July 10, 2006
Epi Update Managing Staff:
"The reason for collecting, analyzing and disseminating information on a disease is to control that disease. Collection and analysis should not be allowed to consume resources if action does not follow."
Foege WH, International
Journal of Epidemiology 1976; 5:29-37
On March 24, 2006, the Florida Poison Information Center notified the Broward regional environmental epidemiologist of two cases of possible ciguatera poisoning associated with eating a large grouper, in persons residing in Islamorada, Florida. Initial interviews of the two cases noted a similar story about receiving a large black grouper as a gift from a fisherman friend and becoming ill with gastrointestinal and neurological symptoms - tingling/ numbness/ burning sensations/itching after consumption of the black grouper. Contact information for the fisherman was obtained. An interview with the fisherman provided information linking both groups along with additional information for two other groups (3 persons total) who had also consumed the fish and become symptomatic.
Samples of the leftover black grouper were collected by Monroe CHD Environmental Health staff from two of the groups and sent to the FDA Gulf Coast Seafood Laboratory for analysis. Samples of the black grouper were examined for the presence of ciguatera-related toxins using the sodium channel-specific mouse neuroblastoma (cytotoxicity) assay. Caribbean ciguatoxin-1 (C-Ctx-1) was used as a standard.
A total of 8 cases from 4 separate groups ate the fish that was given to them by the fisherman. Three of the cases were Broward County residents, and five of the cases were Monroe County residents. The cases ranged in age from 47- 71 years, with a median of 62 years. Half of the cases were male, and half were female. Dates of onset ranged from 4 - 12 hours following ingestion of the black grouper with a median incubation period of 6.5 hours. Duration of symptoms was ongoing for four of the cases as of May 9, 2006. Range of symptoms for other cases was 4 - 8 weeks' duration with a median duration of 5 weeks.
The results of the laboratory samples were:
Sample 06-12-1: 0.58 ng C-Ctx 1 Eq/g sample
Sample 06-12-2: 0.04 ng C-Ctx 1 Eq/g sample
Caribbean ciguatoxin-1 was confirmed in both samples of the black grouper. There was a 10-fold difference between the samples. Based on previous experience, reported by the FDA Gulf Coast Seafood Lab, levels of ciguatoxin in both samples were considered marginal for causing illness.
Most of the interviewed patient cases were aware that they should avoid eating large reef fish; however, some said they didn't think about it until after the fact, and some mentioned they assumed since it was caught in deeper waters it was safe. A couple of the cases also ate leftovers of the fish before they realized the fish was making them ill. One of the cases stated they fed some of the fish to their dog, which subsequently developed diarrhea for about a week.
Frequencies of symptoms are summarized in Table 1.
Table 1. Frequency of Symptoms for a confirmed Ciguatera outbreak
Ryan Lowe is a Regional Environmental Epidemiologist in the Food and Waterborne Disease Program for the Bureau of Community Environmental Health assigned to Broward County. He can be reached at 954.467.4841.
1 FDA's Bad Bug Book. Ciguatera. http://www.cfsan.fda.gov/~mow/chap36.html
County Health Department Hurricane
The Bureau of Epidemiology presented its initial roll-out of the County Health Department Epidemiology Hurricane Toolkit, which is now posted at: http://dohiws.doh.state.fl.us/Divisions/Disease_Control/epi/Hurricanetoolkit/index.htm.
In a Grand Rounds presentation on Tuesday,
June 27, a basic training provided an overview of what the hurricane
toolkit contains and how it might be utilized by CHD epidemiology units
and/or epidemiology teams for post-hurricane response.
The Bureau of Epidemiology is also looking for other ways to distribute the toolkit, including providing CHD epidemiology contacts ‘thumb’ drives of the toolkit with the FDOH logo, so that the kit can become part of every CHD epidemiology unit’s go-kit for disasters. Details on the thumb drives will be provided in future email announcements.
For more information concerning this device, contact Janet Hamilton, MPH, at 850.245.4444, ext. 2403.
Janet Hamilton is administrator of the Surveillance Section at the Bureau of Epidemiology in Tallahassee.
Grant Writing Workshop Scheduled
The Grant Institute's Grants 101: Professional Grant Proposal Writing Workshop will be held at the University of Miami - James L. Knight International Center, August 9-11from 8:00 a.m. until 5:00 p.m. each day. Development professionals, researchers and graduate students should register as soon as possible. All participants will receive certification in professional grant writing from the institute.
The Grant Institute’s Grants 101 course is an intensive and detailed introduction to the process, structure, and skill of professional proposal writing. Participants will learn the entire proposal writing process and complete the course with a solid understanding of not only the ideal proposal structure, but a holistic understanding of the essential factors, which determine whether or not a program gets funded. Through the completion of interactive exercises and activities, participants will complement expert lectures by putting proven techniques into practice. This course is designed for both the beginner looking for a thorough introduction and the intermediate looking for a refresher course that will strengthen their grant acquisition skills.
In three days,
students will be exposed to the art of successful grant writing practices,
and led on a journey that ends with a complete grant proposal. Grants 101
consists of three courses: Fundamentals of Program Planning, Professional
Grant Writing and Grant Research.
· The Grant Institute
Certificate in Professional Grant writing
For more information, call 888. 824.4424 or visit The Grant Institute website at http://www.thegrantinstitute.com
Salmonellosis in Collier County:
Because Salmonella infection is preventable with 60%-80% of cases occurring sporadically,2 studies aimed at further understanding the trends and distribution of non-outbreak associated cases are informative when planning public health interventions intended to reduce the spread of this enteric disease.
However, assessing rates on a county-wide level for counties that are expansive geographically and large in population hides variation within the region. To better understand the trends in salmonellosis cases within Collier County, we aimed to assess Salmonella infection rates within smaller geographic areas.
Our overall goal was to define the geographic distribution of salmonellosis rates to identify regions where individuals are at a particularly high risk of infection, so as to better plan and implement interventions that could serve to minimize future transmission.
Initial data from Merlin indicated 376 cases of salmonellosis in Collier during the 5-year period of this study. After the aforementioned exclusion and inclusion criteria were applied, 92 records were eliminated (58 were outbreak-associated, 13 were acquired outside the state, and 21 had missing information about outbreak status or acquisition location), resulting in 284 eligible cases. From this data, 10 cases (3.5%) had to be further eliminated due to unknown or incorrect addresses. As a result, 274 cases were entered into this analysis.
The surveillance data entered into Merlin comes from a variety of sources. Collier CHD utilizes Health Sentry, an electronic lab reporting system whereby all positive Diagnostic Services, Inc. (DSI) lab results are reported directly to the health department daily through downloadable, secure online databases. This system is considered an exceptionally complete lab surveillance method because an estimated 90% of all lab tests ordered in the county are processed by DSI labs. In addition, faxes and calls from hospital infection control nurses and private practitioners alert health department staff about additional cases.
The demographic distribution for all cases was determined. The number of cases reported during this period in each zip code was recorded, and the age-adjusted 5-year average rates of Salmonella infection were calculated for each zip code region.
To calculate the 2002 midyear population for each zip code, we extrapolated data for each zip code from the US Census (available at http://factfinder.census.gov), using the 2000 and 2002 county-wide population estimates (available from the Collier County Government Long-Term Planning Committee) and the 2000 population. This was necessary because the US Census does not estimate the zip code level population yearly. There are two possible limitations to these data. First, population growth rates may vary across the county resulting in some zip codes increasing in population more quickly than others. However, because we were only extrapolating over a 2 year period, this variation was likely to be minimal. Second, the 2000 US Census uses Zip Code Tabulation Areas (ZCTA) as an alternative to zip code regions because the boundaries are more uniform and useful for mapping purposes. In most cases, ZCTA regions are nearly identical to zip code regions, and a comparison of these two boundaries indicates that they are appropriate for this study.
Age-adjusted rates were calculated using three age categories applicable to this study (under 5 yrs, 5 – 64 yrs, and 65 yrs and older). Age classes were combined because of the small number of cases; these specific boundaries were chosen because the case rates are similar within these age groups. The calculated rates were then graphed and mapped for comparison. 95% confidence intervals (CI) were calculated using the formula:
95% CI = (Age-adjusted Rate) +/- 1.96* (Age-adjusted Rate) /mathematical checkmark symbol (Number of Events)
There are many alternate methods for calculating the 95% CI, but this formula is considered to be more than adequate for use in small area analysis for community health planning.3
Because studies involving geographic distributions of cases rely heavily on accurate address information, we compared archived paper records for the 5-year study period to electronic records, to ensure that the address listed in Merlin was correct at the time of diagnosis. Only 4 records could not be confirmed, as their paper records could not be located. These records were included in the analysis. Each address was then verified with MapMarker, a geo-coding program that can determine if an address actually exists. Any addresses that required further verification were checked using the Naples, Florida Cross Search Reference Directory (October 2004 edition). Only individuals with known physical addresses were included in this study.
The one case residing in zip code 34134 was not included in the geographic rate distribution analysis because no population data was available for the section of this zip code that is located in Collier County. The zip code crosses county boundaries and is partly located in Lee County; therefore, calculating the rate was not feasible.
Data was analyzed and graphs were created using Microsoft Excel. MapMarker Plus 10.2 and ArcMap 9.0 were used to map the rates within the zip code regions.
In cases where an infection or a disease disproportionately affects a certain age class, as is the case with salmonellosis, it can be misleading to compare crude rates. This is especially true when the underlying age distributions of the regions being compared are quite different (figure 2).
Therefore, we calculated and graphed the 5-year average age-adjusted rate of salmonellosis cases in each zip code region along with the corresponding 95% confidence interval (figure 3). Zip code region 34120 (Golden Gate Estates) had the highest rate at 56.23 cases per 100,000 people per year. This rate was significantly higher than the rate in 11 other zip code regions of the county. Conversely, zip code 34102 (City of Naples) had the lowest rate with just 7.70 cases per 100,000. The most economically disadvantaged area and the area of the county with the largest migrant community, zip code 34142, had a rate of 26.61 cases per 100,000. No cases were reported during the 5-year time period in zip codes 34137, 34138, 34139, 34140, and 34141, all of which are very rural, sparsely populated areas of the county.
To understand the spatial relationship between these regions and identify trends, salmonellosis rates were mapped (figure 4). This representation suggests that there is a north-south region within the center of the county where rates are higher than in the rest of the county. In particular, it highlights the significantly higher rate of Salmonella infections in 34120.
This study aimed to identify regions of Collier County where unexpectedly high rates of Salmonella infections have been reported. By comparing 5-year average age-adjusted rates for each zip code, we found the Golden Gate Estates region to have the highest reported case rate. This region’s rate was significantly higher than the rate for 11 other zip codes in the county where cases have been reported.
In this analysis, several years’ data were pooled in an attempt to increase the sample size in each region. In addition, by calculating the 95% CI for comparison, the issue of small sample sizes and variability in the rates was directly addressed. It is possible that confounding factors could lead to a higher rate in one region compared to another. The greatest source of confounding would be due to age, since salmonellosis is an infection that disproportionately affects young children. In this study, we accounted for this fact by comparing age-adjusted rates only.
In addition, it is possible that reporting frequencies differ between the zip code regions of the county. This could occur if, for instance, physicians serving one area were more vigilant about reporting positive lab results to the county health department. However, since Collier CHD utilizes the electronic lab reporting system Health Sentry, there is minimal variation in reporting activity within the county.
On the other hand, it is possible that individuals in one area of the county seek out health care for salmonellosis-like symptoms more often then those living in other areas, perhaps due to the cost of health care. If this were the case, we would expect to see the highest rates in the most affluent areas of the county. Instead, we found that the most affluent area, the city of Naples, had the lowest rate, thereby refuting this theory.
This study has the advantage of utilizing a reliable dataset, of having minimized confounding based on age and reporting frequency, and of clearly identifying an area where further interventions aimed at reducing the transmission of Salmonella bacteria could benefit the community. Of course, all communities could benefit from educational information about adequate hand-washing, proper food handling protocols, and infection control measures to minimize person-person transmission. However, if certain high-risk regions are identified based on studies such as this one, efforts should be made to ensure that useful, timely information about how to reduce the risk of transmission is made available.
A case-control study is an excellent way to utilize data about case exposures to identify risk factors in a specific population. An individual with epidemiology experience, an adequate sample size, and a detailed questionnaire would be needed in order to obtain valid results.
Attending health fairs and presenting to community groups would also serve to educate communities about ways to prevent Salmonella and food-borne illness in general.
These interventions could educate the community and identify behaviors leading to exposure to Salmonella bacteria that may be more common in certain areas, such as keeping yard chickens, or consuming raw or undercooked eggs. These suggestions will require greater time and resources than the current case investigation methods, but the high rate of salmonellosis cases warrants the attention and could ultimately reduce the case rates in areas where they are higher than expected.
Nicole Basta is a Florida Epidemic Intelligence Service
fellow assigned to the Collier and Hendry/Glades County Health
Departments. She can be reached by calling 239.774.8234.
CDC Issues Testing Guidelines
CDC Guidance on Who Should be Tested for
Influenza A (H5N1) Virus in Florida
For consultation on testing or management of possible human cases of avian influenza, contact your country health department or the Bureau of Epidemiology at 850.245.4401
However, this revised interim guidance provides an updated case definition of a suspected H5N1 human case for the purpose of determining when testing should be undertaken and also provides more detailed information on laboratory testing. Effective surveillance will continue to rely on health care providers obtaining information regarding international travel and other exposure risks from persons with specified respiratory symptoms as detailed in the recommendations below. This guidance will be updated as the epidemiology of H5N1 changes.
CDC is revising its interim guidance for infection control precautions for avian influenza A (H5N1). These will be issued as soon as they are available.
The total number of confirmed human cases of H5N1 reported as of June 7, 2006 has reached 225. The case fatality rate for these reported cases continues to be approximately 50 percent. As of this date, H5N1 has not been identified among animals or humans in the United States.
The epizootic in Asia and parts of Europe, the Near East and Africa is not expected to diminish significantly in the short term and it is likely that H5N1 infection among birds has become enzootic in certain areas. It is expected that human infections resulting from direct contact with infected poultry will continue to occur in affected countries. Since no sustained human-to-human transmission of influenza H5N1 has been documented anywhere in the world, the current phase of alert, based on the World Health Organization (WHO) global influenza preparedness plan, remains at Phase 3 (Pandemic Alert).*
In addition, no evidence for genetic reassortment between human and avian influenza A virus genes has been found. Nevertheless, this expanding epizootic continues to pose an important and growing public health threat. CDC is in communication with WHO and other national and international agencies and continues to monitor the situation closely.
Reporting and Testing
Testing for avian influenza A (H5N1) virus infection is recommended for:
A patient who has an illness that:
1. requires hospitalization or is fatal; AND
2. has or had a documented temperature of ≥ 38°C (≥100.4° F); AND
3. has radiographically confirmed pneumonia, acute respiratory distress syndrome (ARDS), or other severe respiratory illness for which an alternate diagnosis has not been established; AND
4. has at least one of the following potential exposures within 10 days of symptom onset:
a. History of travel to a country with influenza H5N1 documented in poultry, wild birds, and/or humans,† AND had at least one of the following potential exposures during travel:
i. direct contact with (e.g., touching) sick or dead domestic poultry;
ii. direct contact with surfaces contaminated with poultry feces;
iii. consumption of raw or incompletely cooked poultry or poultry products;
iv. direct contact with sick or dead wild birds suspected or confirmed to have influenza H5N1;
v. close contact (approach within 1 meter [approx. 3 feet]) of a person who was hospitalized or died due to a severe unexplained respiratory illness;
b. Close contact (approach within 1 meter [approx. 3 feet]) with an ill patient who was confirmed or suspected to have H5N1;
c. Worked with live influenza H5N1 virus in a laboratory.
Testing for avian influenza A (H5N1) virus infection can be considered on a case-by-case basis, in consultation with local and state health departments, for:
Clinicians should contact their county health department and the Florida Department of Health, Bureau of Epidemiology 850.245.4401 as soon as possible to report any suspected human case of influenza H5N1 in the United States. Advanced consultation with the Bureau of Epidemiology can be done before testing commences. See specific guidance on the Florida Department of Health Internet site dated 6/8/06 http://www.doh.state.fl.us/disease_ctrl/epi/htopics/BirdFlu.htm.
Specimen Collection and Testing Guidelines
Travel Health Notice
Annual Florida Professionals in Infection
Control Conference to be Held in September
The 31st Florida Professionals in Infection Control Annual Conference will be held in Orlando September 12 - 15, 2006 at the Orlando Airport Marriott.
A pre-conference seminar will be held on September 12th and is approved for 6 contact hours titled "Infection Control Basics Across the Continuum of Care" which will focus on basic infection control in all settings.
On September 13, the conference "Expanding The Horizons" will begin with renowned keynote speaker William Jarvis, MD to address the Evolution of Infection Control across the Continuum. The conference will present state of the art information about infection control practice and disease control. The programs are designed for infection control professionals in all health care settings and are approved for 18 contact hours of continuing education for nursing and clinical laboratory personnel.
Call for abstracts is due by July 26. Here is an excellent opportunity to present your infection control solutions/ strategies at the annual conference!
For more information on the conference and
abstract submission visit www.flpic.com.
Mosquito-borne Disease Summary June 25 - July1, 2006
Rebecca Shultz, MPH, Caroline Collins, Daneshia Roberts, Calvin DeSouza, Carina Blackmore, PhD
During the period June 25 - July 1, 2006, the following arboviral activity (St. Louis Encephalitis virus [SLEv], Eastern Equine Encephalitis virus [EEEv], Highlands J virus [HJv], West Nile virus [WNv], California Group virus [CALv]) was recorded in Florida:
EEE virus activity: One horse from Okaloosa County was reported positive for EEE virus infection this week. A total of 16 counties have reported EEEv activity so far this year, compared to 33 at this time last year.
WN virus activity: There was no West Nile virus activity reported this week. So far, 10 counties have reported WNv activity this year, compared to 11 at this time last year.
No locally-acquired human cases of arboviral infection were reported yet this year.
Horses: One horse from Okaloosa County tested positive for EEE virus infection this week.
Dead Birds: The Fish and Wildlife Conservation Commission (FWC) collects reports of dead birds, which can be an indication of arbovirus circulation in an area. This week, 43 reports representing 76 dead birds were received from 25 counties. Of the reported birds, 5 were corvids, 3 were a type of raptor, and the remaining 68 were other birds. The FWC collects reports of birds that have died from a variety of causes, not only arboviruses. There were no positive test results this week.
See the web page for more information: www.MyFloridaEH.com. The Disease Outbreak Information Hotline offers recorded updates on medical alert status and surveillance at 888.880.5782. Dead birds should be reported to www.myfwc.com/bird/
The Bureau of Epidemiology encourages
Epi Update readers to not only register on the EpiCom system at
https://www.epicomfl.net but to sign up for features such as automatic
notification of certain events
contribute appropriate public health observations related
any suspicious or unusual occurrences or circumstances. EpiCom is the primary method of communication
between the Bureau of Epidemiology and other state medical agencies during emergency situations.
Christie Luce is administrator
of the Surveillance Systems Section in the Bureau of
Epidemiology. She can be reached at 850.245.4444, ext. 2450.
Click here to review the most recent disease figures provided by the Florida Department of Health Bureau of Epidemiology.
D'Juan Harris is a Systems Project Analyst in the Surveillance Systems Section of the Bureau of Epidemiology. He can be reached at 850.245.4444, ext. 2435.
Copyright©2006 State of Florida