Design and Psychometric Evaluation of Mothers' Perception of Postpartum Care in Comprehensive Health Centers

Background: In order to properly evaluate postpartum services in comprehensive health centers, encouraging women to use these services and presence of a reliable and valid tool, is necessary to design and assess mothers' perceptions of postpartum care. Aim: Designed to evaluate the psychometric dimensions of Iranian version of Mothers' Perception of Postpartum Health care Questionnaire at Comprehensive Health Centers. Methods: This is a cross-sectional study. The verbatim translation was used to translate the questionnaire from English to Persian. Face, content, and structural validity as well as reliability of the questionnaire were evaluated. Findings: The results showed that the face, content, and structure validity as well as reliability of this questionnaire to be appropriate. The calculated Cronbach's alpha coecient for the whole questionnaire was 0.668. Interclass correlation coecient was 0.688, which conrmed divergent validity. Conclusion: Mothers' Perception of Postpartum Health care Questionnaire at Comprehensive Health Centers is a valid and reliable questionnaire to measure mothers’ perceptions of these vital services. This study evaluated the psychometric properties of mothers' perception of postpartum care in comprehensive health centers in Iran. The validity and reliability of each questionnaire tool ensure its accuracy and precision. Based on the ndings of this study, the MPQ has appropriate face, content, and construct validity. Load factor of all items in the questionnaire was signicant at α = 0.05. Given that this questionnaire was designed and validated for the rst time, no similar study was found. All other published studies have evaluated psychiatric satisfaction, perception, or misconduct in postpartum care in hospitals. All items in this questionnaire have impact coecient above 1.5, CVI above 0.79, and the questionnaire has face and content validity. The study by Taavoni and colleague aimed to design and test a Respectful Maternity Care Questionnaire. The results of their study showed that all items had impact coecient above 1.5 and the CVR of the questionnaire was 0.74, which was acceptable(6) .Their results were similar to the current study. The CVI, CVR and S-CVR were all acceptable items for the MPQ. Study by Sheferaw and co-workers in Ethiopia to validate 15-item questionnaire that assessed the mothers’ perception of respectful prenatal care in health centers showed that the items in the questionnaire had construct validity of 0.76 to 0.82, and the entire questionnaire had acceptable construct validity ,ndings of this study conrmed the structural validity of the two factors of perception of midwifery care in postpartum and postpartum care (11). The study published by Afulani and others aimed to develop a person-based maternal care assessment tool. This questionnaire have four factors and 38 items. EFA ndings conrmed the validity of 30 items in three factors for urban and rural communities (18). Conrmatory factor analysis of the study by Taavoni and colleague revealed that the Respectful Maternity Care Questionnaire related to mothers' satisfaction with postpartum nursing care had a good relationship between items and acceptable value(6). This research is close to the present research, but since the questionnaire has already been available in some languages, including English and has been validated, conrmatory factor analysis was used. Cronbach's


Introduction
Postpartum care is an essential component of women's reproductive health (1). An important part of this care is provided after discharge from hospital or home birth, in comprehensive health centers and by health care providers, especially midwives (2). Mothers' perception of the quality and quantity of the care that they have received as well as their relationship with health care staff have a signi cant impact on the continuity of care and their subsequent future referrals to receive care (3). Every woman referred to a health care center has the right to compassionate care, which includes dignity, respect for their independence and feelings, and respect for their cultural and family preferences(1) (1).

Background
There have been numerous reports from around the world of unfair, distant, and disrespectful care during pregnancy and later in the health centers (4). Mothers' satisfaction with postpartum care closely depends on how health care providers treat them (2). Unsatisfying experience and a negative understanding of the health care is an important obstacle to access health care (5). Some studies have examined women's negative experiences with care provided during and after pregnancy (6,7). Unacceptable behavior and misconduct in managed care and the disruption in communication between the mothers and nurses are the major challenges of prenatal care and after care, especially in developing countries, to which many maternal mortality is attributed (8). Appropriate and reliable tools must be used for accurate measurement of any study subject (9). The questionnaires available to measure mothers' perceptions of postpartum care are all designed and used to evaluate postpartum nursing care in the hospital only (2,5,10); whereas, most of postpartum and child care is provided at health centers. Patient-centered reproductive care has been recognized as an important factor in improving overall reproductive health. However, the number of studies on how to implement it is very limited (11). A valid and reliable tool is needed to assess mothers' perception of health care and encourage them to use existing care services. In addition, such tools would help management care teams to better assess their patients' satisfaction and make the necessary changes to meet their patients' needs (8). In order to properly evaluate postpartum services in comprehensive health centers, encouraging women to use these services and presence of a reliable and valid tool, is necessary to design and assess mothers' perceptions of postpartum care. The overall purpose of this study was to design a tool and assess its validity and reliability to evaluate mothers' perception of postpartum health care.

Study design:
This cross-sectional study was conducted in 2018 to assess the validity and reliability of the Persian version of the psychometric analysis of "Mothers' Perceptions of Care Provided by Postpartum Health Care Workers". Sample size was different (quantitatively and qualitatively) at different stages of this study. To assess the validity and reliability of samples, ten primary health centers providing postpartum care were randomly selected. Out of the women referred to these centers, 250 women were randomly sampled. This study was conducted in Shiraz, southwestern Iran. The face and content validity of the questionnaire were assessed using the opinions of ten academic professors specializing in psychiatry, reproductive health, and comprehensive health staff.

Data Collection
Analysis:

Translation
The questionnaire items were originally in English. The questionnaire was translated into Persian by a person uent in both English and Persian, and then translated back into English by another person uent in the both languages. In order to validate the questionnaire, another expert (independent of the two individuals) reviewed the translation. Finally, two English and Persian experts compared the questionnaire and con rmed the accuracy of the translation.

Validity
The validity of mothers' perception of caregiving questionnaire was presented. This questionnaire is a part of a 100-item questionnaire of Goldberg's Five Factor Model (12), which was improved based on the characteristics of the original questionnaire. Goldberg questionnaire is valid and reliable worldwide. The scale of satisfaction for this questionnaire was based on the scale designed by Hall and co-workers (13). Although the scale was designed for older population, it has been modi ed by Dimatteo and Hays and is used to assess physicians' satisfaction (14). Because of the nature of the items in these questionnaires and their validity (15), in this study, a part of the questionnaires was selected and psychometrically designed to have a valid and reliable tool for assessing mothers' perceptions of care provided in comprehensive health centers. The mothers' perception questionnaire (MPQ) consists of 18 main items, including participants' satisfaction with caregiver interactions, communication, behavior, and professional competency in educating patient about anxiety management. Additionally, MPQ contains two sub-items including perception of midwifery care in postpartum and postpartum care. This questionnaire was completed and self-reported by the participants within a 3-month postpartum period. Cronbach's alpha of 0.82 was used to determine the internal correlation coe cient of each item. It showed su cient reliability between items and sub-items. The validity of the MPQ was assessed by evaluating face validity, content validity, and construct validity (exploratory factor con rmation). Quantitative and qualitative methods were used to determine the face validity of the MPQ. To assess the face validity using the qualitative method of Persian version of the MPQ, initially few employees of the selected health centers and mothers referring to these centers (n = 10) commented on the questionnaire. Their views on the appropriateness and clarity of the items were applied to the questionnaire items. It is important to consider the target group's judgment of the relevance and comprehensiveness of the questionnaire items and these items should assess the important areas for the target group (16). Impact score was used to determine quantitative face validity. Comprehensive health center staff and reproductive health professionals (n = 10) evaluated the importance of each item using the following formula. Impact Score = Frequency (%) × Importance. A coe cient of 1.5 and above is acceptable (13). At this stage, the importance of each item was measured using the 5-point Likert scale: quite important (5 points), somewhat important (4 points), moderately important (3 points), slightly important (2 points), and not at all important (1 point). The items with an effect size of less than 1.5 score were modi ed. In addition to the opinion of a number of staff working in comprehensive health centers, the views of the literature experts on the writing, wording, and appearance of the questionnaire items were utilized.
Quantitative method was used to determine the content validity of the MPQ. At this stage, the opinions of some (10 individuals) faculty members who specialized in maternal care and reproductive health were used. Content validity index (CVI) was performed based on Waltz and Basel CVI (17). This index assesses the relevance, clarity, and simplicity of the questionnaire questions on a four-point Likert scale. The CVI score for each phrase was calculated through dividing the number of experts agreeing with the rankings 3 and 4 by the total number of experts. The CVI of greater than 0.79 was considered appropriate (16). Exploratory factor analysis (EFA) was used to determine construct validity. 250 women who referred to comprehensive health centers after childbirth were selected by purposeful sampling. The results were analyzed using SPSS software version 22.0. Reliability shows the reproducibility, and accuracy of the tool (14). The reliability of the Persian version of the MPQ was calculated by determining Cronbach's alpha (determining internal consistency) of the entire questionnaire and its sub-items.

Results
250 women participated in this study after referral to comprehensive health centers. Their mean age was 30 ± 5.54 years, and most of them were housewives (90.4%, 226 participants). Most of the individuals had planned pregnancy 86% (215 individuals) and used comprehensive health services during and after pregnancy. Table 1 shows the demographic and characteristics of the samples. The impact coe cient of all the items was above 1.5, and the questionnaire had appropriate and acceptable face validity. The minimum acceptable CVI value was 0.79, which was higher than all the items in this questionnaire. The results showed that the CVI of all the items was 1.8-2.5 with scale-level index (S-CVI) of 0.73. The questionnaire items had S-CVI of 0.73 and content validity ratio (CVR) ranged 0.62-1.0. The CVR of questionnaire items ranged between 0.62 and 1.0. The CVR value of the two items was less than 0.62; these items were modi ed and reevaluated and after reviewing the CVR values, they reached acceptable level. In order to re ne the writing and wording of each phrase, the items were edited by a literary expert. EFA was performed on 250 completed questionnaire. Output results are described below. Kaiser-Meyer-Olkin (KMO) index value at rst output was 0.825. Both positive and negative skewness of the data were within acceptable range; thus, the sample size selected was su cient for factor analysis. According to Bartlett's test, factor analysis was signi cant for identifying factor model structure at P < 0.001, indicating detectable relationships among the analyzed variables. The second output of EFA which comprised of two parts, the initial eigenvalue, and the extraction or rotation sums of squared loadings are listed in Table 2. The results of the rotational correlation matrix table showed that a set of 18 items related to the mothers' perception of postpartum care measured two major factors. The rst factor consisted of six items with their loading factor ranging from minimum of 0.254 to a maximum of 0.698. The second factor consisted of 12 items that were investigated in three dimensions. The rst dimension consisted of seven items with a load factor from minimum of 0.335 to a maximum of 0.835. The second dimension consisted of three items with minimum operational load of 0.612 and maximum of 0.675. Finally, the third dimension consisted of two items with minimum operational load of 0.335 and maximum of 0.637. Chart 1 (see Supplementary Files) descript factor analysis to determine the correlation between items. The rst factor had six items and one of the items had a weak correlation coe cient with the other items and its reliability was low; thus, the item was eliminated by examining research group and Cronbach's alpha reliability coe cient increased from 0.717 to 0.84. These ve items have a load factor of 0.578-0.883. The second factor had 12 items and three items were eliminated because of low correlation coe cient. Load factor of these nine items are 0.686-0.853. Load factor for the nal 12-question model are listed in Table 3. Estimation of all factor loadings was signi cant at 0.05. Table 3: Rotated matrix factor analysis of components of mothers' perception questionnaire Internal consistency coe cient was used to determine the internal validity of the questionnaire. The Cronbach's alpha values for each item and the entire questionnaire, after factor analysis, are presented in Table 4.
Cronbach's alpha coe cient for the questionnaire was calculated to be 0.668. Intraclass correlation coe cient (ICC) was 0.688 which con rms divergent validity. The cross-correlations of the structures were less than the root mean square of the extracted variances, indicating a divergent validity model. The nal results showed su cient reliability of the questionnaire. Table 4 show Correlation, validity, and homogeneity of perceptions questionnaire.

Discussion
This study evaluated the psychometric properties of mothers' perception of postpartum care in comprehensive health centers in Iran. The validity and reliability of each questionnaire tool ensure its accuracy and precision. Based on the ndings of this study, the MPQ has appropriate face, content, and construct validity. Load factor of all items in the questionnaire was signi cant at α = 0.05. Given that this questionnaire was designed and validated for the rst time, no similar study was found. All other published studies have evaluated psychiatric satisfaction, perception, or misconduct in postpartum care in hospitals. All items in this questionnaire have impact coe cient above 1.5, CVI above 0.79, and the questionnaire has face and content validity. The study by Taavoni and colleague aimed to design and test a Respectful Maternity Care Questionnaire. The results of their study showed that all items had impact coe cient above 1.5 and the CVR of the questionnaire was 0.74, which was acceptable (6) .Their results were similar to the current study. The CVI, CVR and S-CVR were all acceptable items for the MPQ. Study by Sheferaw and co-workers in Ethiopia to validate 15-item questionnaire that assessed the mothers' perception of respectful prenatal care in health centers showed that the items in the questionnaire had construct validity of 0.76 to 0.82, and the entire questionnaire had acceptable construct validity , ndings of this study con rmed the structural validity of the two factors of perception of midwifery care in postpartum and postpartum care (11). The study published by Afulani and others aimed to develop a person-based maternal care assessment tool. This questionnaire have four factors and 38 items. EFA ndings con rmed the validity of 30 items in three factors for urban and rural communities (18). Con rmatory factor analysis of the study by Taavoni and colleague revealed that the Respectful Maternity Care Questionnaire related to mothers' satisfaction with postpartum nursing care had a good relationship between items and acceptable value (6). This research is close to the present research, but since the questionnaire has already been available in some languages, including English and has been validated, con rmatory factor analysis was used. Cronbach's alpha for the questionnaire is 0.668, and the internal consistency and reliability of this questionnaire is quite appropriate and demonstrates its reliability at a large level. In Afulani and co-workers' study, Cronbach's alpha was 0.8, and all factors had satisfactory value for reliability and applicability (18). Overall, the modi ed 12 main items and two sub-item questionnaire of mothers' perception of postpartum care in comprehensive health centers is a valid and reliable questionnaire to measure mothers' perceptions of these vital and useful services.

Conclusion
This questionnaire can be used to identify the strengths and weaknesses of the care provided by midwives and health care providers. There have been many studies conducted on mothers' satisfaction, mothers' perception, and malnutrition in prenatal or post-natal care in different hospitals around the world, leading to the development of reliable and valid tools, especially in developing countries where health care is poorer. This has led to the design or con rmation of the validity and reliability of these tools (8,10), offering essential assets to managed care teams and health care providers.

Declarations
Ethics approval and consent to participate This study was approved by the Ethics Committee of Shiraz University of Medical Sciences, Shiraz, Iran (No. IR.SUMS.REC.1397.522). All the participants were informed about the study objectives and signed written informed consents for taking part in the study.

Consent for publication
All presentations of this study must have consent for publication.

Availability of data and materials
Data available by Corresponding author by request

Competing interests
The authors declare that they have no competing interests.

Funding
Funding for this study was provided by Vice-chancellor of research of Shiraz University of Medical Sciences (No: 17112). VCR had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication. All authors declare that they have no con icts of interest.