Effects of a proposed quality improvement process in the proportion of the reported ultrasound findings unsupported by stored images

Objectives: To investigate the proportion of documented ultrasound findings that were unsupported by stored ultrasound images in the obstetric ultrasound unit, before and after the implementation of a quality improvement process consisting of a checklist and feedback.

Methods: A quality improvement process was created involving utilization of a checklist and feedback from physician to sonographer. The feedback was based on findings of the physician's review of the report and images using a check list. To assess the impact of this process, two groups were compared. Group 1 consisted of 58 ultrasound reports created prior to initiation of the process. Group 2 included 65 ultrasound reports created after process implementation. Each chart was reviewed by a physician and a sonographer. Findings considered unsupported by stored images by both reviewers were used for analysis, and the proportion of unsupported findings was compared between the two groups. Results are expressed as mean ± standard error. A p value of < .05 was used to determine statistical significance.

Results: Univariate analysis of baseline characteristics and potential confounders showed no statistically significant difference between the groups. The mean proportion of unsupported findings in Group 1 was 5.1 ± 0.87, with Group 2 having a significantly lower proportion (2.6 ± 0.62) (p value = .018).

Conclusions: Results suggest a significant decrease in the proportion of unsupported findings in ultrasound reports after quality improvement process implementation. Thus, we present a simple yet effective quality improvement process to reduce unsupported ultrasound findings.