Meta-Analysis Techniques and Applications

This assignment requires the student to explore various resources on meta-analysis techniques, including books, articles, and systematic review protocols. The student must demonstrate an understanding of meta-analysis methods, their application in medical research, and how to evaluate comparative effectiveness research. A practical tool for evaluating comparative effectiveness research is also expected to be discussed. Additionally, the student should provide a summary of relevant literature on meta-analysis techniques and applications, including a discussion on the own-age bias in face recognition, exercise as a treatment for depression, and the methodological quality assessment tools for preclinical and clinical studies.

Running head: EVIDENCE BASED NURSING RESEARCH
Evidence Based Nursing Research
Name of the Student
Name of the University
Author note

1EVIDENCE BASED NURSING RESEARCH
Introduction
The paper deals with the critique of the article, “Interventions that increase use of Pap
tests among ethnic minority women: a meta-analysis” by Han et al. (2011). The critique is based
on certain aspects of the systematic review article. It includes the “problem, search strategy, the
sample, quality appraisal, data extraction and data analysis general, data analysis quantitative,
and the conclusion” part of the chosen article.
Critique of the paper using questions
The problem
Han et al. (2011), has clearly stated the research problem to be the lack of sufficient
literature in the area of the ethnic minority women in US having barriers to early redetection of
cervical cancer as well as interventions. Therefore the research focuses on the interventions
methods to promote the screening among the target population in US. However, there is no
clearly formulated research question in the meta-analysis as recommend by the Zeng et al.
(2015). The research question can be understood by the research objectives and is considered a
drawback.
The scope of the project is appropriate and can be considered the strength of the study as
per Schmidt & Hunter (2014). It is well described in the abstract of the paper as well as in the
introduction contextualising readers about the contents of the research process. The scope of the
project is appropriate as per clearly mentioned research objectives, succinct information about
methodology, as well as results and primary outcome.

2EVIDENCE BASED NURSING RESEARCH
The topic of the review is important for nursing as it may help the nurses to decrease the
barriers to detection of cervical cancer among the US ethnic minority women. The nurses can use
the intervention approaches described in the study to improve the screening behaviour of the
ethnic minority women (Han et al., 2011).
The author has well stated that the patients or group of people concerned the strategies of
interventions to be identified, comparison groups, efficacy of the interventions as comparison
and the outcomes (quality of interventions). According to Khambalia et al. (2012) the four
components in any metaanalysis is considered to reduce the bias. However, the main drawback is
the lack of succinct information on the dependent and independent variable. It adds to confusion
and makes the validity doubted.
The integration approach is discussed as the series of studies (clinical trials) to be
reviewed that address the same question and analysed of its quality by different researchers
independently. It is consistent with the most commonly evident approach of the metaanalysis
study. This approach may be considered a gold standard as per the merits it offers over analysing
set of different data. It ensures the validity of the data quality and conduct statistical analysis
Sena et al. (2014). However, it would have been more convenient if the reader had defined the
phenomena more comprehensively.
Search strategy
The search strategy was appropriate and was consistent as per the method described by
Zeng et al. (2015). However, there are no clearly defined criteria for selecting the primary
studies. It is very confusing for the readers as the details related to search are not mentioned
comprehensively. The author had discussed using four electronic databases that are “MEDLINE,

3EVIDENCE BASED NURSING RESEARCH
CINAHL, PsycINFO, and Science Citation Index-Expanded” (Han et al., 2011). It is not very
comprehensive. The author had identified keywords that are not so exhaustive. The selection
criteria were precisely summarized as initial screening of title, abstracts, and references of prior
systematic reviews. It aligns with the method described by Haidich, (2010) as it states that
working from references eliminates bias.
The databases are considered to be appropriate as they contain rich literature both
updated and outdated. However, there are drawbacks related to the databases search which
appeared to be quick search. According to Schmidt & Hunter (2014), a sound review would
occur only if all of the databases are searched. However, author had reviewed only few of them.
It would have been effective to use “The Cochrane Controlled Trials Registry” as it is considered
to be the complete database of clinical trials. In the MEDLINE database the author had not
specified if the publication type was chosen as “clinical trial” (Han et al., 2011). According to
members of the Cochrane Collaboration, MEDLINE database have incorrectly classified many
trials (Schaumberg et al., 2018). It decreases the reliability. The drawbacks are balanced as the
author had provided proper justification for not considering the databases like EMBASE. Overall
criteria were reasonable as also found in other meta-analysis by Shazly et al. (2015). The limited
database search can be justified considering the focus to improve screening of cervical cancer
among US women of ethnic minority.
There were two independent reviewers who assessed the relevancy of the studies and if it
meets inclusion criteria. The details of the adequate supplementary efforts to identify the relevant
studies are not highlighted in the study. It may be considered a drawback in terms of the
reliability and validity of the study (Han et al., 2011). As recommend by Moher et al. (2015)
PRISMA flow chart was included for the literature and the selection process.

4EVIDENCE BASED NURSING RESEARCH
The sample
The inclusion and exclusion criteria were clearly articulated. The eligibility criteria was
chosen as studies that aimed to increase use of Pap tests either exclusively or among other health
behaviors among the asymptomatic women. Studies were chosen that concerns Black, Hispanic
woman or Asian American and it was considered if represented 440% of the total sample. The
criterion was taking studies that were based on outcomes as women adherence to Pap tests.
Further studies with quasi experimental design were considered. The exclusion criteria was
eliminating article that focuses on health professionals oriented interventions. It can be
concluded that the criteria was reasonable as it helped the author meet the research objectives by
obtaining article pertaining to the “interventions that can increase the use of Pap tests among
ethnic minority women”. The articles only published in English language and between 1984 and
April 2009 were included. It can be considered the strength of the study as the inclusion and
exclusion criteria was defensible (Han et al., 2011). The criteria were sufficient to meet the
research objectives (Zeng et al., 2015).
The search strategy should be designed in a manner that it yields comprehensive sample
of studies as per Shazly et al. (2015). In the chosen research article wide range of studies were
included. However, no international databases were used as the focus was on US population.
The studies lacking key information were excluded as per the inclusion criteria and the
exclusion criteria. Therefore, overall only 18 articles were considered. The reviewers did not
attempt to contact the original researchers.
However, the author had provided adequate justification for eliminating some studies as
mentioned by researcher of meta-analysis in medical research Haidich (2010). For instance, the

5EVIDENCE BASED NURSING RESEARCH
author had mentioned that, “Interventions exclusively aimed at health professionals were not
included,” It was justified as the same was highlighted by previous reviews. The justifications
are also supported with relevant literature in the same area.
Quality appraisal
The study used the meta-analysis technique to calculate the affect size (dependent
variables) of the interventions. For multiple studies the authors compared two or more treatment
groups with one comparison group. Using the evidenced based framework, “Research and
Quality Scoring Method, the Jadad scale, and the items published by Cho and Bero”, the quality
of the studies was rated (Han et al., 2011). The author considered the interventions with more
than one component. It may be drawback as considering multiple components may be
misleading. Scores between 0 to 9 were considered ad the studies with quality score of 6–9 were
considered to be high.
The variables coded independently by the reviewers. With detailed discussion among the
team members the discrepancies were eliminated. This set of criteria as per Sena et al. (2014) is
beneficial as it would reduce the risk of bias.
The two raters did come up with coding agreement and calculated the Cohen’s k, average
k was 0.812, using SPSS version 15. The quality scores were documented and illustrated as
tabular charts. However, this method has less power to detect the heterogeneity of review
(Hughes-Morgan et al. 2018.. The study also documented the estimated affect size with 95% CIs.
The drawback arises from lack of presentation of the quality assessment protocol as highlighted
by Zeng et al. (2015). The inter-rater agreement report was not presented.

6EVIDENCE BASED NURSING RESEARCH
For both the analysis of the results and selection of study the appraisal information was
used in a well-defined and defensible manner. At each stage, the information was supported with
relevant literature and evidences from meta-analysis used by other authors.
Date extraction
The reviewers identified the methodological and the administrative aspects of the study.
It is evident from the extracted information section that highlights the coding of the following
information-
 First author, year, study design;
 sample (% ethnic minority women)
 types of intervention and duration
 time to outcome measure
 method of outcome
 proportion of Pap tests for the treatment and comparison groups;
 comparison group (no intervention, minimal intervention/usual care, or other non-
cervical cancer intervention);
 cultural strategies used
 study quality
Sample characteristics extracted were adequate as recommended by Shazly et al. (2015).
It takes the research process one step close to the intervention strategies required for detecting
the cervical cancer among the concerned population.
Sufficient information was extracted about the study findings and is detailed in tabular
form for the readers. It is convenient and easy to interpret. The administrative aspects were not

7EVIDENCE BASED NURSING RESEARCH
clearly highlighted. Instead the author had adopted the typology. The same was used in previous
reviews of cancer screening interventions. It was convenient to find the different categories of
the cultural strategies identified by the authors. The author had defined the different components
of interventions. It can be constituted as the drawback of the study (Han et al., 2011).
The data set integrity can be considered to have been enhanced as the review process
used more than two members for data analysis. It is sufficient enough to eliminate the
discrepancies in the reviewed data. According to Haidich (2010) several steps should be
considered in the investigation of the interventions. Although random models may seem
appropriate, it is recommended to examine the data and make use of common sense in decision
making (Wulff et al. 2018). It will help determine the influence of the dependent and
independent variable.
Data analysis- general
According to Khambalia et al. (2012) the most effective method of examining the pattern
of the summary-effect size is the cumulative metaanalysis. It helps identify the impact of a
specific study on the overall conclusion. In the given study the author did not comprehensively
explain the pooling of the results. The pooling is evident from the affect size at (95%) confidence
interval that signified the efficacy of the interventions to increase the Pap test among US women
of ethnic minority. The results also demonstrated the intervention that is access enhancement to
have largest effect size. Next to it was community education and individual counseling. The
statistical significance of the data was also given. The result section were well described the
studies in the meta –analysis as also recommended by Murad et al. (2014). It detailed the types

8EVIDENCE BASED NURSING RESEARCH
of interventions in different studies, followed by the effects of interventions on the use of Pap
tests at follow up. The author ensured that it met certain criteria.
It can be concluded from the results presented in the paper that the data is credible and
addressed the research objective. The researchers were affective in summarising the findings
using tables and figures.
Data analysis-quantitative
Narrative review is the summary of all the primary studies collected that is modified with
the reviewer’s perspective and practical knowledge. It gives broader picture of the problem and
is considered useful to develop a conceptual framework. However, there are several drawbacks
as it is based on the personal beliefs of the reviewer. Therefore, it is considered lowest evidence
type. On the other hand the meta-analysis is the superior technique. It is most effective when
compared to narrative review in integrating literature. With literature review it would not have
been possible to collect the affect size or correct studies. Also it would not have been possible to
collect quantitative synthesis (Pae, 2015). Therefore, it was justified to conduct metaanalysis.
There is no adequate justification for using the narrative review it would not have been possible
to collect different interventions with efficacy for cervical cancer.
Procedures for computing effect size estimates for all relevant outcomes were highlighted
and appropriate. It includes interventions at different settings, intervention types, and cultural
strategies, ethnic groups, and intervention of different quality. Using the evidenced based
framework, “Research and Quality Scoring Method, the Jadad scale, and the items published by
Cho and Bero”, the quality of the studies was rated. Thus, each affect size was calculated using
the meta-analyses techniques. For data analyses, STATA software was used and raw-score

9EVIDENCE BASED NURSING RESEARCH
metric was opted to define the effect size. It was used to calculate the overall mean weighted
effect size for the selected studies. It is considered beneficial as per Miles & Crisp (2014) for its
ease in interpretation and usefulness in clinical purposes. For studies with odd ratios post-
intervention Pap test rates were calculated.
DerSimonian and Laird Q-statistic was used to test heterogeneity. If significant results
were observed then the random effects model was use to poll affect sizes. The author conducted
sensitivity analysis for influence from studies, and fail-safe analysis for publication bias. Further,
the author used leave-one-out approach to remove potentially influential studies. The author also
used Rosenthal’s fail-safe N to rule out of potential bias. However, it was not adequately
addressed. Excluding the unpublished data may lead to inaccurate estimation of effect sizes and
cause bias. These models can be considered sound as per similar approach used in study by
Schuch et al. (2016). The author used pooled proportions for ethnic subgroup analysis and is
justified considering the research objective and target population.
Conclusion of review
The author has drawn clear conclusions from the review process. Readers have detailed
understanding about the consistency of the evidence. It is evident from the discussion well
supported with the relevant literature, that the ethnic minority women are highly likely to use Pap
test if access-enhanced strategies were combined. The results were found to be consistent with
the research objectives and purpose of the research appeared to be fulfilled. The conclusion
related to single and multiple cultural strategies of interventions targeting ethnic minority women
and those related to social network satisfied the curiosity of the readers. In regards to the quality
of the results, the author clearly stated that there was insignificant number of studies signifying

10EVIDENCE BASED NURSING RESEARCH
increase in Pap test through cultural strategy involving members of target community. At the
same time highlighted social network interventions were of high quality. The conclusion
indicates need of robust studies.
The limitations of the review were noted. According to Haidich (2010), the researcher
must highlight the limitations of the paper succinctly to ensure the reliability of the results and
must be well justified. The researcher in the meta-analysis has comprehensively discussed the
heterogeneity found. The author also highlighted about the publication bias and lack of
generalisability of the results to other countries having ethnic minority women. The provision of
appropriate justification to the limitations can be considered the strength of the study.
The drawback of the section arises from the lack of succinct implications of the data
obtained in nursing practice (Rhodes & Anastasi, 2012). It only concerns the care provider and
the patient’s outcomes. However, recommending further research on the tightly controlled
studies for interventions pertaining to the cervical cancer screening adds to reader’s convenience.
Conclusion
The meta-analysis of the chosen paper is critically reviewed. The critical analysis has
enlightened the knowledge of evidence based practice. It enhanced the awareness of way to
collect the research evidence on a particular problem area. In future studies similar research can
be conducted to address the problem in the nursing practice.

11EVIDENCE BASED NURSING RESEARCH
References
Haidich, A. B. (2010). Meta-analysis in medical research. Hippokratia, 14(Suppl 1), 29.
Retrieved from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3049418/
Han, H. R., Kim, J., Lee, J. E., Hedlin, H. K., Song, H., Song, Y., & Kim, M. T. (2011).
Interventions that increase use of Pap tests among ethnic minority women: a meta‐
analysis. Psycho
‐oncology, 20(4), 341-351.
Hughes-Morgan, M., Kolev, K., & Mcnamara, G. (2018). A meta-analytic review of
competitive aggressiveness research. Journal of Business Research, 85, 73-82.
Khambalia, A. Z., Dickinson, S., Hardy, L. L., Gill, T. A., & Baur, L. A. (2012). A synthesis
of existing systematic reviews and meta‐analyses of school‐based behavioural
interventions for controlling and preventing obesity. Obesity Reviews, 13(3), 214-233.
Miles, E., & Crisp, R. J. (2014). A meta-analytic test of the imagined contact
hypothesis. Group Processes & Intergroup Relations, 17(1), 3-26.
Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., ... & Stewart, L.
A. (2015). Preferred reporting items for systematic review and meta-analysis
protocols (PRISMA-P) 2015 statement. Systematic reviews, 4(1), 1.
Murad, M. H., Montori, V. M., Ioannidis, J. P., Jaeschke, R., Devereaux, P. J., Prasad, K., ...
& Meade, M. O. (2014). How to read a systematic review and meta-analysis and
apply the results to patient care: users’ guides to the medical literature. Jama, 312(2),
171-179.

12EVIDENCE BASED NURSING RESEARCH
Pae, C. U. (2015). Why Systematic review rather than narrative review?. Psychiatry
investigation, 12(3), 417-419.
Phan, K., Tian, D. H., Cao, C., Black, D., & Yan, T. D. (2015). Systematic review and meta-
analysis: techniques and a guide for the academic surgeon. Annals of cardiothoracic
surgery, 4(2), 112.
Rhodes, M. G., & Anastasi, J. S. (2012). The own-age bias in face recognition: a meta-
analytic and theoretical review. Psychological bulletin, 138(1), 146.
Schaumberg, D. A., McDonald, L., Shah, S., Stokes, M., Nordstrom, B. L., & Ramagopalan,
S. V. (2018). Evaluation of comparative effectiveness research: a practical
tool. Journal of comparative effectiveness research, (00).
Schmidt, F. L., & Hunter, J. E. (2014). Methods of meta-analysis: Correcting error and bias
in research findings. Sage publications. Retrieved from:
https://books.google.co.in/books?
hl=en&lr=&id=sU8XBAAAQBAJ&oi=fnd&pg=PR1&dq=Meta-
analysis+in+medical+research&ots=hUhK0-Itxy&sig=KYrAqgWTVY3CJOIZ_lLs-
H9Z6Fs#v=onepage&q=Meta-analysis%20in%20medical%20research&f=false
Schuch, F. B., Vancampfort, D., Richards, J., Rosenbaum, S., Ward, P. B., & Stubbs, B.
(2016). Exercise as a treatment for depression: a meta-analysis adjusting for
publication bias. Journal of psychiatric research, 77, 42-51.
Sena, E. S., Currie, G. L., McCann, S. K., Macleod, M. R., & Howells, D. W. (2014).
Systematic reviews and meta-analysis of preclinical studies: why perform them and

13EVIDENCE BASED NURSING RESEARCH
how to appraise them critically. Journal of Cerebral Blood Flow &
Metabolism, 34(5), 737-742.
Shazly, S. A., Murad, M. H., Dowdy, S. C., Gostout, B. S., & Famuyide, A. O. (2015).
Robotic radical hysterectomy in early stage cervical cancer: a systematic review and
meta-analysis. Gynecologic oncology, 138(2), 457-471.
Wulff, D. U., Mergenthaler-Canseco, M., & Hertwig, R. (2018). A meta-analytic review of
two modes of learning and the description-experience gap. Psychological
bulletin, 144(2), 140.
Zeng, X., Zhang, Y., Kwong, J. S., Zhang, C., Li, S., Sun, F., ... & Du, L. (2015). The
methodological quality assessment tools for preclinical and clinical studies,
systematic review and meta‐analysis, and clinical practice guideline: a systematic
review. Journal of evidence-based medicine, 8(1), 2-10.