About the Author: Professor David CA Candy, Consultant Paediatric Gastro-Enterology and Nutrition, Royal West Sussex NHS Trust, Chichester, Visiting Senior Lecturer, Department of Child Health, University of Southampton and Visiting Professor, University of Chichester
The information in this article is correct at date of publication: September 2009
Opinions expressed by the author are not necessarily those of the publisher or editorial staff.
This is part two of a three part article. For the other articles in the series click here‘Doing a bit of research’ has always been seen as a way to improve the career prospects of Healthcare Professionals. This approach to research as a means to career advancement was always sneered upon by ‘dyed-in-wool’ academics. Their jobs depended upon obtaining research grants in open competition, and publishing in prestigious journals after having their work rigorously scrutinised by rival researches (a process described as peer review).
Having one’s name on a publication enabled the necessary ‘boxes to be ticked’ when undergoing appraisal, applying for jobs and salary increments. Since the NHS Constitution was published in January 2009, this has changed with commitment to research laid down on the first page (http://www.dh.gov.uk/en/Healthcare/NHSConstitution/index.htm). Research has become part of the ‘core business’ of all Trusts. The establishment of the National Institute for Health Research (http://www.nihr.ac.uk/) has meant that research within the NHS is conducted in a more co-ordinated fashion. Priorities are established (e.g. Stroke, Diabetes, Cancer, Medicines for Children), dedicated funding is available and all Trusts are able to contribute by recruiting patients. In this way, trials can be completed sooner and the large numbers of research subjects guarantees that clear-cut conclusions can be drawn from the results (that is, the trials are adequately powered).
It therefore pays to have an insight into how clinical research is conducted in the UK before you find yourself swept along by this tide of enthusiasm. Once you have hands-on experience of research you will begin to understand the problems and pitfalls underlying the production of the research literature. You may have already have an academic interest in some aspect of your job, and take a pride in keeping up with some aspect of the latest research. After all, many aspects of our job must have started off as someone’s research project. Before you know it, colleagues are consulting you as the local expert in your field. You may have ideas of your own about how practice could be improved. Or you may come about developing an interest in developing guidelines, when you will rapidly come to appreciate the shortage of first rate evidence to back up your guideline. To understand research, you just need to understand the common-sense principles that underpin all research. These enable you to interpret and question the research for practical application. Even if you don’t conduct your own research, a basic understanding of the jargon will help you keep up with the latest results and evaluate the research of others. You should be able to prevent others from blinding you with science. One of my personal drivers is the thought that, at the end of a clinic we all hope to have influenced the health of a handful of people in a positive way. By contributing to a research project (for example testing a new vaccine, or recruiting subjects for a trial) you can influence the health of thousands.
To download a PDF of the `Common terms used when discussing research`1 click here
Another factor is the addictive nature of research, which is a bit like prospecting for gold. You only have to come across a single ‘nugget’ and that will keep you going for years! It’s also good to know that patients treated as part a clinical trial will have a better outcome than those treated outside of a trial because of the greater attention to detail. This process of obtaining research data is strictly routine, but once gathered in and analysed and a picture starts to emerge of one treatment being better than another, that’s when the excitement begins.
What’s the evidence?
When compiling a guideline, or reading around a topic that interests you, the best place to start is with evidence based reviews. Prior to the acceptance of evidence-based practice, Healthcare Professionals could write review articles or guidelines on their favourite subjects based on their own opinions and personal practice. The resulting guidelines are referred to as GOBAT because they were written by ‘good-old-boys around a table’. A collection of good evidence-based reviews can be found in the Cochrane Database (www.cochrane.co.uk). The usual approach is to combine results from several comparable papers, which is more persuasive than relying the results of individual studies because increasing the number of results increases the power of the conclusion. This technique is called meta-analysis. The systematic review is another form, whereby studies are only included if they meet strict quality criteria. Systematic reviews are good for learning about a subject you want to research. Such articles will highlight gaps in the knowledge about a subject which can be the starting point for a research project.
Figures or feelings?
If you really can’t stand the idea of becoming embroiled in statistics this doesn’t preclude you from doing or evaluating research. Qualitative research is a very appropriate technique for evaluating the impact on our patients and clients, and the way we interact with each other. The technique involves semi-structured interviews, which are then transcribed and analysed, looking for common themes or threads running through the responses. As no numbers are generated, statistics are not required. The technique analyses the complexities of healthcare in a way that facts and figures never could.
Careful observations
Another type of clinical research not based on trials is called Observational. Observational studies follow up cohorts of subjects over years, for example to see what illnesses they develop. These studies may still be controlled, where groups of individuals, matched as closely as possible, are followed. Observational studies can test a hypothesis (e.g. do nurses who eat more fruit and vegetables have less cancer than nurses who consume less fruit and vegetables?) but can only establish an association, rather than cause and effect (the nurses who consumed more fruit and vegetables had less cancer, indicating an association between fruit and vegetable intake with reduced cancer, not that the diet caused less cancer).
Give me the facts
By contrast, much healthcare research is quantitative and does produce numeric results which require statistical analysis. Quantitative research is based on placebo-controlled trials, whereby participants are assigned randomly to receive active or dummy treatment. If subjects are selected randomly, that is, via a procedure in which all individuals in a study population have equal chance of selection, results may be generalisable to that population. If enough subjects are available to randomise, the groups should be comparable for age, sex, severity of illness and other variables, so that the only difference between groups is the treatment undergoing evaluation. This ensures that all subjects have an equal chance of being in either an experimental or control group (which has to be confirmed statistically). The researchers should record whether anyone was excluded from the trial, or dropped out, and why. There may be valid reasons for exclusions and drop-outs and it is important to document this. To avoid bias, the researcher should not know which patients are receiving treatment or placebo –and neither should the subjects.
It has been suggested that the double blind, placebo-controlled trial was Britain’s greatest contribution to Medicine. An early exponent of a randomized trial was a Scottish Naval Surgeon, James Lind, who wrote up his trial on the use of citrus fruit for seamen with scurvy in 1753 (he must have had difficulties producing placebo lemons). The double-blind placebo controlled clinical trial is often referred to as the ‘gold standard’ of clinical trials – it avoids bias being introduced, demonstrates a cause and effect and tests a hypothesis. It is still the subject of some controversy as to whether a treatment should be compared with a placebo, or with the current best treatment.
Is bigger always better?
If you are researching a miracle treatment (think of when penicillin before MRSA were invented) you only need a small trial to show a significant effect. With penicillin, the results were so spectacular that it was introduced without formal trials.
Generally speaking the larger the trial, the less likely a positive result is due to chance. This is one reason why research should go on in all hospitals and PCTs, so that large numbers of patients are recruited, producing reliable, applicable results in as short a time as possible.
In the next section, we’ll get started on the value of grounding in statistics and how to avoid using them if you possibly can. References
- Journal of Family Health Care, Volume 15, No 3, 2005 Special Supplement 1
