Childbirth experience questionnaire (CEQ): development and evaluation of a multidimensional instrument

Childbirth is described as a multifaceted experience. Sense of security and perceived control, experienced level of labour pain, personal support, midwifery care, experience of earlier deliveries, intrapartum analgesia, information given and involvement in decision-making contribute to the childbirth experience [1–9]. Unplanned medical interventions during childbirth, e.g. oxytocin augmentation, emergency caesarean and operative vaginal deliveries, intrapartum complications and need of neonatal intensive care are related to maternal dissatisfaction [8–13].

First-time mothers are particularly vulnerable for negative experiences and in Sweden approximately 9% of all primiparous women are dissatisfied with the birth experience one year after the birth [13]. Negative experiences increase the risk for maternal postpartum depression and may negatively affect attitudes to future pregnancies and births and may prompt women to request caesarean delivery [14]. Therefore, knowledge of factors affecting maternal satisfaction is important in order to improve childbirth care. For this purpose, postpartum questionnaires covering multidimensional aspects are needed to comprehensively explore women's experiences of childbirth.

Existing instruments for assessing maternal experiences during labour and birth [15–18] either assess isolated aspects of the childbirth experience or aggregate diverse aspects into single overall scores. For example, the Labour Agentry Scale (LAS) [15, 19] covers only one childbirth dimension, i.e. perceived personal control during pregnancy and childbirth. Another questionnaire used by Lavender et al. [16] aggregates ratings of different aspects of childbirth experience to a single score of maternal experience. Likewise, the Wijma Delivery Expectancy/Experience Questionnaire (W-DEQ) [17] measures fear of childbirth in one domain. The Childbirth Self-efficacy Inventory (CBSEI) is also a unidimensional instrument measuring self-efficacy and coping during childbirth [18]. An exception is the Quality from the Patient's Perspective Intrapartal (QPP-I) instrument which assesses women's perceptions of childbirth care [20] in both one general factor and 10 subfactors. However, this questionnaire was developed primarily for evaluating quality of care and many of the domains assess aspects of the physical environment or the performance of different healthcare professionals. Hence, validated postpartum questionnaires covering multidimensional aspects that comprehensively explore women's perceptions of and feelings about their first labour and birth are lacking.

The aim of this study was therefore to develop and evaluate a questionnaire measuring different dimensions of childbirth experience in first time mothers.

Permission to conduct the study was obtained from the Regional Ethics Board in Gothenburg in November 1997 (L586-97).

Factor analysis of the 22 item Childbirth Experience Questionnaire (CEQ) yielded four major dimensions of the childbirth experience, namely Own capacity, Perceived safety, Professional support and Participation

These domains confirmed the areas of the childbirth experience found in our literature searches [4–9] and the interviews with mothers one month postpartum. The psychometric properties of the CEQ were also found to be good. Specifically, scaling success was achieved in all four sub-scales and the instrument discriminated well between groups hypothesized to differ in experience of childbirth.

The dimension that explained most of the variance, Own capacity, included items regarding experienced emotions and sense of control, together with experienced labour pain. This dimension corresponds with the Labour Agentry Scale [15], measuring perceived personal control; however, our factor also included an item tapping perceived pain. Labour pain is considered to be a very significant part of childbirth [9, 27] and experienced pain is likely to be strongly associated with the mother's sense of control in childbirth. Own capacity also has a component of self-efficacy (I felt capable) and coping (I felt that I handled the situation well) and thus likens the domain-specific Childbirth Self-efficacy Inventory scale [18]. Unlike the above instruments, the recently validated QPP-I measures 10 domains of intrapartal care [20]. Developed primarily for use in evaluating quality of care, many of the domains assess aspects of the physical environment or the role performance of different professionals. Although these aspects are of importance from a quality improvement perspective, our instrument focuses on the mother's experiences of her own performance and sense of support. There are obvious tradeoffs between domain-specific measures, such as the Labour Agentry Scale, and more generic measures, such as ours, that aim to comprehensively assess the childbirth experience. In future studies we plan to conduct head-on-head comparisons to examine the benefits and weaknesses of our questionnaire in relation to established domain-specific questionnaires.

Items regarding sense of security correlated with statements about memories and formed the dimension labelled Perceived safety. A possible explanation for the intercorrelations between fear, sense of security and memories from the childbirth is that distressing recollections are symptoms after experienced fear [28]. Childbirth is a stressful event and some women have traumatic stress symptoms, e.g. anxiety and fear of childbirth postpartum [29–31]. It has also been shown that feelings of safety during childbirth influence the memories from the birth [32]. This is an important area and should be included in an overall assessment of the childbirth experience. In contrast, the W-DEQ-scale specifically measures fear of childbirth with questions about anxiety, control and personal feelings in one domain [17] and does not include items regarding memory. Instead all questions are asked both before and after the birth, which is a way to include the influence on the memory. In another instrument [33] fear related to childbirth, in both men and women, is measured with 18 items about fear in four factors, all related to fear. One of the factors is about insecurity and danger, e.g. "Childbirth is a very risky situation". The safety dimension in our questionnaire includes items not unlike these items but most of them are formulated as positive statements, e.g. "My impression of the medical competence made me feel secure". Adding more negatively formulated items can possibly further enhance the discriminatory qualities in this scale.

As mentioned above, three items were found to have very high ceiling effects and were therefore not entered into the factor analysis. These items covered two separate domains, namely partner support and maternal worries about the baby. Although these domains have been identified as highly important aspects of the childbirth experience, we reasoned that they would be of little discriminative value [23]. The item Memory lapses from the childbirth did not fit into the factor model but may possibly contribute to the knowledge of this phenomenon. Memory lapses after labour and birth have been reported earlier [34] but this aspect of the birth experience is not well studied.

The strengths of this study are that domains and items comprising the CEQ were derived from patient interviews, discussions among experienced midwives and other staff, and pilot tested for face validity in a group of primiparous women. The psychometric performance of the questionnaire was tested on a large and representative sample of primiparous women with spontaneous onset of active labour (n = 920) living in various settings in Sweden. This heterogeneous sample represents a large group of childbearing women, as approximately 70% of all women expecting their first child are healthy, have a normal pregnancy and experience a spontaneous start of active labour.

The CEQ discriminated between groups of women known to differ in childbirth experiences [8–13]. Specifically, women with labour lasting more than 12 hours had, as expected, significantly lower scores on all scales than did those with shorter labour. Women who had their labour augmented scored lower than those without the treatment. Likewise, women who had delivered operatively had significantly lower scores than did those delivering normally (Table 4). On the other hand, effect sizes were primarily moderate to small, suggesting that further efforts may need to be made to improve the sensitivity of the questionnaire, especially regarding Professional support and Participation, where particularly weak effect sizes were observed (Figure 1).

The discriminative ability of the Participation scale may possibly have been debilitated by the fact that it comprised relatively few items (3 items) and that these items had comparatively poor internal consistency, as reflected by a low Cronbach's alpha (0.62). Adding relevant items and rewording existing items may help to augment the sensitivity of this scale. A reason for the weaker performance of the Professional support scale may be that the distribution of scores was skewed, with substantial ceiling effects. This problem is common in other scales and instruments measuring patient satisfaction with care and has serious implications for their ability to detect differences both between groups and over time [35]. A possible solution to this problem may be to extend the response scale from a four to a 5-step scale. For example, Moret et al. [36] have shown that extending the number of response alternatives substantially reduces or eliminates ceiling effects [37].

In summary, this study reports on the development and testing of the Childbirth Experience Questionnaire to assess different aspects of women's experiences of first childbirth. The four dimensions revealed in the exploratory factor analysis are concordant with earlier research and cover important aspects of the childbirth experience [1–9]. Furthermore, the instrument discriminates well between groups known to differ in childbirth experiences. A normal first childbirth with a positive experience is very important with regard to future pregnancies and childbirths. Women with a negative experience and a severe fear of childbirth often request an elective caesarean delivery [14] which is associated with increased risks for both mother and baby [38–42]. Postpartum questionnaires are needed 1.) to explore how different types of labour management influence the women's childbirth experiences, 2.) to improve care during childbirth and 3.) to identify those in need of support and counselling after a negative birth experience. Although most current instruments assess single domains of the childbirth experience, there is also a need for questionnaires that reflect and comprehensively assess the multidimensional nature of women's experiences of childbirth. We suggest that this four-dimensional model adequately reflects the multidimensionality of the childbirth experience and can be used as a tool to identify women with negative experiences and for evaluating efforts to improve the quality of childbirth care.