See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/327691129

Systematic Review And Meta-Analysis In Oral Medicine-An Update

Article · November 2017

CITATIONS READS
0 396

8 authors, including:

Nallan Chaitanya Parupalli Karunakar

49 PUBLICATIONS   385 CITATIONS    50 PUBLICATIONS   816 CITATIONS   

SEE PROFILE SEE PROFILE

Komali Garlapati
Panineeya Institute of Dental Sciences and Research Centre
37 PUBLICATIONS   231 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ascorbic acid oral supplementation in oral mucositis severity reduction View project

dental caries and probiotics View project

All content following this page was uploaded by Nallan Chaitanya on 17 September 2018.

The user has requested enhancement of the downloaded file.

 Journal of Rural Development Society
THE TRENDS IN SCIENCE AND TECHNOLOGY
(Published by : Rural Develeopment Society - Govt. Recognized R & D Institution)
www.rdsjournals.com

Systematic Review And Meta-Analysis In Oral

Medicine-An Update
Nallan Csk Chaitanya* , D.Reshma Priyanka, P.Karunakar, Komali Garlapati, Mamata Boringi,
Sujatha Mekala Mary, Sushma K,
*
Department of Oral Medicine and Radiology, Panineeya Institute of Dental Sciences and Research Centre, Hyderabad-500082,

Abstract-The amount of information being generated in medical (randomized control trials) being done. The results of most of the
research is becoming overwhelming. Clinicians are finding it studies are different, vary among same population and even
nearly impossible to stay current, even in their own area of specialty conflicting, which leads to difficulty in clinical decision-making
as new studies are constantly being published. Meta-analysis is a and puts the clinician in a helpless situation as to what to consider
review which is done systematically, following certain criteria, and what to discard.
with the pooling of results and quantitative analyzation. Valuable Evidence based medicine may be defined as the systematic,
information is provided by a well-designed meta-analysis for quantitative, preferentially experimental approach to obtaining and
researchers and clinicians. Systematic reviews (SRs) represent a using medical information. Central role in evidence based medicine
specific type of medical research in which the units of analysis are is played by meta-analysis, which is a statistical procedure that
the original primary studies. They are essential tools in combines the results of all other studies conducted independently.
synthesizing available scientific information, increasing the In the hierarchy of evidence, meta-analysis occupy the top position
validity of the conclusions of primary studies, and identifying areas where clinical evidence is ranked according to the strength of the
for future research. They are also indispensable for the practice of freedom from various biases. (1)
evidence-based medicine and the medical decision making
The terms "meta-analysis" and "systematic review" are
process. SRs are an essential tool in synthesizing available
occasionally used interchangeably in the literature. However, they
scientific information; they enhance the validity of the findings
are two different processes with complementary approach. A
from individual studies and identify areas of uncertainty where
systematic review is defined as a methodic review of the literature,
research is needed. They are essential to the practice of evidence-
often accompanied by a scale or grading algorithm that assigns
based medicine. However, conducting high quality systematic
values to the presence or absence of evidence (defined by the
reviews is not easy and they can sometimes be difficult to interpret.
publication of randomized trials, observational studies, and case
A strict methodology and applying quality control is to be
reports) that answers the question posed by the systematic review.(2)
implemented in SRs to avoid biased conclusions. Regarding a
particular patient it is the clinician who must make decisions Glass first defined meta-analysis in the social science literature
and; SRs help us in making those decisions.. as "The statistical analysis of a large collection of analysis results
from individual studies for the purpose of integrating the findings".
Keywords-Evidence based medicine, Meta-analysis, Precision, (3)
Statistics, Systematic review..
In simpler words, when systematic review is subjected to
I. INTRODUCTION statistical analysis what we get is meta-analysis.
We live in the information age and the practice of medicine is Meta-analysis systematically assesses the results of various
becoming increasingly specialized. The rapid advancement in the researches which are done previously to get to a definitive
field is attributed to integration of health care with research which conclusion regarding what the research hypothesis is. Most of the
has paved way for proper protocols and update information on studies included in meta-analysis fall into the category of
various aspects of health care management. randomized, controlled clinical trials. (1)
With the myriad of problems, we can estimate the amount of RATIONALE
research work being carried out and the multiple small studies The available scientific literature on a particular research, enables

Nallan Csk Chaitanya, D.ReshmaPriyanka, P.Karunakar, Komali Garlapati, Mamata Boringi, Sujatha Mekala Mary And Sushma K,
"Systematic Review And Meta-Analysis In Oral Medicine-An Update,"Journal of Rural Development Society
The Trends in Science And Technologies, Volume 1, Issue 1, November - 2017, ISSN No:……….
www.rdsjournals.com Page No : 35
us to get to a simplified clinical decision which is based on the 
evidence available, helping us to get a good clinical outcome and
Assemble most comprehensive data set
also in reducing the cost of the treatment.
Meta-analysis has the advantage of detecting small but clinically

important effects, as it encompasses large subject numbers. It also Analyze results of RCT using statistics
helps us to find significant inter-study differences which may 
generate new hypotheses for future research, and analyzing of Compare alternative analyses if appropriate and possible
group/subset effects increases precision. (2)

Meta-analysis also helps in establishing whether scientific Prepare critical summary, stating aims, materials and results (7)
findings are consistent in various studies which can be generalized
across populations, settings, and treatment variations, or whether Generally one has to follow certain prescribed format of
findings vary by particular subsets. It enhances precision, provides questionnaire when contemplating on a systematic review in any
robust estimates, and answers subject such as those given as outline below.
calculatedquestions that single trials are under-powered or were 1. Did the review address a clearly focused question?
not designed to address. (4) 2. Did the review include the right type of the study?
3. Did the reviewers try to identify all relevant studies?
Explicit methods used in Meta -analysis limit bias and help
4. Did the reviewers assess the quality of all the studies included?
improve the reliability (precision) and accuracy (validity) of
5. If the results of the study were combined, was it reasonable?
conclusions. As data is available from more than one study the
6. How the results presented and what are main results?
'sample size' increases leading to increase in "power". In clinical
7. How precise are the results?
trials and cohort studies, it helps in observing more events of interest
8. Can results be applied to local population?
in different groups considered. So, when incidence or mortality is
9. Were all important outcomes considered?
rare, meta-analysis helps in giving combined estimates which are
10. Should policy or practice change as a result of the evidence?(8)
likely to be more precise. (5)
Supporting arguments for performing a meta-analysis include ASSESSMENT OF TRIAL METHODOLOGY
the increase in power and precision of the estimates, quantification a) Searching data base for relevant studies in question -
of effect sizes, and the assessment of the consistency of results
across trials. The most convincing reason for conducting meta- A single data base may not effectively provide all the studies
analysis is when similar clinical studies are present with results of in question equally. Dickersin et al showed that only 30-80%
different effect sizes. (2) of all known published RCTS were identifiable using
MEDLINE. (9)Thus a comprehensive search is important to
Finally, meta-analysis identifies various questions which have reduce selection bias. Article search should not be limited to
not been adequately addressed with past research. Thus it English language and those indexed in MEDLINE.
documents the need for a major clinical trial. Similarly, it also confirms
the sufficiency of available literature on a particular topic. Thus, it b) Quality assessment of trials -
helps to avoid the time and expense of conducting another clinical There should be a predetermined method of assessment
trial. It has been estimated that for any clinical setting which involves including eligibility and quality of studies. At least two
a 10% risk of an event, 10,000 patients need to be randomized to reviewers should independently assess the quality of included
demonstrate a 25% reduction in the risk with a more than 90% power studies to minimize the risk of selective bias. (10)
and a p value less than 0.01. (6) c) Result pooling -
How to conduct meta-analysis in medicine? Determine whether the pooling was reasonable. A clinical
The flow chart shows the methodology of a systematic review judgment should be made about whether it was reasonable
of RCTs. Thus any such methodology in oral medicine should for the studies to be combined based on the individual trials
include: differed considerably in population studies, interventions and
comparisons used or outcome achieved. The Meta analytic
State objectives of the review and outline eligibility criteria validity looking for homogeneity of outcome should be
 stressed upon. This is done by inspecting the graphical display
Comprehensively search for trials that meet eligibility criteria of results looking for similarities in direction of results. If there
 is heterogeneity, it is not proper to pool such results. (7)
Tabulate each of the characteristics and assess the methodology d) Meta-analysis:
 Following a systematic review, data from individual studies
may be pooled quantitatively and reanalyzed using established
Apply eligibility criteria and justify exclusion
statistical methods.(11) This technique is called meta-analysis.
Nallan Csk Chaitanya, D.ReshmaPriyanka, P.Karunakar, Komali Garlapati, Mamata Boringi, Sujatha Mekala Mary And Sushma K,
"Systematic Review And Meta-Analysis In Oral Medicine-An Update,"Journal of Rural Development Society
The Trends in Science And Technologies, Volume 1, Issue 1, November - 2017, ISSN No:……….
www.rdsjournals.com Page No : 36
 It is done by combining the samples of individual studies, the The "3" stands for the control group in the example.
overall sample size thus increase improving the statistical power
Interpretation:
of the studies as well as the precision of the estimates of
treatment effects. (12) If the ODDS ratio and relative risk are greater than 1, it indicates
increased likelihood of good intervention in group. If it is less than
It is a two stage process. 1, then it indicates decreased change with intervention. If it is equal
e) Stage 1: Involves the calculation of the measure of the treatment to 1, then there is no change.
effect with its 95% confidence intervals for each individual More widely placed confident intervals show us that there is
study. The summary statistics used to measure treatment effect less precision outcome. If the confidence interval for relative risk
include odds ratio (OR) relative risk (RR). or odds ratio for an estimate includes 1, then there is no statistic al
f) Stage 2: Over all treatment effect is calculated as a weighted significance between the groups compared. (7)
average of individual summary statistics. (13)Greater weights Advantages of doing a meta-analysis: (15)
are given to the results from studies providing more
The main advantage is we can able to detect small but clinically
information because they are likely to be closer to true effect
relevant effects of an intervention. Diverse groups of subjects are
of estimate. It is thus taken as inverse of variance. The typical
present across studies, which permits a transparent solution to a
graph to display this data is called "forest plot". (14)(Figure 1)
question or hypothesis. The conclusion has the potential advantage
g) The picture below is a typical representation of forest plot. of being based on most, if not all, existing evidence from relevant
The alphabets of the left side of the graph represent first author studies. If the evidence from studies are available which are
of the primary study. The black squares represent the odds statistically well handled, then meta-analysis provides a conclusive
ratio of individual studies. The horizontal lines their 95% result and helps in avoiding time and expense of doing a large
confidence intervals. The area of black space reflects the study for confirmatory purpose.
weight the each trial contributed to meta-analysis. The 95%
For many studies as the data is too old or too expensive to
confidence interval represents the true underlying effect in 95
retrieve, meta-analysis helps in collecting the data needed. The
% of the occasion if the study was repeated again and again.
statistical tools are such that most sources of bias and heterogeneity
The solid vertical line represents to no effective treatment,
can be statistically examined. Sensitivity of statistically significant
with p Value (P<0.05). The overall treatment effect calculated
results can be measured. The statistical techniques to analyze results
as weighted average of individual odds from meta-analysis
from meta-analyses are generally not new and many standard
and its CI at bottom represents a diamond. The center of the
techniques apply. The resources to help accomplish a systematic
diamond represents combined effect. If the diamond shape is
review of studies to be included in the meta-analysis are available.
present towards left of the line it shows lesser change in results
and if on right side of the line, it shows no change of the For Eg., To know the association of HPV with oral cancer, whether
results. If it touches the line at Centre, it is considered as no it is just an incidental finding or a sole etiologic factor one can have
effect. (OR=1) that is p is not significant. a look into the "Systematic meta analysis on association of human
Methods of expression of results can be done using odds papilloma virus and oral cancer"(16) where it was concluded that,
ratio, relative risk and mean differences. If the result is two there was definite prevalence of HPV in cancer group patients when
ways, for example, no effect Vs effect, presence of infection Vs compared to control group but that association wasn't clearly proven
no infection, odds ratio or relative risks are used. If the result is to be existing.
a continuous variable (like blood glucose), mean differences Some of the more prominent issues facing Meta analyses include:
may be used. (17)
What is ODDS and ODDS ratio : An assumption that the evaluation techniques carried out in
doing meta-analysis are consistent across various studies. There is
ODD-it is calculated as number of patients in a group who
something called as "file-drawer" effect- that is difficulty in locating
achieved end point ÷ number of patients who did not.
all the articles needed. Some statistics must be adjusted accordingly
Example: Treatment with vitamin a reduces mucositis in 6 of leading to bias. Heterogeneity in the studies leads to inconsistent
10 patients: ODD is here 6:4=1.5. In control group the odds may coding of variables. Moreover, the exact source of heterogeneity is
be 3:7 = 0.43. not detected. Because the study is the unit of analysis, the adequacy
The ODDS ratio is the ratio of odds of treatment group to odds of statistical power of any meta-analysis may be in question.
of control group. (1.5÷0.43). Sometimes the data available may be inconsistent leading to
difficulty in interaction and trends. When combining studies for
What is risk and relative risk?
analysis, appropriate weights and scoring must be considered.
Risk- here number of patients who received treatment and found Apart from these, there are some statistical issues which have to be
effective to total number of patients. (6÷10= 0.6) addressed to overcome the limitations of a well-designed Meta
Risk ratio is a risk of two groups ratio = 6/10÷3/10= 0.6/0.3 analysis.eg. If Bayesian approach is being considered, then an

Nallan Csk Chaitanya, D.ReshmaPriyanka, P.Karunakar, Komali Garlapati, Mamata Boringi, Sujatha Mekala Mary And Sushma K,
"Systematic Review And Meta-Analysis In Oral Medicine-An Update,"Journal of Rural Development Society
The Trends in Science And Technologies, Volume 1, Issue 1, November - 2017, ISSN No:……….
www.rdsjournals.com Page No : 37
experienced statistician should be consulted to eliminate the al. (eds.), Evidence-Based Practice: Toward Optimizing Clinical
misinterpretation of a study. Outcomes,Springer-Verlag Berlin Heidelberg 2010.
16. Chaitanya NC, Allam NS, Gandhi Babu DB, Waghray S, Badam RK,
CONCLUSIONS Lavanya R. Systematic meta-analysis on association of human papilloma
The traditional basis of medical practice has changed by the virus and oral cancer. J Can Res Ther, vol 12, pp 969-74, 2016.
use of randomized, blinded, multicenter clinical trials and meta- 17. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic
analysis, leading to the widely used term "evidence-based Reviews of Interventions Version 5.0.2 [updated September 2009]. The
medicine". Any field can benefit from well conducted small research Cochrane Collaboration, 2009.Available from www.Cochrane-
handbook.org.
and compilation of the same in the form of meta-analysis when
large studies are missing.
Single studies, whether meta-analytic or not, will not provide
the precise understanding of responses to treatment, diagnostic tests,
or risk factors that influence the disease process. Meta-analytic
approaches offer benefits such as addressing the limitations of study
size, including diverse populations along with the opportunity to
evaluate new hypotheses than any single study.
At last, we can conclude that systematic reviews and meta-
analysis are definitely a major contribution to the research studies
conducted in area of Medicine, and much more are required for
benefit of the clinician [1].
REFERENCES
1. A.B. Haidich. Meta-analysis in medical research. HIPPOKRATIA, vol
14, issue 1, pp 29-37, 2010.
2. Andres Pinto.Considerations for planning and designing meta-analysis
in oral medicine.OOOO, vol 116, issue 2,pp 194-20, 2013.
3. Glass GV. Primary, secondary and meta-analysis of research.Educational
Researcher, vol 5, pp 3-8, 1976.
4. Lau J, Ioannidis JPA, Schmid CH. summing up evidence: one answer is
not always enough. Lancet, vol 351, pp 123-7, 1998.
5. Dickersin K, Berlin JA. Meta-analysis: state-of-the-science. Epidemiol
Rev, vol 14, pp 154-76, 1992.
6. YusufS,Peto R, lewis J, Collins R, Sleight P. Beta blockade during and
after myocardial infarction: An overview of the randomized trials.
ProgCardiovasc Dis, vol 27, pp 335-71, 1985.
7. A K Akobeng. Understanding systematic reviews and meta- analysis.
Arch Dis Child, vol 90, pp 845-848, 2005.
8. Critical Appraisal Skills Programme. Appraisal tools. Oxford, UK. http:/
/www.phru.nhs.uk/casp/appraisa.htm.
9. Dickersin K, Scherer R, Lefebvre C. Systematic reviews: identifying
relevant studies for systematic reviews. BMJ, vol 309, pp 1286-91,1994.
10. Clarke M, Oxman AD, eds. Selecting studies. Cochrane reviewers'
handbook 4.2.0[updated March 2003]. In: The Cochrane Library, issue
2. Oxford: Update software, 2003.
11. Muir Gray JA. Evidence based health care. How to make health policy
and management decisions. London: Churchill Livingstone, pp 125-6,
2001.
12. Lang TA, Secic M. How to report statistics in medicine. Philadelphia:
American College of Physicians, 1997.
13. Deeks JJ, Altman DG, Bradburn MJ. Statistical methods for examining
heterogeneity and combining results from several studies in meta-
analysis.BMJ publishing group, pp 285-312, 2001.
14. Lewis S, Clarke M. Forest plots: trying to see the wood and the trees.
BMJ, vol 322, pp 1479-80,2001.
15. Alfred A. Bartolucci and William B. Hillegass. Overview, Strengths, and
Limitations of Systematic Reviews and Meta-Analyses. F.Chiappelli et

Nallan Csk Chaitanya, D.ReshmaPriyanka, P.Karunakar, Komali Garlapati, Mamata Boringi, Sujatha Mekala Mary And Sushma K,
"Systematic Review And Meta-Analysis In Oral Medicine-An Update,"Journal of Rural Development Society
The Trends in Science And Technologies, Volume 1, Issue 1, November - 2017, ISSN No:……….
www.rdsjournals.com Page No : 38

View publication stats