Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
29
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
MISCELLANEOUS
Principles of public health surveillance: a revisit to fundamental concepts Chanapong Rojanaworarit, Ph.DDepartment of Epidemiology, Faculty of Public Health,Mahidol University, ThailandEmail:[email protected]
Abstract
Theobjectiveofthisarticlefocusesonfundamentalconceptsofpublichealthsurveillance,especiallytograduatestudentsandhealthprofessionalswhoarenewto thisdiscipline.Thearticle initiallyexplainshowconceptsofsurveillanceinpublichealthhaveevolvedover theperiodofsixdecades.Thenthe threemajorobjectivesofsurveillancearefurtherdiscussedindetail.Withclearspecificationoftheobjective,surveillancesystemcanbeestablishedtofulfillthegoal.Generalconsiderationsforestablishingnewsurveillancesystemare thereforeoutlined.Methodicalprocessof surveillance from informationgeneration to the link topublichealth actions is thoroughly explained.Challenging issues and current technical advance in this field ofepidemiological practice are additionally summarized.
Keywords: public health surveillance, epidemiologicalmethod, public health, disease control
13 � 1.indd 29 6/10/15 4:36 PM
30
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
Introduction Fromtheearlyconceptofsurveillanceconcentra-
tingondetectionofsuspectedactofcrimecommis-
sion,conceptionofsurveillancehaslaterbeenadopted
intomonitoringdiseaseoccurrenceinindividualsand
progressively evolved into the current concept of
ongoingobservationonacertainpopulationtogauge
thechange indefinedhealtheventand todetermine
whether a public health act is promptly needed for
controlorpreventionofthealteredhealthcondition.
Since the global trend of disease has shifted in
termsof burden from infectious diseases to chronic
non-communicablediseases,methodofpublichealth
surveillance has thus been developed in response
to this change. From the conventional surveillance
approach,which concentrates onmeasuring disease
occurrence(e.g.,numberofcaseswithacertaindisease)
and related consequence (e.g.,mortality); extension
toincludeobservationofriskfactors,environmental
hazard,andpositivehealthdeterminants–whichprecede
theoccurrenceofdiseaseorhealthoutcome–hasbeen
made for comprehensive care of population health
with amorepreventiveorientation.
This article focuses on fundamentals concepts
ofpublichealthsurveillanceimportantlyintermsof
its conceptualization andmethodological principles.
Conceptualdevelopment,objectives,generalconsid-
eration for surveillance system establishment, and
currentmethod of public health surveillance are
discussedindetail.Challengingissuesandtechnical
advance in practice of surveillance are additionally
summarized. The aim of this article is to provide
general overview of this epidemiological practice
especiallyforgraduatestudentsinepidemiologyand
healthprofessionalswhomaynotbe familiar to the
discipline.More technical advance in this evolving
fieldof public health surveillance,whichmaybe far
beyond the scopeof this article, can alsobe further
exploredwith the prior foundationprovidedherein.
1. Defining and conceptualizing ‘surveillance’:
frometymologicalbasistocontemporaryrecogni-
tion inpublichealth
‘Surveillance’–aFrenchword inoriginbelieved
tobeadoptedintoEnglishin1802–isanounsimply
defined as ‘the act of oversight orwatching over’.
Thetermcomprisestwowordelementsincluding‘sur’
(overoratop)and‘veiller’(towatch).1‘Surveillance’
was introduced intoEnglish from terror in France
where ‘surveillance committees’were established
in allmunicipalities in 1793 tomonitor the act of
suspected individuals.1
Surveillancewas initially acknowledged as the
act of closemonitoring of individuals exposed to
contagious or communicable diseases for timely
detection ofmanifestationswhich further indicated
controlmeasures–such as quarantine.2Until 1950,
perspective change in surveillance from the act of
individualmonitoringtopopulation-basedsurveillance
ofdiseaseoccurrence.2Thisapproachbecamepromi-
nent following the 1954field trial of poliomyelitis
vaccine(Salk’sinactivatedpoliovaccinetrial)inthe
US.3Populationsurveillancehasalsobeenapproved
asanintegralresponsibilityofpublichealthpractice.4
The approach comprised three fundamental
featuresincludingsystematicdatacollection(atlocal
sources),dataassemblyandanalysis(atsurveillance
centers), anddisseminationof pertinent information
and precisemessage through descriptive epidemio-
logical reports.5
13 � 1.indd 30 6/10/15 4:36 PM
31
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
The notion of extending role of surveillance to
indicate control activitieswas also considered.6 In
someWHOprograms,informationfromsurveillance
hasevidentlybeenapplied tospecifyingactivecon-
trolmeasures–such as extensive vaccination against
smallpox.6Nonetheless, this concept of using the
information from surveillance to determinewhat
activitytobeimplementedwaslaterobjected.6Ithas
beensuggestedthatsurveillanceshouldonlybeahint
of feasible control activitieswhile health authority
should retain the right tomake decision regarding
practical controlmeasure to be implemented in the
actual context.6This suggestion ismore relevant to
the actual context of public health practice since
healthresourcesandcontextualconstraintsaretaken
into consideration, anddecision to act inwhatway
can be justified by local health practitionerswho
understand local circumstancewell.
Other terms including ‘epidemiological surveil-
lance’and‘publichealthsurveillance’havealsobeen
proposed. The term ‘epidemiological surveillance’
was broadly defined as ‘epidemiological study of
disease–even incorporating epidemic investigation
and research–as a continuous and ever-changing
process’.2Nonetheless, this definitionwas opposed
based on the reason that it seemed equivalent to
how epidemiological practice has been defined.2
Even though surveillance datamay reveal gap of
knowledge and hypothesis formulation leading to
research,objectiveof research isdifferent from that
ofsurveillanceandshouldberecognizedseparately.7
ThackerandBerkelmanthuslaterproposedtheterm
‘publichealthsurveillance’whichconservedbenefits
of the initial term ‘epidemiological surveillance’
while excluding confusionwith research.2 Public
healthsurveillancehasthusbeendefinedasasystem
comprising continuousdata collection, data analysis
and interpretation,disseminationofkey information
andmessage to the responsible personnel to timely
urgepublichealth action for control andprevention
of disease or condition.8 Public health surveillance
does not incorporate specification of preventive and
controlmeasures.Publichealthsurveillanceisaimed
essentiallyatprovidinghintofapplicablecontroland
preventive activities. In otherwords, public health
ratherprovidesinformationofpublichealthsituation
forresponsiblepersonnelatalllevels,guidesfeasible
programs, and later eases programevaluationwhen
outcomes are assessable.8 For further discussion in
this article, the term ‘surveillance’ refers to ‘public
health surveillance’. Summaryof conceptual devel-
opment of surveillance andmajor characteristics at
each stage is provided inTable 1.
13 � 1.indd 31 6/10/15 4:36 PM
32
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
Table 1 Summaryofconceptualdevelopmentofsurveillanceateachstageandcorrespondingkeyfeatures
Conceptualdevelopment
Unit ofobservation
Health eventofinterest
Key features
1)Personal surveillance (Individual monitoring)
2)Disease surveillance
3)Epidemiological surveillance
4)Public health surveillance
• Individual
• Population
• Population
• Population
• Disease(especiallyinfectiousdisease inman)
• Disease (communicable and non-communicable)• Other health problems (e.g., environmental hazard)
• Disease• Other health problems
• Disease• Other health problems• Unusual event (e.g., mass animal death)• Riskbehavior• Variousother conditions
• Closemonitoringofindividualsuspectedof havingdisease
• Detectionof disease symptom• Informationisusedforindicatingcontrolmeasure such as quarantine.
• Ongoingobservationofhealtheventwithorganizedprocess
• The process comprises data collection,assembly, analysis and interpretation.
• Extendedscopeofsurveillancetoincludeepidemic investigation and research
• Theconceptwasobjectedsinceitmisledunderstandingofsurveillancepracticetobeequivalenttoepidemiologicalpractice.
• Organized systemcomprisingdata collection, analysis, interpretation, disseminationofinformation,andlinktopublic health action
• Information isusedfordecisionmakingrather thandictatewhat action to take
Although ‘surveillance’ and ‘monitoring’ seem
similarinmeaning,thesetermsareactuallydifferent
andshouldnotbeusedinterchangeably.Monitoring
resemblessurveillanceinthatitsmethodalsofocuses
ontimeliness,practicability,andcontinueddatacol-
lectiononroutine.2Nonetheless;surveillanceallows
impactassessmentofdiseaseinacertainpopulation
prior to and following implementation of health
program,whilemonitoringfocusesonlyonevaluating
post-implementationoutcomes.2Theotherdistinction
of these terms isaccording to theirgroupsof focus.
Surveillance focuses on relatively larger group of
populationswhilemonitoring commonly concerns
individuals(e.g.,monitoringofvitalsignsinacertain
13 � 1.indd 32 6/10/15 4:36 PM
33
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
patient)orspecificgroups(e.g.,effectivenessevalua-
tionofHIV/AIDSpreventiveprograminfemalesex
workersormonitoringoflong-termglycemiccontrol
amongdiabeticpatientsattendingadistricthospital).2
Theother difference to be remarked is between
‘surveillance’and‘survey’.Whendataarenotreadily
availableinroutinedatacollectionofexistinghealth
servicesystem–asthatcommonlyfoundinsurveillance,
suchdataarethenspecificallycollectedforacertain
reason or objective in an organized way called
‘survey’.9Therefore,intermsofhowsurveyisrelated
to surveillance, survey can be one of the possible
methodsofobtainingdataforsurveillance.Nonetheless,
survey adopts active approach of population-based
data collection for a certain purpose;while surveil-
lance, in general, passively adopts readily available
dataofacertain facility.9Survey is thusundertaken
on occasion with specified value of budget and
invested resources;whereas surveillance is rather a
continued and less costly process in the long run.9
Surveillanceusuallyinvolvesallhealthagenciesfrom
local health facilities to health authorities of higher
levels: provincial, national, international.2Theflow
of information is usually in back and forth pattern:
upwarddata reporting, downwardpolicy suggestion
for action, upward report on effectiveness of policy
implementation,downwardfeedbackandsupport,and
others.Unlike surveillance, survey is occasionally
fundedforspecifiedobjectiveanditsresult israther
directed to fundingprovider or concerned agency.9
2. Objectives of surveillance Ingeneral,thepurposeofsurveillanceistoidentify
changesindistributionortrendstolaunchinvestiga-
tionorcontrolactivities.10Tobemorespecific,three
major objectivesof surveillance canbe listed as (a)
todescribethedynamicpatternofdiseaseoccurrence
which links to public health action, (b) to elucidate
naturalhistoryandepidemiologicalprofileofdisease,
and(c)tosupplybaselinedataandrelevantinforma-
tion.2Theirfeaturescanbesummarizedasillustrated
inTable 2.
13 � 1.indd 33 6/10/15 4:36 PM
34
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
Table 2 Objectives of surveillance and corresponding features
Objectives of surveillance Key features
1) Todescribe the dynamic pattern of disease and link to public health action
2)To explain natural history and epidemiological profile of disease
3)To supplybaseline data for planning and evaluationof health program
Descriptionof disease pattern is undertaken to:• detect immediate changes in disease occurrence anddistribution (e.g., disease outbreak, food poisoning),• identifydisease trendandpattern (e.g., increased incidence of ovarian cancer
in younger Thai women),• anticipatepossibilityofgettingdiseasebymonitoringchangeinhostoragent
factors (e.g., study of mutated pathogen), and• identify change in disease burden fromhealthcare practice (e.g., increasing number of patients requiring dental extraction).
Informationobtained from thedescription is used for:• decision to provide prompt action (e.g., identifying agent causing disease
outbreak and guiding control measure)• healthcare planning (e.g., focusing more on oral health prevention to control
the problem of tooth loss)• reorganizinghealth service system• evaluating effectiveness of the implemented program (e.g., effectiveness evaluation of vaccination program)11
• Aid in explaining natural course of a certain diseasewhich has not beenpreviously clarified
• Allowdescriptionof epidemiological profile of disease in termsof ‘persons’ being affected, ‘place’ or risk environment, and ‘time’
of disease occurrence.• Provide information for planningof control andpreventivemeasures• Allow estimation of impact from health program on target health outcome
prior to implementation (e.g., projection or predictive model development to anticipate effect of proposed health policy using disease trend data)
• Supply baseline data for comparison or evaluation of target health outcomeafter the health program is implemented (e.g., measuring marginal benefits after implementing health program)11
13 � 1.indd 34 6/10/15 4:36 PM
35
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
3. General consideration for surveillance system establishment Establishmentofsurveillancesystemisappropri-
ate only for somediseases or health-related events.
This ismainlydue tohealth resourceconstraintand
additionalburdenonoperatingthesysteminthelong
run.Rationalselectionofdiseasesoreventseligible
to be put under surveillance is thus important for
system establishment planning. In general, diseases
or health events suitable for surveillance are those
potentiallycausingseverehealthoutcomes(e.g.,death,
disability,long-termsuffering).Examplesofdiseases
and conditions in this category aremalaria,HIV/
AIDS, tuberculosis, rabies, and hazardous chemical
exposure.12-15
Highly transmissible diseases and epidemic
or pandemic diseases of international concern are
also considered for surveillance.Examples of these
diseases areEbola16, hemorrhagic fever,measles17,
and influenza18. Some conditions or health events
are put under surveillance for evaluation of health
program implementation and related effectiveness.
Examplesinthiscategoryaresurveillanceofinfluenza
vaccinationcoverage19andevaluationofpoliomyelitis
vaccine efficacy infield trial3.
Inadditiontotheselectionofdiseasesorconditions
eligibleforsurveillance,thereareseveralotherrelated
issuestobeconsidered.Dataandinformationrequired
foranalysisandformulationofapplicablepolicyand
implementationmustbewellspecified.Therequired
datamust also be feasible in terms of collection
fromappropriate sources andquality ascertainment.
Regardingcostconsideration,benefitofsurveillance
for a certain event shouldoutweigh the cost burden
arises from its operation.Cost containment–control
ofexpensesaccordingtobudgetconstraints–isalsoa
criticalissueinfluencingsustainabilityofsurveillance
systemoperation, especially in the long run.
Box1. Considerations for establishing
surveillance system
• Health resource constraints (critical issue especially setting up system requiring long-term operation)
• Diseaseorhazard • causing severehealthoutcomes (e.g., rabies, radiation hazard) • highly transmissiblediseases (e.g. Ebola) • epidemics / pandemic (e.g., influenza) • diseaseoutcome indicating success or failure in evaluationof interventional effectiveness (e.g., vaccine efficacy trial)
• Feasibility of obtainingqualitydata • Cost containment
4. Methodical process of surveillance Methodical process of surveillance comprises
several consecutive steps including data collection,
data analysis, interpretation of analysis results,
dissemination of information, and link to public
healthmeasures.Detailsofthesestepsareexplained
as followed.
4.1 Data collection
Data collection is themost critical phase of
surveillanceintermsofthegreatestamountofbudget
13 � 1.indd 35 6/10/15 4:36 PM
36
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
toinvestanddifficultyinobtainingqualitydata.Data
for surveillance primarily vary according to ‘health
events’ofinterest.Dataregardingsomeeventscanbe
obtainedthroughpopulationregisterorvitalstatistics
system(e.g.,mortality,cause-specificmortality,infant
mortality).20 Some events are recorded on routine
medicalpractice(e.g.,maternalandneonataloutcomes,
congenital defects, communicable diseases, chronic
diseases,mental illness, health practice, nosocomial
infection,surgicalsiteinfections21).Surveillanceisnot
only limited to detection of health problemonce it
occurs,butitsrangeofdatacanalsobeextendedto
include factors influencing the health problemprior
toitsoccurrenceormanifestation.22Theseinfluencing
factorscanbehealthriskbehaviors22,environmental23
andoccupationalhealthhazards24,foodcontamination,
diseasevectors,animal reservoirs,andothers.Some
eventsareshort-termconsequencesofcatastropheand
thesecanbetemporarilyputundersurveillanceuntil
the disaster is resolved. Examples of these events
areoccurrenceof infectiousdiseases, injuries,effect
onmental health, anddemandofmedical care after
earthquakeandtsunami.25,26Otherunusualeventscan
alsobenotified through tourgeprompt information
capture,assessmentofpotentialrisktopublichealth,
andimmediateactionifrequired.Examplesofthese
eventsareclusterofdisease,deathsofunidentifiable
cause, atypically severe case,mass animal deaths,
andmigrationofwild animals.
Surveillanceofsomediseasescanbeundertaken
atdifferentstages; rangingfromsurveillanceofrisk
behaviors,subclinicalstages,clinicalevents,treatment
outcomes,andtoultimateconsequenceofrecoveryor
death.An example of diseaseswith comprehensive
surveillance system isHIV/AIDS.27 Surveillance of
HIV/AIDScanbe commenced from surveillance of
riskbehaviors.HIVsero-surveillanceisconductedto
determineHIV infection in laboratory investigation,
early before the disease clinicallymanifests.HIV
cases–classifiedbydifferentcasedefinitionsforsur-
veillance–arereportedonroutinepatientcareservice
to reflect burdenof the disease and for planningof
patientcare(e.g.,multipleantiretroviraldrugsprovi-
sion).Drugresistancesurveillanceisalsoanintegral
partofHIV/AIDSpatientcareandfrequencyofreport
primarilydependsonlevelofinfectionandavailable
healthresourcesforconductingthisperiodicsurveil-
lance.
Numerous sources of data can be adopted for
publichealth surveillance.Variationofdata sources
isonaccountofseveralfactors;includingavailability
andaccessibilityofdata(e.g.,routinely-collecteddata),
budgetconstraint(e.g.,fundingforspecialeffortofdata
collection), characteristics of health service systems
(e.g., cooperationofpublicorprivatehealth service
providers), quality and availability of facility (e.g.,
laboratoryfacilities,computernetwork),andperson-
nel(e.g.,medicalspecialistsverifyingcasebasedon
surveillance case definition).2To specifywhat data
are needed, objective of surveillancemust be first
considered.Theobjectiveofsurveillancewouldguide
whichdataarerationaltobecollected–priortocheck-
ing feasibility toobtain thedata inpractical setting,
specifyingdatacollectionapproach,andanticipating
action to be recommendedby surveillance informa-
tion. For instance, if the objective of surveillance
is to detect a foodborneoutbreak, evidences related
to the suspected outbreak (e.g., abrupt increase in
numberofpatientsdiagnosedwithacutegastrointes-
tinaldisease,typeoffoodsharedincommonamong
13 � 1.indd 36 6/10/15 4:36 PM
37
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
cases)areneeded.Approachofdatacollection(e.g.,
active surveillance approach to directly collect food
specimen for laboratory investigation of causative
agent) can then be specified. Surveillance data
and their corresponding sources are exemplified in
Table 3.
Table 3 Surveillance data and corresponding sourcesor settings of data collection
Surveillancedata Sources or settingofdata collection
1. Mortality data2. Morbidity data
3. Data for epidemic detection
4. Data for laboratory surveillance
5. Data for analyzingdisease occurrence and risk factors
6.Data for healthcare andhealth system surveillance
7.Environmental data
Death registrywith recordof accurate causeof deathCase reporting (from routinemedical service),Individualcasereport(e.g.,casereportofraredisease,caseofemergingdisease,usual variant of commondisease),Compulsory report of casefindingby legal regulation for somediseases(e.g., cholera,Ebola hemorrhagic fever)Epidemicfield investigation,Cluster of illness reportedby community-based surveillance,Findingof unusual event (e.g., cluster of animalmass deaths)Serological survey,Laboratory report forHIV sero-surveillance,Laboratory identification of etiologic agent for diagnosis of a certain diseaseRecordof demographic data,Riskbehavior survey,Findingof disease vector or animal reservoirMedical care statistics,Record fromcooperative networkof health professionals,Recordof specifiedhealth indicators,Recordof specified administrative dataWater quality assessment,Measurement of hazardous chemical exposure in factory
13 � 1.indd 37 6/10/15 4:36 PM
38
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
There are five data collection approaches in
surveillance including passive surveillance, active
surveillance,sentinelsurveillance,community-based
surveillance,andsyndromicsurveillance.Keyfeatures
of these approaches are summarized inTable 4.
Data collectionapproaches insurveillance
Key features
1)Passive surveillance
2)Active surveillance
3)Sentinel surveillance
4)Community surveillance
5)Syndromic surveillance
• Data reportedbyprovidersworking in routinepractice• Aim tomeasuremagnitudeof health problems• Economical and likely to be sustainable in a long run• Obtain only the data of individuals utilizing facility• Data canbeused as baseline for further considerationof additional active surveillance.• Directdatacollectionfromoriginalsourcebypersonnelinsurveillancesystem
themselves of a certain objective• Aimtoprovidebetterestimateofprevalenceorburdenofdiseasefromcases
living in a specified community.• Requiremore budget, time, personnel, and other resources to conduct than
required for passive surveillance• Preferablyperiodical undertaking• Data specifically collected for analysis of disease pattern• Focusondataofacertainsubgroupratherthanthewholepopulation,acertain
catchment area, or a certain disease factor• Canbe either passiveor active surveillance (dependingon the role of surveillancepersonnel in data collection)• Community residents act as data collector.• Report of unusual event or outbreak in community and further urge investigation.• Notablyuseful for initiation of outbreak investigation• Focusondataofclinical syndromeoccurringbeforecompleteprogression to
disease status,or signsandsymptomsoccurringprior todiagnosisofdisease• Some signs and symptoms canbeput under surveillanceprior to certain diagnosis of disease.Examples of syndromeput under surveillance are Influenza-like illness (ILI) andAcuteFlaccidParalysis (AFP).
Table 4 Data collection approaches in surveillance and correspondingkey features
13 � 1.indd 38 6/10/15 4:36 PM
39
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
4.2Dataanalysis
Data analysis approach in descriptive study is
usuallyadoptedforanalysisofsurveillancedata.Typical
aimof analysis is to revealmagnitude, pattern, and
trend of a certain health problem.To achieve this
goal, data are analyzed to primarily identifymag-
nitude using appropriate indicator for health event
of interest. Prevalence is a commonmeasurement
ofmagnitude in caseofnon-communicabledisease.
Incidencecanalsobeofchoicetoidentifymagnitude
incaseof injuryandcancer. Inaddition to identify-
ingmagnitudeofproblem,pattern and trendcanbe
analyzedbyperson(personatrisk),place(riskarea),
and time (time at risk).
Affectedgroupofpersons,orthe‘person’element,
is generally describedbydemographic factors (e.g.,
age,gender,socioeconomicstatus).Othercharacter-
istics can also be described, based on objective of
surveillance.These characteristics are, for instance,
presenceofriskfactor(e.g.,smokingstatus),presence
of protective factor (e.g., history of immunization),
underlying systemic disease, and personal hygiene
practice (e.g., toothbrushinghabit).
The ‘place’ element of analysis is geographical
analysis in relation to the health event.Urban and
ruralareasaredifferentintermsoflivingenvironment,
population density, availability and accessibility to
healthfacility,andseveralotherconditions.Geographi-
cal analysiswould additionally provide information
on area-specific factorswhich potentially influence
health anddisease in residentsdwelling in the area.
Specifyingcatchmentareaofinterestwouldalsoallow
measurement of disease frequency (e.g., prevalence
andincidence).Diseaseoccurrenceonly inacertain
placewould imply risk areawhich further urges
investigationof area-specific factorwhich influence
suchoccurrence.
Analysisof‘time’elementisalsoimportantsince
natural course of disease requires time period to
progress(e.g.,inductionperiodfornon-communicable
disease occurrence and incubation period for infec-
tiousdisease).Moreover,dynamicchangeofdisease
occurrencecanbeobservedovertime.Therearefour
time trends commonly analyzed in epidemiology;
including secular trend, seasonal pattern, cyclical
trend,andepidemicdiseaseoccurrence.2Analysisof
health eventover a longperiodof time (e.g., years,
decade) is recognized as ‘secular trend analysis’.
Graphicaldisplayofdataisusuallyappliedtoreveal
how occurrence of the health event in a defined
population changes over the long observed period.
‘Seasonalpattern’ canbe revealedwhenoccurrence
of health event exhibits a certain seasonal pattern.
Denguehemorrhagic fever is anexampleofdisease
with seasonal variation in terms of transmission.
‘Cyclical trend’ can be analyzed for health event
hypothesized to have cycle of occurrence or repeat
a certain pattern again and again overtime.An ex-
ampleofhealtheventwithcyclical trend iscyclical
vomitingsyndrome.‘Epidemicdiseaseoccurrence’is
characterizedbydiseaseoutbreakordiseaseoccurrence
whichexceedsregularlyexpectedoccurrencerate in
a particular period of time. The outbreak of acute
foodbornegastrointestinaldiseaseinOswegoCounty
is a classic example for the epidemic occurrence of
disease.28Elements in analysis of surveillance data
canbe summarized in the followingBox2.
13 � 1.indd 39 6/10/15 4:36 PM
40
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
Box2.Elements inanalysis of surveillancedata
• Measuring ‘magnitude’ ofproblem
(using epidemiological measurements–incidence and prevalence)
• Descriptionofpatternand trend
• Analysisby ‘person’ characteristics
• Age (e.g., disease incidence by age)
• Gender (e.g., lung cancer risk behavior by gender)
• Ethnicgroups (e.g., incidence of Tuberculosis among different ethnic group in the USA)
• Marital status (e.g., cervical cancer incidence among single and married women)
• Occupation (e.g., symptoms indicating pesticide exposure among chili farmers)
• Socioeconomicstatus(e.g., malnutrition among children in low socioeconomic status families)
• Analysisby ‘place’ characteristics’
• International comparison (e.g., estimated numbers of HIV/AIDS cases among different Asian
countries)
• Intra-country comparison (e.g., comparison of infant mortality rate by regions in Thailand)
• Urban-ruralcomparison (e.g., incidence of diabetes mellitus by area of residence in Thailand)
• Localdistributionofdisease (e.g., spot map of dengue hemorrhagic fever cases in a certain
district)
• Analysisby ‘time’ characteristics’
• Timeonset (e.g., incubation period in infectious disease)
• Secular trend (e.g., twenty-year trend incidence and mortality of cardiovascular disease in the
United States)
• Cyclical variation (e.g., mode of occupational injuries by month)
• Seasonalpattern (e.g., seasonal pattern of dengue hemorrhagic fever)
• Point epidemic (e.g., foodborne outbreak)
13 � 1.indd 40 6/10/15 4:36 PM
41
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
4.3 Data interpretation
Interpretationofsurveillancedataiscriticalsince
theinterpretationwouldfurtherleadtoconsideration
whether public health action is really needed.Key
issues in data interpretation is the identification of
accurate increase in disease occurrence–to a certain
extentthatextrapublichealthactionispromptlyneeded
to timely control or prevent the disease.2Observed
increaseindiseaseoccurrencecanbeconfoundedor
influenced by other factors. The observed increase
maybe due to the larger size of population inves-
tigated.2Disease screening campaign and improved
diagnostic techniquewith better sensitivity can as
well increase case finding.2Reporting system also
determine the number of case to be found.2 These
factors should be rationally ruled out beforemak-
ingconclusive interpretationofdata that thedisease
occurrence is actually increased.
Different epidemiological measures provide
different information. Rational selection of these
measures to give an answer to a specific question
regarding situationof interestedhealthevent is thus
crucial.Epidemiologicalmeasurescommonlyusedin
surveillanceareincidencerate,incidenceproportion,
period prevalence, point prevalence,mortality rate,
and case fatality rate.Thesemeasuresmust bewell
selected since they allowdifferent implications and
interpretationof themeasuresmust be scientifically
sound.
4.4 Data dissemination and link to public
healthaction
Major issue in surveillance data dissemination
is ‘who need to know?’. Since the primary aim of
surveillanceistoprovideinformationforpublichealth
action,authoritativehealthpersonnelwhorequirein-
formationfordecisionwhethertoactarethustheones
mustbeinformed.Thepatternofdatadissemination
fromlocalorsubsequent levels to thehigherhierar-
chies can be viewed as a ‘down-top’ dissemination
ofdata.Thisdirectionofdatadisseminationusually
link to public action in termsof ‘planned response’
or planning for control andpreventionof disease.
Nevertheless,thehealthpersonnelatsubsequent
levelsmust also bewell informed of the situation.
This is due to the fact that local personnelmust
provide ‘acute response’ in case that prompt action
is needed, such as the case of outbreak.Moreover,
implementationofpolicyatlocallevelsrequiresjudg-
ment and applicationwhich is relevant to the local
contextorsettingbythesepersonnel.This‘top-down’
dissemination of data and policy can also provide
feedbackand stimulate improvement in surveillance
datareportingatthelocallevels.Theoverallsystem
of surveillance can be illustrated in the following
Figure 1.
13 � 1.indd 41 6/10/15 4:36 PM
42
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
5. Challenging issues and technical advancement in surveillance Practice of surveillance continuously evolves in
responsetothedynamicchangeinhealthanddisease
condition in population. From the early concept of
detectinginfectedcase,focusofmodernapproachin
infectiousdiseasesurveillancehasshiftedtoforecasting
futureincidenceoroutbreakwithadvancedtechniques
such asmathematicalmodeling.29,30 This approach
enablestimelywarning,preparationofhealthfacilities,
andpreparednessofprofessionalstoproperlymanage
abruptly increaseddemand.Nonetheless, challenges
in forecastingexist especially in termsofpredictive
accuracyof futureoccurrence.31
Surveillanceofemergingandre-emerginginfec-
tiousdiseases isanotherfieldwhich rapidlyevolves
in accordance with dynamic change in lifestyle
(e.g., drug abuse)32, altered ecosystem (e.g., land
Figure1Overall systemof surveillanceand link topublichealthaction
use, deforestation, pesticide use)33, food production
(e.g.,EscherichiacoliO104:H4contamination)34,and
globalization (e.g., Ebola pandemic)35. Since the
disease emergence is influence by various factors,
interdisciplinarycollaborationandapplyingmultiple
strategies to strengthen corresponding surveillance
system can be a solution. ‘One health’ disease
surveillance,forexample, isanapproachcombining
expertise in several disciplines (e.g., veterinary and
environmental sciences) to control diseases (e.g.,
zoonoses).36Advances in laboratory-based surveil-
lancealsocontributetonearreal-timerecognitionof
outbreakincommunity.37Inresponsetoglobalization,
InternationalHealthRegulations(IHR)hasbeenestab-
lishedfor internationalcommunity toco-operatively
build up international surveillance systemwhich
timely detects, notify and response to public health
risks–such as surveillance in international airports.38
Policymaking for control andprevention
Acuteaction to control andprevent transmissionor
progression
13 � 1.indd 42 6/10/15 4:36 PM
43
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
Major challenges in implementing the IHR include
requirement of expertise and resource, governance,
international collaboration, andpolitical barrier.38
Surveillance of non-communicable diseases is
growing in its importance. Itsmodern paradigm
has shift towards health promotion and prevention.
With advances in causal research in epidemiology,
risk factors ofmany non-communicable diseases
havebeenrevealed.Suchknowledgefurtherenables
establishment of risk factor surveillance formajor
non-communicable diseases (e.g., cardiovascular
disease).39,40Behavioralriskfactorsarealsoofinterest
inmodernsurveillancesincemanynon-communicable
diseases are related to lifestyle and personal health
behavior.41
Conclusion The concept of surveillance has long been
developed from individual-based monitoring to
population-based surveillancewith organized sys-
tem to generate quality informationwhich links to
publichealthaction.Objectiveofsurveillancemustbe
specifiedprior to data collection since the objective
wouldindicatewhichdataareneededfortherequired
information.Varioushealtheventscanbeputunder
surveillanceandnotonlylimitedtodiseaseoccurrence.
Differentdata collection approaches canbe adopted
to suit the context of surveillance. In analysis of
surveillancedata,descriptiveepidemiologicalmethod
isimportantlyadoptedtorevealmagnitudeandpattern
ofhealthproblem.Longitudinaldatacollectionwould
additionallyallowanalysisoftrendofsuchproblem.
Informationobtainedfrominterpretationisprimarily
disseminated to those in need of such information.
Link to public health action can be either informa-
tion for prompt action or information for planning.
Since health and disease condition in population is
dynamic; practice of surveillance is also advanced
withmedical and information technology, and the
paradigmshifttowardspreventivehealthorientation.
Acknowledgement
Author wished to express deep gratitude for
Prof.Dr.JayantonPatumanond(FacultyofMedicine,
ThammasatUniversity) andMs. Jongkol Podang
(Senior lecturer, Faculty of PublicHealth,Mahidol
University)fortheirconstructivecommentsondrafts
of this article. This academic article is supported
byMahidolUniversity throughTalentManagement
Project to promote research activity.
References 1. Etymonline.com.OnlineEtymologyDictionary
[Internet].2014[cited29October2014].Avail-
able from: http://www.etymonline.com/index.
php?term=surveillance
2. Declich S, CarterAO. Public health surveil-
lance: historical origins,methods and evalua-
tion.BulletinoftheWorldHealthOrganization,
1994;72(2):285–304.
3. Francis T. Evaluation of the 1954 poliomy-
elitisvaccinefieldtrial:furtherstudiesofresults
determining the effectiveness of poliomyelitis
vaccine(Salk)inpreventingparalyticpoliomyeli-
tis.JournaloftheAmericanMedicalAssociation.
1955;158(14):1266-1270.
4. Thacker S, BerkelmanR. Public health sur-
veillance in theUnited States. Epidemiologic
Reviews.1988;10:164-90.
13 � 1.indd 43 6/10/15 4:36 PM
44
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
5. LucasAO.The surveillance of communicable
diseases.WHOChronicle,1968;22:439-444.
6. LangmuirA.WilliamFarr: founder ofmodern
conceptsofsurveillance.InternationalJournalof
Epidemiology.1976;5(1):13-18.
7. LangmuirA.Evolutionoftheconceptofsurveil-
lance in theUnitedStates. Proceedings of the
RoyalSocietyofMedicine.1971;64(6):681.
8. HallH,CorreaA,YoonP,BradenC.Lexicon,
definitions, and conceptual framework forpub-
lichealthsurveillance.MorbidityandMortality
WeeklyReport.2012;61(Suppl):10-14.
9. AbdallahS,PanjabiR.Epidemiologyandsurveil-
lance.In:RandE,ed.by.TheJohnsHopkinsand
RedCrossRedCrescentPublichealthguide in
emergencies.2nded.Geneva:InternationalFed-
erationofRedCrossandRedCrescentSocieties;
2008.p.220-283.
10. LastJ.Adictionaryofepidemiology.NewYork:
OxfordUniversityPress;2001.
11. NsubugaP,WhiteM,ThackerS,AndersonM,
Blount, S,BroomeC et al. PublicHealthSur-
veillance:AToolforTargetingandMonitoring
Interventions.In:JamisonD,BremanJ,Measham
A,AlleyneG,ClaesonM,EvansDetal.,ed.by.
DiseaseControlPriorities inDevelopingCoun-
tries.2nded.WashingtonDC:WorldBank;2006.
p.997-1015.
12. DouganS,EvansB,MacdonaldN,GoldbergD,
GillO,FentonKetal.HIVingayandbisexual
menintheUnitedKingdom:25yearsofpublic
healthsurveillance.EpidemiologyandInfection.
2008;136(02):145-156.
13. HedtB,LauferM,CohenT.Drugresistancesur-
veillanceinresource-poorsettings:currentmeth-
odsandconsiderationsforTB,HIV,andmalaria.
TheAmericanJournalofTropicalMedicineand
Hygiene.2011;84(2):192-199.
14. [Internet].2014[cited30October2014].Available
from: http://www.who.int/rabies/epidemiology/
Rabiessurveillance.pdf
15. [Internet].2014[cited30October2014].Available
from:http://www.safeworkaustralia.gov.au/sites/
SWA/about/Publications/Documents/765/Hazard-
ous-chemicals-requiring-health-monitoring.pdf
16. Pillais,Nyenswaht,RouseE,ArwadyM,For-
rester J,Hunter Jet al.Developingan Incident
ManagementSystemtoSupportEbolaResponse
—Liberia, July–August 2014.Morbidity and
MortalityWeeklyReport.2014;63(41):930-933.
17. HershB,TambiniG,NogueiraA,CarrascoP,de
QuadrosC.Reviewofregionalmeaslessurveil-
lancedataintheAmericas,1996–99.TheLancet.
2000;355(9219):1943-1948.
18. GensheimerK,FukudaK,BrammerL,CoxN,
PatriarcaP, StrikasR.Preparing forPandemic
Influenza:TheNeedforEnhancedSurveillance.
EmergingInfectiousDiseases.1999;5(2March-
April):297-298.
19. LuP,SantibanezT,WilliamsW,ZhangJ,Ding
H,BryanLetal.SurveillanceofInfluenzaVac-
cinationCoverage—United States, 2007–08
Through2011–12InfluenzaSeasons.Morbidity
andMortalityWeeklyReport.2013;62(4):1-28.
20. XuJ,KochanekK,MurphyS,AriasE.Mortal-
ityintheUnitedStates,2012.NCHSdatabrief.
2014;(168):1-8.
21. EuropeanCentre forDisease Prevention and
Control.Surveillanceof surgical site infections
inEurope2010–2011.Stockholm:ECDC;2013.
13 � 1.indd 44 6/10/15 4:36 PM
45
Journal of Public Health and Development Vol. 13 No. 1 January - April 2015
22. XuF,TownM,BalluzL,BartoliW,Murphy
W,ChowdhuryPetal.SurveillanceforCertain
HealthBehaviorsAmongStates and Selected
LocalAreas—UnitedStates,2010.Morbidityand
MortalityWeeklyReport.2013;62(1):1-247.
23. AbelsohnA, Frank J, Eyles J. Environmental
publichealthtracking/surveillanceinCanada:a
commentary.HealthcarePolicy.2009;4(3):37-52.
24. LowryS,BleckerH,CampJ,DeCastroB,Hecker
S,ArbabiSetal.Possibilitiesandchallengesin
occupational injury surveillance of day labor-
ers.American Journal of IndustrialMedicine.
2010;53(2):126-134.
25. IwataO,OkiT,IshikiA,ShimanukiM,Fuchimu-
kaiT,ChosaTetal.Infectionsurveillanceaftera
naturaldisaster:lessonslearntfromtheGreatEast
JapanEarthquakeof2011.BulletinoftheWorld
HealthOrganization.2013;91(10):784-789.
26. NoharaM.ImpactoftheGreatEastJapanEarth-
quakeand tsunamionhealth,medical careand
publichealthsystemsinIwatePrefecture,Japan,
2011.WesternPacificSurveillanceandResponse
Journal.2011;2(4):e1-e1.
27. SweeneyP,GardnerL,BuchaczK,GarlandP,
MugaveroM,BosshartJetal.ShiftingthePara-
digm:UsingHIVSurveillanceDataasaFounda-
tionforImprovingHIVCareandPreventingHIV
Infection.MilbankQuarterly. 2013;91(3):558-
603.
28. GrossM.OswegoCountyrevisited.PublicHealth
Report.1976;91(2):168–170.
29. Chretien J, GeorgeD, Shaman J, Chitale R,
McKenzie F. Influenza Forecasting inHuman
Populations:A ScopingReview. PLoSONE.
2014;9(4):e94130.
30. Siettos C, Russo L.Mathematicalmodeling
of infectious disease dynamics. Virulence.
2013;4(4):295-306.
31. HyderA,BuckeridgeD, LeungB. Predictive
ValidationofanInfluenzaSpreadModel.PLoS
ONE.2013;8(6):e65459.
32. El-BasselN,ShawS,DasguptaA,StrathdeeS.
DruguseasadriverofHIVrisks.CurrentOpinion
inHIVandAIDS.2014;9(2):150-155.
33. McFarlaneR,SleighA,McMichaelA.Land-Use
ChangeandEmergingInfectiousDiseaseonanIs-
landContinent.IJERPH.2013;10(7):2699-2719..
34. GradY,GodfreyP,CerquieraG,Mariani-Kurkd-
jianP,GoualiM,BingenEet al.Comparative
GenomicsofRecentShigaToxin-ProducingEs-
cherichiacoliO104:H4:Short-TermEvolutionof
anEmergingPathogen.mBio.2012;4(1):e00452-
12-e00452-12.
35. GilsdorfA,MorganD,LeitmeyerK.Guidance
forcontacttracingofcasesofLassafever,Ebola
orMarburghaemorrhagic feveronanairplane:
resultsofaEuropeanexpertconsultation.BMC
PublicHealth.2012;12(1):1014.
36. Karimuribo E, Sayalel K, Beda E, Short N,
WamburaP,MboeraLetal.TowardsOneHealth
diseasesurveillance:TheSouthernAfricanCentre
for InfectiousDisease Surveillance approach.
Onderstepoort Journal ofVeterinaryResearch.
2012;79(2).
37. ViñasM,TuduriE,GalarA,YihK,PichelM,
Stelling J et al. Laboratory-BasedProspective
SurveillanceforCommunityOutbreaksofShig-
ellaspp.inArgentina.PLoSNeglectedTropical
Diseases.2013;7(12):e2521.
13 � 1.indd 45 6/10/15 4:36 PM
46
วารสารสาธารณสุขและการพัฒนา ปีที่ 13 ฉบับที่ 1 มกราคม - เมษายน 2558
38.GreshamL, SmolinskiM, Suphanchaimat R,
KimballA,WibulpolprasertS.CreatingaGlobal
Dialogue on InfectiousDisease Surveillance:
ConnectingOrganizationsforRegionalDisease
Surveillance(CORDS).EmergingHealthThreats
Journal.2013;6(0).
39. NgN,MinhH,JuvekarS,RazzaqueA,BichT,
KanungsukkasemUetal.Using theINDEPTH
HDSStobuildcapacityforchronicnon-commu-
nicablediseaseriskfactorsurveillanceinlowand
middle-incomecountries.GlobalHealthAction.
2009;2(0).
40. BloomfieldG,MwangiA,ChegeP,SimiyuC,
AswaD,OdhiamboDetal.Multiplecardiovas-
cularriskfactorsinKenya:evidencefromahealth
anddemographic surveillance systemusing the
WHOSTEPwise approach to chronic disease
riskfactorsurveillance.Heart.2013;99(18):1323-
1329.
41. LiC,ZhaoG,OkoroC,WenX,FordE,Balluz
L. Prevalence ofDiagnosedCancerAccording
toDurationofDiagnosedDiabetesandCurrent
InsulinUseAmongU.S.AdultsWithDiagnosed
Diabetes: Findings from the 2009Behavioral
RiskFactorSurveillanceSystem.DiabetesCare.
2013;36(6):1569-1576.
13 � 1.indd 46 6/10/15 4:36 PM