Division of Disease Control and Health Protection

"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 et al. Int. J of Epidemiology 1976; 5:29-37.

1. Retirement Party for Bill Bigler to be Held October 9,2000

2. Communicable Disease Epidemiology Regional Training to be Held October 26^th - 27^th in West Palm Beach

3. Infant Mortality Racial Disparities in Florida by County

4. Rabies Information

5. Weekly Arbovirus Activity Summary

6. Editor’s Note: Weekly Disease Table

Bill Bigler, PhD, Senior Epidemiologist with the Bureau of Epidemiology, has retired after 35 years of distinguished service with the Florida Department of Health. Bill has served the State of Florida and its citizens in several positions including Research Director, Deputy State Epidemiologist, and Chief of the Tuberculosis Control Program. He is currently the supervisor of the Field Epidemiology section.

A retirement party will be held in his honor on October 9, 2000, from 2 to 4 PM in Room 310A of the Prather Building, Capital Circle Office Complex in Tallahassee. We are looking for pictures (originals will be returned), stories (written or oral), and cards/other greetings. Join us for a reunion and farewell and relive memories of how Bill has survived snakebites, overturned vehicles, attempted robbery, and reorganizations, all in the line of duty. For additional information, please contact Melanie Black.

Melanie Black, LCSW, Bureau of Epidemiology

The Bureau of Epidemiology is pleased to announce the next Principles of Epidemiology training program for county health department staff members, which will be held in West Palm Beach, Florida on October 26-27, 2000. The target audiences for the regional training programs are county health department staff members and partner agencies who are involved in epidemiology. County health directors and administrators are welcome to attend.

This program will provide an overview of epidemiological principles such as disease surveillance and reporting and communicable disease outbreak investigation. On-line registration will be available Monday October 2, 2000 through October 20, 2000 on the Bureau of Epidemiology Intranet web page.

Space is limited to 45, so please register as soon as possible.

Information will be provided in the Epi Update and on the Bureau of Epidemiology Intranet web page. We intend to offer one other training session this year in southwest Florida. If you are interested in hosting a training session or have questions related to this program, please feel free to contact Melanie Black, LCSW, Professional Training Coordinator for the Bureau of Epidemiology.

Dan Thompson, MPH and Richard S. Hopkins, MD, MSPH

One measure of racial disparity is the rate ratio. A rate ratio above 1 indicates the black rate is higher than the white rate, a ratio less than 1 indicates the black rate is lower than the white rate, and a rate ratio close to 1 indicates little or no disparity in the rates. The infant mortality rates and the rate ratios for Florida are similar to the statistics for the U.S. (see table 1). In the earlier period (1985-87) the rate ratio for Florida is slightly higher than the U.S. In the two later time periods, the U.S. rate ratio increased while the Florida rate ratio increased and then decreased so the rate ratio for Florida is lower than the U.S in the last two time periods in table 1.

The difference between the white and black rates is another way to measure racial disparities. The rate differences follow a pattern similar to the rate ratios. In the earliest time period (1985-87) the rate difference is smaller for the U.S. compared to Florida. This is reversed in the two later time periods where the rate differences are smaller for Florida. This is also reflected in the percent decrease for the rates. This statistic shows a consistently greater decrease for Florida compared to the U.S. for both the white and black infant death rates. As a result, the rates for both races are substantially lower than the U.S. rates in the latest time period, even though they are close to the U.S. rate in the earliest time period.

In summary, based on the data in table 1, the racial disparities in the infant death rates have improved more in Florida than in the U.S. and the rates for both races have also improved more for Florida.

However, within Florida the rate ratios may be uniform across all of the counties of the state or there may be counties, or groups of counties, where the rate ratio is much different than the rest of the state.

The purpose of this paper is to examine the infant mortality rate ratios for the 67 counties in Florida, to find out if the racial disparities in infant death rates are consistent throughout the state or, alternatively, if there are areas where racial disparities are more pronounced.

The data for this analysis were obtained from the Florida Department of Health PHIDS system which is available on the department web site at: www.doh.state.fl.us. The data retrieved through the PHIDS system are Vital Statistics birth and infant death data. These were obtained in electronic form and placed in an Excel Spreadsheet.

Infant death rates were calculated for white and non-white infants, for each county, and the rate ratios (non-white rate to white rate) were then calculated for each county. The

The rate ratios for each county were compared to the statewide rate ratio to determine if the county rate ratio is significantly different from the statewide ratio. Monte Carlo tests of the statistical methods used here have shown the methods are reliable when the numerators used to calculate the rates are greater than 4. For this reason the statistical test is not done when this condition is not met, i.e. when the county has fewer than 5 infant deaths in either the white or non-white category.

This entire method was applied to two five-year time periods; 1989 through 1993 and 1994 through 1998. The five-year time periods were used because in many counties the number of infant deaths would be too small to analyze for shorter time periods. The 2 time periods were examined to assess the variation in results over time and to examine the change in the rates and rate ratios over time.

The results are given in tables 2 and 3. Table 2 contains data for the years 1989 through 1993 and table 3 is for years 1994 through 1998. The statewide rate ratio is lower for the 1994-98 time period compared to the 1989-93 period. (2.03 and 2.15 respectively). The infant death rates for both white and non-white infants are also lower for the later time period.

Among the 67 counties, there were 45 in the 1989-93 time period (table 2) that could be compared statistically to the statewide rate ratio. Of these 45 counties, 3 had rate ratios that were significantly higher than the statewide ratio (Broward, Columbia, and Flagler) and one county had a rate ratio significantly lower than the statewide ratio (Pasco). In all three counties, where the rate ratio was significantly higher than the state ratio, the white rate was lower than the state rate for white infants and the non-white rate was higher than the state rate for non-white infants. In short, the low white rate and high non-white rate resulted in a high rate ratio.

In the 1994 through 1998 time period (table 3) there were 43 counties that could be compared statistically to the statewide rate ratio. Of these 43 counties, two had rate ratios that were significantly higher than the statewide ratio (Bradford and Palm Beach) and one county had a rate ratio significantly lower than the statewide ratio (Volusia). As in the earlier period, the two counties with the significantly higher rate ratios had relatively low white rates and relatively high non-white rates compared to the state rates. This resulted in the high rate ratios.

There are several conclusions that can be drawn from this data. One positive feature is the decline in the statewide rate ratio for the later time period compared to the earlier time period. This may indicate the racial disparity in infant death rates is declining. However, the statewide death rate for non-white infants is still 2.03 times the rate of white infants in the latest five-year period (1994 – 1998), so the disparity is still large.

It is also worth noting the data in table 1 which shows that for the period 84-87 racial disparity, as measured by the infant death rate ratio, was higher for Florida compared to the U.S. In the two later time periods the pattern is reversed and the rate ratios are lower for Florida. This may indicate that Florida has improved more than the U.S. in regard to racial disparities in infant death rates.

Another conclusion is the racial disparity in infant death rates does not appear to be consistently more pronounced in any particular county or group of counties. There were a few counties with significantly higher rate ratios in each of the 2 time periods but these were different counties for each of the time periods and the counties were not geographically contiguous in either time period. It should be noted that when a statistical test is applied, using the 5% level of significance, to 67 counties it is not unusual to get statistically significant results in two or three counties due to chance.

Dr. Lisa Conti, State Public Health Veterinarian

During the first eight months of 2000, 113 rabid animals which had exposed at least one person or pet were reported by the Department of Health Laboratories (148 animals were reported during this period in 1999). While rabies is a rare human disease, it is not uncommon in animals in the state and considerable time and effort is devoted to animal bite investigation for determination of rabies management.

The Department of Health has several tools for rabies education, control and prevention. These include documents which appear on the Department's website (www.doh.state.fl.us, click Epidemiology from the pull down menu, then Health Topics, then Rabies). On this site you will find the state rabies guidebook, surveillance information and a brochure for the general public. A limited number of rabies brochures are available from the Bureau of Epidemiology (859-245-4401). We invite suggested changes to any of these documents to make them more effective and user-friendly.

Recently, the Journal of the American Medical Association (JAMA) offered a new patient education piece titled "Rabies" in its issue dated August 23/30, 2000. Please note that one sentence in this document may be misleading, "Also contact a doctor or seek emergency care right away if you think you have been exposed to the virus that causes rabies in the air, such as after visiting a cave where bats live or waking to find a bat in your room, tent, or cabin." Documented aerosol transmission of rabies in people has occurred in 1957 and 1960, attributable to exposure in a cave containing millions of free-tailed bats (one of these individuals also occasionally worked with rabies in a laboratory).1 Two other nonbite-associated rabies cases were in rabies laboratory workers (1973 and 1977).2 Given the thousands of spelunkers and visitors to caves with no additional documented human rabies due to aerosols, this potential route of transmission is considered extremely unlikely. Finally, this document should have listed local and state health departments as sources for additional information about rabies

1. Baer, The Natural History of Rabies, Academic Press, New York, 1975, 115-120.

2. CDC. Rabies in a laboratory worker - New York. MMWR 1977; 26:183-4.

Robin Oliveri, Arbovirus Surveillance Coordinator and Lisa Conti, DVM, MPH, State Public Health Veterinarian

There are currently no Arbovirus Medical Alerts issued for the state. During the period September 17 through September 23, 2000 the following arbovirus* activity was recorded for Florida:

(*Mosquito-borne virus including St. Louis encephalitis virus, Eastern Equine encephalitis virus, West Nile encephalitis virus and dengue virus)

Sentinel chickens: Three sentinel chicken seroconversions to SLE were identified; one each in Lee, Hendry and Charlotte counties (517 chickens tested). (Source: DOH Tampa Laboratory from mosquito control agencies and county health departments). See attached figure.

The Weekly Disease Table will return to this space next week.