Passive surveillance as an instrument to identify risk factors for fatal Rocky Mountain spotted fever: is there more to learn?

Abstract

Abstract. National surveillance for Rocky Mountain spotted fever (RMSF) dates from 1920; however, the collection of detailed epidemiologic, clinical, and laboratory data on RMSF by using case report forms began in 1970. Despite issues with compliance and changes in case definitions, surveillance data have permitted researchers to assess risk factors for fatal RMSF quantitatively. Factors consistently associated with increased risk of death include severity of disease, older age, lack of tick bite, absence of classic symptoms, delay in diagnosis and initiation of appropriate antibiotic treatment, and treatment with chloramphenicol only. In several studies, treatment with a tetracycline has been shown to be protective. The continuation of current passive surveillance activities may allow researchers to refine their estimates of risk but is unlikely to produce novel results. Modified surveillance activities could focus on evaluating the risk for fatal RMSF among special populations, monitoring appropriate antibiotic use, and assessing new diagnostic tests.

Assessing the magnitude of fatal Rocky Mountain spotted fever in the United States: comparison of two national data sources.

“…However, the remarkable similarity of the geographic and demographic characteristics from each source, and similar seasonal and temporal trends between crude case counts during the study period support the validity of the RMSF diagnoses in the MCD data source as measured against the standard epidemiologic profile of this disease established by CRF data. 7,16,30 In a similar manner, capture-recapture estimates of RMSF deaths established by this study appear to be robust to variations in matching criteria: estimates obtained when criteria for matching were made more or less stringent were each contained within the 95% CI established for the original estimate.…”

“…This is suggested by the observation that only 22% of reported deaths were classified as unconfirmed, while 42% of all (fatal and non-fatal) CRF cases reported during the period 1981−1996 were categorized as unconfirmed. 7 Median annual concordance between CRF and MCD data for RMSF-associated deaths during the 1990s was approximately half of that observed during the preceding decade. Concordance between annual numbers of deaths identified by the CRF and MCD sources provides the basis for estimates of reporting completeness, and this study suggests a gradual but steady erosion of reporting of fatal RMSF during the study period, a trend particularly apparent with CRFs.…”

“…7 Some characteristics were not reported on all CRFs, including age (2 cases), sex (1), and month-of-death (24). cases of RMSF were reported to CDC through CRFs, that approximately 400 deaths were missed by the principal instrument of national RMSF surveillance, and that passive reporting through CRFs deteriorated progressively during the study period.…”

“…Using clinical and laboratory criteria, cases are classified as being confirmed, probable, or unconfirmed. 7 In this study, RMSF deaths from all three categories were compared with mortality records for RMSF compiled by the National Center for Health Statistics (NCHS, CDC). The NCHS mortality data represent a second independent source of national RMSF death information and are abstracted from all death certificates collected under the civil registration system of each state in the United States.…”

“…In this context, the quantity and quality of nationally reported RMSF data have been historically problematic and have hampered efforts to accurately estimate the true numbers of deaths attributable to this disease. 7 Capture-recapture methods, originally developed to estimate the size of mobile animal populations, compare the results of two or more surveillance systems for the same event.…”