(H)DSS CONCEPTS AND METHODS - CU Population Center€¦ · Importance of HDSS Provide health...
Transcript of (H)DSS CONCEPTS AND METHODS - CU Population Center€¦ · Importance of HDSS Provide health...
(H)DSS CONCEPTS AND (H)DSS CONCEPTS AND METHODSMETHODS
[(HEALTH) AND DEMOGRAPHIC [(HEALTH) AND DEMOGRAPHIC SURVEILLANCE SYSTEM]SURVEILLANCE SYSTEM]
Nizam KhanNizam KhanInstitute of Behavioral Sciences (IBS)Institute of Behavioral Sciences (IBS)
University of Colorado, BoulderUniversity of Colorado, BoulderJuly 17, 2008July 17, 2008
What is HDSSWhat is HDSSSet of field and computing operations to handle the longitudinalSet of field and computing operations to handle the longitudinal followfollow--up of wellup of well--defined entities or primary subjects (individuals, households, adefined entities or primary subjects (individuals, households, and residential units) nd residential units) and all related demographic and health outcomes within a clearlyand all related demographic and health outcomes within a clearly circumscribed circumscribed geographic area.geographic area.
•• Site selection:Site selection:•• Census: Define and register the target populationCensus: Define and register the target population
Collected at prescribed intervalsCollected at prescribed intervals
•• Core system: Monitor population dynamics through routine collectCore system: Monitor population dynamics through routine collection and ion and processing of information on births, deaths, and migrations processing of information on births, deaths, and migrations
•• Complementary systems: Collects various other data sets that proComplementary systems: Collects various other data sets that provide vide important social and economic correlates of population and healtimportant social and economic correlates of population and health dynamics:h dynamics:
•• Union formationUnion formation•• Asset, incomeAsset, income•• EducationEducation•• MorbidityMorbidity
Importance of HDSSImportance of HDSS
Provide health information that more accurately reflects Provide health information that more accurately reflects the prevailing disease burden;the prevailing disease burden;
Assist in monitoring and tracking new health threats, Assist in monitoring and tracking new health threats, such as emerging and reemerging infectious disease such as emerging and reemerging infectious disease and drug resistance, and alert the health community to and drug resistance, and alert the health community to prepare response;prepare response;
Serve as a platform for actionServe as a platform for action--oriented research to test oriented research to test and evaluate health interventions, such as new vaccines and evaluate health interventions, such as new vaccines or drugs, health education messages and costor drugs, health education messages and cost--effectiveness of initiatives.effectiveness of initiatives.
HDSS SitesHDSS Sites
During the past 30 years, DSSs have been established During the past 30 years, DSSs have been established in a number of field research sites in various parts of the in a number of field research sites in various parts of the developing world.developing world.
Currently there are 37 DSS sites in 19 countries that are Currently there are 37 DSS sites in 19 countries that are member of INDEPTH network.member of INDEPTH network.
25 sites in 25 sites in AfricaAfrica10 sites in 10 sites in Asia Asia 1 in 1 in Oceania Oceania and and 1 in 1 in Central Central America. America.
HDSS Sites (contd.)HDSS Sites (contd.)Asia, C/America & OceaniaAsia, C/America & Oceania
Chilillab DSS, VietnamChilillab DSS, VietnamFilabavi DSS, VietnamFilabavi DSS, VietnamChakaria DSS, BangladeshChakaria DSS, BangladeshHSID DSS, BangladeshHSID DSS, BangladeshMatlab DSS, BangladeshMatlab DSS, BangladeshWatch HDSS, BangladeshWatch HDSS, BangladeshKanchanaburi DSS,ThailandKanchanaburi DSS,ThailandPuworejo DSS, IndonesiaPuworejo DSS, IndonesiaVadu DSS, IndiaVadu DSS, IndiaWosera, Papua New GuineaWosera, Papua New GuineaLeon HDSS, Nicaragua Leon HDSS, Nicaragua
Africa Africa ACDIS HDSS, South AfriACDIS HDSS, South AfricacaAgincourt HDSS, South AfricAgincourt HDSS, South AfricaaDikgale HDSS, South AfricaDikgale HDSS, South AfricaDodowa HDSS, GhanDodowa HDSS, GhanaaKintampo HDSS, GhanKintampo HDSS, Ghana a Navrongo HDSS, GhanNavrongo HDSS, Ghanaa
Africa Africa Ifakara HDSS, TanzaniaIfakara HDSS, TanzaniaMagu HDSS, TanzaniaMagu HDSS, TanzaniaRufiji HDSS, TanzaniaRufiji HDSS, TanzaniaIganga/Mayuge HDSS UgandaIganga/Mayuge HDSS UgandaRakai HDSS, UgandaRakai HDSS, UgandaKilifi HDSS, KenyaKilifi HDSS, KenyaKisumu HDSS, KenyaKisumu HDSS, KenyaNairobi HDSS, KenyaNairobi HDSS, KenyaNouna HDSS, Burkina FasoNouna HDSS, Burkina FasoOubritenga HDSS, Burkina FasoOubritenga HDSS, Burkina FasoOuagadougou HDSS, Burkina FasoOuagadougou HDSS, Burkina FasoSapone HDSS, Burkina FasoSapone HDSS, Burkina FasoButajira HDSS, EthiopiaButajira HDSS, EthiopiaManhica, MozambiqueManhica, MozambiqueBandim HDSS, Guinea BissauBandim HDSS, Guinea BissauFarafenni HDSS, GambiaFarafenni HDSS, GambiaNiakhar HDSS, SenegalNiakhar HDSS, SenegalKaronga HDSS, MalawiKaronga HDSS, Malawi
HDSS Sites (contd.)HDSS Sites (contd.)
What is INDEPTH networkWhat is INDEPTH networkINDEPTH NetworkINDEPTH Network: : IInternational nternational NNetwork for the etwork for the DDemographic emographic EEvaluation of valuation of PPopulations and opulations and TTheir heir HHealth in ealth in developing countries. developing countries.
http://www.indepthhttp://www.indepth--network.net/network.net/
The vision of INDEPTH is to harness the collective potential The vision of INDEPTH is to harness the collective potential of the world's communityof the world's community--based longitudinal demographic based longitudinal demographic surveillance initiatives in resource constrained countries. surveillance initiatives in resource constrained countries. Collective work will provide a better, empirical Collective work will provide a better, empirical understanding of health and social issues, and enable understanding of health and social issues, and enable scientists to apply this understanding to alleviate the most scientists to apply this understanding to alleviate the most severe health and social challenges.severe health and social challenges.
CU, Boulder has institutional links CU, Boulder has institutional links with two DSS siteswith two DSS sites
MATLAB HDSS, BANGLADESHMATLAB HDSS, BANGLADESH
AGINCOURT HDSS, SOUTH AFRICAAGINCOURT HDSS, SOUTH AFRICA
MATLAB HDSS, BANGLADESHMATLAB HDSS, BANGLADESH
It is recognized as the largest and longest It is recognized as the largest and longest sustained DSS.sustained DSS.
Matlab HDSS started operations in 1966 as Matlab HDSS started operations in 1966 as major component of the field research program major component of the field research program of ICDDR,B.of ICDDR,B.
It has made significant contributions to health It has made significant contributions to health development in both Bangladesh and the rest of development in both Bangladesh and the rest of the world.the world.
MATLAB: ObjectivesMATLAB: Objectives
• To provide a small-area registration system which is suitable for assessment of the effectiveness, safety and acceptability of maternal and child health and family planning interventions.
• To undertake research related to diarrhea diseases, and on the measurement and determinants of fertility and mortality.
• To develop a demographic field site that can be used for training of program planners, researchers and implementers.
Matlab: Site and Population (2005)Matlab: Site and Population (2005)Located in Chandpur district. About 55 km southeast of the country’s
capital, Dhaka.
~184 sq. km.142 villages~46,716 households~46,716 households~224,762 people~224,762 peopleAvg. Hhold size:4.8Avg. Hhold size:4.888% Muslim88% Muslim12% Hindu*12% Hindu*TFR: 2.8TFR: 2.8ee0:0: Male~66.8 yearsMale~66.8 years
Female~ 70.8 years
70% of Hholds are either landless 70% of Hholds are either landless or functionally landless (land<.050 or functionally landless (land<.050 acres)acres)30% has no formal schooling30% has no formal schooling12% farming (head of the 12% farming (head of the household)household)Field visit every monthsField visit every months
Female~ 70.8 years
Matlab Site selectionMatlab Site selectionIn 1960 Cholera Research Laboratory was In 1960 Cholera Research Laboratory was established in Dhaka to develop a vaccine which established in Dhaka to develop a vaccine which would prevent, an eventually eradicate, cholera. would prevent, an eventually eradicate, cholera. An essential element of this program was the An essential element of this program was the implementation of large scale controlled vaccine implementation of large scale controlled vaccine field trials. This required selection of a field site field trials. This required selection of a field site where certain conditions could be met, including where certain conditions could be met, including the presence of endemic cholera and year round the presence of endemic cholera and year round
access to villages. Matlab Thana was chosen. access to villages. Matlab Thana was chosen.
Matlab: TFR (per woman)Matlab: TFR (per woman)
0
1
2
3
4
5
6
7
1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005
Treatment
Govt area
Matlab: Under five mortality Matlab: Under five mortality (per 1000 live births)(per 1000 live births)
0
50
100
150
200
250
1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005
Treatment
Govt area
Matlab: Main environmental issuesMatlab: Main environmental issues
Embankment divide the study area Embankment divide the study area into two equal parts.into two equal parts.
Arsenic problem.Arsenic problem.
DataData
All vital events (birth, death, All vital events (birth, death, marriage, migration).marriage, migration).
Census and SES, 1974, 1982, 1996, Census and SES, 1974, 1982, 1996, 2005.2005.
Site of various health and population Site of various health and population studies.studies.
Agincourt HDSS, South AfricaAgincourt HDSS, South Africa
A rural subdistrict of Bohlabelo District, A rural subdistrict of Bohlabelo District, Limpopo Province, near Mozambican Limpopo Province, near Mozambican border border
!
!
!
DURBAN
CAPE TOWN
JOHANNESBURG
NambiaBotswana
Zimbabwe
Moz
ambi
que
South Africa
#AgincourtStudy Site Kruger
National Park
Started in 1992 as part of a program of Started in 1992 as part of a program of decentralized health systems research decentralized health systems research and development.and development.
By late 1990s, the Agincourt program was By late 1990s, the Agincourt program was transformed to a site of advanced transformed to a site of advanced communitycommunity--based research program to based research program to better understand the health and better understand the health and population dynamics in rural South Africa.population dynamics in rural South Africa.
Agincourt site selectionAgincourt site selection
Its location, some distance from any tar road or township Its location, some distance from any tar road or township settlement;settlement;
The presence of health center, with its satellite clinics The presence of health center, with its satellite clinics and its potential to function as a referral networkand its potential to function as a referral network
The need to develop rational pattern, delinked The need to develop rational pattern, delinked constraints imposed by homeland boundaries; andconstraints imposed by homeland boundaries; and
The presence of large numbers of Mozambicans, The presence of large numbers of Mozambicans, displaced by the recent civil war. displaced by the recent civil war.
Agincourt site and population Agincourt site and population (2003)(2003)
ee0:0: Male~52 yearsMale~52 yearsFemale~ 60 yearsFemale~ 60 years
CDR (`92CDR (`92--`93): 5.0/1000 `93): 5.0/1000 (`03(`03--`04): 10.9/1000`04): 10.9/1000
TFR (1970TFR (1970--74): 674): 62003: 2.32003: 2.3
36% female headed household36% female headed householdVisit annually
400 sq km400 sq km21 villages21 villages~11,600 Hholds~11,600 Hholds~72,000 people~72,000 peopleAvg. Hhold Avg. Hhold size:6size:631% Mozambican31% Mozambican69% South African69% South African
Visit annually
AHDSS data modulesAHDSS data modulesIndividual level dataIndividual level data
Baseline census, 1992 : updated annuallyBaseline census, 1992 : updated annuallyVital events (births, death, in and out migration): updated annuVital events (births, death, in and out migration): updated annuallyallyEducation (updated 1992, 1997, 2002 and new individuals)Education (updated 1992, 1997, 2002 and new individuals)Cough (1999 Cough (1999 –– one screening question)one screening question)Stroke (2001 Stroke (2001 –– two screening questions)two screening questions)Health Care Utilization (2003)Health Care Utilization (2003)Temporary migrations (2002)Temporary migrations (2002)Labor participation (2000 and 2004)Labor participation (2000 and 2004)Marital status Marital status –– current and historical (2005)current and historical (2005)
Household level dataHousehold level dataAssets (2001, 2003, 2005)Assets (2001, 2003, 2005)Child Care Grants (2002, 2005)Child Care Grants (2002, 2005)Food Security (2004)Food Security (2004)
AHDSS data storageAHDSS data storageAll data are stored in Relational database which holds data in All data are stored in Relational database which holds data in multiple interrelated tables;multiple interrelated tables;
Every Table represents either an object (for example, a person, Every Table represents either an object (for example, a person, location) or an event (for example, birth, death, marriage);location) or an event (for example, birth, death, marriage);
Data in different tables is connected not physically but logicalData in different tables is connected not physically but logically. The ly. The logical connection between two tables is a field which both havelogical connection between two tables is a field which both have in in common;common;
Every Table has a primary key which uniquely identifies each recEvery Table has a primary key which uniquely identifies each record ord in a given table.in a given table.
Use SQL or Microsoft Access to create a dataset for a research Use SQL or Microsoft Access to create a dataset for a research project or use programs like State or SAS to create dataset;project or use programs like State or SAS to create dataset;
Very complex configurationVery complex configuration……need to very familiar with conceptual need to very familiar with conceptual and concrete definitions within database to export datasets fromand concrete definitions within database to export datasets fromrelational database. relational database.
Core TablesCore TablesEpisodes TablesEpisodes Tables-- ResidencesResidences
* Person at geographic location* Person at geographic location* Can start with a Enrollment* Can start with a Enrollment(1992), birth or in migration(1992), birth or in migration* Can end with death or out* Can end with death or outmigrationmigration
-- MembershipsMemberships* Person* Person’’s membership to as membership to a
householdhousehold* Can start with a Enrollment (1992), * Can start with a Enrollment (1992),
birth or in migrationbirth or in migration* Can end with death or out migration or* Can end with death or out migration or
change of household headchange of household head
-- Union EpisodesUnion Episodes* Person in union with one and only one * Person in union with one and only one
other person
Object TablesObject Tables-- IndividualsIndividuals-- LocationsLocations-- HouseholdsHouseholds
Event TablesEvent Tables-- DeathsDeaths-- BirthsBirths-- PregnanciesPregnancies-- ImmigrationsImmigrations-- Out MigrationsOut Migrations
other person
ASsetStatusHousehold
ObsYear
Construct
Walls
Roof
Floor
SepKit
SepLiv
ToiletFac
ToiletType
WaterSup
WaterAvail
DistMetre
DistMin
PowerLight
PowerCook
Fridge
TV
CellPhone
Cattle
Goats
Poultry
BirthsResidence
Pregnancy
Birthweight
BreAStEver
BreAStDurType
BreAStDuration
RoadToHealthCard
BirthRegistration
ChildGrantFormsChildGrantFormID
Household
ObsYear
Respondent
ChildGrantFilter
Newspaper
Radio
TV
Welfare
ClinicsHospital
FriendsRelative
PensionQueue
CRW
HeardOther
HeardOtherSpec
AdultsThisHouseReceived
ChildReceivedAdultOutOfHouse
DidNotKnow
DidNotWant
NotEligibleIncome
NoEligibleChild
CannotAccess
NoRSAID
NoRTHCard
NoBirthCert
NoDeathCert
LackOther
LackOtherSpec
Comments
ChildGrantsChildGrantsID
ChildGrantFormID
AdultID
AdultInDSS
ChildID
ChildInDSS
ChildAge
ChildSex
ChildRelToAdult
GrantType
Disability
NotProcessed
NotProcessdSpec
PaymentStopped
PaymentStoppedSpec
FirstVisit
FirstVisitYear
FirstVisitMonth
SubmittedNoSlip
SubmittedNoSlipYear
SubmittedNoSlipMonth
SubmittedSlip
SubmittedSlipYear
SubmittedSlipMonth
AwardedNoPay
AwardedNoPayYear
AwardedNoPayMonth
Received
ReceivedYear
ReceivedMonth
PayStopped
PayStoppedYear
PayStoppedMonth
Refused
RefusedYear
RefusedMonth
Comments
CoughStatusId
ObsYear
Cough
DeathsResidence
DeathLocal
DiedAt
Hospital
Province
WomanDeath
DeathRegistration
MainCause
EducationStatusID
Education
ObsYear
HealthCareUtilizationID
CardColour
BCG
Polio
DTP
Hib
HepB
MeASles
Immunized
AdmittedLASt12Months
Admissions
Hospitals
ObsYear
HouseholdsHousehold
Location
IndividualsID
Gender
DoB
DoBEstimated
MotherID
MotherStatus
FatherID
Refugee
DoD
DoDEstimated
InMigrationsResidence
InternalMigration
MoveType
MovePlace
Village
Reason
LocationsLocation
Village
MarriageAttUnionID
Q119LobolaPaid
Q120ReligiousCeremony
Q121MarriageRegistered
MaternityHistoryID
DoB
Gender
Outcome
Multiple
Living
ChildStatus
DeathAge
DeathAgeType
DeathDate
MembershipsID
Membership
Household
StartDate
StartObsDate
EndDate
EndObsDate
HHRelation
OutMigrationsResidence
InternalMigration
MoveType
MovePlace
Village
Reason
PregnanciesID
Pregnancy
AntenatalClinic
AntenatalVisits
Education
Scholar
BacktoSchool
PregnancyPlanned
ContraceBe
ContraceAf
DeliveryDate
DateEstimated
DeliveryLocal
DeliveryPlace
Hospital
Attendant
Complication
ComplSpecified
Outcome
StillBorn
LiveBorn
Duration
ResidencesResidence
Location
ID
StartDate
StartObsDate
EndDate
LAStObsDate
ResidentStatusID
ResMonths
ResStatus
ObsYear
SampleVillHHResultsItem
Value
Village
Tries
StrokeStatusID
ObsYear
Weakness
Stroke
UnionEpisodesUnionID
RespondentID
PartnerID
StartDate
InitiatingUEventType
StartObsDate
EndDate
LAStObsDate
TerminatingUEventType
Q122CohabitationStatus
VillagesVillage
LabourStatusId
ObsYear
EverWorked
CurrentlyWorking
Unemployment
UnemploymentSpec
Unemwf
PensionStudent
PWorkType
PWorkCat
PSector
PEmployer
PPeriod
PPlaceName
PPlaceCode
PTax
SWorkType
SWorkCat
SSector
SEmployer
SPeriod
SPlaceName
SPlaceCode
STax
Pop PyramidsPop PyramidsAgincourt Population Pyramid 1992
5,000 4,000 3,000 2,000 1,000 0 1,000 2,000 3,000 4,000 5,000
0-4
10-14
20-24
30-34
40-44
50-54
60-64
70-74
80-84
90-94
>100
Age
Gro
up
Males Females
Permanent ResidentsMigrant Residents
Agincourt Population Pyramid 2005
5,000 4,000 3,000 2,000 1,000 0 1,000 2,000 3,000 4,000
0-4
10-14
20-24
30-34
40-44
50-54
60-64
70-74
80-84
90-94
>100
Age
Gro
up
Males Females
Permanent Residents
Migrant Residents
Trend and Age Profile of MortalityTrend and Age Profile of MortalityA. Trends in Expectation of Life at BirthA. Trends in Expectation of Life at Birth B. AgeB. Age--Specific Mortality Rates, 1992Specific Mortality Rates, 1992--20052005
50.0
52.5
55.0
57.5
60.0
62.5
65.0
67.5
70.0
72.5
75.0
77.5
80.0
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Year
e 0 (Y
ears
)
Female
Male
Male fit (4th Poly.)
Female fit (4th Poly.) 0
20
40
60
80
100
120
140
0 1-4 5-910
-1415
-1920
-2425
-2930
-3435
-3940
-4445
-4950
-5455
-5960
-6465
-6970
-7475
-7980
-84 85+
Age (Years)
Mor
talit
y R
ate
(per
1,0
00)
FemaleMale
Figure 2: Trend and Age Profile of Mortality. Trends in ExpectatFigure 2: Trend and Age Profile of Mortality. Trends in Expectation of Life at Birth B. Ageion of Life at Birth B. Age--Specific Specific Mortality Rates, 1992Mortality Rates, 1992--2005Over the study period mortality has worsened significantly, 2005Over the study period mortality has worsened significantly, Figure 2 Figure 2 panel A. For women expectation of life at birth has fallen frompanel A. For women expectation of life at birth has fallen from something around 70 to about 60 something around 70 to about 60 years between the mid 1990s and 2005. For men the drop was fromyears between the mid 1990s and 2005. For men the drop was from about 65 to 52 years. Box about 65 to 52 years. Box sexes lost about 12 years of life. Almost certainly the mortalisexes lost about 12 years of life. Almost certainly the mortality of women (and probably men ty of women (and probably men too) in the early period of the study us underestimated resultintoo) in the early period of the study us underestimated resulting from errors in data management g from errors in data management and recording of death, in particular data for 1992 and 1993 shoand recording of death, in particular data for 1992 and 1993 should not be trusted. uld not be trusted. The age profile of mortality between 1992 and 2005 is displayed The age profile of mortality between 1992 and 2005 is displayed in Figure 2 panel B. Child in Figure 2 panel B. Child mortality is comparatively low with little sex differential. Thmortality is comparatively low with little sex differential. The risk of dying from childhood e risk of dying from childhood through roughly age 25 is roughly equal for women and men, and tthrough roughly age 25 is roughly equal for women and men, and then the mortality for men hen the mortality for men rises sharply while the mortality of women plateaus. At all agerises sharply while the mortality of women plateaus. At all ages older than 25 male mortality s older than 25 male mortality exceeds female by a exceeds female by a
Figure 1: Trend and Age Profile of Fertility A. Trends in TFR, CFigure 1: Trend and Age Profile of Fertility A. Trends in TFR, CBR and CDRB. BR and CDRB. AgeAge--Specific Fertility Rates, 1992Specific Fertility Rates, 1992--2005 Over the study period fertility has 2005 Over the study period fertility has fallen by about one child per woman from a TFR of roughly 4 to lfallen by about one child per woman from a TFR of roughly 4 to less than 3, ess than 3, Figure 1 panel A. The crude birth rate has similarly fallen froFigure 1 panel A. The crude birth rate has similarly fallen from about 32 to m about 32 to 25 per thousand. These overall reductions are reflected at all 25 per thousand. These overall reductions are reflected at all ages, Figure ages, Figure 1, panel B. 1, panel B.
2.50
2.75
3.00
3.25
3.50
3.75
4.00
4.25
4.50
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Year
TFR
0
5
10
15
20
25
30
35
CB
R a
nd C
DR
(per
1,0
00)
TFRCBRCDRCBR fit (4th Poly.)CDR fit (2nd Poly.)TFR fit (4th Poly.)
0
20
40
60
80
100
120
140
160
15-19 20-24 25-29 30-34 35-39 40-44 45-49
Age (Years)
ASF
R p
er 1
,000
1995200592-05
B. Age-Specific Fertility Rates, 1992-2005A. Trends in TFR, CBR and CDR
Trend and Age Profile of Fertility