Key elements in a surveillance system Patrick Rolland EPIET / EUPHEM Intro Course 2012 &

Key elements ina surveillance system

Patrick Rolland

EPIET / EUPHEM Intro Course 2012

&

2

&

Le fil rouge in surveillance?

Surveillance isInformation for

action!

3

&

Good reminder: surveillance?

Langmuir AD., 1963

“Systematic collection, consolidation, analysis and dissemination of data on specific disease”

Thacker SB., 1996

“The final link is the application of these data to prevention and control”

4

&

Good reminder: surveillance loop

Data collection

Data analysis

Information

Action!

Evaluation

Objectives

5

&

Aim and content of the lecture

Aim: To understand key elements in a surveillance system

- From a clear understanding of the purpose

- To the right dissemination of information

Content- Main objectives of a surveillance system

- Checklist of key elements, step by step

6

&

Main objectives ofa surveillance system

7

&

Main objectives

Describe: produce information in terms of TPP

Alert: detect epidemics or emerging events

Evaluate: assess prevention or control measures

And also: Generate hypotheses for research

Detect changes in health practice

Plan public health actions and resources

8

&

But keep in mind the goal!

Surveillance isInformation for action!

By implementation of prevention and control measures

In order to reduce morbidityand mortality

9

&

Describe: Time = Graph

Source: GP network “Réseau unifié”

Week

Inci

denc

e (p

er 1

00,0

00)

Weekly incidence rate of flu-likesyndromes, 2009-2012, France

10

&

Describe: Place = Map

Weekly incidence rateof acute diarrhea(per 100,000)

1 to 7 march 2012, France

Source: GP network “Réseau sentinelles”

11

&

Describe: Person = Table

Severe cases of influenza, winter 2011-2012, France

Source: Intensive care units Network

Characteristicsof persons

Burden

12

&

Alert: detect an epidemic

Exceedence

Week

Weekly incidence of flu-like syndromes, 2007-2012, FranceSource: GP network “Réseau sentinelles”

Inci

denc

e (p

er 1

00,0

00)

IncidenceThreshold

13

&

Emergence of W135

0

50

100

150

200

250

300

350

400

450

85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06

années

No

mb

re d

e c

as d

écla

rés

B C W 135 Y

Alert: detect an emerging phenomena

Invasive meningococcal infectionsby serogroup B, C, W135 et Y, 1985-2006, France

Source: InVS, NRC for meningococci

Year

Num

ber o

f cas

es B C W135 Y

14

&

0

20

40

60

80

100

120

140

160

180

200

1940 1950 1960 1970 1980 1990

Année

No

tifi

cati

on

s (x

100

0)C

ou

verture (%)

Evaluate: prevention/control measure

Cases of Pertussis, England and Wales, 1940-1999

Num

ber o

f cas

es (x

100

0)

Year

0

80

40

Vacc

inat

e co

vera

ge (%

)

Vaccinationimplementation

15

&

Evaluate: prevention/control measure

Source: Notifiable disease, InVS Cases of legionellosis, 1988-2011, France

Reinforced surveillanceNew investigation guide

Urine antigen testing

Num

ber o

f cas

es

Year

Inci

denc

e (p

er 1

00,0

00)

No casesIncidence

16

&

Key elements in asurveillance system

17

&

Key elements

Justification of implementation Objectives of the system Types of system Case definition Population under surveillance Data providers and data sources Data elements and data collecting tools Flow chart and data transmission Data validation and data analysis Dissemination and communication data Human and financial resources Data security and confidentiality Evaluation of system

18

&

Key elements


19

&

Justification of implementation

Ask yourself two main questions:

- Should this event be under surveillance?• High frequency? • High severity?• High socioeconomic cost?

- Do some prevention or control measures exist?

And these two additional questions:

- Do we have existing data that already answer?

- Do we have resources to set up a new system?

20

&

Health events under surveillance

Infectious diseases- Notifiable diseases- Health-care associated infections (HCAI)- Vaccine-preventable diseases- Food and waterborne diseases- Sexually transmitted infections (STI)- Zoonoses- Vector-borne diseases

But also?

21

&

Health events under surveillance

Chronic diseases and injuries: cancers, accidents, traumas, cardiovascular diseases, etc.

Occupational health: cancers, musculoskeletal disorders, respiratory diseases, mental health, etc.

Environmental hazards: air pollution, ionizing radiations, heat/cool waves, water/soil pollution, etc.

22

&

Key elements


23

&

Objectives of the system

Primary objective: To describe a health-related event in population-based or in a specific population

With two possible aims (sometimes both):

To alert by early detection of epidemics or emerging pathologies that need timely action

To evaluate the efficiency of prevention orcontrol measures

24

&

Key elements


25

&

Passive and active system

Consideration useful for discussion

Passive: “The data come to you” Data collection based on existing information Simple, not burdensome but could be

incomplete E.g.: notifiable diseases, deaths, emergency data

Active: “You go towards the data” Data collection specially set up Good quality of data but required resources E.g. : injuries, non-notifiable infectious diseases, etc.

26

&

E.g. of passive system

0,0%

0,5%

1,0%

1,5%

2,0%

2,5%

3,0%

3,5%

4,0%

4,5%

S16

S19

S22

S25

S28

S31

S34

S37

S40

S43

S46

S49

S52

S02

S05

S08

S11

S14

2011-2012 2010-2011

% of gastroenteritis among emergency consultations, seasons 2010-2011 and 2011-2012, Aquitaine, France

Prop

ortio

n of

cas

es

Week

27

&

E.g. of active system

Cases of envenomation by Physalia Physalis reported bylifeguards, Aquitaine Atlantic Coast, Summer 2011, France

0

10

20

30

40

50

60

70

80

90

100

1-ju

in //

30-ju

in

4-ju

il.

8-ju

il.

12-ju

il.

16-ju

il.

20-ju

il.

24-ju

il.

28-ju

il.

1-ao

ût

5-ao

ût

9-ao

ût

13-a

oût

17-a

oût

21-a

oût

25-a

oût

29-a

oût

2-se

pt.

6-se

pt.

10-s

ept.

14-s

ept.

18-s

ept.

22-s

ept.

26-s

ept.

30-s

ept.

Période de surveillance

No

mb

re d

e ca

s d

'en

ven

imat

ion

sN

umbe

r of c

ases

Day

28

&

Exhaustive vs. sentinel system

Important consideration for data analysis

Severe diseases or low-frequency diseases requiring timely action

Exhaustive system (= all providers)E.g.: cancers registries, notifiable diseases

High-frequency diseases or less-severe diseases

Sentinel system (= selected providers)E.g.: seasonal flu, occupational diseases (except cancers)

29

&

E.g. of exhaustive system

Cancers among women, Year 2005, Gironde, FranceSource: Cancer

registry of Gironde

Number of cases

1. Breast cancer

2. Colorectal cancer

3. Skin cancer

Can

cer c

ause

s

30

&

E.g. of sentinel system

Prevalence of occupational diseases (except cancers),Year 2010, Region of Aquitaine, France

Male (%) Female (%) Total (%)

Musculoskeletal disorders

3.4 3.5 3.4

Psychic suffering 1.0 2.1 1.5

Skin rashes and allergies 0.3 0.4 0.3

Hearing disorders 0.4 / 0.2

Total 5.2 6.0 5.6

Source: Sentinel Network of occupational physicians (n=92)

31

&

Case vs. syndromic system

Case system (traditional system)

Targets a defined health-related event

E.g.: mesothelioma, Lyme disease, diabetes, etc..

Syndromic system (“before diagnosis”)

For early detection, evaluation of event impact

Based on existing activity data, real-time collection, analysis and interpretation data

E.g.: emergency services, virology labs, deaths certificates, medicine sales, absence in schools, etc.

32

&

E.g. of case system

Occupations at risk for mesothelioma, FranceSource: Program of Mesothelioma Surveillance (1998-2012), InVS

Occ

upat

ions

Odds-ratio (95% CI)

33

&

E.g. of syndromic system

0,0%

0,5%

1,0%

1,5%

2,0%

2,5%

3,0%

3,5%

4,0%

4,5%

S16

S19

S22

S25

S28

S31

S34

S37

S40

S43

S46

S49

S52

S02

S05

S08

S11

S14

2011-2012 2010-2011

% of gastroenteritis among emergency consultations, seasons 2010-2011 and 2011-2012, Aquitaine, France

Prop

ortio

n of

cas

es

Week

34

&

Key elements


35

&

Case definition

The “heart” of the system!

Must be clear and simple

Based on criteria:clinical, biological, epidemiological

May include:- Classification (possible, probable, confirmed)

- TPP (Time-Place-Person) information

36

&

Source:

Case definitions of notifiable diseases

Commission Decision 2008/426/EC – 28-IV-2008

E.g. for measles definition as notifiable disease

Clinical criteria

Laboratory criteria

Epidemiological criteria

Case classification

37

&

Key elements

Justification of implementation Objectives of the system Types of system Case definition Population under surveillance Data providers and data sources Data elements and data collecting tools Flow chart of data and data transmission Data validation and data analysis Dissemination and communication data Human and financial resources Data security and confidentiality Evaluation of system

38

&

Population under surveillance

Depends on characteristics of health-event

E.g.: Hemolytic Uremic Syndrome (HUS)

- Rare disease that predominantly affects children, needs timely action in outbreak

Population under surveillance (France): children (< 15 years) hosted in pediatric and nephrology hospital services (N=31)

39

&

Key elements


40

&

Data providers and data sources

Data providers: health professionals, laboratories, health insurance funds, civil status offices, etc.

Data sources

Administrative: death certificates, hospital systems, etc.

Medical: patients folders, notifiable diseases, etc.

Biological: virology, bacteriology, toxicology samples

Environmental: air pollution sensors, individual ionizing radiations card for exposed workers, etc.

41

&

E.g. data provider: NRC (France)

Anaérobies et Botulisme (LA)

Arbovirus (LA)

ATNC

Brucella

Borrelia (LA)

Campylobacter & Helicobacter

Charbon (LA)

Chlamydiae

Cytomégalovirus

Enterovirus

Escherichia coli & Shigella (LA)

Francisella tularensis

Gonocoques

Haemophilus influenzae

Legionella

Leishmania

Mycobactéries et résistance des mycobactéries

Paludisme (2 co-responsables)

Pneumocoques

Résistance aux antibiotiques (LA)

Rickettsia, Coxellia & Bartonella Virus de la rougeole

Staphylocoques VIH

Streptocoques Virus Influenza

Syphilis Virus entériques

Trichinella Virus des hépatites A et E

Toxoplasmose Virus des hépatites B, C et Delta

Arbovirus

Arbovirus & influenza virus en AG

Charbon

Borrelia

Anaérobies et Botulisme

Chimiorésistance du paludisme

en Antilles Guyane

Coqueluche et autres bordetelloses

Corynebactéries toxinogènes

Escherichia coli & Shigella

Fièvres hémorragiques virales

Leptospires

Listeria

Méningocoques

Mycologie et antifongiques

Peste et autres yersinioses

Rage

Salmonella

Résistance aux antibiotiques

Streptocoques (LA)

Virus des Hépatites B & C (LA)

Vibrions et cholera

Virus Influenza

32 NRC and 31 AL

Within hospitals, universities, other research institutes

15 NRCand 3 AL

Pasteur Institute(Paris)

47 National Reference Centers and 34 Associated Laboratories

One NRC perinfectious disease

42

&

E.g. data source: ND (France)

• Botulisme• Brucellose • Chikungunya• Dengue• Fièvres typhoïdes et paratyphoïdes• Hépatite aiguë A• Infections invasives à méningocoques• Légionellose• Listériose• Rougeole• Saturnisme de l’enfant mineur• MCJ et ESST• Toxi-infection alimentaire collective• Tuberculose

• Choléra (RSI)• Diphtérie• Fièvres hémorragiques africaines• Fièvre jaune (RSI)• Paludisme autochtone et paludisme importés

dans DOM• Peste (RSI)• Poliomyélite• Rage• Typhus exanthématique

• Charbon, Tularémie, Variole

• Infection à VIH quel qu’en soit le stade• Hépatite aiguë B• Tétanos

Bioterrorism-related ND (n=3)

Frequent ND (n=14) Infrequent ND (n=10)

27 notifiable diseases (ND) require surveillance and timely action

• Mésothéliome(entrée en 2012)

4 ND require surveillance only

43

&

Key elements

Justification of implementation Objectives of the system Types of system Case definition Population under surveillance Data providers and data sources Data elements and data collecting tools Flow chart of data and data transmission Data validation and data analysis Dissemination and communication data Human and financial resources Data security and confidentiality Evaluation of system

44

&

E.g. French formfor mesotheliomanotiable disease

Anonymous data on patient diagnosis (by clinician&pathologist)

Nominative data on patient and provider

Dates and signatures of provider and sanitary authority

45

&

Key elements


46

&

E.g. of flow chart: ND (France)

Patient

Health professionals, laboratories, etc.

ReferenceCenters

Experts NetworksClinicians/Pathologists

PartnersNational/International

Public Health High Council

Politic of Health

Alert

Dec

lara

tio

n

Ministry of Health

HealthRegional Agencies

Alert

47

&

Data transmission

Reliable and fast Electronic: e-mails, websites And: phone, fax and mail

Low frequency Daily Weekly Monthly

Secure Regularity, punctuality, exhaustiveness And don’t forget the “Zero reporting”

48

&

Key elements


49

&

Data validation and data analysis

Data validation Missing data Duplicates Data quality

Data analysis Indicators: frequency, severity Methods: descriptive, analytical Illustrations: tables, graphs and maps Keep in mind: Time-Place-Person!

50

&

Daily deaths, summers 1999-2002 vs. 2003, France

E.g. of data analysis: time series

0

50

100

150

200

250

300

350

25-juin 30-juin 05-juil 10-juil 15-juil 20-juil 25-juil 30-juil 04-août 09-août 14-août 19-août

Décès journ

alie

rs

0

5

10

15

20

25

30

35

Tem

péra

ture

(°C

)

Décès 1999_2002 Décès 2003 T moy 1999_2002 T moy 2003

Num

ber o

f dea

ths

Day (25 June to 19 august)

Source: Syndromic Program, InVS

Tem

pera

ture

(°C

)

Temperature

Deaths

1999-2002

2003

1999-2002

2003

51

&

Key elements


52

&

Dissemination/communication data

Systematic and regular For stakeholders, decisions makers For other professionals, public, medias

Adapted To “Those who need to know” To the urgency of the situation

Format Report, article, flash news, press release etc. But also: phone, e-mail, fax, mail

Factor of motivation/improvement

53

&

E.g. report at local level

“Point épidémio”

France, Bordeaux(InVS, south west)

Weekly report on health status of population

Topics:

OutbreaksSeasonal pathologiesNotifiable diseasesVirology surveillanceDeaths

54

&

E.g. report at international level

WHO (Switzerland)

CDC (USA)

55

&

E.g. press release

At international level (e.g. Europe)

At local level(e.g. France, Bordeaux)

56

&

Key elements


57

&

Human and financial resources

From the budget to the expenses Salaries, travels, equipments, furniture, etc. Useful to negotiate resources, check the

workplan, assess the cost of the system

58

&

Key elements


59

&

Data security and confidentiality

“Critical point” in surveillance system

Public heath data

Often personal and private in nature Might directly identified individuals

Take into account policiesand protection forpersonal data

Data collectionoften under laws

60

&

Key elements


61

&

Aims of evaluation

Ask yourself the fundamental question:

Have the objectives been met?

And these additional questions:- Did the system generate needed answers?

- Was the information timely?

- Was it useful for decisions makers?

- How was the information used?

- What could be done to improve the “attributes”?

- Is it useful to continue the surveillance?

62

&

Simplicity: refers to structure and ease operation

Flexibility: ability to adapt to changing needs

Data quality: refers to data completeness&validity

Acceptability: willingness of persons and organizations to participate

Sensitivity: refers to proportion of detectedcases by the system (“A / A + C”)

The “9” attributes of a system (1)

63

&

PVP: refers to proportion of reported the cases that actually have the event (“A / A + B”)

Representativeness: ability to describe the event over time and its distribution in the population

Timeliness: reflects the speed between system steps of the system

Stability: refers to reliability and availability

The “9” attributes of a system (2)

64

&

Conclusion

65

&

Main messages

Justification- Relevance of the event to be under surveillance- Existence of prevention and control measures

Objectives and aims To describe – To alert – To evaluate To produce information for action In order to reduce morbidity and mortality

Requirements- “Good data”, involvement of stakeholders- Regular evaluation of the system

66

&

One useful book

3rd edition

September 2010

Lisa M. Leeet al.

67

&

Don’t forget the surveillance loop!

Data collection

Data analysis

Information

Action!

Evaluation

Objectives

68

&

Le fil rouge in surveillance?

Surveillance isInformation for

action!

Key elements in a surveillance system Patrick Rolland EPIET / EUPHEM Intro Course 2012 &

Documents

data sources data elements

dissemination of data

surveillance system

surveillance data providers

existing data

france slide

data analysis dissemination

control slide