Table 4

Studies that tested system interventions to address dimensions of diagnostic error

StudyStudy DesignInterventionDuration of interventionIntervention subjectsScopeOutcome measuresEffective? (Y/N) and Mechanism
Provider–patient encounter
Perno et al15UBADesignated paediatric trauma response team48 monthsCare teamUnknown care teams; a total of 3265 patients were included; no patients were excludedIncidence of delayed diagnosis of injury among paediatric trauma patientsY; speculated reasons included team dedicated only to trauma
Howard et al16Post-test onlyComprehensive re-evaluation (ie, tertiary examination) of trauma patients within 24 h of admission6 monthsA trauma clinical nurse specialist, 2 emergency physicians, and the trauma medical director4 healthcare providers, 90 patientsIncidence of missed injuriesY; tertiary ‘repeat’ examination and review of all lab and radiology studies
Diagnostic tests
Weatherburn et al17CBAImplementation of picture archiving and communications system (PACS), which acquires, transports, and stores radiographical images electronically, with accident and emergency (A&E) cliniciansPre-PACS data collection period based on conventional film images: 31 March 1992 to 30 October 1992; post-PACS data collection period: 1 April 1996 to 30 October 1996A&E departmentNo. of A&E attenders: 14 256 (film), 17 071 (PACS)Misdiagnosis (false negative) rates for adults and childrenY; speculated reasons include clinicians could manipulate soft copy images in PACS; potential for images to be viewed simultaneously in A&E and radiology prompting more consultations
Follow-up and tracking
Singh et al 24UBAAdded a code to the software configuration that links patients to their primary care physician for tests ordered by others10 monthsPrimary care physiciansOne large urban facility and satellite clinics; 490 alertsRates of timely follow-up of positive faecal occult blood tests pre and post interventionY; improved electronic communication of abnormal test results
Poon et al25Post-test onlyImplementation of Result Notification via Alphanumeric Pagers (ReNAP), an application that notifies clinicians of patient laboratory results via an alphanumeric pager once results are filed onto the patient database12 monthsInpatient and clinic physiciansDuring the 12-month period between February 2000 and January 2001, 780 different clinicians used ReNAP; a total of 22 775 requests were made during this time periodNo. of laboratory notification requests made, user satisfaction scoresY; improved electronic communication of test results
Piva et al26UBAImplementation of a computerised notification system for critical lab values (email, text message, video alert)2 monthsClinicians14 departments (including the emergency department) in one large hospitalPercentage of successful notifications (acknowledged within 1 h), time to notificationY; improved electronic communication of abnormal test results
  • CBA, controlled before and after study; post-test only, measures only taken after intervention was implemented; UBA, uncontrolled before and after study.