The reporting efficiency of measles by hospitals in Los Angeles County, 1986 and 1989

Anticipating epidemic transitions with imperfect data

Epidemic transitions are an important feature of infectious disease systems. As the transmissibility of a pathogen increases, the dynamics of disease spread shifts from limited stuttering chains of transmission to potentially large scale outbreaks. One proposed method to anticipate this transition are early-warning signals (EWS), summary statistics which undergo characteristic changes as the transition is approached. Although theoretically predicted, their mathematical basis does not take into account the nature of epidemiological data, which are typically aggregated into periodic case reports and subject to reporting error. The viability of EWS for epidemic transitions therefore remains uncertain. Here we demonstrate that most EWS can predict emergence even when calculated from imperfect data. We quantify performance using the area under the curve (AUC) statistic, a measure of how well an EWS distinguishes between numerical simulations of an emerging disease and one which is stationary. Values of the AUC statistic are compared across a range of different reporting scenarios. We find that different EWS respond to imperfect data differently. The mean, variance and first differenced variance all perform well unless reporting error is highly overdispersed. The autocorrelation, autocovariance and decay time perform well provided that the aggregation period of the data is larger than the serial interval and reporting error is not highly overdispersed. The coefficient of variation, skewness and kurtosis are found to be unreliable indicators of emergence. Overall, we find that seven of ten EWS considered perform well for most realistic reporting scenarios. We conclude that imperfect epidemiological data is not a barrier to using EWS for many potentially emerging diseases.

Subacute Sclerosing Panencephalitis: The Devastating Measles Complication That Might Be More Common Than Previously Estimated

Background

Subacute sclerosing panencephalitis (SSPE) is a fatal complication of measles. We reviewed California cases from 1998-2015 to understand risk factors for SPPE and estimate incidence.

Methods

SSPE cases had clinically compatible symptoms and measles antibody detection in cerebrospinal fluid (CSF) or medical record documentation of SSPE. Cases were identified though a state death certificate search, Centers for Disease Control and Prevention reports, or investigations for undiagnosed neurologic disease. Measles detection in CSF was performed by serology at the California Department of Public Health or at clinical laboratories.

Results

Seventeen SSPE cases were identified. Males outnumbered females 2.4:1. Twelve (71%) cases had a history of measles-like illness; all 12 had illness prior to 15 months of age. Eight (67%) children were exposed to measles in California. SSPE was diagnosed at a median age of 12 years (3-35 years), with a latency period of 9.5 years (2.5-34 years). Among measles cases reported to CDPH during 1988-1991, the incidence of SSPE was 1:1367 for children <5 years, and 1:609 for children <12 months at time of measles disease.

Conclusions

SSPE cases in California occurred at a high rate among unvaccinated children, particularly those infected during infancy. Protection of unvaccinated infants requires avoidance of travel to endemic areas, or early vaccination prior to travel at age 6-11 months. Clinicians should be aware of SSPE in patients with compatible symptoms, even in older patients with no specific history of measles infection. SSPE demonstrates the high human cost of "natural" measles immunity.

The Costs and Valuation of Health Impacts of Measles and Rubella Risk Management Policies

National and global health policymakers require good information about the costs and benefits of their investments in measles and rubella immunization programs. Building on our review of the existing measles and rubella health economics literature, we develop inputs for use in regional and global models of the expected future benefits and costs of vaccination, treatment, surveillance, and other global coordination activities. Given diversity in the world and limited data, we characterize the costs for countries according to the 2013 World Bank income levels using 2013 U.S. dollars (2013$US). We estimate that routine immunization and supplemental immunization activities will cost governments and donors over 2013$US 2.3 billion per year for the foreseeable future, with high-income countries accounting for 55% of the costs, to vaccinate global birth cohorts of approximately 134 million surviving infants and to protect the global population of over 7 billion people. We find significantly higher costs and health consequences of measles or rubella disease than with vaccine use, with the expected disability-adjusted life year (DALY) loss for case of disease generally at least 100 times the loss per vaccine dose. To support estimates of the economic benefits of investments in measles and/or rubella elimination or control, we characterize the probabilities of various sequelae of measles and rubella infections and vaccine adverse events, the DALY inputs for health outcomes, and the associated treatment costs. Managing measles and rubella to achieve the existing and future regional measles and rubella goals and the objectives of the Global Vaccine Action Plan will require an ongoing commitment of financial resources that will prevent adverse health outcomes and save the associated treatment costs.

Probabilistic case detection for disease surveillance using data in electronic medical records

This paper describes a probabilistic case detection system (CDS) that uses a Bayesian network model of medical diagnosis and natural language processing to compute the posterior probability of influenza and influenza-like illness from emergency department dictated notes and laboratory results. The diagnostic accuracy of CDS for these conditions, as measured by the area under the ROC curve, was 0.97, and the overall accuracy for NLP employed in CDS was 0.91.

[Measles, pertussis, rubella and mumps completeness of reporting. Literature review of estimates for industrialized countries]

BACKGROUND

Surveillance data for infectious diseases in industrialized countries have been providing useful epidemiological information for several years. However, surveillance is complex and notifications underestimate the true disease incidence. The goal of this paper is to determine the completeness of reporting of measles, pertussis, rubella and mumps through notification in industrialized countries.

METHODS

A thorough literature review of papers published (in French or English) on surveillance data of measles, pertussis, rubella and mumps was undertaken with PUBMED. The review was limited to studies conducted in industrialized countries that quantitatively assessed the completeness of the reporting of these diseases.

RESULTS

Fourteen studies published from the 1920s met the inclusion criteria. For measles, the studies suggest that 64-85% of symptomatic patients sought medical care, 13-57% of those were diagnosed with measles and 22-81% of the diagnosed cases were reported for a completeness of reporting ranging from 7% to 63% between 1920 and 1980 and 3% to 41% in recent years. For pertussis, it was estimated that 49-59% of symptomatic patients sought medical care, 12% of those were diagnosed and 19-47% of the diagnosed cases were reported for a completeness of reporting ranging from 5% to 25% between 1920 and 1980 and from 3% to 12% in recent years. Very few studies on rubella and mumps were found and none allowed the determination of the global completeness of reporting.

CONCLUSION

Reporting has been found to be considerably incomplete. Continued efforts to improve the recognition and reporting of these diseases are needed.

Measles hospitalizations, United States, 1985-2002

Cases of measles that require hospitalization are a good marker of the burden of clinically severe measles in the United States. Measles hospitalizations routinely are monitored by the National Notifiable Disease Surveillance System (NNDSS). Our objectives were to describe measles hospitalizations reported to the NNDSS in 1985-2002, to use hospital discharge data from independent data sets (the National Hospital Discharge Survey [NHDS] [data available for 1985-1999] and the Health Care Investment Analysts [HCIA] hospital discharge database [data available for 1985-1996]) to provide additional estimates of total measles hospitalizations, and to compare trends in measles-associated hospitalizations. In 1985-2002, a total of 13621 patients with measles reported to the NNDSS were hospitalized (annual average, 757; range, 19-5856 patients). In 1985-1996, a total of 13472 measles hospitalizations were reported from NNDSS, compared with 28047 estimated from the NHDS and 19352 extrapolated from HCIA data. In the NNDSS, the annual total number declined after 1992 to <or=45 measles hospitalizations per year; this trend was closely paralleled by both the NHDS and the HCIA data. We demonstrate a decline in reported measles hospitalizations since the 1989-1991 measles resurgence, with the lowest numbers ever reported in the NNDSS. These numbers are corroborated by the very low numbers of measles hospitalizations in the NHDS and HCIA data.

Completeness of Measles Case Reporting: Review of Estimates for the United States

Measles surveillance is complex: the patient must seek health care, the diagnosis must be recognized by the physician, and the case must be reported to health departments. The portion of total (incident) measles cases that is reported to health departments is termed "completeness of reporting." Few studies describe this measure of the quality of surveillance in the United States; these studies use different methods, but they are all limited because the actual number of measles cases needed to derive completeness of reporting could not be determined. Estimates of completeness of reporting from the 1980s and 1990s vary widely, from 3% to 58%. One study suggests that 85% of patients with measles sought health care, the proportion of compatible illnesses for which measles was considered varied from 13% to 75%, and the proportion of suspected cases that were reported varied from 22% to 67%. Few cases were laboratory-confirmed, but all were reported. Surveillance in the United States is responsive, and its sensitivity likely increases when measles is circulating. Continued efforts to reinforce the clinical recognition and reporting of measles cases are warranted.

The interruption of transmission of indigenous measles in the United States during 1993

BACKGROUND

The United States has a goal to eliminate all indigenous cases of measles by the year 2000. Initial interruption of indigenous measles transmission would be expected during a period of very low measles incidence as occurred during late 1993.

METHODS

Indigenous measles cases (i.e. cases acquired in the United States and not traceable to any imported case) from 1993 were investigated to determine their source of infection. The probability of sustained undetected measles transmission between isolated indigenous cases was estimated.

RESULTS

Of the 312 measles cases reported for 1993, only 25 (8%) occurred after September 19. Of these only 4 cases (16%) could be classified as indigenous. The estimated probability that any of these 4 cases resulted from indigenous measles transmission in theirs or any adjoining counties was 0.05 or less.

CONCLUSIONS

Interruption of indigenous measles transmission appears to have occurred for the first time throughout the United States in 1993. This event provides strong support for the current national strategy for measles elimination. However, complete elimination of indigenous measles will require maintaining high population immunity to prevent spread from imported cases and attaining global measles control to prevent the importation of measles.

Measles reporting completeness during a community-wide epidemic in inner-city Los Angeles

A study was undertaken to determine the extent of measles underreporting among preschool-age children. In two community surveys conducted in inner-city Los Angeles during 1990 and 1991, respondents were asked whether preschool-age children in their households had ever been ill with measles. Information about measles episodes was obtained and medical records were reviewed, when available. A probable measles case was defined as having 3 or more days of rash with fever of 38.3 degrees centigrade or greater, and either cough, coryza, or conjunctivitis. To determine the proportion of cases reported, probable measles cases identified were matched with measles cases reported to the Los Angeles County Department of Health Services. Of the 947 children ages 6 weeks through 59 months included in the surveys, 35 children had experienced an illness episode which met the probable measles case definition. Ten (29 percent) of the 35 probable measles cases were reported to the health department. Hospitals reported 9 (69 percent) of 13 probable measles cases evaluated while private physicians' offices reported 0 (0 percent) of 12 evaluated (Fisher's exact test, P < 0.001), although 5 children were seen by private physicians before rash onset. Reporting was more complete for cases occurring during 1990 and 1991 (33 percent) than from 1987 through 1989 (18 percent). The hospitalization rate for preschool-age children with probable measles cases in the catchment area was estimated to be 8 percent (95 percent confidence interval = 0 to 18 percent). Although measles is a serious communicable disease which is almost completely preventable, cases of it among preschool-age children in this high incidence area were substantially underreported,especially by private physicians. Due to reporting bias, reported measles cases were representative of more severe cases than all the cases that occurred.