In spite of advances in pain management (Carr and Goudas, 1999; Donovan, 1990, Klopfenstein, Hermann, Mamlie et al., 2000, Long 2000; Svensson, Sjstrm and Haljmae, 2000), pain continues to be a major clinical problem with over 80% of patients experiencing moderate to severe pain after surgery (Acute Pain Management Guideline Panel, 1992). This can result in prolonged hospitalisation with its associated cost when efforts are not taken to deal with it properly (Heitz, Symreng and Scamman, 1992). In an attempt to improve and standardise care, several guidelines have been developed to deal with this menace (Svensson, Sjostrom and Halhamae 2000; National Health and Medical Research Council 2005; NHS, 2004); nonetheless, the problem still persists. In recent times, the right to pain management (IASP, 2004) is placing a lot of demands on health care professionals to find alternative ways of improving quality of life, while minimising analgesic side effects. As a registered general nurse who is working at a general surgical ward, I deem it as a responsibility to effectively manage this unpleasant symptom. Surgery is an undesirable experience for most patients and causes pain resulting from tissue damage (Taylor and Stanbury, 2009). The use of nonpharmacological techniques have been shown in several studies to be complementary to analgesics (McCaffery, 1990; McCaffery and Beebe, 1989). Specifically, the use of music has been shown to be very effective in reducing pain, anxiety and fear Mullooly, Levin and Feldman, 1988; Heitz, Symreng and Scamman 1992; Laurion and Fetzer 2003). Notwithstanding, little of these techniques is practised in Ghana due to the lack of knowledge and the heavy work load on nurses. In our surgical departments, there are limited opportunities for providing music as a non-pharmacological pain intervention before and after surgery. This sometimes affect patients both psychologically and physically. By virtue of these issues, I consider it very essential to contribute in the management of pain through pharmacologically and non-pharmacologically ways. Effective pain management has the potential of enhancing the quality of nursing care given and improving patient outcomes such as early ambulation, timely wound healing, reduced infections and enhancing comfort (Wasylak, Abott, English and Jeans, 1990; Watwill, 1989; Sydow, 1989). The aim of this essay is to develop a research question and to find out alternative methodology and methods that can be used in answering the question. I will commence with a review of the literature to find out the gaps in the available evidence. The research question will then be proposed, followed by an evaluation of the various designs and methods that can be used in answering the question. Lastly, the essay will end with my recommendations of the most suitable approaches to address the problem
Evidence-based nursing has been defined as the delivery of nursing care using an integration of research evidence, clinical expertise and patient's preferences in making clinical decisions (Mulhall, 1998). This definition is more acceptable as compared to the definition of evidence-based medicine by Sackett, Rosenberg, Gray et al. (2000) which ignores patient's preferences in decision making. In this section, the literature surrounding music as a postoperative pain management will be reviewed to assist in identifying the areas that need further clarification. Pain is the most common reason why people seek hospital care (McLean, Domeier, DeVore et al., 2004). According to the International Association for the Study of Pain (1979: 250), "pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage described in terms of such damage". Thus, McCaffery (1983: 14) asserts that, "it is whatever the experiencing person says it is, existing whenever she says it does". Apart from pain signifying some tissue damage, it is of no significance and leads to detrimental effects (Apfelbaum, Chen, Mehta et al., 2003). Pain disturbs emotional states, sleeping pattern, physical and social performance, and thus impairs the quality of life (Reyes-Gibby, 2002; Strassels, 2000). The importance of pain management cannot be overemphasised; in addition to reducing unwarranted suffering, it enhances function and improves the quality of life for patients (Goudas, 2001; Reyes-Gibby 2002; Rogers, 2000a; Rogers, 2000b; Strassels, 2000). Due to its significance, pain management has gained international recognition (Apfelbaum et al., 2003; Royal College of Surgeons and College of Anaesthetists, 1990; Svensson, Sjstrm and Haljmae, 2000) leading to the formulation of policies to improve and standardise care (NHMRC, 2005; NHS, 2004; Svensson et al., 2000). Postoperative pain constitutes a major clinical problem confronting health care professionals (Good, 1999), with over 80% of patients experiencing moderate to severe pain after surgery (Acute Pain Management Guideline Panel, 1992). This leads to enormous effects such as stress, infections, restricted mobility (Yeager, Glass, Neff, Brink-Johnsen 1987; Murray 1990), sleep deprivation, delayed wound healing, increased costs and patient dissatisfaction (Shang and Gan, 2003). The use of both pharmacologic and non-pharmacologic interventions have been found to be effective in relieving postoperative pain (McCaffery, 1990). Thus, the nurse has a significant role in providing a variety of non-pharmacologic strategies in addition to drug treatments (McCaffery, 1990; McCaffery and Beebe, 1989). For many decades, non-pharmacological interventions have been found to be valuable, simple, and inexpensive adjuvants to pharmacological pain interventions (Hyman, Feldman, Harris, Levin and Malloy, 1989). Nevertheless, some of these strategies lack efficiency and for others, the analgesic effect is at its best, moderate (Goudas, 2001). Listening to music has been found to reduce pain intensity and/ or analgesic requirements (Jacox, Carr, Payne et al., 1994) during brief interventions (as either a supplement or an alternative for drugs). Music therapy offers potential benefits of being inexpensive, easy to provide and safe to use. Nevertheless, the efficiency of music to reduce pain intensity or analgesic requirements has not yet been established (Cepeda, 1998; Good 1996; Koch 1998). The use of music as a postoperative pain management strategy has been used in several studies; notwithstanding, these studies have produced mixed results, with some showing improved pain relief (Mullooly, Levin and Feldman 1988; Heitz, Symreng and Scamman 1992; Laurion and Fetzer 2003; Nilsson, Rawal, Enqvist and Unosson 2003; Nilsson, Rawal and Unosson 2003; Good, Anderson, Ahn et al., 2005; Masuda, Miyamoto and Shimizu, 2005), while others have reported no improvement in pain (Heiser, Chiles, Fudge etal., 1997; Ikonomidou, Rehnstrm and Naesh, 2004). The main approaches used in these studies have mainly been experimental, however, most of them lack strict control with various outcome measures ranging from psychological (pain, anxiety), physical (sleep) to physiological parameters (heart rate, respiratiory rate, blood pressure). Moreover, some of the studies limited the music only to the operating theatre (Koch, Kain, Ayoub, and Rosenbaum, 1998), while others limited the music to just after the operation (Shertzer and Keck, 2001).
THE RESEARCH QUESTION
An initial and often important step in evidence-based nursing is the formulation of clearly-focused question. This facilitates the literature searching process and saves time (Snowball, 1997). The PICO framework represents patient population, intervention, comparison and outcome (Richardson, Wilson, Nashikawa, et al., 1995). An extended form of is PICOTT, which adds the type of question and the most suitable type of study (Schardt, Adams, Owens et al., 2007). Following a review of the literature, I have developed my research question as follow: Does listening to music before and after an abdominopelvic surgery reduce pain and analgesic consumption more than ordinary rest periods? Using the PICOTT framework (Schardt et al., 2007), my research question involves the following: Population: Postoperative patients after an abdominal surgery. Intervention: Music Comparison: Rest period Outcome: Postoperative pain and analgesic consumption. Type of question: Prognosis Type of design suitable: Randomised controlled trials (RCTs)
METHODOLOGY AND METHODS
Common among all research strategies is the systematic and organised pattern of answering a question or finding solutions to a problem (Polit and Beck, 2008). My research question can be answered through various approaches such as an empirical research, a systematic review of primary studies, and also through an evaluation research.
Clearly, the choice of a research design depends mainly on the research question and its associated aims, which ranges from assessing an intervention, investigating a process, engaging users and/ or transforming practice (Green and Thorogood, 2004). Others factors that might affect the research typology include the researcher's preferred paradigm, available resources, time, ethical and practical feasibility (Brewer and Hunter, 1989; Diers, 1979; Green and Thorogood, 2004). According to Green and Thorogood (2004), an experimental design is best for measuring the effectiveness of an intervention; a survey is suitable for establishing variable relationships; observational studies for discovering social units; a case study for in-depth subject analysis; or an action research for studying and changing practice. Notwithstanding, some studies can be classified under various typologies, for instance, an ethnographic study can be classified as a case study as well as an observational design [ibid]. In the conduct of a research, ethical principles such as respect for human rights (right to self determination, full disclosure), beneficence (doing good), non-maleficence (avoiding harm) and justice (fair treatment, right to privacy) should be carefully taken into consideration (Beauchamp and Childress, 1983). Moreover, informed consent (Homan, 1991), confidentiality (WMA, 2000) , adhering to cultural norms, professional code of conducts, ethics approvals, and national legal frameworks are very crucial before the commencement of a research study. Apart from the benefits of a research to the general population, the investigator also have a duty to be safe from risks such as physical abuse, psychological trauma, exposure to infectious diseases and accidental injury (Green and Thorogood, 2004). Empirical research can be categorised mainly into quantitative and qualitative paradigms. Quantitative studies, unlike qualitative designs, are more objective, deductive and generalisable. It is considered as the benchmark of science, and centres on hypothesis testing, reliability, validity, numbering and statistical inference (Robson, 2002). Its study approaches ranges from highly controlled experimental designs through to quasi-experimental studies, and then to less controlled nonexperimental (descriptive and correlational) designs (Polit and Beck, 2008). Considering the nature of my research question, a randomised controlled trial would be most suitable in determining the effectiveness of music as a pain management strategy. Postoperative study participants who meet the inclusion criteria would be randomly allocated into three groups; with one group listening to music through headphones from a compact disc, another group listening to a blank compact disc and the third group receiving nothing. My choice for dividing the groups into three is to eliminate as much as possible any confounding factors that might mimic the research outcomes. In addition, randomisation will help to eliminate self selection bias which can greatly affect the study results. Baseline pain scores will be taken with the McGill Pain Questionnaire (Short Form) since this measures multidimensional aspects of pain and also saves time (Zalon, 1999). This would be done before and after the intervention, during the preoperative and postoperative periods. Patients would be informed about the intervention to avoid interruptions in the study, however, investigators would be blinded about particular interventions so as to reduce any personal biases. After data collection, descriptive and inferential statistical techniques will be used in analysing the data in terms of frequencies, percentages, means, odds ratio, standard deviations, confidence interval (95%), probability value, test of significance (using t-test and analysis of variance), chi-square test (to verify hypothesis in terms of group differences) and power analysis (to estimate the probability of making a type II error or sample size requirements) (Polit and Beck, 2008). Based on these findings, conclusions can then be drawn about the effectiveness of music as a pain management strategy. Apart from a randomised controlled trial, a quasi-experimental study design using pre-post test design can be conducted using the three groups without randomly allocating them. An advantage of this is that, it is more likely to be acceptable by a large group of people, however, this would lead to unreliable and inconclusive results. In spite of the fact that quantitative designs are highly regarded than qualitative designs, it is not without critics. Approaches are rigid and prespecified in terms of processes and range of information (Robson, 2002). It focuses on a relatively small part of human experience and tend to control or eliminate complexities; thus, leading to a narrow focus and sometimes doubtful insights (Polit and Beck, 2008). Conversely, a qualitative study design should not be overlooked, in spite of the fact that, they are not the appropriate approaches for my research question. Qualitative designs are subjective, context-specific (Baker, 1998; Rice and Ezzy, 1999), naturalistic (Hoinville and Jowell, 1978; Pope and Mays, 1995), inductive and cannot be generalised. (Polit and Beck, 2008). They provide insights into the lives , experiences and understandings of research participants, and explores their relationship (Babbie and Mouton, 2001; Denzin and Lincoln, 1994; Du Ploy, 2001; Henning, 2004; Robson, 2002). Data collection and analysis proceed together (Miles and Huberman, 1994; Smith, 1988), and can lead to theory development (Henning, 2004). Qualitative designs is based on frameworks such as phenomenology, grounded theory and ethnography (Robson, 2002). With regards to my question, a phenomenological study could be conducted to explore the experiences and meanings of listening to music as a pain management strategy among patients (after an abdominopelvic operation). Some of the benefits of qualitative designs include its flexibility, in-depth understanding and naturalistic nature (Wimmer and Dominick, 1983; Dillon et al., 1993). It can be very useful in studying new areas to allow for future generalisations using quantitative designs. Nevertheless, it has also got its own setbacks: the use of small sample sizes does not allow for generalisation (Chelimsky and Shadish, 1997; Patton, 1990), it lacks rigour (Mays and Pope, 1995), provides a biased view due to the researcher's reflexivity (Robson, 2002) and leads to unreliable results(William and Dominick, 1983; Du Plooy, 2001).
Systematic review is an approach used in summarising, evaluating and communicating the findings of studies (NHS CRD 1996) that address a clearly defined question (Mulrow, 1987). It involves the use of explicit methods in identifying, selecting, appraising, collecting and analysing studies included in a review(Greenhalgh, 1997; NHS CRD, 2001). The precise use of methods minimise bias(Greenhalgh, 1997; Mulrow, 1987; Oxman and Guyatt, 1988) and results in more reliable and accurate conclusions (Greenhalgh, 1997; L'Abbe, Detsky and O'Rourke, 1987; Oxman, 1993; Sacks, Berrier, Reitman et. al, 1987; Thacker, 1988). Systematic reviews (SRs) can take the form of meta-analyses (using statistical methods), meta-syntheses or mixed method approach (Polit and Beck, 2008). Specifically, meta-analyses increases power and precision of overall results in terms of effects and risks (Greenhalgh, 1997; Mulrow, 1987). My research question can be answered through a systematic review of high quality randomised controlled trials . This can be done by identifying studies conducted on the "effects of music on postoperative pain" from electronic databases, hand searching, internet and so on. Those studies that meet my inclusion criteria would be evaluated and summarised using statistical methods. For consistency and validity of conclusions, the review can be conducted with the collaboration of an experienced researcher or team. Notwithstanding, as an international student on this course, a modified systematic review would be most suitable. The advantages of SRs are enormous; it manages large amounts of data (Greenhalgh, 1997) to establish generalisability and consistency (Morgan, 1986; Mulrow, 1987; Bossel, Blanchard, Panak et al., 1989; Greenhalgh, 1997). It is usually quicker and less expensive than conducting a new study (Chalmers, Hetherington, Newdick et al., 1986; Greenhalgh, 1997). SR helps to identify issues of importance (Petticrew, 2003; Petticrew and Roberts, 2006) and can be used in directing future studies (Brown, Brunnhuber, Chalkidou et. al, 2006). Unlike traditional reviews, it can detect small but significant effects (Cooper and Rosenthal, 1980). More importantly, SRs assists policy makers to become more knowledgeable (Glass, 1976) in developing guidelines and enacting laws (Greenhalgh, 1997). In spite of the numerous benefits, SRs come with its own challenges. Apart from the difficulty in exploring quality studies, synthesis may oversimplify important aspects of studies (such as inclusion/ exclusion criteria or the nature of an intervention). As a result, difficulties arise when applying findings in certain circumstances (Greenhalgh, 1997). Some critics (Eysenck, 1995) argue that if the benefits of a treatment are so concealed that it requires meta-analyses to determine its effectiveness, then both the treatment and its underlying mechanism needs further improvement.
Evaluation research seeks to explore the worth of an ongoing program with the aim of accepting, modifying or rejecting it (Polit and Beck, 2008). As a result, it leads to professional accountability and development (Hawe, 2005). This research approach usually does not require ethical approval as it normally does not introduce an intervention. The process (formative) or outcome (summative) of an intervention can be analysed using experimental, quasi-experimental and nonexperimental designs (Polit and Beck, 2008). In evaluation research, cost/ benefit analysis and cost-effectiveness are the most widely used approaches in cost analyses (Chang and Henry, 1999). With regards to my question, formative and summative evaluation with quasi-experimental design could be used to examine the processes (duration, intensity, rhythm of a preferred music) and outcomes of music therapy as a pain management strategy. Data concerning pain levels and analgesic consumption would be retrieved from previous records and compared to an ongoing program. Alternatively, baseline data would be collected from two groups of postoperative patients (without random allocation) before the introduction of the intervention to one of the groups. After a period of implementation, outcomes (such as pain intensity, analgesic consumption) would be assessed to determine the cost-effectiveness of the music therapy. In a similar study, the quality and duration of the music could also be assessed to determine its effectiveness on pain levels and analgesic consumption. Although the findings of an evaluation study cannot be generalised, interventions can be implemented in other settings to enhance understanding (Polit and Beck, 2008).
With the increasing emphasis on evidence-based practice, it has become imperative for nurses to engage in research and based their decisions on the most appropriate, cost effective and patient preferred interventions. Every research question can be answered in diverse ways. Notwithstanding, the selection of an appropriate design depends on the various factors such as the aims, researcher's preference, available resources, time, ethical and practical feasibility. Although systematic reviews of randomised controlled trials are deemed as the "gold standard" in recent times, no single methodology or method can completely answer a research question. While each is subject to being infallible, each approach adds to the world of evidence