Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Abstract
This article provides a step-by-step approach to conducting and reporting systematic literature reviews
(SLRs) in the domain of healthcare design and discusses some of the key quality issues associated with
SLRs. SLR, as the name implies, is a systematic way of collecting, critically evaluating, integrating, and
presenting findings from across multiple research studies on a research question or topic of interest. SLR
provides a way to assess the quality level and magnitude of existing evidence on a question or topic of
interest. It offers a broader and more accurate level of understanding than a traditional literature review.
A systematic review adheres to standardized methodologies/guidelines in systematic searching, filtering,
reviewing, critiquing, interpreting, synthesizing, and reporting of findings from multiple publications on a
topic/domain of interest. The Cochrane Collaboration is the most well-known and widely respected
global organization producing SLRs within the healthcare field and a standard to follow for any researcher
seeking to write a transparent and methodologically sound SLR. Preferred Reporting Items for Sys-
tematic Reviews and Meta-Analysis (PRISMA), like the Cochrane Collaboration, was created by an
international network of health-based collaborators and provides the framework for SLR to ensure
methodological rigor and quality. The PRISMA statement is an evidence-based guide consisting of a
checklist and flowchart intended to be used as tools for authors seeking to write SLR and meta-analyses.
Keywords
systematic literature review, healthcare design, evidence based design
As the name suggests, systematic literature study as a single snapshot in time and a longitu-
review (SLR) is a systematic way of collecting, dinal study as a single video. As the number of
critically evaluating, integrating, and presenting snapshots and videos available on a question or
findings from across multiple research studies on topic increases, it offers a better and more accu-
a research question or topic of interest. It is rate grasp on the phenomena of interest. Since a
“systematic” since it adopts a consistent, widely single research is just one snapshot or video,
accepted, methodology. The methodology should
address concerns regarding quality issues, such as
1
bias, replicability, credibility, et cetera. Texas Tech University, Lubbock, TX, USA
2
SLR provides a way to assess the quality level University of Florida, Gainesville, FL, USA
and magnitude of existing evidence on a question
Corresponding Author:
or topic of interest. It offers a broader and more Debajyoti Pati, PhD, FIIA, IDEC, LEED AP, Texas Tech
accurate level of understanding than a traditional University, Lubbock, TX 79409, USA.
literature review. Imagine a single cross-sectional Email: d.pati@ttu.edu
16 Health Environments Research & Design Journal 11(1)
basing design or policy decisions on a single research (Dunn, 1997). By the 1970s, meta-
research is not as robust as basing decisions on analysis was formalized as a process of research
findings from multiple studies. There are, of synthesis (Glass, 1976) and evidence-based policy
course, issues related to the quality of snapshots and practice gained momentum in the 1990s.
or videos, and the viewpoint from which those SLRs are only one type of review of literature.
were taken. Poor quality could result in less clarity The type of review one needs to conduct depends
and, hence, less confidence. Changes in viewpoint on the objective of the review. Broadly speaking,
could offer a different perspective of a phenomena the two most commonly known types of literature
or introduce bias. Even with appropriate measures reviews are the “traditional” or narrative review
for quality, findings from individual studies may and the “gold standard” or SLR (Booth, Sutton, &
vary owing to differences in population, setting, Papaioannou, 2016). Traditional reviews, also
instruments, random chance, and so on. A properly known as conventional or nonsystematic reviews
conducted SLR considers the quality of issues and (Greenhalgh, 2014), are generally faster and eas-
incorporates multiple viewpoints, providing a ier to conduct and are sometimes appropriate due
broader and more complete picture. to a short time frame or lack of resources. Grant
SLRs are particularly gaining importance in and Booth (2009) provide a more comprehensive
the healthcare design field. Starting from about classification of 14 different types of reviews in
600 published studies in 2004 (Ulrich, Zimring, the fields of medicine and library sciences. In gen-
Quan, Joseph, & Choudhary, 2004), the number eral, reviews differ in the question being asked,
of studies in the field has increased to a level that including, among others: What is the current the-
facilitates SLRs on important questions. This oretical framework/understanding on a topic/
growing body of knowledge is making SLRs in domain—based on either just peer-reviewed pub-
the healthcare design field both possible and cri- lications or all sources? What are some gaps, in
tically important. From single-patient rooms to theory or evidence, in existing published litera-
decentralization, there is a wide range of topics ture? Is there sufficient justification for conducting
of importance to efficiency, safety, care quality, a study? What is the most appropriate study ques-
cost, patient/family/staff experience, and so on, tion/topic? Which research question/topic will pro-
which could benefit from SLRs. Articles on SLRs vide the best return on investment? What is the
are available in other fields of studies. The pur- effect size of an intervention, based on quantitative
pose of this article is to provide a step-by-step findings from previous studies? How did a certain
approach to conducting and reporting SLRs in the past intervention (design, policy, practice, etc.)
domain of healthcare design and discuss some of work? What is the volume of literature available
the key quality issues associated with SLRs. on a topic/domain of interest?
The key variations among these review types are
The purpose of this article is to provide a in the primary audience group being targeted
step-by-step approach to conducting and (intended end use of the review), scope, level of
reporting SLRs in the domain of interpretation, level of evaluation of individual
healthcare design and discuss some of the study quality, type of information included (pub-
key quality issues associated with SLRs. lished literature vs. other information sources), type
of data included (qualitative, quantitative, or both),
depth of analysis of published data/findings, level
of synthesis, level of investigator bias, level of pub-
Types of SLRs lication bias, extent of time commitment by the
SLR began within the realm of healthcare and can review team, and the type and amount of resources
be traced back to the work of Scottish naval surgeon needed to conduct a review, among others.
James Lind who, in 1753, conducted the first ran- Among these 14 different review types articu-
domized controlled trial (RCT) and recognized sys- lated by Grant and Booth (2009), 3 are of relevance
tematic methods as a key component in avoiding to the current discussion, namely: (1) meta-analysis,
bias, which remains a pivotal goal of systematic (2) qualitative systematic review or qualitative
Pati and Lorusso 17
evidence synthesis, and (3) systematic review. Key Steps in Conducting SLRs
These three types of reviews share several common
The first and most crucial step in conducting
attributes: need for systematic search, need for com-
an SLR is to familiarize oneself with generally
prehensiveness, need for study quality assessment,
accepted standards on the subject. Today, the
need for interpretation, and, most importantly, the
Cochrane Collaboration is the most well-
need for transparency of the methodology such that
known and widely respected global organization
the study can be replicated. The variations between
producing SLRs within the healthcare field and a
the three types of review originate from the long
standard to follow for any researcher seeking to
history and maturation of science and publication
write a transparent and methodologically sound
in medicine and library sciences, resulting in more
SLR. Founded in 1993, Cochrane consists of
targeted review types.
over 37,000 contributors from more than 130
Meta-analysis uses statistical analysis proce-
countries who make up over 50 subject review
dures on quantitative data to estimate the effect
groups (Cochrane, n.d.). Together, this network
size of a specific intervention, where the input
of contributors has transformed the healthcare
data are the end products of the statistical proce-
industry by publishing SLRs that gather and
dures of a number of studies with vastly similar
summarize high-quality evidence to aide
questions, setting, population, intervention, and
researchers, physicians, caregivers, and others
time period. The authors believe that the health-
in making well-informed decisions that positively
care design research field does not currently
impact human health globally. Cochrane provides
enjoy sufficient volume of quantitative studies
guidance on methodological standards which are
with the above similarities, which may, however,
available at methods.cochrane.org/mecir and are
change as the field matures. The intent of a meta-
compliant with the Preferred Reporting Items for
analysis is similar to those of a systematic review.
Systematic Reviews and Meta-Analysis
The objective of a qualitative systematic review
(PRISMA) which provides the framework for
or a qualitative evidence synthesis is similar to
SLR (Moher, 2009; PRISMA, n.d.).
meta-analysis, except that these reviews deal pri-
PRISMA, like the Cochrane Collaboration,
marily with qualitative studies, and hence the out-
was created by an international network of
comes and resulting end uses are different.
healthcare-based collaborators seeking to
A systematic review adheres to standardized
strengthen and streamline the methodological
methodologies/guidelines in systematic searching,
rigor and quality of SLRs. First published
filtering, reviewing, critiquing, interpreting, synthe-
in 2009 as the PRISMA statement, it is an
sizing, and reporting of findings from multiple pub-
evidence-based guide consisting of a checklist
lications on a topic/domain of interest. It attempts to
and flowchart intended to be used as tools for
capture the broadest set of available literature on the
authors seeking to write SLR and meta-analyses
topic of interest. After assessing the quality of indi-
(Moher, 2009; PRISMA, n.d.). Several exten-
vidual studies, SLRs may eliminate low-quality
sions to the original document were published
studies from further consideration. Moreover,
in 2015 and can all be found at www.prisma-sta
owing to the extensive documentation and reporting
tement.org/extensions/default.aspx. Although
of the steps and assumptions, SLRs renders itself
originally created for the review of randomized
amenable for replication. It demands a team effort
trials, PRISMA is broadly applicable to myriad
(at least two) to eliminate bias, among other issues.
types of research. The actual SLR then proceeds
SLRs are inherently time intensive. Owing to the
systematically through various sequential steps,
aforementioned quality demands, SLRs is regarded
including question formulation, team formula-
as the most robust level of evidence available on a
tion, identification of search domains and publi-
topic or question of interest.
cation sources, systematic search, systematic
SLRs is regarded as the most robust level critical analysis, systematic interpretation, and
of evidence available on a topic or systematic reporting. Table 1 outlines key steps
in the SLR process.
question of interest.
18 Health Environments Research & Design Journal 11(1)
(continued)
Pati and Lorusso 19
Table 1. (continued)
11 Within the JCR, create a tab for each database, including snowballing, hand You can use viewing panes
searching, and gray literature on their own tabs, if included in the search. within Excel to help
Organize headers for relevant information including author name, abstract, condense the cells to only
publication journal, DOI, volume, year of publication, etc. Include columns at what is relevant during this
the right pertaining to your exclusion criteria that can be answered by yes or stage (abstract and
no. Remove any duplicates by striking through and hiding the cells. This way if screening questions).
you make a mistake, the information is still there and you can reincorporate
it into your JCR if needed. In case you are not sure whether to include or
exclude an article based on the title, either retain it or consult the other
team members.
12 Abstract review: Up to this point you have screened only titles, so now you Note, once you have deleted
will begin to read through the abstracts of each article and screen them articles from inclusion in
against the exclusion criteria. Instead of inserting “yes” or “no,” it is helpful the JCR you can remove
to create separate columns for the two and insert a number “1” (for yes or them from your reference
for no, as applicable) so that you can run a sum total at the bottom of each citation manager for
spreadsheet to keep track of how many duplicates were removed, how consistency.
many were excluded and how many were included per database. Once you
have completed this for each separate database, create a tab for all included
articles. Copy and paste all the included articles into this sheet and sort by
title to quickly find and remove any remaining duplicates. This information
will be important data for the PRIMSA flowchart.
13 Abstract phase screening quality check: To strengthen validity and reliability You may review the research
for your search results, work with at least one other researcher to question, exclusion
conduct an inter-rater reliability assessment. For instance, you can give criteria, and definitions
the second reviewer 20% of the total number of abstracts you reviewed, again to ensure uniformity
randomly chosen, and leave your exclusion fields blank so that the second of understanding.
reviewer can run a parallel screening process on the subset of abstracts.
Compare your findings and aim for at least 90% agreement. Discuss any
areas of disagreement with the third team member reviewer until you
reach an inclusion/exclusion agreement.
13 On the last tab in the Excel spreadsheet, organize your process totals in the PRISMA flowchart can be
JCR and update your PRISMA flowchart with the information. Keep the found at http://www.prisma-
JCR as a tool and record of your process. statement.org/
14 Critical review of full articles: Once you have the list of final included articles,
conduct in-depth critical review of each article. Extract all relevant
information from each article and insert in the appropriate place in the Excel
document. As part of the critical review, rate the level of quality of evidence
for each article using the framework you have adopted for your study.
15 Critical review assessment quality check: To strengthen validity and reliability You may review the research
for your critical review findings, work with at least one other researcher to question, the quality of
conduct an interrater reliability assessment. For instance, you can give the evidence scale, quality
secondary reviewer 20% of the total number of articles you critically issues, and definitions again
reviewed, randomly chosen, and leave your assessment fields blank, so that to ensure uniformity of
the second reviewer can run a parallel critical evaluation process on the understanding.
subset of articles. Compare your findings and aim for complete agreement.
Closer the agreement to 100%, more robust is the critical review process.
A level of at least 90% agreement may be acceptable. Discuss any areas of
disagreement with the third team member.
(continued)
20 Health Environments Research & Design Journal 11(1)
Table 1. (continued)
16 Update your PRISMA flowchart with any changes from the inter-rater You can also calculate
reliability analysis. Now you will create a new, clean spreadsheet that will interrater reliability with
become your matrix of included studies. This spreadsheet will be simpler the level of quality of
than the JCR with headings that are tailored specifically to the purpose of evidence to further
your overall SLR. These might include first author, year, level of evidence, strengthen validity and
methods, participants, environmental intervention, outcomes, setting, reliability of your SLR.
assessments, results, etc. This will likely be published with your article
and should be well organized. Now you are ready to compose your
report, following the guidelines from the Cochrane Collaboration and the
PRISMA checklist, and using Table 2.
Note. SLRs ¼ systematic literature reviews; PRISMA ¼ Preferred Reporting Items for Systematic Reviews and Meta-Analysis.
A Suggested Model for Introduction. The two critical elements within the
Systematically Composing introduction are the author’s rationale and objec-
tives. The rationale for the review should be
an SLR Report
written in a way that explains the topic in light
Key Sections in Writing an SLR Following the of what is currently understood in the existing lit-
PRISMA Checklist erature. The objectives should follow and clearly
identify the question(s) to be addressed and how
Contents within the 27 PRISMA checklist items they relate to the various concepts of interest. In
include guidelines regarding key steps in the pro- interventional studies, a typical articulation of con-
cess and are broken into title, abstract, methods, cepts of interest is participants/interventions/com-
results, discussion, and funding related to the SLR parisons/outcomes (PICO) and study design
(Moher, 2009; PRISMA, n.d.). According to the (Figure 1). The PICO process comes from the field
PRISMA checklist, authors should first clearly of evidence-based medicine (EBM) to address a
identify the report as a systematic review, meta- healthcare-based question. In SLRs, PICO can be
analysis, or both. This is the only aspect of PRISMA used to frame the search terms and key words to
not required by the Cochrane Collaboration but is a the specific topic being studied (Centre for
good rule to follow for healthcare design research- Evidence-Based Medicine, 2014; Centre for
ers who wish to make the nature of their work Reviews and Dissemination, 2009; Lorusso &
easily identifiable for publishers and readers. The Bosch, 2016). Investigators should identify or
following is a summary of the PRISMA checklist, develop their specific list of concepts of interest,
which can be found and downloaded at http:// such as population/setting/issue, based on the
www.prisma-statement.org/ for use as a guide. study objective and report it in the introduction.
Abstract. The abstract of the report should consist Methods. The methods section should outline the
of a detailed summary including the following ele- overall process followed within the SLR. It is
ments, as applicable to each report: background, important to maintain clarity and transparency
objectives, data sources, study eligibility criteria, throughout the description of the process fol-
participants and interventions, study appraisal and lowed in the SLR in order to reduce bias and
synthesis methods, results, limitations, conclusions improve the translation of the overall findings.
and implications of key findings, and inclusion of This section will include information on the pro-
a systematic review registration number if appli- tocol and registration, eligibility criteria, infor-
cable. The overall goal is to be clear and concise in mation sources, search strategy, study selection
the abstract so as to transparently communicate the and data collection process, data items studied,
contents of the paper to readers. risk of bias in individual studies, a summary of
Pati and Lorusso 21
measures, synthesis of results, risk of bias across publication, publication date, volume, abstract,
studies, and any additional analysis. This section etc.). It is often used as a tool by the author or
provides an overview explanation preceding the team writing the SLR to rate the found articles
results section which will provide information against the inclusion criteria and keeps a record of
culminating from a deeper dive into many of the duplicates, excluded and included articles. Data
same topics for the studies included in the final items are listed within the methods section and
review. Protocol and registration, if it exists and include variables and all assumptions or simplifi-
where it can be found, should be reported. Elig- cations made by the SLR author(s). Risk of bias
ibility criteria should be provided of included in individual studies is provided and discussed in
studies such as study characteristics, time frame terms of how the risk was assessed for each article
of studies, language, publication, and so on, and a and how it will be used in the data synthesis of the
rationale for why the author used these specific SLR. A summary of measures should be outlined in
eligibility criteria. Information sources included terms of the primary measures used within the arti-
in the review should be clearly described. These cles included in the SLR. This helps to clarify the
are typically the databases targeted for the review overall analysis and helps readers understand how
with an explanation of the dates used within the outcomes were achieved. A synthesis of results is
review and an explanation of any contact with explained by a description of the methods used by
authors for additional included studies. The the SLR author(s) to review data and evaluate the
search strategy should be clearly defined for one level of evidence. Risk of bias across studies is
or more databases. This should be provided in discussed that may impact the overall findings. This
great detail to allow for future updates or replica- is different from the risk of bias within individual
tion and should include any limits used by the studies because it discusses a broader range of risk
author. The study selection process should be out- including publication bias or limited reporting of
lined regarding screening of findings, ratings of data throughout studies. Additional analyses are
eligibility, and a breakdown of what articles were important particularly in SLRs, which might
included in the review. This is supported by the include sensitivity or subgroup analyses, meta-
PRISMA flowchart, which provides a clear, gra- regression, and so on. It may also include any other
phic representation of the process from included additional analysis conducted by the SLR author or
articles through to the final selection of included team relevant to the synthesis of findings.
articles (Figure 2). A template is available at
http://www.prisma-statement.org/PRISMAState Results. The results section of the SLR is presented
ment/FlowDiagram.aspx in descriptive, graphic, and detailed tabular for-
The data collection process and precise mats. Combined, these representations should
method used by the authors to gather information carefully describe the study selection, characteris-
from the literature should be clearly provided. tics, risk of bias within and of individual studies,
This is often done through an excel document synthesis of results, risk of bias across studies, and
known as the journal citation report (JCR) that any additional analysis of findings. Whereas the
collates information extracted from reference methods section provides an introduction, and
managers in an organized way (author name, outline of how these topics will be discussed, the
22 Health Environments Research & Design Journal 11(1)
Figure 2. Example of Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) flow
diagram.
results section provides great detail specifically on and so on. This information is supported graphi-
the articles included in the final review. Study cally in a matrix of the included studies. Risk of
selection, supported by the PRISMA flowchart, bias within studies is identified with outcome-level
provides detailed information on the number of assessments. Results of individual studies are pro-
studies screened, assessed for eligibility, and vided for all outcomes including information on
excluded or included for review with an explana- the intervention and any group differences. This
tion for exclusion throughout the process. Study can be done in written format, separated into com-
characteristics are provided for each included arti- mon themes or areas of affinity, and supported
cle along with information on the data that were graphically by the SLR in a matrix of the included
extracted. This could include participant or site studies. A synthesis of results will provide infor-
descriptions, intervention, method, time frames, mation on meta-analysis of included articles and
Pati and Lorusso 23
assessed levels of evidence. Risk of bias across elements are best achieved by following the
studies and additional analysis provides informa- Cochrane Collaboration’s guidelines and the
tion on broad risk of bias across all included stud- PRISMA statement checklist closely. Adhering
ies and detail any additional analysis conducted on to these frameworks helps ensure a level of uni-
included literature. formity in the SLRs, supporting transparency, and
reducing overall risk of bias. This uniformity
Discussion. Lastly, the discussion section of the incorporates key elements including an assessment
article will bring all the information together into of the evidence, the concepts of interest and
a final summary of the evidence, listing limita- PRISMA flow diagrams, a JCR, and a matrix of
tions of the SLR and offering conclusions stemming included studies.
from the review. In healthcare design SLRs, this Both the quality level and magnitude of evi-
section may also include a description of possible dence need to be analyzed in an SLR. The quality
implications for practice that can be used by health- level of evidence pertains to the quality of an indi-
care designers, researchers, and administrators in vidual study. The magnitude of evidence pertains
future applications. The summary of evidence will to the volume of quality evidence available on a
restate the main findings and discuss the overall topic or question of interest. Referring back to our
quality level and magnitude of evidence and how analogy of snapshots in the introduction, the qual-
each finding relates to key user groups within the ity level of evidence can be viewed as the quality
healthcare industry. A discussion of limitations is of each photograph whereas the magnitude of evi-
an important component of the SLR to overcome dence can be viewed as the number of good quality
bias. All limitations including any risk at any level photos available for stitching together.
of the review process should be identified. All The purpose of critically assessing the evidence
reviews will have some form of risk of bias and so in an SLR is important because it provides the
this is not a section that should be avoided or seen as reader with a description of its quality. This assess-
a weakness to the overall report. To the contrary, an ment is often tied directly to the inclusion/exclusion
open and straightforward discussion of any risk of criteria when an author will state that only certain
bias serves to increase the credibility of an SLR. quality of evidence will be included. This may be
done in an SLR of a topic where the author(s) goal is
Conclusions. The concluding section of the discus-
to report only the highest quality of evidence and
sion will include the author(s) perspective on the
there is enough available literature to be selective.
findings and their role related to other reported find-
Sometimes, however, this is not possible particu-
ings and how they may impact future research or
larly in the healthcare design domain, currently,
healthcare design overall. Any sources of funding
because the body of knowledge regarding a topic
should be reported including a description of any
may not yet be so extensive as to allow it. Whatever
other support from sources supplying data and the
the decision of the author or team conducting the
subsequent role of funding sources related to the SLR.
SLR, it is important to rate the quality level of evi-
Table 2 provides a suggested step-by-step,
dence. Examples can be found in EBM (Balshem
generic model for composing an SLR report. It
et al., 2011) and also in healthcare design literature
includes important questions that need to be
(Hamilton, 2011; Marquardt, Bueter, & Motzek,
responded to in each section of an SLR report.
2014; Pati, 2011; Stichler, 2010) and are commonly
One may ignore questions that do not apply to
described in a simple table format outlining the
one’s specific SLR study.
descriptions of each quality level.
Other key elements within the review are the
concepts of interest (such as PICO) and PRISMA
Explanation of Key Elements Within
flowchart both of which relate to the overall
the Review search strategy of the SLR. It is important to note
As discussed earlier, critical factors separating that when designing the specific search strategy
SLR from traditional reviews are clarity, validity, for each SLR, it is incredibly valuable to enlist the
and auditability (Booth et al., 2016). These help of a research librarian with expertise within
24 Health Environments Research & Design Journal 11(1)
Main
Sections Subtopics Key Questions to Respond to
(continued)
Pati and Lorusso 25
Table 2. (continued)
Main
Sections Subtopics Key Questions to Respond to
Discussions Summarization What new information was generated from this systematic review?
What is the general level (quality and magnitude) of evidence in various
thematic areas? Which of the thematic areas enjoy higher quality and
magnitude of evidence? Which areas currently lack quality and/or
magnitude of evidence?
Limitations What are the limitations of this SLR?
Future studies What topics/questions should be examine in future studies?
Conclusions What are the key take-away messages of this SLR for your target readers?
References [include complete list of articles reviewed and/or cited in the SLR, denote
articles included in the SLR matrix with an asterisk*]
Appendix [as appropriate]
the specific topic to be reviewed. Research librar- marked with an asterisk to clearly identify them for
ians with a focus on healthcare, for example, have the reader. The matrix will be tailored specifically to
extensive knowledge of the broad range of liter- each SLR based on the unique objectives of the
ature and are adept in assisting in the develop- paper. Typical components of the matrix include the
ment of the PICO model serving as a guide for the last name of the first author and year of publication,
search terms and key words relating the patient, assigned quality level of evidence, methods, parti-
population or problem, intervention, and outcome cipants’ description, intervention summary, list of
to be evaluated in the SLR, if the SLR is focused outcomes, description of setting and assessments,
on interventional studies. Once established, the and a brief recap of the results of each study. The
PICO search model for the SLR is often illu- intention of the matrix is to serve as a tool providing
strated with a table and it is helpful to document an easily digestible format for readers to understand
the exact key words and terms used in a separate the purpose and outcomes of the SLR.
appendix for future revision or replication of the
search. In case of SLRs not restricted to interven-
tional studies, librarians can also assist in devel- Important Issues in
oping the right search terms for population,
Conducting SLRs
settings, and issues of interest. The PRISMA flow
diagram is also used to document the search strat- There are several issues that warrant close atten-
egy. A template is available from the PRISMA tion while conducting and reporting an SLR. The
statement’s website that can be edited and key question one should ask while conducting an
inserted directly into a Word document. The SLR is whether the intended audience will con-
flowchart is simple and easy to follow and pro- sider the findings credible (irrespective of the
vides a consistent way to communicate the results actual nature of SLR findings). If trust is not gen-
of the search process from the initial findings, erated among the readers, then findings may not be
inclusion of gray or snowballed literature, exclu- perceived as credible, which will be counterpro-
sions, and final inclusion counts for the SLR. ductive to the objectives of the SLR study. The
One of the most important elements within the following subsections outline some of the issues
SLR is the matrix of included studies. The matrix is that investigators need to accord special attention
a tabular representation of the results section of the to in order to generate trust and confidence.
paper made in a spreadsheet document that sum-
marizes the details of each included study and the The key question one should ask while
overall findings of the SLR. It is important to note conducting an SLR is whether the intended
that the articles included in the final review are audience will consider the findings
included in the reference list of the paper and are credible.
26 Health Environments Research & Design Journal 11(1)
Term Definition
(continued)
28 Health Environments Research & Design Journal 11(1)
Table 3. (continued)
Term Definition
Appendix
Table A1. Search Strategy Template.
Table B1. Primary Literature Searching Template. guidelines: 3. Rating the quality of evidence. Journal
Population Intervention Outcome of Clinical Epidemiology, 64, 401–406.
Concept: Concept Concept Booth, A., Sutton, A., & Papaioannou, D. (2016). Sys-
tematic approaches to a successful literature review
Mesh/ (2nd ed.). Los Angeles, CA: Sage.
subheadings Centre for Evidence-Based Medicine. (2014, June 10).
Text words
Asking focused questions. Retrieved February 22,
Note. Database: (Do this for each of your databases and gray 2016, from http://www.cebm.net/asking-focused-
literature). questions/
Centre for Reviews and Dissemination. (Ed.). (2009).
PubMed Search Strategies CRD’s guidance for undertaking reviews in healthcare
(3rd ed.). York, England: York Publishing Services.
POPULATION (Example below, do the same Cochrane. (n.d.). Cochrane Collaboration. Retrieved
for intervention/outcome as well) October 10, 2017, from www.cochrane.org/about-us
(“Dementia/drug therapy”[Mesh] OR “Dementia/ Dunn, P. M. (1997). James Lind (1716-94) of Edin-
rehabilitation”[Mesh] burgh and the treatment of scurvy. Archives of Dis-
INTERVENTION ease in Childhood—Fetal and Neonatal Edition, 76,
OUTCOME F64–F65. Retrieved from https://doi.org/10.1136/
Yield: fn.76.1.F64
Duplicates Removed: Glass, G. V. (1976). Primary, secondary, and meta-
Total Added analysis of research. Educational Researcher, 5,
3–8. Retrieved from https://doi.org/10.3102/
0013189X005010003
Declaration of Conflicting Interests
Grant, M., & Booth, A. (2009). A typology of reviews:
The author(s) declared no potential conflicts of An analysis of 14 review types and associated meth-
interest with respect to the research, authorship, odologies. Health Information and Libraries Jour-
and/or publication of this article. nal, 26, 91–108.
Greenhalgh, T. (2014). How to read a paper: The
Funding basics of evidence-based medicine (5th ed.). Chi-
The author(s) received no financial support for chester, England: John Wiley.
the research, authorship, and/or publication of Hamilton, D. K. (2011). What constitutes best practice
this article. in healthcare design? Health Environments
Research & Design Journal, 4, 121–126. Retrieved
References from https://doi.org/10.1177/193758671100400210
Balshem, H., Helfand, M., Schunemann, H., Oxman, A., Healey, J. (2014). Statistics: A tool for social research
Kunz, R., Brozek, J. . . . Guyatt, G. (2011). GRADE (10th ed.). Belmont, CA: Wadsworth.
30 Health Environments Research & Design Journal 11(1)