Investigating determinants for patient satisfaction in women receiving epidural analgesia for labour pain: a retrospective cohort study

Background: Epidural analgesia is a popular choice for labour pain relief. Patient satisfaction is an important patient-centric outcome because it can significantly influence both mother and child. However, there is limited evidence in the correlations between clinical determinants and patient satisfaction. We aim to investigate clinical covariates that are associated with low patient satisfaction in parturients receiving labour neuraxial analgesia. Methods: After institutional ethics approval was obtained, we conducted a retrospective cohort study using electronic and corresponding hardcopy records from 10,170 parturients receiving neuraxial analgesia between the periods of January 2012 to December 2013 in KK Women's and Children's Hospital in Singapore. Demographic, obstetric and anesthetic data were collected. The patient satisfaction scores on the neuraxial labour analgesia was reported by the parturient at 24 to 48 h post-delivery during the post-epidural round conducted by the resident and pain nurse. Parturients were stratified into one of three categories based on their satisfaction scores. Ordinal logistic regression models were used to identify potential covariates of patient dissatisfaction. Results: 10,146 parturients were included into the study, of which 3230 (31.8%) were 'not satisfied', 3646 (35.9%) were 'satisfied', and 3270 (32.2%) were 'very satisfied'. Multivariable ordinal logistic regression analysis showed that instrument-assisted vaginal delivery (p = 0.0007), higher post-epidural pain score (p = 0.0016), receiving epidural catheter resiting (p < 0.0001), receiving neuraxial analgesia at a more advanced cervical dilation (p = 0.0443), multiparity (p = 0.0039), and post-procedure complications headache (p = 0.0006), backache (p < 0.0001), urinary retention (p = 0.0002) and neural deficit (p = 0.0297) were associated with patient dissatisfaction. Chinese, compared with other ethnicities (p = 0. 0104), were more likely to be dissatisfied. Conclusions: Our study has identified several clinical determinants that were independent associated factors for low patient satisfaction. These covariates could be useful in developing a predictive model to detect at-risk parturients and undertake time-sensitive precautionary measures for better patient satisfaction.