Title: Experiences in the Use of NonTraditional Data Sources in a Rural State South Carolina
1Experiences in the Use of Non-Traditional Data
Sources in a Rural State South Carolina
- Dan Drociuk, MT(ASCP), MSPH Director
- Epidemiological Response / Enhanced Surveillance
Section - Division of Acute Disease Epidemiology
- South Carolina Department of Health and
Environmental Control
2Outline
- Part I
- The People
- How we are organized to develop, implement,
monitor and respond to EED system needs. - Part II
- The Push
- How we disseminate information about EED systems
and general epidemiological activities. - Part III
- The Products
- Case studies showing the interaction and
intersection of EED systems currently in use in
South Carolina.
3Part IThe People
4A bit about South Carolina
- Approximate population 4,350,500 (sans golf
courses and beaches) - Three MSAs in the Top 100
- Columbia, Charleston, Rock Hill/Charlotte
- Tourism a main industry along the costal areas
- Centralized health department structure
- All public health employees are state employees
regardless of location - No local boards of health
5Current Systems in Use
- Palmetto Poison Center (electronic daily feeds to
public health), - National Retail Data Monitoring (NRDM) for OTC
sales (using home-grown C1, C2, C3 indicators), - BioSense for DoD and VA ambulatory care
procedures and ambulatory care diagnosis, - Essence data monitored and crossed with
Biosense alerts and indicators. - Sentinel providers with Influenza-like Illness
reporting (number/wk). - Two (2) pilot hospitals providing chief-complaint
data in home-grown categories.
6(No Transcript)
7(No Transcript)
8(No Transcript)
9Regional Epi Teams
- Services are delivered and surveillance happens
at the local/community level - Surveillance, case and outbreak investigation
response starts with the Regional Epi Team - 6-20 members
- Mix of skills nurses, Env. Health, PIO
- Established 1995 to provide trained integrated
team-based response to routine surveillance, case
investigations, and outbreak investigations - Available for Urgent Reports 24/7 via on-call
rotation.
10Epidemiological Responses to Events of Public
Health Significance
- White powder events
- Ricin in a U.S. Postal Facility in Greenville
- Chlorine tanker train derailment, Graniteville,
SC - Numerous point-source outbreak investigations
- Food-borne (Salmonellas),
- Water-borne (Legionella cluster, rash-illness),
- Respiratory (Pertussis)
- Hurricane Katrina Evacuation Center surveillance
activities, Greenville, SC
11Part IIThe Push
12Dissemination of Epi-related information
- Daily
- Provided to the South Carolina Intelligence
Fusion Center - Reports from our Division of Acute Disease
Epidemiology on-call staff for overnight calls
received, - Summary information from our Early Event
Detection Systems (BioSense, OTC sales, Palmetto
Poison Center) - Previous day on-call events of public health
significance - Weekly
- Provided via the Epidemiological Weekly Report
(EWR) - Summary information from our Regional and Central
Office - Monthly
- Meetings with Regional Epidemiology Response
Staff - As Needed
- Distribution via the SC Health Alert Network
- CDC and DHEC Health Alerts/Advisories/Updates
13Example of a Daily SCIEx report
14Example of a weekly EWR
15EWR Epidemiological Weekly Report
- The EWR is compiled weekly from Regional and DADE
reports. - It includes basic summary information related to
an ongoing or concluding epi-investigations. - Submissions usually include what is known or
hypothesized about etiology, geographic location,
and numbers of persons affected.
16EWR Epidemiological Weekly Report
- Circulation
- The EWR is submitted to the Deputy Commissioner
for Health Services, members of the Senior
Leadership Team, Central and Regional Office
Leadership, the Office of Public Health
Preparedness, etc. - Goal
- Provide situational awareness regarding Epi
activities across the state.
17EWR Epidemiological Weekly Report
- Statistics
- In 2005, EWRs documented 288 new
investigations of potential outbreaks or single
cases of disease or events of public health
significance. - In 2005 documented outbreaks totaled 62
stratified as follows (by modes of transmission) - Respiratory 20 (32)
- Foodborne 16 (26)
- Unknown 11 (18)
- Person-to-person 7 (11)
- Fecal-oral 5 ( 8)
- Contact 2 ( 3)
- Nosocomial 1 ( 2)
182004-2005 SC Foodborne Outbreak Statistics
19Part IIIThe Products
20Case Study 1Chlorine Release in Graniteville,
South Carolina
21(No Transcript)
22What Happened?
- At 239 AM, a 42-car Norfolk and Southern train
derailed - A tanker car containing chlorine was punctured,
releasing approximately 60 tons of chlorine - The chlorine was released in the immediate
vicinity of Avondale Mills, the commercial
district, and residential areas of Graniteville SC
23(No Transcript)
24Objectives of a Rapid Epidemiological Assessement
- Assess
- Extent of exposure
- Morbidity
- Health services used
- Identify persons at risk for long-term sequelae
- Risk factors for severe outcomes
- Location of exposure
25Case Definition
- Person treated for symptoms or complications from
chlorine exposure - Identified by
- Hospital emergency department logs
- Physician reporting
26Emergency Department Visits within 24 Hours
ED Visits, N272 ()
27Signs or Symptoms Reported
28280
280
260
260
100
100
Number of Events
Number of Events
80
80
60
60
40
40
20
20
0
0
1
3
5
7
9
11
13
15
17
19
21
23
25
Day(s) Since Accident
Day(s) Since Accident
29Hospitals that Treated Patients for Chlorine
Exposure
January 6-7, 2005
40
Barnwell
Edgefield
35
Lexington
St. Josephs
30
Doctors
25
MCG
Number of Patients
University
20
Aiken
15
10
5
0
2-3 am
4-5 am
6-7 am
8-9 am
2-3 pm
4-5 pm
6-7 pm
8-9 pm
12-1 pm
12-1 am
10-11 am
10-11 pm
Time
30(No Transcript)
31(No Transcript)
32Situational Awareness Issues
- Healthcare providers are going to treat people
first, and worry about the coding later - Review of charts showed a broad span of chief
complaints (n81), primary diagnosis (n48) and
ICD-9 codes (n51) used. - Data used for situational awareness, many are
going to be relying upon the data to make
decisions. This is a paradigm shift beyond
being simple early event detection system.
33Opportunities
- Use of Early Event Detection systems for
monitoring an emerging event must be flexible and
sensitive to be able to answer the questions
being asked, an EED must have the flexibility to
allow local users to create ad hoc syndrome
categories. - The wealth of information present via an
electronic medical record only tells part of the
story. The ability to identify the needs of the
people impacted greatly contributes to the
overall situational awareness of an emerging
event.
34Next Steps
- Within the 3-digit ZIP code area (ZIP3) 298 that
surrounds Graniteville, BioSense identified a
data anomaly based on the graphical visualization
in the Respiratory syndrome category that
corresponded to day 2 of the exposure. - However this anomaly was only apparent for 1 day
(1/7/05) and the source of this anomaly has not
been validated to determine both the source of
this anomaly and its relationship to chlorine
toxicity.
35Case Study 2Use of Poison Control Center data
36Palmetto Poison Center
- 1-800-222-1222
- Located in Columbia, SC
- Serves the entire state of South Carolina
- Staff consists of pharmacists
- and nurses
- Medical Director is trained in toxicology and
emergency medicine
37Palmetto Poison Center
- Services Provided
- 24 hours/7 days a week
- Phone service to provide poison treatment
information - For the public and health care professionals
- Information requests
- Medications
- Pesticides
- Plants
- Food poisoning
- Chemicals
38Statistics
- South Carolina
- 2005 Over 37,000 calls to the Center
- 74 of total calls involved human poison
exposures - 30 involved adults over 21yo
- 80 managed by Poison Center without further
medical evaluation needed
39Documentation
- Every incoming call documented
- Follow up calls included
- Record includes name, phone number and zip code
of caller - Patient data
- Age, gender weight
- Exposure substance, route and amount
- Reason for exposure
- Time of exposure
- Symptoms
- Treatment provided
- Outcome
40(No Transcript)
41(No Transcript)
42(No Transcript)
43State Surveillance - DHEC Notification
- Require notification of Public Health for
- Potential food poisoning/food tampering cases
- Occupational pesticide exposures
- Questionable exposure to current public health
concerns, i.e. anthrax (suspicious powder
calls), West Nile Virus, chlorine toxicity - Animal bites
44State Surveillance - DHEC Notification
- Currently faxing cases and providing downloads of
toxidromes every 24hrs into Toxitrak - Case information from Poison Center is included
in the upload - Reports can be generated
- Ex. How many cases had vomiting, diarrhea, fever
in Clarendon county in the last 4 weeks number
of suspected food poisoning cases from
restaurants per county, city or zip code
45The Epi of Poisonings in SC
- Who?
- 62 of calls involve children lt5
- Where and When?
- 92 of all accidental exposures occur in the home
- 0.8 at school
- Early evening at meal time most likely, followed
by late morning.
46The Epi of Poisonings in SC
- What?
- 38.8 Prescription and OTC medications, e.g.
analgesics, cough/cold - 9.0 Household cleaning substances
- 9.1 Cosmetics and personal care products
- 9.1 Plants
- 8.5 Insecticides, herbicides, rodenticides
- 4.5 Bites/envenomations
47Case Study 3Parris Island, South Carolina
Respiratory Sentinel Alerts
48Disease Surveillance at Marine Corps Recruit
Depot (MCRD), Parris Island
- Marine Corps Recruit Depot, Parris Island
- Marine Corps Air Station, Beaufort SC
- Naval Hospital, Beaufort SC
49Recruit Training Regiment
Within 1st, 2nd, and 3rd Battalions, there are
four companies. Each company contains an average
of six platoons with 60 to 80 recruits in each.
360 480 per company. 1440 1920 per
Battalion. Within 4th Battalion (350-400), there
are three companies, N, O, and P. Each company
contains an average of two platoons with 50 to 60
recruits in each. Fourth Battalion trains only
female recruits.
50(No Transcript)
51Beaufort Naval Hospital
- Provides general medical services to all
Active Duty Navy and Marine Corps Personnel, as
well as Retired military personnel and all
military dependents residing in the Beaufort
area, a total population of approximately 30,000
beneficiaries.
Officers 28 Enlisted 158 Civilians 22 Total
208
Hospital
Branch Health Clinic, MCAS
Branch Health Clinic, MCRD Parris Island
Director Clinical Support Services
1st BAS
Director Clinical Support Services
2nd BAS
3rd BAS
Preventive Medicine
4th BAS
52Methods of Disease Detection
- Traditional reportable disease surveillance
- State processes.
- Navy Disease Reporting System (NDRS)
- Sentinel Surveillance
- Naval Health Research Centers FRI surveillance
at recruit training centers - Astute clinicians
- Syndromic surveillance
- ESSENCE IV
5314 admissions to BMH in September and October
2005
- 2 from 1st Battalion
- Charlie company 2
- 7 from 2nd Battalion
- Foxtrot company - 3
- Gulf company 4
- 5 from 3rd Battalion
- Lima company - 2
- Kilo company 3
54Patients are admitted to BMH when a chest tube is
needed
55(No Transcript)
56(No Transcript)
57(No Transcript)
58(No Transcript)
59Why the rise?
Crowded living conditions
60Why the rise? We considered
- Crowed Living Conditions.
- Poor hygiene?
- Always a struggle, but the Marine leadership is
more cooperative than ever in this area. - Random interviews of recruits in the clinic
indicate that they have been properly instructed
re hand hygiene. - Recruits presenting too late?
- Interviews of the admitted recruits indicate that
they were encouraged vs. discouraged to seek
medical care early.
61Why the rise? We considered
- Something in the buildings?
- Mold and moisture is a problem in some buildings
aboard MCRD PI. Toxic mold was discovered in one
building where all recruits are massed together
for training. - Slow recognition of cases by providers?
- Some recruits admitted to BMH have been ill with
respiratory symptoms for 3-4 weeks prior to
admission, and they had been seen in medical
two or more times prior to admission. - Too high a turnover in medical staff to ensure
effective leadership and staff competency? - 3 SMOs in last 4 months
- Nearly 100 turnover of BAS staff in Sept-Oct.
62Lab Results
- Nine of the 17 admission to BMH have grown GABS
from pleural fluid, including 5 out of last 6
admitted in October. - Nothing atypical about resistance patterns
63Actions Taken
- Chemo prophylaxis with Bicillin provided to all
recruits in high incident Companies, to their
DIs and to their medical staff (BAS). - Presented data to MCRD Providers
64(No Transcript)
65Early Event DetectionOver-the-Counter Drug Sales
- In South Carolina as of February 2006
- 536 distinct sources are represented in the data
- Which is approx 2.6 of the National data sources
- Possible Uses for OTC Sales Data
- If OTC sales data indicate an increase in use of
anti-diarrheals, this could point to a foodborne
outbreak. - An increase in the use of cold/flu remedies might
give information on severity of the influenza
season or an acute respiratory event (i.e.
release of chlorine).
66(No Transcript)
67(No Transcript)
68(No Transcript)
69(No Transcript)
70(No Transcript)
71(No Transcript)
72(No Transcript)
73(No Transcript)
74(No Transcript)
75Next Steps
- Currently engaged in retrospective studies for
all documented outbreaks from 2005, by mode of
transmission, to determine if an temporal
association can be made and at what level.
76Bottom-line
- Syndromic surveillance is another arrow in our
quiver of ways to detect and respond to both our
routine outbreaks of public health significance
and also those large-scale events that will
require exceedingly complex reporting and
situational awareness requirements.
77Acknowledgements
- SC DHEC
- Amy Belflower
- Claire Youngblood
- Marya Barker
- Amy Roach
- Michelle Myer
- Mary Anne Wenck, EIS Officer
- Palmetto Poison Center
- Jill Michaels, PharmD
- William Richardson, MD
- Marine Corps Recruit Depot (MCRD), Parris Island
- LT Allen D. Wright, Preventive Medicine
Department, Naval Hospital Beaufort - CDC
- Leslie Z Sokolow
- Roseanne English
- Haobo Ma
78Questions?