HEALTH SYSTEM

RESEARCH
Term “Research”

Therefore, research means to observe the phenomena again and again from different
dimensions.
Health Research
• Health Research, as defined by the World Health Organization (WHO),
involves "investigating ways and means of achieving the most
effective, economic and durable impact on health as a state of
complete physical, mental and social well being, through the
development of tools, methods and approaches appropriate primarily
to national and local needs and the promotion of national self
reliance"
Health Systems Research (Health Services
Research
• The following definition was adopted by the WHO Western Pacific
Working Group on Health Services Research (1978):
• “Health services research is an integral part of health services
development and can be defined as any group of activities which
involve the generation of information or the application of knowledge
on a scientific basis with a view to providing more effective, efficient
and equitable health care for defined populations. Giving due regard
to the social, epidemiological, economic, legal, operational,
theoretical and political dimensions, health services research aims to
assist in the improved design and management (planning,
implementation and evaluation) of health service.
• Health systems research → "The people, institutions, and activities
whose primary purpose is to generate high quality knowledge that
can be used to promote, restore, and or maintain the health status of
populations. It can include the mechanisms adopted to encourage the
utilization of research.“ ( WHO)
Objectives
• Play role in improving health and Stimulating economic growth.
• Gathers data, information and facts for advancement of knowledge.
• Help in delivering better, fairer and more equitable healthcare
• Identify challenges and provide their solution.
• Provide valid scientific evidence for formulating health policies.
Types of Research
Descriptive
Analytical
Basic
Applied
Quantitative
Qualitative
Empirical
Theoretical
Descriptive research
• It is basically to find out the facts regarding any query. The basic purpose is
to describe the existent situation. The researcher has no control over the
variables. Surveys are descriptive studies, conducted to gain knowledge and
find out facts regarding any particular problem.
Analytical research
• Analytical research is concerned with in-depth study and evaluation of
available information in order to explain complex phenomenon. The
researcher has to use the facts or information already available and analyse
these to make a critical evaluation of the material. For analytical research
you need a hypothesis.
Basic research
• Basic research is usually considered to involve a search for knowledge
without a defined goal of utility or specific purpose. It helps in
formulation of a theory and generalization. Natural phenomenon and
mathematics are the examples of basic research.
Applied research
• Applied research is problem oriented, and is directed toward a
defined and purposeful end; it is frequently generated, by a perceived
need, and is directed toward the solution of an existing problem.
Health system research is a good example of applied research.
Quantitative research
• Quantitative research as the name implies is based on measurement
of quantity or amount.
(a) measurement of Variables;
(b) estimation of population parameters (the determination and
comparison of rates, ratios, proportions, etc.);
(c) statistical testing of hypotheses, or estimating the extent to which
chance alone may account for our findings.
Qualitative research
• Qualitative research is multi-method in focus, involving an interpretive,
naturalistic approach to its subject matter. This means that qualitative
researchers study things in their natural settings, attempting to make
sense of or interpret phenomena in terms of the meanings people bring
to them.
• Qualitative research involves the studied use and collection of a variety
of empirical materials case study, personal experience, introspective,
life story interview, observational, historical, interactional, and visual
texts-that describe routine and problematic moments and meaning in
individuals' lives.
The philosophical approach to research is basically of two types:
1. Empirical
2. Theoretical.
• Health research mostly follows the empirical approach
Empirical Research
• Empirical research in the health sciences necessarily involves
quantification; for the most part this is achieved by 3 related
numerical procedures:
(a) measurement of variables;
(b) estimation of population parameters (the determination and
comparison of rates, ratios, proportions, etc.);
(c) statistical testing of hypotheses, or estimating the extent to which
chance alone may account for our findings.
Theoretical research
• It is based upon theory and abstraction. Even though it is possible in
Epidemiology and other health sciences to study phenomena in
abstraction with mathematical models, advances in understanding of
disease occurrence and causation cannot be made without a
comparison of the theoretical constructs with that which we actually
observe in populations.
Health research triangle:
• Health research, be it empirical or theoretical, basic or applied,
generally falls under three operational interlinking categories of
biomedical, health services and behavioural research, the so called
health research triangle.
Components of a Research Project
• Any research project basically comprises five components.
1. Selection And Formulation Of The Research Problem - A Research Question
2. Literature Search
3. Research Objectives
4. Budgeting
5. Research Methodology
6. Work Plan
7. Data Collection
8. Analysis and Interpretation
9. Reporting
1. Selection And Formulation Of The
Research Problem - A Research Question
• The statement of the research problem is the basis for the
development of a research proposal, including research objectives
and hypotheses, method, work plan and budget. It also allows the
investigator to describe the problem to reflect on its importance, its
priority in the country and on the locality and to point out why the
research problems should include all essential points.
Research design should focus squarely on problems of priority to local
and national health development.
• magnitude of the problem and its impact;
• urgency of need for solution;
• relevance to the aims of the funding agency;
• amenability of the problem to investigation;
• feasibility of approach;
• chance of success;
• expected impact of successful outcome;
Mill's canons
• In health research, hypotheses are often constructed and tested to
identify causes of disease and to explain, the distribution of disease in
populations. Mill's canons of inductive reasoning are frequently utilized
in the forming of hypotheses which relate association and causation.
Briefly stated these methods include:
(a) the method of difference - when the frequency of a disease is
markedly dissimilar under two circumstances and a factor can be
identified in one circumstance and not in the other, this factor, or its
absence may be the cause of the disease (e.g. - the difference in
frequency of lung cancer in smokers and non-smokers);
(b) the method of agreement - if a factor or its absence, is common to a
number of different circumstances that are found to be associated with the
disease (e.g. the occurrence of hepatitis A is associated with patient contact,
crowding, and poor sanitation and hygiene, each conducive to the
transmission of the hepatitis virus;
(c) the method of concomitant variation, or the dose response effect - the
increasing expression of endemic goiter with decreasing levels of iodine in the
diet, the increasing frequency of leukemia with increasing radiation exposure,
the increase in prevalence of elephantiasis in areas of increasing filarial
analogy - the distribution and frequency of a disease or effect may be similar
enough to that of some other disease to suggest commonality in cause, e.g.,
hepatitis B virus infection and cancer of the liver.
Criteria of a Good Research Question
 It should be:-
1. Feasible
• Adequate number of subjects can be recruited
• Adequate Technical Expertise is available
• Affordable in Time and Money
• Manageable in scope, i.e. it can be completed within the time frame and one should avoid to
be too ambitious.
2. Interesting to Investigator and the scientific community.
3. Novel
• Confirms or refutes previous finding
• Extends previous findings
• Provides new findings
4.Ethical
• It should not conflict with the society's cultural, moral, religious and
legal values.
5. Relevant
• To advancement of scientific knowledge.
• To influencing clinical health and policy
• To guidance for future research direction
2. Literature Search
• Literature searching today is an art which means, "developing an ability to search
in a purposeful and systematic manner through the range of literature or
information relevant to your particular field".
Why search the Literature?
• There are a number of reasons to search the literature.
1. To help identify a research topic
2. To avoid any duplication
3. To facilitate a full understanding of the work that has been done in the past.
4. To be updated with new developments
5. To have an evidence (documented base) for the work that is planned
6. To formulate a hypothesis and provide discussion for the topic.
3. Research Objectives
• The objectives of a research project summarize what is to be achieved by the
study. The general objective of a study states what is expected to be
achieved by the study in general terms. It is advisable to break down a
general objective into smaller, logically connected parts. These are normally
referred to as specific objectives. The specific objectives should specify what
you will do in your study, where and for what purpose.
• The formulation of objectives help in:-
1. Focusing the study.
2. Avoiding collection of data that are not strictly necessary for understanding
and solving the identified problem.
3. Organizing the study in clearly defined parts or phases.
4. Budgeting
Why?
• It is important that we know what we, as researchers are getting into
and how much money we will be spending on conducting and
analyzing the study and in final submission of the
dissertation/research paper.
What?
The heads that we should take into consideration include:-
• Literature Searching: This includes the money spent on Internet
browsing, getting photocopies of text, library membership fee etc.
• Statistical Help: One would be lucky to get good statistical help at no
cost.
• Material: We may be needing medicine kits, or, any other disposable
equipment for the study which may not be available in your
institution (free of cost or subsidized). Photocopying of forms may
also need to be done by you for the study.
• Typing of the Manuscript: If one is not computer literate or do not
know typing or have access to a computer, he/she will need someone to
help with this.
• Printing of manuscript for submission: The manuscript will have to be
printed in quadruplicates for submission. The photographs will have to
be copied (if you have colored photographs, that will add to the budget)
and micrographs of x-rays may have to be made.
• Postage: The manuscript may have to be revised and sent a number of
times before final acceptance. These are only some of the essential
budget heads, one can nave more depending on the topic or design of
the study being conducted.
5. Research Methodology
The selection of a research strategy depends on the study objective.
• descriptive, validating and surveillance strategies, using an interview,
survey or mailed questionnaire or existing data;
• observational or analytical strategies, including prospective (cohort)
studies, historical (or reconstructed) cohort studies, retrospective
(case-control) studies, cross-sectional studies and follow-up studies;
• experimental strategies, including animal studies, therapeutic clinical
trials, prophylactic clinical trials
• operational strategies, which include observation and time-motion
studies.
6. Work Plan
• A realistic work schedule is equally important for both beginning
researchers working on a thesis or dissertation and for experienced
researchers working under the deadlines of a research grant or
contract. The existence of deadlines typically necessitates careful
budgeting of time.
 A work plan is a schedule, chart or graph that summarizes in a clear fashion, various
components of a research project and how they fit together.
 It includes:
• The tasks to be performed
• The dates at which each task should begin and be completed.
• Who will perform the task and the time, which will be spent on each task? How to
develop a work plan?
• Compile a list of tasks needed to do for the study in chronological order if possible.
• Identify the amount of time needed per task
• Review to see if the plan is realistic
• Fix the dates (in weeks) indicating the period in which each task will be completed and
draw it in the form of a chart
7. Data Collection
• A short description of plans for collecting data should be included in
the research proposal in order to minimize the possibility of
confusion, delays and errors. Pilot testing of the research methods
and research design, when appropriate, should be included as part of
the project. The reliability and validity of measurements should be
adequately addressed.
8. Analysis and Interpretation
• Plans for analysis are an integral part of the research de-sign, since
they can prevent the investigator from discovering at the end of the
study that certain required information has not been collected, that
other information cannot be included in the analysis or that some
data have not been gathered in a form appropriate for statistical
analysis.
• Ethical issues such as taking informed consent, maintaining
confidentiality, etc. have to be taken into account. Pretesting or
undertaking a pilot study helps in identifying unseen problems and
gives an opportunity to rectify.
9. Reporting
• Tentative plans for disseminating research results should be outlined.
Major emphasis should be placed, in these plans on distribution of
results to potential users at both national and local levels.
LIMITATIONS OF HSR

• Difficult in randomization and equal participation of subject,

• High turnover: retirement, transfer, change, etc

• Difficult to match with relevance policy and community’s expectation,

• Cost and benefits: control and distribution

• Implementation of results with scare resources

• Donor interest and influence

• Access/permission , publication and dissemination of results

HEALTH STATISTICS
Health Statistics

• Health Statistics provide information for understanding, monitoring,

improving and planning the use of resources to improve the lives of
people, provide services and promote their well being.

• Numerical information (data) is everywhere.

• Health statistics promote population health and enhance societal well

being .

• Statistical techniques are used to make decisions that affect our daily lives.
• No matter what line of work you select, you will find yourself faced with decisions where an

understanding of data analysis is helpful.

• In order to make an informed decision, you will need to be able to:

1. Determine whether the existing information is adequate or additional information is required.

2. Gather additional information, if it is needed, in such a way that it does not provide misleading

results.

3. Summarize the information in a useful and informative manner.

4. Analyze the available information.

5. Draw conclusions and make inferences while assessing the risk of an incorrect conclusion.
Types of Statistics
1. Descriptive statistics
2. Inferential statistics
Descriptive statistics
• Collection, presentation and analysis of numerical data.
• Describe characteristics of data set under study in a meaningful way.
• Mean, median, mode standard deviation, variance, percentage etc
are the methods used in descriptive statistics.
Inferential statistics
• Inferential statistics use measurements from the sample of subjects in
the experiment to compare the treatment groups and make
generalizations about the larger population of subjects.
• Hypothesis testing, confidence intervals, regression analysis, analysis
of variance (ANOVA), and chi-square tests are examples of inferential
statistics tools.
Types/Variables of Data
1. Categorical/Qualitative Data
• Categorical data represents characteristics. Therefore, it can represent
things like a person's age, gender, language, etc. categorical data can
also take on numerical values (e.g. 1 for female and 0 for male). Note
that those numbers don't have any mathematical meaning.
i. Nominal Data
ii. Ordinal Data
Nominal data
• Nominal values represent discrete units and are used to label variables, that
have no quantitative value. Data may be divided into groups e.g. male/female,
black /white, rural/urban.
• Nominal data that have only 2 categories is known as dichotomous. Example of
nominal data divided into more than two categories is: What language do you
speak?
English
French
Arabic
Urdu
Punjabi
Ordinal Data
Putting data in some order or rank e.g.
• Upper class, middle class, lower class
• People with income below 10000 Rs/month, 10,000 to 25,000
Rs/month and above 25000Rs/month
• Educational background elementary, highschool, undergraduate,
graduate and masters
Presentation
Qualitative data is presented by:
• Bar Diagram• Map diagram• Pictogram• Pie diagram
2. Numerical/Quantitative Data

1. Discrete Data
When you just count the number - e.g. 10 pencils, number of live births in a
hospital, accidents in a city. One variable is constant; other changes e.g. count
students with pulse rate of 72/minute.
2. Continuous Data
This is measured on some scale. It has no ends. So it is also called continuous
data. It may be interval or ratio. Both variables change e.g. counting the pulse
rate of the class, measuring BP.
Presentation
Quantitative data is represented by:• Frequency Polygon• Frequency Polygon
Curve• Histogram• Line Diagram• Scattered Diagram
What test should I use?
• What is the purpose of study?
• What type of data?
• Are the data normally distributed?
• How many groups/samples?
• Are the groups independent or dependent?
• What is the sample size
When Should I Use Chi-Square Test ?
• You would like to compare between two variables
• Both variables are categorical
When Should I Use Independent t-test ?
• You have 2 groups of subjects or programs or services
• You would like to compare the means outcome (dependent)
variable for two independent groups
• Outcome variable: continuous measurement
When Should I Use Paired t-test ?
• A paired (samples) t-test is used when you have two related
observations (i.e. two observations per subject)
• You want to see if the means on these two normally distributed
interval variables differ from one another.
When Should I Use Correlation Analysis?
• Correlation analysis measures the relationship between two
items (x and y variables)
• The 2 variables are continuous measurement
When Should I Use One-way ANOVA?
• You have 3 or more groups
• You wish to test for differences in the means of the outcome
(dependent) variable
Causality Assessment
• Causality assessment of ADRs is a method used for estimating the
strength of relationship between drug(s) exposure and occurrence of
adverse reaction(s).
• Causality assessment of ADRs may be undertaken by clinicians,
academics, the pharmaceutical industry and regulators, and in
different settings, including clinical trials.
• How close is the relationship between medicine and event?
• relationship

• Was the event caused by the medicine?

• causality
• Individual case safety reports (ICSRs)
• Single reports of suspected reactions
• (spontaneous reporting)

• Epidemiological
• Large numbers of reports of events
• Includes CEM
• Adverse reaction: “a response to a medicine which is noxious and
unintended, and which occurs at doses normally used in man”

• Adverse event: any new clinical experience that occurs after

commencing a medicine, not necessarily a response to a medicine,
and is recorded without judgement on its causality.
Measurement of Statistical
Data
TYPES OF MEASUREMENTS APPLIED
ON DATA
There are following types of measurements, which are applied on data:
Mean Median
Mode
Standard Deviation
Coefficient of Variation
Standard Error
Confidence Interval
Student’s t-Test
Z-test
Chi-Square Test
Correlation Analysis
Arithmetic Mean
• It is the value which is obtained by dividing the sum of all observations by their total
number.
• It is denoted by "x̄" (X-bar) and calculated as x̄=∑X/n
Where
• ∑= Summation
• X = Variable
• n = Total number of observations
Calculation of Mean (X):
• The diastolic blood pressure of 10 individuals was 8375, 81, 79, 71, 95, 75, 77, 84
and 90.
x̄ = ∑X/n = 810/10 = 81.0
Advantage
• It is easy to calculate and understand
Disadvantage
• It is influenced by some extreme values in the distribution
• Sometimes it may be ridiculous e.g., average no. of children born to
woman in a certain area is 5.6, which never occurs in reality
Median

The middle most item of data is called median. It does not depend
upon the total and the number of items.
To obtain median:
• Arrange data in an ascending or descending order
• If total number of observations or data is an odd number, then the
middle item is the median
• If total number of observations or data is an even number, then find
median by taking the average of middle two values
• Example:
• The diastolic BP of 5 individuals is 60, 70, 75, 90 & 80. Calculate
median
• a. Arrange data in an ascending or descending order.
• b. As number of items is 5, an odd no. the middle item or median is
75
Mode
The most commonly occurring value in a data is mode. e.g., 1, 3, 4, 5, 4,
4, 9, 6 & 3. Mode is 4. The most common disease is mode.
Advantage
It is easy to understand and not affected by extreme items.
Disadvantage
The exact location is often uncertain and is not often clearly defined. In
small number of cases, there may be no mode at all because no value
may be repeated. Therefore, it is not used in medical sciences.
Standard Deviation/ Root Mean Square
Deviation
• SD is defined as the positive square root of the arithmetic mean of
the square deviation from the mean of distribution.
• It is denoted by S.D or σ(Sigma). It is the most frequently used
measure of deviation.
• It is also called root mean square deviation.
• If sample size is more than 30, then above formula is used without
modification
• If sample size is less than 30, then the formula used is:
Coefficient of Variation
• Standard deviation expressed as percentage of mean.
• It compares two quantities irrespective of units.
STANDARD ERROR
• This relationship of standard deviation to sample size is known as
Standard Error (SE).So Standard Error tells us that how much away
would be our sample mean from population mean
• Take a sample. Calculate mean. Take similar sample again & again and
calculate mean. We find that every sample has different mean. The SD
of means is the measure of Standard Error.
• The standard error is dependent on the size of the sample: standard
error is inversely related to the square root of the sample size, so that
the larger the "n" becomes, the more closely will the sample means
represent the true population mean.
Confidence Interval
• Suppose we want to know population mean(µ) from the sample mean
can we estimate that? We can not exactly determine the population
mean with the help of sample.
• However we can construct an interval between which the population
mean (µ) may be. This interval is known as confidence interval. Now
the question is, are we one hundred percent confident about
population mean? As the results are sample based we can not be
100% sure. Though we would like to be 100% confident.
• As the estimate of µ with the help of sample mean depends upon the
relationship of S.D. and sample size (known as SE).
• The formula for population mean will be

• As we would construct an interval for u so the value would be both in

+ and - direction.
CHI-SQUARE TEST
• Chi-square test offers a method of testing the significance of
difference between two proportions.
• It can also be used when two groups are to be compared:

• ∑= Summation
• 0 = Observed value
• E = Expected value
STUDENT'S t-TEST
• T compares two averages (means) and tell that if they are different
from each other.
• It is a test of significance (tells how significant differences are),
designed by Williams Sealy whose pen name is Student.
• Formula
Z-TEST
• This test of significance applied when sample size is more than 30.
• Calculated by the following formula:
Correlation
• Assessing the strength of the association between two variables. The
correlation coefficient between two variables.
• The correlation coefficient denoted symbolically as r, defines both the
strength and the direction of the linear relationship between two
variables.
• The correlation coefficient is an index number between-1 and +1
References
• Pharmaceutical Practice by Raees, Smith, Watsan. Chapter 22
• Public Health and Community Medicine by Muhammad Irfanullah
Sadiqui, chapter 8.
• Clinical Pharmacy by Irfan Bashir.
• Foundation of Clinical Research, 4th Edition, chapter 38.