1

Research Methods for Medical Graduates

Chapter Abstracts

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
1 Basics of Medical Research
Research in any field is an enterprise that carries its own risks and benefits. One may make
heavy investment in terms of time, money and expertise yet the returns are not assured in this
endeavor. This is particularly so for medical research where we deal with unpredictable human
beings and vitals such as health and life are at stake. First time research is daunting anyway but it
is more so in medicine. This chapter first describes what medical research is and what it is about
in our context, so that the contents of this book are properly demarcated. This chapter gives an
overview of medical research endeavors, including the pre-eminent role of empiricism, the
dominance of uncertainties, broad steps, and the essential ingredients of good research. This
would help in maintaining high standards in the research process so that the findings are
believable and replicable. Details of all these aspects are provided in the subsequent chapters.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
2 The Topic of Medical Research
The first problem faced by a researcher, particularly at graduate level, is the selection of an
appropriate topic for a thesis. This chapter lists the parameters on which the selection can be
made after review of literature and after examining the available resources, and then describes
steps that sharpen the topic by precise statement of the objectives and hypothesis.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
3 Study Designs: An Overview
All empirical studies are based on a set of evidence collected from different sources, and this
process is expected to follow a design to optimise the resources and to achieve sufficient validity
and reliability. This chapter gives an overview of various designs that are generally followed in
medical studies: details are in subsequent chapters. The meaning of a design of a study is
explained at the outset and the types of design such as descriptive and analytical are also
explained. Essential features of further classification of analytical studies into interventional and
observational are subsequently described. Next section discusses reliability and validity of a
design to give believable results, and the last section gives a guideline of where to use which
design.
 2

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
4 Clinical Trials
Medical experiments on human beings are called clinical trials. The objective of a clinical trial
is to discover or verify the clinical or pharmacological effect of an intervention and to identify
its adverse effects. The end-point could be safety and/or efficacy. The intervention must be
potentially beneficial and not harmful. This could be a drug, a surgical procedure, a medical
device, some behavioral change, a process of care, or any such intervention.
Regimen research environment is rapidly changing. Safety is increasingly getting
precedence over efficacy. The current paradigm seems to be that a new regimen should be free
of toxicity while also be efficacious.When carefully conducted, clinical trials are the fastest and
the safest way as of today to evaluate efficacy and safety of a new regimen under controlled
conditions. Since the subjects are humans in clinical trials, a large number of issues crop-up
ranging from ethics to profound variations. A human experimentation cannot be conducted
unless sufficient reasons for doing so are present. This chapter describes various types of
clinical trials, their basics, and the conditions under which they provide valid results. The
chapter ends with tips for choosing a right design of a trial depending on the resources.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
5 Observational Studies
Nature is a great experimenter. Many interventions or changes occur naturally that require no
human intervention. Some people are naturally exposed to iodine deficiency in water and some
women naturally have low haemoglobin (Hb) level. Study of such naturally occurring events
can provide invaluable help in studying cause–effect relationship with conviction. Such a study
can be based on records or on actual observations, or a combination of both. This is generally
categorized as an observational study because it requires recording of observations only:
sometimes also referred to as an epidemiological study. Whereas both beneficial and harmful
‘natural interventions’ can be studied by observational studies, such studies are particularly
helpful in studying the effect of harmful processes and conditions because, for them, human
intervention is ruled out.
There are three basic formats in which an observational study can be carried out –
prospective, retrospective and cross-sectional. This chapter describes these formats and
discusses their merits and demerits so that a judicious choice can be made depending on the
problem at hand.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
6 Assessment of Medical Factors
Medical factors can be broadly classified into (i) etiological factors that give rise to the risk
 3

factors, (ii) risk factors that promote or inhibit a health condition, (iii) diagnostic factors that
help to identify the presence or absence of a disease, (iv) treatment modalities that aim to
alleviate the suffering and bring the patient back to health, and (v) prognostic factors that
determine the outcome (outcome can be complete relief, dissatisfaction, discomfort, disability,
disease or death). In a research setup, all or a subset of these may require assessment.
A ‘factor’ in our terminology is a characteristic of the subjects, and ‘indicator’ is its
measurement. Obesity is a factor and body-mass index (BMI), waist-hip ratio (WHR), and skin-
fold thickness its indicators. Indicator converts a factor into its operational definition. This
distinction is important for proper implementation of research.
This chapter first discusses the intricacies of assessment of medical factors such as
univariate and multivariate assessments, and the assessment in the implementation and the
results phase of the study. Next section is on the type of medical factors and the choice of
indicators, and the last section provides the details of the assessment of mortality, various
durations, and quality of life that require special handling.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
7 Methodology of Data Collection
All empirical research is based on data. These are not necessarily quantitative: they could be
qualitative also. Signs and symptoms are qualities yet are ‘measurements’ in statistical sense.
Qualitative data too convert to numerics when summarized for a group.
Where from to get the data for a medical research? In a clinical setup, they come from
patients’ interview and examination, and their laboratory and radiological investigations. These
are called primary data and recorded in a questionnaire, schedule or proforma such as an Excel
sheet. Secondary data are existing records of the previous patients and other subjects. These can
also be used for meaningful research.
All measurements are easily understood statistically as variables since they do vary from
person to person. Thus there are qualitative variables and there are quantitative variables. These
could be on nominal, ordinal or metric scale. This distinction is important since the method of
inference is different for different types of variables. Merits and demerits of various types of
measurements will help to decide which ones to use for different kinds of assessment in a
research. All these issues are discussed in this chapter.
Types of measurements such as different scales and qualitative–quantitative, discrete–
continuous, and other types are discussed in the first section. Tools of data collection such as
questionnaire, schedule, record, interview, and examination are in the second section. Quality of
data such as reliability and validity is discussed in the third section, and assessment of validity
of medical tools such as pilot study and pretesting, sensitivity–specificity and ROC curves are
presented in the last section.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
 4

8 Sampling and Sample Size

Medical decisions are almost invariably based on samples. Sample of blood, urine, sputum,
stool, and biopsies are everyday occurrences. Sampling is the only feasible method in this
situation since complete material of these biological substances for anybody cannot be studied.
The thrust in this chapter is on sample of individuals. Whether the research is descriptive
or analytical, it is conducted on a sample of subjects. Sample is the statistical term for the group
of subjects included in the study, and that almost invariably would be a fraction of the target
population.
There are two dimensions of adequacy of a sample. First, it should represent the full
spectrum of subjects in the target population. Various methods of sampling are adopted in
different situations to meet this objective. These methods are discussed in the first section for
descriptive studies and in the second section for analytical studies. Second, the sample size must
be reasonably large to give reliable results, without being excessively large. Sample size is
discussed in the last section. In between there is a discussion on non-sampling errors in contrast
to sampling errors. Controlling non-sampling errors many times assumes more importance
because these have potential to vitiate the results beyond redemption, whereas sampling errors
can be handled by choosing the representative sample of adequate size.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
9 Research Protocol
Once through with the steps mentioned in the preceding chapters, a researcher is ready to
develop a protocol for the research project. Protocol is the document that outlines the proposed
research, gives its full rationale including the objectives and the hypotheses, and the complete
methodology including the system for reaching to an unbiased and credible conclusion. This is
the final step in planning a research and is the backbone that supports it in all steps of its
execution. It serves as a reference for the members of research team whenever needed.
This chapter describes the structure and contents of a medical research protocol, and is
divided into two sections. The first section is on the structure of the protocol such as title,
collaborators and supervisors, executive summary, and logistics, and the second section is on the
contents of the main body of the protocol such as background, review of the literature, objectives
and hypothesis, target population, sampling and sample size, method of eliciting the information,
data collection and analysis, and the system of reaching to a valid conclusion.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
10 Processing of Data
After a research protocol is prepared and approved by a competent authority comes the stage of
actual collection of information from the subjects selected for this purpose on the lines proposed
in the protocol. All precautions stated in the protocol must be taken to ensure data quality and
 5

completeness. The data are generally entered into a spreadsheet, called the master chart. Some
of the individual data are sometimes converted to a more useful quantity such as body mass
index from height and weight. Some researches require calculation of scores for assessing
severity of the disease. These are discussed at the outset in this chapter.
The next step is to use the entire dataset and prepare group summaries. The methods to
obtain epidemiological summary indices such as rates and ratios, including relative risk and odds
ratio, and for measures such as mean and standard deviation (SD) are presented. The last section
is on tabular and graphical representation of data, including a brief on Gaussian and non-
Gaussian distributions. A medical research many times requires that the quantitative
measurements obtained for individual subjects be classified as normal or abnormal. This also is
discussed at the end as it requires an understanding of the shape of the distribution of the values.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
11 Data Analysis
After the data are collected, entered, cleaned and collated, they are ready for the rigors of
analysis. This includes procedures such as generating correct estimates of various parameters
(e.g., incidence and prevalence), building up confidence intervals, testing statistical significance,
and assessing the strength and type of relationship. Actual methods depend on the nature of data,
the type of hypotheses to be examined, and the theoretical conjectures that form the foundation
of the study.
Statistical analysis of data is done in several ways. For descriptive studies and for some
analytical studies, the primary objective is to find the percentage of cases that have a particular
outcome, or the average level of a medical measurement. This is called estimation. This estimate
is accompanied by, what is called, the confidence interval that delineates the range of values
beyond which a sample summary is unlikely to lie in repeated samples. Common methods for
obtaining this are provided in this chapter. The other important activity under data analysis is the
test of hypothesis whereby we find how likely the values obtained in our study can be from a
presumed population. This requires the concept of P-value and power. These also are briefly
discussed. The basic methods for testing significance (this term seems to be on way out) such as
chi-square for qualitative data and Student’s t-test for quantitative data are presented. The next
discussion is on regression which is used to study the relationship between two or more
characteristics. This includes both the ordinary least square where the dependent is a quantitative
measurement and the logistic where the dependent is binary. This section also contains a brief on
correlation coefficient. The methods for assessing the cause–effect relationship and for validation
of results are presented later in this chapter and statistical fallacies that so commonly arise in
medical research are discussed in the end.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
12 Writing a Thesis or a Paper, and Oral Presentation
 6

After completing a research, it is time to let the world know what new has been achieved or
could not be achieved. No research is complete unless it is read, discussed, and evaluated. The
dissemination could be in a conference through PowerPoint or any other kind of presentation
but most medical researches culminate into a written report that could take a form of a paper, a
thesis, a dissertation, or a full-fledged project report. Successful researchers are skillful
biomedical communicators too and they celebrate their research.
This chapter starts with the general principles of effective writing. The actual contents of
a paper or a thesis in terms of what to write and how to write from the beginning to the end,
including the IMRaD format, are presented in subsequent sections. The last section discusses
oral or poster presentation in a conference. Reporting ethics, including plagiarism, are separately
discussed in the last chapter.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
13 Reporting Guidelines
A large number of guidelines are available for reporting of various kinds of studies. There is one
for clinical trials, called CONSORT, one for observational studies, called STROBE, and one for
diagnostic accuracy studies, called STARD. In addition, for statistical reporting, there are
SAMPL guidelines because many authors falter on statistical reporting and are unable to
properly draft the thesis or paper despite doing a good work. We discuss all these guidelines in
this chapter so that a researcher does not have to face inconvenient questions when the research
is reported for publication in a reputed journal, or when a thesis is examined.

Chapter Author – Abhaya Indrayan

https://orcid.org/0000-0002-5940-9666
14 Reporting Ethics and Peer Reviews
Ethics is conforming to the standards of conduct. In the context of medical research, it has
several facets. Research ethics is in terms of sufficient reasons to start a new investigation and
using an appropriate design including adequate sample size so that the efforts do not go waste
and there is no unnecessary exposure to the subjects. This was discussed in Chapter 8 and
concern for the welfare of the study subjects was expressed in Chapter 2. Now we discuss about
reporting ethics in this chapter. Some of it too has already been discussed but some aspects of
reporting ethics have still remained undiscussed. These include duplication and plagiarism,
copyrights and permission, conflict of interest, and peer review. All these are the subject matter
of this chapter.