Pragmatic approach towards understanding music therapy in post-operative pain management is twofold: meta-analysis of evidence-based research papers and randomized controlled experimental analysis. Databases like Article First, Medline, ScienceDirect, LexisNexis, PsychInfo can be used to identify scholarly research articles and those presenting empirical studies should be short-listed.
For randomized controlled study, 24 hour post-operative patients (n = 40) of both sexes (age 18-65) from a trauma unit should be divided in two groups: control and music (n = 20), preferably with equal number of male and females within each group. After taking consent letter and nursing referral forms, the research design should envisaged as pre-test and post-test data collection.
The control group will not have intervention whereas musical group will listen for 30 minutes pre-recorded sedative music without lyrics of 60-80 beats and with soft background of flute or guitar. Those subjects not interested in music or are medically advised to avoid such tests should be excluded.
The variables to be tested in pre- and post-test are vital signs (systolic and diastolic blood pressure and pulse rate) to be tabulated according to frequency and percentage, level of pain to be monitored on a 0-10 Numeric rating scale (NRS) and number of times pain/nausea medication is used.
The experimenter should take semi-structured interviews of the respondents to prepare a pain diary, describing the overall nature of pain and treatment, and 87-descriptor McGill Pain questionnaire to poll sensory, affective and evaluative experience of pain. NRS provides an ordinal data on relative intensity of pain, which can not be given absolute number.
Hence this is a qualitative parameter. Should there be necessity, similar qualitative/semi-quantitative instruments like self report, comprising verbal rating scale (VRS) and Visual analog scale (VAS), should be used. If patient is not able to understand word- pain or synonym or is not able to express in absolute scale, it is advised to avoid VRS or NRS and to switch to VAS as it depicts a range of expression of pain.
Here 100 mm line is drawn as limit from no to severe pain. The patient draws a line below to mark the extent of pain in a continuum. Wong-Baker FACES pain rating scale is used to provide pictures to depict the pain, mostly used in pediatrics but also useful for those prefer to express graphically.
Neuropathy pain scale is a questionnaire measuring in graded scale of different aspects of pain like intensity, sharpness, sensation, itchiness, ups-and downs or background, internal vs. external pain etc. and this should be used as quantitative measure. Behavioral checklist is also a quantitative instrument which the experimenter records.
How many times during the session the patient changes the facial and bodily expression (red face, enlarged eye, loss of eye contact, hyperventilation, whimpering, screaming, protesting, tongue licking etc). In this study another quantitative parameter used will be number of times pain-relieving medication is consumed.
The quantitative biophysical parameters must be continuously recorded during intervention. It is necessary to understand which statistical treatments for ordinal and numeric records are to be used for small sample size.
For qualitative/semi-quantitative indices, the trend should be numbered as grades of ordinal numbers, like low pain as (0), moderate (5) and severe as (10). Mean and SD values should not be calculated, and numbers can be subjected to Mantel-Haenszel test for trend and nonparametric Wilcoxon rank sum test (analogous to student’s t-test).
For the semi-quantitative variables, Wilcoxon rank sum can be obtained from Mann-Whitney U-test. This statistical treatment translate qualitative trends into quantitative scale to enumerate level of significance between pre- and post-tests.
Quasi- and non-experimental design may not be suitable for such study. For the numeral data, mean and SD of each variable should be calculated for pre-test and post-test groups.
The most appropriate test of significance between pre- and post-tests, and between the two groups would be univariate one-way ANOVA computed from mean and SD’s. In pair-wise analysis, pre-test parameters should not be significantly different between the two groups, whereas significant difference should be attained in music pre-and post-test parameters.
The gender-wise test of significance should also be carried out to assess any demographic variation. The limitation of sample size does not permit to make age-dependent/ethnic analogy.
There are limitations in the study, a) in correlation analysis of different dependent variables, the self report may not correlate with behavioral or the biophysical parameters and the three criteria are different, and b) medication may subdue the pain sensation and there is no control to negate this.