Analysis of statistical data in heath information management

Analysis of statistical data in
Heath information management
HIMI, JOS
By
Saleh Ahmed
School of Midwifery, Bauchi.

Analysis of statistical data in
Heath information management
Introduction
 For proper and effective health planning and decision
making in any health institution adequate and reliable
data must be collected and fully analyzed.
 Well analyzed data will make the decision makers to
detect and control emerging and endemic health
problem, monitor progress towards health goals and
promote equity.

Objectives
 At the end of this presentation the participants should
be able to:-
- Define statistics.
- Describe types of health statistics in health
information management.
- Describe various ways of analyzing statistical data in
health information managemen

-
 Meeting the eight (8) millennium Development goal
targets can only be achieved by analysis of health
statistical data such as vital statistics, morbidity
statistics, and health service statistics. Thus analysis of
data is an indispensable tools in health information
management system globally.

Statistics
Definition
 The word statistics is derived from latin – status
meaning state or condition
 According to English advanced Dictionary,
 statistics is the branch of mathematics that deals with
analysis and interpretation of numerical data in terms
of sample and population
 Statistics is a science that is concerned with the
collection, compilation, presentation , analysis and
interpretation of numerical data (Nwabuokei,2001)

-
 Based on the above definition statistics deals with data
and population(both living e.g. human being and non
living e.g. measurement of aggregate results)

Branches of statisticsi) Descriptive statistics
ii) Inferential statistics
Descriptive statistics –it is type of statistics that
studies a body of statistical data and no generalization
is made from the results obtained. It involves
collection of data, classification of data, drawing of
histogram, polygon, charts, percentages, range and
other types of statistical graphs, computation of
sample characteristics such as the mean, mode,
median , standard deviation etc.

-
 Descriptive statistics provide precise, standard ways to
summarize, understand and communicate complex
information.
 It summarizes the population of data by describing
what was observed numerically or graphically
 Numerical –eg, mean, standard deviation.
 NB, frequency and percentages are useful in
describing categorical data e.g race , sex, age.

-
 Inferential statistics – it is the branch of statistics
that studies a group of data in order to use the results
obtained in making generalization on a larger group
of data. It uses sample results to reach conclusions
about populations from which the samples have been
drawn from. For instance testing of hypothesis by
taking the form of answering yes or no questions.

-
 It also performs the followings:-
 Estimating numerical characteristics of
data=Estimation
 Describing association within data =Correlation
 Modeling(a representative set of data) relationships
within the data =Regression, extrapolation,
interpolation or other modeling.

Statistical data
 Data is an information - facts - knowledge
 Statistical data can be classified based on the nature or
the source.
 A) based on the source of the data
i) primary data- data collected by the investigator
himself for the purpose of statistical analysis
ii) secondary data- data collected from existing
data(records )

-
 Based on the nature of the data
 i) Qualitative data – data used for describing
characteristics which can not be defined in numerical
terms e.g, colour of hair, colour of the
eyes, performance grade- good, average, poor
 ii) Quantitative data- data that are capable of
numerical description e.g. weight , height in
meter, scores of students. It consists of measures that
take numerical values for description such as means
and standard deviation. if countable =discrete data
and if measurable express in scale =continuous data eg
ht

Health Statistical Data
 Types
A) Vital statistics
B) Morbidity statistics
C) Health service statistics
A) Vital statistics – it is the branch of statistics that
deals with the changes and most basic events of
human population e.g, birth, marriage, mortality and
illness. Such data are gathered from census and
registration reports.

-
 Vital statistics is collected for the purpose of
generating birth and mortality rates for the whole
population or subgroup.
 Method of collection is through ongoing recording or
registration of vital event such as birth, adoption,
death, marriages, divorces,
legal separation etc

-
 B) Morbidity statistics:-These are data on occurrences of
severity of sickness in a community.
 It is collected for the purpose of analysis of ill-health within
human population and for provision of detailed analysis of
health status and services.
 Method of collection is through medical services such as
medical institutions from:-
 a) outpatient clinic,
 b) special clinics(maternal)
 c)inpatient services (general hospital, specialist hospital)

-
 C) Health services statistics – these are data that are
obtained from operation of the health services. Two types
of data obtained are
 i) Resource data
 ii) institutional records
 i) Resource data – are data on human and material
resources. Human resource are details of number of various
types of health personnel (Doctors, Nurses, Midwives,
community health workers etc). Consideration is made on
their distribution in relation to the population.

-
 ii) Institutional records – these are records
generated from health facilities. For example records
of client/patient that attended a health centre in
certain period of time.(how many pregnant women
received ANC). This serves as a means of providing
information about the demand for and utilization of
health services and about the extent to which target
groups within the population are being served.

Sources of health statistical data
1) Census of the population.
2) Registration of births and deaths
3) Notification of Diseases
(surveillance, epidemiological surveys)
4) Medical institutions (hospital, health
centers, clinical laboratories)

Analysis of statistical data
Health statistics analysis is based on two measures
tools:- rates and ratio.
 Rate is the frequency (number) of events that occur in
a defined period, divided by the average population.

-
 Denominator data – population at risks
 Numerator data – events or condition of concern.
 Constant multiplier is either 100 to make a % or else
1,000, 10,000, or 100,000 to make the numerator larger
than 1 for easy discussion
 Thus rate can be expressed as follows:-
.

-
 -Thus, Health statistical are expressed in
rates, i.e number of events that are related to
population.
Rates are expressed in arbitrary total
e.g, 1,000, 10,000, 100,000, 1,000,000.
Rates is the most important tool for
measuring disease or death. It is used to
measure events that are related to the
population or subgroup in which they occur.

Types of rates
 Rates are grouped into two that is based on the nature of
population used as follows:-
 Crude rates
 Specific rates
 Crude rates –these are rates that are applied to entire
population without reference to any characteristics of the
individuals in it. The rates are calculated with the total
population in an area as the denominator.

-
 Crude rates is used when for instance
 i) the frequency of death or disease(numerator) is not
known for the subgroups of population
 ii) the size of the subgroups(denominator) is not
known
 iii) the number of person at risk is too small to provide
a stable estimate.

-
 Specific rates :- the rates are calculated after
population has been categorized into groups with
particular characteristics e.g age-specific rates or sex-
specific rates.

Calculation of vital statistics
 This calculation is for the purpose of determining
mortality rate.
 Crude rates-

Calculation of morbidity statistics
 Morbidity rates are divided into two major types as
follows;
1) Incidence rate:- it indicates the rate at which new
disease occurs in a defined, previously disease-free
population.
NB. period of time can be 1 year period or more

-
 Prevalence rates:-it measures number of people in a
population that have a disease at a given time.
 Prevalence depends on the number of people that have
been ill in the past and the duration of their illness.

presentation of data.
 Health information management can only be achieved
when data collected are explicitly stored in meaningful
forms which can be understood by stakeholder and
policy makers
 Data are presented in the following methods
 The text method
 The semi-text method
 The tabular method
-Diagrammatic and graphic method(use of
graphs, charts and diagrams)

The text method
 This is the simplest method of presentation of data in
form of written report
 E.g the number of health personnel in Abubakar
Tafawa Balewa teaching Hospital as at 2009,is 1,000.
out of which 600 were male and 400 were females.
 Its shortcomings- it can’t give effective and clear
interpretation of statistical data and there is
possibilities of omission and repetition,

The semi-text method
 At times called the partial-tabular method. It
combines the text and tabular forms of data
presentation.
 It facilitates easy comparison because figures are
presented separately from the text.
 Eg. number of patients admitted in Specialist Hospital
for the period of 2005 – 2007
 2,000 in 2005,
 2,500 in 2006,
 3,000 in 2007

Tabular method
 This involves the systematic arrangement of facts and
figures in series of boxes made up of rows and
columns.
 According to Nwabuokei (2001) the Components of
statistical tables are;
 The title
 The caption or box head(column labeling)
 The stub (row labeling)
 The source and or footnotes(abbreviation or symbol
not universally known)

Types of tables
 Simple table- it consists of merely a list of items. it
provides statistical data in one or two column
 Table 1: Enrollment into SON,Monze in 2009
 Source –SON Monze
LGA No of Candidates enrolled
Muzabuka 14
Choma 28
Chukuni 35

-
 Complex table:-it shows division of total into two
or more. It is useful in making comparisons
 Table II Sex distribution of students enrolled
 Source :-SON Monze
LGA No Male female withdraw
Muzabuk
a
14 5 9 3
Choma 28 10 18 4
Chukuni
Total
35
77
15
30
20
47
6
13

Frequency distribution tables
 This is the most commonly used in presentation of
health statistic in public health. The process of
drawing of the table is a follows
 First , the data –ie the raw data are arranged based on
their magnitude-ascending or descending order. This
is known as the array of data
 Eg , 3,2,8,6,8,3,10,8,10,2,3,5,8,5,8 }raw data
 2,2,3,3,3,5,5,5,6,8,8,8,8,8,8,10,10 }array of data.

-
 Table III: Anatomy test scores over 10
score tally frequency
2
3
5
6
8
10
II
III
III
I
IIII
II
2
3
3
1
5
2
total 16

Graphic representation
 This is the display of data in the forms of graphs, geometric
features or pictures.
 The purpose is to provide a simple, visual aid that the
reader will readily appreciate the important features of the
data
 The common examples are-
 A) Bar chart- it contains bars of which the length is
proportional to the frequency of events and each
representing each items in the group. It is of 3 types
simple,compound &component. It is useful in
representing discrete variable(data).Below are the
examples bar charts.

Causes of maternal deaths in
Gashaka
Causes No of
deaths
percentage
Eclampsia 15 21.4%
infection 25 35.7%
Post partum
hemorrhage
30 42.9%
Total 70 100%

Causes of maternal deaths in
Sabon Kaura village
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
40.00%
45.00%
50.00%
Eclampsia Infection Postpartum
haemorrhage
Series1

Causes of maternal deaths from
2004 – 2007 in Ganjuwa LGA
0% 20% 40% 60%
2004
2005
2006
2007
eclampsia
infection
post partum haemorrge

Histogram
 It is a special type of Bar chart used for displaying
numerical variables
 Variables of interest are shown on one axis as a
continuous scale split into class.
 The bars are joined to each other and their areas
representing frequency of events.
 For instance the age and sex distribution of population
may be displayed in the form histogram to produce
population pyramid

Pie chart
 It consists of a circle divided into sector which area of
each sector is proportional to the value of each
variable.
 It is used for presenting data of proportion or
percentage of whole.
 In order to present information using pie chart, the
following should be done,
 1) calculate the size of each sector based on the
magnitude in degree from 360 circle.
 2) Draw the circle

Major causes of sudden deaths.
15%
25%
8%
52%
hypertension
heart attack
diabetes

-
 There are other various methods of data analysis that
are not dealt with.

Conclusion.
 The objective of health information management is to
generate information that decision makers and
managers can use to support health programs. This
can be achieved by proper record system and well
established institution for registration of vital events
such as births, deaths etc. it therefore the
responsibility of all health
workers, epidemiologists, and statisticians to ensure
that data are collected, analyzed interpreted and
communicated for optimum health services delivery
to all citizens of Nigeria

Analysis of statistical data in heath information management

More Related Content

What's hot (20)

Viewers also liked (12)

Similar to Analysis of statistical data in heath information management (20)

More from Saleh Ahmed (8)

Recently uploaded (20)

Analysis of statistical data in heath information management

Editor's Notes