Topic: A Spatial Analysis of the Leading Causes of Childhood Morbidity in Florida: Aggregates, Incidence Clusters, and the Implications for Prevention in Populations
Presenter: Matt Laidler, MA, MPH, Sarasota County Health Department, Florida Epidemic Intelligence Service Program
Date: Tuesday, March 29, 2005

Abstract:
This presentation describes a spatial analysis of the 5 leading causes of childhood morbidity in Florida as reported to the Bureau of Epidemiology through the Merlin system. The units of analysis were rates of morbidity within zip code areas throughout the state. A spatial scan test was used to identify where local clusters of incidence occur, the rates identified within these clusters, and the amount of aggregate incidence explained by the cases within the detected clusters. Statistically significant incidence clusters were detected for all 5 of the leading causes of morbidity in persons under 18 years of age. Incidence in some of the significant clusters accounts for nearly 50% of the aggregate (state-wide) incidence for particular diseases. It is suggested that addressing prevention/intervention in these areas could reduce a substantial proportion of aggregate incidence, although typical approaches may vary.

Additional Information:
The grand rounds presentation will begin promptly at 11:00 a.m. EDT on Tuesday, March 29, 2005. The PowerPoint slides and dial-in number will be on the Bureau of Epidemiology intranet Website on Friday, March 25. CEUs will be provided for nursing and environmental health professionals. If additional information is needed, contact Professional Training Coordinator Melanie Black, MSW, at 850.245.4444 ext. 2448, or via email at Melanie_Black@doh.state.fl.us

Matt Laidler is an EIS fellow in Sarasota County and can be reached at 941.861.2916.

Decline in Varicella Hospitalizations and Mortality, Florida 1995-2003       
 by Richard Hopkins, MD, MPH                                       
 Lilian Kigonya, MBChB, MPH

In a recent journal article¹, a group of CDC scientists showed that nationally, mortality from varicella (chickenpox) has dropped dramatically since introduction of the varicella vaccine in 1995, both as the underlying cause of death and as an associated cause. We examined Florida death certificate data for varicella for 1995 through 2003 using similar methods, and also in-patient hospital discharge data maintained by the Florida Agency for Health Care Administration. The health impact of varicella in Florida has declined precipitously since the vaccine came into widespread use. Varicella is not currently a notifiable disease in Florida, but it has been proposed to be added to the list of notifiable diseases in the next revision cycle.

Methods
We selected records for the varicella mortality analysis according to the 9th (ICD-9) and 10th (ICD-10) version of the International Classification of Diseases. We used ICD-9 (code 052), for the period 1995 to 1998 and ICD-10 (code B01.0 through B01.9) for 1999 through 2003. The International Classification of Diseases 9th revision Clinical Modification (ICD9-CM) was used to identify varicella hospital discharges (code 52 through 52.9).

Hospital discharge and mortality data were analyzed and tabulated using the Statistical Program for the Social Sciences (SPSS).

Vaccination coverage data for Florida two-year-olds were obtained from the Department of Health’s annual immunization coverage survey for the years 1999 through 2004.

Vaccine coverage
The percent of two-year-old children who have completed varicella immunization has progressed from 32% in 1999 to 90% in 2004.

Mortality
The total number of deaths in which varicella was mentioned as a cause of death on the death certificate (whether or not it was the underlying cause) for the nine year period was 66. Of these, 40 had varicella as the underlying cause and 26 did not. There was a steady decline in annual number of total deaths from 12 in 1995 to 3 in 2003, and of deaths with varicella as underlying cause from 7 in 1995 to 2 in 2003. The decline in total deaths represents a 75% decrease.

Fig. 1. Varicella deaths by year and diagnostic group, Florida, 1995-2003

Among the 66 deaths with any mention of varicella, 45 (66.2%) were in whites and 21 in blacks. Eighteen (27.2%) of the deaths were in Hispanics and 48 in non-Hispanics. There was a similar distribution of the cases with varicella as underlying cause.

Deaths occurred at all ages (see table 1), even though varicella itself is primarily a disease of young children. This may reflect the often more severe course of varicella in adults, especially those with chronic health problems such as cancer or immune system deficits. It could also reflect misclassification of some deaths from or with herpes zoster (shingles) as varicella. Also, in recent years, as more children are immunized against varicella, the proportion of the cases still occurring that are in adults is increasing.

Table 1. Varicella deaths by age group and diagnostic category, Florida, 1995-2003

Hospitalizations
Over the nine-year period, there were a total of 5,153 hospital discharges for varicella, of which 2,715 had varicella as the principal diagnosis. The annual total number of discharges fell from 907 in 1995 to 219 in 2003, a 75.8% decrease (Figure 2). Similarly, the annual number of discharges with varicella as the principal diagnosis fell from a maximum of 438 in 1998 to 118 in 2003, a 73.1% decline. Recent years, since 1999, have shown the greatest rate of decline.

The remainder of this report will consider all 5153 varicella discharges, as the patterns are similar when those with and without varicella as the principal diagnosis are considered.

The pattern of discharges by season followed the known epidemiology of varicella, with 30.0% occurring in January through March and 36.3% in July through September. Males and females were equally represented. By race, 70.1% were white and 25.4% black. Hispanics accounted for 20.8%.

Figure 2. Varicella discharges from acute care hospitals by year, Florida 1995-2003

Most persons discharged (89.5%) were discharged home without need for further follow-up. There were 79 who were discharged after dying, for a case-fatality rate among hospitalized persons of 1.5%. This number of deaths is somewhat larger than the number (66) who had varicella mentioned on the death certificate. The reason for this discrepancy is not known. Hospital discharge records would have to be matched to death certificates to sort this out.

The declines in numbers of cases were high in all race/ethnicity groups: 73.7 % for non-Hispanic whites, 65.8% for Hispanic whites, and 86.9% for non-Hispanic blacks.

Cases were distributed across the entire age range, with the majority in children under age 15 (Table 2). The age distribution in cases in which varicella was not the principal diagnosis was somewhat shifted to older ages. There were more discharges in young adults 25 to 34 years old than in those aged 15 to 24, perhaps because those aged 25 to 34 have more contact with young children.

Table 2. Varicella discharges from acute care hospitals, by age group and diagnostic category, Florida, 1995-2003

For all ages combined, as shown above, the reduction in number of cases was 75.8% over the period under study. This reduction was much greater for the youngest groups. For children aged under 5 years, the reduction in discharges was from 397 in 1995 to 26 in 2003, or 93.4%, while for those aged 35, where there were relatively few discharges, there was little or no reduction (see Table 3).

Table 3. Discharges from Florida acute care hospitals with a diagnosis of varicella, by age group, 1995 and 2003, with percentage change from 1995 to 2003

Comment: The improvement in numbers of hospitalizations and deaths from varicella since introduction of the varicella vaccine in 1995, especially in children, is dramatic and an important public health success story. The timing of the improvement and the consistency between hospitalization and mortality data are consistent with varicella immunization being the major cause of the improvement.

Varicella is not currently a reportable disease in Florida. The Centers for Disease Control and the Council of State and Territorial Epidemiologists have recommended that states begin collecting individual varicella case reports. This would not have been practical or meaningful in the pre-vaccine era, but now that varicella appears to be an uncommon infection, case-based surveillance can assist in identifying populations at risk of varicella in whom additional immunization efforts are needed. Continued monitoring of hospitalization and mortality data will also play a part in documenting the success and remaining challenges of the immunization effort.

References.

1. Nguyen HQ, Jumaan AO, Seward JF: Decline in mortality due to varicella after implementation of varicella vaccination in the United States. NEJM 2005, 352:450-458.
                      
Richard Hopkins is assistant chief for science at the Bureau of Epidemiology in Tallahassee and can be reached at 850.245.4444, ext. 4412. Lilian Kigonya is a surveillance and reporting epidemiologist and can be reached at 850.245.4444, ext. 2417.

 Background: Receiving regular cancer screening plays a vital role in the early detection and treatment of cancer. Colorectal cancer can be detected through the use of screening methods, such as Blood Stool Test and sigmoidoscopy.

Objective: This study identifies racial disparities in colorectal cancer screening and its related factors among adults ages 50 years and older in Florida.

Methods: SAS Version 9.0 was used to analyze data from the 2002 Florida County Behavioral Risk Factor Surveillance System (BRFSS). Cancer screening behaviors were compared among racial/ethnic groups and also among people with or without health insurance or health care providers.

Results: According to the 2002 Florida County BRFSS data, there exists substantial racial disparities in the prevalence of colorectal cancer screening among adults 50 years and older in Florida. The prevalence of having a Blood Stool Test in the past two years among both non-Hispanic Blacks (28.0%) and Hispanics (21.9%) was lower than among non-Hispanic Whites (36.2%). Similarly, the prevalence of ever having a sigmoidoscopy among both non-Hispanic Blacks (44.0%) and Hispanics (37.6%) was lower than among their non-Hispanic White counterparts (56.2%). Among non-Hispanic Blacks and Hispanics, the rates of having either health insurance or a personal healthcare provider were lower than among non-Hispanic Whites. Compared to people with health insurance, those without health insurance had a lower prevalence of having a blood stool test in the past two years (19.6% versus 35.0%), or sigmoidoscopy (31.1% versus 55.0%). Similarly, in comparison to people who had a personal healthcare provider, those who did not have a personal healthcare provider had a lower prevalence of having a blood stool test in the past two years (18.0% versus 35.9%) or sigmoidoscopy (28.3% versus 56.5%).

Conclusion: This study cites specific racial disparities in colorectal cancer screening among adults ages 50 years and older in Florida. Results of this study could be used to reduce the racial disparity of cancer screening, in addition to developing efforts to increase the prevalence of cancer screening among all adults in Florida.

Zhaohui Fan is an epidemiologist in the Chronic Disease Surveillance and Epidemiology Section at the Bureau of Epidemiology in Tallahassee. He can be reached at 850.245.4444, ext. 2418.

Annual Bureau of Epi Seminar Arrangements Underway
by Melanie Black, MSW

The Bureau of Epidemiology is excited to announce the dates for the Tenth Statewide Epidemiology Seminar, “Emerging Issues in Epidemiology”, May 17-18, 2005 in Lake Mary, Florida at the Orlando Marriott Hotel. Last year’s meeting brought together over 200 epidemiologist and other public health professionals representing the state department of health, county health departments, and other partners to discuss current issues in communicable and chronic disease prevention and control. The Department of Health disease prevention staff and other medical providers will be exposed to ideas and methods presented by disease prevention experts from Florida, other states and the Centers for Disease Control and Prevention.

We are in the process of developing an interesting, informative and challenging agenda, a list of exciting speakers and an excellent poster session, not to mention time and occasion for colleagues to interact. Be sure to take advantage of this once-a-year opportunity!

Further details about this program and accommodations will be made available in the Epi Update and on the Bureau of Epidemiology Internet website. Melanie Black, MSW, will be managing this activity and can be reached at 850. 245.4444 ext. 2448 or SunCom 205.4444 ext. 2448.

Melanie Black is a professional training coordinator at the Bureau of Epidemiology in Tallahassee. She can be reached at 850.245.4444, ext. 2448.

     

           This Week on EpiCom
                                      by Pete Garner