MIME-Version: 1.0
Content-Type: multipart/related; boundary="----=_NextPart_01C8328C.98241190"
This document is a Single File Web Page, also known as a Web Archive file. If you are seeing this message, your browser or editor doesn't support Web Archive files. Please download a browser that supports Web Archive, such as Microsoft Internet Explorer.
------=_NextPart_01C8328C.98241190
Content-Location: file:///C:/2D693CD3/CoreLesson3.htm
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html; charset="us-ascii"
PowerPoint Presentation
This presentation contains content that your browser may not be able to =
show
properly. This presentation was optimized for more recent versions of Micro=
soft
Internet Explorer.
=
The
materials in this lesson are based on an injury prevention and control curriculum originally =
developed by the World Health =
Organization and a global network =
of injury experts.
General lesson d=
escription
This lesson focuses=
on the
definition of injury surveillance. It provides a rationale for its use in
injury research and policy development. It provides an overview of potent=
ial
sources of data for injuries and different types of injury surveillance s=
ystems,
including their strengths and limitations. It emphasizes the importance of
existing sources of data on injuries and promotes the use of such data as=
a
basis for the improvement or creation of surveillance systems. It explains
the different steps required for the development of an injury surveillance
system. For each step, data quality control issues are discussed in the
context of seven basic principles: simplicity, flexibility, acceptability,
reliability, utility, sustainability and timeliness. This is illustrated =
with
examples from different geographic regions.
To
understand the importance of injury surveillance systems, general definitions,
types and applications of surveillance systems, and basic principles <=
/span>=
for
preserving data quality.
Core
competencies that students should attain by the end of the session
1. Describe=
the
concept of an injury surveillance system.
2. Identify=
the
multiple uses of surveillance for research, control and policy
development.
3. Recognize
the different types of injury surveillance systems including:
whole population surveillanc=
e
surveillance based on
sampling
institutional review of
registries.
4. Recognize
the attributes of major coding systems such as ICD and ICECI and
=
important data items =
such
as mechanism, intent and setting.
5. Understa=
nd
the sequence of necessary steps required for surveillance including:
<=
/b>
how surveillance can define =
an
injury problem
the process of minimum data
collection for surveillance
the definition of data entry=
and
processing protocols
the basic elements for
interpretation of data
clear reporting of results&#=
13;
the potential use of results=
to
plan interventions
the functionality of a
surveillance system
the basic steps necessary to
maintain a surveillance system.
Injury surveillance is the continuing scrutiny of al=
l aspects of occurrence and patte=
rns of
injury that are
pertinent to effective prevention and control.
In man=
y countries
routine collection of health information is uncommon. Moreover, vital
statistics are often incomplete or non-existent and in some ways not
reliable. The few data that exist are usually available from hospitals and
may not reflect or capture the actual occurrence of fatal and non-fatal
injuries in the community.
Inform=
ation systems
can be used for planning purposes, for assessing the magnitude of an inju=
ry
problem by determining injury rates or estimates of prevalence, for
describing an injury problem in the population, for highlighting the risk=
and
protective factors related to injuries, for following trends in injuries =
and
for evaluating interventions or policies.
There =
are many ways
for collecting information, for example through existing registries, thro=
ugh
injury-specific information systems, occupational services, hospitals,
mortuaries or medical examiner or forensic offices, police and judicial
services. Important information can also be obtained directly from househ=
olds
through surveys in the community and serve as an adjunct for existent
hospital data.
<=
span
style=3D'font-family:Arial;font-size:133%'>•Types =
of
systems
In man=
y countries
routine collection of health information is uncommon. Moreover, vital
statistics are often incomplete or non-existent and in some ways not
reliable. The few data that exist are usually available from hospitals and
may not reflect or capture the actual occurrence of fatal and non-fatal
injuries in the community.
Inform=
ation systems
can be used for planning purposes, for assessing the magnitude of an inju=
ry
problem by determining injury rates or estimates of prevalence, for
describing an injury problem in the population, for highlighting the risk=
and
protective factors related to injuries, for following trends in injuries =
and
for evaluating interventions or policies.
There =
are many ways
for collecting information, for example through existing registries, thro=
ugh
injury-specific information systems, occupational services, hospitals,
mortuaries or medical examiner or forensic offices, police and judicial
services. Important information can also be obtained directly from househ=
olds
through surveys in the community and serve as an adjunct for existent
hospital data.
–=
injury
related to other country/regional conditions
–(socioeconomic level)
–determine changes in characteristi=
cs of
population
– over
time and geographical area
–=
evaluate
interventions
Reasons for doing surveillance?
In ord=
er to be able
to explain an injury problem better, define its characteristics, the
populations affected and the risk and protective factors associated with
occurrence of events, there needs to be some type of collection or gather=
ing
of information. The same is required for evaluating interventions and
policies aimed at prevention or control of injuries.
Survei=
llance systems
contribute to defining the magnitude of an injury problem, allowing for
comparisons of incidence and prevalence of injuries over time, between pl=
aces
or between different groups of population. Surveillance systems can
contribute to the identification of changes in trends of certain injuries=
and
thus can help inform whether or not interventions and policies have worke=
d,
have not produced changes or have been more harmful for the population.
The la=
ck of
standardization of data has until relatively recently, prevented injury d=
ata
comparisons between nations. Within nations the multiple sources of injury
data also pose a challenge.
The av=
ailability of
injury data at the local level appears to be a powerful driver of
community-based injury prevention – such as the WHO Safe Communities
model.
There =
are multiple
ways to classify injuries. These classification schemes have lead to comp=
lex
systems used in national and international contexts to facilitate the
identification of different types of injuries as well as to standardize t=
he
nature of injuries. Injuries have been commonly classified in several way=
s.
The me=
chanism of
injury focuses on the circumstances that lead to an injury. Thus one can =
talk
about injuries due to burns, falls, drowning, firearms or because of an
automobile crash among many others (1). The intent, nature of injury and =
body
region injured, settings (such as home, workplace, sports injury and road
traffic injury) are also important modes of classification.
A purp=
ose-designed
injury surveillance system typically and routinely collects data on the
following variables:
age, g=
ender
place =
injury
occurred
activi=
ty at time of
injury
part o=
f body injured
injury=
diagnosis
severi=
ty (severity
scaling system or admitted to hospital or discharged)
intent=
mechan=
ism of injury
narrat=
ive account of
circumstances of injury, products involved, what went wrong.
<=
span
style=3D'font-family:Arial;font-size:83%'>•=
Other disciplines<=
span
style=3D'font-family:Arial;font-size:83%;mso-special-format:lastCR;display=
:none'>
Injury classification systems
As men=
tioned
previously there are multiple internationally used classification systems.
The most used and widely accepted one is the International Classification=
of
Diseases and Injuries (ICD) of WHO. A special section and special codes h=
ave
been created to define and code different types of injuries. The purpose =
of
these classification systems is to facilitate comparability between count=
ries
and allow for universal agreed definitions. A newer system focusing
exclusively on injuries is the International Classification of External
Causes of Injuries (ICECI) which has recently been accepted by WHO into t=
he
WHO Family of International Classifications (2). To assess its usefulness=
, it
is currently being pilot-tested in some countries. Other more regionally =
used
classifications where injuries are included are the Nordic Medico-Statist=
ical
Committee (NOMESCO) classification, the European Home and Leisure Injury
Surveillance System (EHLASS), and country-specific classifications such as
the Canadian Hospitals Injury Reporting and Prevention Program (CHIRPP).<=
/font>
Other =
disciplines
have specialized classifications of injury-related events. Examples inclu=
de
police-reported road traffic injury (RTI) data, occupational injuries and=
in
the legal disciplines, there are multiple definitions and classifications=
of
behaviours that lead to injuries that have been incorporated into the
different national and international criminal justice systems.
The pr=
imary focus of
data systems that may be available for injury surveillance may not be for=
the
prevention of injuries. For example, trauma registries are designed to
monitor and improve trauma patient management and the primary function of
some hospital data systems is for administrative purposes.
- e.g. hospitals (public he=
alth,
surgery,&=
#13;
emergency room<=
/span>)
As men=
tioned before
some of the major advantages of using injury classification systems are t=
hat
comparisons can be made between places or over time. By using commonly
accepted injury classification systems, the evaluation of interventions by
place and time are more easily made.
The ma=
in
disadvantages are related to the different systems currently used. Some of
the classification systems are difficult to link because their use has
different purposes. While potentially richer information about an injury
event can be obtained using different classification systems, many times
linking these different forms of coding is not possible.
Survei=
llance is
defined as the ongoing systematic collection, analysis and interpretation=
of
health data essential to the planning, implementation and evaluation of
health practice, closely integrated with the timely dissemination of these
data to those who need to know. The final aspect of the surveillance chai=
n is
in the application of these data to prevention and control (3, 4).=
A surv=
eillance
system includes a capacity for collecting data, analysing it and
disseminating it for public health interventions. There are several types=
of
surveillance systems. These can be universal (whole population), based on
sampling (e.g. one week of each month), based on registries or based on
settings (road) or jurisdictions (justice).
In reg=
ards to the
process for collecting information there are several activities:
active=
data
collection occurs when those in charge of the surveillance system go to t=
he
primary data source to gather information;
passiv=
e data
collection is carried out when information is sent to the surveillance sy=
stem
from institutions that routinely collect information;
sentin=
el
surveillance is a mixture of both and occurs when some aspects of informa=
tion
are collected actively to complement others collected passively.
•<=
/span>all injuries or a few very specific
injuries
Examples:
•hospital emergency department-bas=
ed
•compensation databases for occupational and traffic injuries
<=
/span>
•<=
/span>
Severa=
l steps are
needed to create a successful injury surveillance system proposed by WHO.
These steps, in order, include:
defini=
tion of the
problem
collec=
tion of
relevant data
enteri=
ng the data
into the system
proces=
sing or
analysing the data
interp=
reting the
data
report=
ing the
results
using =
this
information to plan or evaluate injury prevention or treatment activities=
.
This w=
hole process
needs to also end in an evaluation of the system itself to improve its
accuracy, reliability and effectiveness. This evaluation helps improve the
data processing as well as in redefining the problem. A definition of an
injury problem could be general or specific or it could encompass all
injuries or one type of injury. The definition will depend greatly on wha=
t is
known about injuries in a specific setting and what priorities have been
defined or addressed. Injury problems may be identified by their magnitud=
e,
severity, impact, how they affect people’s lives, the political or
economic costs, etc. The definition will determine the structure of the
system and will be influenced by the location of potential data sources.<=
/font>
One ex=
ample of
reporting and disseminating surveillance information systems is Bogota. D=
ata
is collected, analysed and a bulletin is published and placed on the Inte=
rnet
for public use.
•a hospital-based syst=
em -
multiple =
sources
within institution
Collec=
ting data
implies defining what information is most useful to address the objective=
s of
the system. This means having an idea of the needed variables and hence of
the data collection forms. It also implies defining how the information c=
an
be collected and what procedures need to be established for this data
collection to take place. This will include identifying sources (institut=
ions
and people) who are best placed to gather the data and report to the syst=
em.
An exa=
mple of this
could be thinking of a hospital-based system where data on injuries (from
causes to treatment and rehabilitation) can be collected from many
departments within the hospital.
There =
are many
advantages in collecting data in compatible form to major existing injury
data systems such as ICD-10 (modified to provide additional detail) and I=
CECI
(see Slide 6).
<=
span
style=3D'font-family:Arial'>•multiple forms aggreg=
ated
into one?
<=
span
style=3D'font-family:Arial'>•paper or electronic f=
ormat?
Once d=
ata is
collected, a previously defined process for entering the data needs to be
activated. When injury data comes from multiple sources (which is frequen=
t),
all the arriving forms need to be entered appropriately into a final form=
at.
This can entail the creation of new linkage variables, of re-categorizati=
on
of data or of merging of multiple sources of data. It can also imply
transforming data from hardcopy into electronic formats. Databases for
electronic entry and analysis can be established in programs such as Epi =
Info
(public domain software), and Microsoft Excel or Access.
The pr=
ocess of data
entry needs to have quality checks to assure reliability and periodic data
entry schedules need to be established.
Feedba=
ck on data
quality to the data source and strategies for correcting any deficiencies=
is
an essential component of any injury surveillance system.
<=
span
style=3D'font-family:Arial'>•Use of exposure
measurements of injury to
determine characteristics and mag=
nitude
of the =
problem
<=
span
style=3D'font-family:Arial'>•Use of categories of
population
<=
/div>
Example:
•reporting information=
on
firearm homicide for all
=
populations compared to reporting it by sex and age
=
groups
Once d=
ata is entered
and cleaned it needs to be interpreted. Raw or basic statistical descript=
ions
are usually not enough. Further analysis and calculations of exposure
measures might imply the creation of new analysis variables. The key to
successful interpretation of data lies in the ease of using the databases=
and
how clearly the results are presented. This step must closely follow the
objectives of the surveillance system as it is here that some of the
questions initially posed, need to be answered.
<=
span
style=3D'font-family:Arial'>•Usually a written rep=
ort
format which can vary
=
depending on the population for whom it is intended
Example:
<=
span
style=3D'font-family:Arial'>•more technical data c=
an be
included for scientific
=
readers as opposed to simpler, more concise forms
=
aimed for policy-makers or the public.
After =
the analysis
of the data, the reporting is also important. While data can be complete =
and
well analysed, if it is not correctly and clearly reported, the system mi=
ght
not achieve its intended goals. Reports can suit all readers or can be
tailored to specific audiences depending on the objectives of the injury
prevention specialists.
The ex=
ample provided
illustrates this concept (see exercise).
<=
span
style=3D'font-family:Arial'>•Usually relates to the
objectives and intent of the
=
information from the surveillance system
Example:
<=
span
style=3D'font-family:Arial'>•calculations of rates=
of
injury among pedestrians
=
before and after modifications of road crossings
In ess=
ence, all
previous steps are directed towards using the results to plan interventio=
ns.
This is the ultimate purpose of a surveillance system. Using a surveillan=
ce
system for the sake of having information is a poor choice. This informat=
ion
must be shared and must be provided as a basis for key persons to make
informed decisions on what are the best health options for the population=
.
The ex=
ample above
given to road traffic planners can provide useful information on whether =
or
not preventive or corrective approaches are yielding better health result=
s.
<=
span
style=3D'font-family:Arial'>•A system is dynamic a=
nd
over time all its
=
attributes should be evaluated to ensure data quality
Example:
<=
span
style=3D'font-family:Arial'>•an injury mortality s=
ystem
based on forensic
=
information
The ev=
aluation of a
surveillance system is essential as such systems need to respond to ever
changing problems, needs, definitions and circumstances. The dynamic natu=
re
of these problems needs to be reflected in the capacity of a system to ad=
apt
to information changes. The evaluation also entails verifying whether or =
not
a system is responding to the objectives and needs to which it was initia=
lly
meant to respond. There are multiple ways for assessing whether this is
occurring or not. Comparison of the system with other sources of informat=
ion
or with complete samples can provide useful data on the accuracy of the
system itself.
The ev=
aluation does
not only refer to the data per se but also to the processes put in place =
to
gather the information. Thus after an evaluation, better forms or procedu=
res
can be designed. Also sources of information ca be dropped or added.
<=
span
style=3D'font-family:Arial'>•criteria to evaluate =
data
might require changes
<=
span
style=3D'font-family:Arial'>•changes in available
technology
A fina=
l element
within this dynamic process is to keep the system updated. The evaluation=
of
definitions, of the flow of information and of its interpretation are
important. Changes in definitions or technology might lead to changes in =
the
system. Keeping a system up to date ensures that it responds to actual
problems in as close to real time as possible.
Many c=
ountries have
developed emergency department injury surveillance systems in recent years
including Columbia, El Salvador, Jamaica, Mozambique, Nicaragua and Thail=
and.
Even the most established systems in high-income countries have existed f=
or
relatively few years.
This lesson focuses on the definition of
injury surveillance. It provides a rationale for its use in injury research
and policy development. It provides an overview of potential sources of da=
ta
for injuries and different types of injury surveillance systems, including
their strengths and limitations. It emphasizes the importance of existing
sources of data on injuries and promotes the use of such data as a basis f=
or
the improvement or creation of surveillance systems. It explains the diffe=
rent
steps required for the development of an injury surveillance system. For e=
ach
step, data quality control issues are discussed in the context of seven ba=
sic
principles: simplicity, flexibility, acceptability, reliability, utility,
sustainability and timeliness. This is illustrated with examples from
different geographic regions.
In many countries routine collection of health information is uncom=
mon.
Moreover, vital
statistics are often incomplete or non-existent and in some ways not reliable. The few data that ex=
ist
are usually available from hospitals and may not reflect or capture the actual occurrence of f=
atal
and non-fatal
injuries in the community.
Information systems can be used for planning purposes, for assessing
the magnitude of=
an
injury problem by determining injury rates or estimates of prevalence, for describing an injury
problem in the population, for highlighting the risk and protective factors related to injuries, f=
or
following trends in
injuries and for evaluating interventions or policies.
There are many ways for collecting
information, for example through existing registries, through injury-specific information systems, <=
/span>occupational services, hospitals,
mortuaries or medical examiner or forensic offices, police and judicial services. Important informati=
on
can also be obta=
ined
directly from households through surveys in the community and serve as an adjunct for
existent hospital data.
In many countries routine collection of health information is uncom=
mon.
Moreover, vital
statistics are often incomplete or non-existent and in some ways not reliable. The few data that ex=
ist
are usually available from hospitals and may not reflect or capture the actual occurrence of f=
atal
and non-fatal
injuries in the community.
Information systems can be used for planning purposes, for assessing
the magnitude of=
an
injury problem by determining injury rates or estimates of prevalence, for describing an injury
problem in the population, for highlighting the risk and protective factors related to injuries, f=
or
following trends in
injuries and for evaluating interventions or policies.
There are many ways for collecting
information, for example through existing registries, through injury-specific information systems, <=
/span>occupational services, hospitals,
mortuaries or medical examiner or forensic offices, police and judicial services. Important informati=
on
can also be obta=
ined
directly from households through surveys in the community and serve as an adjunct for
existent hospital data.
In order to be able to explain an injury problem better, define its=
characteristics, the populations affect=
ed
and the risk and protective factors associated with occurrence of events, there needs to be some type o=
f collection or gathering of information.=
The
same is required for evaluating interventions and policies aimed at prevention or control of
injuries.
Surveillance
systems contribute to defining the magnitude of an injury problem, allowing for comparisons of
incidence and prevalence of injuries over time, between places or between different groups of population=
. Surveillance systems can contribute to =
the
identification of changes in trends of certain injuries and thus can help inform whether or not =
interventions and policies have worked,
have not produced changes or have been more harmful for the population.
The lack of standardization of data has
until relatively recently, prevented injury data comparisons between nations. Within nations the multipl=
e sources of injury data also pose a
challenge.
The availabi=
lity
of injury data at the local level appears to be a powerful driver of community-based injury preven=
tion
– such as the WHO Safe Communities model.
There are multiple ways to classify injuries. These classification
schemes have lead to
complex systems used in national and international contexts to facilitate the identification of
different types of injuries as well as to standardize the nature of injuries.
Injuries have been commonly classified in several ways.
The mechanism of injury focuses on the circumstances that lead to a=
n injury. Thus one can talk about injuries
due to burns, falls, drowning, firearms or because of an automobile crash among many others (1). T=
he intent, nature of injury and body region
injured, settings (such as home, workplace, sports injury and road traffic injury) are also important
modes of
classification.
A purpose-designed injury surveillance system typically and routine=
ly collects data on the following
variables:
age, gender
place injury occurred
activity at time of injury
part of body injured
injury diagnosis
severity (severity scaling system or admitted to hospital or
discharged)
intent
mechanism of injury
narrative account of circumstances of injury, products involved, wh=
at
went wrong.
As mentioned previously there are multiple internationally used classification systems. The most used a=
nd
widely accepted one is the International Classification of Diseases and Injuries (ICD) of WHO.=
A special section and special codes have =
been
created to define and code different types of injuries. The purpose of these classification
systems is to facilitate
comparability between countries and allow for universal agreed definitions. A newer system focusing
exclusively on injuries is the International Classification of External Causes of Injuries (ICECI)
which has recently been
accepted by WHO into the WHO Family of International Classifications (2). To assess its
usefulness, it is currently being pilot-tested in some countries. Other more
regionally used classifications where injuries are included are the Nordic
Medico-Statistical Committee (NOMESCO) classification, the European Home and Leisure Injury Surveillance System (EHLASS), and
country-specific classifications such as the Canadian Hospitals Injury Report=
ing
and Prevention Program (CHIRPP).
Other disciplines have specialized classifications of injury-related
events. Examples inc=
lude
police-reported road traffic injury (RTI) data, occupational injuries and in the legal
disciplines, there are multiple definitions and classifications of behaviours that lead to injuries
that have been
incorporated into the different national and international criminal justice systems.
The primary focus of data systems that may be available for injury =
surveillance may not be for the prevent=
ion
of injuries. For example, trauma registries are designed to monitor and improve trauma patient management and the primary function of =
some
hospital data systems is for administrative purposes.
As mentioned before some of the major advantages of using injury classification systems are that compari=
sons
can be made between places or over time. By using commonly accepted injury classification syst=
ems,
the evaluation of
interventions by place and time are more easily made.
The main disadvantages are related to t=
he
different systems currently used. Some of the classification systems are difficult to link beca=
use
their use has
different purposes. While potentially richer information about an <=
span
lang=3DEN-GB style=3D'font-family:Arial;mso-ascii-font-family:Arial;mso-bi=
di-font-family:
Arial;mso-hansi-font-family:Arial'>injury event can be obtained using
different classification systems, many times linking these different forms of
coding is not possible.
Surveillance is defined as the ongoing systematic collection, analy=
sis
and interpretati=
on
of health data essential to the planning, implementation and evaluation of health practice, closely
integrated with the timely dissemination of these data to those who need to know. The final as=
pect
of the surveillance
chain is in the application of these data to prevention and control (3, 4).
A surveillance system includes a capaci=
ty
for collecting data, analysing it and disseminating it for public health interventions. There are sev=
eral
types of surveilla=
nce
systems. These can be universal (whole population), based on sampling (e.g. one week of each
month), based on registries or based on settings (road) or jurisdictions (justice).
In regards to the process for collecting
information there are several activities:
active data collection occurs when those in charge of the surveilla=
nce system go to the primary data source to
gather information;
passive data collection is carried out when information is sent to =
the surveillance system from institutions t=
hat
routinely collect information;
sentinel surveillance is a mixture of both and occurs when some asp=
ects
of information are
collected actively to complement others collected passively.
Several steps are needed to create a successful injury surveillance
system proposed by =
WHO.
These steps, in order, include:
definition of the problem
collection of relevant data
entering the data into the system
processing or analysing the data
interpreting the data
reporting the results
using this information to plan or evaluate injury prevention or
treatment activities.
This whole process needs to also end in an evaluation of the system
itself to improve i=
ts
accuracy, reliability and effectiveness. This evaluation helps improve the data processing as well as =
in
redefining the problem. A definition of an injury problem could be general or specific or it
could encompass all
injuries or one type of injury. The definition will depend greatly on what is known about injuries=
in
a specific setting and what priorities have been defined or addressed. Injury problems may be <=
/span>identified by their magnitude, severity,
impact, how they affect people’s lives, the political or economic costs,
etc. The definition will determine the structure of the system and will be
influenced by the location of potential data sources.
One example of reporting and disseminating surveillance information=
systems is Bogota. Data is collected,
analysed and a bulletin is published and placed on the Internet for public use.
Collecting data implies defining what information is most useful to
address the objectiv=
es
of the system. This means having an idea of the needed variables and hence of the data collect=
ion
forms. It also implies defining how the information can be collected and what procedures need to be=
established for this data collection to
take place. This will include identifying sources (institutions and people) who a=
re
best placed to gather the data and report to the system.
An example of this could be thinking of a hospital-based system whe=
re data on injuries (from causes to treatm=
ent
and rehabilitation) can be collected from many departments within the hospital.
There are many advantages in collecting
data in compatible form to major existing injury data systems such as ICD-10 (modified to provide
additional detail) and
ICECI (see Slide 6).
Once data is collected, a previously defined process for entering t=
he
data needs to be
activated. When injury data comes from multiple sources (which is frequent), all the arriving f=
orms
need to be entered appropriately into a final format. This can entail the creation of new linkage
variables, of re-categoriz=
ation
of data or of merging of multiple sources of data. It can also imply transforming data from hardc=
opy
into electronic formats. Databases for electronic entry and analysis can be established in <=
/span>programs such as Epi Info (public domain
software), and Microsoft Excel or Access.
The process of data entry needs to have quality checks to assure reliability and periodic data entry
schedules need to be established.
Feedback on data quality to the data so=
urce
and strategies for correcting any deficiencies is an essential component of any injury surveillan=
ce system.
Once data is entered and cleaned it needs to be interpreted. Raw or
basic statistical
descriptions are usually not enough. Further analysis and calculations of exposure measures might
imply the creation of new analysis variables. The key to successful interpretation of data li=
es
in the ease of using
the databases and how clearly the results are presented. This step must closely follow the
objectives of the surveillance system as it is here that some of the questions
initially posed, need to be answered.
After the analysis of the data, the reporting is also important. Wh=
ile
data can be compl=
ete
and well analysed, if it is not correctly and clearly reported, the system might not achieve =
its
intended goals. Reports can suit all readers or can be tailored to specific audiences depending=
on
the objectives of
the injury prevention specialists.
The example provid=
ed
illustrates this concept (see exercise).
In essence, all previous steps are directed towards using the resul=
ts
to plan intervention=
s.
This is the ultimate purpose of a surveillance system. Using a surveillance system for the sake of hav=
ing
information is a poor choice. This information must be shared and must be provided as a basis for key =
persons to make informed decisions on w=
hat
are the best health options for the population.
The example above given to road traffic planners can provide useful=
information on whether or not preventiv=
e or
corrective approaches are yielding better health results.
The evaluation of a surveillance system is essential as such system=
s need to respond to ever changing proble=
ms,
needs, definitions and circumstances. The dynamic nature of these problems needs to be reflected in the capacity of a system to
adapt to information changes. The evaluation also entails verifying whether or not a system is respon=
ding
to the objectiv=
es
and needs to which it was initially meant to respond. There are multiple ways for assessing whether
this is occurring or not. Comparison of the system with other sources of information or with =
complete samples can provide useful dat=
a on
the accuracy of the system itself.
The evaluation does not only refer to the data per se but also to t=
he processes put in place to gather the
information. Thus after an evaluation, better forms or procedures can be desig=
ned.
Also sources of information ca be dropped or added.
A final element within this dynamic process is to keep the system <=
/span>updated. The evaluation of definitions,=
of
the flow of information and of its interpretation are important. Changes in definitions or technology
might lead to chan=
ges
in the system. Keeping a system up to date ensures that it responds to actual problems in as cl=
ose
to real time as possible.
Many countries have developed emergency department injury surveilla=
nce systems in recent years including Colum=
bia,
El Salvador, Jamaica, Mozambique, Nicaragua and Thailand. Even the most established systems in high-income countries have
existed for relatively few years.
![](3D"slide0001_image002.gif")
![](3D"slide0041_image003.gif")
![](3D"slide0038_image004.gif")
![](3D"notes_flag.gif")
No=
tes![](3D"buttons.gif")
![](3D"master03_image001.jpg"){kind=link}