Reference ranges of the WHO Disability Assessment Schedule (WHODAS 2.0) score and diagnostic validity of its 12‐item version in identifying altered functioning in healthy postpartum women

Abstract Objectives To compare scores on the 36‐item WHO Disability Assessment Schedule 2.0 tool (WHODAS‐36) for postpartum women across a continuum of morbidity and to validate the 12‐item version (WHODAS‐12). Methods This is a secondary analysis of the Brazilian retrospective cohort study on long‐term repercussions of severe maternal morbidity. We determined mean, median, and percentile values for WHODAS‐36 total score and for each domain, and percentile values for WHODAS‐12 total score in postpartum women divided into three groups: “no,” “nonsevere,” and “severe” morbidities. Results The WHODAS‐36 mean total scores were 11.58, 18.31, and 19.19, respectively for no, nonsevere, and severe morbidity. There was a dose‐dependent effect on scores for each domain of WHODAS‐36 according to the presence and severity of morbidity. The diagnostic validity of WHODAS‐12 was determined by comparing it with WHODAS‐36 as a “gold standard.” The best cut‐off point for diagnosing dysfunctionality was the 95th percentile. Conclusion The upward trend of WHODAS‐36 total mean value scores of women with no morbidity compared with those with morbidity along a severity continuum may reflect the impact of morbidity on postpartum functioning.


| INTRODUCTION
Progress in maternal health and the consequent reduction in maternal mortality are considered important goals worldwide, as part of the fifth Millennium Development Goal and, presently, of the third Sustainable Development Goal. 1,2 Since the 1990s, there has been significant improvement in maternal and perinatal health indicators, with a decline of about 50% in the overall maternal mortality rate. 3 Improvements in antenatal care, 4 as well as access to institutional deliveries, using clean delivery kits, 5 and interventions to prevent and manage hypertensive disorders and postpartum hemorrhage 6,7 have likely contributed to that reduction. However, there are still many challenges to ascertain women's health and well-being during pregnancy and the postpartum period, even among those women with no medical complications, especially in low-and middle-income countries.
The postpartum period is characterized by multiple concerns involving self-confidence, mother-infant interaction, body image experiences, adjustment to maternal roles, and attitudes. 8 These concerns-which are not only dependent on the presence or diagnosis of morbidity, but also on the inadequate assessment of such issues or even their misinterpretation-may contribute to the deterioration of maternal and newborn health. 9 The occurrence of severe maternal morbidity (SMM), defined as having a potentially life-threatening condition (PLTC) and/or maternal near miss (MNM), has been studied over the past decade: impacts on maternal and child health are unquestionable, [10][11][12] as are their effects on women's functionality (her ability to perform everyday tasks, including social and economic responsibilities). 13 Non-life-threatening or nonsevere maternal morbidity (non-SMM) is also a theme of concern and is currently defined as "any health condition attributed to and/or complicating pregnancy and childbirth that has a negative impact on the woman's well-being and/or functioning". 3 However, even among new mothers without morbidity, the inherent complexity of pregnancy and the postpartum period should be considered when providing care to this specific population.
Monitoring more than just the traditional indicators of health, pregnancy, and childbirth requires an expanded approach. Information on disability and functioning is an important component of health assessment and has provided helpful evidence for measuring disease burden across different settings, 14,15 through a tool developed by the WHO called the WHO Disability Assessment Schedule 2.0 (WHODAS 2.0). 16 This tool seeks to measure functionality, and considers six domains (cognition, mobility, self-care, getting along with people, life activities, and participation) as they apply to daily living activities in the 30 days preceding the tool's application. The complete version has 36 questions (WHODAS-36), while a shorter 12-item version is also available (WHODAS-12). However, the WHODAS has not often been applied to women of reproductive age, during pregnancy, or the postpartum period.
To better understand WHODAS scores, and their distribution, among postpartum women, with the intention of suggesting a cut-off point for screening, we applied this instrument to women with no morbidity and those on a continuum of morbidity (any morbidity to SMM).
We further validated the shorter WHODAS-12 using the full 36-item version as a reference for the assessment of disability and functioning in nonmorbid postpartum women.

This is a secondary analysis of the Brazilian Cohort on Severe Maternal
Morbidity (COMMAG)-a retrospective cohort study that included women who delivered between July 1, 2008, and June 30, 2012, at the Women's Hospital of the University of Campinas, Brazil. The study was a long-term evaluation of the consequences of PLTC and MNM incidents using a multidimensional approach. 17 The details of the methods used for the main study and primary results on WHODAS have been published elsewhere. 13,18 Briefly, cases of SMM were identified using the standardized WHO criteria, 19 as PLTC and MNM, and compared with a randomly selected control group, without SMM. One of the instruments used to assess women's functioning status was the For the current analysis, we aimed to select, from among the control group, all cases with no morbidity at all (during gestation, delivery, or postpartum). This involved applying the broad WHO definition for and criteria of maternal morbidity. 3 Cases with any previous medical condition (hypertension, diabetes, anemia, cardiovascular disease, autoimmune disease, smoking, etc.) and cases with any complications documented during any pregnancy (pre-eclampsia, hemorrhage, and infection, among others) were excluded from the no morbidity group.
It is relevant to consider that our study included evaluations/interviews with women any time from 1 to 5 years postpartum and participants could have had other pregnancies during the intervening period.
We hypothesized that any complication could impact the woman's WHODAS score. Considering this approach, we were able to identify three groups: SMM, non-SMM, and no morbidity. We performed our analyses using these three groups. We used a virtual database built for the main study using the LimeSurvey platform (www.limesurvey.org; LimeSurvey GmbH, Hamburg, Germany).
Our group translated the instrument into Brazilian Portuguese. 13 The total score for WHODAS ranges from 0-100, where a high score indicates major living limitations. 16 The shorter WHODAS-12 consists of two questions from each domain (called sentinel key questions) of the full 36-item version of the WHODAS. 16  Sociodemographic, obstetric, and perinatal characteristics were described for each group considered. We determined mean, median, and percentile values for WHODAS-36 total score, and separately for each domain. For each domain we excluded cases with missing data in any question. We also determined percentile values for WHODAS-12 total score. For the short version, the detailed analysis on domains is not possible. 16 The results were compared using the Kruskal-Wallis test between groups. A P value of 0.05 or below was considered statistically significant.
We further validated the WHODAS-12 (using the WHODAS-36 as the "gold standard" for diagnosing disability and measuring functioning) for three percentiles: P90, P95, and P97.5. Finally, the best cut-off points for screening accuracy of impaired functioning for both WHODAS-36 and WHODAS-12 were determined using receiver operating characteristic (ROC) curves for each endpoint. Our goal was to identify, from our sample, a group of women with no morbidity and to study their baseline WHODAS results in comparison with those of women presenting with maternal morbidity.

| RESULTS
Overall, there were 128 women with no morbidity (Fig. 1). Table 1 presents the sociodemographic and obstetric characteristics of these women as well as their perinatal outcomes. Women were generally young (mean age, 28.0 ± 6.4 years), multiparous, non-white, and with a partner. The majority of deliveries were vaginal and nonoperative (57%), with low rates of prematurity and neonatal deaths.
Considering the three groups of women on a continuum of severity, the WHODAS-36 total mean scores were 11.58, 18.31, and 19.19 for no morbidity, non-SMM, and SMM, respectively (Fig. 2). Table 2 shows mean and median values for each of the six WHODAS-36 domains by morbidity category. For the no morbidity group, domains 1 (cognition) and 4 (getting along with people) presented the high- To evaluate the performance of the 12-item version of the instrument, we compared WHODAS-36 and WHODAS-12 total scores for the nonmorbid group. Results were very similar (Table 3) when F I G U R E 1 Flow chart of women included in the study.
T A B L E 1 Sociodemographic and obstetric characteristics of women with no maternal morbidity (n=128).

No. (%)
Age, y 28.0 ± 6.4 were considered ( Table 4). The best cut-off point to diagnose significantly reduced function among postpartum women without any morbidity or complication was the 95th percentile, with the largest area under the curve of 0.996 (Fig. 3).

| DISCUSSION
Given that, as a result of successful efforts to decrease maternal mortality, more women than ever are now surviving childbirth, and that well-being is understood as a broad spectrum, it is relevant to China, a recently published study shows that maternal satisfaction and postpartum well-being were correlated with giving birth to male infants, and the preference for male children has been well described in Chinese culture. 24 Cultural diversity determines conceptual diversity of well-being and this may influence WHODAS score as well.
The increasing trend of WHODAS-36 total score mean value across the three groups analyzed (without morbidity, non-SMM, and SMM) may reflect the impact of complications on the new mother.
Our analysis suggests that regardless of type, complications influence a woman's perception of her functionality, which correlates with the magnitude of the complication. Otherwise, the differences between the mean scores across the three groups would be larger. Recognizing T A B L E 2 WHODAS-36 total score and specific domain values presented as mean, median, and standard deviation, for the three groups, no morbidity, nonsevere maternal morbidity, and severe maternal morbidity.    Index to a Brazilian cohort of women with and without SMM, and found similar mean scores in the population studied. 12 These issues may be inherent to the pregnancy and postpartum period regardless of the presence or absence of any morbidity.