Detection of unusual events

1
Detection of unusual events

• Integrated Disease Surveillance Programme (IDSP)
  district surveillance officers (DSO) course

2
Preliminary questions to the group

• Have you ever detected an outbreak?
• If yes, what difficulties did you face?
• What would you like to learn about early outbreak
  detection?

3
Outline of the session

1. From denial to vigilance
2. Early warning signals
3. Triggers in the Indian Integrated Disease
   Surveillance Programme (IDSP)

4
Cases of cholera by date of onset, South 24
Parganas, West Bengal, India, May 2006
When to investigate?
30
25
20
Never?
Number of cases
15
10
5
0
24
25
26
27
28
29
30
1
2
3
4
5
April
May
5
When are outbreaks investigated in your district?

• When the first cases occur?
• An indication of vigilance
• When there is a large cluster?
• Shows reactivity
• When most cases have occurred?
• The team only acts under pressure
• Never
• Denotes denial

Fighting denial
6
Reasons why public health professionals might
deny outbreaks

• No skills to investigate
• Additional work
• No intervention to offer
• Press pressure
• Fears of sanctions

Fighting denial
7
Turn challenges into opportunity

• Investigation methods can be learned
• Investigations become easier with practice
• Data guides effective prevention measures
• Good investigations impress the press
• Leaders appreciate signs that the situation is
  under control (e.g., ongoing investigation)

Fighting denial
8
Improving the outbreak detection situation

• The number one obstacle to fight is denial
• Once there is a willingness to address outbreaks,
  technical methods may be used to ensure the
  earliest possible detection
• The system will not go from denial to early
  detection overnight
• But every small progress matters
• It is never too late to investigate an outbreak
• You will still learn, but it is more difficult

Fighting denial
9
Two ways to detect unusual events

• Event-based surveillance
• Reports of events
• Case-based surveillance
• Routine surveillance data analysis

Early warning
10
Components of early warning surveillance
Surveillance
Case-based surveillance
Event-based surveillance
Data
Reports
Alert
Response
Assess
Public health alert
InvestigateControl measures
Post-outbreak strengthening
Evaluate
Early warning
11
Two ways to detect unusual events

• Event based surveillance
• Collect reports
• (e.g., toll free line, internet search, press
  scan)
• Filter
• Identify signals
• Verify
• Case base surveillance

Early warning
12
Early warning signals for event-based
surveillance

• Clustering of cases or deaths
• Single case of disease of epidemic potential
• Acute febrile illness of unknown etiology
• Two or more linked cases of meningitis, measles
• Unusual isolate
• Shifting in age distribution of cases
• High vector density
• Natural disasters

Early warning
13
Sources of information for event-based
surveillance

• Rumour register in each institution
• Standardized format
• Investigation for each entry
• Tracking system to document follow up
• Community informants
• Private and public sector
• Media
• Important source of information, not to neglect
• 24 x 7 call centre

Early warning
14
Two ways to detect unusual events

• Event based surveillance
• Case base surveillance
• Collect data
• Analyze data
• Detect signals
• Validate signals

Early warning
15
Challenges to surveillance data analysis to
detect unusual events

• Imperfect data
• Changes over time
• Multiple sources of information
• Problem of quality and completeness
• Need an in-depth knowledge of the system
• Evaluation
• To know what is unusual, you need to know what is
  usual

Early warning
16
Sources of false alarm

• Changes in the numerator
• Increased awareness about a disease
• Appointment of a new surveillance person
• Acute reporting of old, chronic, cases
• Changes in the denominator
• Population movements

Early warning
17
Triggers in the context of the Indian Integrated
Disease Surveillance Programme (IDSP)

• Threshold for diseases under surveillance that
  trigger pre-determined actions at various levels
• Based upon the number of cases in weekly report
• Trigger levels depend on
• Type of disease
• Case fatality (Death / case ratio)
• Number of evolving cases
• Usual trend in the region

Triggers
18
Levels of response to different triggers
Trigger Significance Levels of response
1 Suspected /limited outbreak Local response by health worker and medical officer
2 Outbreak Local and district response by district surveillance officer and rapid response team
3 Confirmed outbreak Local, district and state
4 Wide spread epidemic State level response
5 Disaster response Local, district, state and centre
Triggers
19
Malaria triggers

• Trigger 1
• Single case of smear positive in an area where
  malaria was not present for a minimum of three
  months
• Slide positivity rate doubling over last three
  months
• Single death from clinically /microscopically
  proven malaria
• Single falciparum case of indigenous origin in a
  free region
• Trigger 2
• Two fold rise in malaria in the region over last
  3 months
• More than five cases of falciparum of indigenous
  origin

Triggers
State may set their own triggers
20
Cholera triggers

• Trigger 1
• A single case of cholera / epidemiologically
  linked cases of diarrhea
• A case of severe dehydration / death due to
  diarrhea in a patient of gt5 years of age
• Clustering of cases in a particular village /
  urban ward where more than 10 houses have at
  least one case of loose stools irrespective of
  age per 1000 population
• Trigger 2
• More than 20 cases of diarrhea in a
  village/geographical area of 1000 population

Triggers
21
Typhoid fever triggers

• Trigger 1
• More than 30 cases in a week from the entire
  primary health centre area
• 5 or more cases per week from one sub-centre of
  5,000 population
• More than 2 cases from a single village/urban
  ward/1000 population
• Clustering of cases of fever
• Trigger 2
• More than 60 cases from a primary health centre
  or more than 10 cases from a sub-center

Triggers
22
Polio trigger

• One single case

Triggers
23
Plague triggers

• Trigger 1
• Rat fall
• Trigger 2
• At least 1 probable case of plague in community

Triggers
24
Japanese encephalitis triggers

• Trigger 1
• Clustering of two or more similar case from a
  locality in one week
• Trigger 2
• More than four cases from a PHC (30,000
  population) in one week

Triggers
25
Dengue triggers

• Trigger 1
• Clustering of two similar case of probable Dengue
  fever in a village
• Single case of Dengue hemorrhagic fever
• Trigger 2
• More than four cases of Dengue fever in a village
  with population of about 1000

Triggers
26
Triggers for syndromic surveillance

• Fever
• More than 2 similar case in the village (1000
  Population)
• Diarrhea
• See cholera
• Acute flaccid paralysis
• 1 case
• Jaundice
• More than two cases of jaundice in different
  houses irrespective of age in a village or 1000
  population

Triggers
27
Basic responses to triggers

• There are triggers for each condition under
  surveillance
• Various trigger levels may lead to local or
  broader response
• Tables in the operation manual propose
  standardized actions to take following various
  triggers
• Investigations are needed in addition to
  standardized actions

Triggers
28
Progressive response

• Levels of alert are progressively increasing
• Unusual signals require filtering / validation
• The best chance of detection is to
• Analyze regularly
• Be familiar with the time, place and person
  characteristics of the diseases in your area

Triggers
29
Public health events of international concern
need to be reported as per new International
Health Regulations (IHRs)

• Public health events
• Sudden serious and unexpected event that require
  immediate action
• Outbreak of epidemic prone diseases
• Any other event that may have impact on the
  health of a community (natural or man-made
  catastrophes)
• International concern
• Risk of spread beyond the borders of the affected
  country (agent/host/environment or capacity to
  contain the event)

Triggers
30
Take home messages

• Fight denial by showing what can be usefully done
  about outbreaks
• Investigations
• Control measures based on evidence
• Consider case-based and event-based surveillance
• Recognize triggers and respond to them as per
  guidelines

31
Additional reading

• Section 4 of IDSP operations manual
• Module 8 of training manual

Triggers