Section 1 of this assignment introduces the concept of
evidence-based practice in Mental Health Nursing. To do this a research
question will be formulated and then the search strategy will be developed using
the appropriate search terms, which will be evidenced and then analysed. It is
important for nurses to be aware of the research process to be able to identify
the best evidence to update their knowledge and improve practice. The NMC Code
(Nursing & Midwifery Council, 2015) repeatedly mentions using the ‘best’ or
most current evidence available to prevent harm to patients and to practice
effectively. This reinforces the idea that evidence-based practice is
particularly important in nursing to deliver the best treatment for patients.
The research question chosen ‘is family therapy more
effective than CBT in treating adolescents?’
To develop a good research question Patino and Ferreiro
(2016) suggest that it should be able to fit into a framework, for example
PICO. This framework is outlined and applied to the question in the table
Table 1: PICO Framework
with eating disorders
These words can then be searched in a database such as
CINAHLPlus in order to find relevant studies and articles that discuss the
identified question. Synonyms and truncations can also be used to widen the
search. The list of synonyms used to widen this search are listed in Table 2. I
decided to split adolescents and eating disorders to make the most of the
databases and research as any combination of the synonyms highlighted could be
used to title a study on the subject. The use of truncations both saves time in
the search process and avoids the risk of missing out a relevant search term as
the ‘*’ indicates that any possible ending to a word will be included in the
search results (Herrlich, 2018).
with Eating Disorders
people; young person; teenage*; adolescen*
Disorder; anorexi*; bulimi*; EDNOS
Therapy; Family Counselling
Cognitive Behavioural Therapy
good; improv*; better; better success; increased success
Table 2: Search terms
Table 3: Search results
or 2 or 3 or 4)
or 7 or 8 or 9)
or 19 or 20 or 21 or 22 or 23)
and 14 and 17 and 24)
Table 3 summarises the use of the Boolean search strategy of
‘AND’ and ‘OR’ where I inputted the search terms into CINAHLPlus. The use of
‘AND’ allowed the results to contain all inputted terms in order to condense
the results to follow the question asked whilst the use of ‘OR’ allowed for
alternative synonyms to broaden the search to studies that have the same core
themes but alternative wordings, (Herrlich, 2018).
A further improvement could be making use of the other
databases available as using only one can limit the search results. Another
relevant database is MedLine, using more than one database order allows us to a
larger selection of articles and studies, they may provide some that are not
available on CINAHL Plus and vice versa. Another improvement could have also
been to make use of the limitations on CINAHL Plus, for example limiting the
year that the studies were published to further condense the articles to the
most recent, although only 7 pieces of work were found on CINAHL Plus,
condensing the results to the most relevant in this way would be less time
consuming than reading every article and judging which are more likely to be
relevant and not disproved.
Section 2 of this assignment demonstrates the ability to
critically analyse research. This section of the essay will analyse the paper
“Autogenic training to reduce anxiety in nursing students: randomized controlled
trial” (Kanji et al, 2006). I will analyse this research paper using the
Critical Appraisal Skills Programme (CASP) Randomised Controlled Trials
Checklist (CASP, 2017).
Did the trial address a clearly focused issue?
The aim of the study was written clearly in the abstract, it
stated that the aim was to study the effect of autogenic training (AT) with the
purpose of reducing anxiety in nursing students aged 19-49. Upon further
reading into the abstract it was clear that the comparatives for this were the
‘attention control group’ who received laughter therapy (B) and the ‘time
control group’ (C) who received no therapy. The time frame for this study was
11 months, this included 2 months of therapy and 3 follow ups. To measure the
effectiveness of the treatments the participants had to take the State-Trait
Anxiety Inventory, the Maslach Burnout Inventory, blood pressure and pulse rate
which were completed before and after treatment and at the follow ups.
Was the assignment of patients to treatments randomised?
In order to randomise the sample into 3 groups each
participant had to take an envelope which either had A, B, or C inside
allocating them to a group to ensure that every participant had an equal chance
of being allocated to a group (random sampling). Randomising the sample is
important for external validity. According to Steckler and McLeroy (2008) RCTs
have a liability of low external validity meaning that they often can’t be
generalised to the wider population, due to this it is essential to minimise
factors that would affect external validity, primarily the sample by having a
large sample size that is randomised as efficiently as possible.
Were all of the patients who entered the trial properly accounted for at
All participants were accounted for at the study’s
conclusion. The authors stated the dropout rates of each and even going as far
to state the reasons behind these dropouts including medical issues and lack of
time (some did not provide reasons, this was also recorded). It is important to
account for all participants at the end of a study in order to comment on
dropout rates, the reasons for dropout, the effects that this may have on the
results. A significant dropout rate in a randomised control trial could lead to
inequalities in group sizes and further issues with generalisability (Bell et
Were patients, health workers, and study personnel ‘blind’ to treatment?
It appears that all participants and health workers and study
personnel due to the nature of each varied trials, for example some received no
treatment or contact time suggesting that they would know they are part of the
control group. Also it is not otherwise stated that anyone involved in the
study was blind to treatment. Whilst it would be difficult to hide this from
participants, it is thought that when participants are aware of what is being
studied they can change their behaviour or answers to what they think the
researchers are looking for (demand characteristics) (Robinson et al 2014). In
this case it may be that some participants over-exaggerated the improvement in
their anxiety when taking the tests leading the results to seem that AT is more
effective than it is in reality.
Were the groups similar at the start of the trial?
It seems that the groups were rather unequal at the start of
the trial. When the groups were divided into gender and age it appears that
there were no under 20 year olds in the control group and no one aged 30-39 in
the AT group. However the groups were similar in terms of their qualifications,
only 1 University College was used in the study suggesting that everyone would
have been on a similar course with many also having similar backgrounds as
might come from the areas close to their place of study, however background of
the participants was not effectively recorded so it cannot be determined if the
groups really were similar. Akobeng (2005) states that the groups should be
similar to eachother and to the population that is being studied to rule out
any other factor that may affect the outcome of the study for example culture.
Aside from experimental intervention, were the groups treated equally?
The groups do appear to be treated equally, the trial is the
same length for all participants and they are all tested at the same intervals
for the follow up testing. It is essential that all participants are treated
equally and that the independent variable (the experimental intervention) is
the only thing that changes in the study (Viera and Bangdiwala 2007) as the
results of the dependent variable (the level of anxiety in the student nurses)
can be affected by a number of extraneous variables such as personal
differences or how well the variables are controlled within the study.
How large was the treatment effect?
The effect of the treatment appears to be significant. Both
the AT group and laughing therapy showed a large reduction in anxiety tested
through the State-Trait inventory immediately after finishing treatment whilst
the AT group also showed a reduction in blood pressure. Kanji et al (2006)
conclude that AT provides short term benefits, although long term benefits seem
to be undetermined as many participants stopped using the techniques. It was found that those who received AT were
more likely to retain some of the techniques they were given so that they could
use their training in the future, meaning that there could be some continuing
benefits after the training is complete. With this being said, the study
appears to have a too small time frame to measure the effect AT has on future
How precise was the estimate of the treatment effect?
To measure the precision of the results Kanji et al (2006)
uses levels of marginal significance (P values) and standard deviation to
present the results from the study. Confidence limits do not seem to have been
set in this experiment however the level of significance was set at P<0.05. When the P value is less than the level of significance, the researcher rejects the null hypothesis (Nieswiadomy, 2011). Generally the conclusions appear to be positive however when calculating the different P values it appears to be more complicated and less precise. Certain areas of the study appeared to show a significant difference between groups A and B for example on the State Anxiety test the significance was P<0.005 but Trait Anxiety which was P<0.084 meaning that for Trait Anxiety the null hypothesis should be accepted, but rejected in the case of State Anxiety. 9. Can the results be applied in your context (or to the local population)? The results of this study are difficult to apply to the wider population for a number of reasons. The study only used nursing students from 1 University College, suggesting that most of the participants would be similar in terms of where they are from, background etc. The age range of 19-49 is a positive as this represents the fact that there are a number of mature students who are training to become nurses however not all age groups were accurately represented in all 3 groups. The sample size was also relatively small, starting with 93 but by the end of the study 22 had dropped out leaving only 71 participants to draw conclusions from. A small sample size is undesirable (Faber and Fonseca 2014) as this limits the type of person in the trial and thus limiting its representation of the population and as a result its external validity (Polit and Beck 2012). 10. Were all clinically important outcomes considered? Other information that could have been considered is whether or not some of the participants had a diagnosable anxiety disorder or if they were experiencing stress. Kanji et al (2006) wrote that participants who were participating in other stress-reducing exercises were excluded from the study however this would not include any participants with anxiety that is not being actively treated. This may affect the outcomes of the study as Mind (2017) points out, whilst they are similar, stress and anxiety are still very different. In this case the AT may have had a more significant effect on those with anxiety over those with stress or vice versa and therefore this is a factor that would have been interesting to consider, either to exclude some participants or as an idea for future research. 11. Are the benefits worth the harm and costs? The benefits of the training appear to be long-term, some participants reported to be using the same techniques at the 8 and 11 month follow-ups suggesting that those who stop using the techniques could easily start again by themselves whenever they feel stressed. AT and other similar options appeared to give the participants the means to help themselves in the future. It is important that during studies ethical issues should be considered. Participants should be protected from harm as much as possible and when this is unavailable further efforts should be put in place to minimise this. Ingham-Broomfield (2017) also highlights the importance of informed consent which includes avoiding the need to deceive participants etc. Reflecting on the study it appears that the benefits do outweigh other factors. The only ethical consideration that was mentioned in the report was anonymity and all other ethical consideration such as psychological harm (e.g. embarrassment) do not seem to be applicable here. Kanji et al (2006) also specifically research the costs effectiveness and how it is relatively cheap, and therefore good to use as an intervention due to the possible long term benefits that AT can have. 12.