My notes by Rupali Sabale

Epidemiological Wheel Theory

rupalisabale — 2019-01-17 15:18:51 UTC

The theory visualizes human disease in the form of a wheel,
which has a central hub representing the genetic components
and the peripheral portion representing the environmental
component. Like any wheel, the outer part (environmental
component) has spokes (3 in this model) and the environmental
component is thus divided into 3 sub components, representing
the social, biological and physical components of the
environment. For every disease, the genetic, social, biological
and physical environmental components take different sizes

The Theory of “Necessary” and “Sufficient” cause

rupalisabale — 2019-01-17 15:21:23 UTC

necessary cause is one whose presence is essential for
disease causation, but which alone, by itself may or may not
be able to finally cause the disease.
At
times, the necessary cause may by itself be a sufficient cause
for the disease (e.g. infection with HIV is a “necessary” as well
“sufficient” cause for AIDS). More often, however, in most of
the infectious diseases, the necessary cause by itself may not
be a sufficient cause (example of Tubercular disease given
earlier). Finally in most of the non-communicable diseases one
may not be able to identify any “necessary” cause at all; what
is important in these diseases is whether the optimum mix of
various factors has occurred in a way to produce the “sufficient
cause”

Induction and Latent Periods

rupalisabale — 2019-01-17 15:30:54 UTC

induction period is the time interval that elapses between
the point when “sufficient cause” becomes operative to the
point when the first pathological changes of disease start in
the body. On the other hand latent period is the time period that
elapses between the initiation of pathological change till the
point when the earliest diagnosis of the disease is possible with
the currently available clinical and diagnostic armamentarium.
The counterpart in infectious disease epidemiology is the
“incubation period” which refers to the time period that elapses
between the entry of the infectious agent into the body till the
point the earliest manifestation of the disease are evident. The
“generation time” on the other hand refers to the time between
entry of the infectious agent into the body and the maximum
infectivity of the human host

Predisposing factors

rupalisabale — 2019-01-17 15:37:58 UTC

factors which create a state of
susceptibility, so that the host becomes vulnerable to
the agent or to necessary cause, e.g. age, sex, previous
illness

Enabling factors

rupalisabale — 2019-01-17 15:42:35 UTC

those which assist in the development
of (or in recovery from) the disease; e.g. housing conditions,
socio-economic status.

Precipitating factors

rupalisabale — 2019-01-17 15:44:35 UTC

those which are associated with
immediate exposure to the disease agent or onset of
disease, e.g. drinking contaminated water, close contact
with open case of pulmonary TB.

Reinforcing factors

rupalisabale — 2019-01-17 15:45:35 UTC

those which aggravate an already
existing disease, e.g. malnutrition, repeated, exposures,
etc.

Risk factors

rupalisabale — 2019-01-17 15:47:04 UTC

: A risk factor is defined as a condition, quality
or attribute, the presence of which increases the chances
of an individual to have, develop or be adversely affected
by a disease process.

Risk Marker

rupalisabale — 2019-01-17 15:50:08 UTC

There is a fine difference between a risk
factor and a risk marker.
It is only a surrogate or a
proxy indicator of the risk of the disease, mostly because
of it’s association with another risk factor. For example,
development of a transverse crease on the ear lobe has
been shown, in many studies, to indicate an increased risk
for IHD. However, even if we correct the ear lobe crease
by a plastic surgery procedure, the risk of IHD will not
reduce!

Risk Groups and Risk Approach

rupalisabale — 2019-01-19 05:42:40 UTC

A high risk group can be defined as a group of
people or a subsection of the population, who, by virtue
of certain characteristics, are likely to have a higher
probability of suffering from one or more diseases or from
general ill health

Natural History of Disease

rupalisabale — 2019-01-20 19:07:24 UTC

Health Promotion

rupalisabale — 2019-01-21 13:39:22 UTC

These include all steps undertaken to
improve the level of general health and well being so that
conditions for initiation of disease process are prevented.
However, these steps are not specific for any disease or a group
of diseases.

Specific Protection

rupalisabale — 2019-01-21 13:40:53 UTC

These include measures to prevent the
initiation of specific diseases or a group of diseases. Examples
include immunization to protect against specific diseases,
fortification of foods with specific nutrients (as salt with
iodine), use of condoms to protect against sexually transmitted
diseases (STDs),

Primary prevention

rupalisabale — 2019-01-21 13:42:24 UTC

focuses on steps so that the disease process
is not initiated, i.e. the stage of pre-pathogenesis is not crossed,
though the adverse agent, host and environmental factors may
be present. For instance, in the example of the natural history
of IHD given above, if we educate and motivate the young
man not to start smoking, despite the fact that cigarettes
are available, we are practicing primary prevention

primordial prevention

rupalisabale — 2019-01-21 13:43:46 UTC

if we create conditions in which cigarettes
are not available or even tobacco is not produced, then we are
preventing the development of those very conditions which
lead to the state of pre-pathogenesis; hence this would be
primordial prevention.

Secondary Prevention

rupalisabale — 2019-01-21 13:44:59 UTC

The person is not aware of any
signs or symptoms and the routine clinical methods also may
not be able to detect a disease at this stage, since the disease
process is in the very preliminary stage. However, by using
special procedures we can uncover a disease in such early stages.
The classical example of this level of prevention is “screening
for disease” as for breast cancer (using mammography) and
cervical cancer (using pap smear).

Tertiary Prevention

rupalisabale — 2019-01-21 13:46:05 UTC

These include all measures undertaken when the disease
has become clinically manifest or advanced, with a view to
prevent or delay death, reduce or limit the impairments and
disabilities, minimize suffering and to promote the subject’s
adjustment to irremediable conditions.

Disability Limitation

rupalisabale — 2019-01-21 13:47:33 UTC

These include all measures to
prevent the occurrence of further complications, impairments,
disabilities and handicaps or even death
When we give complete rest, morphine, oxygen and streptokinase
to a patient of Acute MI, we are actually trying to prevent death
or complications like arrhythmias / CHF.

Rehabilitation

rupalisabale — 2019-01-21 13:48:52 UTC

Rehabilitation stands for the combined and
coordinated usage of all the available medical, social,
educational and vocational measures, for training and
retraining the person to the highest level of functional ability.
Medical rehabilitation : This is done through medical / surgical
procedures to restore the anatomy, anatomical functions and
physiological functions to as near normal as possible.
Vocational rehabilitation : This includes steps involving
training and education so as to enable the person to earn a
livelihood.
Social rehabilitation : This involves steps for restoration of
the family and social relationships.
Emotional and Psychological rehabilitation : This involves
steps to restore the confidence, personal dignity and confi

The Iceberg Phenomena in Human Diseases

rupalisabale — 2019-01-21 13:51:20 UTC

public health professionals should be
careful while making an assessment of the load of diseases and
their associated risk factors in the community, remembering
that the real load lies hidden in the community and is not
clinically apparent. This will help preventing failures which
may otherwise occur consequent to inadequate planning due
to underestimation, in public health programmes.

rupalisabale — 2019-01-23 13:14:52 UTC

Epidemiology.. eg of malaria

rupalisabale — 2019-01-23 14:53:38 UTC

Epidemiology

rupalisabale — 2019-01-23 14:59:13 UTC

“The study of the frequency, distribution and determinants
of diseases and health - related states and events in human
populations” and the application of this knowledge in
prevention, control and mitigation of these problems (9 - 12) .
(Greek; Epi = upon, Demos = populations, Logos = scientific
study).

Data

rupalisabale — 2019-01-25 13:38:42 UTC

“an
organised collection of information, containing the ‘values’
of the various variables, obtained from a sample of subjects,
and which would be subsequently used to derive conclusions
through the process of scientific analysis and reasoning”.

Variables

rupalisabale — 2019-01-25 13:39:40 UTC

any
quality, characteristic or constituent of a subject which can be
measured and which ‘varies’, i.e. likely to have a different value
from one subject to another.

Research hypothesis

rupalisabale — 2019-01-25 13:41:15 UTC

The research question, when expressed in a statement form
(rather than an interrogative form) becomes the research
hypothesis, e.g.
would be “Drinking 5 litres of water can reduce the risk
of renal stone formation by 25% among young, physically
active males, in tropical climate, after duly considering the
confounding effects of primary renal disease, dietary factors,
racial background, occupation, hypercalciuria, hyperoxaluria,
and hyperuricosuria”. The research hypothesis is a tentative
guess which we wish to test and finally accept or reject in our
proposed research work.

null hypothesis

rupalisabale — 2019-01-25 13:42:38 UTC

it is simply the
research hypothesis stated in a negative manner. For example,
the null hypothesis for the above mentioned research hypothesis
would be “Drinking of 5 litres water per day does not have any
effect in reducing the risk of renal stones (by 25% at least) among
young, physically active males, in tropical climate, after duly
considering the confounding effects of primary renal disease,
dietary factors, racial background, occupation, hypercalciuria,
hyperoxaluria, and hyperuricosuria”.

Concept of Random Error

rupalisabale — 2019-01-25 13:45:10 UTC

This is an error which occurs
because, as long as you are studying a sample, your results
from the sample will be different from the actual reality that
exists in the total population from which you have drawn the
sample. This is a natural phenomena which will occur despite
your most honest and meticulous efforts.

Research questions

rupalisabale — 2019-01-25 13:56:21 UTC

These should be:
• Specifi c with respect to time/place/subjects/condition as appropriate
• Answerable such that the relevant data are available or able to be
collected
• Novel in some sense so that the study either makes a contribution to
knowledge or extends existing knowledge
• Relevant to current medicine

Convenience sample

rupalisabale — 2019-01-25 14:01:17 UTC

sample of patients available at a particular time/place

Quota sample

rupalisabale — 2019-01-25 14:03:11 UTC

the researcher aims to identify a representa-
tive sample by choosing subjects in proportion to their numbers in the
population of interest. For example if age, marital status, sex, and employ-
ment status were important characteristics, then the researcher would
select a number of subjects with each combination of these characteris-
tics so that the overall proportions with the characteristics refl ected the
proportions in the population.

Random sample (simple random sample)

rupalisabale — 2019-01-25 14:04:35 UTC

A random sample is chosen so that each member of the population has
an equal chance of being chosen and so the selection is completely inde-
pendent of patient characteristics. In order to draw a random sample a list
of the population is needed, the sampling frame.

Stratifi ed sample

rupalisabale — 2019-01-25 14:05:29 UTC

Stratifi ed samples are used when fi xed numbers are needed from partic-
ular sections or strata of the population in order to achieve balance across
certain important factors. For example a study designed to estimate the
prevalence of diabetes in different ethnic groups may choose a random
sample with equal numbers of subjects in each ethnic group to provide a
set of estimates with equal precision for each group. If a simple random
sample is used rather than a stratifi ed sample, then estimates for minority
ethnic groups may be based on small numbers and have poor precision.
In terms of effi ciency, a stratifi ed sample gives the most precise overall
(weighted) estimate, where the overall estimate is weighted according to
the fractions sampled in each stratum.

Cluster sample

rupalisabale — 2019-01-25 14:07:27 UTC

Cluster samples may be chosen where individuals fall naturally into groups
or clusters.Having chosen the clusters, the
researcher can either select all subjects in the cluster or take a random
sample within the cluster. Cluster sampling is less effi cient statistically than
simple random sampling and so needs to be accounted for in the sample
size calculations and subsequent analyses

Type 1 error

rupalisabale — 2019-01-25 14:10:42 UTC

We conclude that there is a difference between the groups in the
target populations when in fact there is not.

Type 2 error

rupalisabale — 2019-01-25 14:12:52 UTC

We conclude that there is no difference between the groups in the
target population when in fact a real difference of a given size does
exist.

Face validity

rupalisabale — 2019-01-25 14:16:13 UTC

Is the scale measuring what it sets out to measure? This is a subjective
assessment and is achieved by consensus among experts.

Internal consistency: Cronbach’s alpha

rupalisabale — 2019-01-25 14:17:13 UTC

If a scale has several questions or items which all address the same issue
then we usually expect each individual to get similar scores for those
questions, i.e. we expect their responses to be internally consistent.
A statistical quantity, Cronbach’s alpha, is often used to assess the degree
of internal consistency.
Cronbach’s alphais calculated as an average of all correlations among
the different questions in the scale. It can be interpreted as follows:
• Alpha lies between 0 and 1
• Values are usually expected to be above 0.7 and below 0.9
• Alpha below 0.7 broadly indicates poor internal consistency
• Alpha above 0.9 suggests that the items are very similar and perhaps
fewer items could be used to obtain the same overall information

Convergent/criterion/concurrent validity

rupalisabale — 2019-01-25 14:21:05 UTC

This can be tested by comparing it with another similar scale, where one
exists, to see if both give a similar result. For example in developing a
shortened version of an existing but longer questionnaire it is important
to ensure that the short version gives comparable results with the longer
version.

Construct validity

rupalisabale — 2019-01-25 14:21:53 UTC

Construct validity

To assess construct validity where there are no similar scales to compare

with, the researchers apply a series of tests where the answer is known to

check that the scale is behaving as expected.
Example - In developing a new questionnaire scale to measure respiratory symp-
toms we would expect that patients from a chronic obstructive pul-
monary disease (COPD) clinic would score higher than patients from a
fracture clinic, and that patients’ scores would change before and after
exercise etc.
These comparisons show that the scale is working as expected and so
has construct validity.

Research articles: guidelines

rupalisabale — 2019-01-25 15:46:30 UTC

Equator network
The Equator network (Enhancing the QUAlity and Transparency Of health
Research) is ‘an international initiative that seeks to enhance reliability and
value of medical research literature by promoting transparent and accu-
rate reporting of research studies.’ (M www.equator-network.org)
The Network brings together a wide range of resources relating to
health research. Up-to-date lists and links can be found on the website,
and some of those available at the time of writing are as follows:
TREND: non-randomized controlled trials
M www.trend-statement.org
STARD: studies of diagnostic accuracy
M www.stard-statement.org
PRISMA: systematic reviews and meta-analyses (replaces QUOROM)
M www.prisma-statement.org/
STROBE: observational studies in epidemiology
M www.strobe-statement.org
MOOSE: meta-analyses of observational studies in epidemiology
M www.consort-statement.org/mod_product/uploads/MOOSE%20
Statement%202000.pdf

Disease outbreak..

rupalisabale — 2019-01-25 18:05:46 UTC

Disease under surveillance

rupalisabale — 2019-01-25 18:07:52 UTC

Internal validity

rupalisabale — 2019-01-26 11:07:44 UTC

if the study itself has been undertaken
using wrong methods (untrained laboratory workers,
haemolysed blood samples, outdated kits, etc.), then it leads
to loss of internal validity. If, in a study, internal validity has
been compromised or lost (i.e. ‘bias’ has come in) , the study
becomes fit for rejection only

External validity

rupalisabale — 2019-01-26 11:10:21 UTC

External validity, if lost, leads to loss of generalizability of
results to the total population which the investigator has in
mind. However, the results can still be generalised to another
“population” of which the sample, so drawn, is representative

Ratio

rupalisabale — 2019-01-26 11:27:59 UTC

“ratio” is a measure of frequency in which the numerator
is not included in the denominator. The ratio also gives us the idea of another very
important measure - the “odds”

Incidence Density

rupalisabale — 2019-01-26 11:50:21 UTC

ID gives the “forces of morbidity” (or
mortality) due to a disease at given point of time. ID is also
often called as “person - time in epidemiologic practice”. Very
frequently, ID is referred to as “Incidence Rate” and sometimes
as “Hazard”, since it also gives an idea of the hazard that a
subject has of developing an outcome at a given point of time,

Difference between cumulative incidence and incidence Density

rupalisabale — 2019-01-26 11:55:38 UTC

Types of incidence measures

rupalisabale — 2019-01-26 12:11:19 UTC

1) Cumulative incidence or incidence proportion 2) Incidence Density or incidence rate

Incidence proportion or risk or Cumulative incidence

rupalisabale — 2019-01-26 12:18:12 UTC

Incidence proportion is the proportion of an initially disease-free population that develops disease, becomes injured, or dies during a specified (usually limited) period of time. Synonyms include attack rate, risk, probability of getting disease, and cumulative incidence. Incidence proportion is a proportion because the persons in the numerator, those who develop disease, are all included in the denominator (the entire populat

rupalisabale — 2019-01-26 12:44:06 UTC

In an outbreak of gastroenteritis among attendees of a corporate picnic, 99 persons ate potato salad, 30 of whom developed gastroenteritis. Calculate the risk of illness among persons who ate potato salad.

Numerator = 30 persons who ate potato salad and developed gastroenteritis
Denominator = 99 persons who ate potato salad
10ⁿ = 10² = 100

Risk = "Food-specific attack rate" = (30 ⁄ 99) × 100 = 0.303 × 100 = 30.3%

Incidence rate or person-time rate

rupalisabale — 2019-01-26 12:52:42 UTC

Incidence rate or person-time rate is a measure of incidence that incorporates time directly into the denominator. A person-time rate is generally calculated from a long-term cohort follow-up study, wherein enrollees are followed over time and the occurrence of new cases of disease is documented. Typically, each person is observed from an established starting time until one of four "end points" is reached: onset of disease, death, migration out of the study ("lost to follow-up"), or the end of the study. Similar to the incidence proportion, the numerator of the incidence rate is the number of new cases identified during the period of observation. However, the denominator differs. The denominator is the sum of the time each person was observed, totaled for all persons. This denominator represents the total time the population was at risk of and being watched for disease. Thus, the incidence rate is the ratio of the number of cases to the total time the population is at risk

Reporting of Incidence Density or Incidence rate

rupalisabale — 2019-01-26 13:00:52 UTC

if you report the incidence rate of, say, the heart disease study as 2.5 per 1,000 person-years, epidemiologists might understand, but most others will not. Person-time is epidemiologic jargon. To convert this jargon to something understandable, simply replace "person-years" with "persons per year." Reporting the results as 2.5 new cases of heart disease per 1,000 persons per year sounds like English rather than jargon. It also conveys the sense of the incidence rate as a dynamic process, the speed at which new cases of disease occur in the population.

Example of Incidence Density

rupalisabale — 2019-01-26 13:09:50 UTC

Now, the total time of follow up given by our subjects was
(10+10+5+ …… +4+6) = 66 years (or better called as 66
person years). Out of these 66 “person - years” of follow up,
2 new cases of cervical cancer occurred among our subjects,
who were initially “at risk” of developing cervical cancer. The
Incidence Density of Cervical Cancer from our study is thus,
ID = 2/66 Person years = 0.03 per person year (or multiplying
by 100, it becomes 3 per 100 PY). (The figure is also written as
0.03 per PY - 1 or as 3 PY - 100).
This means that in that particular age group of ladies, there
will be 3 cases of cervical cancer if 100 person - years of follow
up is done.

Types of Prevalence Measures

rupalisabale — 2019-01-26 13:18:30 UTC

1) Point prevalence 2) Period Prevalence

Point Prevalence

rupalisabale — 2019-01-26 13:19:30 UTC

the number of cases of the disease refer
to a single point of time, though, of course, the actual study
on such prevalence may take a long time, may be months
or even years. The essential feature is that while the actual
process of assessment for presence or absence of the disease
in the sample may take many months, the study subjects are
examined only once, and the study is deemed to have taken
place at a particular point of time. Most of the comparisons in
medical research are made using “point prevalence”.

Period Prevalence

rupalisabale — 2019-01-26 13:29:47 UTC

It essentially involves calculation of
both, the point prevalence at the time of starting the study, as
well as the incidence over the subsequent period of follow up. A
combination of the initial point prevalence and the subsequent
incidence gives us the period prevalence.

collection of data in a census

rupalisabale — 2019-01-27 05:56:23 UTC

(a) de - facto method : Persons are enumerated according to
their location at the time of enumeration. This method is used
in developing countries like India.
(b) de - jure method : This method is used in developed
countries like U.S.A. The persons are assigned according to their
“usual” place of residence and not according to their location
at the time of census, as practiced in de - facto method.

Terms in research

rupalisabale — 2019-01-27 06:12:59 UTC

the terms
“exposure”, “predictor”, “independent” and “cause” variable are
used interchangeably; similarly, the terms “outcome”, “effect”
and “dependent” variable stand for the same meaning.

Relative Risk

rupalisabale — 2019-01-27 06:17:22 UTC

The Relative Risk (RR) gives an idea of the number of times
that the Incidence is likely to be more among those who have
the exposure compared to those who do not have the exposure.
An increasing value of RR indicates increasing quantum of
risk. If the RR is very high (say, 5 or more), we are convinced
that, particular exposure carries a very high risk of the outcome
to develop. In other words, higher the RR, there is a higher
likelihood of an association between the exposure and the
outcome. The RR indicates, in addition to the magnitude, the
“strength of association” between an exposure and the outcome
variable

What happens if the RR is equal to one or less than one ?

rupalisabale — 2019-01-27 06:18:36 UTC

If RR is equal to one, then risk of getting disease in exposed and non exposed people is same. If it is less than one, it
indicates declining risk, i.e. a protective effect.

Odds Ratio (OR)

rupalisabale — 2019-01-27 06:52:34 UTC

there are two
assumptions which should be met with :
(a) The outcome, i.e. the disease of interest (e.g. IHD in this
case) should be a ‘rare disease”, i.e. the incidence of the
outcome should be <5% in the total population. Luckily,
most of the human diseases are uncommon ones, with this
parameter of <5%.
(b) The “controls” i.e. the comparison group (the 50 healthy
persons) should be “representative” of the same total
population that gave rise to the “cases”.

Why rare diseases?

rupalisabale — 2019-01-27 07:05:15 UTC

when the disease outcome of interest is rare, the odds ratio of exposure can be used to estimate the relative risk

Interpretation of odds ratio

rupalisabale — 2019-01-27 07:08:06 UTC

The odds ratio in a case control study basically measures the
“exposure odds ratio”.
The “odds” of a case being “exposed” = Ratio of “exposed”
cases to “non - exposed” cases = a/c and similarly, the odds of
a control being exposed = b/d. Thus, the “Ratio” of the “odds of
exposure among cases” to “odds of exposure among controls”

Technically speaking therefore, the odds ratio from a case
control study means that the “odds” of cases being exposed are
“so many times higher” compared to the “odds of controls being
exposed”. In our above example, the correct interpretation of
OR = 1.94 is that a case of IHD is 1.94 times more likely to
be having the exposure (i.e. history of smoking) as compared
to a person not having IHD. However, if the 2 assumptions,
enumerated above are met, the OR approximates the RR and
so we conclude that the risk of IHD is 1.94 times more if one
smokes

Interpreting Increase or Decrease in RR or OR in Terms of Multiples or Percentage Change

rupalisabale — 2019-01-27 07:24:44 UTC

In general, if we
have to calculate the percent increase (or decrease) in risk for a
given value of RR (or OR), the general equation is :
“Percentage increase or decrease in risk among the exposed
group = {(RR or OR) ≈ 1} X 100” (where the symbol ≈
means “difference, ignoring the minus or plus sign”). Thus if
OR is 1.85, the percentage increase in risk will be (1.85 ≈ 1) X
100 = 85%; if RR is 0.4 the % decrease in risk will be (0.40 ≈
1) X 100 = 60%. Similarly if RR is 8, it would also be stated as
an increase in risk by 700%.

Measures of Risk difference (impact) in a forward study

rupalisabale — 2019-02-02 14:10:05 UTC

(a) Measure of Absolute Effect b) (b) Measures of Attributable Fractions (Impact)

Measure of Absolute Effect

rupalisabale — 2019-02-02 14:13:06 UTC

The difference in two
incidence rates therefore gives us the “absolute effect” and
is measured in terms of “Attributable Risk” (AR).
Thus, AR = Ie - Io. example, AR = 0.02 - 0.01 (or
1%). We would describe this finding as “there is likely to be one
‘additional’ case of IHD among every 100 smokers as compared
to 100 non smokers”.

Cohort Studies

rupalisabale — 2019-02-24 16:05:19 UTC

In a cohort study, the investigator selects a group of
exposed individuals and a group of nonexposed
individuals and follows up both groups to compare
the incidence of disease (or rate of death from
disease) in the two groups

Prospective cohort study (also a concurrent cohort or longitudinal study).

rupalisabale — 2019-02-24 16:26:57 UTC

It is concurrent because the investigator
identifies the original population at the beginning
of the study and, in effect, accompanies the subjects
concurrently through calendar time until the point
at which the disease develops or does not develop.

Retrospective cohort or historical cohort study (also called a nonconcur-rent prospective study).

rupalisabale — 2019-02-24 16:30:21 UTC

we are still comparing
exposed and nonexposed groups; what we have
done in the retrospective cohort design is to use
historical data from the past so that we can tele-
scope the frame of calendar time for the study and
obtain our results sooner.

Epidemiological study design

rupalisabale — 2019-02-24 16:46:50 UTC

EPI 30 Cluster survey

rupalisabale — 2019-05-08 05:21:59 UTC

Validity of test

rupalisabale — 2019-05-17 08:20:32 UTC

The validity of a test is defined as its ability to dis-
tinguish between who has a disease and who does not. Validity has two components: sensitivity and
specificity.

sensitivity of the test

rupalisabale — 2019-05-17 08:21:57 UTC

The sensitivity of the test is defined as
the ability of the test to identify correctly those who
have the disease

specificity of the test

rupalisabale — 2019-05-17 08:22:45 UTC

The specificity of the test is defined
as the ability of the test to identify correctly those
who do not have the disease.

Calculation of sensitivity and specificity

rupalisabale — 2019-05-17 08:27:11 UTC