Georgia epidemiology report, Vol. 14, no. 5 [i.e. 6] (June 1998)

June 1998

volume 14 number 5

Division of Public Health
http://www.ph.dhr.state.ga.us
Director Kathleen E. Toomey, MD, MPH

The Georgia Epidemiology Report is a publication of the Epidemiology Section of the Epidemiology and Prevention Branch, Division of Public Health, Georgia Department of Human Resources
Summary of the 1998 Georgia Stroke Report
Introduction

Epidemiology and Prevention Branch State Epidemiologist
Acting Director Kathleen E. Toomey, MD, MPH
Epidemiology Section
Chief Paul A. Blake, MD, MPH
Public Health Advisor Mel Ralston
Notifiable Diseases
Jeffrey D. Berschling, MPH, Katherine Gibbs McCombs, MPH, Carol Hoban, MS, MPH, Jane E. Koehler, DVM,
MPH, Laura Gilbert, MPH Amanda Reichert, RN
Chronic Disease
Ken Powell, MD,MPH- Program Manager, , Patricia M. Fox, MPH, Rana Bayakly, MPH, Mary P.
Mathis, PhD, MPH, Alexander K. Rowe, MD, MPH
Tuberculosis
Rose Marie Sales, MD, MPH- Program Manager Naomi Bock, MD, MS, Beverly DeVoe, MS
HIV/AIDS/Sexually Transmitted Diseases
John F. Beltrami, MD, MPH- Program Manager Andrew Margolis, MPH, Lyle McCormick, MPH
Ann Buckley, MPH, Amy Hephner, MPH, Laura Axelson, MPH
Perinatal Epidemiology
James W. Buehler, MD - Program Manager Leslie E. Lipscomb, MPH, Cheryl Silberman, PhD, MPH Hui Zhang, MD, MPH, Mohamed Qayad, MD, MPH
Preventive Medicine Residents
Mark E. Anderson, MD, MPH & Anthony Fiore, MD, MPH
EIS Officers
Julia Samuelson, RN, MPH, & Keoki Williams, MD
Graphics Dept.
Jimmy Clanton Jr. & Christopher Devoe
Georgia Epidemiology Report Editorial Board
Editorial Executive Committee
Andrew Margolis, MPH - Editor Paul A. Blake, MD, MPH
Jane E. Koehler, DVM, MPH Jeffrey D. Berschling, MPH Kathleen E. Toomey, MD, MPH Angela Alexander - Mailing List Christopher Devoe - Graphics

Stroke is the third leading cause of death in Georgia, accounting for more than 4,200 deaths in 1996. Stroke is also a major cause of hospitalization and disability. In Georgia, there were over 21,000 hospitalizations for stroke in 1996. Nationally, stroke is the leading cause of serious, long-term disability. This report describes the burden of stroke in Georgia. Its purpose is to present a brief overview of stroke death rates in the past two decades and to report the approximate cost of stroke for 1996 in terms of hospitalizations.
Methods

Stroke was defined by the International Classification of Diseases, Ninth Revision, codes 430-438. The source of the number of deaths was the Georgia Division of Public Health. The source of stroke hospitalizations was the 1996 Georgia Hospital Inpatient Data System. The source of the population estimates was the US Bureau of the Census. Stroke death rates were age-adjusted using the direct method; the US 1990 population was used as the standard.
Trends in Stroke Mortality in Georgia

Throughout the 1980s and early 1990s, death rates from stroke declined dramatically in

both Georgia and the US. In 1995, Georgias rate was 19 percent higher than the US rate.

From 1981 to 1993, the average annual decrease of the Georgia stroke death rate was 3.9

percent per year (Figure 1). This decline is presumably due to improvements in medical care

and healthier lifestyles.

In the early 1990s, how-

ever, the trend began to re- Fig. 1 - Stroke Death Rates in Georgia and the US, 1981-1996

verse direction. From 1993 to Deaths per 100,000 population
1996, the Georgia rate actu- 140

ally increased by an average of 0.7 percent per year. This dis- 120
turbing trend occurred in all 100
sex and race groups (Figure 2). This figure also reveals strik- 80

Georgia's death rate decreased by 3.9% per year
Georgia

Georgia's death rate increased by 0.7% per year

ing differences between black 60

and white stroke death rates.

United States

In 1996, the risk of dying 40

from stroke was 37 percent 20 higher for black females than

white females, and 60 percent higher for black males than

0 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
Year

white males. Furthermore,

Source: GA DPH. ICD9 codes f or stroke are 430-438. Rates are age-adjusted to the 1990 US population.

these differences have not changed much over

the past two decades. The reasons for these differences are not well understood, but may be a

result of a higher percentage of blacks with high blood pressure, or factors related to poverty,

such as poor diet or decreased access to health care.

Males have a higher risk than females of dying from a stroke. However, in terms of

absolute numbers, there are actually 50 percent more deaths due to stroke among females than

males. In 1996, there were 1,704 stroke deaths among males, but 2,548 stroke deaths among

females. This difference exists because women live to older ages when strokes are more

common. Females also die in greater numbers than men each year from other cardiovascular

diseases, such as heart disease. However, stroke is not just a disease of old age: One in five

Georgians who died from stroke in 1996 was younger than 65 years of age.

Georgia Department of Human Resources Division of Public Health Epidemiology & Prevention Branch, Epidemiology Section Two Peachtree St., N.W., Atlanta, GA 30303 - 3186 Phone: (404) 657-2588 Fax: (404) 657-2586

Hospitalizations for Stroke in Georgia Conclusions

Each year there are more nonfatal than fatal strokes. Stroke survi-

After years of steady decline, the death rate for stroke is on the

vors are often left with permanent, severe disabilities. We can estimate rise. Black Georgians should be especially concerned because their

the burden of non-fatal strokes in Georgia by examining hospital ad- risk of dying from stroke is particularly high. All Georgians should

missions. In 1996, there were 21,601 hospitalizations1 for stroke be concerned by the discrepancy between the death rates of stroke

among Georgia residents, or a hospitalization for stroke every 24 min- among blacks and whites. It is not possible to explain conclusively

utes. Of these stroke victims, 92.4 percent were discharged alive from reasons for the change in stroke death rates, but two factors are prob-

hospitals. However, after excluding those who were institutionalized ably important contributing causes. First there is either a leveling off

before their stroke, more than one-third of stroke survivors were left or an increase in the percentage of Georgians with unhealthy behav-

unable to care for themselves without assistance. Most of these people iors that put them at risk for a stroke. Second, owing to advances in

required care in a skilled nursing facility or other institution.

medical care, the population of Georgians who have already had a

Hospitalizations for stroke consume considerable health and finan- previous heart attack or stroke is growing. These cardiovascular dis-

cial resources. In 1996, Georgia residents spent 137,931 days in the ease survivors are at particularly high risk for dying from stroke, and

hospital because of stroke, and the charges for these admissions totaled thus contribute to the growing number of stroke deaths.

almost $265 million. The average charge

for a hospital stay was $12,264.

Fig. 2 - Stroke Death Rates in Georgia by Sex and Race, 1981-96

The practical implications of these

Reducing the Risk of Stroke

Deaths per 100,000 population
200

150

Black males

From 1981 to 1993,

From 1993 to 1996,

Sex-race group death rate changed by death rate changed by

Black males Black females White males White females

- 4.1% per year - 3.9% per year - 4.1% per year - 3.6% per year

+ 2.0% per year + 0.5% per year + 1.4% per year + 0.2% per year

findings are clear. Because most strokes result from a process of arterial blockage that begins at an early age, a greater effort should be made to reduce the preva-

The increase in stroke death rates
should alert Georgians to the importance of modifying their personal lifestyles to re- 100

Black females White males

lence of risk factors among all Georgians. Reducing stroke risk factors involves not smoking, eating a healthy

duce the rate of stroke. The following risk

White females

diet, engaging in regular physical activ-

factors can be modified to reduce the risk 50 of stroke:

Death rates decreasing

Death rates increasing

ity, controlling high blood pressure, and reducing blood cholesterol.

Hypertension is a leading risk factor for stroke. The National Institutes of Health have updated their guidelines for diagnosing and treating hypertension.

0 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
Year
Source: GA DPH. ICD9 codes f or stroke are 430-438. Rates are age-adjusted to the 1990 US population.

Sustained behavioral changes are often difficult to make, even for highly motivated individuals. Therefore, it is crucial to engage entire communities in

According to these new guidelines, blood pressure should be main- a process to encourage policy and environmental changes that make it

tained at less than 130/85. The following lifestyle modifications can easier for people to change their behavior. Physicians, community

help keep blood pressure under control and even prevent hypertension leaders and parents can act as role models. Additionally, local envi-

from developing in the first place: 1) Stay active; 2) Eat less fat; 3) Use ronments, such as schools and workplaces, can be changed to facilitate

less salt; and 4) Dont drink too much alcohol.

healthy behaviors.

Tobacco Use. Tobacco is a major preventable cause of stroke. The message is simple: if you use tobacco, stop; if you dont use tobacco, dont start.
Diabetes. Having diabetes can seriously increase the risk of stroke and heart disease. Diabetics can prevent or delay heart and blood vessel disease by controlling their weight, cholesterol, and blood pressure.
Elevated cholesterol. A person with a normal cholesterol level (less than 200 mg/dl), should get their level checked every five years. Eating foods that contain no cholesterol, such as fruits and vegetables, and staying physically active are two easy ways to keep cholesterol low.
1In 1996, data were reported for 75 percent of acute care hospital beds in Georgia. Thus, these statistics represent only about 75 percent of the total number and cost of hospitalizations for stroke. Hospitals with acute care beds which did not report data in 1996 were probably similar to those which did.

Much of the death and disability from stroke in Georgia is preventable. If we focus attention on realistic ways of reducing risk, we can renew the decline in stroke death rates that Georgia has experienced in much of the past two decades. Georgians can work together to reduce the number of individuals who suffer and die from strokes.
For More Information
The complete 1998 Georgia Stroke Report is available from the Georgia Division of Public Health (404-657-2588) and the American Heart Association, Southeast Affiliate (770-952-1316). The report can also be accessed via the Internet at www.ph.dhr.state.ga.us.
Contributors This article was contributed by Tim Dignam, MPH and Alex Rowe, MD, MPH, Chronic Disease and Injury Epidemiology Unit, Epidemiology and Prevention Branch, Division of Public Health, Georgia Department of Human Resources.

Outbreak of Gastrointestinal Illness at Rome Middle School

Background

By Preeti Pathela, MPH & Debbie Abercrombie, RN Notifiable Disease Unit, Epidemiology and Prevention Branch Georgia Division of Public Health, DHR.

On Wednesday, April 29, 1998, a Communicable Disease Specialist, reported to the Georgia Epidemiology and Prevention Branch (EPB) that approximately 40 students at Rome Middle School in Rome, Georgia, had been ill with symptoms of nausea, headache, malaise, vomiting, and diarrhea. The ill persons were part of a group of 64 students and 27 adult chaperones that had traveled to Washington, D.C., on a school trip on April 24-26.
Most reported illnesses occurred late Sunday, April 26, with others occurring in the subsequent few days. The symptoms and suspected incubation period were suggestive of illness caused by Norwalk virus. Suspect meals consumed by the group included dinner on Saturday at a dinner theater in Virginia, and breakfast on Sunday at the hotel.

- 2 -

Methods

Discussion

On April 30, investigators interviewed approximately 30 students at Rome Middle School who had experienced illness about their clinical symptoms as well as food and drinks consumed at the dinner theater. The students were asked to provide stool samples for bacterial and viral testing in order to determine the potential etiologic agent.
Laboratory testing for bacterial pathogens was arranged with the Georgia Public Health Laboratory in Atlanta. Laboratory testing for Norwalk virus was arranged with the Division of Viral Diseases of the Centers for Disease Control and Prevention.

The laboratory investigation showed that the outbreak was caused by a Norwalk-like virus. The symptoms were consistent with this etiologic agent. The incubation period and the occurrence of similar illnesses in the 5 other groups from other states incriminated the evening meal served at the Dinner Theater on April 25. The reports of onset of illness during the evening of April 25, within a few hours after the meal, probably reflect errors in completion of questionnaires or poor recall by the children.
The specific food that caused the outbreak was probably the salad. It could have been contaminated by a food handler in the restaurant,

Results

or one of the ingredients may have been contaminated before it reached the restaurant. Because of the El Nino-associated flooding in

Sixty-four children and 27 adults participated in the school trip to Washington, D.C. Of these, 62 (97%) students and 16 (59%) adults were interviewed. All 78 ate dinner at the Dinner Theater on the evening of April 25. Fifty-one (65%) reported having had at least one symptom after the event. Illnesses were reported to have begun on

California, lettuce was both difficult to obtain and expensive at that time. It is possible that the lettuce came from an unusual source, or that it was flooded and contaminated in the field. Investigation of the restaurant has not found any ill food handlers who helped prepare the suspect meal.

April 25, with the largest numbers occurring during the evening of April 26 and on April 27 (Table 1). The symptoms were headache (88%), nausea (63%), vomiting (55%), cramps (51%), diarrhea (41%),

There are several key elements to consider when conducting a foodborne outbreak investigation.

fever (41%), and chills (37%). A case was defined as an illness in a person 12 to 14 years of age
who experienced vomiting (since it is a memorable and unmistakable symptom) OR who experienced at least two of the following 3 symptoms: nausea, headache, and diarrhea (>3 stools in 24 hours). Thirty-

Questionnaire - Before constructing a questionnaire, general hypothesis-generating interviews should be conducted. These will give enough information about what unique questions need to be put on the questionnaire in addition to demographics, illness onset dates and times, symptom complex, food/drink consumption, treatment, etc. Ideally, questions should

two of these 51 persons met the case definition. The median incubation period for symptoms among the cases was 28 hours and 30 minutes, and the median duration of illness was about 56 hours.

be asked and forms should be filled out by members of the investigating team to maintain uniformity and minimize misinterpretation of the questions. Interviews need to be conducted with both ill and well participants, in order to have a comparison group and thus effectively implicate a

source of illness.

Table 1. Date and Time of Onset of Illness in Participants of the

Specimen collection - A specimen or culture is valuable if it is collected soon in the investigation, kept in the proper container and at the

Rome Middle School Washington D.C. Trip. April 24-May 1, 1998.

right temperature, and transported appropriately. The Georgia Public Laboratory should be consulted. The laboratory can be reached at 404/327-7900
Case definition - Before data analysis, it is important to define a case

Number of Cases 20

CASES ANY SYMPTOM

of illness, as all people with symptoms may not have illnesses relating to the outbreak. For instance, some people report symptoms that are too subjective, or occur too early or too late in the outbreak, and it must be decided if

they are truly cases or not. In this investigation, an age restriction was

15

included in the definition, as students had different food exposures than the

adults.

Rapid investigation - Interview affected persons while events are

10

fresh and collect laboratory specimens quickly while they still harbor the

pathogen. Accurate epidemiologic data can effectively guide an environmen-

5

tal investigation. If all phases of the investigation are successful, a source of disease or vehicle of transmission will be identified, and the mission of public

health (to interrupt disease transmission and/or prevent similar outbreaks in

0

the future) will be accomplished. Investigators in the Epidemiology and Prevention Branch of the

4/25 AM 4/25 PM 4/26 AM 4/26 PM 4/27 AM 4/27 PM 4/28 AM 4/28 PM 4/29 AM 4/29 PM 4/30 AM 4/30 PM

Georgia Division of Public Health are always willing to assist district and

Date and Time

local health authorities in investigations. The EPB is available 24 hours a day to provide consultation by phone and fax regarding epidemiologic and labo-

ratory aspects of an investigation; the contact number is 404/657-2588.

The association of illness with foods eaten was examined for the

32 persons with illnesses which met the case definition (cases), and the

17 persons 12-14 years of age who did not have any symptoms (con-

trols) Three items were significantly associated with illness: salad (odds

ratio [OR]=10.50, p=0.009), spaghetti and meatballs (OR=7.15, p=0.007), and ice in drinks (OR=10.50, p=0.009). However, both the

EPIDEMIOLOGY 101

lettuce and the ice could account for 30 of 32 cases, while the spaghetti could account for only 22 of 32. To further weaken the spaghetti as an implicated vehicle, it was discovered that for the four adults who met the case definition (except for the age limitation), all 4 ate salad, 3

Public Health Surveillance is the on-going, systematic collection, management, analysis, interpretation, and dissemination of data for the purpose of evaluating a public health concern. There are two types of public health surveillance.

consumed ice, and none ate spaghetti. In the laboratory investigation, no Salmonella, Shigella,
Campylobacter, E. coli O157:H7, or Aeromonas bacteria were detected

Passive Surveillance: (provider/lab initiated) The form of data collection where health care providers and laboratories send reports to a health department based upon a known set of rules and regulations.

in the 10 stools cultured for bacterial pathogens. On May 6, CDC reported that electron microscopic examination of 6 stool specimens had detected small round structured viruses (SRSV, or Norwalk-like viruses) in 5 of 6 stool samples tested, and the sixth stool had equivocal

Active Surveillance: (health department initiated) The form of data collection where health departments solicit reports from providers and laboratories that would probably otherwise not have been reported. Active surveillance is usually limited to specific diseases over a defined time period.

results.

- 3 -

The Georgia Epidemiology Report Epidemiology and Prevention Branch Two Peachtree St., NW Atlanta, GA 30303-3186

June 1998

Volume 14 Number 5

Reported Cases of Selected Notifiable Diseases in Georgia Profile* for March 1998

Selected Notifiable Diseases

Total Reported for March 1998
1998

Previous 3 Months Total

Ending in March

1996

1997

1998

Previous 12 Months Total

Ending in March

1996

1997

1998

Campylobacteriosis

28

161

113

116

1019

745

769

Chlamydia genital infection

2411

3257

3284

6120

12113

13631

19029

Cryptosporidiosis

7

13

8

18

118

88

84

E. coli 0157:H7

0

4

14

2

31

49

34

Giardiasis

69

161

214

203

614

873

905

Gonorrhea

1856

5377

4342

4969

21954

18907

19164

Haemophilus influenzae (invasive)

1

21

18

18

40

49

42

Hepatitis A (acute)

58

55

112

148

101

471

801

Hepatitis B (acute)

23

8

31

71

60

84

266

Legionellosis

0

1

0

0

10

2

6

Lyme Disease

0

0

1

2

9

2

10

Meningococcal Disease (invasive)

8

58

36

43

129

125

115

Mumps

0

1

5

0

9

13

6

Pertussis

1

6

8

2

30

37

12

Rubella

0

0

0

0

0

0

0

Salmonellosis

57

261

244

201

1708

1452

1340

Shigellosis

73

132

254

219

1083

1247

1171

Syphilis - Primary

12

55

43

33

287

187

149

Syphilis - Secondary

18

147

97

55

627

453

316

Syphilis - Early Latent

77

369

330

233

1542

1303

1014

Syphilis - Other

74

277

376

211

1138

1154

1048

Syphilis - Congenital

2

14

8

3

65

28

17

Tuberculosis

44

219

189

132

813

761

638

* The cumulative numbers in the above table reflect the date the disease was first diagnosed rather than the date the report was received at the state office; and therefore are subject to change over time due to late reporting. The 3 month delay in the disease profile for a given month is designed to minimize any changes that may occur. This method of summarizing data is expected to provide a better overall measure of disease trends and patterns in Georgia.

** Other syphilis includes latent (unknown duration), late latent, late with symptomatic manifestations, and neurosyphilis.

Report Period

Total Cases Reported *

AIDS Profile Update

Percent Female

Risk Group Distribution (%)

MSM IDU MSM&IDU

HS Blood

Unknown

Race Distribution (%) White Black Other

Latest 12 Months:
06/97 to 05/98 Five Years Ago: 05/92 to 05/93 Cumulative:
7/81 to 05/98

1465 1825 19263

19.2 14.1 15.0

39.8

18.1

3.7

16.0

1.2

21.3

48.5

21.6

6.2

11.8

2.2

9.6

51.1

19.2

5.8

11.9

2.0

10.0

23.5 73.8

2.7

36.1 62.1

1.8

38.9 59.0

2.0

MSM - Men having sex with men

IDU - Injection drug users

HS - Heterosexual

* Case totals are accumulated by date of report to the Epidemiology Section

- 4-