Summary
Introduction
The Obstetric Quality of Recovery-10 (ObsQoR-10) is a patient-reported outcome measure for evaluating postpartum recovery. However, the minimal clinically important difference for ObsQoR-10 remains unknown, limiting clinical and research interpretability. The primary aim of this study was to determine the minimal clinically important difference of ObsQoR-10 through systematic review and network meta-analysis of studies and to evaluate its psychometric properties.
Methods
We conducted a systematic review and network meta-analysis of studies and searched relevant databases using keywords relating to ‘obstetric quality of recovery’. We included observational and interventional studies reporting more than one psychometric property of ObsQoR-10 at ≥ 1 inpatient time-point. Two reviewers independently screened and evaluated study quality using the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) risk of bias checklist. Minimal clinically important difference was estimated through triangulation of distribution-based (half of standard deviation of scores) and anchor-based (differences in scores between vaginal delivery and caesarean delivery, and between planned and unplanned caesarean delivery) methods.
Results
We included 12 studies with a total of 2017 parturients. Mean (SD) ObsQoR-10 score ranged from 48 (16.3) (caesarean delivery) to 89 (9.3) (vaginal delivery). Distribution-based estimates ranged from 6.7 (95%CI 6.0–7.6) to 8.7 (95%CI 7.9–9.6), while anchor-based estimates ranged from 9.7 (95%CI 2.7–16.5) to 12.4 (95%CI 3.1–21.5). Most studies reported statistically significant differences in scores between delivery modes. Overall, there was moderate evidence for sufficient validity, reliability and responsiveness of ObsQoR-10 across the postpartum period.
Discussion
We recommend using a minimal clinically important difference of 7–10 points for ObsQoR-10, with 7 points as the lower-bound estimate. This can be used to help interpret ObsQoR-10 scores in clinical settings and to power future studies using ObsQoR-10 as a primary outcome measure.
Keywords: postpartum period, minimal clinically important difference, psychometrics, patient-reported outcome measures
Plain Language Summary
What we did
Researchers looked at 12 studies involving 2017 women who had recently given birth. They studied the Obstetric Quality of Recovery-10 item (ObsQoR-10) survey, which asks women about their postpartum recovery. They compared scores from different patient groups and used several methods to determine how much a score needs to change before it represents a meaningful difference in recovery.
Why did we do it
Clinicians and researchers use ObsQoR-10 to measure how well women are recovering after childbirth. However, they did not know what change in the score is important to women. The researchers wanted to find a useful number that would show when a change in recovery was likely to be meaningful rather than just a small change that might not matter.
What we found
The researchers found that a change of about 7–10 points on the ObsQoR-10 score is likely to be meaningful. A change of 7 points can be used as the minimal clinically significant change in score. The questionnaire also appeared to be reliable and useful for measuring recovery after both vaginal and caesarean birth. These findings will help doctors and patients understand changes in recovery and help researchers decide how many women need to be included in future studies.
Introduction
Effective postpartum care is associated with short- and long-term maternal and infant outcomes [1–3]. Quality of postpartum care is complex, spanning multiple domains including physical recovery; maternal psychological and social wellbeing; sleep; fatigue; and infant care [1,4]. Patient-reported outcome measures such as the Obstetric Quality of Recovery-10 item measure (ObsQoR-10) can facilitate patient-centred assessment and care by directly capturing multiple domains of inpatient recovery from the patient-perspective. The ObsQoR-10 consists of 10 items encompassing physical, mental, infant-related and psychosocial domains, and yields a total score ranging from 0 to 100, with higher scores indicating better recovery.
The ObsQoR-10 has been identified as the best available patient-reported outcome measure for evaluating inpatient postpartum recovery and has been validated in multiple healthcare settings and in different languages [5]. Inpatient ObsQoR-10 scores are associated with outpatient 6-week depression scores and quality of outpatient recovery [6]. The psychometric properties have been explored and described in multiple countries [7,8]. Clinical and research implementation has been recommended, including evaluation of enhanced recovery after caesarean delivery protocols [9]. However, the minimal change in ObsQoR-10 score signifying a clinically important difference has not yet been established [9]. Determining the minimal clinically important difference (MCID) would help guide clinical decision-making. It would also support the identification of risk factors and interventions that produce clinically meaningful recovery differences in quality improvement and research initiatives.
Our primary aim was to estimate the MCID in ObsQoR-10 score through a systematic review and network meta-analysis of relevant studies. The secondary aim was to summarise the reported psychometric properties of ObsQoR-10 across different languages and contexts.
Methods
A medical librarian (LB) searched PubMed MEDLINE, Embase, Web of Science, Scopus, CINAHL and PsycINFO using keywords relating to obstetric quality of recovery (online Supporting Information Appendix S1) with no date or language limiters. We conducted our search on 11 February 2025 with an updated search on 23 February 2026. We included observational and interventional studies and all publications that reported more than one psychometric property of ObsQoR-10 (definitions are outlined in online Supporting Information Appendix S2). In brief, these cover patient-reported outcome metrics of validity, reliability and responsiveness [10]: validity asks whether the tool measures the construct of postpartum recovery; reliability asks whether the score is reproducible; and responsiveness asks whether the measure can detect change in recovery status over time. We did not include studies that used ObsQoR-10 without evaluation of psychometric properties; outpatient studies; studies that used ObsQoR-11; qualitative studies; case reports; letters; and conference abstracts/posters. This study follows the PRISMA reporting guidelines.
We uploaded articles to Rayyan [11], removed duplicate publications and two reviewers (PC and AJ) independently screened by title, abstract then full text; PS and MT resolved any discrepancies. Where there were multiple publications from the same study sample, the reviewers selected the key article after discussion.
Two reviewers (MT and SG) extracted data into a standardised table. We extracted the following: study characteristics and design (including country of study, language, year of publication, study aims, setting, sample size, participant inclusion and exclusion criteria, timing of ObsQoR-10 administration); psychometric aspects including validity analysis, reliability analysis and responsiveness; feasibility; and interpretability of ObsQoR-10.
Minimal clinically important differences are values of patient-reported outcome measures that correspond to a minimal within-person change over time that patients consider important. Clinically significant differences are defined as a difference in score that is large enough to have implications for the treatment or care of the patient, such as change in clinical management [12]. These concepts are defined further in online Supporting Information Appendix S2.
Three authors (PC, AJ, SG) assessed the quality of the included studies using the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) risk of bias checklist [10]. The reviewers graded the methodology used to assess a measurement property in each study as ‘very good’, ‘adequate’, ‘doubtful’, ‘inadequate’ or ‘not applicable’ and selected the overall risk of bias rating based on the worst score [10]. At least two authors evaluated each study and a fourth (MT) resolved discrepancies. The COSMIN checklist contains standards for 10 domains: patient-reported outcome measure development; content validity; structural validity; internal consistency; cross-cultural validity; reliability; measurement error; criterion validity; hypothesis testing for construct validity; and responsiveness [10]. We focused on the latter seven domains, as the first three domains (development of ObsQoR-10, its content validity and structural validity) fell outside the scope of our study and have been described previously [7,13]. Previous studies have evaluated structural validity of ObsQoR-10, and showed unidimensionality of the patient-reported outcome measure [14]. We did not assess criterion validity as there is no established gold standard for postpartum recovery evaluation.
The primary outcome of the study was to determine the MCID for ObsQoR-10 scores obtained within 24 ± 6 h of delivery [15]. Secondary outcomes were to summarise measures of validity; reliability; responsiveness; feasibility; and interpretability related to ObsQoR-10 (online Supporting Information Appendix S2).
The MCID of a health-related quality of life patient-reported outcome measure such as ObsQoR-10 can be broadly divided into two categories: distribution-based; and anchor-based methods [16,17]. For this study, both methods were implemented. We used the distribution-based method as primary evidence and the anchor-based method as supportive evidence.
We calculated half the standard deviation of ObsQoR-10 scores as a distribution-based MCID. This is the statistical method used most widely for health-related quality of life outcome measures [17]. To calculate it, the ObsQoR-10 score is assumed to follow a normal distribution so that log-transformed empirical standard deviation follows a normal distribution with a variance of 0.5/(n-1), where n is the sample size of an individual study. The random effects model is used to combine the reported standard deviation from each individual study to estimate the mean standard deviation for vaginal delivery, planned caesarean delivery and unplanned caesarean delivery separately.
For the anchor-based method, we used the mean group differences in ObsQoR-10 between vaginal delivery, planned caesarean delivery and unplanned caesarean delivery as anchors, since clinically significant recovery differences between these groups are well established [15]. Caesarean delivery is associated with worse pain, higher and longer opioid consumption [18–20], and greater functional limitations [19,21] compared with vaginal delivery. Unplanned caesarean delivery is linked to greater maternal morbidity [22,23], peripartum psychological distress [24] and negative childbirth experience [25] compared with planned caesarean delivery. To estimate the mean between group differences, we collected data from studies reporting the mean and standard deviation of the ObsQoR-10 score for patients delivering via vaginal, planned caesarean or unplanned caesarean. For studies which only reported median and quartiles of the ObsQoR-10 score, we calculated the mean and SD as the reported median and interquartile range divided by 1.349 [26], assuming that the ObsQoR-10 approximately follows a normal distribution. We then conducted the network meta-analysis in R (version 4.5.1, R foundation for Statistical Computing, Vienna, Austria) using the default setting of the ‘pcnetmeta’ package. We report the between group differences as well as the 95% credible intervals and the mean ObsQoR-10 scores for each delivery mode.
To evaluate the impact of distributional assumptions, we performed a sensitivity analysis and did not include studies that required the conversion of median-to-mean to account for potential skewness in ObsQoR-10 scores that could bias pooled estimates. The ObsQoR-10 scores for operative vaginal deliveries were not included in the meta-analysis as studies have shown that there are significant differences in postpartum recovery between operative and spontaneous vaginal deliveries [7], and operative vaginal delivery rates are considerably lower than other delivery modes [27,28]. However, given that operative vaginal delivery is performed far less commonly than spontaneous vaginal delivery, studies that included mixed vaginal delivery cohorts were included. To evaluate the impact of this assumption, we also performed a sensitivity analysis omitting studies with mixed vaginal delivery cohorts (i.e. studies that combined spontaneous and operative vaginal delivery groups).
We rated the psychometric measurement properties reported in the included studies using COSMIN methodology [10,29]. Briefly, we assessed the results of each psychometric property for every study and evaluated them collectively to derive an overall performance assessment for the property assessment. We assigned a level of evidence for each psychometric property using a modified GRADE approach [10].
Results
Twelve studies met inclusion criteria (Fig. 1) and included 2017 parturients across 12 countries. The ObsQoR-10 had been validated in 10 languages (Arabic, Dutch, French, German, Hebrew, Japanese, Portuguese, Spanish, Turkish and Thai). Study characteristics are summarised in Table 1 with an extended summary and results in online Supporting Information Appendices S3 and S4. Reported mean (SD) ObsQoR-10 scores ranged from 48 (16.3) for planned caesarean delivery to 89 (9.3) for vaginal delivery (Table 2).
Figure 1:
Study flow diagram. ObsQoR-11, Obstetric Quality of Recovery 11.
Table 1.
Included study characteristics
| Study (location) | Study site | Study population | Number of deliveries | Time of ObsQoR-10 completion |
|---|---|---|---|---|
| Dusitkasem et al. [34] (Thailand) | Single tertiary centre | Planned caesarean delivery c ≥ 18 y Primiparous and multiparous |
110 | Day before surgery 24 h postpartum 48 h postpartum |
| Fantin et al. [30] (Austria) | Multicentre tertiary | Planned caesarean and spontaneous vaginal delivery Age ≥ 18 y Primiparous |
Planned caesarean (n = 100) Spontaneous vaginal (n = 100) |
24 ± 6 h postpartum |
| Guevara et al. [38] (Colombia) | Single tertiary centre | Planned caesarean, spontaneous, surgical vaginal delivery Age ≥ 18 y Primiparous |
Planned caesarean (n = 50) Spontaneous vaginal or surgical (n = 50) |
24 ± 6 h postpartum |
| Hayoun et al. [39] (Israel) | Single tertiary centre | Planned and unplanned caesarean, spontaneous or surgical vaginal delivery Age > 18 y Primiparous Gestational age ≥ 37 weeks |
Spontaneous vaginal (n = 152) Operative vaginal (n = 53) Planned caesarean (n = 64) Unplanned caesarean (n = 51) |
24–48 h postpartum |
| Kozanhan et al. [35] (Turkey) | Single tertiary centre | Planned caesarean and spontaneous vaginal delivery Primiparous |
Planned caesarean (n = 50) Spontaneous vaginal (n = 50) |
24 h postpartum |
| Mathias et. al. [40] (Brazil) | Single tertiary centre | Planned caesarean delivery Age ≥ 18 y Primiparous and multiparous |
113 | 24 ± 6 h postpartum |
| Mazda et al. [15] (Japan) | Single tertiary centre | Planned caesarean and spontaneous vaginal delivery Age ≥ 18 y Primiparous |
Planned caesarean (n = 48) Spontaneous vaginal (n = 50) |
Postpartum 24 h, 48 h, 72 h, 96 h, 120 h, 144 h, 168 h |
| Mazoué et al. [14] (France) | Single tertiary centre | Planned and unplanned caesarean, spontaneous, induced vaginal delivery Age ≥ 18 y Primiparous and multiparous |
Planned caesarean (n = 26) Unplanned caesarean (n = 53) Spontaneous vaginal (n = 228) Induced vaginal (n = 124) |
0 h - during labour 24 h ± 4 h postpartum 48 h ± 4 h postpartum |
| Mukarram et al. [31] (Saudi Arabia) | Single tertiary centre | Unplanned caesarean delivery Age ≥ 18 y Primiparous and multiparous |
100 | 24 h postpartum 48 h postpartum |
| Sultan et al. [32] (USA) | Single tertiary centre | Planned and unplanned caesarean and spontaneous vaginal delivery ≥ 18 y Primiparous and multiparous |
Caesarean (n = 109) Spontaneous vaginal (n = 106) |
Within 72 h postpartum |
| Sultan et al. [7] (UK) | Single secondary centre | Operative and spontaneous vaginal delivery ≥ 18 y Primiparous and multiparous |
Spontaneous vaginal (n = 68) Operative vaginal (n = 55: forceps, n = 18; vacuum, n = 37) |
Postpartum day 1, within 40 h of delivery |
| van den Bosch et al. [33] (the Netherlands) | Single tertiary centre | Planned and unplanned caesarean and spontaneous vaginal delivery ≥ 18 y Primiparous and multiparous |
102; delivery: Planned caesarean (n = 60) Unplanned caesarean (n = 35) Repeat caesarean (n = 49) |
24 h postpartum |
ObsQoR-10, Obstetric Quality of Recovery 10
Table 2.
Average Obstetric Quality of Recovery 10 (ObsQoR-10) scores for each delivery mode. Values are number, mean (SD) or median [IQR].
| Study | Planned caesarean delivery | Unplanned caesarean delivery | Vaginal delivery |
|---|---|---|---|
| Dusitkasem et al. [34] | 57 (11.6) (n=110) | ||
| Fantin et al. [30] | 70 (16.5) (n = 100) | 83 (13.0) (n = 100) | |
| Guevara et al. [38] | 81 (12.4) (n = 50) | 89 (9.3) (n = 50) | |
| Hayoun et al. [39] | 69 (16.3) (n = 64) | 56 (16.9) (n = 51) | 81 (12.1) (n = 105, no episiotomy) 69 (13.6) (n = 52, episiotomy) |
| Kozanhan et al. [35] | 83 [76-89] (n = 50) | 83 [69-90] (n = 50) | |
| Mathias et al. [40] | 84 [74-92] (n = 113) | ||
| Mazda et al. [15] | 48 (16.3) (n = 48) | 73 (17.7) (n = 50) | |
| Mazoué et al. [14] | 75 (14.7) (n = 26) | 65 (18.4) (n = 53) | 81.7 (13.1) (n = 352) |
| Mukarram et al. [31] | 75[64-87] (n = 100) | ||
| Sultan et al. [32] | 76 [62-84] (n = 80) | 81 [65-89] (n = 29) | 86 [77-94] (n = 106) |
| Sultan et al. [9] | n = 68, 80.2 (15.5) | ||
| van den Bosch et al. [33] | 73 (12.0) (n = 60) | 64 (17.0) (n = 42) |
There was moderate quality evidence for sufficient construct validity, reliability and responsiveness of ObsQoR-10, but there was low-quality evidence for insufficient cross-cultural validity (Table 3). There was serious risk of bias among the three studies that evaluated cross-cultural validity of ObsQoR-10, with considerable inconsistency between study results. The quality of the included studies and risk of bias ratings are described in greater detail in online Supporting Information Appendix S5.
Table 3.
Risk of bias assessment and GRADE level of evidence for overall performance rating of each psychometric property using COSMIN guidelines
| Quality of evidence for measurement property of ObsQoR-10 | Methods (risk of bias) assessment | Results |
|---|---|---|
| Hypothesis testing for construct validity | Summarised result | Forty-five supporting hypothesis for convergent validity with comparator instrument, discriminant validity between ‘good’ and ‘poor’ recovery groups and known-groups hypothesis between delivery modes. Five not supporting hypothesis for convergent validity with comparator instrument, discriminant validity between ‘good’ and ‘poor’ recovery groups or known-groups hypothesis between delivery modes. |
| Rating summarised result | + | |
| Quality of evidence | Moderate | |
| Cross-cultural validity | Summarised result | One supporting sufficient cross-cultural validity Two suggesting insufficient cross-cultural validity |
| Rating summarised result | − | |
| Quality of evidence | Low | |
| Reliability | Summarised result | Ten supporting sufficient intraclass correlation coefficient |
| Rating summarised result | + | |
| Quality of evidence | Moderate | |
| Internal consistency | Summarised result | Fourteen supporting sufficient internal consistency One suggesting insufficient internal consistency |
| Rating summarised result | + | |
| Quality of evidence | High | |
| Responsiveness | Summarised result | Four supporting responsiveness One not supporting responsiveness |
| Rating summarised result | + | |
| Quality of evidence | Moderate |
ObsQoR-10, Obstetric Quality of Recovery 10; COSMIN, COnsensus-based Standards for the selection of health Measurement INstruments. Methods (risk of bias) assessment: very good; adequate; doubtful; or inadequate. Rating summarised results: sufficient (+); insufficient (−); inconsistent (+/−); or indeterminate (?). Quality of evidence rated using GRADE level of evidence: high; moderate; low; or very Low.
Ten studies reported descriptive statistics for ObsQoR-10 scores obtained within 24 ± 6 h of delivery for each delivery mode, including 1426 parturients (Table 4). Network meta-analysis established highest mean (SD) ObsQoR-10 scores among patients following vaginal delivery (81 (13.6)), followed by planned caesarean delivery (71 (13.4)), and lowest among patients who underwent unplanned caesarean delivery (69 (17.4); online Supporting Information Appendix S6). The differences between scores among patients following planned and unplanned caesarean deliveries were not statistically significant. The between-group differences between patients who underwent vaginal and planned caesarean delivery, as well as between vaginal and unplanned caesarean delivery, were statistically significant. Distribution-based estimates of the MCID ranged from 6.7 to 8.7 (Table 4) and anchor-based MCID estimates ranged from 9.7 to 12.4 (Table 4). In a sensitivity analysis omitting studies that required median-to-mean conversion, the MCID estimates using the distribution-based method remained stable (6.9 vs. 6.7 for planned caesarean delivery, 6.7 vs. 6.8 for vaginal delivery). The known-groups comparison estimate was 13.0 vs. 9.7 with a wider 95% credible interval. A further sensitivity analysis which did not include mixed spontaneous and operative vaginal delivery cohorts showed that MCID estimates remained stable using the distribution-based method (6.9 vs. 6.8 for vaginal delivery) and known-groups comparison (8.7 vs 9.7). Full sensitivity analysis results are summarised in online Supporting Information Appendix S7.
Table 4.
Summary of minimum clinically important difference (MCID) estimates from distribution- and anchor-based methods. Values are SD (95%CI) or median (95%Crl).
| Distribution-based MCID (ObsQoR-10 scores by delivery mode) | |
|---|---|
| Delivery mode | 0.5 SD |
| Vaginal delivery | 6.8 (5.9 – 7.8) |
| Planned caesarean delivery | 6.7 (6.0 – 7.6) |
| Unplanned caesarean delivery | 8.7 (7.9 – 9.6) |
| Anchor-based MCID (ObsQoR-10 score differences between delivery modes) | |
| Known group comparisons | Between group difference |
| Vaginal vs. planned caesarean delivery | 9.7 (2.7 – 16.5) |
| Vaginal vs. unplanned caesarean delivery | 12.4 (3.1 – 21.5) |
| Planned vs. unplanned caesarean delivery | 2.7 (−5.9 – 11.2) |
ObsQoR-10, Obstetric Quality of Recovery 10.
All studies supported convergent and discriminant validity of ObsQoR-10. Moderate or strong correlation between ObsQoR-10 and the global health visual analogue scale, the EuroQol Five Dimension, and the global health numerical rating scale was shown across all studies. Discriminant validity was supported in all 10 studies that assessed discriminant validity, with a statistically significant difference in ObsQoR-10 scores between patients who reported a global health visual analogue scale or global health numerical rating scale of ≥ 70 vs. < 70.
A statistically significant difference in ObsQoR-10 scores between different delivery modes was reported in 8/9 studies. Vaginal delivery was associated with significantly higher scores compared with caesarean delivery in 5/6 studies; planned caesarean delivery was associated with significantly higher scores than unplanned caesarean delivery in 2/3 studies; and two studies established higher scores following spontaneous vaginal delivery compared with operative vaginal delivery (Table 5). There was a statistically significant correlation between ObsQoR-10 scores and maternal duration of hospital stay (ρ = −0.305, p < 0.001 [30]; r = −0.177, p < 0.001 [14]; r = −0.21, p = 0.03 [31]; r = −0.2, p < 0.001 [32]; r = −0.264, p = 0.006 [33]), time to mobilisation (ρ = −0.323, p < 0.001 [30]; r = −0.57, p < 0.001 [15]) and postpartum opioid consumption (ρ = −0.346, p < 0.001 [30]; r = −0.283, p = 0.001 [15]) in the majority of studies that assessed these variables. Studies reported inconsistent correlation between ASA physical status, gestational age, maternal age, estimated blood loss and APGAR scores with ObsQoR-10 (Table 5).
Table 5.
Summary of known groups validity hypothesis testing results
| Known groups comparison assessed | Studies that support the hypothesis | Studies that do not support the hypothesis |
|---|---|---|
| Supports known-groups validity | ||
| ObsQoR-10 scores differ significantly between delivery modes: | 8 | 1 |
| Caesarean vs. vaginal delivery | 5 | 1 |
| Unplanned vs. planned caesarean delivery | 2 | 1 |
| Spontaneous vs. surgical vaginal delivery | 2 | None |
| ObsQoR-10 scores correlate with: | ||
| Duration of maternal stay in hospital | 6 | 2 |
| Postpartum inpatient opioid consumption | 2 | 1 |
| Time to mobilisation | 2 | None |
| Does not support known-groups validity | ||
| ObsQoR-10 scores correlate with: | ||
| Gestational age | 3 | 3 |
| ASA physical status | 2 | 2 |
| Estimated blood loss | 3 | 7 |
| Maternal age | 2 | 3 |
| Apgar scores | 1 | 2 |
ObsQoR-10, Obstetric Quality of Recovery 10.
Cross-cultural validity was assessed in three studies. Of these, two reported differential item functioning between items. The Portuguese translation of ObsQoR-10 identified bias in two items (‘shivering’ and ‘feeling in control’), while the Japanese translation found differential item functioning in all items except pain.
Reported Cronbach’s α values ranged from 0.67 to 0.88 across all studies [31,34], indicating acceptable to good internal consistency. Spearman-Brown correlation values indicated acceptable to very good reliability (range 0.75 to 0.94) in all 10 studies, where assessed [15,31]. Test-retest reliability was good or excellent in all 10 relevant studies (median = 0.90; online Supporting Information Appendix S4).
Four studies evaluated change in ObsQoR-10 scores over time. Two reported statistically significant increase in scores from 24 to 48 h postpartum, with over 10-point differences in median scores between the two time-points among parturients who had vaginal deliveries [15], and almost 20-point differences between the two time-points among parturients who delivered by caesarean delivery [15,31]. One study reported approximately 10-point increase in median scores from 24 to 48 h after delivery [14], although they did not report statistical significance. One study did not identify any statistically significant change in scores over 72 h postpartum but was limited by small sample size [32].
One interventional study assessing the effects of intrathecal morphine on post-caesarean delivery quality of recovery showed a statistically significant difference in ObsQoR-10 scores following intrathecal morphine administration with mean difference of 9.0 points (95% CI 3.1–15, p = 0.002) in the pre- vs. post-implementation groups [33].
Survey completion rate ranged from 86% to 100% across all studies. Median ObsQoR-10 completion time was 1.2–3 min [30,35].
Floor and ceiling effects were not observed in any of the studies at 24 h postpartum. Ceiling effect was observed at 48 h postpartum in one study, where the maximum possible ObsQoR-10 score was reported by 16/100 (16%) parturients, 1% above the 15% threshold used to determine floor/ceiling effects [31].
One study evaluated MCID and clinically significant difference. The MCID was estimated using the average of the MCID obtained by the distribution method (1.96 x standard error of measurement) and anchor method (difference in mean score values between the ‘same’ and ‘slightly good’ statuses on a seven-item Likert scale ranging from ‘very poor’ to ‘very good’ reported at the time of ObsQoR-10 completion) [14]. The MCID was 5.0 (95%CI 2.3-7.8) and 4.5 (95%CI 1.1-8.0) at 24 h and 48 h, respectively. Clinically significant difference was estimated using the difference in ObsQoR-10 score between the group of women who answered ‘yes’ to the question “Do you consider that you have recovered well?” and those who answered ‘no’ at 24 and 48 h postpartum. The clinically significant difference was 18.2 (95%CI 16.9-19.5) and 20.3 (95%CI 19.40-21.18) at 24 h and 48 h, respectively.
Discussion
The smallest MCID estimate in our meta-analysis was 7 points using distribution-based methods as primary evidence. The ObsQoR-10 performed well in measures of validity, reliability, responsiveness and feasibility among inpatients within 48 h of childbirth following vaginal and caesarean delivery.
The lower bound of our MCID estimate is closely aligned with previous reports. Mazoué et al. reported an MCID of 5 points at 24 h postpartum, while our triangulated estimates range from 6 to 10 points [14]. The discrepancies are likely attributable to two key methodological differences. First, Mazoué et al. used a patient-perceived anchor (a seven-item Likert scale ranging from ‘very poor’ to ‘very good’ at 24 h and 48 h postpartum), whereas our meta-analysis relied on a known-groups comparison between delivery modes as the anchor. Our anchor likely captures a more substantial clinical difference than the ‘minimal’ threshold, which explains our higher anchor-based MCID estimate of 10 points. Second, Mazoué et al. calculated the MCID using a simple arithmetic average of the distribution-based and anchor-based estimates [14]. In contrast, we triangulated the MCID estimates derived from both approaches across a larger multi-study population. When considered alongside previous reports, our findings suggest that the ObsQoR-10 MCID likely falls between 6 and 10 points, with 10 representing a conservative upper-bound estimate that may reflect a moderate-to-large clinical difference.
The application of these MCID thresholds across different contexts must account for the cross-cultural validity of the specific ObsQoR-10 translation. Our findings indicate considerable differential item functioning depending on the translation, suggesting that a given score difference in one context may not be clinically equivalent to the same score difference in another. The significance of differential item functioning is determined by the number of affected items, the magnitude of the differential item functioning and the extent to which the relevant items contribute to the total score. For example, patients based in the USA scored 1.48 units higher than expected compared with patients from Portuguese study cohorts in the item regarding ‘feeling in control’ for the same level of overall recovery. While this represents a small impact on a 0-100 scale, it represents a non-trivial proportion of our 6–10 point MCID estimate. Furthermore, where differential item functioning is shown across the majority of items, such as in the Japanese translation, the MCID is unlikely to be universal and would be inappropriate to apply indiscriminately. For clinicians and researchers in multinational trials, using a single MCID across languages and contexts may result in misinterpretation of treatment effects if cross-cultural validity is not established. Consequently, further prospective studies are needed to determine the necessity of language-specific MCIDs, particularly where the cross-cultural validity remains unknown or insufficient.
Our network meta-analysis has shown that vaginal delivery was associated with the highest mean ObsQoR-10 scores (best recovery), while unplanned caesarean delivery was associated with the lowest scores (worst recovery). The differences between vaginal and planned caesarean delivery as well as between vaginal and unplanned caesarean delivery were statistically significant; however, the difference between planned and unplanned caesarean deliveries was not statistically significant, likely due to small sample sizes. This is consistent with previous findings. Caesarean delivery has been associated with worse pain and increased opioid consumption compared with vaginal delivery [15,18–21]. Similarly, unplanned caesarean delivery has been associated with worse recovery compared with planned caesarean delivery [22–25].
The MCID facilitates interpretability of ObsQoR-10 in clinical settings. This can help clinicians to identify patients with clinically significantly worse recovery that may benefit from clinician review, guide when to refer to subspecialties such as physical therapy, social work and lactation teams, and may also aid monitoring response to management. Moreover, given the differences in the quality of recovery between delivery modes across multiple settings, understanding the reasons for these differences could help inform patient counselling, clinical decision-making regarding delivery mode and intra- and postpartum management.
In research, the MCID aids interpretability of research findings where ObsQoR-10 is used as an outcome measure. The ObsQoR-10 has been recommended in a core outcome set of measures for evaluating enhanced recovery after caesarean delivery [9]. The MCID can be used to evaluate effectiveness in quality improvement projects and interventional studies. Further prospective longitudinal studies using multiple independent anchors, such as global ratings of change, are necessary to confirm our MCID estimates. At the same time, it is also necessary to establish what change in each anchor constitutes a MCID.
A strength of our study is the triangulation of distribution- and anchor-based approaches to estimate MCID, using a distribution-based approach as primary evidence. The two approaches produced compatible results, and our sensitivity analysis further reinforced the distribution-based estimate. While median-to-mean conversion can introduce bias, the distribution-based MCID estimate remained stable after omitting studies that required median-to-mean conversion (online Supporting Information Appendix S7). This consistency adds evidence to the robustness of our distribution-based estimates. Moreover, by carrying out a comprehensive search of the literature, we have captured findings from a broad range of languages and geographical and cultural settings across four continents. The consistency of psychometric findings supports the validity and reliability of this patient-reported outcome measure.
However, several limitations must be acknowledged. While there is no clear consensus on the ‘criterion standard’ methodology for MCID estimation [17,36,37], patient-perceived anchors have been recommended as the primary source of evidence [16]. In our study, delivery mode was the most robust anchor available in the existing literature. However, it represents a known-groups comparison and likely captures a moderate-to-large clinical difference rather than a truly ‘minimal’ threshold. This is reflected in our anchor-based estimate of 10, which was substantially higher than our distribution-based estimate of 7, and nearly double the patient-anchored estimate of 5 reported by Mazoué et al [14]. Consequently, we weighted the distribution-based estimate as primary evidence for the MCID, as it likely reflects the MCID more accurately in the context of this meta-analysis.
In addition, while we have attempted to mitigate the impact of study heterogeneity by limiting our meta-analysis to ObsQoR-10 scores collected within 24 ± 6 h of delivery, considerable variability in protocols, study settings and populations remains. Patient clinical status can also change considerably in this 12-h time window, introducing variance, which can potentially inflate the standard deviation used for distribution-based MCID estimates (calculated as half of the standard deviation). Moreover, the number of studies eligible for inclusion in the meta-analysis was limited by variability in study protocols and reported results (online Supporting Information Appendix S6). Our selection of this window of 24 ± 6 h postpartum for our meta-analysis was a pragmatic balance. A narrower time window of 24 ± 4 h postpartum would have excluded almost a third of the studies eligible for the meta-analysis and this time frame is consistent with clinical practice. Furthermore, our study showed considerable differential item functioning, particularly in the Japanese translation. This suggests that the MCID is unlikely to be universal and needs to be applied selectively with caution across translations and contexts. Larger prospective, multicentre longitudinal studies using standardised protocols, narrower ObsQoR-10 assessment windows and patient-reported global rating of change anchors are necessary to confirm these estimates and determine requirement for language specific thresholds.
In conclusion, the MCID of ObsQoR-10 ranged from 7 to 10, with 10 points likely representing an upper-bound estimate. The ObsQoR-10 showed robust psychometric properties, took < 5 min for patients to complete and validated translations were available in at least 10 languages. This underscores its potential utility for clinical applications to evaluate and monitor inpatient postpartum recovery, as well as in research settings.
Supplementary Material
Appendix S1. Search strategies
Appendix S2. Measurement properties and other relevant properties and definitions.
Appendix S3. Included study characteristics.
Appendix S4. Reported measurement properties of ObsQoR-10
Appendix S5. Risk of bias assessment and GRADE level of evidence for overall performance rating of each psychometric property using COSMIN guidelines
Appendix S6. Summary of mean ObsQoR10 scores from meta-analysis
Appendix S7. Summary of MCID estimates from anchor- and distribution-based methods.
Acknowledgements
Pervez Sultan has received funding from National Heart, Lung, and Blood Institute (R01HL166253-01A1) and National Institute of Child Health and Human Development (U54HD113142-01). Moe Takenoshita is supported by funding from the National Heart, Lung, and Blood Insitute (R01HL166253-01A1). Aleesha J Jethwa is supported by funding from the National Institute of Child Health and Human Development (R90HD118650).
References
- 1.McCauley H, Lowe K, Furtado N, Mangiaterra V, van den Broek N. Essential components of postnatal care – a systematic literature review and development of signal functions to guide monitoring and evaluation. BMC Pregnancy Childbirth; 2022; 22:448. 10.1186/s12884-022-04752-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lawn JE, Blencowe H, Oza S et al. Every Newborn: progress, priorities, and potential beyond survival. Lancet; 2014; 384:189–205. 10.1016/S0140-6736(14)60496-7. [DOI] [PubMed] [Google Scholar]
- 3.Sacks E, Langlois ÉV. Postnatal care: increasing coverage, equity, and quality. Lancet Glob. Health; 2016; 4:e442–3. 10.1016/S2214-109X(16)30092-4. [DOI] [PubMed] [Google Scholar]
- 4.Association of College of Obstetricians and Gynecologists. Optimizing Postpartum Care, Committee Opinion, Number 736. 2018. https://www.acog.org/clinical/clinical-guidance/committee-opinion/articles/2018/05/optimizing-postpartum-care (accessed 05/11/2024). [Google Scholar]
- 5.Sharawi N, Klima L, Shah R, Blake L, Carvalho B, Sultan P. Evaluation of patient-reported outcome measures of functional recovery following caesarean section: a systematic review using the consensus-based standards for the selection of health measurement instruments (COSMIN) checklist. Anaesthesia; 2019; 74: 1439–55. 10.1111/anae.14807. [DOI] [PubMed] [Google Scholar]
- 6.Stump CM, Peralta FM, Sultan P et al. Association of inpatient postpartum recovery with patient-reported outcome measures following hospital discharge: a prospective cohort study. BMC Pregnancy Childbirth; 2024; 24: 618. 10.1186/s12884-024-06805-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sultan P, Kormendy F, Nishimura S, Carvalho B, Guo N, Papageorgiou C. Comparison of spontaneous versus operative vaginal delivery using Obstetric Quality of Recovery-10 (ObsQoR-10): An observational cohort study. J. Clin. Anesth; 2020; 63:109781. 10.1016/j.jclinane.2020.109781. [DOI] [PubMed] [Google Scholar]
- 8.Ciechanowicz S, Setty T, Robson E et al. Development and evaluation of an obstetric quality-of-recovery score (ObsQoR-11) after elective Caesarean delivery. Br. J. Anaesth; 2019; 122:69–78. 10.1016/j.bja.2018.06.011. [DOI] [PubMed] [Google Scholar]
- 9.Sultan P, George R, Weiniger CF, El-Boghdadly K, Pandal P, Carvalho B. Expert Consensus Regarding Core Outcomes for Enhanced Recovery after Cesarean Delivery Studies: A Delphi Study. Anesthesiology; 2022; 137:201–11. 10.1097/ALN.0000000000004263. [DOI] [PubMed] [Google Scholar]
- 10.Mokkink LB, de Vet HCW, Prinsen CAC et al. COSMIN Risk of Bias checklist for systematic reviews of Patient-Reported Outcome Measures. Quality of Life Research; 2018; 27:1171–9. 10.1007/s11136-017-1765-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan – a web and mobile app for systematic reviews. Syst. Rev; 2016; 5:210. 10.1186/s13643-016-0384-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wyrwich KW, Bullinger M, Aaronson N, Hays RD, Patrick DL, Symonds T. Estimating clinically significant differences in quality of life outcomes. Quality of Life Research 2005; 14:285–95. 10.1007/s11136-004-0705-2. [DOI] [PubMed] [Google Scholar]
- 13.Ciechanowicz S, Howle R, Heppolette C, Nakhjavani B, Carvalho B, Sultan P. Evaluation of the Obstetric Quality-of-Recovery score (ObsQoR-11) following non-elective caesarean delivery. Int. J. Obstet. Anesth; 2019; 39:51–9. 10.1016/j.ijoa.2019.01.010. [DOI] [PubMed] [Google Scholar]
- 14.Mazoué É, Veret M, Corroënne R et al. Translation and validation of the French version of the ObsQoR-10 questionnaire for the evaluation of recovery after delivery: the ObsQoR-10-French. BJA Open; 2023; 7:100221. 10.1016/j.bjao.2023.100221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mazda Y, Ando K, Kato A et al. Postpartum recovery of nulliparous women following scheduled cesarean delivery and spontaneous vaginal delivery: a prospective observational study. AJOG Global Reports; 2023; 3:100226. 10.1016/j.xagr.2023.100226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Revicki D, Hays RD, Cella D, Sloan J. Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes. J. Clin. Epidemiol; 2008; 61:102–9. 10.1016/j.jclinepi.2007.03.012. [DOI] [PubMed] [Google Scholar]
- 17.Mouelhi Y, Jouve E, Castelli C, Gentile S. How is the minimal clinically important difference established in health-related quality of life instruments? Review of anchors and methods. Health Qual. Life Outcomes; 2020; 18:136. 10.1186/s12955-020-01344-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hardy-Fairbanks AJ, Lauria MR, Mackenzie T, Mccarthy M. Intensity and Unpleasantness of Pain Following Vaginal and Cesarean Delivery: A Prospective Evaluation. Birth; 2013; 40:125–33. 10.1111/birt.12039. [DOI] [PubMed] [Google Scholar]
- 19.Komatsu R, Ando K, Flood PD. Factors associated with persistent pain after childbirth: a narrative review. Br. J. Anaesth; 2020; 124:e117–30. 10.1016/j.bja.2019.12.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Peahl AF, Dalton VK, Montgomery JR, Lai Y-L, Hu HM, Waljee JF. Rates of New Persistent Opioid Use After Vaginal or Cesarean Birth Among US Women. JAMA Netw.; Open 2019; 2:e197863. 10.1001/jamanetworkopen.2019.7863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pereira TRC, Souza FG De, Beleza ACS. Implications of pain in functional activities in immediate postpartum period according to the mode of delivery and parity: an observational study. Braz. J. Phys. Ther; 2017; 21:37–43. 10.1016/j.bjpt.2016.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Yang X-J, Sun S-S. Comparison of maternal and fetal complications in elective and emergency cesarean section: a systematic review and meta-analysis. Arch. Gynecol. Obstet; 2017; 296:503–12. 10.1007/s00404-017-4445-2. [DOI] [PubMed] [Google Scholar]
- 23.Pallasmaa N, Ekblad U, Gissler M. Severe maternal morbidity and the mode of delivery. Acta Obstet. Gynecol. Scand; 2008; 87:662–8. 10.1080/00016340802108763. [DOI] [PubMed] [Google Scholar]
- 24.Saulnier L, Chau A, Barrera J, Massey S. Influence of Planned Versus Unplanned Cesarean Delivery on Postpartum Peritraumatic Distress: A Prospective Observational Study. J. Obstet. Gynaecol. Can.; 2025; 47:103114. 10.1016/j.jogc.2025.103114. [DOI] [PubMed] [Google Scholar]
- 25.Karlström A. Women’s self-reported experience of unplanned caesarean section: Results of a Swedish study. Midwifery; 2017; 50:253–8. 10.1016/j.midw.2017.04.016. [DOI] [PubMed] [Google Scholar]
- 26.Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med. Res. Methodol; 2014; 14: 135. 10.1186/1471-2288-14-135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sultan P, Ando K, Sultan E et al. A systematic review of patient-reported outcome measures used to assess sleep in postpartum women using Consensus Based Standards for the Selection of Health Measurement Instruments (COSMIN) guidelines. Sleep; 2021; 44. 10.1093/sleep/zsab128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Harrison MS, Saleem S, Ali S et al. A Prospective, Population-Based Study of Trends in Operative Vaginal Delivery Compared to Cesarean Delivery Rates in Low- and Middle-Income Countries, 2010–2016. Am. J. Perinatol. 2019; 36: 730–6. 10.1055/s-0038-1673656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Vannevel V, Swanepoel C, Pattinson RC. Global perspectives on operative vaginal deliveries. Best Pract. Res. Clin. Obstet. Gynaecol. 2019; 56: 107–13. 10.1016/j.bpobgyn.2018.09.004. [DOI] [PubMed] [Google Scholar]
- 30.Fantin R, Hütter C, Lichtenberger P, Guo N, Ortner C, Sultan P. Development and validation of a German version of Obstetric Quality of Recovery-10 (ObsQoR-10-German). Int. J. Obstet. Anesth.; 2025; 61:104292. 10.1016/j.ijoa.2024.104292. [DOI] [PubMed] [Google Scholar]
- 31.Mukarram S, Ali S, Zulqurnain M et al. Validation of translated Obstetric Quality of Recovery (ObsQoR-10A) score after nonelective cesarean delivery (CD) in an Arabic-speaking population. Saudi J. Anaesth; 2022; 16: 390–400. 10.4103/sja.sja_52_22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sultan P, Kamath N, Carvalho B et al. Evaluation of inpatient postpartum recovery using the Obstetric Quality of Recovery-10 patient-reported outcome measure: a single-center observational study. Am. J. Obstet. Gynecol; MFM 2020; 2:100202. 10.1016/j.ajogmf.2020.100202. [DOI] [PubMed] [Google Scholar]
- 33.van den Bosch OFC, Rijsdijk M, Rosier SE et al. The implementation of intrathecal morphine for caesarean delivery into clinical practice, and assessment of its impact on patient-reported quality of recovery using the ObsQoR-10-Dutch scale. Eur. J. Anaesthesiol; 2025; 42:332–9. 10.1097/EJA.0000000000002127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Dusitkasem S, Jindapitak J, Arnuntasupakul V et al. Validation of the Thai version of the obstetric quality of recovery score (obsqor-10-Thai) after elective cesarean delivery. BMC Anesthesiol.; 2023; 23: 72. 10.1186/s12871-023-02010-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Kozanhan B, Yildiz M, Polat A et al. Development and Validation of a Turkish Version of Obstetric Quality of Recovery-10. Turk. J. Anaesthesiol. Reanim; 2022; 50:366–72. 10.5152/TJAR.2022.21441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Olsen MF, Bjerre E, Hansen MD, Tendal B, Hilden J, Hróbjartsson A. Minimum clinically important differences in chronic pain vary considerably by baseline pain and methodological factors: systematic review of empirical studies. J. Clin. Epidemiol; 2018; 101:87–106.e2. 10.1016/j.jclinepi.2018.05.007. [DOI] [PubMed] [Google Scholar]
- 37.Franceschini M, Boffa A, Pignotti E, Andriolo L, Zaffagnini S, Filardo G. The Minimal Clinically Important Difference Changes Greatly Based on the Different Calculation Methods. Am. J. Sports Med; 2023; 51:1067–73. 10.1177/03635465231152484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Guevara J, Sánchez C, Organista-Montaño J, Domingue BW, Guo N, Sultan P. Development and validation of a Spanish version of the Obstetric Quality of Recovery-10 item score (ObsQoR-10-Spanish). BJA Open; 2024; 10:100269. 10.1016/j.bjao.2024.100269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Hayoun DH Ben, Sultan P, Eshel R, Guo N, Carvalho B, Weiniger CF. Inpatient postpartum recovery following episiotomy: secondary outcomes of a prospective observational study. Int. J. Obstet. Anesth; 2026; 65:104798. 10.1016/j.ijoa.2025.104798. [DOI] [PubMed] [Google Scholar]
- 40.Mathias LAST, Carlos RV, Siaulys MM et al. Development and validation of a Portuguese version of Obstetric Quality of Recovery-10 (ObsQoR-10-Portuguese). Anaesth. Crit. Care Pain Med; 2022; 41:101085. 10.1016/j.accpm.2022.101085. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix S1. Search strategies
Appendix S2. Measurement properties and other relevant properties and definitions.
Appendix S3. Included study characteristics.
Appendix S4. Reported measurement properties of ObsQoR-10
Appendix S5. Risk of bias assessment and GRADE level of evidence for overall performance rating of each psychometric property using COSMIN guidelines
Appendix S6. Summary of mean ObsQoR10 scores from meta-analysis
Appendix S7. Summary of MCID estimates from anchor- and distribution-based methods.