Caesarean section audit to improve quality of care in a rural referral hospital in Tanzania

2018 
Caesarean section (CS) is often a life-saving procedure, but can also lead to serious complications, even more so in low-resource settings. Therefore unnecessary CS should be avoided and optimal circumstances for vaginal delivery should be created. In this study, we aim to audit indications for Caesarean sections and improve decision-making and obstetric management. Audit of all cases of CS performed from January to August 2013 was performed in a rural referral hospital in Tanzania. The study period was divided in three audit blocks; retrospective (before auditing), prospective 1 and prospective 2. A local audit panel (LP) and an external auditor (EA) judged if obstetric management was adequate and indications were appropriate or if CS could have been prevented and yet retain good pregnancy outcome. Furthermore, changes in modes of deliveries, overall pregnancy outcome and decision-to-delivery interval were monitored. During the study period there were 1868 deliveries. Of these, 403 (21.6%) were Caesarean sections. The proportions of unjustified CS prior to introduction of audit were as high as 34 and 75%, according to the respective judgments of LP and EA. Following introduction of audit, the proportions of unjustified CS decreased to 23% (p = 0.29) and 52% (p = 0.01) according to LP and EA respectively. However, CS rate did not change (20.2 to 21.7%), assisted vacuum delivery rate did not increase (3.9 to 1.8%) and median decision-to-delivery interval was 83 min (range 10 - 390 min). Although this is a single center study, these findings suggest that unnecessary Caesarean sections exist at an alarming rate even in referral hospitals and suggest that a vast number can be averted by introducing a focused CS audit system. Our findings indicate that CS audit is a useful tool and, if well implemented, can enhance rational use of resources, improve decision-making and harmonise practice among care providers.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    24
    References
    8
    Citations
    NaN
    KQI
    []