Search

Suicide and other self-directed violence deaths are likely grossly underestimated, reflecting inappropriate classification of many drug intoxication deaths as accidents or unintentional and heterogeneous ascertainment and coding practices across states. As the tide of prescription and illicit drug-poisoning deaths is rising, public health and research needs would be better satisfied by considering most of these deaths a result of self-intoxication. Epidemiologists and prevention scientists could design better intervention strategies by focusing on premorbid behavior. We propose incorporating deaths from drug self-intoxication and investigations of all poisoning deaths into the National Violent Death Reporting System, which contains misclassified homicides and undetermined intent deaths, to facilitate efforts to comprehend and reverse the surging rate of drug intoxication fatalities.

The search for the cause of the large outbreak of severe pneumonia that affected attendees of a convention of war veterans in Philadelphia in 1976 was complicated by a number of factors. The 2-10 day incubation period meant that cases were dispersed across Pennsylvania at time of onset, and matters were further complicated by the lack of a centralised record of attendance of the convention. Known methods for culture, serology, tissue staining, and toxicology gave negative results. The reliance on a clinical case definition, and the need to coordinate local, state, and federal jurisdictions both hindered efforts. Intense public scrutiny, anxiety, and promotion of implausible theories, researchers' memories of prior unsolved major investigations, and the inherent difficulty of documenting airborne spread also had an impact. The novel bacterial agent Legionella pneumophila was identified and its spread defined by persistent application of basic epidemiological and laboratory principles, including the willingness to re-examine one's assumptions.

As of April 30, 1980, 83 nosocomial cases of sporadic legionellosis had been reported to the Centers for Disease Control (CDC). In all 83 cases the patients had pneumonia; the median age of the patients was 56.5 years. All but one patient were hospitalized at the time of onset. Of 71 patients for whom the outcome is known, 22 (31 percent) died of causes directly attributed to their infection. Eleven patients had end-stage renal disease, 28 were receiving systemic immunosuppressive medications, 17 had cancer, 12 had chronic bronchitis or emphysema, 29 were smokers, and four had diabetes mellitus. Risks of acquiring nosocomial sporadic legionellosis for patients with these conditions relative to the general United States population = 340, 26, 11, 3.7, 1.9 and 1.3, respectively. These risk factors are similar to those identified for sporadic community-acquired legionellosis and for epidemic nosocomial legionellosis. Methods for preventing nosocomial legionellosis are not known, but comparing Legionella to other water-associated organisms which have been spread from medical devices to cause pneumonia may be fruitful.

As of 30 September 1979, 1005 confirmed cases of sporadic legionellosis caused by Legionella pneumophila serogroups 1 to 4 in U.S. residents had been reported to the Centers for Disease Control; 19% were fatal. All but 2% of the 1005 cases were associated with pneumonia documented by chest radiograph. About 75% of the cases occurred in June through October. The risk of acquiring sporadic legionellosis was increased among males and persons 50 years or older; persons with renal disease necessitating dialysis or transplantation, with chronic bronchitis or emphysema, with diabetes mellitus, and with cancer (10 selected sites or types); persons who smoke; and persons being treated with immunosuppressive drugs. Increasing age and chronic bronchitis or emphysema were associated with increased risk of death. The sensitivity of culturing L. pneumophila from specimens positive by direct immunofluorescence was estimated to be 45%. The distribution of serogroups 1, 2, 3, and 4 of L. pneumophila in 57 fresh, not previously examined direct fluorescent antibody-positive specimens was 84%, 11%, 4%, and 2%, respectively; all 26 strains isolated from these specimens were of one of these four serogroups.

Epidemiology has features that resemble those of the traditional liberal arts. This makes it fit both for inclusion in an undergraduate curriculum and as an example in medical school of the continuing value of a liberal education. As a "low-technology" science, epidemiology is readily accessible to nonspecialists. Because it is useful for taking a first look at a new problem, it is applicable to a broad range of interesting phenomena. Furthermore, it emphasizes method rather than arcane knowledge and illustrates the approaches to problems and the kinds of thinking that a liberal education should cultivate: the scientific method, analogic thinking, deductive reasoning, problem solving within constraints, and concern for aesthetic values.

The efficacy of pneumococcal vaccine in groups of patients in the United States at high risk for pneumococcal disease was estimated by comparing distributions of serotypes of Streptococcus pneumoniae isolated from vaccinated and unvaccinated persons. Between May 1978 and March 1984, 187 blood isolates and 62 cerebrospinal fluid isolates from vaccinated patients, and 1447 blood isolates and 191 cerebrospinal fluid isolates from unvaccinated patients were serotyped at the Centers for Disease Control. The study did not include patients who were less than 2 years old or who had Hodgkin's disease, multiple myeloma, or immunoglobulin deficiency. In patients with bacteremic disease, the overall efficacy of pneumococcal vaccine was estimated at 64% (95% confidence limits, 47% to 76%); efficacy did not differ significantly with age. In persons over 65 years of age with diabetes mellitus, chronic heart disease, pulmonary disease, or no underlying illnesses, efficacy was 61% (95% confidence limits, 1% to 85%). These findings support the use of pneumococcal vaccine in selected populations in the United States.

Programs administered by health departments and heart associations for control of streptococcal and poststreptococcal disease differ greatly from state to state. The year 1974 was used as a representative year when studying the variety of state policies and programs regarding surveillance of streptococcal disease, primary and secondary prevention of rheumatic fever, and prevention of poststreptococcal glomerulonephritis. A uniform national policy for streptococcal control is desirable but should be based on evidence not now available concerning the impact of various elements of control programs on morbidity and mortality.

From 1977 to 1981, 18,642 cases of bacterial meningitis were reported to the Centers for Disease Control. We analyzed data from 27 states with full participation from 1978 through 1981. Hemophilus influenzae was the most frequent cause of bacterial meningitis (48.3%), followed by Neisseria meningitidis (19.6%) and Streptococcus pneumoniae (13.3%). Overall attack rates for males were greater than for females (3.3 v 2.6 cases per 10(5) population per year). Attack rates were highest in children under 1 year of age (76.7 per 10(5) population per year). Case-fatality ratios were highest for gram-negative and miscellaneous causes of bacterial meningitis (33.7%) and lowest for meningitis caused by H influenzae (6.0%). Neisseria meningitidis and S pneumonia meningitis occurred preponderantly during the winter, while H influenzae meningitis had peak activity in the spring and fall. Ampicillin resistance among H influenzae increased from 18.7% in 1978, to 23.9% in 1981. Serogroup B Neisseria meningitidis was the most common serogroup identified during the reporting period (51.1%), followed by serogroup C (22.3%), serogroup Y (5.8%), and serogroup A (4.7%) infections.

Two outbreaks of peritonitis caused by a Mycobacterium chelonei-like organism--a previously unrecognized pathogen--occurred among patients receiving intermittent chronic peritoneal dialysis (CPD). In one center, five of 22 patients who had undergone CPD during a one-month period developed peritonitis caused by an M. chelonei-like organism acquired from a single contaminated automated CPD machine. In a second center, five of eight patients who had received CPD during a several-week period became infected, apparently as a result of cross-infection from contaminated machines. Seven sporadic cases of peritonitis due to an M. chelonei-like organism were also found. M. chelonei-like organisms can survive and proliferate in water and are relatively resistant to formaldehyde. Defects in the design of CPD machines and disinfection procedures were identified that may have permitted M. chelonei-like organisms to survive attempted disinfection.

Using decision analysis, we estimated the benefits, risks and costs of routine childhood immunization against pertussis. Without an immunization program, we predict that there would be a 71-fold increase in cases and an almost fourfold increase in deaths (2.0 to 7.6) per cohort of one million children. With a vaccination program, we predict 0.1 case of encephalitis associated with pertussis and five cases of post-vaccination encephalitis; without a program, there would be only 2.3 cases of encephalitis associated with pertussis. Community vaccination would reduce by 61 per cent the costs related to pertussis. Our analysis supports continuation of vaccination in routine childhood immunization programs, but suggests the need for more reliable data on complications from the vaccine, further study of the epidemiology of pertussis and development of a less toxic vaccine.

To determine the risk of severe Haemophilus influenzae illness among household contacts of patients with H. influenzae meningitis, we studied prospective data obtained in 19 states from January 1, 1977, to June 30, 1978. H. influenzae meningitis was reported in 1403 patients, and 1147 (82 per cent) of the exposed families were investigated for the occurrence of H. influenzae disease within 30 days after its onset in the index patient. During this interval, nine of 1687 household contacts (0.5 per cent) under the age of six years had systemic disease confirmed to be caused by H. influenzae Type b. The risk in children less than one year of age was 6 per cent, and the risk in those less than four years of age was 2.1 per cent. None of 2624 contacts above the age of five was affected. In the 30 days after onset of meningitis, the risk of this infection alone, aside from other types of serious H. influenzae disease, is 585 times greater in household contacts than the age-adjusted risk in the general population. The risk of H. influenzae disease in household contacts under six years of age is similar to the risk of secondary meningococcal disease in all household contacts--indicating a need for effective antimicrobial prophylaxis.