The title of paper 2 is “A Music Intervention to Reduce Anxiety Prior to Gastrointestinal Procedures”. It is a quantitative analysis of how a single 15 minutes of self selected music intervention affects anxiety. It is a randomised controlled trial of 198 participants and uses the State Trait Anxiety Inventory to measure the participant’s anxiety. The aim of the study was to evaluate whether a music intervention reduced patients anxiety prior to GI procedures. The research questions were Will listening to music for 15 mins prior to a GI procedure reduce patient’s anxiety? Will listening to music for 15 mins prior to GI procedure decrease patient’s blood pressure and pulse rates? The sample was made up of 193 men and 5 women with the average age of 61. They all attended the West coast (USA) Veterans Affairs Medical Centre. The results showed that patients who listened to music reduced there anxiety score from 36.7 to 32.3 while those patients who did not listen to music reduced their anxiety score from 36.1 to 34.6 these differences were significant at p= .007.
The title of paper 3 is “The effect of music interventions on anxiety in the patient after coronary artery bypass grafting”. The aim of the study was to examine the influence of music intervention during the early postoperative period on mood and anxiety of patients undergoing heart surgery. It used a prospective, repeated measures design i.e. the same group of participants were compared on separate occasions. It was a quasi experimental method using random assignment of participants. The independent variable was type of intervention Music, Music/ video or scheduled rest. The dependent variable was anxiety as measured by the STAI Mod measured by NRS and physiologic variables i.e. Heart rate and BP. The results were analysed using Chi-square and the ANOVA test of multi variance. The major finding of this study was that the use of music, music video or rest periods did not significantly reduce anxiety of patients after CABG.
The chosen papers use a Quantitative method, this is a term used for studies which are experiments using a large sample size and statistical analysis (Beyea, 1997). Quantitative researchers tend to produce and test hypothesis rather than look at problems and questions. They describe their work statistically in the representation of numbers; the researchers would argue that their technique is more scientific, reliable and open to checking by other researchers (Cormack, 2001). Randomised control trials were used in all papers; theses are experiments to measure and compare the outcomes of two or more clinical interventions (Parahoo, 1997). A Quasi experiment was also used in paper 3; this type of experiment resembles some of the characteristics of an experiment but not with the quality of a “true” experiment. Draw backs with this experiment are that there can be know direct manipulation, so the experimenter does not have complete control, also random allocation isn’t possible (Cormack, 2000).
Validity is a vital key to effective research, as research that is invalid is worthless (Cohen, 2000). One important aspect of validity is that if research is valid then it is also considered reliable. There are several types of validity that contribute to the overall validity of a study. The two main dimensions are Internal and External validity, Internal Validity is concerned with the degree of certainty that observed effects in an experiment are actually the result of the experimental treatment or condition (the cause), rather than intervening, extraneous or confounding variables. Internal validity is enhanced by increasing the control of these other variables (Abbot & Sapsford, 1997).
External Validity Is concerned with the degree to which research findings can be applied to the real world, beyond the controlled setting of the research. This is the issue of generalisability. Attempts to increase internal validity are likely to reduce external validity as the study is conducted in a manner that is increasingly unlike the real world (Abbot & Sapsford, 1997).
The internal validity of each of the papers can be questioned on a number of counts. The researchers of each paper have been caught between the high desire to make there research quantitative and scientific. Paper 1 used an independent measures design as the experimental group and control group were made up of different people. This leaves the study open to the criticism that it was individual differences and not the independent variable that led to differing outcomes for the Dependent Variable. Variables are measurable entity or characteristic; in experiments variables are manipulated to establish cause and effect (Polit, 1995). The design states in paper two the research used repeated measures design but proceeded to compare two distinct groups one experimental and one control it also states that individuals were randomly allocated, this was not a counter balance but an allocation of participant to groups, an experimental or control group, each participant only experienced the procedure once suggesting an independent measures design. This casts some doubt on the validity and reliability of the procedures reported. If the research used repeated measures design then order effects must be controlled for (Powers & Knapp, 1995). However it appears from the procedure that they used independent design and are open to the same critic as paper one i.e. that it is individual differences and not the Music which is responsible for any differences in the Dependent Variable. Paper three does use repeated measures design measuring the difference between pre and post anxiety, however this leaves the research with out a real control group to compare any effects against therefore the validity of any findings will be affected. It is not possible to really establish cause and effect without an adequate control group (Cormack, 2001).
All three papers have only a limited amount of controls. Paper one does not effectively control for factors which could affect the Dependent variable such as pain tolerance lifestyle factors etc. Although in support of the study pre treatment anxiety showed no significant differences between the two groups. Paper two used many standardised procedures and controls i.e. the 15 mins music in the experimental group was controlled with 15 mins quiet time, each participant completed the same tests at the same point in the experimental procedure. There was however no controls for use of anti hypertensions and cardiac medication. This would seem particularly important in this sample group. This would also seem important as the research is measuring blood pressure and pulse rate as part of the Dependent Variable. Another procedural problem is that only 15 minutes of music was allowed this could be too short a time for the music to have any affect. In all papers the music was self selected but does not give an opportunity to adequately investigate the difference made by different types of music. Paper 3 as well as having no control group there are very few controls for this paper other than no interruption for any group during the intervention interval. There were no controls for pain management, type or amount of medication or environment. This shows a very low level of internal validity as if they did find a difference in the dependent variable it can not be claimed that this is due to the independent variable.
All three of the studies fail to adequately control for demand characteristics. The participants are aware of the purpose of the study and as with any questionnaire measurement participants could just tell the researchers what they think they want to hear (Parahoo, 1997). Paper one and two standardise much of the procedure so direct comparison was more likely i.e. same measures same instructions random allocation (Brink & Wood, 1997). In paper one the same nurse carried out the treatment eliminating any individual differences in this part of the study The dependant variable in all three papers is Anxiety the researchers use STAI to measure it, this has established some validity and test re-test reliability (Trochim, 2000). It also controls for trait anxiety but fails to control for perceptual differences in how individuals perceive and cope with this type of invasive procedure. Two of the studies also incorporate some physical measures which are much less subjective measures of anxiety. They measure Pulse rate and Blood pressure although paper three does not use them as a measure of the dependent variable. This evidence could be used to further add validity to the STAI should the two measures show a high level of correlation (which they did not).
The external validity is affected by the internal validity, if a study is not able to establish truthful conclusions due to poor controls or problematic procedures it is not likely to be applied successfully into the real world or into practise (Locke, LF et al, 1998). The first limitation of all three papers in terms of external validity is that they are American papers and therefore may have limited generalisation to the practise in the UK. This may be due to Cultural differences in the perception of pain, appropriateness of music in this setting or effects of the music itself on individual experiences (Bejes, Marvel, 1992). The external validity of each of the papers is further damaged by the sampling technique. In each paper the sample is too small to generalise from, is not representative of the population and has been achieved by opportunity sampling (Parahoo, 1997). For these reasons each paper can be said to have low external validity. The sample in paper 1 is very small comprising of only 64 participants. The sample was gained via an opportunity sample from one centre in the mid west of America it is predominantly made up of American middle aged white women. It did have exclusion criteria but was an unrepresentative bias sample which has no possibility of being representative of the population that a Nurse working in the UK is likely to administer this treatment to. The sample in paper 2 is made up of 198 participants 193 male and 5 female. The average age is 61 and they are all taken from the Veterans affaires medical centre. As in paper one Non English speakers, People with Mental illness and Deaf clients were eliminated from the population of study. This is a small non representative sample and any conclusions based on this research alone will not be generalisable to English practice. It over represents older males and the sample is made up of individuals who have all shared a life altering event which is unparalleled in English history. Random allocation was used to select the control and experimental conditions but like paper one the sampling technique was opportunity sampling therefore has low external validity (Parahoo, 1997) Paper 3 also users an opportunity sample, it comprised of 96, however it over represents Males 65 to 31 females, all participants are white but excludes individuals who are under 19, do not speak English or are deaf. This again represents low external validity (Polit & Hungler, 1997).
All three studies were carried out in hospital wards or Units this represents a very close match to the situation in “real life”. This shows much higher external validity than if they were conducted in laboratories (Polgar & Thomas, 2000, Brink & Wood, 1998). The draw back of such naturalistic studies is that it becomes much harder to control for confounding and extraneous variables affecting the internal validity (Polit & Hungler, 1997).
The reliability of the STAI has been established using the test re-test method for trait anxiety. This shows that this test is reliably measuring these stable traits (Hays et al, 2003). This can not be said for state anxiety as this is much more difficult to show it obviously is affected by the severity of the situation and trait anxiety as well as perception of such a situation. This was used as a measure of anxiety in all three papers. The important measure of course in these studies was state anxiety as music is unlikely to affect stable traits. In paper 1 additional measure include NRS (Numerical rating scale), PS (patient satisfaction) and PFC (Perceived future compliance) have not been subjected to any reliability checks and are likely to be highly subjective and may not be stable over time. In paper two an additional measure was the researchers own creation the music enjoyment scale there was no reliability checks done on this measure and as this is the first time it has been used its reliability can not be demonstrated. In paper three additional measures were NRS again this has not been established as a reliable measure.
In conclusion paper 1 found mixed result it showed significant reduction of State Anxiety and NRS as shown by the ANOVA test of multi variance. The significance was set at .05 which is fine for such a low risk intervention. However there was no significant effect on Patient Satisfaction or crucially PFC. It is the PFC statistic which is the most important for practice as this is the determinant of regular check ups which is the most important aspect in the prevention of cancer in this case. The results of paper 2 found that using STAI the participants in the experimental group reported a significant reduction in anxiety and had reduced state anxiety significantly more than the control group. This was the case even after trait anxiety has been controlled for. However no significant differences were found in the vital signs (pulse rate and blood pressure). This said there is no evidence to suggest that blood pressure needed to be reduced; there were no controls for anti hypertensive or cardiac medication. It leaves this research with the same problem as paper one, it has shown a significant reduction of state anxiety but has not found any other evidence to corroborate and validate the claims made.
Overall looking at papers 1 and 2 and recognising they are both floored but have both found very similar results using completely different samples this adds validity to their claims. The non pharmacological nature and cheapness warrants at the very least more research in to this area. Paper three found no significant affects of music in reported anxiety NRS or in the STAI. This paper did not find any real benefits that would suggest the implementation of music therapy. However due to the cheapness and non invasive nature of the experiment I feel that it would be worth using music therapy in practice. These studies have also been cross referenced with many other studies which have similar conclusions this adds to the likely external validity of the findings and the level of external validity which would be accepted.
References
Barnason, S. Zimmerman, L. Nieveen, J. (1995) The Effects of Music interventions on anxiety in the patients after coronary artery bypass grafting. Heart & Lung. Vol.24 (2) pp124-32.
Bejes, C. Marvel, M. (1992) Offering colorectal cancer screening to appropriate patients. Family Practice Research Journal. Vol. 12 (1) pp83-90.
Beyea, S. (1997) Qualitative and Quantitative Approaches to nursing research: AORN journal. Vol. 66 (2).
Brink PJ & Wood MJ, (1998), Advanced design in nursing research, 2nd edition, London: SAGE Publications
Burns, N. Grove, SK. (1999) - Understanding nursing research, 2nd ed Philadelphia; London
Cohen L, et all (2000). Research Methods in Education, 5th Edition, London: Routledge
Chlan, L. et al (2000) Effects of a Single Music Therapy Intervention on Anxiety, Discomfort, Satisfaction and Compliance with Screening Guidelines in Outpatients Undergoing Flexible Sigmoidoscopy. Gastroenterology Nursing. Vol. 23 (4) pp148-156.
Cormack, D. (2001) The research Process In Nursing. 4th Ed. UK. Blackwell Science.
Dane, C. (1990) Research Methods, Brooks/Cole: California.
Evans,D. (2001) Music as an Intervention for Hospital Patients. A Systematic
Review, Evidence Based Nursing and Midwifery. Vol 5, (4).
Flemming K, 1998, Asking answerable questions, Evidence-based Nursing Vol 1 No 2 pp 36-37
Hayes, A. et al (2003) A Music Intervention to Reduce Anxiety Prior to Gastrointestinal Procedures. Gastroenterology Nursing. Vol. 26 (4) pp145-149
Ingersoll GL. (2000) Evidence-based nursing: what it is and what it isn’t. Nursing Outlook.; 48: 151–152
Locke, LF. Silverman, SJ. Spirduso, WW. (1998) Reading and Understanding Research, London. Sage.
McKenna, H, Cutcliffe, John, McKenna, P., 2000, Evidence-based practice: demolishing some myths. Nursing Standard. 14 (16): 39-42.
NHS Direct (2004) online 22/04/04
http://www.nhsdirect.nhs.uk/
NMC, 2002, Code of Professional Conduct, London: NMC
Parahoo, K. (1997), Nursing Research: Principles, Process and Issues, Macmillan, Basingstoke.
Polit, DF. Hungler, BP (1997) Nursing Research. Principles and methods. 5th Ed. Philadelphia: J.B. Lippencott.
Polgar s & Thomas SA, (2000) Introduction to research in the Health Sciences 4th edition, London: Churchill Livingstone
Powers, BA. Knapp, TR. (1995) A Dictionary of nursing theory and research. 2nd Ed. London. Sage Publications.
Sarantakos, S. (1993), Social Research, Basingstoke, Macmillan
Thomas B, 1999, Research and audit in effectice health services, Nursing Standard Vol 13 No 33 pp 40-42
Trochim WM, (2000), The Research Methods Knowledgebase 2nd edition, online 22/04/04
http://trochim.human.cornell.edu/kb/index.htm
Youngblut, J. Brooten, D. (2001) Evidence-based Nursing Practice: Why Is It Important? American Association of Critical-Care Nurses Vol 12(4) pp 468-476
Zinberg S. (1997) Evidence-based practice guidelines: a current perspective. Obstet Gynecol Survey.; 52: 265–266