Article Text


Models of care delivery
P231 “Safety-net” for abnormal chest radiographs with a low index of suspicion for malignancy
  1. C Wall,
  2. P Arora,
  3. P M Turkington,
  4. R Chisholm,
  5. P M Turkington
  1. Salford Royal NHS Foundation Trust, Manchester, UK


Introduction and Objectives Guidelines dictate that where a chest x-ray is incidentally suggestive of lung cancer, a copy of the radiologists report is sent to a designated member of the lung cancer multidisciplinary team. A similar system was implemented for abnormal chest radiographs demonstrating significant non-malignant changes or abnormalities with a low index of suspicion of malignancy. This was used to assess if abnormal chest radiographs are adequately followed-up by the requesting clinicians.

Methods Reporting radiologists coded radiographs that warranted further action such as repeat imaging within a specified time frame. These were relayed to a designated coordinator. If action did not occur as suggested on the report, the requesting clinician was prompted on a weekly basis until the issue was resolved or else referred to a member of the chest team for further evaluation. At 6 months the cases were reviewed and information extracted including indication for further action, whether action occurred without coordinator intervention and outcome.

Results All coded films over a 6-month period (n=331) were analysed. The most frequent indications for further action included the assessment of pneumonia resolution (n=167), further review of possible hilar pathology (n=25), small nodules (n=22) and other unexplained abnormalities. 143 cases (42.6%) required action by the coordinator. In total, 4% (n=13) of cases yielded significant diagnoses including eight malignancies, pulmonary embolism, cobalt induced fibrosis and asbestosis, of these one case required action by the coordinator.

Conclusions This study suggests that a large proportion of abnormal chest radiographs are not followed-up as suggested by the reporting radiologist. Significant diagnoses may therefore be delayed unnecessarily. There is little reason to assume that similar practice is uncommon throughout the country. While the number of significant diagnoses detected by the “safety-net” system was perhaps lower than expected, some may consider a delay in 7.7% (1/13) unacceptably high. Further exploration of this area is needed to decide whether this is an effective use of resources.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.