Understanding literature reviews: Implications for music therapy

ABSTRACT

Social science and health care researchers, including music therapy researchers, have conducted various types of literature reviews; these types have largely been understood as either traditional or systematic reviews. However, authors, both within and outside of music therapy, have used mixed terminology, provided questionable rationales for categorization, disagreed about categorization, or used ambiguous language to describe broad review methodologies. As an alternative to categorization, one could view literature review processes and methodologies from the standpoint of historical development and methodological value; each methodology answers different types of questions through review’s characteristics. The purpose of this article is to present a comprehensive overview of literature review processes and methodologies by (a) describing the general purposes of narrow and broad literature reviews, (b) providing a historical overview of broad reviews, and (c) describing broad review methodologies in relation to their respective definitions, histories, methodological characteristics, purposes, example questions, and study examples from healthcare and music therapy literature. After providing a general overview of categories and histories of reviews, the author analyzes a total of 25 broad literature review methodologies and 2 salient broad review procedures. The author then discusses implications for music therapy research and practice, specifically in relation to current broad review publications, policy development, information flow, stakeholder engagement, and tertiary reviews. Furthermore, this article may act as an initial reference for researchers and educators.

KEYWORDS:

• Literature review
• systematic review
• music therapy
• overview
• research
• pedagogy

Introduction

Research can be understood as a methodical procedure of inquiry that leads to new ways of understanding. Research produces some type of outcome that “contributes to or modifies existing knowledge or practice” (Bruscia, 1995, p. 21). In other words, research informs and reorganizes an evidence base through a multiplicity of perspectives.

History plays an important role in research. One engages with present inquiry by understanding prior evidence. An evidence-based practice that includes clinician expertise, patient experience, and context will also take into account extant literature (Rycroft-Malone et al., 2003). Regardless of research questions, philosophical positions, or methodology within a study, one investigates effectively by understanding what has been investigated beforehand.

The term literature review refers to a documented summary of prior research, and is therefore also known as an evidence summary. Literature review exists as a foundational task in almost any research project. Reviews connect past studies to present and future inquiry, and connect individual studies with the collective body of knowledge.

Researchers have recognized the importance of evidence summary and evidence synthesis for more than two centuries. Naval surgeon James Lind described the cumulative research review in relation to 18th century scurvy treatment (Chalmers, Hedges, & Cooper, 2002; Grant & Booth, 2009; Hampton, 1998). Statistician Legendre developed the method of least squares to promote statistical synthesis in the early 19th century. Strutt (1885) acknowledged the importance of those instances “in which discovery and explanation go hand in hand, in which not only are new facts presented, but their relation to old ones is pointed out” (p. 20). In some cases, authors have noted the challenges in promoting quality evidence summary. Wright Mills stated “There are never enough bricks and there are too few good synthesizers to search out the bricks and thus put the wall together” (1959, p. 65). More specific to health care, Solesbury noted that “Social science is very bad at the cumulation and re-use of past research results” (2002, p. 92).

Two overarching categories—narrow and broad—differentiate general approaches to literature review. Where a narrow literature review occurs within a primary study, a broad literature review acts as a stand-alone evidence summary (Cooper, 1998). Authors have historically categorized broad review methodologies into dichotomous types: traditional and systematic. Disagreements regarding categorization, as well as inconsistent and ambiguous language in the literature, have likely promoted confusion. The broad range of review methodologies demonstrate nuance and complexity; they can be more clearly articulated when acknowledged as a set of singular historical contributions, each with a set of characteristics that address a particular type of need. Healthcare disciplines, including music therapy, benefit by understanding the histories, values, and characteristic differences between broad review methodologies.

Purpose

The purpose of this article is to provide a comprehensive overview of literature review processes and methodologies. To achieve this purpose, the author will (a) describe general purposes of narrow and broad literature reviews; (b) provide a historical overview of broad literature reviews; and (c) describe 25 broad review methodologies and 2 salient broad review procedures in relation to their respective definitions, histories, characteristics, purposes, example questions, and study examples from healthcare and music therapy literature. The article will conclude with implications for music therapy research and practice.

Narrow and broad literature reviews described

Most researchers write literature reviews in the narrow sense. Narrow reviews summarize past literature in order to inform, justify, and substantiate primary studies (Hanson-Abromeit & Sena Moore, 2014). In contrast to the narrow type, a broad literature review is a stand-alone study. A broad review inquires about the relationships between prior studies to create an evolved whole (Onwuegbuzie, Collins, Leech, Dellinger, & Jiao, 2010). In this sense, the broad review generally seeks to synthesize some amount of past literature for a particular purpose. Figure 1 describes the purposes of narrow and broad literature reviews.

Figure 1. Literature reviews: overarching types and their purposes.

Historical categorization of broad literature review methodologies

Authors have categorized broad methodologies into dichotomous categories: “traditional” (or “non-systematic”) reviews and “systematic” reviews (Gough, Oliver, & Thomas, 2012; Hanson-Abromeit & Sena Moore, 2014). Researchers have defined the systematic review as a literature review that includes rigorous literature selection, appraisal, analysis, synthesis, and reporting in order to answer specific research questions (Gough, Oliver, & Thomas, 2012; Hanson-Abromeit & Sena Moore, 2014; Khan, Kunz, Klijnen, & Antes, 2011; Porta, 2008). However, categorizations have not always appeared straightforward.

Challenges with categorization, terminology, and consensus

Researchers have historically used the terms systematic review, integrative review, research synthesis, knowledge synthesis, and meta-analysis interchangeably (Kastner et al., 2012; LoBiondo-Wood, 2014). More recently, authors have sought to clarify and differentiate these terms. According to A Dictionary of Epidemiology (Porta, 2008), a systematic review and a meta-analysis are differentiated in that meta-analysis is the application of strategies that limit bias in the assembly, critical appraisal, and synthesis of all relevant studies on a specific topic. Hanson-Abromeit and Sena Moore (2014) also differentiated meta-analysis from systematic review, noting the use of a “specialized statistical technique that results in an effect size” (p. 11) in order to locate a weighted average effect for the particular intervention or event. The meta-analysis procedure may be, but is not necessarily, used as a part of the systematic review process. However, any systematic review that seeks to assess efficacy must aim to include a meta-analysis.

Authors may occasionally disagree about how to categorize particular broad review methodologies, or may not provide adequate rationales for exclusion. Where Hanson-Abromeit and Sena Moore (2014) note that meta-ethnography does not qualify as a systematic review, other authors articulate differently (Chalmers, Hedges, & Cooper, 2002; Gough et al., 2012; Harden, 2010). Hanson-Abromeit and Sena Moore (2014) note that an integrative review includes diverse methodologies, and is therefore not a systematic review. However, some systematic review types include examination and synthesis of heterogeneous methodologies (see Table 1).

Table 1. Broad literature review types: methodological characteristics.

	Methodological Characteristics
	Researchers Search Appraisal Analysis Primary Outcome Emphases
Review Types	Single /Multiple	Multiple Encouraged or Required	Stakeholder Participation Encouraged	Comprehensive Search Requirement	Primary Studies	Secondary Studies	Inclusion/Exclusion Required	Formal Quality Assessment	Informal/Internal Quality Assessment	Homogenous Methodologies	Heterogeneous Methodologies	Homogenous Contexts	Heterogeneous Contexts	Homogenous Outcomes	Heterogeneous Outcomes	Efficacy /Effectiveness	Impact /Meaning	Concept /Theme Creation	Theory Development/Analysis	Gaps in Literature	Risk /Heterogeneity Assessment	Effect Critique/Bias Reduction	Theoretical Critique/Problematizing	Language or Content Outcome	Information Linkage and Flow	Policy Guideline Development	Mapping of Content	Author Expertise
Content Analysis	✓				✓	✓			✓	✓	✓	✓	✓	✓	✓			✓						✓
Narrative Review	✓				✓	✓			✓	✓	✓	✓	✓	✓	✓													✓
Theoretical Review	✓				✓				✓	✓	✓	✓	✓	✓	✓				✓				✓
Systematized Review	✓			✓	✓		✓		✓		✓	✓		✓		✓	✓		✓	✓
Scoping Review		✓			✓	✓	✓				✓	✓		✓				✓	✓							✓	✓
Rapid Review	✓				✓	✓			✓		✓	✓		✓												✓
State of the Art Review	✓				✓	✓				✓	✓	✓		✓						✓						✓		✓
Mapping Review		✓	✓	✓	✓	✓					✓		✓		✓				✓	✓				✓	✓		✓
Meta-ethnography	✓			✓	✓		✓	✓		✓	✓		✓		✓		✓	✓	✓
Integrative Review	✓			✓	✓		✓	✓				✓	✓				✓		✓	✓
Stand-alone Systematic Review	✓	✓		✓			✓	✓		✓	✓	✓		✓		✓	✓		✓	✓						✓
Meta-analysis (procedure)	✓	✓		✓	✓		✓	✓		✓		✓		✓		✓					✓	✓
Cochrane Review		✓	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓		✓	✓			✓
Narrative Synthesis	✓	✓		✓	✓		✓	✓		✓	✓		✓		✓		✓	✓	✓					✓		✓
Meta-synthesis (procedure)	✓	✓		✓	✓		✓	✓			✓	✓		✓			✓	✓	✓
Meta-narrative Review	✓	✓		✓	✓		✓	✓			✓	✓	✓	✓	✓		✓		✓
Critical Interpretive Synthesis	✓	✓		✓	✓		✓		✓		✓	✓	✓	✓	✓			✓	✓				✓
Thematic Synthesis	✓	✓		✓	✓		✓	✓			✓	✓	✓	✓	✓			✓	✓
Framework Analysis	✓	✓		✓	✓		✓	✓			✓	✓	✓	✓	✓		✓	✓	✓								✓
Best-fit Framework Synthesis	✓	✓		✓	✓		✓	✓			✓	✓	✓	✓	✓			✓	✓							✓	✓
Realist Synthesis	✓	✓	✓	✓	✓		✓	✓		✓	✓	✓	✓	✓	✓	✓		✓	✓
Qualitative Research Synthesis	✓	✓		✓	✓					✓	✓	✓	✓	✓	✓		✓	✓	✓				✓
Metasummary	✓	✓		✓	✓		✓	✓			✓		✓	✓			✓	✓
Mixed-methods Systematic Review	✓	✓		✓	✓		✓	✓			✓	✓	✓	✓		✓	✓	✓	✓
Meta-reviews	✓	✓		✓		✓				✓	✓			✓	✓	✓	✓				✓					✓
Meta-epidemiology	✓	✓				✓				✓	✓	✓	✓	✓		✓					✓	✓
Umbrella Review	✓	✓		✓	✓	✓	✓	✓			✓	✓		✓	✓	✓	✓			✓						✓

Understanding components of broad literature reviews

Dichotomous categorization can limit a broader and deeper understanding of individual methodologies. The author contends that each literature review methodology contains unique value due to the various components each possesses. Each component, as related to research questions and methodological characteristics, deserves further consideration.

Research questions

Questions guide each step of the investigative process, and therefore play a role in determining a researcher’s chosen methodology. Researchers can investigate a range of questions, including but not limited to outcomes, methods, theories, applications, or combinations thereof (Cooper, 1998). Each broad review type caters to particular types of research questions and particular types of outcomes (see Table 2).

Table 2. Broad literature review methodologies: example questions.

Subtype	Example Questions
Content Analysis	What are the most frequently discussed music intervention types used within special education according to peer-reviewed journals?
Narrative Review	In what ways has music therapy been used in health care research to promote health in cancer patients?
Theoretical Review	What theories underpin the use of music to promote language development?
Systematized Review	(see stand-alone systematic review)
Scoping Review	What types of evidence are available regarding music therapy in hospice setting? What concepts appear most common within that evidence?
Rapid Review	What is known about music interventions to promote improved gait?
Mapping Review	Where and how have anxiety-reducing music interventions been conducted?
Meta-ethnography	What are the primary concepts employed in case studies discussing the use of Creative Music Therapy?
Integrative Review	What gaps in the literature exist regarding music therapy interventions in neonatal intensive care units?
Stand-alone Systematic Review	How do different types of music therapy interventions compare in terms of anxiety reduction with pre-operative medical patients?
Meta-analysis (Procedure)	How effective is patterned sensory enhancement for improving range of motion in stroke patients?
Cochrane Review	What is the effect of music therapy interventions on motor skills in people with stroke?
Narrative Synthesis	What factors should be considered when implementing relaxation interventions in a medical setting?
Meta-synthesis (Procedure)	What are the experiences of caretakers in relation to music interventions in hospice settings?
Meta-narrative Review	How do various theoretical orientations of music therapy practice compare when implemented within special education practice?
Critical Interpretive Synthesis	What decisions regarding theoretical frameworks and implementation have been made by music therapy researchers in randomized controlled trials?
Thematic Synthesis	How do clients with Parkinson’s disease perceive their own health after receiving medical music therapy interventions?
Framework Analysis	What social or cultural factors play a role in the performance of music therapy undergraduate students, and how may this affect pedagogy?
Best Fit Framework Synthesis	Does the current literature present a theoretical framework of cultural humility that promotes success in undergraduate music therapy pedagogy, and if so, what specifically does that framework look like?
Realist Synthesis	What context-mechanism-outcome configurations appear to promote emotional expression with at-risk youth in music therapy sessions?
Qualitative Research Synthesis	What are the themes within clinical improvisation case studies regarding the promotion of increased communication?
Metasummary	What are the implications of synthesized survey results in relation to music therapist job satisfaction?
Mixed-methods Systematic Review	What is the level of effectiveness and quality of personal impact of music therapy to reduce post-operative anxiety?
Meta-reviews	According to systematic reviews, how effective are music-based gait training interventions with stroke patients?
Meta-epidemiology	What potential flaws exist in current meta-analyses of music therapy interventions with people who have autism spectrum disorder?
Umbrella Review	How do music interventions, that promote emotional regulation in children who are at-risk, compare in terms of intervention types?

Methodological characteristics

A methodology frames a set of research methods in order to gain a specific result (Crotty, 1998). Methods can be understood as a set of specific steps, whether predefined or iterative, that move a study through data retrieval, analysis, outcome, and reporting (Crotty, 1998). Various methodologies approach methods differently, resulting in a set of unique methodological characteristics. These distinctions include but are not limited to: (a) researchers, (b) search, (c) appraisal, (d) analysis, (e) and primary outcome emphases, (see Table 1). Each characteristic is discussed below.

Researchers

Most broad review methodologies allow for single or multiple researchers; some require multiple researchers. Some methodologies encourage stakeholder participation. A stakeholder is a person or organization that has direct interest in research outcomes, and may include clients, caretakers, or policy makers (Hovland, 2005).

Literature search

Some broad methodologies require a comprehensive literature search, while others do not. Some methodologies review only primary studies while others focus on analyzing secondary studies (i.e. other literature reviews). Some tertiary reviews analyze both primary and secondary studies.

Appraisal

Some review methodologies require that both inclusion and exclusion procedures are conducted and reported. Some methodologies require formal quality assessment for each included study, while others encourage informal quality assessment. Some methodologies may not require study exclusion or promote any assessment of study quality, in some cases because the research questions that the methodology addresses do not benefit from such.

Analysis

The purpose of each methodology reflects an emphasis on different types of analysis. Analysis characteristics note whether or not the methodology focuses on assessing highly similar or purposefully diverse (a) methodologies, (b) study contexts, and (c) study outcomes. These three areas are described below.

Primary study methodologies include but are not limited to controlled trials, phenomenological studies, ethnographies, descriptive studies, and case studies. Meta-analysis focuses only on randomized controlled trials (RCTs) (homogenous) while realist synthesis is able to examine different (heterogeneous) methodologies.

Context, for the sake of this article, refers to the sameness or diversity of environmental underpinnings. Context can include the population, setting, the need being addressed, and even the intervention itself. Meta-analysis again provides a useful example, in that the primary studies must be addressing the same contextual problem and intervention. A content analysis, however, does not require that the same intervention be studied, because it focuses on the communication patterns used to describe an intervention or process. For example, a content analysis may analyze how often research articles report the instruments being used.

Some broad review methodologies must analyze the same types of outcomes, while other methodologies are able to examine diverse types of outcomes. Meta-analysis will assess similar outcomes related to a particular intervention for a particular problem. Critical interpretive synthesis may look at the same problem and outcomes from the same RCTs, but may also analyze primary studies discussing personal and social experiences.

Synthesis outcome focus

Different broad review methodologies report different types of outcomes. Outcome focus characteristics include the following:

• Efficacy/effectiveness: measures the capability of an intervention to either promote efficacy in an ideal laboratory setting, or promote effectiveness in a realistic clinical setting (Godwin et al., 2003).

• Impact/meaning: explains how clients, caretakers, clinicians, or others experience changes through a phenomenon/intervention.

• Concept/theme creation: promotes the construction or organization of words or ideas in order to better understand a particular phenomenon.

• Theory development/analysis: focuses on developing theory through construction, analysis, comparison, or a combination thereof.

• Gaps in literature: analyzes and reports areas that have not yet been studied.

• Theoretical critique/Problematizing: a unique type of theory analysis that warrants its own consideration; focus on critical inquiry (Barnett-Page & Thomas, 2009).

• Effect critique/bias reduction: seeks to understand the occurrence or cause of a problem by locating issues with study bias.

• Risk/Heterogeneity assessment: presents information, usually in descriptive table format, related to control group risk levels. Describes differences between study methods.

• Language or content outcome: presents results of communication patterns.

• Policy /guideline development: Seeks to inform administrators and stakeholders and promote policies on a health care need, or to promote guidelines in clinical practice or assessment.

• Information linkage and flow: describes what evidence there is, what subsets of evidence exist, how subsets relate to each other, and how information regarding this evidence is accessed.

• Mapping of content: Results require a visual map or flowchart to provide complete understanding of results to reader.

• Author expertise: focuses on promoting awareness through an individual’s knowledge on a topic.

Types of outcome reporting

Every methodology can report outcomes through tables, figures, and narrative results. However, some methodologies will tend to focus on one of these three.

Each of the above components—questions and methodological characteristics—help to comprise singular broad review methodologies. The following section promotes a greater contextual and historical understanding of each methodology.

Broad literature review methodologies: descriptions and histories

Content analysis

Content analysis reviews human communication within researcher-selected literature in order to locate frequencies, patterns, and trends (Babbie, 2010). Word frequency may reveal empirical tendencies in practice (e.g. frequency of “piano” instead of “guitar”) or political tendencies within communication (e.g. the use of person-first language in journal articles). Holsti (1969) notes three purposes of content analysis: to describe trends in communication content, to compare content to standards, and to describe patterns of communication. Examples of communication patterns include (a) what was said, (b) who said it, (c) to whom was it said, (d) where or on what platform was it said, and (e) what function did the communication serve. The researcher can choose the units of analysis, including but not limited to words, concepts, phrases. The researcher can also design the content analysis to measure the objective use of units, to construct and analyze themes from units, or both. For example, Matney (2015) analyzed the use of percussion in therapy in peer-reviewed articles over seven decades. Matney used descriptive statistics to present trends in instrumental use, and populations using percussion. He also employed concept construction and analysis of themes to communicate how and why percussion has been used in the field.

The exact origin for content analysis appears unknown (Hanson-Abromeit & Sena Moore, 2014). However, the written history of content analysis suggests associations with political analysis (Kaplan, 1943, p. 230), hermeneutics and psychology (Rust, 1983), sociology (Glaser & Strauss, 1967), and grounded theory (Crotty, 1998). Berelson (1952) wrote the first book on the methodology.

Stand-alone narrative review

A stand-alone narrative review, also known as a stand-alone literature review, may be understood as a summary of author-selected literature. Results demonstrate an author’s expertise on a particular topic (Cipriani & Geddes, 2003). Narrative reviews largely rely on the experience of the author and do not promote transparent procedures. For these reasons, narrative reviews may suffer from selection bias (Gough et al., 2012) and publication bias (Cipriani & Geddes, 2003).

The stand-alone narrative review does not appear to have a definitive historical origin (Hanson-Abromeit & Sena Moore, 2014), but is the likely precursor of all other broad reviews. Researchers appear to have originally used narrative reviews for “academic debates” and “think pieces” (Gough et al., 2012, p. 5), but the methodology evolved to promote general theory development, identify the state of current knowledge, and promote historical analysis (Baumeister & Leary, 1997). For example, Tang and Vezeau (2010) completed a narrative review to identify the implementation of music interventions in music therapy. The authors sought to understand with whom interventions were completed, what outcomes were sought, how interventions were operationalized, what comprised the music interventions, what measurement tools were used, and how effective were the interventions.

Theoretical review

The theoretical review, or “critical review,” analyzes a theory or theories in relation to phenomenon, idea, concept, or issue. Researchers may analyze the “breadth, internal consistency, and the nature of their (theoretical) predictions” (Cooper, 1998, p.4). Researchers may examine coherence (does the theory make sense for practice), fidelity (are people actually and faithfully applying the theory) or reliability (does the theory promote a particular effectiveness or type of meaningful outcome). In some cases, two or more theories may be compared to each other (Hanson-Abromeit & Sena Moore, 2014). For example, Dimitriadis and Smeijsters (2011) analyzes treatment theories related to autism spectrum disorder (ASD). The authors present select psychological theories (e.g. Damasio, Stern) and provide linkages to select music therapy theories (e.g. Stige, Ansdell, Lee) and applied techniques (i.e. Wigram’s descriptions of clinical improvisation techniques). Results suggested particular types of connection that may be relevant to the educator or researcher.

Cooper (1998) implied a common separation between a theoretical review and a “research synthesis” (p. 4). The theoretical review has historically assessed critical experiments—those studies that seek to directly test the strength and/or superiority of a theory for the social science community. Interestingly, literature does not clarify whether the theoretical review originated in scientific theory (fact-supported theories within the natural world), critical theory (addressing historical, social, and political components of the world), or a combination of the two. The theoretical review may have acted as a precursor to current theory-driven “systematic review” methodologies.

Rapid review, scoping review, state-of-the-art review, and mapping review

Four review methodologies use methods that inform evidence-based policy while dealing with time constraints. Researchers, administrators and government officials may seek to make informed decisions based on these evidence summary types. Each review type warrants brief description. A rapid review, or rapid evidence assessment, uses abridged systematic review techniques to provide policymakers evidence-based information within a five-week timeline (Grant & Booth, 2009; Khangura, Konnyu, Cushman, Grimshaw, & Moher, 2012). A scoping review offers a preliminary evaluation of available research literature on a topic or issue (Grant & Booth, 2009), and may be used to inform the development of a systematic review (Gough, Oliver, & Thomas, 2013). A state-of-the-art review highlights a contemporary issue in regard to current research. A mapping review, or systematic map, maps out current literature to identify gaps in the literature; this information may be used for either a primary study or a more rigorous evidence summary.

The author of this study located one scoping review related to music therapy. Scott and Kidd (2016) summarized the current research regarding music participation and anxiety, depression, and agitation with people who have dementia. The authors review suggested insufficient evidence in this area, and recommended that researchers become more specific about interventions in future studies.

British Prime Minister Tony Blair and his administration popularized, and then Adrian Smith substantiated, evidence-based policy just before the turn of the century (Boaz, Ashby, & Young, 2002). The scoping review originated in 2001 (Dijkers, 2015; Mays, Roberts, & Popay, 2001). Arksey and O’Malley (2005) created a methodological framework for the scoping review. The Evidence for Policy and Practice and Coordinating Centre in London developed the mapping review (Grant & Booth, 2009; James, Randall, & Haddaway, 2016), and Khangura et al. (2012) provided a detailed set of methods for the methodology.

Systematized review

A systematized review abridges a “systematic review” methodology for educational purposes; changes in methods often seek to shorten time to complete. A master’s degree student may complete a systematized review for a research assignment or thesis; such may also facilitate a future complete systematic review. Some journals have published systematized reviews (see Table 3) (Grant & Booth, 2009).

Table 3. Broad literature review methodologies: journal article examples.

	Article Examples
Methodology	Music Therapy	Related Fields
Content Analysis	Jones, 2006; Matney, 2015; Roberts & McFerran, 2008	Ramanadhan, Mendez, Rao, & Viswanath, 2013; Schmidt, Raque-Bogdan, Piontkowski, & Schaefer, 2011
Narrative Review	Tang & Vezeau, 2010	Afshar, Raju, Ansell, & Bleck, 2011; Holmes & Padgeham, 2011
Theoretical Review	Dimitriadis & Smeijsters, 2011	Carl, Soskin, Kerns, & Barlow, 2013; Rohrmeier & Koelsch, 2012
Systematized Review		Barr-Walker, 2017
Scoping Review	Scott & Kidd, 2016	Jepson, Blasi, Wright, & Riet, 2001
Rapid Review		Kilgour, Grundy, & Monrouxe, 2016
State of the Art Review		Kales, Gitlin, & Lyketsos, 2015
Mapping Review		Gordon et al., 2012
Meta-ethnography	O’Callaghan et al., 2016	Adams et al., 2011; Franzel, Schiegerhausen, Heusser, & Berger, 2013
Integrative Review	Burns, 2012	Bailey et al., 2010
Stand-alone Systematic Review	Brown & Jellison, 2012; Carr, Odell-Miller, & Priebe, 2013; Sena Moore, 2013	Choudry, Fletcher, & Soumerai, 2005; Doherty & Stravropoulou, 2012
Meta-analysis (procedure)	Dileo, 2006; Gold, Solli, Krüger, & Lie, 2009; Lee, 2016; Standley, 2002, 2012	Bray, Cowell, & Hinde, 2011; Virues-Ortega, Julio, & Pastor-Barriuso, 2013
Cochrane Review	Geretsegger et al., 2014; Bradt, Magee, Dileo, Wheeler, & Mcgilloway, 2010; Mossler, Chen, Heldal, & Gold, 2011	Van Dalen et al., 2012; Rueda, Pascual, & Subirana Casacuberta, 2011
Narrative Synthesis	Carr, Odell-Miller, & Priebe, 2013; Draper, 2016; McDermott et al., 2013	Schrank et al., 2013
Meta-synthesis (procedure)	Solli et al., 2013	Bennion, Shaw, & Gibson, 2012; Hodge, Horvath, Larkin, & Curl, 2012
Meta-narrative Review		Brown et al., 2015; Murad et al., 2013; Van Belle, 2016
Critical Interpretive Synthesis	Comte, 2016; McFerran, Garrido, & Saarikallio, 2013; Skewes McFerran, Hense, Medcalf, Murphy, & Fairchild, 2017	Dixon-Woods et al., 2006; Fleming, 2010; Yazdani, Lakeh, Ahmady, Foroutan, & Afshar, 2015
Thematic Synthesis		Latchem & Greenhalgh, 2013; Rathbone, Todd, Jamie, Bonam, Banks, & Husband, 2016
Framework Analysis		Brunton, Oliver, Oliver, & Lorenc, 2006; Oliver et al., 2008; Tallentire, Smith, Skinner, & Cameron, 2015
Best-Fit Framework Synthesis		Carroll, Booth, & Cooper, 2011; Shaw et al., 2016
Qualitative Research Synthesis	Meadows & Wimpenny, 2016	Hansen, Draborg, & Kristensen, 2011
Metasummary		Childress, 2013; Williams, 2006
Realist Synthesis		Kane, Gerretsen, Scherpbier, Poz, & Dieleman, 2010; Leeman et al., 2010; Thomas & Thomas, 2014
Mixed-methods Systematic Review		Arrowsmith et al., 2016; Thomas et al., 2003; Wang & Yeh, 2012
Meta-review	Kamioka et al., 2014	Harald & Gordon, 2012; Matjasko et al., 2012
Meta-epidemiology		Caldwell et al., 2010; Tzoulaki, 2011
Umbrella Review		Krebs Seida et al., 2009

Meta-ethnography

Researchers use meta-ethnography to build a comparative social understanding from “interpretive” studies (Barnett-Page & Thomas, 2009; Noblit & Hare, 1988). A meta-ethnography generally moves through three stages: (a) identifying concepts in the literature, (b) description of the primary study concepts as provided by the original author, and (c) constructing the reviewer’s synthesis. A meta-ethnography seeks to develop an awareness of underlying concepts found within interpretive literature through three distinctive types of synthesis: reciprocal translation (focusing on agreement), refutational synthesis (focusing on conflicts), and line of argument synthesis (building up parts of information to create a wider narrative, such as building up a greater understanding of culture or organization) (Gough et al., 2012; Noblit & Hare, 1988).

Ethnographic research originated as a participant-observer, fieldwork methodology within the discipline of anthropology. Educational researchers initially created meta-ethnography in the early 1980s to synthesize ethnographic studies (Noblit & Hare, 1988). However, authors have since used the methodology with various primary study types (Gough et al., 2012). Meta-ethnography has produced theories with “greater explanatory power” than narrative reviews through the use of higher-order synthesis constructs (Barnett-Page & Thomas, 2009).

Integrative review

The integrative review synthesizes information from both empirical and theoretical literature. The integrative review analyzes a phenomenon to highlight gaps in the literature, construct new concepts and theoretical frameworks, and locate new issues (Broome, 2000; Hanson-Abromeit & Sena Moore, 2014). For example, Burns (2012) conducted an integrative review to assess the use of theories underpinning music selection for interventions. She found that most studies did not provide a rationale or an intervention theory related to music selection. Furthermore, intervention reporting lacked in general.

The term “integrative review” was used interchangeably with, and then largely replaced by “systematic review” in the Cumulative Index of Nursing and Allied Health Literature (Stevens, 2001, p. 536). One may wonder whether the change in terminology occurred merely due to semantics, a change in methods, or a new focus on rigor.

Stand-alone systematic review

Stand-alone systematic reviews answer questions related to populations, interventions, outcomes, study designs, and theories (Okoli, 2015). Authors note the review’s capability to locate gaps within the literature, and to design new theory for future research through the synthesis of evidence (Hanson-Abromeit & Sena Moore, 2014). For example, Sena Moore (2013) sought to synthesize research related to music experiences and their impact on the neural mechanisms of emotional regulation. Results suggested neural activation related to regulation occurred with familiar and preferred music and when singing. More complex and unexpected music structures were more likely to activate neural mechanisms related to stress.

The stand-alone systematic review originated in health care (Hanson-Abromeit & Sena Moore, 2014; Khan et al., 2011). However, a particular origin is difficult to locate, largely due to the term being used interchangeably with integrative review and meta-analysis. Several authors have provided updated methods for the methodology (Gough et al., 2012; Whittemore & Knafl, 2005). Key to understanding the versatility of the systematic review is its relationship to the particular research question or questions being asked. As mentioned earlier, any systematic review that seeks to assess efficacy must include a meta-analysis.

Meta-analysis

Researchers and stakeholders often wish to understand the efficacy or effectiveness of interventions, as based in numerical analysis, probability, and generalizability. The meta-analysis procedure promotes statistical synthesis of data amongst studies judged sufficiently similar (Thomas, Harden, & Newman, 2012). Researchers achieve results by locating or computing effect sizes of individual studies for specific outcomes, and using statistical analysis to compute a pooled effect size per selected outcome. A meta-analysis provides more accurate estimates of treatment efficacy than a single study (Bradt, 2016). Implementation of statistics for the purpose of summary apparently originated in gambling (O’ Rourke, 2007), math (O’ Rourke, 2007), and astronomy Airy (1861). However, these concepts and mathematical operations were not employed in the medical sciences until the 20th century (Chalmers et al., 2002). Karl Pearson (1904) published a key paper in the British Medical Journal, using statistical analysis to synthesize information from 11 separate studies measuring the effects of a typhoid vaccine.

Standley (1986) conducted the first meta-analysis in music therapy, analyzing the efficacy of music in medical and dental treatments; the author included applications for clinical practice and medical programming. More recently, Lee (2016) synthesized current research on music and pain management. Results suggested that music interventions could decrease pain through various self-report and biomedical measures.

Cochrane review

The Cochrane review—an umbrella of collaborative, comprehensive systematic review types—focuses on the power of multiple experts and stakeholders to synthesize and communicate evidence (Porta, 2008). There exist five Cochrane review types that are published and maintained by the Cochrane Library (www.cochrane.org): (a) intervention reviews, (b) diagnostic test accuracy reviews, (c) methodology reviews, (d) qualitative reviews, and (e) prognosis reviews (see www.cochranelibrary.com/about/about-cochrane-systematic-reviews.html for further details)

Cochrane reviews originated in health care (Khan et al., 2011). Archie Cochrane’s Effectiveness and efficiency: Random reflections on health services (1972) challenged the medical community to obtain better evidence for their practices, most specifically by critically analyzing and collectively summarizing RCTs (Starr, Chalmers, Clarke, & Oxman, 2009). The Cochrane Collaboration was formed in 1992. Whereas the Cochrane Collaboration focuses on health sciences reviews, the Campbell Collaboration is a research network that promotes evidence production and dissemination regarding the social sciences (https://www.campbellcollaboration.org).

Geretsegger, Elefant, Mössler, and Gold (2014) assessed the effects of music therapy interventions for children with autism spectrum disorder. The authors analyzed a total of 10 RCTs. Results suggest music therapy interventions promoted development in social interaction, social reciprocity, and communication, with either moderate or low quality evidence.

Narrative synthesis

Narrative synthesis, or “textual narrative synthesis” (Barnett-Page & Thomas, 2009, p. 7), provides researchers a way to compare and contrast diverse studies through the language used. Narrative synthesis “relies primarily on the use of words and text to summarize and explain the findings of the synthesis” (Popay et al., 2006, p. 5). While statistical analysis may be included in narrative synthesis, the study must include how and why an intervention works. Popay et al. (2006) note that narrative synthesis, as a form of integrated storytelling, “is one of the ways in which the gap between research, policy and practice can start to be bridged” (p. 5), and stakeholders and policy makers may locate greater resonance with research results. Narrative synthesis can help clarify the context and characteristics of a study, promoting implications for theory and future research (Barnett-Page & Thomas, 2009). Popay and colleagues (2006) provided guidance for narrative synthesis through the following four steps: (a) developing a theoretical model, (b) developing a preliminary synthesis, (c) exploring relationships in the data, and (d) assessing the robustness of the synthesis product. Interested readers can refer to the Popay article for more details about these steps. McDermott, Crellin, Ridder, and Orell (2013) followed these four steps to assess music therapy interventions with people who have dementia. Results noted the commonality of singing amongst studies, and that short-term improvement with mood and behavior were consistent. However, the authors noted the need for strong, long-term studies and more developed theoretical frameworks.

The origin of narrative synthesis appears unclear (Hanson-Abromeit & Sena Moore, 2014). Popay et al. (2006) provided a set of guidelines that was later tested by Rodgers, Sowden, Petticrew, Arai, and Roberts (2009). Results suggested that the guidelines increased transparency and promoted the original functions of narrative synthesis.

Meta-synthesis

Like meta-analysis, meta-synthesis can be understood as an example of a procedure that uniquely contributes to a systematic review methodology. Where meta-analysis assesses efficacy or effectiveness through statistical analysis of quantitative data within studies, meta-synthesis analyzes qualitative data to assess emergent phenomena (Walsh & Downe, 2005), or to construct and communicate a new theory (Khan et al., 2011). Research questions may (a) provide a broad description of a phenomenon, (b) build theory, (c) reconceptualize and explain theory, or (d) any combination thereof (Noblit & Hare, 1988). Researchers compare and contrast studies by locating, evaluating, and translating the following: phrases, keywords, concepts, metaphors, and relations (Walsh & Downe, 2005).

Within the meta-synthesis procedure, one may choose to approach analysis through reciprocal translation (analyzing similarities), refutational translation (analyzing contrasts), or an emerging approach specific to meta-synthesis known as ecological triangulation (Gough et al., 2012). Ecological triangulation employs two concepts: ecological perspective, and triangulation. Ecological perspective acknowledges a mutually interdependent system of behaviors, individuals, and environments that can play a role in intervention success or failure (Barnett-Page & Thomas, 2009); objective results are balanced with individual and social perspectives. Triangulation refers to looking at a phenomenon from “multiple vantage points” (Banning, n.d., p. 1), including multiple researchers, multiple sources of data, or the consideration of multiple theories during analysis.

The author located one meta-synthesis specific to music therapy. Solli, Rolvsjord, and Borg (2013) analyzed service users’ experiences of music therapy within mental health settings. Findings suggested four themes: (a) having a good time, (b) being together, (c) feeling, and (d) being someone; the authors concluded that these themes reflect music therapy as a recovery-oriented practice.

Meta-synthesis originated in healthcare (Khan et al., 2011). Stern and Harris (1985) appear to have been the first to use the phrase “qualitative meta-synthesis.” Jim Banning authored ecological triangulation in order to better understand complex interventions, originally for youth with disabilities.

Meta-narrative review

Meta-narrative review looks at problems from a historical and philosophical perspective. Meta-narrative reviews “often look historically at how particular research traditions or epistemic traditions have unfolded over time and shaped the ‘normal science’ of a topic area” (Wong, Greenhalgh, Westhorp, Buckingham, & Pawson, 2013a, p. 6). This historical positioning promotes “sense making of a complex (and sometimes controversial) topic area” (Wong et al., 2013a, p. 6). For example, Murad, Chatterley, and Guiruis (2013) assessed the quality of communications between pharmacists and patients. The overwhelming majority of studies presented evidence of a biomedical model for communication, while eight studies presented a patient-centered focus. Authors discuss how the chosen methodology of each primary study played a role in the type of communication model.

Meta-narrative review originated within the health sciences (Wong et al., 2013a). Greenhalgh, Robert, Macfarlane, Bate, and Kyriakidou (2004) developed the meta-narrative review to examine a problem from a multiplicity of viewpoints, assumptions, and contexts. Wong et al. (2013a) suggested preliminary publication standards for meta-narrative reviews, known as Realist and Meta-narrative Evidence Syntheses: Evolving Standards (RAMESES).

Critical interpretive synthesis

Critical interpretive synthesis (CIS) focuses on theory development through a socio-critical perspective. CIS “generates a coherent and illuminating theory of a body of evidence that is based on a detailed critical study of the theoretical contribution of the evidence” (Gough, Oliver, & Thomas, 2012, p. 258). Paramount to CIS is the inclusion of the reviewer’s voice as a critical guide. CIS focuses on inclusion and exclusion criteria related to theoretical quality of studies, rather than methodological quality. The primary methods in CIS include formulating the question, searching the literature, sampling, determination of study quality, data extraction, interpretive synthesis including critique, and reporting results (Dixon-Woods et al., 2006). For example, Comte (2016) examined music therapy with refugees who have experienced trauma. The author synthesized and critiqued the following themes: (a) an initial, anchoring theme of neocolonialism, a subsequent theme communicating “refugees” as a homogenous group, and the notion of improvisation as a universal language.

Critical interpretive synthesis originated in the 21st century. Mary Dixon-Woods and colleagues proposed CIS (2006), largely as an outgrowth of meta-ethnography and grounded theory methodologies. Critical interpretive synthesis historically aligns with critical theory and social justice perspectives and uses rigorous techniques to promote synthesis.

Thematic synthesis

Thematic synthesis explores whether clients need an intervention, whether an intervention type is appropriate to addressing a need, and whether clients find the intervention acceptable (Thomas & Harden, 2008). This methodology can thematically identify potential barriers and facilitators to intervention success. For example, a medical intervention may be less successful in some contexts, perhaps due to cultural norms or socioeconomic status. Rathbone, Todd, Jamie, Bonam, Banks, and Husband (2016) conducted a systematic review and thematic synthesis to assess patients’ experiences regarding adherence to medicine. The authors located both descriptive themes and analytic themes, showing that cost-benefit considerations of medicine took into account patient identity as mediated by interactions with professionals, friends, and family.

Thomas and Harden (2008) developed thematic synthesis. The methodology adapts ideas from meta-ethnography and grounded theory. From meta-ethnography, thematic synthesis borrows thematic analysis and higher order interpretations. From grounded theory, thematic analysis borrows the constant comparison method.

Framework analysis and best-fit framework synthesis

Framework analysis, also known as “framework summary,” promotes the emergence of thematic frameworks through the analysis of qualitative data within studies (Gough et al., 2012). The methodology explores general research questions, analyzes heterogeneous methodologies, and facilitates the emergence of iterative themes (Ward, Furber, Tierney, & Swallow, 2013). Framework analysis consists of five stages: (a) familiarization through immersion in the data, (b) developing a theoretical framework by identifying recurrent and important themes, (c) indexing and pilot charting, (d) summarizing data in an analytical framework, and (e) synthesizing data through mapping and interpreting.

Best-fit framework synthesis is a type of framework analysis that uses thematic analysis techniques and directs focus toward a more specific research question (Carroll, Booth, Leaviss, & Rick, 2013; Dixon-Woods, 2011). Where a framework analysis promotes a framework of themes, a best-fit framework synthesis seeks to develop a theoretical framework. “Best fit” refers to the creation of inclusion criteria specific enough to directly answer the chosen research question or questions. Carroll et al. (2013) described the steps of framework synthesis: (a) systematically identify relevant studies in accordance with the research question; (b) generate a priori theory and appraise quality of studies; (c) code evidence, (d) create new themes through thematic analysis; (e) produce a new framework with developed themes; and (f) revisit literature to explore new relationships and to synthesize a new model. For example, Shaw, Holland, Pattison, and Cooke (2016) assessed patient experiences of prevention programs for cardiovascular disease and diabetes. The authors began with an a priori selected Theoretical Domains Framework to assist with themes. Results for theory development included themes regarding the individual (knowledge, beliefs), and the social (cues, social influence, identity, and context).

Social policy researchers in the UK developed framework analysis (Ritchie & Spencer, 1994). Oliver et al. (2008) provided the earliest example of a best-fit framework synthesis. Health care researchers have increasingly used framework analysis and best-fit framework synthesis.

Qualitative research synthesis (QRS) and arts-informed qualitative research synthesis (AI-QRS)

QRS analyzes “qualitative” studies to synthesize concepts, categories, or themes related to research questions (Major & Savin-Baden, 2010; Meadows & Wimpenny, 2016). QRS recognizes “the social, historical, and ideological context of the (reviewed) research” (Meadows & Wimpenny, 2016, p. 2). AI-QRS study uses the QRS model, while including an artistic interpretation of higher order themes (Savin-Baden & Wimpenny, 2014). Meadows and Wimpenny (2016) assessed the ability of clinical improvisation to promote meaning-making. The authors concluded with an emergent conceptual framework.

Descriptions and citations related to QRS suggest a historical relationship to meta-ethnography and the ethnographic tradition. Major and Savin-Baden introduced the QRS methodology in 2010 in order promote both an inclusive and reflexive perspective within broad literature review. The first AI-QRS publication appeared in 2016 (Meadows & Wimpenny, 2016).

Metasummary

Metasummary, also known as “qualitative metasummary,” provides a unique alternative to typical systematic reviews, in that it synthesizes results of reports, surveys, observational studies, interviews, and a range of intervention studies. This methodology employs methods that aggregate descriptive results, not dissimilar to those that may be used in a survey (Barnett-Page & Thomas, 2009) or in manifest content analysis; “higher frequency findings” promote both statistical validity, as well as promote the claim of having “discovered a pattern or theme” (Sandelowski, Barroso, & Voils, 2007, p. 101). In order to promote sound synthesis, metasummary analyzes and promotes convergence in sampling, data collection, and analysis, as well as calculates frequency and intensity effect sizes (Sandelowski et al., 2007).

Metasummary originated within the field of nursing (Sandelowski et al., 2007). The original study focused on antiretroviral adherence in women who were HIV positive (Sandelowski et al., 2007). Authors originally developed the methodology to address the features of survey findings, but also allowed flexibility to assess a range of study types. For example, Childress (2013) assessed literature related to domestic violence. The author located recurring themes related to effects of violence, the cyclical nature of violence, the normalization of violence, resilience, barriers, and the role of substance abuse. The author also noted cross-cultural differences in experiences, and a lack of study regarding domestic violence with people from particular countries.

Realist synthesis

Realist synthesis seeks to discern what works for whom in what environment. Realist synthesis therefore exists as a theory-building methodology, promoting understanding of how and why an intervention is or is not effective (Gough et al., 2012). The researcher locates connections between a context (e.g. population, location, and other factors), a mechanism (all components of an intervention), and an outcome (CMO). The resulting CMO configurations “can amplify or mute the fidelity of an intervention theory” (Pawson, Greenhalgh, Harvey, & Walshe, 2004, p. iii). Thomas and Thomas (2014) conducted a critical appraisal and realist synthesis of studies discussing safe patient transfer. The realist synthesis component of the study located important programmatic mechanisms for safety: (a) management support, (b) policy, (c) ergonomic assessment of spaces, (d) appropriate transfer equipment, (e) risk assessment, and (f) training.

Realist synthesis originated in health care, first under the names “realistic synthesis” and “realist evaluation” (Pawson & Tilley, 1997, 2004). Pawson and Tilley (1997) introduced the realist synthesis methodology in order to address the design and implementation of services covering a “multiplicity of goals” (p. iii). Wong, Greenhalgh, Westhorp, Buckingham, and Pawson (2013b) promoted realist synthesis guidelines through RAMESES.

Mixed methods systematic review

Researchers may seek to answer complex questions related to both effectiveness/efficacy and impact/meaning (Bradt, Burns, & Creswell, 2013; Pearson et al., 2014). Mixed methods research in general refers to the collection, analysis and integration of two sets of data, one quantitative and the other qualitative, in order to “maximize” findings. A mixed methods systematic review includes three separate syntheses. The researcher first synthesizes quantitative and qualitative data through two broad review methodologies (e.g. meta-analysis and meta-ethnography), and then completes a final synthesis of the two sets of results (Gough et al., 2012). For example, Arrowsmith, Lau-Walker, Norman, and Maben (2016) studied nurses’ perceptions of work role transitions. Synthesis of statistical and narrative data suggested differences between the experiences of novice nurses and more experienced nurses.

Current trends in research promote rigorous inquiry of complex questions. The mixed methods systematic review originated in health care (Harden, 2010). Pearson et al. (2014) offered a specific guide with which to conduct mixed methods systematic reviews.

Tertiary reviews: meta-reviews, umbrella reviews, overviews of reviews, and meta-epidemiology

Broad literature reviews generally exist as a type of secondary literature. However, three review types are considered tertiary review methodologies: meta-reviews, meta-epidemiology, and umbrella reviews (Biondi-Zoccai, 2016; Gough et al., 2012). Tertiary reviews can analyze either secondary reviews or both primary and secondary studies. A primary strength of these tertiary studies is their ability to compile evidence for policy makers (Biondi-Zoccai, 2016). Similar to secondary review types, there exists some lack of clarity regarding each type. Each of these reviews is therefore briefly described below.

Meta-reviews, also known as overviews of reviews, synthesize information found in systematic reviews and other literature (Biondi-Zoccai, 2016). The methodology therefore provides a broader view of a research question, and requires a consistent and coherent body of literature to be worthwhile (Caldwell, Welton, & Ades, 2010). Meta-reviews may assess systematic reviews of similar methodologies, or may assess mixed methodologies (Gough et al., 2012). A meta-review includes the following steps: (a) identify the question, (b) establish the research team, (c) identify the outcomes of interest, (d) literature search and analysis, and (e) presentation of findings, primarily through narrative format. Meta-reviews are often registered with the Cochrane Library (Biondi-Zoccai, 2016).

Kamioka et al. (2014) completed a meta-review of systematic reviews that focused on RCTs. Twenty-one systematic reviews met inclusion criteria, which included 16 Cochrane reviews. Results suggested that music therapy treatment improved global and social functioning in certain mental health disorders, gait in people diagnosed with Parkinson’s disease, general depressive symptoms, and general sleep quality.

Meta-epidemiology is a unique type of meta-review that specifically seeks to better understand cause by examining heterogeneity in otherwise homogenous studies, and reducing bias within those studies (Zhang, 2010). Meta-epidemiology focuses on two related areas: studying patterns of illness incidence and distribution in order to locate causes and effects, and controlling bias and confounding variables within studies (Sterne et al., 2002). There exist three subclassifications: simple meta-epidemiology (study of meta-analyses), meta-meta-epidemiology (study of meta-epidemiological studies), and network epidemiology (study of network meta-analyses) (Biondi-Zoccai, 2016; Trinquart, Dechartres, & Ravaud, 2013).

Umbrella reviews often focus on competing interventions to compare effectiveness, impact, and theoretical development (Biondi-Zoccai, 2016). An umbrella review is a review of systematic reviews that may include both primary and secondary studies on a topic, and that focuses on competing interventions for a specific problem or condition (Grant & Booth, 2009).

Researchers began considering literature reviews and other secondary literature as specific objects of study in the 1980s (Biondi-Zoccai, 2016; Oxman & Guyatt, 1991). Meta-epidemiology appears to have emerged in the literature in 1997 (Naylor, 1997; Zhang, 2010), with the evolution of the methodology beginning in the 21st century (Zhang, 2010). Meta-reviews appear to have been pioneered by the Child Health Cochrane Group (Biondi-Zoccai, 2016). The Cochrane group conducted its first umbrella review in 2015 (Biondi-Zoccai, 2016). Authors have published information regarding methods for tertiary reviews (Biondi-Zoccai, 2016; Smith, Devane, Begley, & Clarke, 2011).

Implications for music therapy practice

Broad literature reviews can employ a range of characteristics. Each methodology seeks to address different needs through specific characteristics. Table 1 provides the research consumer an opportunity to compare and contrast different methodologies. Table 2 provides example research questions, also for the sake of comparison. These two tables may assist in the formulation of research review, either at the student or professional level.

Authors from the field of music therapy have conducted a range of broad review types (Table 3). However, the author could not locate peer-reviewed published articles of particular methodologies; when this information is cross-referenced with Table 1, particular themes arise in regard to theory, policy, information flow, stakeholder engagement, and tertiary reviews.

Theory

To date, there appear to be no peer-reviewed published music therapy articles using the following theory-contributing methodologies: meta-narrative, thematic synthesis, framework analysis/synthesis, realist synthesis, metasummary, or mixed methods. These methodologies may promote the development of stronger conceptual and theoretical frameworks to inform interventions.

Policy development

To date, there appears to be one peer-reviewed, published music therapy article using a quick-review methodology that emphasizes policy/guideline development (Scott & Kidd, 2016). While the field has published Cochrane reviews, stand-alone systematic reviews, and narrative syntheses, there appear to be no studies using rapid review, state of the art review, meta-review, or umbrella review methodologies. Development and proper dissemination of diverse policy-oriented reviews (i.e. scoping review, rapid review, state of the art review) may benefit advocacy efforts.

Information flow and stakeholder engagement

Reviews that promote a better understanding of information flow (mapping reviews) and promote stakeholder engagement may also enhance efforts to empower the profession. Mapping reviews uniquely offer the opportunity to study how research information is gathered and disseminated. Mapping reviews and realist syntheses encourage stakeholder participation on the research team, while many Cochrane reviews appear to require such. Furthermore, other systematic review methodologies can choose to employ stakeholder participation.

Tertiary reviews

Evidence synthesis continues to evolve, and some health care fields are engaging in tertiary reviews. The field of music therapy has primarily pursued updated secondary reviews (Bradt, Dileo, Magill, & Teague, 2016; Geretsegger et al., 2014; Standley, 2012); the author located one meta-review specific to music therapy (Kamioka et al., 2014). The field of music therapy, due to both its youth and diversity, may more frequently employ tertiary reviews as the depth and breadth of the evidence base grows.

The need for effective and efficient evidence summaries continues, and has perhaps increased over time. Health care research continues to proliferate at a quick pace. Given the broad reach of music therapy practice, researchers and consumers benefit from a deeper and clearer understanding of literature review processes.

Conclusion

Research, as a historical act, includes review. Review can either support a primary study or be conducted through a broad methodology. This article discussed narrow and broad literature reviews in the following areas: descriptions, history, research questions, methodological characteristics, and example articles from music therapy and related fields. Implications for music therapy include the ability to compare methodologies for future research, and to envision what types of broad investigations may strengthen the evidence base and advocacy efforts; at this juncture, the past may meet the present and future to address the needs of the field.

Acknowledgment

The author wishes to acknowledge Deanna Hanson-Abromeit, Ph.D., MT-BC, and Cynthia Colwell, Ph.D., MT-BC for their engaged mentorship during the nascent stages of this manuscript.

Additional information

Notes on contributors

Bill Matney

Bill Matney is an assistant professor of music therapy at the University of Kansas. His research line includes the employment of broad literature review methodologies, as well as engagement with theoretical problems related to research and practice.