Introduction
The coronavirus outbreak was declared as a pandemic in March 2020 (1). The pandemic caused health problems and negatively affected societies biopsychosocially (2,3). The pandemic may affect perinatal women more because they are more vulnerable to environmental stressors (4-6). In addition, pregnant women are more affected by the adverse consequences of coronavirus disease 2019 (COVID-19) concerning decreased pulmonary capacity in pregnancy, altered immunity, and response to viral infections (7). According to the Centers for Disease Control and Prevention, pregnant women are more likely to be hospitalized, be admitted to intensive care units, receive mechanical ventilation support, and die from COVID-19 than women of reproductive age with COVID-19 (8). COVID-19 obstetric complications in pregnant women result in an increased need for emergency cesarean section and potential neonatal infection (5,7).
In addition to all these adverse conditions, women in the perinatal period are particularly disadvantaged during the COVID-19 pandemic due to more frequent emotional fluctuations and psychological changes (8). Pregnant women experience more mental health problems because they are afraid of contracting the coronavirus and are concerned about the potential effects of the virus on the fetus and newborn. Research has shown that stress, depression, and anxiety are more common among pregnant and postpartum women during the COVID-19 pandemic than before (6,9-13). During this period, almost seven in ten pregnant women experienced anxiety and depression (14,15). The prevalences of anxiety and depression among Turkish pregnant women during the pandemic were 63.9-64.5% and 27.7-56.3%, respectively (3,16).
In the context of perinatal health care during the pandemic, healthcare professionals should physiologically assess women and understand COVID-19-related stressors that affect their mental health. However, there is no Turkish scale to measure the perinatal perception of COVID‑19 stressors. The COVID-19 Perinatal Perception Questionnaire (COVID-19-PPQ) was developed by Hulsbosch et al. (6) to determine COVID-19-related stressors affecting prenatal and postnatal Dutch women. This study was evaluated the Turkish validity and reliability of the questionnaire.
Methods
This methodological study was conducted between October 1, 2021, and June 30, 2022, in the antenatal and newborn clinics of a Health Services Practice and Research Hospital in Turkey.
Participants
It is recommended to take a sample of 5 to 10 times the number of items on a scale for adaptation (17). Therefore, the sample consisted of 150 pregnant women and 150 postpartum women admitted to the antenatal and newborn clinics of the hospital.
Inclusion and Exclusion Criteria: The inclusion criteria were; speaking Turkish, living in Turkey, being over 18 years of age, being in the 12th gestational week to 8-10 weeks postpartum (18), having no mental problems (anxiety disorder, depression, severe psychosocial problems, etc.). The exclusion criteria were; failing to fill out the data collection tools and wanting to withdraw from the study.
Data Collection Tools
The Individual Information Form (IIF) was based on a literature review (4-6,11-13,15). The form consisted of two parts, each with 11 items on individual and obstetrical characteristics. Pregnant women filled out the first part, while postpartum women filled out the second part. A pilot study was conducted, and the items were revised based on the results.
The COVID-19 Perinatal Perception Questionnaire (COVID-19-PPQ): The scale was developed by Hulsbosch et al. (6) to measure COVID‑19 perinatal perception during pregnancy and postpartum period. The questionnaire allows researchers to determine COVID-19-related stressors. The four-point Likert-type scale consists of 19 items. It has two scales: pregnancy (nine items; min score =0, max score =27) and postpartum (ten items; min score= 0, max score =30). The pregnancy scale has three subscales: risk of infection, contact, and future. The postpartum scale also has three subscales: first postpartum week, COVID-19 measures, and fear of infection. Seven postpartum scale items (3-9) are reverse scored. Higher scores indicate higher rates of negative COVID-19-related stress perceptions during pregnancy and postpartum. The pregnancy and postpartum scales have Cronbach’s alpha (a) values of 0.71 and 0.64, respectively (6).
The Fear of COVID-19 Scale (FCV-19S): It was developed by Ahorsu et al. (19). It was adapted to Turkish by Bakioğlu et al. (20). The five-point Likert-type scale consists of seven items. Higher scores indicate greater fear of COVID-19 (min score= 7, max score= 35). The Turkish version’s Cronbach’s alpha value is 0.84 (20). Cronbach’s alpha values were 0.87 (pregnancy) and 0.90 (postpartum period) in the present study.
The Edinburgh Postpartum Depression Scale (EPDS): It was developed by Cox et al. (21), and adapted to Turkish by Aydin et al. (22). The American College of Obstetricians and Gynecologists (2015) recommends using the EPDS to screen for perinatal depression during pregnancy and postpartum period (23). The instrument consists of ten items, with higher scores indicating a higher risk for depression (min score =0, max score =30) (22). The Turkish version’s Cronbach’s alpha value is 0.72 (22). The Cronbach’s alpha values were 0.84 (during pregnancy) and 0.87 (postpartum period) in the present study.
Ethical Approval
This study was approved by the National Ministry of Health, the Health Services Practice and Research Hospital (approval number: E-93596471-010.01-116695) and Ethics Board (ethics approval number: 2022/02/11). Electronic written consent for this study’s use of the scales were taken for the COVID-19-PPQ and the FCV-19S. Informed consent was obtained from all participants included in this study.
Procedure
This study had three stages: language and content validity, implementation, and analysis (Figure 1).
Language and Content Validity: Two translators translated the scale from English into Turkish. A third translator reviewed the original scale and the two translated versions and translated the scale from English into Turkish again. A fourth translator (a native English speaker) translated the document from Turkish back into English (24). Finally, a draft was created.
Lawshe’s (25) method was used to assess content validity. Thirteen experts were consulted and asked to rate each item on a scale of 1 to 3 for items clarity/essentiality (1= not clear/essential, 2 =item needs some revision/useful, but not essential, and 3 =very clear/essential) (25). The items were revised based on expert feedback and were evaluated by experts in terms of conformity to Turkish.
Implementation: A pilot study was conducted with 15 pregnant and 15 postpartum women randomly selected based on the inclusion criteria. The sample of the pilot study was not included in the main study. The purpose of the pilot test was to evaluate the intelligibility and relevance of the IIF and the COVID-19-PPQ. After the pilot study, the researchers finalized the scale.
The researchers briefed all women in the antenatal (n=330) and newborn clinics (n=355) about the research purpose and procedure. They explained the inclusion criteria after they were invited to the study. The pregnant women filled out the IIF, the pregnancy scale of the COVID-19-PPQ, the FCV-19S, and the EPDS (n=150). The postpartum women filled out the IIF, the postpartum scale of the COVID-19-PPQ, the FCV-19S, and the EPDS (n=150).
Analysis: The data were analyzed using the Statistical Package for Social Sciences (SPSS, IBM version 26, Chicago, IL, USA) and Analysis of Moment Structures at a significance level of 0.05. Number (n), percentage (%), mean, and standard deviation were used for descriptive statistics. Content and construct validity were assessed. Cronbach’s alpha, equivalent-form reliability, and item analysis methods were used to assess the scale’s reliability.
Content validity ratios (CVRs) allow researchers to keep or remove items from a scale. The minimum values of the CVRs and the content validity index (CVI) should be 0.54 because the number of experts was 13. After determining which items to include in the scale, the CVI was computed for the total scale. The CVI is the CVR’s average value of the retained items (25).
A confirmatory factor analysis (CFA) was conducted to construct validity. The CFA revealed factor loadings and path coefficients, which should be greater than 0.50 to obtain an acceptable scale structure (26). The ratio of chi-square to the degree of freedom (x2/df) is used to identify the conformity of the model to the data. x2/df should range from 2 to 5. The Root Mean Square Error of Approximation (RMSEA) should range from 0 to 0.08. The Goodness of Fit Index (GFI) has a cut-off point of 0.90. A Standardized Root Mean Square Residual (SRMR) as high as 0.08 is deemed acceptable. A Comparative Fit Index (CFI) ≥0.95 indicates a good fit (27). The Tucker Lewis Index (TLI)/Not-Normed Fit Index (NNFI) close to 1 indicates a good fit, while values below 0.90 indicate a need to respecify the model. The Incremental Fit Index (IFI) should be ≥0.90 for an acceptable model (28).
Cronbach’s alpha indicates internal consistency. Methodologists recommend a minimum Cronbach’s alpha value of 0.65 and regard a Cronbach’s alpha of less than 0.50 as unacceptable. A maximum value of 0.90 is recommended (29).
Equivalent forms have identically functioning items. The FCV-19S and the EPDS were equivalent forms in this study. Each participant’s score should be correlated to achieve a stable tool. A Pearson’s correlation coefficient (r) of less than 0.3 indicates a weak correlation, and 0.3< r <0.5 indicates a moderate correlation. An r greater than 0.5 indicates a strong correlation (30).
An item analysis allows researchers to determine which items to keep, revise, or remove (31). This study evaluated the relationship between the items and the total score in item analysis (especially corrected item-total correlation). The item discrimination index (>0.30= good, 0.10-0.30= fair, and <0.10= poor) is the correlation between an item and a test (32).
Results
Participants
All participants were Turkish. Pregnant women had a mean age of 29.33±6.94 years. More than half of the pregnant women had high school degrees (70%). Most pregnant women did not work in a paid job (82.7%). Eleven pregnant women experienced COVID-19. When the characteristics of postpartum women were analyzed, postpartum women had a mean age of 29.91±6.04 years. More than half of the postpartum women had high school degrees (63.9%). Most pregnant women did not work in a paid job (84.7%). Four postpartum women experienced COVID-19 (2.6%) (Table 1).
Pregnant women had a mean pregnancy scale score of 12.22±5.68 (range 0-24). Postpartum women also had a mean postpartum scale score of 19.63±5.42 (range 0-30) (Table 2). Table 3 shows all participants’ FCV-19S and EPDS scores.
Validity
Content validity: On the pregnancy scale, all items but 9 had CVRs of 0.54 to 1.00. Item 9 had a CVR of 0.23. The acceptable minimum value of CVR and CVI was 0.54. Therefore, item 9 was removed from the scale, and CVI was recalculated. The scale had a CVI of 0.77. Of the postpartum scale, all items had CVR and CVI values greater than 0.54 (Table 4).
Construct validity: The pregnancy and postpartum scales of the COVID-19-PPQ consist of three factors. Therefore, a CFA was performed for the three-factor structure. The measurement model established to confirm the scale structures was analyzed (Figure 2). The pregnancy scale had factor loadings of 0.52 to 0.90. The postpartum scale had factor loadings of 0.69 to 0.93 (Table 4). There was no item factor loading lower than 0.5. Therefore, model fit indexes were examined. c²/sd >2.0, RMSEA and SRMR ≤0.08, GFI, TLI/NNFI, and IFI >0.90, CFI ≥0.95 for the pregnancy and postpartum scales. These results confirmed the model (Table 5).
Reliability
Cronbach’s alpha coefficient: The total pregnancy scale and subscales had Cronbach’s alpha values of 0.65 to 0.87. The total postpartum scale and its subscales had Cronbach’s alpha values of 0.81 to 0.90 (Table 2). The Cronbach’s alpha was acceptable as it was greater than 0.65 and less than 0.90.
Equivalent-form reliability: There was a positive correlation between the scores of the pregnancy scale and the FCV-19S (r=0.459, p<0.001). Moreover, there was a positive correlation between the scores of the postpartum scale and the EPDS (r=0.459, p<0.001). Table 6 shows the correlations between the subscale scores and FCV-19S or EPDS scores.
Item analysis: The corrected item-total correlation values ranged from 0.33 to 0.68, indicating that all items had “good” discrimination. Table 4 shows the scale means if the item is deleted, the scale variance if the item is deleted, and the squared multiple correlation values of the pregnancy and postpartum scales.
Discussion
The COVID-19 pandemic significantly impacted the mental well-being of vulnerable women during the perinatal period (2,33). This study adapted the COVID-19-PPQ to Turkish to measure COVID-19-related stressors during pregnancy and postpartum period.
Some experts assessed item 9 of the pregnancy scale as “not essential or item needs some revision.” They thought that item 7 and item 9 measured the same construct, and that item 7 was not about pregnancy. Item 9 was removed from the scale because it had a CVR of 0.23. Content validity was achieved because the total scale had a CVI of 0.77 (CVI >0.54) (25).
The CFA results showed that the models for the pregnancy scale (eight items; three subscales) and postpartum scale (ten items; three subscales) had acceptable fit indices. Therefore, the models seemed fit (27,28). The original pregnancy and postpartum scales also have adequate, and excellent model fits, respectively (6).
Item 3 (factor loading: 0.52), and item 1 (factor loading: 0.57) had factor loadings close to the lower limit. The items were acceptable because they had factor loadings of greater than 0.50 (26). Hulsbosch et al. (6) reported that items 1 and 3 had factor loadings of 0.64 and 0.51, respectively. These items are about antenatal visits. Pregnant women are hesitant to attend antenatal visits because they are worried about the “risk of infection” and “contact” (item 3 in factor 1. Risk of infection, and item 1 in factor 2. contact). Turan et al. (34) also found that half of the pregnant women made fewer antenatal visits during the COVID-19 pandemic, which might affect women’s responses to these items.
In addition, these factors had Cronbach’s alpha values close to the cut-off value (Factor 1. Risk of infection a=0.74, Factor 2. Contact a=0.65). In this study, all subscales had Cronbach’s alpha values of 0.65 to 0.90, indicating high internal consistency for the pregnancy and postpartum scales (29,35). Hulsbosch et al. (6) reported similar Cronbach’s alpha values, suggesting that the COVID-19-PPQ was reliable.
The pregnancy scale has three subscales. The subscale “the risk of infection” concerns testing positive for COVID-19 during pregnancy. The subscale “contact” concerns cancellations of ultrasounds, having ultrasounds alone, and family and friends not visiting. The subscale “future” is about financial and work-related concerns. All three subscales focus on fear or worry (6). Our results showed that the pregnancy scale and its subscales significantly correlated with the FCV-19S. Hulsbosch et al. (6) also reported that the pregnancy scale and its subscales significantly correlated with pregnancy-specific distress symptoms. Research also showed that pregnant women were more worried and afraid of COVID-19 (10,11). However, the present study has no significantly correlated total pregnancy scale and the EPDS.
Similarly, Boekhorst et al. (36) did not document any findings regarding the increased prevalence of depression among pregnant women during the pandemic. There was a significant correlation between the “infection” subscale and the EPDS. However, the other two subscales (contact and future) were not correlated with the EPDS. This suggests that pregnant women are more negatively affected by the possibility of having COVID-19 than by the indirect effects of COVID-19, such as lack of visits from friends and financial problems.
The postpartum scale has three subscales. The subscale “first postpartum week” refers to the perception of COVID-19-related changes due to fewer visits. The subscale “COVID-19 measures” concerns the perception of measures and guidelines during delivery. The subscale “fear for infection” is about concerns regarding getting infected (self, baby, or partner) with COVID-19 during delivery or in the first postpartum week (6). These subscales were correlated with the FCV-19S. The subscale “fear for infection” was significantly correlated with the EPDS, suggesting that the first postpartum week and COVID-19 measures were about visits and the birth period. In the postpartum period, labor is over, and the woman focuses on her baby and breastfeeding. Therefore, she might move away from the stressors of COVID-19. Hulsbosch et al. (6) stated that the “first postpartum week” subscale was not correlated with symptoms of postpartum depression and anxiety.
Strengths and Limitations of the Study
The study had three strengths. First, this is the first study to adapt the COVID-19-PPQ to Turkish. Thus, this scale can be used in clinical practices in perinatal health services during the COVID-19 pandemic. Second, the COVID-19-PPQ is a user-friendly scale that includes both pregnancy and postpartum scales. Therefore, it can be used as a screening tool in all areas where antenatal and postnatal healthcare is offered in Turkey during the COVID-19 pandemic. Thus, this study contributes to develop a national perinatal care policy and strategy for this vulnerable group. Third, validity and reliability were established face-to-face in the hospital during the COVID-19 pandemic.
Study Limitations
The study had three limitations. First, the results were sample-specific and could not be generalized. Second, the study was conducted in a city in Turkey. Third, women with healthy pregnancies and healthy babies were included in the study. Perception might change in high-risk pregnancies, and this was also true in the postpartum period. It was also suggested to validate the COVID-19-PPQ in other regions of Turkey.
Conclusion
The COVID-19-PPQ is a valid and reliable scale used to assess perinatal COVID-19-related stress perception. Future research should examine the use of the scale in clinical practice during the COVID-19 pandemic. It is recommended to adapt the COVID-19-PPQ to other countries.
Ethics
Ethics Committee Approval: This study was approved by the National Ministry of Health, the health Services Practice and Research Hospital (approval number: E-93596471-010.01-116695) and Ethics Board (ethics approval number: 2022/02/11). Electronic written consent for this study’s use of the scales were taken for the COVID-19-PPQ and the FCV-19S.
Informed Consent: Informed consent was obtained from all participants included in this study.
Peer-review: Externally peer reviewed.
Authorship Contributions
Concept: H.A., S.K., Design: H.A., S.K., Data Collection or Processing: B.Ö.G., Analysis or Interpretation: H.A., B.Ö.G., S.K., Literature Search: H.A., Writing: H.A., S.K.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.