Validation of an Obstetric QoR Score and to Establish Its MCID.

The investigators aim to perform a prospective observational study of term parturients undergoing caesarean section. Patients undergoing elective and emergency caesarean section will be eligible for inclusion. Patients will be invited to fill out the survey before their caesarean section if feasible and then again at 24 and 48 hours. A subset of patients will be asked to repeat the second survey 30-60 minutes later to assess test-retest reliability. The patient demographics will be recorded at the initial survey. The time taken to fill out the 11-item ObsQoR-11 will be recorded. The QoR-15 score will also be completed at the same time and will allow comparison of the scores. The patient's Global health score will be assessed on a 0-100 scale from worst imaginable health state to best imaginable health state. At 24 and 48 hours patients will be asked to rate their overall recovery from surgery yesterday on a 15-point scale ranging from -7 to +7. This will facilitate an anchor-based determination of the MCID.

The validation of the ObsQoR-11 score will follow as per the Consensus based Standards for the selection of health Measurement Instruments (COSMIN) initiative. The assessment tool will be evaluated under the domains of validity, reliability, responsiveness and feasibility. The QoR-15 score will be similarly assessed to allow a comparison.

Validity will be assessed using two separate methods:

• Convergent and discriminant validity - Comparison of the ObsQoR-11 score to 100mm Global health score.
• Content validity - ObsQoR-11 will be assessed for correlation with patient age and length of hospital stay.

Reliability as a measure of consistency will be assessed based on the following methods:

• Internal consistency - the averaged correlation between each of the items within theObsQoR-11.
• Split half reliability - assessed by evaluating the correlation between random split segments.
• Test-retest reliability - a subset of patients will be reassessed at 30-60 minutes and the correlation between scores assessed.
• Floor and ceiling effects - this will be evaluated by whether <15% of patients achieved the highest (110/110) or lowest (0/110) possible scores.

Responsiveness is a measure of the ability to detect a clinically important change.

• Cohen effect size - the mean change in scores from before surgery to 24 hours after surgery divided by the standard deviation at baseline.
• Standardised response mean - the change in the scores before surgery to 24 hours after surgery divided by the standard deviation of the change in scores.
• Mean ObsQoR-11 scores - mean scores compared before surgery to 24 hours after surgery.

Feasibility

• Assessment of recruitment rate.
• Successful recruitment rate.
• Time taken to complete the ObsQoR-11.

The MCID will be determined based on an anchor-based method. The change in the mean score of the ObsQoR-11 score will be compared to the patient's own assessment of her change in health status using a Likert scale. We opted to follow a similar statistical method from an influential paper for estimation of the MCID. This will be based on calculation of the 0.3 SD (standard deviation), SEM (standard error of the mean) and the 5% range.

Inclusion criteria

• Caesarean section.
• Age >18 years.

Exclusion criteria

• Inability to read or understand English.
• Patient refusal.