Contents lists available atScienceDirect
Sexual & Reproductive Healthcare
journal homepage:www.elsevier.com/locate/srhcDevelopment and validation of the Midwife Pro
filing Questionnaire
assessing women
’s preferred perinatal care professional and knowledge
of midwives
’ legal competences
Joeri Vermeulen
a,⁎, Wim Peersman
b, Linda Quadvlieg
c, Maaike Fobelets
a,d, Gerlinde De Clercq
a,
Eva Swinnen
e, Katrien Beeckman
a,f,gaDepartment Health Care, Midwifery Department, Knowledge Centre Brussels Integrated Care, Erasmus University College Brussels, Laarbeeklaan 121, 1090 Brussels, Belgium
bFaculty of Medicine and Health Sciences, Department of Family Medicine and Primary Health Care, Ghent University, De Pintelaan 185, 9000 Ghent, Belgium cVerloskundepraktijk De Tantes, Franeker, Hertog van Saxenlaan 36C, 8801 ES Franeker, The Netherlands
dFaculty of Medicine and Pharmacy, I-CHER (Interuniversity Centre for Health Economics Research), Vrije Universiteit Brussel (VUB), Laarbeeklaan 103, 1090 Brussels, Belgium
eFaculty of Physical Education and Physiotherapy, Rehabilitation Research, Vrije Universiteit Brussel (VUB), Pleinlaan 2, 1050 Brussels, Belgium fFaculty of Medicine and Pharmacy, Department Medical Sociology, Vrije Universiteit Brussel (VUB), Laarbeeklaan 103, 1090 Brussels, Belgium gNursing and Midwifery Research Unit, University Hospital Brussels, Laarbeeklaan 101, 1090 Brussels, Belgium
A R T I C L E I N F O
Keywords: Validation study Midwives’ competences Women’s preferences Women’s knowledge Uncomplicated pregnancyUncomplicated labour and uncomplicated childbirth
Midwifery
A B S T R A C T
Objectives: Currently maternity care organisation is developing worldwide. Therefore insight in the position of the midwife is important. The 'Midwife Profiling Questionnaire′ (MidProQ) measures women’s preferred peri-natal care professional and their knowledge of midwives’ legal competences. MidProQ is based on the European legal framework and was tested in a pilot study. This study aims to determine its content and face validity. Study design.: A two-phase validation study with a Delphi method questioning content experts (n = 10) on items relevance and clarity as well as its scale and face validity. Further semi-structured interviews were performed with lay experts (n = 10) to evaluate the questionnaire’s clarity, layout, phrasing and wording.
Results: After round one, most questions (42/47) were considered content valid for relevance and clarity (Item Content Validity Index 0.80–1.00). Scale (Scale Content Validity Index 0.92) and face validity (Face Validity Index 0.89) of the entire instrument was obtained. Five questions were revised until item content (0.83–1.00), scale content (0.92) and face validity (1.00) were appropriate. Lay experts’ suggestions for improving the readability and usability were taken into account.
Conclusions: We developed a valid instrument to elicit women’s preferred health professional for uncomplicated pregnancy, labour and childbirth and to determine their knowledge about midwives’ legal competences. Our instrument can be valuable in identifying knowledge gaps and improving the knowledge of the general popu-lation about the midwifery profession and maternity care. Finally, the MidProQ may improve research in the domain of maternity care culture, scale up midwifery and facilitate a more women-centred care.
Introduction
The contribution of midwifery in the public healthcare field has gained international attention[1]. A recent systematic review demon-strates the positive outcomes and cost containment of the so-called midwife-led care model, in which the midwife is the lead professional in
the planning, organisation and delivery of care given to a woman from initial booking to the postnatal period[2]. Midwife-led care is based on the belief of normality in childbirth, advocating continuity, autonomy and building relationships with mothers[3], and can therefore play a central role in women-centred care. Care by midwives has been found to be cost-effective, affordable and sustainable [4]. Hence, national
https://doi.org/10.1016/j.srhc.2018.01.003
Received 19 September 2017; Received in revised form 19 December 2017; Accepted 14 January 2018 ⁎Corresponding author.
E-mail address:[email protected](J. Vermeulen).
Abbreviations: MidProQ, Midwife Profiling Questionnaire; HP, health professional; I-CVI, Item Content Validity Index; S-CVI/Ave, Scale Content Validity Index
1877-5756/ © 2018 Elsevier B.V. All rights reserved.
governments should invest in deploying midwives and national health plans need to include a strategy to scale up midwifery.
Although young women and their partners are the prospective users of maternity services, there is still little research on their preferences in, knowledge of[5], and opinions on[6]midwifery and maternity services. In a recent study, for example, women reported considerable uncertainty and a lack of information about provider options, thus revealing sub-stantial knowledge deficits in this area[7]. It has been internationally acknowledged that women’s choices in maternity services are affected by the organisation of care[8], care provider influence[9], culturally em-bedded habits[10]and psychosocial and demographic factors[11]. This means that the decisions of prospective users of maternity services may be influenced in several ways. In accordance with the current shift towards women-centred care, women's preferences with regard to maternity ser-vices have become increasingly interesting for policy-makers[10]. Un-derstanding women’s preferences and their knowledge about midwives’ competences can help midwifery organisations, women’s groups and policy-makers understand maternity care culture and encourage a tran-sition to more women-centred care.
In the Belgian context, such information is crucial, especially in the light of the structural changes in maternity services that have recently been initiated by health authorities [12,13]. These changes focus on shortening hospital length of stay in postnatal care[13], the develop-ment of primary maternity care services [13,14] and a shift to the midwife-led care model[14]. Internationally, maternity care typically consists of a medical model of care with varying levels of midwifery input [15], and is usually not able to provide the same midwife throughout[2]. As changes in maternity care culture are initiated by concrete structural changes[16]those changes will impact the role of health professionals (HP) in maternity services as well as women’s views on their roles, and it has the potential to facilitate woman-centred care. Today, the obstetrician is the main HP in maternity care in Bel-gium[17]. The medical model of care is responsible for the fact that obstetricians are regarded as the central perinatal care professional [18], whereas midwives remain unknown until the day of birth[19]. As the organisation of care has an impact on women’s preferences[8,10], it is safe to assume that it might also affect their knowledge of mid-wives’ and other HP’s in maternity services’ legal competences. Insight into women’s preferences and knowledge about the legal competences of midwives can help to understand maternity care culture. This has become important since the shift toward home care with a more central role for the midwife in Belgium. Although one qualitative study in Flanders, Belgium[20]and one observational study from the Brussels metropolitan region, Belgium[21]have been conducted on this topic, further research is needed to uncover trends in women’s opinions, preferences and knowledge across time, regions and healthcare sys-tems. To ensure comparability in such endeavours, however, one and the same questionnaire should be used. Because no such instruments were available, we recently developed and employed the 'Midwife Profiling Questionnaire' (MidProQ version I) [21]. The MidProQ de-termines women’s preferred HP for uncomplicated pregnancy, labour and childbirth, and assesses their knowledge of midwives’ legal com-petences. Note that labour is defined as ‘the process of giving birth’ and childbirth as‘the act of giving birth’[22]. From our pilot study it was concluded that for Brussels women, obstetricians (88%) were preferred over midwives (68%) for care during labour and childbirth, only one in five of the respondents considered midwives to play a central role in the care for an uncomplicated pregnancy. Knowledge on the legal compe-tences of the midwives varied widely, least known were compecompe-tences related to the medical autonomy of the midwife.
The aim of present study was to optimize the Midwife Profiling Questionnaire[21]through a validation study.
Methods
Description of MidProQ questionnaire Development of the MidProQ version I
The instrument was developed in three steps, as suggested by Zamanzadeh et al.[23]: (1) we identified the content domain through a comprehensive literature review, (2) we generated the instrument items, and (3) we constructed the entire instrument. The MidProQ version I was based on Belgian legislation on midwives’ responsibilities [24]as well as European legislation[25], ensuring a frame of reference within a European context. Furthermore, six midwives with a clinical or educational work experience of minimumfive years were asked to re-view the questionnaire, as is also recommended in research literature [26]. No ambiguity was expressed, and only minor changes were made to improve the readability of the instrument. The MidProQ version I was piloted in 2014 and 2015 with women in their reproductive age, living in the Brussels metropolitan region (n = 830)[21].
Optimisation of the MidProQ version I
To expand our scope from Brussels to Flanders, the MidProQ version I was optimised after the pilot study in April 2016. The changes in-cluded adding a short glossary of medical terms, turning one specific question on preferences into a single answer question, adding a ques-tion on preferences in pregnancy, and deleting a quesques-tion that mea-sured participants’ expectations rather than their knowledge or pre-ferences. This revision resulted in the MidProQ version II, consisting of three components with a total of 47 closed-ended questions:
Component 1:‘Preferences’; three questions that measure women’s preferences for the HP to follow up on an uncomplicated labour and childbirth.
Component 2: ‘Knowledge’; 41 questions determining women’s’ knowledge of midwives’ legal competences during pregnancy, la-bour and childbirth.
Component 3:’Opinion’; three questions to explore women’s opi-nions on the central HP for an uncomplicated pregnancy.
Design
A two-phase validation study with - a Delphi method with content experts, and
- semi-structured face-to-face interviews with lay experts.
Content experts are professionals with research or work experience in thefield of interest, while lay experts are potential research subjects [23], ensuring that the population from whom the instrument is being developed is represented. To incorporate the content experts in this study, a Delphi method was chosen. As stressed by Keeney and col-leagues this is an important method for achieving consensus on issues where no previously existed. The Delphi method is a structured process that gathers information in a series of rounds which are continued until consensus is reached[27]. For the involvement of lay experts, semi-structured interviews were conducted. This gives the lay experts the opportunity to evaluate the questions in terms of clarity, phrasing and wording and to make suggestions[28]. Through the interviews, the lay experts may express the need for construct refinement and increase the likelihood that items are valid for their intended purpose[29](Fig. 1).
Participants
Phase one: content experts
A panel of experts was asked to evaluate the content of the new instrument[26]. In validation studies, at least three and maximum ten content experts are usually recommended[28]. In our study we strived for a participation of ten content experts[23], who were all recruited because of their work experience and/or expertise as a midwife[28]in one of the specific domains described in the professional and compe-tency profile of Belgian midwives[30]. Maximum variation sampling was used, meaning that experts from all midwifery domains were re-cruited. Additionally, we aimed to include one obstetrician.
Participation of the experts was voluntary. Potential content experts were invited by e-mail in October 2016. The invitation included in-formation about the study, an informed consent form and a link to the survey site.
Phase two: Lay experts
Inclusion criteria for lay experts were; women in their reproductive age (15–44 years of age), native Dutch speakers, without a medical background, nulli-, primi- or multipara, from all levels of education (i.e. primary, secondary and tertiary education level). As such, they share the characteristics of the potential subjects under study. Other potential users of maternity services such as partners and youngsters were ex-cluded from this study. We strived for a participation of lay experts until data saturation would be achieved, meaning that no new themes emerged from the semi-structured interviews.
Potential candidates were randomly approached by a researcher [LQ] in public places (i.e. a high-school cafeteria, a bus station, a shopping mall) in the city of Ghent, Belgium's third largest munici-pality, in December 2016. They were informed about this study and were invited to participate. If they were likely to do so, they were asked to read the information letter and give their consent. As recommended, additional socio-demographic information was gathered[28]. Evaluating content and face validity
To draw any conclusions about the quality of the scale, information
Since content validity, defined as ‘the degree to which an instrument has an appropriate sample of items for the construct being measured’ [26]is a prerequisite for other (construct and criterion-related) validity, it should be the highest priority[23]. Face validity indicates that the instrument appears to be valid‘on its face’[28]. It refers to whether the designed instrument is apparently related to the construct underlying the study, whether participants agree with the items and wording used to realise the research objectives. Face validity does not consider what is measured but instead focuses on the appearance of the instrument [23]. Nevertheless, it is a desirable quality for any instrument since it increases acceptance by potential users[32].
Both content and face validity are obtained when the content ex-perts and subjects under study recognise that the instrument is suitable for measuring pertinent attributes.
Data collection
Phase one: evaluation of content and face validity by content experts The Delphi study involved a survey with a brief introduction, ele-ment definitions, contact information, instructions and scoring me-chanisms. Thefirst section elicited demographic and professional in-formation, while the second section gathered feedback on the relevance and clarity of the items in terms of content and face validity. During this evaluation, the experts used a Likert-type rating scale. Using at least five experts and a 4-level, Likert-type scale has been argued to over-come the limitations of the proportion agreement procedure, because a neutral and ambivalent midpoint is avoided[33].
We were able to quantitatively measure content validity by estab-lishing what proportion of experts agreed on the relevance and clarity of the items[34]. An item’s clarity was evaluated on how clearly it was formulated, while its relevance refers to how relevant it seemed for the research objectives. As suggested by Zamanzadeh et al., responses in-cluded a rating of 1 = not relevant/not clear, 2 = somewhat relevant/ somewhat clear, 3 = quite relevant/quite clear, 4 = very relevant/very clear [35]. As recommended, experts assessed clarity and relevance consecutively on the same scale, encouraging them to evaluate each item completely[28].
For face validity, the content experts indicated whether items were appropriately formulated to realise the research objectives [23]. The statement‘The instrument realises the research objectives’, could be rated 1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree. Moreover, the experts could comment and suggest additional items.
Because of the iteration process inherent in the Delphi technique, the responses of individual panel members were only known to one researcher [LQ] and remained unknown to the other participants. The researcher in question was thus able to follow up and encourage non-responders, as has been suggested in previous research[27]. To the other researchers, all information was de-identified during the analysis and in thefinal reports and was stored on a secured server at Erasmus University College Brussels and in a lockedfiling cabinet.
As the validation process is iterative and uses carefully controlled feedback to progressively seek consensus, ideas are usually generated in thefirst round, in which experts may draw on existing information and put forward additional ideas. In our study, these ideas were fed back to the experts in the subsequent rounds, in which the emphasis was more on reducing the number of items to an agreed cut-off point[36]. Phase two: evaluation of face validity by lay experts
In the second phase of this validation study, after an appropriate content validity and face validity of the questionnaire was obtained by content experts, lay experts were asked to evaluate the questionnaire in terms of clarity, layout, phrasing and wording. The interviews were semi-structured, an interview guide with open-ended questions was used. The researcher [LQ] elicited further explanations if necessary and Recruitment
of content experts (n=16)
Online evaluation of content and face validity
by content experts (n=10)
Appropriate content and face validity reached
Recruitment of lay experts (n=18) Feedback to seek consensus (Delphi survey) Phase one Phase two Semi-structured interviews with lay experts (n=10) to evaluate clarity, layout, wording and give suggestions
Data analysis
Phase one: evaluation of the content and face validity by the content expert panel
Two content validity indices (CVIs) were computed: the content validity of individual items and that of the overall scale[35].
There is a general consensus on how the item-level CVI, also re-ferred to as the I-CVI, can be determined[31]. To obtain a CVI for the relevance and clarity of each item (I-CVI), the number of experts jud-ging it as relevant or clear (rating 3 or 4) is divided by the total numbers of experts. I-CVI expresses the degree of consensus with a value be-tween 0 and 1.00. In our study, these values were interpreted as fol-lows: if the I-CVI was higher than 0.79, the item was deemed appro-priate, and if it was between 0.70 and 0.79 it would need revision, in which case it was adjusted based on the content experts’ advice and then included in a following Delphi round. If the I-CVI was below 0.70, the item would have to be eliminated[23].
The second CVI, scale-level CVI (S-CVI), can be defined as the proportion of total items deemed to have content validity[33]. In line with Polit et al., the S-CVI was calculated using the‘average proportion of items rated as 3 or 4 across the various judges’ (hereafter referred to as S-CVI/Ave). It has been suggested that the S-CVI/Ave should be conceptualised as the average I-CVI value, because this value focuses on average item quality rather than average performance by the experts [31]. As research literature indicates[31], the S-CVI/Ave should be ≥0.90 to reflect the content validity of the entire scale.
Besides the CVI, the face validity index was also computed. The number of content experts who‘agreed’ or ‘strongly agreed’ (i.e. rating 3 or 4) to the statement‘The instrument realises the research objectives’ was divided by the total number of experts. As suggested by Amondela et al.[37]a face validity index of≥0.80 shows that the instrument has face validity.
To determine face validity, researchers need to use expert view-points[35]. Typically, content analysis techniques are used to identify the major themes generated by the initial unstructured questionnaire [38]. In our case, the content experts’ written suggestions were ana-lysed by two researchers [JV and LQ], using a thematic content analysis approach. The suggestions were coded into recurrent or common themes as suggested in research literature[39].
Phase two: Evaluation of face validity by lay experts
The transcripts of the interviews with lay experts were analysed and recurrent themes were identified until no new themes emerged. Data saturation and themes were confirmed among the interviewer [LQ] and a second investigator [JV].
Results
Phase one: Evaluation of content and face validity by content experts Characteristics of the content experts
From the 16 invited content experts, 10 experts agreed to partici-pate in the validation process. Maximum variation sampling was ex-erted as defined in the professional and competency profile of Belgian midwives, meaning the inclusion of experts from professional organi-sations/regulatory body (n = 3), research (n = 3), management (n = 3), primary care (n = 3), clinical midwifery (n = 4) and education (n = 4). Seven experts combined two domains of expertise, with the combination of research and education as the most common one (n = 3). One obstetrician holding a PhD also agreed to participate (Table 1).
Item Content Validity Index I-CVI (Relevance and clarity)
Thefirst round revealed that three questions related to the com-ponent ‘Knowledge’ needed to be revised (I-CVI between 0.70 and 0.79). These were then submitted to an additional content expert round. Of the three questions, two did not get an appropriate score for I-CVI Clarity (both I-I-CVI clarity 0.70; questions ‘Know2.1.’ and ‘Know3.1.’) and one did not reach the cut-off point for relevance (I-CVI 0.70; question‘Know2.19.’). As suggested by Zamanzadeh et al.[23], two other questions, also related to the component‘Knowledge’, had to be eliminated (I-CVI < 0.70; questions ‘Know1.’ and ‘Know2.7.’). However, as these questions obtained an appropriate score for re-levance (I-CVI 0.90), they were modified according to the experts’ re-commendations and included in the second Delphi round as well. All questions from the components ‘Preferences’ (n = 3) and ‘Opinion’ (n = 3), had I-CVIs above the cut-off point for relevance (0.80–0.89) and clarity (0.80–1.00), thereby confirming the content validity of the questions related to these two components (Table 2).
In the second round, six of the ten content experts agreed to parti-cipate. An acceptable variation sampling was still achieved, and experts from professional organisations (n = 2), research (n = 2), management (n = 2), primary care (n = 3), clinical midwifery (n = 2) and education (n = 3) were still represented. At this point, the obstetrician also withdrew (SeeTable 1).
Thefive optimised questions related to the component ‘Knowledge’, were modified according to the content experts’ suggestions. Questions with an initial I-CVI for clarity between 0.70–0.79 and < 0.70 obtained 1.00 in the second round, and the question with an I-CVI of 0.70 for relevance now obtained 0.83. It is safe to conclude that the MidProQ version II achieved an acceptable content validity after two rounds (See Table 3).
Table 1
Area of expertise of the content experts. Content expert ID Area of expertise
Professional organisations-Regulatory body Researcher Management Primary care Clinical midwifery Education Obstetrician
1 X X 2* X* 3* X* 4 X X X X 5 X X 6 X X 7* X* X* 8* X* X* X* 9 X X X 10 X
2 Validity Index I-CVI after round one. Received answers (total n = 10) n Clear (rating 3o r 4 ) n Not clear (rating 1 or 2) n I-CVI Clarity Interpretation Received answers (total n = 10) n Relevant (rating 3o r 4 ) n Not relevant (rating 1 or 2) n I-CVI Relevance Interpretation related to Component ‘Preferences ’ (n = 3) Which HP do you prefer to follow up an uncomplicated pregnancy? 9 8 1 0.89 Appropriate question 9 8 1 0.89 Appropriate question Which HP do you prefer to follow up an uncomplicated labour? 9 8 1 0.89 Appropriate question 9 8 1 0.89 Appropriate question Which HP do you prefer to follow up an uncomplicated childbirth? 9 9 0 0.89 Appropriate question 9 8 1 0.89 Appropriate question related to Component ‘Knowledge ’ (n = 41) Which HP ’s can independently follow up an uncomplicated pregnancy? 10 6 4 0.60 * Inappropriate question 10 9 1 0.90 Appropriate question Which HP ’s are allowed in an uncomplicated pregnancy to autonomously: Diagnose the pregnancy 10 7 3 0.70 * Inappropriate question 10 9 1 0.90 Appropriate question Inform about the progress of the pregnancy 10 10 0 1.00 Appropriate question 10 8 2 0.80 Appropriate question Listen to the foetal heart beat 10 10 0 1.00 Appropriate question 10 9 1 0.90 Appropriate question Apply an electrocardiography (explained in the accompanying glossary) 10 9 1 0.90 Appropriate question 10 10 0 1.00 Appropriate question Active preparation towards childbirth 10 9 1 0.90 Appropriate question 10 10 0 1.00 Appropriate question Perform a vaginal examination (explained in the accompanying glossary) 10 9 1 0.90 Appropriate question 10 10 0 1.00 Appropriate question Prescribe examinations necessary to evaluate the progress of pregnancy 10 5 5 0.50 ** Inappropriate question 10 9 1 0.90 Appropriate question Perform a vaginal smear 10 10 0 1.00 Appropriate question 10 10 0 1.00 Appropriate question Perform an ultrasound 10 10 0 1.00 Appropriate question 10 9 1 0.90 Appropriate question Perform an abortion 10 10 0 1.00 Appropriate question 10 8 2 0.80 Appropriate question Perform an amniocentesis (explained in the accompanying glossary) 10 10 0 1.00 Appropriate question 10 8 2 0.80 Appropriate question Check if the membranes are ruptured 10 8 2 0.80 Appropriate question 10 9 1 0.90 Appropriate question Perform a blood sample 10 10 0 1.00 Appropriate question 10 10 0 1.00 Appropriate question Interpret blood results related to pregnancy 10 9 1 0.90 Appropriate question 10 10 0 1.00 Appropriate question Prescribe medicaments related to pregnancy 10 9 1 0.90 Appropriate question 10 10 0 1.00 Appropriate question Vaccinate the pregnant woman 10 10 0 1.00 Appropriate question 10 10 0 1.00 Appropriate question Weigh the woman 10 10 0 1.00 Appropriate question 10 9 1 0.90 Appropriate question Measure the blood pressure 10 10 0 1.00 Appropriate question 10 10 0 1.00 Appropriate question Perform a breast examination 10 9 1 0.90 Appropriate question 10 7 3 0.70 * Inappropriate question Prepare the couple for parenthood 10 10 0 1.00 Appropriate question 10 9 1 0.90 Appropriate question (continued on next page )
Table 2 (continued ) Questions Received answers (total n = 10) n Clear (rating 3o r 4 ) n Not clear (rating 1 or 2) n I-CVI Clarity Interpretation Received answers (total n = 10) n Relevant (rating 3o r 4 ) n Not relevant (rating 1 or 2) n I-CVI Relevance Interpretation Know3. Which HP ’s are legally allowed in an uncomplicated pregnancy and labour to: Know3.1. Follow up the labour 10 7 3 0.70 * Inappropriate question 10 9 1 0.90 Appropriate question Know3.2. Perform a delivery 10 8 2 0.80 Appropriate question 10 8 2 0.80 Appropriate question Know4. Which HP ’s are allowed in an uncomplicated labour and childbirth to autonomously: Know4.1. Listen to the foetal heart beat 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Know4.2. Early detect abnormalities in the progress of labour or childbirth 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.3. Apply an electrocardiography (explained in the accompanying glossary) 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Know4.4. Perform a vaginal examination (explained in the accompanying glossary) 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.5. Arti fi cially rupture the membranes 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.6. Provide medication for the stimulation of contractions 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.7. Place an epidural anaesthesia 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Know4.8. Provide non-medical pain relief methods 9 8 1 0.89 Appropriate question 9 8 1 0.89 Appropriate question Know4.9. Administer an infusion 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.10. Cut the umbilical cord 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Know4.11. Perform an episiotomy (explained in the accompanying glossary) 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Know4.12.Suture an episiotomy (explained in the accompanying glossary) 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.13. Suture a tear (explained in the accompanying glossary) 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.14. Perform a vacuum extraction (explained in the accompanying glossary) 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Know4.15. Perform a forceps extraction (explained in the accompanying glossary) 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Know4.16. Perform a Caesarean Section 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.17. Examining the new-born 9 9 0 1.00 Appropriate question 9 9 0 1.00 Appropriate question Know4.18. Resuscitation of the new-born 9 9 0 1.00 Appropriate question 9 8 1 0.89 Appropriate question Questions related to Component 'Opinion' (n = 3) Op1. Which HP is in your opinion the central HP to follow up an uncomplicated pregnancy? 10 8 2 0.80 Appropriate question 10 8 2 0.80 Appropriate question Op2. Which HP is in your opinion the central HP to follow up an uncomplicated labour? 10 8 2 0.80 Appropriate question 10 8 2 0.80 Appropriate question Op3. Which HP is in your opinion the central HP to follow up an uncomplicated childbirth? 10 8 2 0.80 Appropriate question 10 8 2 0.80 Appropriate question * I-CVI between 0.70 and 0.79. ** I-CVI < 0.70.
Scale Content Validity Index S-CVI/Ave (Relevance and clarity)
The S-CVI/Ave, determining the‘average proportion of items rated as 3 (quite relevant/quite clear) or 4 (very relevant/very clear) across the various judges’, was 0.92. As a S-CVI/Ave of at least 0.90 is needed for content validity, we can conclude that the entire MidProQ version II has the required content validity[31]. Although the amendments made after the results of the Delphi of round one, the S-CVI/Ave remained appropriate; 0.92 thus confirming content validity of the entire in-strument.
Face validity index (‘Does the instrument realise the research objectives?’) To confirm face validity, the content experts were asked to indicate whether they felt that the items and wording of the questionnaire would allow the research objectives to be realised. After thefirst round, face validity was obtained, with a rating of 0.89. However, the amendments that were made in response to round one could have in-fluenced the initial face validity calculation. As a result, content experts had to evaluate the MidProQ version II on its face validity again in the second round, at which point all content experts (n = 6) confirmed that the MidProQ version II contributes to the research objectives, resulting in a face validity score of 1.00 (Table 4).
Suggestions from the content experts
Content experts could comment and suggest new items to be added. From a thematic content analysis of the suggestions, two themes emerged: wording and content.
Most suggestions were related to the wording of the questions. Although the remarks considered questions with an acceptable I-CVI for clarity > 0.70 and I-CVI for relevance > 0.70, the suggestions were discussed among the researchers [JV, LQ and KB]. As soon as a con-sensus had been reached, the amendments (i.e. reformulations and additional explanations of medical terms) were included in the second Delphi round. In this round, all amended questions scored between 0.83–1.00 I-CVI for clarity and 0.83–1.00 I-CVI for relevance, meaning that the instrument had content validity at this point.
Another recurring theme in the suggestions was content. The con-tent of the MidProQ was perceived as medical and task-focused and some content experts stressed the need to include additional midwifery competences. As a result, two additional questions were included after round one, both of which involved the component‘Knowledge’: the first dealt with determining the position of the baby (Question ‘Knowl2.21.added’) and the second with resuscitating the mother (Question‘Knowl4.19.added’). Both added questions were considered as content valid in the second round, as the I-CVI for clarity and re-levance was 1.00 for thefirst question and 0.83 for the second (See Table 3).
Phase two: evaluation of face validity by lay experts Characteristics of the lay experts
Eighteen women were invited to participate, six of which refused due to a lack of time and two of which did not meet the inclusion criteria (i.e. one was not sufficiently proficient in Dutch and another was not between 15 and 44 years of age). In total, ten women agreed to participate, representing an adequate sample of the subjects under study. After ten interviews, data saturation was achieved as no new themes emerged from the semi-structured interviews (Table 5). Evaluation from the lay experts
An analysis of the semi-structured interviews with the lay experts demonstrated that some questions in the MidProQ version II were seen as complex and long (this was more prominent in nullipara). Some lay experts also found it confusing that there was no‘I don’t know’ option. Furthermore, not all mentioned HP’s were considered relevant for the aims of the study, especially the physiotherapist and the general
prac-3 Validity Index I-CVI after round two. Received answers (total n = 6) n Clear (rating 3 or 4) n Not clear (rating 1o r 2 ) n I-CVI Clarity Interpretation Relevant (rating 3o r 4 ) n Not relevant (rating 1 or 2) n I-CVI Relevance Interpretation related to Component ‘Knowledge ’ which had an I-CVI clarity between 0.70 and 0.79 (n = 2) Which HP ’s are allowed in an uncomplicated pregnancy to autonomously: Know2.1. Diagnose the pregnancy 6 6 0 1.00 Appropriate question Not applicable Not applicable Not applicable Not applicable Which HP ’s are legally allowed in an uncomplicated pregnancy and labour to: Know3.1. Follow up the labour 6 6 0 1.00 Appropriate question Not applicable Not applicable Not applicable Not applicable related to Component ‘Knowledge ’ which had an I-CVI clarity < 0.70 in round one (n = 2) Which HP ’s can follow up an uncomplicated pregnancy? 6 6 0 1.00 Appropriate question Not applicable Not applicable Not applicable Not applicable Which HP ’s are allowed in an uncomplicated pregnancy to autonomously: Know2.7. Prescribe examinations necessary to evaluate the progress of pregnancy 6 6 0 1.00 Appropriate question Not applicable Not applicable Not applicable Not applicable related to Component ‘Knowledge ’ which had an I-CVI relevance between 0.70 and 0.79 (n = 1) HP ’s are allowed in an uncomplicated pregnancy to autonomously: Know2.19. Perform a breast examination (explained in the accompanying glossary) 6 Not applicable Not applicable Not applicable Not applicable 5 1 0.83 Appropriate question questions after round one (n = 2) Determining the position of the baby 6 6 0 1.00 Appropriate question 6 0 1.00 Appropriate question Resuscitation of the mother 6 5 1 0.83 Appropriate question 5 1 0.83 Appropriate question
amendments were made due to some persistent remarks from our target group.
To improve the readability and usability of the instrument, we shortened the text on the introduction page as well as any questions that were evaluated as too long or complex. Keywords were highlighted in the questions and an‘I don’t know’ answer option was added. Discussion
It is increasingly recognised that users of maternity services views and preferences should be taken into account in the provision of healthcare[40]. To achieve optimal women-centred care, insight into women’s preferences is essential [10]. The aim of this study was to determine the content and face validity of the MidProQ version II, a questionnaire to determine women’s preferred HP for an uncomplicated pregnancy, labour and childbirth and their knowledge about midwives’ legal competences. Our instrument achieved content and face validity after two Delphi rounds, which is in line with the two to four rounds usually required as described by Keeney et al.[27].
One of the content experts’ critical remarks was that the instrument is rather task-focused. We acknowledge that the practice of midwifery includes a wide range of competences, as defined by the International Confederation of Midwives [41]. However, we chose to restrict the questionnaire to the legal framework of midwives’ responsibilities in Belgium and Europe, as this would contribute to its validity. Never-theless, we agree that investigating women’s knowledge about a broader range of midwifery competences is warranted. All suggestions given by the experts were thoroughly discussed among the researchers, with a clear focus on the research objectives. Two additional questions were added and both scored appropriately for clarity and relevance. Some suggestions related to psychosocial support and information were not included, as these themes were already sufficiently covered. In accordance with the comments from the lay experts, we decided that expanding the questionnaire would be undesirable.
We observed adverse opinions from both the content and lay experts with regard to the desirability to include the physiotherapist and gen-eral practitioner into the instrument. Although as both the phy-siotherapist[42]and the general practitioner[17]are legally compe-tent in thefield of pregnancy, labour and childbirth in Belgium[43]as
well as internationally[44]it was decided not to exclude them from the MidProQ version II.
Some limitations of validity studies should be noted. As the feed-back from the experts is subjective, our study is vulnerable to potential bias among the experts. We also recognise the limitations of the Delphi methodology, as a group consensus does not necessarily produce the best or correct results and represents expert opinion, rather than in-disputable fact[27,36]. Nevertheless, the Delphi method is widely ac-cepted in validation studies in health research[27]. Another potential limitation is that this type of study does not necessarily identify content that might have been omitted. To minimise this, limitation experts were able to suggest other items[28].
As our instrument is based on the European legal framework of midwives’ responsibilities, it has the potential to advance research in Europe in the domain of maternity care culture. Although the legal framework is the same across Europe, some countries might have dif-ferences in legislation and organisation of maternity care, meaning that the MidProQ component constructs may vary from one culture to an-other [45]. However, the increasing demand for cross-cultural com-parisons in the international setting and the use of valid and cultural adapted instruments is stronger than ever[46]. To obtain reliable re-sults, stringent back-translation and validation for the local context is necessary. Therefore the evaluation of the use of MidProQ in other countries is highly recommended for future cross-cultural research [47].
Conclusion
We have developed a valid instrument that can help determine women’s preferred HP for an uncomplicated pregnancy, labour and childbirth and their knowledge about midwives’ legal competences. The MidProQ can offer valuable insights into the maternity care culture of (prospective) maternity users in Belgium.
Designing and delivering maternity services in which women’s voices are heard is essential and should be recognised as a key indicator of safe and quality care. As we contemplate the future of maternity care in Belgium, it is crucial that we understand the preferences and knowledge of the next generation of maternity care users. Maternity services must be women-centred and responsive to women’s’ needs and Table 4
Face validity index.
Received answers Relevant (rating 3 or 4) n
Not relevant (rating 1 or 2) n
Face Validity Index Interpretation
Does the instrument realises the research objectives (round 1) 9 8 1 0.89 Appropriate
Does the instrument realises the research objectives (round 2) 6 6 0 1.00 Appropriate
Table 5
Characteristics of the lay experts. Lay expert ID Characteristics
Age category (years) Education level (highest completed education) Parity
15–19 20–29 30–39 40–44 No education/Primary education only Secondary education Tertiary education 0 1 2 ≥3
1 X X X 2 X X X 3 X X X 4 X X X 5 X X X 6 X X X 7 X X X 8 X X X 9 X X X 10 X X X
preferences[48]. By identifying knowledge deficits on the competences of HP’s in maternity care, we may have the opportunity to influence future decisions [7]. Efforts to change maternity care systems must consider how the cultural embeddedness of women's preferences can be countered[10]. As midwife-led care has been proven to be cost-effec-tive, affordable and sustainable, it is pivotal that we identify women’s preferred HP for an uncomplicated pregnancy, labour and childbirth. These insights can offer valuable information to women’s groups and health authorities that wish to familiarise the public with the im-portance and added value of midwifery care.
The MidProQ will be of special use in afield that has a scarcity of valid instruments to explore women’s preferences for and knowledge of maternity care culture. Moreover, this instrument can be of great value in identifying knowledge gaps and improving the knowledge of the general population about the midwifery profession and maternity care. Finally, the MidProQ may improve research in the domain of maternity care culture, scale up midwifery and facilitate a more women-centred care in Belgium and internationally.
Acknowledgements
The authors wish to thank all the content and lay experts for their indispensable contribution to this study and for their supportive role in guiding the researchers. Additionally we express our gratitude to all women and student midwives that participated in our pilot study in 2014 and 2015. We thank Eva Tyteca for her help during the revision process of this paper.
Conflict of interest
The authors declare no conflict of interest. References
[1] Sandall J. Every woman needs a midwife, and some women need a doctor too. Birth: Issues Perinatal Care 2012;39(4):323–6.http://dx.doi.org/10.1111/birt.12010. [2] Sandall J, Soltani H, Gates S, Shennan A, Devane D. Midwife-led continuity models
versus other models of care for childbearing women. The Cochrane Database Of Systematic Reviews 2016: CD004667. doi: 10.1002/14651858.CD004667.pub5. [3] Soltani H, Fair F, Duxbury A. Exploring health professionals’ and women’s
aware-ness of models of maternity care evidence. British J Midwif 2015;23(1):22–31.
http://dx.doi.org/10.12968/bjom.2015.23.1.22.
[4] ten Hoope-Bender P, de Bernis L, Campbell J, Downe S, Fauveau V, Fogstad H, et al. Series: improvement of maternal and newborn health through midwifery. Lancet 2014;384:1226–35.http://dx.doi.org/10.1016/S0140-6736(14)60930-2. [5] Newick L, Vares T, Dixon L, Johnston J, Guilliland K. A midwife who knows me:
women tertiary students' perceptions of midwifery. New Zealand College Midwives J 2013;47:5–9.http://dx.doi.org/10.12784/nzcomjnl47.2013.1.5-9.
[6] Lavender T, Chapple J. An exploration of midwives’ views of the current system of maternity care in England. Midwifery 2004;20:324–34.http://dx.doi.org/10.1016/ j.midw.2004.01.005.
[7] Edmonds J, Cwiertniewicz T, Stoll K. Childbirth education prior to pregnancy? Surveyfindings of childbirth preferences and attitudes among young women. J Perinat Educat 2015;24(2):93–101.http://dx.doi.org/10.1891/1058-1243.24.2.93. [8] Christiaens W, Verhaeghe M, Bracke P. Childbirth expectations and experiences in
Belgian and Dutch models of maternity care. J Reproduct Infant Psychol 2008;26(4):309–22.http://dx.doi.org/10.1080/02646830802350872. [9] Cooper T, Lavender T. Women's perceptions of a midwife's role: An initial
in-vestigation. British J Midwife 2013;21(4):264–72.
[10] Haaren-ten Haken T, Pavlova M, Hendrix M, Nieuwenhuijze M, Vries R, Nijhuis J. Eliciting preferences for key attributes of intrapartum care in the Netherlands. Birth 2014;41(2):185–94.http://dx.doi.org/10.1111/birt.12081.
[11] Wilson K, Sirois F, Fuschia M. Birth attendant choice and satisfaction with antenatal care: the role of birth philosophy, relational style, and health self-efficacy. J Reproduct Infant Psychol 2010;28(1):69–83.http://dx.doi.org/10.1080/ 02646830903190946.
[12] Van de Voorde C, Van den Heede K, Beguin C, Bouckaert N, Camberlin C, de Bekker P, et al. Required hospital capacity in 2025 and criteria for rationalisation of complex cancer surgery, radiotherapy and maternity services. Belgian Health Care Knowledge Centre (KCE). Retrieved fromhttps://kce.fgov.be/sites/default/files/ page_documents/KCE_289_Required_hospital_capacity_Report.pdf. 2017. [13] Benahmed N, Devos C, San Miguel L, Vinck I, Van kelst L, Lauwerier E, et al. Caring
for mothers and newborns after uncomplicated delivery: towards integrated post-natal care. Belgian Health Care Knowledge Centre (KCE). Retrieved fromhttp://
kce.fgov.be/sites/default/files/page_documents/KCE_232_Postnatal_care_Report_0. pdf. 2014.
[14] Vandenbroeck P, Ooms K, Benahmed N, Wickert R, Christiaens W, Stordeur S. Strategies for improving the medical workforce projection model: a stakeholder consultation. Belgian Health Care Knowledge Centre (KCE). Retrieved fromhttps:// kce.fgov.be/sites/default/files/page_documents/KCE_290A_projectiemodel_ medisch_aanbod_Rapport.pdf. 2017.
[15] Wernham E, Gurney J, Stanley J, Ellison-Loschmann L, Sarfati D. A comparison of midwife-led and medical-led models of care and their relationship to adverse fetal and neonatal outcomes: a retrospective cohort study in New Zealand. PLoS Med 2016;9:e1002134.http://dx.doi.org/10.1371/journal.pmed.1002134. [16] Christiaens W, Nieuwenhuijze M, de Vries R. Trends in the medicalisation of
childbirth in Flanders and the Netherlands. Midwifery 2013;29:e1–8.http://dx.doi. org/10.1016/j.midw.2012.08.010.
[17] Eggermont M. The Belgian, French and Dutch midwife on trial: a critical case study. Midwifery 2015;31:547–53.http://dx.doi.org/10.1016/j.midw.2015.02.008. [18] Mead M, Bogaerts A, Reyns M. Midwives’ perception of the intrapartum risk of
healthy nulliparae in spontaneous labour, in The Flanders. Belgium Midwife 2007;23:361–71.http://dx.doi.org/10.1016/j.midw.2006.05.003.
[19] Christiaens W, Bracke P. Place of birth and satisfaction with childbirth in Belgium and the Netherlands. Midwifery 2009;25:e11–9.http://dx.doi.org/10.1016/j. midw.2007.02.001.
[20] Van kelst L, Spitz B, Sermeus W, Thomson A. Perspectives on maternity care in Belgium. A hermeneutic study. Leuven: KU Leuven. 2012.
[21] Vermeulen J, Swinnen E, D’haenens F, Buyl R, Beeckman K. Women's preferences and knowledge about the legal competences of midwives in Brussels, Belgium. A descriptive observational study. Midwifery 2016;40:177–86.http://dx.doi.org/10. 1016/j.midw.2016.07.002.
[22] Turnbull J. Oxford Wordpower Dictionary for learners of English. Oxford: Oxford University Press; 2006.
[23] Zamanzadeh V, Rassouli M, Abbaszadeh A, Alavi-Majd H, Nikanfar A,
Ghahramanian A. Details of content validity index and objectifying it in instrument development. Nursing Practice Today 2014;1(3):163–71.
[24] Federal Government of Belgium, 1991. Royal Decree concerning the practice of the profession of midwife.
[25] European Parliament Council of the European Union, 2013. Amending Directive 2005/36/EC on the recognition of professional qualifications and Regulation (EU) No 1024/2012 on administrative cooperation through the Internal Market Information System.
[26] Beck C, Polit D. Essentials of nursing research. Appraising evidence for nursing practice. 8th edition. Philadelphia: Wolters Kluwer; 2014.
[27] Keeney S, Hasson F, McKenna H. Consulting the oracle: ten lessons from using the Delphi technique in nursing research. J Adv Nurs. 2006;53(2).http://dx.doi.org/ 10.1111/j.1365-2648.2006.03716.x. 205-12 8p.
[28] Rubio D, Berg-Weger M, Tebb S, Lee S, Rauch S. Objectifying content validity: conducting a content validity study in social work research. Soc Work Res 2003;27(2):94–104.http://dx.doi.org/10.1093/swr/27.2.94.
[29] Vogt D, King D, King L. Focus groups in psychological assessment: enhancing content validity by consulting members of the target population. Psychol Assess 2004;16(3):231–43.http://dx.doi.org/10.1037/1040-3590.16.3.231.
[30] Federal Government of Health, 2015. The professional and competency profile of the Belgian midwife.
[31] Polit D, Beck C. The content validity index: are you sure you know what's being reported? Critique and recommendations. Res Nurs Health. 2006;29(5). 489-97 9p. [32] Polit D, Hungler B. Nursing research principles and methods. Philadelphia:
Lippincott Williams and Wilkins; 1999.
[33] Lynn M. Determination and quantification of content validity. Nurs Res 1986;35:382–5.
[34] Wynd C, Schaefer M. The osteoporosis risk assessment tool: Establishing content validity through a panel of experts. Appl Nurs Res 2002;15(3):184–8.http://dx.doi. org/10.1053/apnr.2002.34243.
[35] Zamanzadeh V, Ghahramanian A, Rassouli M, Abbaszadeh A, Alavi-Majd H, Nikanfar A. Design and Implementation content validity study: development of an instrument for measuring patient-centered communication. J Caring Sci 2015;4(2):165–78.http://dx.doi.org/10.15171/jcs.2015.017.
[36] Trevelyan E, Robinson N. Delphi methodology in health research: how to do it? Europ J Integrat Med 2015;7(4):423–8.http://dx.doi.org/10.1016/j.eujim.2015. 07.002.
[37] Amendola F, Alvarenga M, Gaspar J, Yamashita C, Oliveira M. Face validity of an index of family vulnerability to disability and dependence. Rev Esc Enferm USP 2011;45(SPE2):1736–42.http://dx.doi.org/10.1590/S0080-62342011000800017. [38] Powell C. The Delphi technique: myths and realities. J Adv Nurs
2003;41(4):376–82.http://dx.doi.org/10.1046/j.1365-2648.2003.02537.x. [39] Green J, Thorogood N. Qualitative Methods for Health Research. 2nd ed Sage
Publications; 2009.
[40] Van Der Pol M, Shiell A, Au F, Jonhston D, Tough S. Eliciting individual preferences for health care: a case study of perinatal care. Health Expect 2010;13(1):4–12.
http://dx.doi.org/10.1111/j.1369-7625.2009.00551.x.
[41] Fullerton J, Thompson J, Johnson P. Competency-based education: the essential basis of pre-service education for the professional midwifery workforce. Midwifery 2013;29:1129–36.http://dx.doi.org/10.1016/j.midw.2013.07.006.
[42] Vyt A. Interprofessional and interdisciplinary collaboration in health and well-being [Interprofessioneel en interdisciplinair samenwerken in gezondheid en welzijn]. Garant; 2015.
[43] Eggermont M. The choice of child delivery is a European human right. Europ J Health Law 2012;19(3):257–69.http://dx.doi.org/10.1163/157180912X639125. [44] Kowalewski M, Jahn A. Health professionals for maternity serrvices: experiences on
covering the population with quality maternity care. In: De Brouwere V, Van Lerberghe W, editors. Safe motherhood strategies: a review of the evidence. Antwerp, Belgium: ITG Press; 2001. p. 131–50. 450 pp.
[45] Hyrkäs K, Appelqvist-Schmidlechner K, Oksa L. Validating an instrument for clinical supervision using an expert panel. Int J Nurs Stud 2003;40:619–25.http://dx.doi. org/10.1016/S0020-7489(03)00036-1.
[46] Gourounti K, Sandall J. The validation and translation of multidimensional measure
of informed choice in greek. Midwifery 2011;27:170–3.http://dx.doi.org/10.1016/ j.midw.2009.04.006.
[47] Gourounti K, Kouklaki E, Lykeridou K. Validation of the Childbirth Attitudes Questionnaire in Greek and psychosocial characteristics of pregnant women with fear of childbirth. Women Birth 2015;28(3):e44–451.http://dx.doi.org/10.1016/j. wombi.2015.02.004.
[48] Stoll K, Hauck Y, Hall W. Home or hospital? Midwife or physician? preferences for maternity care provider and place of birth among Western Australian students. Women Birth 2016;29(1):e33–8.http://dx.doi.org/10.1016/j.wombi.2015.07.187.