Hypothesis Objectives GD Final 2020

You might also like

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 81

Formulating Research Objectives and

Hypothesis

Professor, NAMS
Dr. Ganesh Dangal (MBBS, MD, FICS)
Senior Consultant Obstetrician and Gynecologist,
Kathmandu Model Hospital/phect-NEPAL
Member: NAME, WAME, ICMJE, COPE, APAME
Editor-in-Chief: J Nep Health Res Counc
Emeritus Editor-in-Chief: J Nep Soc Obstet Gynecol
Editor: JNMA, NJOG, Plos One, J Obs Gyne Acta Scand.
President-elect: Nepal Soc Obs Gyn (NESOG)
IRC Chair: phect-NEPAL, IRB Member: NAMS
Objectives of this session

At the end of the session, learners will be able to:

•Define Objectives-Meaning of research objectives and its importance


•Describe types of objectives
•Describe the qualities and elements of specific objectives
•Formulate objectives using action verbs
•Describe and Formulate hypothesis of research

HYP OBJ RM
• Research Question
• Hypothesis
• Objective

HYP OBJ RM
Research?

Research is an ORGANIZED and


SYSTEMATIC way of FINDING
ANSWERS to QUESTIONS.
What is research?
Discovery
To learn or find out for the first time

Research (Scholarly or scientific investigation or inquiry)


Close, careful study or investigation, especially to
discover new facts or information
Research
• Systematic Collection,
• Analysis and
• Interpretation of data
• To answer a certain question
• Or solve a problem

HYP OBJ RM
HYP OBJ RM
HYP OBJ RM
What are the steps in doing research?

Choose the research question

Develop a protocol

Carry out the study and collect data

Analyze the findings

Draw and disseminate conclusions


What is a research question?

Uncertainty about something in a group (population)


which the investigator wants to resolve by

making measurements (observational study)


or
interfering with natural course (interventional study)

in a subset of the group (sample: his study subjects)

… … in a planned manner
Research questions: examples

• What causes dental caries?

• Is wearing an X mask better than Y mask for preventing


chronic lung disease in traffic policemen of Kathmandu?

• Do children eating an O diet not become obese as adults?

• How can we help patients with high blood pressure, who


cannot take currently available drugs?
Usual research questions

• To find out the frequency of a disease


– Incidence
– Prevalence
– Disease burden

• To identify and define a disease


– What are the features of a disease

• To find out who gets disease


– Time
– Place
– Person
Usual research questions: continued

• To identify the determinants of disease


– Cause
– Modifiers

• What does the disease lead to? How does it behave?


– Outcomes, natural history

• How can a disease be treated?

• How can a disease be prevented?

• How can a disease be diagnosed (early)?


Research Hypothesis

Specific version of a research question which one wishes


to address––in a form that establishes the basis for test of
statistical significance
What is a hypothesis

• Testable tentative statement that proposes a possible explanation to a


phenomenon or event

• May include a prediction

• ‘… logical supposition, a reasonable guess, an educated conjecture. It


provides a tentative explanation for a phenomenon under investigation.’

• The hypotheses should state a relationship between at least two


variables
Hypothesis: formulation

Participants Who is the patient or what problem is being


addressed?
Intervention What is the intervention or exposure?
Comparison What is the comparison group?
Outcome What is the outcome or endpoint?

E Effect size
+ Time frame
+ Study design
B
J

R
M

The PICO model for clinical question 


P Patient, Population, How to describe a group of
or Problem patients similar to yours?
I Intervention, Prognostic Main intervention, prognostic
Factor, or Exposure factor, or exposure to consider?
C Comparison or Intervention What is the main alternative to
(if appropriate) compare with the intervention?
O Outcome you would like to What you hope to accomplish,
measure or achieve measure, improve, or affect?
  What Type of question are Diagnosis, Etiology/Harm,
you asking? Therapy, Prognosis, Prevention
  Type of Study you want to What would be the best study
find design/methodology?
Study question

To determine whether laparoscopic fundoplication is good for patients with


hiatus hernia and reflux esophagitis (as compared to open fundoplication)
Hypothesis

Does [intervention] lead to [outcome] more often in [subjects] than the


[comparator] does?
Hypothesis

• Participants (Subjects)

• Intervention

• Comparator

• Outcome (nature, effect size)


Hypothesis

• Subjects: Patients with hiatus hernia and reflux esophagitis (Los


Angeles grade C or D)

• Intervention

• Comparator

• Outcome
Hypothesis

• Subjects: Patients with hiatus hernia and reflux esophagitis (Los


Angeles grade C or D)

• Intervention: Laparoscopic fundoplication by an experienced surgeon

• Comparator

• Outcome
Hypothesis

• Subjects: Patients with hiatus hernia and reflux esophagitis (Los


Angeles grade C or D)

• Intervention: Laparoscopic fundoplication by an experienced surgeon

• Comparator: Open (surgical) fundoplication

• Outcome
Hypothesis

• Subjects: Patients with hiatus hernia and reflux esophagitis (Los


Angeles grade C or D)

• Intervention: Laparoscopic fundoplication by an experienced surgeon

• Comparator: Open (surgical) fundoplication

• Outcome: Reduction of 3 points in symptom score on XYZ scale 6 months


after surgery
Hypothesis

Does [laparoscopic fundoplication done by an experienced surgeon] lead to


[relief of symptoms by >3 points on XYZ scale] more often [80% vs. 60%] in
[persons with hiatus hernia and reflux esophagitis, LA grade C or D] than
[open surgery] does?
Outcomes

• ‘Clinically important’ reduction in symptoms


– at 6 mo after procedure
– at 4 wk after procedure
– at 5 y after procedure

• Others
– Lower esophageal sphincter pressure (manometry)
– Grade of esophagitis (endoscopy)
– Total time pH < 4.0 (pH metry)
Primary vs. secondary outcomes

Primary outcome measure


One (or a few) pre-specified outcome of greatest importance and is usually
the one used in the sample size calculation

Secondary outcomes
Other outcomes of interest (including unanticipated or unintended effects of
the intervention)
Question versus Hypothesis: An example

• Is it worth spending 2 days teaching a course on ‘Study design and


statistics’?
Question versus Hypothesis: An example

• Is it worth spending 2 days teaching a course on ‘Study design and


statistics’?

• Do such courses help the participants?


Question versus Hypothesis: An example

• Is it worth spending 2 days teaching a course on ‘Study design and


statistics’?

• Do such courses help the participants?

• Do people prepare better research proposals after attending such workshops


than before?
Question versus Hypothesis: An example

• ‘Study design and biostatistics’ workshops may help improve the quality of
research proposals that researchers write

• If workshops influence the quality of research proposals, then the research


proposals written by those who have participated in a workshop will be
better than from those who have not attended such a workshop.
Question and Hypothesis

• ‘Study design and biostatistics’ workshops may help improve the quality of
research proposals that researchers write

• Participation in the workshop will lead to an increase (by at least 10 points)


in the average score of research proposals submitted by the participants over
the next 3 years as compared to that of proposals submitted over the 3
years before attending the workshop.
Formulation of a therapy question

Intervention Outcome

Is zinc effective in treating cold?

Patient/problem Intervention

In children with common cold, is oral zinc effective in


reducing the duration of symptoms as compared to placebo?
+RCTs
Outcome Comparison
Study design
Formulation of a diagnostic question

Test/intervention Outcome

Is 64-slice CT a good test for coronary artery disease?

Test/intervention Outcome

Is 64-slice CT as sensitive and specific in diagnosing


coronary artery disease as is coronary angiography?

Problem/patient Comparison

Study design +cross-sectional studies


Formation of an etiology question

Exposure Outcome

Is snoring a risk factor for diabetes?

Patients (subjects) Exposure

Are people who snore regularly at a higher risk of


developing type II diabetes mellitus than those who do not?

Outcome Comparison

Study design +cohort and case control studies


Hypotheses: Examples

• Chocolate may cause pimples.


• Eating sugar may cause dental caries
• Excess salt in soil may affect plant growth.
• Color of incident light may affect plant growth.
• Bacterial growth may be affected by temperature.
• Ultra violet light may cause skin cancer.
• Drug A may reduce diastolic blood pressure.
Research Proposal Writing

HYP OBJ RM
The output of a research is at best as good as
its research objective
Setting concrete and well-considered research objectives
is imperative to ensuring that your research is going to be
meaningful, and that efforts aren't wasted on objectives
that don't make sense or are impractical.

The key question here is: how do I


formulate such objectives?
WHY Objective?
The formulation of objectives will help us to:
•Focus the study (narrowing it down to essentials)
•Avoid collection of data that are not necessary for
understanding and solving the identified problem
•Organize the study in clearly defined parts

The explicit formulation of study objectives is an essential


step.

HYP OBJ RM
We have to make sure that our objectives:
• Cover the different aspects of the problem and its contributing
factors in a coherent way and in a logical sequence
• Are clearly expressed in measurable terms
• Are realistic considering local conditions, fulfils the purpose

• Use action verbs that are specific enough to be measured


- to determine - to compare - to verify – to analyse- to discuss
- to calculate - to describe - to find out- to establish - to identify
NOT: - to appreciate - to understand - to study- to believe – to know- to
learn- to see, etc
HYP OBJ RM
Objectives
This is a very important and pivotal section
and everything else in the study is centered
around it
•The objective stated should be specific, achievable and
measurable
•Too many objectives are to be avoided
•Even just one clearly stated relevant objective for a study
would be good enough
•If there is more than one objective the objectives can be
presented in the appropriate order of importance ( 2-4)
B
J

R
M

Avoid Objectives

• Over ambitious objectives


• Vague objectives
• Irrelevant objectives
Types of Objectives
General Objectives
• Clear indication of expected outcome in general terms
• Means of achieving the objectives: usually not specified

Specific Objectives
• Clear indication of expected outcome
• Specification of means of achieving the objectives
• Way of measuring outcome

HYP OBJ RM
Formulation of General objective
• Framing the questions according to general and specific objectives;
• Developing a testable hypothesis to achieve the objectives;

General objective: aim of the study in general terms


- matches with the title
• Example: In a study on missed opportunities for EPI in Kathmandu
the general objective was: “to assess missed opportunities for EPI
in Kathmandu”.

HYP OBJ RM
Formulation of Specific Objectives
• Specific objectives: measurable statements on the specific questions to
be answered. It is statement of precise outcome that is measurable.

• Unlike the general objective, the specific objectives are more specific and
are related to the research problem situation.
• They indicate the variables to be examined and measured.

• The objective should be simple, specific, achievable and measurable


• Too many objectives to be avoided

HYP OBJ RM
Specific Objectives
Example:
In the study of missed opportunity for EPI in Kathmandu the specific
objectives could be:

•To find out the magnitude of missed opportunities for children who
attend OPD, MCH, CDD, etc. in Kathmandu,

•To examine the reasons for children not being immunized while
attending the OPD, MCH, etc. services.

HYP OBJ RM
General or Broad Objective
• States what is to be achieved by the study in general terms.
• Usually only one per study
• Related to core problem and topic of the study
• Shows target population and place
• Uses action verb

Example:
• To determine the factors associated with increased number of
diarrhea cases among under fives in X village

HYP OBJ RM
Specific objectives
Describe the variables that are measured by the study
• Several in one study (2-4)
• Use action verb

Example:
1. To find out methods used for excreta disposal

2. To find out methods used for water treatment

3. To determine the feeding habits of the under fives

HYP OBJ RM
Qualities of specific objectives
• Stated in logical sequence
• In order of importance

• Clear and unambiguous


• What is to be done?

• Specific
• Addressing one specific thing only

• Measurable
• Can be evaluated
• Use proper action verbs (Determine, Find out, Identify, etc)
HYP OBJ RM
Qualities of specific objectives-cont
• Attainable, given the resources available

• Realistic, given the local conditions

• Time bound, can be achieved in the given time

HYP OBJ RM
Qualities of Specific Objectives
SMART
•Specific 
•Measurable
•Achievable/ Attainable
•Reliable / Relevant/ Realistic/Reasonable
•Time bound/(feasible)

HYP OBJ RM
HYP OBJ RM
Problems Regarding Objectives
• Stating too many study objectives which are not
appropriately addressed
• Some of the specific objectives are not addressed in the
results section/ dummy tables at all.

• On the basis of these specific objectives - the methods,


results and discussion sections will have to be presented.
• For example, sample size calculations for each stated
objective is not covered /done

HYP OBJ RM
What is a research question?
Uncertainty about something in a group (population)
which the investigator wants to resolve by

making measurements (observational study)


or
interfering with natural course (interventional study)

in a subset of the group (sample: his study subjects)

… … in a planned manner

HYP OBJ RM
Research Question

HYP OBJ RM
HYP OBJ RM
PICO Model

HYP OBJ RM
Research questions: More examples
• What causes dental caries?

• Is wearing an N95 mask better than KN95 mask for


preventing COVID-19 in frontline health workers of NAMS?

• Do children eating an O diet not become obese as adults?

• How can we help patients with high blood pressure, who


cannot take currently available drugs?

HYP OBJ RM
Formulation of a therapy question
Intervention Outcome

Is zinc effective in treating cold?

Patient/problem Intervention

In children with common cold, is oral zinc effective in


reducing the duration of symptoms as compared to placebo?
+RCTs
Outcome Comparison
HYP OBJ RM Study design
Formulation of a diagnostic question
Test/intervention Outcome

Is 64-slice CT a good test for coronary artery disease?

Test/intervention Outcome

Is 64-slice CT as sensitive and specific in diagnosing


coronary artery disease as is coronary angiography?

Problem/patient Comparison

Study design +cross-sectional studies


HYP OBJ RM
Primary vs. secondary outcomes
Primary outcome measure
One (or a few) pre-specified outcome of greatest importance and is
usually the one used in the sample size calculation

Secondary outcomes
Other outcomes of interest (including unanticipated or unintended
effects of the intervention); by products—side effects if efficacy study

HYP OBJ RM
What is a hypothesis
• It is a prediction of a relationship between one or more variables
and the problem under study.

• That is, it specifies the relationship among variables.

• These variables are to be statistically tested at a later stage.

• In order to measure the relationship among variables to be studied


the dependent and independent variables need to be identified.

HYP OBJ RM
Hypothesis
• Idea or explanation Based on known facts
- which is not proven but leads to further study / OR
Discussion / OR
Testing (Statistically Testing the Hypothesis)

• Scientific hypothesis “ has statistical methods to test” for significance.


• It is also predictor of the outcome- Speculates or predicts some outcome

HYP OBJ RM
Criteria for Hypothesis
Must fulfill two criteria:

• It is a statement about relation between variables (or


arms)
E.g. Paracetamol is as equally good analgesic as Ibuprofen.

• It carries clear implications of testing the stated


relationships

HYP OBJ RM
Types of Hypothesis
Two types of hypothesis

• Null hypothesis (HO): Hypothesis of no difference;

• Alternate hypothesis (H1): alternate to null hypothesis,


( there is a difference)

HYP OBJ RM
Null hypothesis
• It states that there is no relation between the phenomena whose
relation is under investigation

Examples:
• There is no difference between Tramadol & Fentanyl for
post –operative pain via epidural route.
• Paracetamol is as equally good analgesic as Ibuprofen.
There is no difference between smoker & non-smoker in post–operative healing
after periodontal surgery

HYP OBJ RM
Alternate Hypothesis
• It states that there is some kind of relation.

• (or there is a difference between two arms or


variables or phenomena)

E.g. Tramadol > Fentanyl


• Paracetamol < Ibuprofen

HYP OBJ RM
Examples of Hypotheses
• The health of children living in rural villagization projects is
better than those living in traditional rural communities.

• To examine whether there is any significant difference


between district “A” and district “B” with respect to their
malaria prevalence rates. (bi-directional)

• An increase in the frequency of hand washing is followed by a


reduction in Corona prevalence. (unidirectional)
HYP OBJ RM
Testing Hypotheses
Analytic studies are used to test hypotheses concerning the
relationship between a suspected risk factor and an outcome and
to measure the magnitude of the association and
its statistical significance.

Analytic study designs have two broad designs: Observational and


interventional.
•Examples of observational studies: comparative cross-sectional,
cohort and case-control studies, with follow ups.

HYP OBJ RM
Hypothesis

Does [intervention] lead to [outcome] more often in [subjects] than


the [comparator] does?

HYP OBJ RM
Hypothesis
• Participants (Subjects)

• Intervention

• Comparator

• Outcome (nature, effect size)


HYP OBJ RM
Hypothesis
• Subjects: Patients with hiatus hernia and reflux esophagitis (Los
Angeles grade C or D)

• Intervention

• Comparator

• Outcome
HYP OBJ RM
Hypothesis
• Subjects: Patients with hiatus hernia and reflux esophagitis (Los
Angeles grade C or D)

• Intervention: Laparoscopic fundoplication by an experienced surgeon

• Comparator: Open (surgical) fundoplication

• Outcome

HYP OBJ RM
Hypothesis
• Subjects: Patients with hiatus hernia and reflux esophagitis (Los
Angeles grade C or D)

• Intervention: Laparoscopic fundoplication by an experienced surgeon

• Comparator: Open (surgical) fundoplication

• Outcome: Reduction of 3 points in symptom score on XYZ scale 6


months after surgery (symptom improvement)
HYP OBJ RM
Hypothesis

Does [laparoscopic fundoplication done by an experienced surgeon]


lead to [relief of symptoms by >3 points on XYZ scale] more often
[80% vs. 60%] in [persons with hiatus hernia and reflux esophagitis, LA
grade C or D] than [open surgery] does?

HYP OBJ RM
HYP OBJ RM
THESIS TOPIC

FETOMATERNAL OUTCOME IN WOMEN WITH PREGNANCY INDUCED


HYPERTENSION VERSUS NORMOTENSIVE PREGNANCY
Comparison of Maternal and Fetal outcome in
Pregnancy Induced Hypertensive with
normotensive pregnant women
1. To find out the frequency of PIH according to age group
and parity.
2. To compare the maternal complications like preterm
labor, antepartum hemorrhage (APH), retinopathy, eclampsia,
CVA, HELLP syndrome, DIC, etc between PIH and
normotensive pregnancy
3. To compare rates of fetal complications like low birth
weight, prematurity, IUGR, birth asphyxia and others
4. To compare the frequency of mode of delivery like normal
delivery, preterm delivery, vacuum or forceps delivery and
LSCS
Summary
• Objectives: 
• What researchers should be able to do…
• Types: 
• General and specific
• Qualities: SMART, FINER

• Elements: 
• Activity, Content, Condition & Criteria

HYP OBJ RM
Summary-contd.
 A hypothesis refers to a provisional idea whose merit
requires evaluation.

 Can be Null or Alternate

 A hypothesis requires more work by the researcher in order


to either confirm or disprove it. (statistical tests)- testable

HYP OBJ RM
Thank You

Acknowledgement:

Some Slides Taken from AuthorAid, INASP, NAMS and


NAME Presentations

Questions ?
HYP OBJ RM

You might also like