The DNP Project: Literature Synthesis ?For this discussion post, you will practice writing your: 1) Literature Search Methodology, and 2) Provide an example of one full

Created by Cosette Taylor, Communications Instructor for the Faculty of Nursing at the University of

Manitoba, [email protected]

Can you give me

examples of the process

of analysis to synthesis

of the literature?

When your kitchen table is covered in a mountain of research literature and you do not know what to do next, an example of the process of analysis to synthesis can be helpful in getting you started. I have provided these examples below.

The thinking at this stage consumes a lot of energy and time (more hands would not necessarily be helpful, trust me). How you interpret the significance of the literature is what’s most important. You will do more than simply reporting the findings (facts); you will make statements about what is “known” or “not known” (the gaps) on the topic and identify the controversies among the academics who research the topic.

While reading through your sources, you first need to determine the:

• general trends or themes in the literature

• similar findings in some of the different studies

• contrasting findings among the different studies.

Below, the original sources are provided and the successful synthesis of these findings. Read through these examples carefully. Next go to the sample literature reviews found on this site.

Reporting General Trends in the LiteratureReporting General Trends in the LiteratureReporting General Trends in the LiteratureReporting General Trends in the Literature

Original sourcesOriginal sourcesOriginal sourcesOriginal sources to analyzeto analyzeto analyzeto analyze

In Davis, Hershberger, Ghan, and Lin (1990), you found that:

“In this study about the personal qualities for a good nurse, caring was reported 84% of the time in good

nurses and kindness was reported 80% of the time”.

In Resnick (2002), the patient shared that she:

“… was fortunate to have a positive experience with my nurse. The nurse was thoughtful and empathetic.

Created by Cosette Taylor, Communications Instructor for the Faculty of Nursing at the University of

Manitoba, [email protected]

The nurse comforted me, listened to my fears, shared the experiences of other women in this situation with

me, and gave me information when I needed it. This nurse touched my heart and made the illness easier”

In their study, Rush and Cook (2006) reported that:

“Five hundred and twenty five comments were recorded on the main requirements of a good nurse. More

than 70 percent reported that ‘having a caring attitude’ or ‘caring nature’ was extremely important for the

nurse. Taking time to listen and talk with patients was also associated with caring. Many comments were

about ‘what a good nurse was not’ including someone who ignores patients, makes jokes about patients,

shouts at patients, for instance. Thus, being caring and kind to patients is integral to being a good nurse”

Smith and Godfrey (2002) claim that:

“The positive attitudes of the good nurse are linked to attitudes of the nurse as a person. Many good nurses

are caring, kind, and compassionate people. A good nurse ‘truly cares about people’, ‘likes the patients and

wants to help them’, ‘listens to the patients carefully and respectfully’, and ‘sees the person, not just the

disease’”.

Successful Synthesis:

The literature consistently states that the important personal characteristics of a good nurse are related to the nurse’s ability to be caring and kind to patients (Davis, Hershberger, Ghan, & Lin, 1990; Resnick, 2002; Rush & Cook, 2006; Smith & Godfrey, 2002).

Two or three sources with the saTwo or three sources with the saTwo or three sources with the saTwo or three sources with the same meaning or main ideame meaning or main ideame meaning or main ideame meaning or main idea

Original sources to analyzeOriginal sources to analyzeOriginal sources to analyzeOriginal sources to analyze

Ruth and Cook (2006) found that:

“The strongest theme in the research was communication. There were 102 comments on communication

with particular emphasis on listening skills and taking the time to communicate with patients. Another

important theme was knowledge. While specific nursing knowledge was important, the participants reported

that good nurses use their common sense to make good decisions about care (and then communicate

those decisions well)”

Created by Cosette Taylor, Communications Instructor for the Faculty of Nursing at the University of

Manitoba, [email protected]

Smith and Godfrey (2002) also came to the conclusion that:

“… good nurses make good decisions and appropriate judgments about their patients’ care. This requires

that the nurse should continually ask questions, clarify information, and consider a variety of options before

choosing the best, most logical, decision. Often the nurse will need to consider very different perspectives

and opinions. Therefore, the nurse must be able to effectively communicate the reasons why he/she made

this particular decision”.

Successful Synthesis:

The ability to communicate and have common sense are also desirable characteristics of a good nurse (Rush & Cook, 2006; Smith & Godfrey, 2002).

Or

Rush and Cook (2006) and Smith and Godfrey (2002) argue that the ability to communicate and have common sense are also desirable characteristics of a good nurse.

Two or three sources with Two or three sources with Two or three sources with Two or three sources with some some some some contrasting or differcontrasting or differcontrasting or differcontrasting or differinginginging ideaideaideaideassss

Original sources to analyzeOriginal sources to analyzeOriginal sources to analyzeOriginal sources to analyze

Davis, Hershberger, Ghan, and Lin (1990) report that:

“In this study about the personal qualities for a good nurse, caring was reported 84% of the time in good

nurses and kindness was reported 80% of the time. Good/moral character was found in 66% of the sample

and responsibility was determined in 66 percent of the research participants”.

Similarly, Smith and Godfrey (2002) conclude that:

“The larger implications from the research are that good nurses are kind, compassionate, caring men and

women who use their knowledge to do the right thing or make the best decisions. The difficult question

becomes: how do nursing educators create or encourage nursing students to have these qualities?”

However, Rush and Cook (2006) point out that:

“The importance of showing respect to the patients and other health care professions was a consistent

theme in the research. This respect comes in many forms, including respect for age, cultural diversity,

class, and disability. Respect means making each patient feel valued and important”.

Created by Cosette Taylor, Communications Instructor for the Faculty of Nursing at the University of

Manitoba, [email protected]

Successful Synthesis:

While Davis, Hershberger, Ghan, and Lin (1990) and Smith and Godfrey (2002) state that kindness, compassion, and good character are the most frequently identified qualities of a good nurse, Rush and Cook (2006) claim that being respectful is one of the most significant qualities of a good nurse.

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Evidence level and quality rating:

Article title: Number:

Author(s): Publication date:

Journal:

Setting: Sample (composition and size):

Does this evidence address my EBP question?

Yes

No-Do not proceed with appraisal of this evidence

Is this study:

QuaNtitative (collection, analysis, and reporting of numerical data)

Measurable data (how many; how much; or how often) used to formulate facts, uncover patterns in

research, and generalize results from a larger sample population; provides observed effects of a

program, problem, or condition, measured precisely, rather than through researcher interpretation of

data. Common methods are surveys, face-to-face structured interviews, observations, and reviews of

records or documents. Statistical tests are used in data analysis.

Go to Section I: QuaNtitative

QuaLitative (collection, analysis, and reporting of narrative data)

Rich narrative documents are used for uncovering themes; describes a problem or condition from the

point of view of those experiencing it. Common methods are focus groups, individual interviews

(unstructured or semi structured), and participation/observations. Sample sizes are small and are

determined when data saturation is achieved. Data saturation is reached when the researcher identifies

that no new themes emerge and redundancy is occurring. Synthesis is used in data analysis. Often a

starting point for studies when little research exists; may use results to design empirical studies. The

researcher describes, analyzes, and interprets reports, descriptions, and observations from participants.

Go to Section II: QuaLitative

Mixed methods (results reported both numerically and narratively)

Both quaNtitative and quaLitative methods are used in the study design. Using both approaches, in

combination, provides a better understanding of research problems than using either approach alone.

Sample sizes vary based on methods used. Data collection involves collecting and analyzing both

quaNtitative and quaLitative data in a single study or series of studies. Interpretation is continual and

can influence stages in the research process.

Go to Section III: Mixed Methods

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Section I: QuaNtitative

Level of Evidence (Study Design)

Is this a report of a single research study?  Yes  No

Go to B

1. Was there manipulation of an independent variable?  Yes  No

2. Was there a control group?  Yes  No

3. Were study participants randomly assigned to the intervention and control groups?  Yes  No

If Yes to questions 1, 2, and 3, this is a randomized controlled trial (RCT) or experimental study.

LEVEL I

If Yes to questions 1 and 2 and No to question 3 or Yes to question 1 and No to questions 2 and 3, this is quasi-experimental. (Some degree of investigator control, some manipulation of an independent variable, lacks random assignment to groups, and may have a control group).

LEVEL II

If No to questions 1, 2, and 3, this is nonexperimental. (No manipulation of independent variable; can be descriptive, comparative, or correlational; often uses secondary data).

LEVEL III

Study Findings That Help Answer the EBP Question

Skip to the Appraisal of QuaNtitative Research Studies section

A

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Section I: QuaNtitative (continued)

Is this a summary of multiple sources of research

evidence?  Yes

Continue

 No

Use Appendix F

1. Does it employ a comprehensive search strategy and rigorous appraisal method?

If this study includes research, nonresearch, and experiential evidence, it is an integrative review (see Appendix F).

 Yes

Continue

 No

Use Appendix F

2. For systematic reviews and systematic reviews with meta-analysis

(see descriptions below):

a. Are all studies included RCTs? LEVEL I

b. Are the studies a combination of RCTs and quasi-experimental,

or quasi-experimental only? LEVEL II

c. Are the studies a combination of RCTs, quasi-experimental, and

nonexperimental, or non- experimental only? LEVEL III

A systematic review employs a search strategy and a rigorous appraisal method, but does not

generate an effect size.

A meta-analysis, or systematic review with meta-analysis, combines and analyzes results from

studies to generate a new statistic: the effect size.

Study Findings That Help Answer the EBP Question

Skip to the Appraisal of Systematic Review (With or Without a Meta-Analysis) section

B

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Appraisal of QuaNtitative Research Studies

Does the researcher identify what is known and not known about the problem and how the study will address any gaps in knowledge?

 Yes  No

Was the purpose of the study clearly presented?  Yes  No

Was the literature review current (most sources within the past five years or a seminal study)?  Yes  No

Was sample size sufficient based on study design and rationale?  Yes  No

If there is a control group:

 Were the characteristics and/or demographics similar in both the control and intervention groups?

 Yes  No

N/A

 If multiple settings were used, were the settings similar?  Yes  No

N/A

 Were all groups equally treated except for the intervention group(s)?  Yes  No

N/A

Are data collection methods described clearly?  Yes  No

Were the instruments reliable (Cronbach’s [alpha] > 0.70)?  Yes  No N/A

Was instrument validity discussed?  Yes  No N/A

If surveys or questionnaires were used, was the response rate > 25%?  Yes  No

N/A

Were the results presented clearly?  Yes  No

If tables were presented, was the narrative consistent with the table content?  Yes  No

N/A

Were study limitations identified and addressed?  Yes  No

Were conclusions based on results?  Yes  No

Complete the Quality Rating for QuaNtitative Studies section

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Appraisal of Systematic Review (With or Without Meta-Analysis)

Were the variables of interest clearly identified?  Yes  No

Was the search comprehensive and reproducible?

 Key search terms stated  Yes  No

 Multiple databases searched and identified  Yes  No

 Inclusion and exclusion criteria stated  Yes  No

Was there a flow diagram that included the number of studies eliminated at each level of review?  Yes  No

Were details of included studies presented (design, sample, methods, results, outcomes, strengths, and limitations)?  Yes  No

Were methods for appraising the strength of evidence (level and quality) described?  Yes  No

Were conclusions based on results?  Yes  No

 Results were interpreted  Yes  No

 Conclusions flowed logically from the interpretation and systematic review question  Yes  No

Did the systematic review include a section addressing limitations and how they were addressed?  Yes  No

Complete the Quality Rating for QuaNtitative Studies section (below)

Quality Rating for QuaNtitative Studies

Circle the appropriate quality rating below:

A High quality: Consistent, generalizable results; sufficient sample size for the study design; adequate control; definitive conclusions; consistent recommendations based on comprehensive literature review that includes thorough reference to scientific evidence.

B Good quality: Reasonably consistent results; sufficient sample size for the study design; some control, and fairly definitive conclusions; reasonably consistent recommendations based on fairly comprehensive literature review that includes some reference to scientific evidence.

C Low quality or major flaws: Little evidence with inconsistent results; insufficient sample size for the study design; conclusions cannot be drawn.

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Section II: QuaLitative

Level of Evidence (Study Design)

Is this a report of a single research study?

 Yes this is Level III

 No go to II B

Study Findings That Help Answer the EBP Question

Complete the Appraisal of Single QuaLitative Research Study section (below)

Appraisal of a Single QuaLitative Research Study

Was there a clearly identifiable and articulated:

 Purpose? ❑ Yes ❑ No

 Research question? ❑ Yes ❑ No

 Justification for method(s) used? ❑ Yes ❑ No

 Phenomenon that is the focus of the research? ❑ Yes ❑ No

Were study sample participants representative? ❑ Yes ❑ No

Did they have knowledge of or experience with the research area? ❑ Yes ❑ No

Were participant characteristics described? ❑ Yes ❑ No

Was sampling adequate, as evidenced by achieving saturation of data? ❑ Yes ❑ No

Data analysis:

 Was a verification process used in every step by checking and confirming with participants the trustworthiness of analysis and interpretation?

❑ Yes

❑ No

 Was there a description of how data were analyzed (i.e., method), by computer or manually?

❑ Yes ❑ No

Do findings support the narrative data (quotes)? ❑ Yes ❑ No

Do findings flow from research question to data collected to analysis undertaken? ❑ Yes ❑ No

Are conclusions clearly explained? ❑ Yes ❑ No

A

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Skip to the Quality Rating for QuaLitative Studies section

For summaries of multiple quaLitative research studies

(meta-synthesis), was a comprehensive search strategy and

rigorous appraisal method used?

 Yes Level III

 No go to Appendix F

Study Findings That Help Answer the EBP Question

Complete the Appraisal of Meta-Synthesis Studies section (below)

Appraisal of Meta-Synthesis Studies

Were the search strategy and criteria for selecting primary studies clearly defined? ❑ Yes ❑ No

Were findings appropriate and convincing? ❑ Yes ❑ No

Was a description of methods used to:  Compare findings from each study?

❑ Yes ❑ No

 Interpret data? ❑ Yes ❑ No

Did synthesis reflect: ❑ Yes ❑ No

 New insights? ❑ Yes ❑ No

 Discovery of essential features of phenomena? ❑ Yes ❑ No

 A fuller understanding of the phenomena? ❑ Yes ❑ No

Was sufficient data presented to support the interpretations? ❑ Yes ❑ No

Complete the Quality Rating for QuaLititative Studies section (below)

B

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

1 https://www.york.ac.uk/crd/SysRev/!SSL!/WebHelp/6_4_ASSESSMENT_OF_QUALIT ATIVE_RESEARCH.htm 2 Adapted from Polit & Beck (2017).

Quality Rating for QuaLitative Studies

Circle the appropriate quality rating below:

No commonly agreed-on principles exist for judging the quality of quaLitative studies. It is a subjective

process based on the extent to which study data contributes to synthesis and how much information is

known about the researchers’ efforts to meet the appraisal criteria.

For meta-synthesis, there is preliminary agreement that quality assessments should be made before synthesis to screen out poor-quality studies1.

A/B High/Good quality is used for single studies and meta-syntheses2.

The report discusses efforts to enhance or evaluate the quality of the data and the overall inquiry in

sufficient detail; and it describes the specific techniques used to enhance the quality of the inquiry.

Evidence of some or all of the following is found in the report:

 Transparency: Describes how information was documented to justify decisions, how data were reviewed by others, and how themes and categories were formulated.

 Diligence: Reads and rereads data to check interpretations; seeks opportunity to find multiple sources to corroborate evidence.

 Verification: The process of checking, confirming, and ensuring methodologic coherence.

 Self-reflection and self-scrutiny: Being continuously aware of how a researcher’s experiences, background, or prejudices might shape and bias analysis and interpretations.

 Participant-driven inquiry: Participants shape the scope and breadth of questions; analysis and interpretation give voice to those who participated.

 Insightful interpretation: Data and knowledge are linked in meaningful ways to relevant literature.

C Lower-quality studies contribute little to the overall review of findings and have few, if any, of the

features listed for High/Good quality.

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

Section III: Mixed Methods

Level of Evidence (Study Design)

You will need to appraise both the quaNtitative and quaLitative parts of the study independently, before appraising the study in its entirety.

1. Evaluate the quaNitative part of the study using Section I. Level Quality

Insert here the level of evidence and overall quality for this part:

2. Evaluate the quaLitative part of the study using Section II. Level Quality

Insert here the level of evidence and overall quality for this part:

3. To determine the level of evidence, circle the appropriate study design:

 Explanatory sequential designs collect quaNtitative data first, followed by the quaLitative data; and their purpose is to explain quaNtitative results using quaLitative findings. The level is determined based on the level of the quaNtitative part.

 Exploratory sequential designs collect quaLitative data first, followed by the quaNtitative data; and their purpose is to explain quaLitative findings using the quaNtitative results. The level is determined based on the level of the quaLitative part, and it is always Level III.

 Convergent parallel designs collect the quaLitative and quaNtitative data concurrently for the purpose of providing a more complete understanding of a phenomenon by merging both datasets. These designs are Level III.

 Multiphasic designs collect quaLitative and quaNtitative data over more than one phase, with each phase informing the next phase. These designs are Level III.

Study Findings That Help Answer the EBP Question

Complete the Appraisal of Mixed Methods Studies section (below)

Johns Hopkins Nursing Evidence-Based Practice

Appendix E Research Evidence Appraisal Tool

©2017 The Johns Hopkins Hospital/ Johns Hopkins University School of Nursing

3 National Collaborating Centre for Methods and Tools. (2015). Appraising Qualitative, Quantitative, and Mixed Methods Studie s included in Mixed Studies Reviews: The MMAT. Hamilton, ON: McMaster University. (Updated 20 July, 2015) Retrieved from http://www.nccmt.ca/ resources/search/232

Appraisal of Mixed Methods Studies3

Was the mixed-methods research design relevant to address the quaNtitative

and quaLitative research questions (or objectives)? ❑ Yes ❑ No ❑ N/A

Was the research design relevant to address the quaNtitative and quaLitative

aspects of the mixed-methods question (or objective)? ❑ Yes ❑ No ❑ N/A

For convergent parallel designs, was the integration of quaNtitative and

quaLitative data (or results) relevant to address the research question or

objective?

❑ Yes ❑ No ❑ N/A

For convergent parallel designs, were the limitations associated with the

integration (for example, the divergence of quaLitative and quaNtitative data or

results) sufficiently addressed?

❑ Yes ❑ No ❑ N/A

Complete the Quality Rating for Mixed-Method Studies section (below)

Quality Rating for Mixed-Methods Studies

Circle the appropriate quality rating below

A High quality: Contains high-quality quaNtitative and quaLitative study components; highly relevant study design; relevant integration of data or results; and careful consideration of the limitations of the chosen approach.

B Good quality: Contains good-quality quaNtitative and quaLitative study components; relevant study design; moderately relevant integration of data or results; and some discussion of limitations of integration.

C Low quality or major flaws: Contains low quality quaNtitative and quaLitative study components; study design not relevant to research questions or objectives; poorly integrated data or results; and no consideration of limits of integration.