BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
HEALTH INDICATORS SOURCES OF DATA FOR HEALTH
INDICATORS
It is a way of measuring specified health
characteristics in a given population A. CENSUS OF THE COUNTRY’S
Defined as summary measures that capture POPULATION
relevant information on different attributes - Complete count of the population of a given place
and dimensions of health status and - First recorded value of the population of Philippines
performance of a health system was documented by Fr. Buzeta in 1799 with a count
An indicator may provide a direct estimate of of 1, 504, 574
the underlying condition of interest -First known Philippine census was conducted in
It can also serve as a proxy for a condition or 1877 with a count of 5,984,727
phenomena which is difficult to measure - The most recent census was conducted in 2015
with a count of 100,979,303
USES OF HEALTH INDICATORS
a) To define public health problems at a VITAL REGISTRATON SYSTEM
particular point of time -VRS, requires basic and compulsory registration of
b) To indicate changes over time in the level of all births, deaths and marriages occurring in the
health of an individual or a population country
c) To assess the extent to which the objectives -it was legally instituted in the country in February
of a program are being reached. 27, 1931
d) To inform, educate and empower people -PSA is in-charge of the VRS
about health issues. MAIN PROBLEM: under-registration of births and
death
CHARACTERISTICS OF INDICATORS
National Health Information System
VALID
The indicator measures what is supposed to be Major data sources within NHIS for public health
measured use:
REALIABLE 1. Philippine Integrated Disease Surveillance and
The indicator will have the same value even if it is Response (PIDSR)
measured by other people at different times - Enhanced surveillance system that monitors
under similar conditions notifiable diseases and other health-related events
SENSITIVE of public health importance
Able to detect small changes in the phenomenon Category 1: 10 diseases: measles, human avian
being measured in a significant way influenza, anthrax, neonatal tetanus, rabies and
SPECIFIC SARS among others
The indicator reflects changes only in the situation Category 2: 17 diseases: acute bloody diarrhea,
concerned cholera, dengue, leptospirosis, malaria, typhoid and
FEASIBLE paratyphoid fever, and pertussis among others
Collection of data to determine its value if
technically, financially and operationally possible. 2: Field Health Service Information System
(FHSIS)
• Intended to address the short-term needs of DOH
and LGU staff with
managerial or supervisory functions in facilities and
program areas.
• Provides summary on health service delivery and
selected program accomplishment indicators.
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
Hospital Operation and Management TYPES OF HEALTH INDICATORS
Information System (HOMIS) ACCORDING TO:
-Computer based system developed by DOH with the 1. Aspect of health being measured
objective of supporting the hospital management for 2. Group referred to
effective and quality health care 3. Time covered by the events
-To systematically collect, process, store, present 4. Components of the evaluation framework
and share information in support of hospital
functions ACCORDING TO THE ASPECT OF HEALTH
BEING MEASURED
Specialized Information Systems within the HEALTH STATUS INDICATORS
Health Sector - Measure health outcomes and/ or their
-Event/condition-specific information and risk factors
surveillance systems HEALTH SERVICE PERFORMANCE
-Surveillance in post-extreme emergencies and INDICATORS
disasters (SPEED) - Measure aspects of the performance of
-Online National Electronic Surveillance System health services or public health
(ONEISS) programs
-Phil. Malaria Information System (PhilMIS)
Disease Registries ACCORDING TO GROUP REFERRED TO
- Cancer Registry CRUDE DEATH RATES
- HIV/AIDS Registry - Describes the total population which is
used as the denominator in the
Sample Surveys computation
AGE SOECIFIC RATES
National Demographic and Health Survey - Describes only a specific sub-group of
(NDHS) the total population being considered.
-Collecting accurate and nationally representative
data on health, population and nutrition in EXAMPLE COMPUTATION:
developing countries
-Funded by USAID
-Started in 1984 and has covered 90 countries
including Philippines
-Important in policy making, program planning,
monitoring and evaluation of programs
Multiple Indicator Cluster Surveys (MICS)
-Series of household surveys conducted by UNICEF
covering important indicators describing the
situation of women and children
ACCORDING TO THE TIME COVERED BY
-First was conducted in 1995
THE EVENTS
*last MICS in the Philippines was in 2014
PREVALENCE TIME
- Occurred during the specific point in
Young Adult Fertility and Sexuality Study
time being considered
(YAFS)
-series of national surveys on Filipino youth aged 15- INCIDENCE TIME
24 - Occurred over a period of time
-one of primary sources on sexual and non-sexual ACCORDING TO THE COMPONENTS OF
risk behaviors and its determinants in the Philippines THE EVALUATION FRAMEWORK
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
1. INPUT INDICATORS MEASURES OF FERTILITY
- resources needed to deliver the FORMULA WHAT IT
INDICATOR
essential services to the population or MEASURES
to achieve project objectives # of children
2. Output indicators born per 1000
Crude Birth
- direct products of project activities. population in
Rate (CBR)
Generally, in the form of activities and a given place
processes undertaken and time
3. Outcome indicators # of children
- immediate result of the services or born per 1000
activities implemented women in the
4. Impact indicators General reproductive
- intended or unintended long-term Fertility Rate age groups in
organizational/community changes (GFR) a given place
during a
TYPE OF INDICATOR INDICATOR particular
-number of posters and time
brochures on family # of children
planning born per 1000
INPUT -number of community women in a
volunteers trained to particular age
conduct health group
Age-specific
education measures
Fertility rate
-number of couples fertility level
(ASFR)
who have seen or of each sub-
received the posters group of
OUTPUT
-number of health women
education classes according to
conducted age
-% increase in the level # of children
of knowledge among TFR= ∑ 𝑨𝑺𝑭𝑹 who will be
couples on family (for single year born per
OUTCOME planning age group) woman if she
Total Fertility
-% increase in number pass through
Rate (TFR)
of family planning ∑
TFR=𝟓 𝑨𝑺𝑭𝑹 the
acceptors (For 5 year childbearing
-decrease in the group) years bearing
average number of children
IMPACT children per family
-decrease in the MEASURES OF MORBIDITY
population growth rate
MEASURES OF DISEASE FREQUENCY
CONVENTIONAL HEALTH STATUS Magnitude of the disease in a given
INDICATORS population
MEASURES OF MORBIDITY
MEASURES OF MORTALITY
MEASURES OF FERTILITY
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
CHARACTERISTICS OF A NORMAL
ATTRIBUTE INCIDENCE PREVALENCE
DISTRIBUTION
Question How many people How many 1. Bell shaped and symmetric about the mean
being developed the people have 2. Mean=median=mode
answered disease? the disease?
Events in Number of new Number of
the cases of disease existing cases
numerator of a disease
Types of Acute Chronic
diseases
usually
referred to
Usual area Disease Service
of etiology/causation delivery
application
KURTOSIS- measure of whether the data are heavy-
Types of incidence measures tailed or light-tailed relative to a normal distribution
Incidence proportion (IP) or Cumulative Total area under the curve is 1 or 100%
Incidence (CI) Has long tapering tails that extend infinitely
Incidence Rate (IR) or Incidence Density (ID) in either direction but never touch the x-axis
Completely determined by two parameters,
MEASURES OF ASSOCIATION its mean and standard deviation
Quantifies the strength of the relationship
between the disease/condition of interest
and the corresponding exposure/risk factors
being assessed
LESSON 2:
NORMAL DISTRIBUTION AND
HYPOTHESIS TESTING
NORMAL DISTRIBUTION
Normal distribution is a distribution that is
symmetric about the mean
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
many statistical test assume normality of the
distribution
THE STANDARD NORMAL DISTRIBUTION
has mean=0 and SD=1
the capital Z is traditionally used to
represent the standard normal random
variable
small letter z is used to represent a particular
value of Z
any value x from the normal distribution can
Standard deviation increases, the be transformed into a standard normal value
distribution becomes wider’ of z using the formula
𝑋−𝜇
Standard deviation decreases, the 𝑍=
𝛼
distribution becomes thinner Areas under the standard normal are
tabulated
SOLVING THE NORMAL DISTRIBUTION
STEPS:
Compute for the equivalent z value of the
given x value
Sketch the graph
Get the area from the standard normal table
Z-SCORE FORMULA
1 SD covers 68% of the distribution 𝑋−𝜇
2 SD covers 95% of the distribution 𝑍=
𝜎
3 SD covers 99.7% of the distribution EXPLANATION
X= individual value
IMPORTANCE OF THE NORMAL DISTRIBUTION 𝜇 = 𝑚𝑒𝑎𝑛
𝜎 = 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
1. Useful for explaining many biological
phenomena EXAMPLES:
even if the distribution of the variable is not
normal, can easily transform using log, 1. suppose the height of 5 yr. old Filipino boys is
square root or other transformation to make normally distributed with a mean of 110 cm and
it approximate the normal distribution standard deviation of 5.61 cm. What is the
2. Plays an important role in statistical inference proportion of these boys that are taller than 121 cm?
because
the other important distributions (binomial,
t-distribution, can be approximated by the
normal esp. when the sample size is large
enough
The sampling distribution of the mean is
approximately normal if the sample size is
large enough (central limit theorem)
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
HOW TO USE A Z-TABLE.
To find the corresponding area under the curve
(probability) for a z-score:
1. Go down to the row with the first two digits
of your z-score.
2. Go across to the column with the same third
digit as your z- score.
3. Find the value at the intersection of the row
and column from the previous steps.
USE THE STANDARD NORMAL DISTRIBUTION TO
FIND PROBABILITY
Z-tests and p-values
The z-score is the test statistics used in a z-test. The
z-test is used to compare the means of two or
compare the mean of a group to a set value. Its null
hypothesis typically assumes no difference between
groups.
The area under the curve to the right of a z-score is
the p-value, and it's the likelihood of your
observation occurring if the null hypothesis is true. EXAMPLE: USING THE Z-DISTRIBUTION TO FIND
PROBABILITY
Usually, a p-value of 0.05 or less means that your
results are unlikely to have arisen by chance; it We've calculated that a SAT score of 1380 has a z-
indicates a statistically significant effect. score of 1.53. Using the full z-table, we find that for
a Z-score of the p-value is 0937.
By converting a value in a normal distribution into a
z- score, you can easily find the p-value for a z-test. This is the probability of SAT scores being 1380 or
less (93.7%), and it’s the area under the curve left of
the shaded area.
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
STEP BY STEP EXAMPLE OF USING THE Z-
DISTRUBUTION
To assess whether your sample mean significantly
differs from the pre-lockdown population mean, you
perform a z-test:
1. First, you calculate a z-score for the sample
mean value
2. Then, you find the p-value for your z-score
using a z-table.
STATISTICAL HYPOTHESIS
Statement about the value of a population
parameter
o Mean, median, mode, variance, sd,
proportion, total
Assertion or proposition about the
relationship between 2 or more variables
Formulated as a result of years of
observation and research
HYPOTHESIS TESTING
Method of making decision using data,
whether from a controlled experiment or an
observational study (not controlled)
Set of procedures that culminates in either
rejection or non-rejection of null hypothesis
Involves the comparison of two hypotheses:
o Null
o Alternative
HO and H1 or Ha
NULL HYPOTHESIS (HO): There’s no effect
on the population
ALTERNATIVE HYPOTHESIS (H1 or Ha):
There’s an effect on the population
PROCESS OF HYPOTHESIS TESTING
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
STEPS OF HYPOTHESIS TESTING among those using models for
1. State the null (Ho) & alternative (H1 or Ha) demonstration, PM , is equal to the
hypothesis proportion among those using
2. State the level of significance cadavers, Pc
3. Select the appropriate test statistic
4. Determine the critical region HA: PM ≠ Pc ("two-tailed test")
5. Compute the test-statistics o The proportion of students who
6. Make a statistical decision obtain a grade of 2.0 or better
7. Draw conclusion among those using models for
demonstration, PM , is not equal to
State the null (Ho) & alternative (H1 or Ha) the proportion among those using
hypothesis] cadavers, Pc
NULL HYPOTHESIS (Ho) HA : Pm> Pc ("one-tailed test")
Hypothesis of no difference o The proportion of students who
Statement of equality obtain a grade of 2.0 or better
Often used to signify zero treatment effect among those using models for
or the equivalence of population parameters demonstration, PM is greater than
or to a specified value to the proportion among those using
cadavers, pc
NON-DIRECTIONAL (TWO-TAILED TEST)
o Simply states that there is difference
in the group being compared
o That the true value of the parameter
is not equal to the hypothesized
value
o Ho: 𝜃 = 𝜃𝑜 𝑣𝑠 𝐻𝑎: 𝜃 ≠ 𝜃𝑜
LEVEL OF SIGNIFICANCE (a)
o The probability level that is
considered too low to warrant
ALTERNATIVE HYPOTHESIS (H1 or Ha) support of the hypothesis being
Usually the research hypothesis (the tested
hypothesis that the investigator believes in) o Probability of committing a Type 1
Uses <, >, ≠ sign error
o Usually set as 0.05, 0.1 or 0.001
Suppose a proposed study was used to
compare the performance in anatomy of DECISION MAKING ERRORS
two groups of students, those using Statistical True state of affairs
cadavers for demonstration and those using decision in the population
models. If the parameter for evaluating based on
student performance is the proportion who sample data
obtain a grade of 2.0 or better, how do we Ho is true Ho is not true
write HO and Ha Do NOT reject Correct 𝜷 𝒐𝒓 𝒕𝒚𝒑𝒆 𝑰𝑰 𝒆𝒓𝒓𝒐𝒓
Ho: PM —Pc ("no difference") decision
o The proportion of students who Reject A or type I Correct decision
obtain a grade of 2.0 or better error
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE
TYPE I ERROR (𝜶)
Error of rejecting a true null hypothesis
Alpha is kept at a low level when it is
important not to make a mistake of rejecting
a true Ho
TYPE II ERROR (𝜷)
Error of not rejecting a false null hypothesis
Beta is kept at a low level when it is
important not to accept a false Ho
BIOSTATISTICS AND EPIDEMIOLOGY- LECTURE MERRA CYRINE VENZUELA
