific, social, political, and economic factors affecting health. (BMJ, 2004).
From here on in, this article will be referred to as 'Morrell et al' and is exhibited as 'Appendix III'.
Thurlby and Griffith (2002) chose not to include the article by Simon et al (1996) within their mini-review; however, in order to compare a non-random study with a RCT, the author has also chosen to critique the article:
Simon DA, Freak L, Kinsella A, Walsh J, Lane C, Groarke L, McColum C, (1996) Community leg ulcer clinics: a comparative study in two health authorities. BMJ, 312: 1658-61.
From here on in, this article will be referred to as 'Simon et al' and is exhibited as 'Appendix IV'.
Both articles have been cited on numerous occasions by other authors and appear to be well substantiated. Utilising the Cormack framework (2000), the author will critique the above articles and in doing so will attempt to establish the quality of the research underpinning the cost-effectiveness of community leg ulcer clinics and whether they are, indeed, based on uncontrolled studies and narrative accounts, as suggested by Bosanquet et al (1993); Fletcher (1995); Stevens et al (1997); Thorne (1998); and the Audit Commission (1999).
Title
The Cormack framework (2000) states that a title should be concise and informative and indicate the content and research approach used. Eachus (2003) agrees, stating a title should be concise, giving a good indication of the content of the report.
The title utilised by Morrell et al is concise and informative, stating the research approach used - a RCT. Likewise, the title utilised by Simon et al is also concise and informative and states that the research approach is a 'comparative study'. Both articles, therefore, fulfil the aims within Cormack's framework (2000).
According to Parahoo (1997), there is no real right or wrong title, only a misleading or confusing one and neither title is misleading or confusing.
Author(s)
Whether the authors have appropriate academic qualifications and appropriate professional qualifications and experience are questioned within the Cormack framework (2000). Eachus (2003) informs us that the authors of a research article may or may not be well known in the field and that the institute to which they are attached may give an indication as to the quality of the article.
The authors of the first article are; Jane Morrell (research fellow), Stephen Walters (statistician), Simon Dixon (lecturer), Karen Collins (research associate), Louise Brereton (research associate), Jean Peters (research fellow) and Charles Brooker (professor of nursing). The institutes to which the authors are attached are; the School of Health and Related Research, University of Sheffield; the School of Nursing and Midwifery, Samuel Fox House, Northern General Hospital, Sheffield; and the School of Nursing, University of Manchester. It is clear that the authors have appropriate academic qualifications and appropriate professional qualifications. The authors are well known in the field of leg ulcer management and the institutes to which they are attached indicate that the quality of the article is likely to be high.
The authors of the second article are; Deborah Simon (research nurse specialist), Louise Freak (research nurse specialist), Annette Kinsella (data manager), Julia Walsh (community research sister), Chris Lane (community research sister), Louise Groarke (research nurse specialist) and Charles McCollum (professor of surgery). The institutes to which the authors are attached are; the University Department of Surgery, University Hospital of South Manchester; Stockport District Health Authority and; Trafford District Health Authority. Simon et al have appropriate academic qualifications and professional qualifications; however, their qualifications in the field of research do not equal those of the authors of Morrell et al. However, the Simon et al are also well known in the field of leg ulcer management, albeit to a lesser extent. The institutes to which they are attached give an indication that the quality of the article may be poorer - as only one university is stated.
Overall, in comparison to Morrell et al, Simon et al seem less likely to attain good quality research.
Abstract
Cormack (2000) enquires as to whether the abstract identifies the research problem. Crookes and Davies (1989) believe that an abstract is a brief summary of the article and usually identifies what type of information is being presented. Eachus (2003) believes that an article's abstract should be a concise statement that gives the reader a clear idea of what the researcher(s) were trying to achieve, how they were trying to achieve it and whether it was achieved.
...
This is a preview of the whole essay
Overall, in comparison to Morrell et al, Simon et al seem less likely to attain good quality research.
Abstract
Cormack (2000) enquires as to whether the abstract identifies the research problem. Crookes and Davies (1989) believe that an abstract is a brief summary of the article and usually identifies what type of information is being presented. Eachus (2003) believes that an article's abstract should be a concise statement that gives the reader a clear idea of what the researcher(s) were trying to achieve, how they were trying to achieve it and whether it was achieved.
Morrell et all do not clearly identify the research problem within their abstract, this is given within the introduction section. The abstract gives a brief summary of the article and identifies what type of information is being presented - time to complete ulcer healing, patient health status, recurrence of the ulcers, satisfaction with care, use of services and personal costs. The abstract in this case is concise and gives the reader a clear idea of what Morrell et al were trying to achieve - to establish the relative cost-effectiveness of community leg ulcer clinics versus usual care provided by district nurses, how they were trying to achieve it - through the use of a RCT, and whether it was achieved - the abstract states the ulcers of patients in the clinic group tended to heal sooner than those in the control group.
Likewise, Simon et al do not clearly identify the research problem within their abstract. Again, this is given within the introduction section. The abstract also gives a brief summary of the article, identifies what type of information is being presented and is also concise, giving the reader a clear idea of what Simon et al were trying to achieve - to compare the outcome and cost of care for leg ulcers in community leg ulcer clinics in Stockport District Health Authority with Trafford district health authority as a control, how they were trying to achieve it - using detailed cost and efficacy studies, and whether it was achieved - the abstract states the healing of leg ulcers was improved whereas costs were reduced.
Both abstracts fulfil the aims within the Cormack framework (2000) and also contain 'key messages' regarding the research and its outcomes.
Introduction
The framework (Cormack, 2000) states an introduction should clearly identify the problem, give a rationale and state the study's limitations. An introduction needs to be just that, it should introduce the aims of the study, identify the problem or issue being researched and give a rationale for the study being carried out (Cormack, 2000). According to Hardey and Mulhall (1994), the introduction should explain why the research project is important, relevant and worthwhile.
The introduction given by Morrell et al clearly identifies the problem, giving a rationale - that an uncontrolled study provided the basis for the introduction of a new leg ulcer service. The introduction also explains why the research project is important, relevant and worthwhile - evaluating the cost effectiveness of the two interventions. However, it does not state the study's limitations.
In comparison, Simon et al in their introduction, also clearly identify the problem - that a previous approach to measure the cost effectiveness of care was both expensive and ineffective, giving a clear rationale for undertaking the research. The study's limitations are not made explicit to the reader and it is unclear to the reader how the research project is important, relevant and worthwhile.
The introductions vary in their quality and neither one wholly fulfils the aims within the Cormack framework (2000).
Literature Review
According to Cormack (2000), a literature review should be up-to-date, identify underlying theoretical frameworks, present a balanced evaluation of material both supporting and challenging the position being proposed and identify the need for the research proposed. A literature review is a critical review of previous literature relating to a research topic, the aim of which is to prepare the ground for new research. It provides the researcher and the reader with knowledge of the field being researched and conceptualises the research problem being considered (Cormack, 2000). A literature review examines and summarises articles within a chosen topic and is an essential stage in any research process. It attests to discover what has already been written about the chosen topic so that research will not be repeated unnecessarily (Hek, Judd and Moule, 2002). Cormack (2000) concludes that reviewing and evaluating the literature is central to the research process.
The article from Morrell et al does not contain a section entitled 'literature review', however, it does contain vast and varied references in an attempt to review previous literature relating to the topic, including the Charing Cross (Blair et al, 1998) and Riverside (Moffatt et al, 1992) studies, thus preparing the ground for new research. The references provide the reader with knowledge of the field being researched. The literature is not reviewed and evaluated systematically by Morrell et al, although this is central to the research process. Morrell et al have not examined nor summarised articles within the topic area as a separate component but they do attest to discover what has already been written, throughout by referring to past material. There seems to be no essential references omitted.
Likewise, the article from Simon et al does not contain a separate 'literature review' component and contains fewer references. This may be that in 1996 (when the study was undertaken), other literature on the subject was scarce, compared to 1998 (when Morrell et al undertook their study). Simon et al fail to prepare the ground for new research, provide the reader with knowledge of the field being researched or conceptualise the research problem being considered. However, such a lack of past material gives a good rationale for undertaking the study.
Neither article wholly fulfils the aims within the Cormack framework (2000) regarding a literature review.
Hypothesis
Cormack's framework (2000) suggests the hypothesis should be capable of testing, unambiguous, and state whether the study uses an experimental approach. A hypothesis can be defined as the prediction the researcher makes at the beginning of the study, which links an independent variable to a dependent variable (Rees, 1997). Hardey and Mulhall (1994) believe that devising concisely defined objectives and aims for a project can be a particular challenge, they are essential however, in providing the reader with a comprehensive and coherent statement of what the researcher hopes to achieve.
Morrell et al offer their hypothesis within the abstract under the subheading 'objectives' their hypothesis being - to establish the relative cost effectiveness of community leg ulcer clinics that use four-layer compression bandaging versus usual care provided by district nurses. This hypothesis appears to be capable of testing, unambiguous, and states whether the study uses an experimental approach, in this case a RCT.
Likewise, the hypothesis of Simon et al is also given within the abstract - to compare the outcome and cost of care for leg ulcers in community leg ulcer clinics in Stockport District Health Authority. Again, this hypothesis appears to be capable of testing and unambiguous, it, however, does not state whether 'detailed cost and efficacy studies' use an experimental approach.
Operational Definitions
Whether all terms used in the research question/problem are clearly identified is questioned within the Cormack framework (2000). Crookes and Davies (1998) define operational definitions as a description of how variables or concepts will be measured or manipulated in a study.
Terms used by Morrell et al in the research question/problem are unambiguous. Some operational definitions are identified and explained for example; measures of self-perceived health status - the 36-item short form health survey, the EuroQol, the McGill short form pain questionnaire and the Frenchay activities index - are referenced appropriately. However, within their statistical analysis section, 't tests', 'Mann-Whitney tests', 'X2 tests', the 'Kaplan-Meier method' and the 'log rank test' are all quoted within the text but are not referenced, nor explained to the reader.
Likewise, terms used by Simon et al in the research question/problem are unambiguous and unlike Morrell et al, all terms used within the study by Simon et al are clearly identified and unambiguous.
Methodology
Cormack (2000) suggests the methodology section should clearly state the research approach to be used and asks whether the method is appropriate to the research problem and whether the strengths and weaknesses of the chosen approach are stated. Methodology concerns questions about the manner in which we can gain knowledge about what exists (Cormack, 2000). According to Hardey and Mulhall (1994), the methodology section should include an overall description of the research design and details of the proposed methodology.
Morse and Field (1996) believe smart researchers are adept at both quantitative and qualitative methods and use the appropriate method at the appropriate time according to the research question, the goal of the research and other considerations. When selecting a research design, the methods used may be the personal preferences of the researcher. Quantitative research methods are often chosen in favour of qualitative methods due to their scientific nature, and surveys in favour of interviews due to their time and cost implications. However, Parahoo (1997) asserts, short cuts should not be made if the best knowledge is to be gained. Morse and Field (1996) state that each of the qualitative and quantitative paradigms has its own set of assumptions, established methodologies and set of experts.
In reviewing research designs applicable to nursing, it is essential to evaluate all research paradigms, qualitative as well as quantitative. There exists within nursing research, a hierarchy of evidence with quantitative research being more superior to qualitative. It is merely a convenient rule of thumb that, all other things being equal, RCTs are more able to attribute effects to causes. The RCT and, especially, systematic reviews of several of these trials are traditionally the gold standard for judging the benefits of treatments, mainly because it is conceptually easier to attribute any observed effect to the treatments being compared (Barton, 2000).
RCTs are the most common design in the field of experimental research that, as the name suggests, involves conducting a study in which the experimenters do not know the outcome. It involves comparing one group the intervention or treatment group, to a control or placebo group (Cormack, 2000). It is not surprising that high quality RCTs and high quality observational studies can sometimes produce similar answers. Quasi-experimental studies are not quite so powerful in that they lack the intensity of control which is an inherent feature of experiments.
Alongside the RCT within the paradigm of quantitative research are non-randomised studies. The role of non-randomised (observational) studies in evaluating treatment is conscientious; deliberate choice of the treatment for each person implies that observed outcomes may be caused by differences among people being given the two treatments, rather than the treatment alone (Barton, 2000). Observation techniques are frequently used in descriptive research studies (Cormack, 2000). As with other types of research, descriptive research begins with the identification of a problem or problematic situation. The description and analysis of that situation may reveal relevant factors or relationships hitherto undetected which, in turn, could form the basis of further research. Descriptive studies vary enormously in their scope and complexity. For example, a large sample of subjects drawn from a defined population may be studied. This is referred to as survey research (Cormack, 2000).
Qualitative research has a different aim in that it is designed to explore the meaning of specific situations by probing more deeply than standardised questions can do. Interviews are most frequently used to collect the information, allowing respondents to answer open rather than standardised questions. Qualitative data can also be obtained from observation but scientific rules have to be followed to provide a measure of validity.
Morrell et al, in reviewing their methodology under the subheading 'Patients and Methods' - clearly demonstrate that the design method is a RCT with 1 year of follow- up. As previously discussed, RCTs are the most common design of experimental research, which is the gold standard method of demonstrating, in a rigorously scientific manner, that a treatment or intervention is effective. In other words, it is the essential tool for a quantitative assessment of the efficacy of an intervention (Cormack, 2000). It seems from this statement that the chosen design method was appropriate to the research problem, however, Morrell et al do not state any strengths and weaknesses of their chosen approach. The quantitative research method of RCT was, perhaps, chosen in favour of qualitative methods due to its scientific nature.
The section also includes an overall description of the research design and details of the proposed methodology - recruitment, inclusion criteria and patient assessment and treatment are explained in detail. Morrell et al state that a random assignment schedule and serially number, sealed, opaque allocation envelopes were prepared in advance for each of the eight clinic sights, thus explaining how the trial was randomised.
A further quantitative research tool - questionnaire - was used to collect data on the use of health services related to care of leg ulcers. This questionnaire utilised four measures of self-perceived health status and were posted 12 weeks and 12 months after recruitment to collect data on health status and satisfaction.
Cormack (2000) deduces that questionnaires are the most common form of instrument in collecting data. However, a weakness of using the questionnaire method is the Hawthorne Effect. Cormack (2000) recognises that if people are aware that they are participating in a study, it is natural that they might alter their responses - the Hawthorne Effect, and the quantitative researcher needs to make every effort to minimise this. Morrell et al do not state how they minimised the Hawthorne Effect and even fail to present the questionnaire within their article. For the reader, reservations regarding the validity and reliability of results from this questionnaire must be made. The methodology section, overall, fulfils the aims set within the Cormack framework (2000).
In contrast, the design methods utilised by Simon et al are not so transparent. They are described as 'detailed cost and efficacy studies' and appear to fall into the classification of case-control study designs, within the sphere of descriptive research methodology. The aim of descriptive research is to discover new facts about a situation, people, activities or events or the frequency with which certain events occur (Cormack, 2000). A case-control study is one in which a group of cases is identified. A second group of controls is then selected. The two groups can then be compared with respect to a particular variable (Cormack, 2000).
It seems from the information that the method of 'case-control study', chosen by Simon et al, was appropriate to the research problem. Simon et al do not state the strengths and weaknesses of the approach, nor do they include an overall description of the research design or specific details of the proposed methodology. Again, quantitative methods were chosen in favour of qualitative methods.
The methodology section from the Simon et al study fails to fulfil the aims within the Cormack framework (2000) by its ambiguity. However, it contains fewer flaws and purports to represent a valid example of descriptive research.
Subjects and Sample Selection
Crookes and Davies (1998) define subjects as 'individuals participating in a study'. The Cormack framework (2000) asks whether the sample selection approach is congruent with the method to be used and whether the approach to sample selection and sample size is clearly stated.
Researchers have many sampling choices available to them that stem from theory or method or from simple practicalities such as time and money. A sample, therefore, should be chosen purposefully and many sampling strategies can be used. Samples vary considerably in the extent to which they represent the population, but the researcher who pays particular attention to the representativeness of the sample increases the possibility of generalizing the findings to a larger group.
Sample size can be affected by the researcher's time, budget, and geographical location. The researchers should determine the number of participants that are needed, as there is no single, correct sample size (Byrne, 2001). It is also important that the authors explain their sampling method and reasons for sample chosen so that readers can assess whether or not the findings can be useful in other settings (Parahoo, 1997).
The subjects within the Morrell et al study were 233 patients with venous leg ulcers allocated at random to intervention (120) or control (113) group and, as stated, the sampling selection took the form of randomisation. Random sampling techniques will help to ensure the representativeness of the sample in that each member of the population will have an equal chance of being included in the study (Cormack, 2000).
Cormack (2000) reminds us that, depending on the purpose of the RCT, subjects who are likely to be representative of those in whom the intervention is to be applied are selected as potential entrants to the study, as was the case within the Morrell et al study. In this case, the sample selection approach was congruent with the method to be used (RCT) and the approach to sample selection and sample size were clearly stated, as recommended by Cormack (2000).
In contrast, the subjects within the Simon et al study were 'all patients receiving treatment for an active leg ulcer irrespective of the profession or location of their carer'. The sampling selection - as it included all patients - was, therefore, non-random.
Some studies use non-random sampling techniques and, while they may produce important relevant findings, such findings cannot automatically be extrapolated to similar situation (Cormack, 2000).
In this case the sample selection approach was congruent with the method to be used and the approach to sample selection and sample size were clearly stated, as recommended by Cormack (2000).
Data Collection
Cormack (2000) suggests that data collection procedures are adequately described. There are many ways to collect research information; the choice of data collection method is influenced by the research approach used and the research question being addressed. The validity and reliability of the data collection method are very important and give the research findings their credibility (Hek, Judd and Moule, 2002).
Two data collection methods were utilised by Morrell et al. Firstly, nurses recorded the date of healing, secondary end points were health status, time to first recurrence of a healed ulcer, and the number of weeks patients were free of ulcers. Secondly, patients were sent questionnaires 12 weeks and 12 months after recruitment. Reminder questionnaires and telephone calls were used for non-responders or missing responses.
Morrell et al, as suggested by Cormack (2000) adequately described data collection procedures. The validity and reliability of the data collection method are, however, questionable. Again, an explanation is not given of how Morrell et al limited the Hawthorne Effect, although data collection methods included questionnaires.
In contrast, Simon et al utilised computerised databases, set up to manage information, as their preferred data collection method. The information managed concerned precise records of the care each patient received, including dressing materials, where these were obtained, nursing or other staff time, use of health authority transport, inpatient care and, pharmaceuticals used. Simon et al also describe, adequately, their data collection procedures that, by limiting the possibilities for bias, appear more valid and reliable than those utilised by Morrell et al.
Ethical Considerations
Regarding ethical considerations, Cormack (2000) suggests areas of importance include ethical committee approval, informed consent, confidentiality and anonymity. Ethical 'issues', according to Cormack (2000) are an integral part of all phases of the research process as well as to the use and application of research. Abbott and Sapsford (1998) state that it is general practice to promise respondents confidentiality - that is the information will never be used in such a way that respondents can be identified. There is great emphasis on documenting any risks, benefits, and safeguards on humans used for research.
Researchers should show any precautions taken for protecting human subjects and meeting ethical research standards. A researcher should also ensure that it is possible for a participant to withdraw at any time (Valente, 2003). It is also essential that all research participants give their voluntary, informed consent. Consent is vital as it respects the autonomy of individuals, their right to privacy and their right to choose (Tingle and Cribb, 2002).
As a result of the Nuremberg trials of nazi war crimes, the World Health Organisation (WHO) developed the 'Declaration of Helsinki' and all nursing research carried out in the UK should comply with the fundamental principles within this declaration. The 1975 revision of the declaration recommended codes of practice for researchers. This resulted in guidelines developed by national bodies such as the Royal College of Nursing (RCN) (Tingle and Cribb, 2003).
Morrell et al state that their study was 'approved by appropriate ethics committees'. Although the RCT is the gold standard for demonstrating efficacy, it is not always easy or even possible to organize such a study. Seeking the consent of a subject to enter a RCT does create difficulties. The patient will usually not be told which intervention he will receive and the health professional must admit ignorance over which treatment is 'best' for the subject. These factors go against the grain for most health professionals, however, they are essential components of a RCT (Cormack, 2000). In contrast, Simon et al do not make any reference whatsoever to any ethical considerations and neither study raises issues of informed consent or confidentiality and anonymity.
Results
Cormack (2000) enquires as to whether results are presented clearly and if enough detail is given to enable the reader to judge how reliable the findings are. Cormack (2000) states that data presentation - or to define it more explicitly, getting your results across and understood by your readers - is a critical element of any study. The method used to present data depends on a number of factors such as the type of data, the target audience and, the study design. According to Polit and Hungler (1997), the quality of the research is closely tied to the kinds of decisions the researcher makes in conceptualising, designing and executing the study and interpreting and communicating the results.
Morrell et al present their results systematically, utilizing a variety of charts and graphs;
* The baseline characteristics of patients recruited (Table 1).
* Mean scores on measures of health status at baseline by allocation to treatment in clinic or at home (Table 2).
* Healing time (Fig 2).
* The range of treatments used by district nurses per visit in control group (Table 3).
* Mean costs per patient for use of all NHS services by group (Table 4).
According to Cormack (2000), it is essential that findings be presented in such a way that they are clearly understood. The target audience (professional nurses) of the Morrell et al study are likely to be able to interpret the figures within each table and, thus, judge how reliable the findings are. The methods utilised by Morrell et al to present the data related effectively to the type of data, the target audience and the study design.
Likewise, Simon et al also present their results systematically, utilizing a variety of charts and graphs;
* Frequency of redressing leg ulcers in community clinics (Table 1).
* Healed leg ulcers at first visit / second visit (Fig 1).
* Expenditure on care of leg ulcers over 13 weeks (Table 2).
Again, the target audience of the Simon et al study are likely to be able to interpret the figures within each table and, thus, judge how reliable the findings are. The methods utilised by Simon et al to present the data related effectively to the type of data, the target audience and the study design.
Data Analysis
According to Cormack (2000) the approach should be appropriate to the type of data collected. Of the data collected, Cormack (2000) enquires as to whether statistical analysis is correctly performed, if there is sufficient analysis to determine whether significant differences are not attributable to variations in other relevant variables and if complete information is reported. Crookes and Davies (1998) define data analysis as conducted to reduce, organise and give meaning to data. Reid and Boore (1987) state that the term 'statistics' has two meanings - it can refer to the practice of collecting numerical data and also to the practice of making inferences from that data.
Often, the analysis in a RCT can be very simple. Using randomisation, the groups in the trial should only differ according to which intervention group they belong and, hence, there should be no need to adjust the analysis for extraneous factors (Cormack, 2000). The data analysis approach used by Morrell et al was appropriate to the type of data collected. However, statistical analysis was not correctly performed. Sixty-five patients within the study were excluded because they were unable to travel to the clinic, and did not undergo any formal assessment; these patients, however, still required home visits. Complete information, therefore, was not reported.
An important principle when designing a RCT is to standardise all aspects of care except the one under evaluation, and Morrell et al, due to the many variables, did not achieve this. Outcomes were very predictable. The control group, according to Cormack (2000), provides a standard set of results for the outcome by which the intervention group outcome can be judged. The selection of the control group, thus, is clearly critical in a RCT.
Morrell et al succeeded in reducing, organising and giving meaning to their data. However, it was unreasonable of them to conclude that a clinic-based service is more cost effective when 16.5% of those requiring treatment were excluded from the study owing to immobility. It would have been more valuable to transport these patients to the clinic and include those costs in the analysis.
Morrell et al have clearly attempted to apply trial methods to the delivery of care for chronic leg ulceration. Unfortunately, the trial design was such that useful conclusions are limited. The study - as a valid and reliable RCT - contains numerous flaws. Patients in the intervention group were treated in dedicated community clinics by 'clinic coordinators' who had undertaken special training and used specific bandaging techniques and materials. Treatment of ulcers was followed by a standard protocol of fitting stockings and surveillance. None of these aspects of care was available to the control group.
Likewise, Simon et al also reported encouraging data and succeeded in reducing, organising and giving meaning to their data. The approach utilised was appropriate to the type of data collected, however once again, statistical analysis was not correctly performed. The study by Simon et al contained several omissions that make interpretation of the findings difficult. The initial size of a leg ulcer has considerable influence on the time to healing. Simon et al should have indicated the size of the leg ulcers in each group at the onset of each study period and whether significant differences were present.
Simon et al stated that if healing did not progress the patient was immediately referred for vascular, surgical or dermatological care. They did not state how many patients were in this category, what there definition of failure of progression of healing was, and whether these more 'difficult' patients, referred elsewhere were included in the final analysis of healing rates. They did not make clear that, if patients with an ABPI of <0.5 were referred on, they were considered to have severe ischaemia and, therefore, excluded from the analysis.
Validity underpins the entire research process and refers to the degree to which an instrument measures what it purports to measure and reliability is the degree of consistency or accuracy with which an instrument measures the attribute investigated (Cormack, 2000). From the information gained, neither study can be proven valid or reliable.
Discussion & Conclusions
Cormack (2000) recommends that conclusions be supported by the results obtained and that implications of the study be identified. Valente (2003) tells us that when evaluating research, the conclusions and implications should summarise the major findings and conclude their application to practice, research, and theory and knowledge development.
Morrell et al discuss various components of the study under the subheading 'discussion'. Morrell et al make generalisations about the findings of their study and this component concludes the paper. Their conclusions are somewhat vague, however, in part, they are supported by the results obtained and the implications of the study are identified. Morrell et al succeed in summarising their major findings and conclude their application to practice, research, and theory and knowledge development.
Likewise, Simon et al discuss various components of their study, concluding the paper under the subheading 'discussion'. The conclusions of Simon et al are unambiguous, and are supported by the results obtained. The implications of the study are also identified. Simon et al, like Morrell et al, succeed in summarising their major findings and conclude their application to practice, research, and theory and knowledge development.
Recommendations
The Cormack framework (2000) asks whether the recommendations made by the researcher suggest further areas for research and whether the researcher identifies how any weaknesses in the study design could be avoided in future research. Implications for further research should be stated and cautions about generalising findings should be considered. Implications should describe how the findings can be applied to nursing practice, suggest further research studies and improve theory and knowledge (Valente, 2003).
Unfortunately, neither study contains any recommendations. Morrell et al and Simon et al fail to suggest further areas for research or identify how any weaknesses in the study design could be avoided in future research. Implications for further research are not stated.
Conclusion of Portfolio
In making conclusions, it is necessary to identify areas of adequacy and inadequacy, virtues as well as faults.
The title utilised by Morrell et al is concise and informative. The authors have appropriate academic qualifications, appropriate professional qualifications and are well known in the field of leg ulcer management. Morrell et all clearly identify the research problem and their abstract gives a brief summary of the article, identifying what type of information is being presented. The introduction given by Morrell et al clearly identifies the problem, gives a rationale and explains why the research project is important, relevant and worthwhile. However, Morrell et al do not present a literature review, although the study contains vast and varied references.
Morrell et al offer a hypothesis that appears to be capable of testing, unambiguous, and states whether the study uses an experimental approach. Terms used by Morrell et al in the research question/problem are unambiguous. However, some operational definitions are not referenced, nor explained to the reader. Morrell et al clearly demonstrate that the design method is a RCT with 1 year of follow- up. Their chosen design method was appropriate to the research problem, however, strengths and weaknesses of the approach were not stated.
Morrell et al state how the trial was randomised, however, they fail to state how they minimised the Hawthorne Effect or to present the questionnaire. Data collection procedures were adequately described, however, the validity and reliability of the data collection method are questionable. Morrell et al state that their study was 'approved by appropriate ethics committees'.
Morrell et al present their results utilizing a variety of charts and graphs and the target audience are likely to be able to interpret the figures and judge how reliable the findings are. The methods utilised to present the data related effectively to the type of data, the target audience and the study design. The data analysis approach used by Morrell et al was appropriate to the type of data collected. However, statistical analysis was not correctly performed and complete information was not reported.
Morrell et al, due to the many variables, did not standardise all aspects of care except the one under evaluation, however, they succeeded in reducing, organising and giving meaning to their data. It was unreasonable of them to conclude that a clinic-based service is more cost effective - the trial design was such that useful conclusions are limited.
Morrell et al make generalisations about the findings of their study. Their conclusions are vague, however, in part they are supported by the results. Implications of the study are identified. Morrell et al fail to suggest further areas for research or identify how any weaknesses in the study design could be avoided in future research. Implications for further research are not stated.
The title utilised by Simon et al is also concise and informative. The authors have appropriate academic qualifications and professional qualifications. Simon et al are also well known in the field of leg ulcer management. Simon et al identify the research problem within their abstract, which gives the reader a clear idea of what Simon et al were trying to achieve. Their introduction, also clearly identifies the problem, giving a clear rationale for undertaking the research. The study's limitations are not made explicit and it is unclear to the reader how the research project is important, relevant and worthwhile.
The study by Simon et al does not contain a literature review and contains fewer references. Simon et al fail to prepare the ground for new research, provide the reader with knowledge of the field being researched or conceptualise the research problem being considered. However, such a lack of past material gives a good rationale for undertaking the study.
Their hypothesis also appears to be capable of testing and unambiguous, it, however, does not state whether 'detailed cost and efficacy studies' use an experimental approach. Terms used by Simon et al in the research question/problem are unambiguous.
The design methods utilised by Simon et al are described as 'detailed cost and efficacy studies' and were appropriate to the research problem. Simon et al, however, do not state the strengths and weaknesses of the approach, nor do they include an overall description of the research design or specific details of the proposed methodology.
The study contains fewer flaws than the Morrell et al study and purports to represent a valid example of descriptive research. The subjects within the Simon et al study were 'all patients receiving treatment for an active leg ulcer irrespective of the profession or location of their carer' and, therefore, the sample selection was non-random. However, the sample selection approach was congruent with the method to be used and the approaches to sample selection and sample size were clearly stated.
Simon et al utilised computerised databases as their data collection method and describe their data collection procedures adequately. Simon et al do not make any reference whatsoever to any ethical considerations and the study raises issues of informed consent or confidentiality and anonymity.
Simon et al also present their results systematically, utilizing a variety of charts and graphs and the target audience are likely to be able to interpret the figures within each table. The methods utilised by Simon et al to present the data related effectively to the type of data, the target audience and the study design. Simon et al reported encouraging data and succeeded in reducing, organising and giving meaning to their data. The approach utilised was appropriate to the type of data collected, however once again, statistical analysis was not correctly performed.
The study by Simon et al contained several omissions that make interpretation of the findings difficult. The study cannot be proven valid or reliable. The conclusions of Simon et al are unambiguous, and are, in part, supported by the results obtained. The implications of the study are also identified. Simon et al succeed in summarising their major findings and conclude their application to practice, research, and theory and knowledge development. Simon et al fail to suggest further areas for research or identify how any weaknesses in the study design could be avoided in future research. Implications for further research are not stated.
The question posed - whether community clinics are more cost effective and improve leg ulcer care in the community - was not answered by either study. It might, for example, be more cost effective not to treat leg ulcer patients in community clinics, but simply to give all community nurses access to the best bandaging materials, train them properly, or both. In addition, the initial benefits noted in the short-term within the studies, may be counterbalanced by the long-term deterioration of the elderly population.
At the time of undertaking the studies critiqued, Moffatt et al (1992) had proved compression bandaging as the most effective form of treatment. The comparisons that perhaps should have been made by the authors, therefore, were between the different compression bandaging systems available and not between a group receiving compression bandaging and a control group being treated with a range of what were clearly, less effective interventions, as was the case within the Morrell et al study.
The articles differ in their research approach and methodology, though both studies utilise quantitative design methods. The studies similarly, fail in their quest to offer quality research in this particular field. The study by Morrell et al was neither valid nor reliable, as is usually the case with a RCT. However, the study by Simon et al contained fewer faults and, in the authors view, is representative of a satisfactory case-control study.
Implications for Practice.
Evidence from lower quality studies has generally supported the effectiveness of leg ulcer clinics, however, most suffer from the same failure to compare like with like. There remains the need for a trial to compare the effectiveness of treatment in leg ulcer clinics with home visits, where the same nurses (or similarly qualified nurses) provide the care in both settings and where all subjects have access to high compression bandaging and effective preventative measures. Any such study should consider costs from a wider perspective (Thurlby and Griffiths, 2002).
More recently, in 2004, Simon et al reviewed the evidence base of leg ulcer management. In this review, they cite their own work along with that of Morrell et al (1996) and others. The references cited are all somewhat outdated, ranging from 1988 with no further RCT of the cost effectiveness of leg ulcer clinics being carried out since Morrell et al in 1996. Furthermore, Cullum (2003) makes the point that so many of the studies conducted regarding leg ulcers are near worthless because of methodological flaws. Cullum (2003), however, provides some light at the end of the tunnel, not just by reciting all the aspects of leg ulcer management that desperately need good research, but actually by telling us how that research should be conducted if it is to be of value (see Appendix V).
It is difficult to comprehend that leg ulcer care has become a flagship for evidence-based practice as, for years, it has been based on a mixture of intuition, improvisation and tradition. Recent systematic reviews and the resulting clinical practice recommendations (NHS Centre for Reviews and Dissemination 1997, RCN Institute Centre for Evidence-Based Nursing University of York and School of Nursing Midwifery and Health Visiting University of Manchester 1998), have established a strong foundation for quality nursing care, to the extent that leg ulcer practice has become a touchstone of the quality of nursing care and is a primary focus of the ongoing National Sentinel Audit.
These recommendations represent an attempt to produce the clearest picture possible with a mosaic of individual pieces of research, each of which has limitations and sometimes, contradictory conclusions. Each study is the culmination of the efforts of a team of researchers and clinicians, representing years of work. The picture, however, is far from complete. The evidence available is of variable quality and it is not always clear how different pieces fit together. Some pieces remain unsubstantiated, or are based on the results of small numbers of patients, and so there is a need for further research to substantiate findings. Indeed, one of the fundamentals of evidence-based practice is that the evidence must be continually renewed so that it does not become 'out of date'.
Since, until fairly recently, the delivery of community leg ulcer care in patients' homes was the norm, the introduction of leg ulcer clinics provided an opportunity to evaluate their intervention. For many, this opportunity has now passed, but those considering establishing such clinics should note that the practice is not clearly supported by evidence and should, ideally, be subjected to evaluation that is more rigorous.
It should be remembered that good nursing practice is based on evidence and that the best evidence comes from high-quality research. It should also be remembered that not all research is high quality. Critical analysis and evaluation are vital. McSherry (2000) believes that the need to base practice on the best possible evidence is vital if nurses are to meet the challenges of providing quality care.
***************************************************
References
Abbot P, and Sapsford R, (1998) Research methods for nurses and the Caring Professions. 2nd Ed. Buckingham, Open University Press.
Audit Commission (1999) First Assessment: A review of District Nursing Services in England and Wales. Audit Commission Publications, London.
BMJ (2004)[Online] http://bmj.bmjjournals.com/cgi (Accessed Dec 2004).
Barton S, (2000) Which clinical studies provide the best evidence ?. British Medical Journal. 321, 255 - 256.
Blair SD, Wright DDI, Blackhouse CM, Riddle E, McCollum CN (1988) Sustained compression and healing of chronic venous ulcers. BMJ 297, 1159 - 1161.
Bosanquet N, Franks PJ, Moffatt CJ, Connoly M, Oldroyd MI, Brown P et al (1993) Community leg ulcer clinics: cost-effectiveness. Health Trends. 25, 146 - 148.
Briggs Report (1972) cited in Clifford C, (1997) Nursing and Healthcare Research. 2nd Ed. London, Prentice Hall.
Byrne M, (2001) Sampling for qualitative research. AORN Online, Vol. 73 (2) Feb 2001. [Online] http://gateway.uk.ovid.com/gw1/ovidweb.cgi (Accessed Dec 2004).
Clifford C, (1997) Nursing and Healthcare Research. 2nd Ed. London, Prentice Hall.
Cormack D, (2000) The Research Process in Nursing. 4th ed. Oxford, Blackwell Science.
Crookes PA, and Davies S, (1998) Research into Practise. Edinburgh, Balliere Tindall.
Cullum N, Fletcher A, Semylen A et al (1997) Compression therapy for venous leg ulcers. Quality Health Care. 6, 226 - 231.
Cullum N, (2004) The Nursing Management of Leg Ulcers in the Community: A Critical Review of the Research. Department of Nursing, University of Liverpool.
Cutliffe J, (1997) Qualitative research in nursing: a quest for quality. BJN 17:47-9.
Department of Health (1989) Strategy for Nursing, cited in Hardey M, and Mulhall A, (1994) Nursing Research. Theory and Practice. London, Chapman and Hall.
Dowsett C, (2004) Our agenda is to raise the standard of venous leg ulcer care. Journal of Wound Care. 13(5) 181-5.
Fletcher A, (1995) review of classic research: community clinics for leg ulcers. Critique III. Journal of Wound Care. 4, 471 - 472.
Hardey M, and Mulhall A, (1994) Nursing Research. Theory and Practice. London, Chapman and Hall.
Hek G, Judd M, and Moule P, (2002) Making Sense of research: An Introduction for Health and Social Care Practitioners. 2nd ed. London, Continuum.
Kenkre JE et al (1996) A randomised controlled trial of electromagnetic therapy in the primary care management of venous leg ulceration. Family Practice. 13, 3, 236-241.
Lawton S, Cantrell J, Harris J, (2000) District Nursing: Providing Care in a Supportive Context. Edinburgh, Churchill Livingstone.
McSherry RW, (2000) Putting research into practise. Nursing Times 97(23) 36-7.
Moffatt CJ, Franks PJ, Odroyd MI, Bosanquet N, Brown P, Greenhalgh RM et al (1992) Community clinics for leg ulcers and impact on healing. BMJ, 305, 1389 - 1392.
Morrell CJ, Walters SJ, Dixon S et al (1998)n Cost-effectiveness of community leg ulcer clinics: randomised control trial. BMJ, 316, 1487 - 1491.
Morse JM, and Field PA, (1996) Nursing Research. The Application of Qualitative approaches. 2nd Ed. London, Chapman Hall.
NHS Centre for Reviews and Dissemination (1997) Compression therapy for venous leg ulcers. Effective Health Care Bulletin. 3, 1-12.
Notter LE, and Holt JR, (1994) Essentials of Nursing Research. New York, Springer publishing Company.
Parahoo K, (1997) Nursing research: Principles, Process and Issues. Hampshire, Palgrave McMillan.
Polit DF, and Hugler BP, (1997) Essentials of Nursing research. Methods, Appraisal and Utilisation. 4th Ed. New York, Lipincott.
RCN Institute Centre for Evidence-Based Nursing University of York and School of Nursing Midwifery and Health Visiting University of Manchester (1998) The Management of Patients with Venous Leg Ulcers: Recommendations for Assessment,
Compression Therapy, Cleansing, Debridement, Dressing, Contact Sensitivity, Training/Education and Quality Assurance. London, RCN.
Reid N, and Boore JRP, (1987) Research Methods and Statistics in Healthcare. London, Edward Arnold.
Sackett DL, Rosenberg WMC, Gray JAM, (1996) Evidence based medicine: what it is
and what it isn't. BMJ 312: 71-2.
Simon DA, Freak L, Kinsella A, Walsh J, Lane C, Groarke L et al (1996) Community leg ulcer clinics: a comparative study in two health authorities. BMJ, 312, 1648 - 1651.
Simon et al (2004) Management of venous leg ulcers BMJ Online, Vol. 328 (7452) 2004. [Online] http://bmj.bmjjournals.com/cgi/content/full/328/7452/1358 (Accessed Jan 2005).
Stevens J, Franks PJ, Harrington M, (1997) A community/hospital leg ulcer service. Journal of Wound Care. 6, 2, 62 - 68.
Tingle J, and Cribb A, (2002) Nursing Law and Ethics. 2nd ed. Oxford, Blackwell Science.
Thorne E, (1998) Community leg ulcer clinics and the effectiveness of care. Journal of Wound Care. 7, 2, 94 - 99.
Thurlby K, and Griffiths P, (2002) Community leg ulcer clinics vs. home visits: which is more effective?. British Journal of Community Nursing. 7, 5, 260 - 264.
Valente S, (2003) Critical Analysis of Research Papers. Journal for Nurses in Staff Development. 19, 3, 130 - 142.