Search

Since human immunodeficiency virus (HIV) transmission from a dentist to six of his patients was first reported in 1990 by the Florida Department of Health and Rehabilitative Services and the Centers for Disease Control and Prevention, controversy and speculation have surrounded the investigation of that case. This controversy has been fueled by the inability to determine exactly how the transmissions occurred. Many theories have appeared in the media and have led to confusion and uncertainty about the facts of this investigation. Recently, a magazine article and a newspaper article, as well as a segment on the television newsmagazine "60 Minutes," presented information that was largely based on findings by investigators hired as part of private litigation and that cast doubt on the conclusion that the patients had been infected by the dentist. However, these reports omitted pertinent epidemiologic and laboratory evidence that shows that no other sources of HIV infection could be documented for the six dental patients. The scientific evidence indicates that the Florida dentist transmitted HIV to six of his patients. [ABSTRACT FROM AUTHOR]

Objective: To investigate why women who use crack cocaine are at increased risk of human immunodeficiency virus (HIV) infection., Methods: One thousand one hundred fifty-two (99.7%) of 1155 consecutive prenatal patients attending a rural public health clinic were interviewed about drug use and sexual practices and tested for HIV infection and other sexually transmitted diseases., Results: Fifty-one (4.7%) of 1096 pregnant women reported ever using crack cocaine, but only five (10%) of the crack cocaine users had ever injected drugs. Eighteen (35%) of the crack users were HIV infected compared with 22 (2%) of the 1045 women who reported never using crack (odds ratio 25, 95% confidence interval 12-52; P < .001). Crack users were more likely to have had a known HIV-infected sex partner, exchanged sex for money or drugs, and tested positive for syphilis than were non-crack users (for each comparison, P < .001). Before using crack, 18% of crack users had exchanged sex for money or drugs and 8% had averaged three or more sex partners per month; in contrast, after beginning to use crack, 76% of crack users exchanged sex for money or drugs and 63% averaged three or more sex partners per month (for both comparisons, P < .001). Crack users who were not HIV infected were more likely to have almost always used condoms and/or had fewer than three sex partners per month than were HIV-infected crack users (P < .01)., Conclusion: Women who reported using crack cocaine were at an increased risk of HIV infection because crack use was associated with a significant increase in unprotected sexual contact.

During the second half of the 1980s, Miami had a syphilis epidemic while gonorrhea rates decreased. To determine whether the direction of these trends truly differed within all population subgroups or whether they resulted from aggregating groups within which trends were similar, records from four sexually transmitted disease clinics from 1986 to 1990 and census data from 1990 were used to compare race-, sex-, age-, and zip code-specific groups. Syphilis and gonorrhea clustering was similar; 50% of cases occurred in the same zip codes, representing 10% of the population. In all groups, gonorrhea decreased (aggregate 48%) while syphilis first increased (aggregate 47%) and then decreased. Determining reasons for these different trends may facilitate controlling these diseases.

Objective: To determine whether dentist-to-patient or patient-to-patient transmission of human immunodeficiency virus (HIV) occurred in the practice of a dentist who had the acquired immunodeficiency syndrome (AIDS)., Design: Retrospective epidemiologic investigation supported by molecular virology studies., Setting: The practice of a dentist with AIDS in an area with a high AIDS prevalence., Participants: A dentist with AIDS, his former employees, and his former patients, including 28 patients with HIV infection., Measurements: Identification of potential risks for acquisition of HIV infection, genetic relatedness among HIV strains, and infection-control practices., Results: A dentist with known behavioral risks for HIV infection, who was practicing in an area of Miami, Florida, that had a high rate of reported AIDS cases, disclosed that he frequently did invasive procedures and did not always follow recommended infection-control procedures. Of 6474 patients who had records of receiving care from the dentist during his last 5 years of practice, 1279 (19.8%) were known to have been tested for HIV infection and 24 of those (1.9%) were seropositive. Four other patients with HIV infection were identified through additional case-finding activities. Of these 28 patients with HIV infection, all but 4 had potential behavioral risk factors for infection. Phylogenetic tree analysis of HIV genetic sequences from the dentist and 24 of the patients with HIV infection showed an absence of strong bootstrap support for any grouping and therefore did not indicate that the virus strains were linked., Conclusions: Despite identifying numerous patients with HIV infection, we found no evidence of dentist-to-patient or patient-to-patient transmission of HIV during dental care. Our findings are consistent with those of all previous studies in this area, with the exception of one that did identify such transmission.

Objective: To study rates of documented HIV seroconversion and syphilis as a cofactor for seroconversion in sexually transmitted disease (STD) clinics. In the main clinic the HIV seroprevalence rate was 12% and most infections had been shown to be acquired by heterosexual contact., Methods: We analyzed computer records of patients who had at least two HIV-antibody tests between 1 December 1987 and 31 December 1990, at STD clinics in Dade County (Miami), Florida., Results: Of 5164 individuals with two HIV tests, 208 (4.0%) seroconverted. The overall seroconversion rate was 3.1 per 100 person-years. Among blacks, who accounted for 77% of seroconversions, the rate was higher for women (4.8) than for men (2.7). The highest rate was in 15-19-year-old black women (7.1 per 100 person-years). The HIV seroconversion rate was 12.8 for patients with primary or secondary syphilis diagnosed between two HIV tests, 3.1 for patients who acquired syphilis before their first HIV test, and 2.3 for patients who had never had syphilis. Eighteen per cent of all HIV seroconversions were attributable to syphilis acquired in the interval between two HIV tests., Conclusions: We found high HIV seroconversion rates, especially among black teenagers and black women, in an STD clinic population in which the majority of HIV infections were shown previously to have been acquired heterosexually. Syphilis was a marker for HIV seroconversion and syphilitic ulcers may facilitate HIV transmission. Innovative prevention programs directed towards women and adolescents should be developed and evaluated.

Florida ranked fourth in the nation with 1,707 tuberculosis cases reported in 1992 for a rate of 12.7 per 100,000 population. Thirteen percent of these patients had AIDS. Recent cases in prisons, shelters, hospitals and schools have stimulated interest and media coverage. Resurgence of strains of multiple-drug resistant tuberculosis is a serious concern. The Florida Department of Health and Rehabilitative Services, in collaboration with allied agencies, has utilized several initiatives in response. The most significant, Tuberculosis Epidemic Containment Plan, details intervention strategies needed to eliminate TB in the state by the year 2010. Successful implementation depends upon local TB prevention and control coalitions that include private and public sector providers.

Acquired immunodeficiency syndrome (AIDS) cases reported as the result of heterosexual contact have been increasing in the United States, with Florida reporting a disproportionate number. We investigated 168 such AIDS cases from southern Florida. After follow-up, 50 (30%) patients were reclassified into other transmission categories. The data suggest that the increased rate of heterosexually acquired AIDS cases reported from southern Florida was partially related to misclassification of risk.

Objectives: The effects of posttest counseling on acquisition of sexually transmitted diseases in patients at a large urban sexually transmitted disease clinic were studied., Methods: Comparisons were made of the percentage of patients who had a positive gonorrhea culture (or any sexually transmitted disease) in the 6 months before and after human immunodeficiency virus (HIV) counseling and testing., Results: For 331 patients counseled about a positive HIV test, the percentage with gonorrhea was 6.3 before and 4.5 after posttest counseling (29% decrease). For 666 patients counseled about a negative test, the percentage with gonorrhea was 2.4 before and 5.0 after posttest counseling (106% increase). With any sexually transmitted disease as the outcome, patients who tested positive for HIV had a 12% decrease and patients who tested negative had a 103% increase after counseling., Conclusions: HIV counseling and testing was associated with a moderate decrease in sexually transmitted diseases among patients who tested positive for the virus, but risk increased for patients who tested negative. This suggests a need to improve posttest counseling in this clinic and to assess the effects of counseling and testing in other clinics.

Objectives: The purpose of this study was to develop a method to identify persons at high risk for acquiring new sexually transmitted infections., Methods: Computerized medical records from sexually transmitted disease clinics in Dade County, Florida, were used to conduct a retrospective cohort study. For all patients who visited in 1987, risk factors were identified for returning to the clinics within a year with a new sexually transmitted infection. Predictor variables were derived from the index visit and any visits in the year prior to the index visit. Logistic regression was used to develop a model that was applied to all patients who attended in 1989., Results: Of 24,439 patients attending in 1987, 18.5% returned within a year with a new infection. Return rates were highest for 15- to 19-year-old Black males (31.8%). The highest odds ratios for returning were a diagnosis or treatment for an infection in the previous year and a diagnosis or treatment for infection at the index visit. The patients predicted to be at highest risk had a 39% return rate. There were as many new infections among the 2893 patients at highest risk as there were among the 13,326 patients at lowest risk., Conclusions: We developed a model that identifies persons at very high risk for sexually transmitted infection. These persons should be targeted for intensive intervention to reduce their risk.

Impact of the human immunodeficiency virus among women in Florida is increasingly recognized in the second decade of the AIDS epidemic. Illicit drug use, sexual activity with multiple partners, lack of access to health care, and socioeconomic disadvantages are greatly contributing to the increase of female AIDS cases. Current salient features of the epidemic are summarized as well as prevention strategies to reduce transmission of the virus.

Background: In the United States, an increasing proportion of women infected with the human immunodeficiency virus (HIV) live in nonmetropolitan areas. Little is known, however, about the risk factors for HIV transmission in women outside large cities., Methods: We interviewed and tested 1082 (99.8 percent) of 1084 consecutive pregnant women who registered for prenatal care at a public health clinic in western Palm Beach County, Florida. This rural agricultural area of about 36,000 people is known to have a high prevalence of HIV infection., Results: The seroprevalence of HIV was 5.1 percent (52 of 1011 women). Black women who were neither Haitian nor Hispanic had the highest rate of infection (8.3 percent [48 of 575]). Only 4 of 1009 women (0.4 percent) reported ever injecting drugs, and the 4 were HIV-seronegative; however, 14 of 43 users of "crack" cocaine (33 percent) had HIV infection. At prenatal registration, 131 of 983 women (13 percent) tested positive for gonorrhea, chlamydial infection, or syphilis. By multivariate logistic-regression analysis, HIV infection was found to be independently associated with having used crack cocaine (odds ratio, 3.3; P < 0.001), having had more than two sexual partners (odds ratio, 4.6; P < 0.001), being black but neither Hispanic nor Haitian (odds ratio, 11; P < 0.001), having had sexual intercourse with a high-risk partner (odds ratio, 5.6; P < 0.001), and testing positive for syphilis (odds ratio, 3.1; P = 0.015). Nevertheless, 11 of the 52 HIV-infected women (21 percent) reported a total of only two to five sexual partners and no known high-risk partners, had never used crack cocaine, and had no positive tests for sexually transmitted disease., Conclusions: In the rural community we studied, most of the women with HIV infection acquired it through heterosexual contact. The increasing seroprevalence of HIV and the increasing incidence of syphilis and use of crack cocaine mean that other women may be at similar risk of acquiring heterosexually transmitted HIV infection.

Objective: To describe the factors underlying an increasing incidence of tuberculosis in children., Design: Descriptive case review., Setting: Palm Beach County, Fla., Participants: Forty-four children with suspected and confirmed pediatric tuberculosis from 1985 through 1989., Interventions: None., Measurements/main Results: From 1988 through 1989, tuberculosis was confirmed in 15 children and suspected in another 16 compared with data from 1985 through 1987 in which the disease was confirmed in nine children and suspected in four. Pediatric tuberculosis occurred primarily in blacks younger than 5 years; the increase in the number of cases reported in 1988 and 1989 occurred only in blacks. One child in whom tuberculosis was confirmed during the recent period was infected with the human immunodeficiency virus (HIV); however, among children with suspected tuberculosis, four of the nine children tested were seropositive for HIV. There was no evidence of increased transmission of tuberculosis to children by HIV-seropositive adults compared with transmission by HIV-seronegative adults with TB. New adult tuberculosis cases in the county increased annually, from 92 cases in 1986 to 169 in 1989, of whom at least 36% were infected with HIV., Conclusions: The largest effect of the HIV epidemic on tuberculosis in children appeared to be indirect, through an increase in the number of adults with active tuberculosis serving as potential sources of tuberculosis infection for children. A direct effect of HIV infection in the progression of tuberculous disease in children is likely, but was not detected in this investigation. Case-finding for tuberculosis among children will need to increase, particularly in areas heavily affected by acquired immunodeficiency syndrome, but may be complicated by the difficulty of diagnosing tuberculosis in HIV-infected children.

Florida is third in the nation in the cumulative number of AIDS cases. The epidemic in the state has grown steadily since 1981, clustering initially in metropolitan South Florida but dispersing to other urban and nonurban areas. Heterosexually transmitted AIDS is fast increasing and is more than twice as common in Florida than the United States; however, homosexual/bisexual men and intravenous drug-using men and women continue to represent the majority of new cases. Special serologic studies of the prevalence of HIV infection provide a timely view of the dynamics and characteristics of the epidemic in various segments of the state's population. Data from a large, ongoing serosurvey suggest that about one in every 220 childbearing women is HIV-infected, a finding consistent with Florida's rank of second nationally in the number of pediatric AIDS cases.

Pneumonia and influenza (P & I) constitute Florida's sixth leading cause of death. The P & I death rate in 1987, 10.5 per 100,000, was the highest since 1978. Major target groups for one or both vaccines used in prevention, as recommended by the Immunization Practices Advisory Committee (ACIP), include persons with chronic diseases of the heart or lungs, residents of nursing homes and other chronic care facilities, and persons aged 65 and older. Despite well-defined recommendations, vaccine coverage rates in Florida are as low as 30% in persons greater than or equal to 65 years of age. Knowledge and attitude surveys demonstrate that low coverage among various population groups may be due largely to insufficient awareness and/or negative attitudes regarding pneumococcal and influenza vaccines. Conversely, recommendations by physicians and other health care providers are strongly associated with receiving either vaccine. If the incidence of P & I is to decrease substantively in Florida, much wider use of the vaccines must occur. Because so many high-risk patients depend on private physicians for health care, their role is critical to the success of Florida public health strategies to reverse P & I trends.

In 1984, a press release by a Miami, Florida, neurologist described a possible cluster of persons with multiple sclerosis in Key West, Florida. The authors examined the cluster using prevalence rates, which are recognized as having a latitudinal gradient for multiple sclerosis, being generally high at high latitudes and low at low latitudes. Case ascertainment showed 32 definite or probable cases among residents of the study area (latitude, 24.5 degrees N) on September 1, 1985, a prevalence rate of 70.1/100,000 population--14 times the rate estimated for this latitude by modeling techniques based on US and international data, 7-44 times the rate for areas at similar latitudes (Mexico City, Mexico; Hawaii; New Orleans, Louisiana; and Charles County, South Carolina), and 2.5 times the expected rate for all US latitudes below 37 degrees N. This finding could not be explained by changes in diagnostic criteria, case ascertainment bias, immigration of people from high-risk areas, an unusual population structure, a large percentage of related cases, or better survival. Prevalent cases (n = 22) were more likely than general population controls (n = 76), matched by sex and 10-year age group, to have: lived longer in Key West, been a nurse, ever owned a Siamese cat, had detectable antibody titers to coxsackievirus A2 and poliovirus 2, and ever visited a local military base (Fleming Key). Key West has an unusually high prevalence of multiple sclerosis that may be related to these risk factors.

A long-term comparative field trial of three live, attenuated rubella vaccines (HPV-77 DE-5, HPV-77 DK-12, and Cendehill) was initiated in 1969 on the islands of Kauai and Hawaii in the state of Hawaii. Rubella hemagglutination-inhibition (HI) tests on prevaccination serum specimens from 7,931 children in the two study areas indicated an overall susceptibility to rubella of nearly 70%. The rates of seroconversion of 5,153 seronegative subjects to HPV-77 DE-5, HPV-77 DK-12, and Cendehill vaccine were 97.5%, 99.9%, and 99.8%, respectively. Over the subsequent four-year follow-up period, during which time natural exposure to rubella was minimal, the percent decline of geometric mean titers did not vary substantially among the three vaccine groups and measured about twofold for all three. A total of only 28 vaccines (0.7%) who seroconverted to one of the vaccines in 1969 lost all measurable antibody by 1974. Measurable antibody persisted in more than 98% of all vaccinees over the four-year period. Reinfection, thought possibly to be an important factor in maintaining titers, did not occur frequently in the study population and could not be related to outbreaks of disease.

The high cumulative incidence of AIDS and the large percentage of AIDS patients with no identified risks in Belle Glade, Florida, were evaluated through case interviews and neighborhood-based seroepidemiologic studies. It was found that of 93 AIDS patients reported between July 1982 and 1 August 1987, 34 could be directly linked to at least one other AIDS patient or to a person with AIDS-related complex by sexual contact, sharing of needles during intravenous drug abuse (or both), or perinatal exposure; of 877 randomly selected adults, 28 had antibodies to HIV; no person over age 60 and none of 138 children aged 2 to 10 years had antibodies to HIV; no clustering of infected persons within households occurred, except in sex partners; and HIV-seropositive adults were more likely than HIV-seronegative adults to be from Haiti, have a lower income, report sex with intravenous drug abusers, and have a history of previous treatment for sexually transmitted diseases. The presence of antibodies to five arboviruses prevalent in South Florida or the Caribbean did not correlate significantly with HIV infection. The high cumulative rate of AIDS in Belle Glade appears to be the result of HIV transmission through sexual contact and intravenous drug abuse; the evidence does not suggest transmission of HIV through insects.

Three years after receiving rubella vaccine, 1,060 elementary school children living on the island of Maui, Hawaii, were revaccinated with either HPV-77 DE-5 or RA 27/3 rubella vaccine given subcutaneously or intranasally in order to compare the effectiveness of these two vaccines in raising antibody titers. RA 27/3 was the more effective booster vaccine, producing fourfold or greater titer rises in 20.1% of recipients, including 80% of children with hemagglutination-inhibiting antibody titers less than or equal to 1:40 at the time of revaccination, intranasal revaccination was not significantly more effective than subcutaneous revaccination, although it did elicit higher titers in children who responded. Responses differed according to the vaccine that children had received three years earlier. Because antibody titers have persisted in vaccinated children, routine administration of a second dose of rubella vaccine is not currently recommended.

The surveillance of clinical cases of acquired immune deficiency syndrome (AIDS) in the United States began in June 1981 when the first case was reported. Since then, state and federal public health officials have continuously monitored progression of the epidemic by the number of persons reported with diagnosed clinical AIDS. The human immunodeficiency virus (HIV) that causes AIDS was identified in 1983 and an antibody test was licensed by the Food and Drug Administration in 1985. Recent studies have shown that the time from HIV infection to development of clinical AIDS is an average of seven or more years. Consequently, the reported cases of clinical AIDS reflect the severity of the epidemic an average of seven years ago or more, not now, and certainly not in the future. The AIDS epidemic is in reality an HIV epidemic. The number of persons with HIV infections is a better measure of the present status and future course of the disease. This paper discusses the development of HIV prevalence and incidence studies and illustrates the use of these data to predict the future number of persons with clinical AIDS and the economic impact of the epidemic.

National morbidity figures show a decline in reported rubella and congenital rubella syndrome since 1969, concurrent with widespread use of rubella vaccine. In addition, no nationwide outbreak, such as the 1963-1964 epidemic, has occurred, though on the basis of long-term secular trends, one would be expected between 1970 and 1974. Recent rubella outbreaks have occurred in unimmunized students in high schools and universities, and there appears to have been a slight upward shift in the age-specific incidence of rubella in the United States since the beginning of widespread immunization. Currently available vaccines have provided durable protection to date, and, although reinfection is known to occur following vaccination, it has not proven a risk to the pregnant woman. There is a small but significant incidence of adverse reactions and a potential risk to the woman who is vaccinated during pregnancy. These data indicate that rubella vaccines are safe and effective. They also imply that rubella vaccines, as they are currently applied, have been successful in reducing the morbidity of congenital rubella syndrome, although continued surveillance will be necessary to confirm this trend.

Florida reported 1858 cases of the acquired immunodeficiency syndrome (AIDS) and 8455 cases of tuberculosis from January 1, 1981, through October 31, 1986. Of the patients with AIDS, 159 (8.6%) also had tuberculosis, and 154 (1.8%) of the patients with tuberculosis also had AIDS. Among patients with both diagnoses, tuberculosis was diagnosed before AIDS by more than 1 month in 50%, was diagnosed within 1 month before or 1 month after the diagnosis of AIDS in 30%, and was diagnosed more than 1 month after the AIDS diagnosis in 20%. Compared with patients with AIDS only, patients with both diagnoses were also more likely to be Haitian, black (other than Haitian), or Hispanic. Compared with patients with tuberculosis only, patients with both diagnoses were more likely to be younger, male, Haitian, black (other than Haitian), and Hispanic, have extrapulmonary tuberculosis and negative tuberculin skin tests, and have noncavitary chest roentgenograms. These data suggest that patients with AIDS may have an increased risk of tuberculosis and that patients with both diagnoses differ in important demographic and clinical characteristics from patients with AIDS only or tuberculosis only.

In May 1983, an estimated 865 cases of epidemic gastrointestinal disease occurred in Greenville, Florida. Surveillance of pharmacy sales of antidiarrheal medicines suggested that the outbreak was confined to Greenville and its immediate vicinity. Surveys demonstrated that the gastrointestinal illness attack rates inside and outside the city limits were 56 per cent (72/128) and 9 per cent (7/77), respectively (relative risk (RR) = 6.2); consumption of city water was associated with illness (RR = 12); and as water consumption increased, the attack rate also increased (p less than 0.001). Four adults were hospitalized and one outbreak-related case of Guillain-Barre' syndrome was identified. Campylobacter jejuni was isolated from specimens from 11 ill persons; serologic studies showed the development of Campylobacter-specific antibodies. Fecal coliforms were found in water samples, but Campylobacter was not recovered from water. The city water plant, a deep well system, had numerous deficiencies including an unlicensed operator, a failure of chlorination, and open-top treatment towers. Birds were observed perching on the open-top treatment tower. Of 38 birds trapped seven weeks later, 37 per cent harbored C. jejuni; however, plasmid and serotyping studies showed that strains were not the same as the common strain from ill persons. This outbreak suggests that water systems that are unprotected from contact with birds may become contaminated and a source of outbreaks of human campylobacteriosis.

Implementation of the rubella immunization policy in the United States has resulted in marked decreases in the number of reported cases of both rubella and CRS. Emphasis on immunization of one- to 12-year-old children has resulted in a dramatic reduction in reported rubella in this age group. Accordingly, the proportion of reported cases occurring in adolescents and young adults has increased. Therefore, it seems appropriate to place more emphasis on immunizing susceptible adolescent and adult females as an adjunct to routine immunization in early childhood. This joint strategy should result in a further decrease in the incidence of rubella and CRS.

Immunodeficient patients who were presumed to be susceptible received zoster immune globulin prophylaxis after exposure to varicella. The highest clinical attack rate (35.9%) was seen in household contacts; the lowest attack rate (0%) was observed in children exposed at school. Among household contacts, 48 of 100 patients who received high titer ZIG (reciprocal complement fixation titer greater than or equal to 2,560) developed fourfold rises in serum CF antibody between pre- and 48-hour post-treatment specimens, compared to only one of 34 patients treated with lower titer ZIG lots (P less than 0.001). Patients who developed fourfold antibody rises were significantly less likely to contract clinical varicella (P less than 0.01). Patients who received high titer ZIG also had significantly lower risks of death (P = 0.025) and complications (P = 0.006). Among ZIG-treated patients who contracted clinical varicella, 80% developed mild disease (less than 100 pox), and the median incubation period was prolonged. Immunodeficient children exposed to varicella benefit from ZIG prophylaxis and higher titer ZIG is of greatest benefit.

Five years after primary infant immunization with trivalent oral poliovirus vaccine, employing either a three-dose primary series as recommended by the U.S. Public Health Service Advisory Committee on Immunization Practices (ACIP) or a four-dose series as recommended by the Committee on Infectious Diseases of the American Academy of Pediatrics. 115 children were serologically tested for persistence of neutralizing antibodies by the microneutralization test. Of the 57 individuals immunized according to the ACIP recommendation, antibody persistence was demonstrated in 92% for type 1 poliovirus, 98% for type 2, and 84% for type 3. Of those 58 individuals originally receiving a four-dose primary infant immunization series, the persistence of antibody was 98% to type 1, 98% to type 2, and 87% to type 3. Twenty-one of 24 negative sera showed neutralizing ability when tested by a more sensitive plaque reduction test. Thus, individuals completing either immunization schedule demonstrated satisfactory persistence of neutralizing antibody to all three poliovirus types over a five-year period.

Histories obtained in 350 of 375 clinically cofirmed cases of subacute sclerosing panencephalitis (SSPE) reported to a national registry showed that 292 patients had measles and 58 had no history of measles. Forty of the latter patients received live, attenuated measles virus vaccine. In patients with a history of measles, measles illness occurred before age 2 years in 46%, and a mean of 7.0 years before onset of SSPE. In contrast, there was no relationship of SSPE with age at vaccination in 35 of the 40 patients historically associated with measles vaccine, and SSPE occurred a mean of 3.3 years after vaccination. Based on estimated national measles morbidity data and national measles vaccine distribution data, the risk of SSPE following measles vaccination (0.5 to 1.1 cases/106) appears to be less than the risk following measles (5.2 to 9.7 cases/106). Because live measles vaccine is highly effective in preventing measles illness and a high proportion of children in the United States have received measles vaccine, these data are consistent with the observed downward trend in SSPE incidence since 1969.