Outbreak and infection-control reports
A report of an outbreak, or of a study of measures taken to control an infection acquired in hospital, tells its reader which kind of paper it is, lays out the setting and the control measures over time, accounts for the other changes that could explain the result, and names the pathogen so the reader can tell which organism is meant.
Outbreak report or intervention study
The ORION statement, which Stone and colleagues published in 2007, sets out guidelines for transparent reporting of outbreak reports and intervention studies of nosocomial infection.
The title and abstract describe the paper as an outbreak report or an intervention study, give the design of an intervention study, such as an interrupted time series with or without a control group or a crossover study, and briefly describe the intervention and the main outcomes. The introduction gives the scientific or local clinical background, or both, and the rationale, and describes the organism as epidemic, endemic, or epidemic becoming endemic. The introduction also states the type of paper, intervention study or outbreak report, and an outbreak report gives the number of outbreaks. The report gives the start and finish dates of the study or report. An outbreak report states its objectives and an intervention study its hypotheses. The study design is classified by the EPOC (Effective Practice and Organisation of Care) classification. The design section says whether the study was retrospective, prospective or ambidirectional, whether the decision to report or to intervene was prompted by any outcome data, and whether the study was formally implemented with a predefined protocol and endpoints.
Setting, intervention phases and outcomes
The setting is described as the unit, ward or hospital and, for a hospital, the units included, with the number of beds and the presence and staffing levels of infection control. The participants are described by the number admitted during the study or outbreak, summaries of the distributions of age and length of stay and, if possible, the proportion admitted from other wards, hospitals, nursing homes or from abroad. Where relevant, the report gives potential risk factors for acquiring the organism, and it gives the eligibility criteria for a study and the case definitions for an outbreak.
A phase is defined by a major change in a specific infection-control practice, with start and stop dates, and a summary table giving precise details of the interventions and how and when they were administered in each phase is strongly recommended.
The report details the culture media, the use of selective antibiotics, and local typing, reference typing or both, and where relevant the environmental sampling.
Primary and secondary outcomes, such as the incidence of infection, colonization or bacteremia, are clearly defined. They are reported at regular intervals, such as daily, weekly or monthly, with denominators such as the numbers of admissions or discharges or patient bed days, rather than as a total for each phase. The ORION checklist asks for at least three data points in each phase and, for many two-phase studies, twelve or more monthly data points in each phase. If possible, the prevalence of the organism and the incidence of colonization on admission are reported at the same intervals. The report gives the criteria for infection, colonization on admission and directly attributable mortality, and reports all-cause mortality.
Confounding, bias and interpretation
A report states which potential confounders were considered, recorded or adjusted for, such as changes in length of stay, case mix, bed occupancy, staffing levels, hand-hygiene compliance, antibiotic use, strain type, processing of isolates, or seasonality. It describes the measures taken to avoid bias, including blinding and standardization of outcome assessment and of the provision of care, and gives power calculations where appropriate.
The statistical methods used to compare groups or phases are described, along with the methods for any subgroup or adjusted analyses, distinguishing planned analyses from unplanned, exploratory ones. Unless the outcomes are independent, statistical approaches that can account for dependence in the outcome data should be used, adjusting where necessary for potential confounders. For an outbreak report, statistical analysis may be inappropriate.
For designs such as crossover studies, or where groups of patients were excluded, the results give the dates of the recruitment and follow-up periods and a flow diagram of participant flow in each phase. The main outcomes are reported with the estimated effect size and its precision, usually as a confidence interval. A graphical summary of the outcome data is often appropriate for dependent data, such as most time series. Any subgroup analyses should be reported, stating whether each was planned, that is specified in the protocol, and whether it was adjusted for possible confounders.
The report gives prespecified categories of adverse events and their occurrence in each intervention group, which might include drug side effects, crude or disease-specific mortality in antibiotic policy studies, or opportunity costs in isolation studies. If a formal economic study was done, the report defines the outcomes it reports and describes the resources used in the interventions, with costs broken down to basic units and the important assumptions stated.
An intervention study assesses the evidence for and against its hypotheses, accounting for potential threats to the validity of inference, including regression to the mean and reporting bias. An outbreak report considers the clinical significance of its observations and the hypotheses generated to explain them. The discussion addresses the external validity of an intervention study's findings, that is how far its results can be expected to generalize to other target populations or settings, and the feasibility of maintaining an intervention long term. The results are interpreted in the context of current evidence.
Naming the pathogen
The outbreak in Germany in 2011, caused by a bacterium that infected thousands and killed dozens, was written up under a jumble of abbreviations, and the most commonly used were these:
- EC: Escherichia coli
- EHEC: enterohaemorrhagic E. coli
- STEC: Shiga toxin-producing E. coli
- VTEC: verotoxin-producing E. coli
- HUS: haemolytic-uraemic syndrome, the range of symptoms the bacterium causes, including haemolytic anaemia, thrombocytopenia and renal failure
- HUSEC: HUS-associated E. coli
- HUS STEC: HUS-associated Shiga toxin-producing E. coli
- STEC HUS: Shiga toxin-producing E. coli-associated haemolytic-uraemic syndrome
Some of these abbreviations name different things, and of two that name the same thing, one is held to be more correct.
According to the Oxford Textbook of Medicine, there is an important epidemiological distinction between EHEC and STEC. In that textbook, EHEC refers to STEC associated with the distinctive clinical syndrome of haemorrhagic colitis, most commonly due to serotype O157:H7, while other STEC can produce a range of diarrhoeal illnesses that do not fit that description. So, in the textbook's words, all EHEC are STEC, only some STEC are EHEC, and STEC is the more comprehensive term.
STEC is synonymous with VTEC, because Shiga toxin is synonymous with verotoxin or verocytotoxin. The same textbook states that STEC is more correct, because it is named for the gene designation of the prototype Shiga toxin from Shigella dysenteriae type 1.
HUS STEC is more specific than HUSEC, and the European Centre for Disease Prevention and Control (ECDC) distinguishes HUS STEC from non-HUS STEC cases by symptom manifestation. HUS STEC and STEC HUS are sometimes used interchangeably, and one medical writer holds that the choice depends on whether the writing is about the symptoms (HUS) or the pathogen (STEC).
The bodies that named the organism during the 2011 outbreak used different names. The ECDC called it STEC or HUS STEC; the German federal Robert Koch Institute, which detected the first case, called it EHEC or EHEC O104:H4, O104:H4 being the serotype; the group of scientists who published a possible treatment called the organism STEC and the illness STEC–HUS; and the US Centers for Disease Control and Prevention called it STEC O104:H4, to distinguish it from the STEC O157:H7 that caused outbreaks in the United States in the 1990s. One medical writer holds that the serotype is the least ambiguous way of naming the organism. Whichever term is used, the medical writer's rule of thumb is to use it consistently throughout the document.