Adverse events in British hospitals: preliminary retrospective record review

Table A

Table B
 
 

Criteria for adverse events

An adverse event is "an unintended injury caused by medical management rather than by the disease process and which is sufficiently serious to lead to prolongation of hospitalisation or to temporary or permanent impairment or disability to the patient at time of discharge"

• Medical management includes both the actions of an individual member of staff or the overall health care system.

• Medical management includes acts of omission (for example, failure to diagnose or treat) and commission (for example, incorrect treatment).

• Causation of adverse event by medical management was judged on a six point scale, where 1 indicates "virtually no evidence for management causation" and 6 indicates "virtually certain evidence for management causation." Only adverse events with a score of 4 or higher, requiring evidence that causation is more likely than not, are reported in the results.

• Adverse events may or may not be preventable; the judgment is separate from that of causation. Preventability was also judged on a six point scale, with only those adverse events scoring 4 or higher being considered preventable.

• Injury may result from intervention or from failure to intervene. Although this is not always made clear, injuries that come about from failure to arrest the disease process are also included provided that standard care would clearly have prevented the injury.

• The injury has to be unintended, since injury can occur deliberately and with good reason (for example, amputation).

• Adverse events include recognised complications, which were judged as leading to harm but being of low preventability.

Operationalisation of adverse event criteria

The review form is structured in such a way that the clinician was obliged to make specific judgments about such issues as causation before making a final decision on the presence or absence of an adverse event. The definition of an adverse event is specified in the questionnaire and referred to during each review. Many of the criteria above are essentially clarifications of the basic definition designed to overcome common problems and ensure a consistent way of working. For instance, clinicians may be reluctant to include known complications as adverse events. In fact, the definition requires that they be included, but clinicians would specify low preventability to indicate that a certain level of complications is to be expected.

In this preliminary study we primarily aimed to follow the methods of previous studies and have used the same broad definition. In the light of our experience, and of further analysis of Australian data, we believe it would now be feasible to compile a list of some of the more important adverse events before a future study and produce standardised criteria for at least a proportion of specified adverse events.

Judgments of preventability

An adverse event was considered preventable if the reviewer judged that it would not have occurred if the patient had received ordinary standards of NHS care, appropriate for the time of the study (1998). For instance omission of prophylactic antibiotics or anticoagulants specified in a department protocol. As specified above a score of 4 or above on a six point scale was required for an adverse event to be identified as preventable.

Reliability and validity of adverse event judgments

For a full discussion of sensitivity and specificity issues, and data from reliability studies, see Brennan TA, Localio AR, Laird NL. Reliability and validity of judgement concerning adverse events suffered by hospitalised patients. Med Care 1989;27:1148-58.
 
 

Table A Number of records screened and number of adverse events by specialty and age

Table B Characteristics of cases who did and did not have adverse event