alexa Patient’s Preferences for Management Options in Relation to Preterm Birth | Open Access Journals
ISSN: 2157-7420
Journal of Health & Medical Informatics
Like us on:
Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on
Medical, Pharma, Engineering, Science, Technology and Business

Patient’s Preferences for Management Options in Relation to Preterm Birth

van der Ven AJ1, van Os M2, van den Wijngaard L3, Mochtar MH3, de Bekker-Grob EW4, Kazemier BM1, de Groot CJM2, Pajkrt E1, Mol BWJ5 and van Wely M3*

1Department of Obstetrics and Gynaecology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands

2Department of Obstetrics and Gynaecology, VU University Medical Center, Amsterdam, The Netherlands

3Center for Reproductive Medicine, Dept. Obstetrics and Gynaecology, Academic Medical Center, Amsterdam, The Netherlands

4Department of Public Health, Erasmus University MC, Rotterdam, The Netherlands

5The Robinson Institute, School of Paediatrics and Reproductive Health, University of Adelaide, 5000 SA , Australia

*Corresponding Author:
M van Wely
Center for Reproductive Medicine
Department of Obstetrics and Gynaecology
Academic Medical Center, University of Amsterdam
Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
Tel: 31642238278
E-mail: [email protected]

Received date: May 27, 2015; Accepted date: June 05, 2015; Published date: June 11, 2015

Citation: van der Ven AJ, van Os M, van den Wijngaard L, Mochtar MH, de Bekker-Grob EW, et al. (2015) Patient’s Preferences for Management Options in Relation to Preterm Birth. J Health Med Informat 6:189. doi:10.4172/2157-7420.1000189

Copyright: © 2015 van der Ven AJ. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at Journal of Health & Medical Informatics


To explore pregnant women’s preferences regarding cervical length measurement by ultrasound and treatment with progesterone in relation to preterm birth prevention we performed a discrete choice experiment. Four hospitals, four ultrasound centers and ten midwifery practices spread over the country participated in this study and distributed questionnaires among pregnant women between 15-36 weeks of gestation. Each questionnaire contained 16 choice sets with two screening or treatment options and one opt-out ‘no screening or treatment’ option. Women were asked to consider the following screening/treatment options 1) transvaginal or abdominal cervical length measurement, 2) vaginal or oral administration of progesterone, 3) short-term health risk and 4) long-term health risk for the child. The relative importance of the choices and trade-offs patients were willing to make were analyzed with panel-based mixed logit regression in STATA. Of the 156 questionnaires that were actually handed out, 138 were returned. Overall most respondents made trade-offs between attributes and all screening/treatment characteristics proved important in their decision making. Transvaginal cervical length measurements were not preferred (p=0.01) and was traded only in exchange for an absolute decrease of 6.5% (95% CI 2.6 – 10.4) in long-term neonatal complication rate. Previous experience with adverse neonatal outcome affected the preferences of the women.


Preterm birth; Discrete choice experiment; Cervical length measurement; Progesterone treatment; Patient’s preferences


Since preterm birth is the main contributor to perinatal mortality and morbidity, the focus of many studies is on the risk assessment on preterm birth [1-3]. Pregnant women with a short cervical length, measured either with abdominal or transvaginal ultrasound, are at increased risk of having a preterm birth [4-7]. Vaginal progesterone administration during pregnancy can potentially decrease the number of preterm births and lower neonatal mortality and morbidity [8-11].

In this context the Dutch Obstetric Consortium performed a nationwide cohort study on screening women with a singleton pregnancy for short cervical length, under the acronym ‘Triple P’ ( [12]. This acronym is an abbreviation formed of the initial P’s in ‘Preventing Preterm birth with progesterone’. Women with cervical length ≤ 30 mm were invited to participate in the embedded randomized clinical trial to receive either vaginal progesterone or placebo capsules until 34 weeks. The Triple P study started in December 2009. The power calculation dictated a sample size of 1920 participants per arm to prove or refute a 50% reduction of preterm birth (from 5% to 2.5%) after treatment with progesterone. During the study period, we noticed that considerably less pregnant women than we anticipated were willing to give informed consent and declined cervical length measurement. Moreover, to our surprise, even if the cervical length was measured ≤ 30 mm, which meant a considerable high risk on preterm birth, quite a few (27.5%, N103/375) women refused a second cervical length measurement or did not consent to randomization (47%, N71/151). Insight in patient’s thoughts regarding the risk of preterm birth and the health impact of their child, compared to the intervention (i.e. vaginal progesterone), is therefore of utmost importance.

A quantitative approach to measure preferences, which is increasingly used in health care, is a discrete choice experiment. This is a formal methodology to evaluate respondents’ preferences to explore trade-offs that patients make between different treatment alternatives. Within a discrete choice experience respondents are asked to choose between two or more alternative’s or treatment options. Discrete choice experiments have become more common and useful to investigate acceptability of interventions before general introduction [13-15].

Understanding the considerations in pregnant women in expressing their preferences can contribute to improvement in patient counseling. We choose a discrete choice experiment to value trade-offs between management options in relation to preterm birth, i.e. cervical length measurement and progesterone administration, versus health outcome of the new born child as a consequence of preterm birth.

Materials and Methods


This patient’s preference study was conducted alongside the Triple P study [12], a multicenter cohort study (Triple P screening) with a subsequent randomized clinical trial (Triple P treat) performed by the Dutch Obstetric Consortium ({HYPERLINK””}). The Dutch Consortium is a collaborative research network of university and general hospitals in The Netherlands. In the Triple P study 23 general and seven university hospitals participated, along with approximately 160 primary care midwifery practices and 29 ultrasound centres. At the time of the patient’s preference study, the Triple P continued for another year.


Mid-2012 four hospitals, four ultrasound centers and ten midwifery practices spread across the country, were asked to distribute at least ten discrete choice experiment questionnaires among pregnant women between 16-36 weeks of gestation. Only pregnant women who had been informed about the Triple P study and who were willing to undergo or already had taken part in the standard anomaly scan at 18- 22 weeks of gestation could participate. Participation was voluntary. The institutional review board of the Academic Medical Center was informed about this study and exempted the study from IRB approval. One hundred and ninety questionnaires were sent to the participating hospitals, ultrasound centers and midwifery practices. The only condition to fill in the questionnaire was sufficient understanding of the Dutch language.


The introduction section of the questionnaire consisted of explaining the purpose of the study as well as general information on preterm birth consequences, cervical length measurement and treatment with progesterone. In the explanation about cervical length measurement, vaginal ultrasound was mentioned as the most accurate and therefore the most preferable technique of cervical length measurement, especially in case of short cervix. Also, information was given regarding the administration of progesterone (oral or vaginal) and about the safety of progesterone application during pregnancy. No difference was mentioned in the efficacy of the administration. The relation between gestational age at delivery and admission to neonatal ward as a result of prematurity was also discussed, i.e. a poorer outcome is to be expected with decreasing gestational age.

Besides the introduction, the questionnaire consisted of two parts. In the first part general data about the women were collected. This included maternal age, native country, experience with prior vaginal examination, obstetric history, treatment of perinatal complications of previous born children, participation in the standard anomaly scan and obstetric care provider (primary or secondary care). In the second part preferences for cervical length measurement, treatment with progesterone and preterm birth complications were studied by means of discrete choice experiment.

At the end of the questionnaire respondents were asked to indicate the difficulty of the questionnaire on a scale of 1 to 10, in which 1 was very difficult and 10 very easy.

Discrete choice experiment

The technique of discrete choice experiment is based on the assumption that health-care interventions (or treatments) can be described by their characteristics. A discrete choice experiment investigates which tradeoffs patients are willing to make between risks and benefits of a certain treatment option. Figure 1 illustrates the concept of this discrete choice experiment and its definitions. The characteristic s or so called ‘attributes’, consisted of treatment options and risks, and are described in the first column. Scenario A and B describe the possibility of choosing a treatment with varying health risks. Scenario C, the opt out, can be selected if the respondent prefers no optioned treatment. The risk outcome in the opt-out scenario was always equal to or worse than in the other scenario’s.


Figure 1: Example of choice set as presented in the questionnaire.

The respondents were asked to make a choice between two or more sets of hypothetical alternatives with systematically varying combinations of attribute levels [15]. The importance of the attributes can be estimated by analyzing the choices the responders made between attributes and attribute levels.

Attribute and attribute levels

The selection of the attributes and levels was based on data from literature on prevalence of preterm birth and short- and longterm impact of prematurity [16-24]. After discussing the short and long-term risk levels in the expert group, these were converted into understandable explanations and risk levels for respondents. From the literature it is known that in discrete choice experiment studies 4-6 attributes is most common and that most studies use 9-16 choices [25,26]. The greater the number of attributes, the more difficult it is to complete a discrete choice experiment. A completion time between 10-15 minutes seemed mostly acceptable [25]. Potential attributes and levels were discussed with specialists on discrete choice experiment development as well as with obstetric care providers and pediatricians. After consensus on the attributes and levels by both expert groups, the questionnaire was presented to two midwives and four pregnant women to test whether the questionnaires were comprehensible. After that, no changes were made. Four attributes were selected: cervical length measurement, treatment with progesterone and short- and long-term preterm birth complications (Figure 1) with three levels per attribute. Attributes and corresponding levels are shown in Table 1.

Attributes Levels
cervical length measurement none
  abdominal ultrasound
  vaginal ultrasound
progesterone administration none
  oral administration
  vaginal administration
risk of short term health care problems due to prematurity 2%
risk of long term health care problems due to prematurity <5%

Table 1: Attributes with corresponding levels.

Short term health care problems due to prematurity were defined as temporarily neonatal hospitalization for instance because of respiratory problems, glucose fluctuations, temperature- or nutritional problems or jaundice.

Long term health care problems due to prematurity were defined as complications requiring permanent care and attention, not only during the period of hospitalization. This included developmental delay, learning- or behavioral problems, impaired concentration, vision- or hearing problems and spasticity. Both short and long-term problems were transformed into attributes as complication rates due to prematurity.

Respondents had to choose their most preferable option in each choice set. Every choice set included one no treatment (opt-out) option. This opt-out was necessary since, as in real life, respondents are not obliged to take a treatment. Thus the women were asked to choose in each choice set between three hypothetical alternatives, two screening/ treatment options differing in 1) vaginal or abdominal cervical length measurement, 2) vaginal or oral administration of progesterone, 3) short-term health risk and 4) long-term health risk for the new born child, and one ‘no screening/treatment’ opt out option.

Development of choice sets

The combination of four attributes and three levels per attribute provided 81 (34) hypothetical alternatives. The alternatives were placed into balanced choice sets with a minimum overlap. We used a fractional factorial design to generate a functional sample of 16 alternatives. The fractional factorial method systematically selects this sample according to an orthogonal design. Orthogonality guarantees an optimal balance of the attribute levels with zero correlation between the attributes [27,28]. The orthogonal design was generated by Orthoplan (Statistical Package for Social Sciences [SPSS] version 16 SPSS Inc., USA).

The 16 alternatives formed ‘treatment A’ in each of the 16 choice sets. To ensure minimal overlap of attribute levels, we created a set of alternatives to form ‘treatment B’ by means of a syntactical fold over technique, based on the profiles of ‘treatment A’ [15]. This resulted in 16 different choice sets, whereby each choice set consists of two alternatives representing hypothetical risks and treatment options in relation to preterm birth. A not changing opt-out or no treatment option was added as a third alternative. The health care consequences in this no screening/treatment alternative were either equal or worse than one of the other screening/treatment alternatives. The 16 choice sets for screening/ treatment option A and B, and C (opt-out) were considered sufficient to estimate all main effects representing the relative importance of each attribute or attribute level. To assess the understanding of the attributes (treatment options and risk effects), the questionnaire contained one dominant question in which all attributes were in favor of one specific choice. This is called a rationality test.


A mixed logit regression model for panel data was employed to analyze the effect of the attribute levels on women’s preferences In STATA 12.1. (Hole AR 2007). Each discrete choice experiment attribute was included in the regression model. The ‘no treatment’ alternative was included as an alternative specific constant to account for any latent or uncontrolled factors when choosing the ‘no treatment’ alternative.

Short-term and long-term complication rates were coded as continuous variables after testing for linearity. As cervical length measurement and the progesterone administration in case of short cervix are categorical variables, these were recoded as -1, 0 and 1 (effects coding).

The constant of the model was set as a random parameter. Subsequently random parameters for attributes were included based on the model fit (AIC and Chi-square tests). A normal distribution was assumed.

The statistical significance of a coefficient (p-value ≤ 0.05) indicates that individuals differentiated between one attribute (or attribute level) and another in making stated choices. The sign of a coefficient reflects whether the attribute has a positive or negative effect on preference score.

The value indicates the relative importance of the attribute to total relative utility. A statistically significant coefficient indicates that respondents considered that attribute important. We expected that the attributes short- and long-term risks would have a negative effect to reflect the preference for low risks/complications. If long-term risk would be valued as more important than short –term risk then this would be reflected in a higher negative preference value.

Trade-offs that respondents are willing to make between attributes was estimated by calculating the ratios of the coefficients of two attributes where we also accounted for preference heterogeneity. As both the constant and the expected outcome attributes were included as random parameters in the analyses, the trade-offs could not be calculated directly. We calculated importance scores to visualize the relative importance of a given attribute by dividing the difference in utility between highest and lowest level for a single attribute by the sum of the differences of all attributes. A simulation (n=1000) was used to estimate the trade-offs.

A sensitivity analysis was done excluding women that failed the rationality test. Subgroup analyses were conducted using two-way interaction terms in the regression model to assess the effect of specific baseline parameters. In case of a significant interaction results are presented sub grouped for that term.


In May 2012, 190 questionnaires were sent to the participating hospitals (4), ultrasound centers (4) and midwifery practices (10) to be distributed to the pregnant women who are under their control. In November 2012 actually 156 of the 190 the questionnaires were handed out, and 138 were returned, resulting in a response rate of 88%. In total 128 respondents filled in the questionnaire after the 20-week ultrasound examination was completed, six women before the 20-week ultrasound examination was performed and four of the respondents had not filled in whether their cervical length was measured or not.

One questionnaire was excluded from the analysis as none of the discrete choice experiment questions were answered. All other 137 questionnaires were fully completed with only some exceptions in the general information section, i.e. once postal code was missing, once date of birth, once date of completion and twice gravidity. The baseline characteristics of the respondents are shown in Table 2.

Baseline characteristics N=138
Age in years,  mean (SD) 31 (4.3)
Native country, N (%)  
Netherlands 130 (94.2)
Gravidity  N (%)
8  (5.8)
First pregnancy 60 (43.5)
Second pregnancy  or more 76  (55.1)
Gestational age(weeks), mean (SD) 23 (3.9)
Gestational age at time responding N (%)  
Before anomaly scan 8 (5.8)
After anomaly scan and < 32 weeks gest. 121 (87.7)
≥ 32 weeks of gestation 7 (5.1)
Previous vaginal examination N (%)  
No previous vaginal examination 26  (20.3)
Discomfort/pain during previous examination 30 (21.7)
Cervical length measurement N (%)  
Agreed 65  (47.1)
Refused 49  (35.5)
Unknown/not offered 24  (17.4)
Previous child needed extra care, N (%) 12  (8.7)
Antenatal care provider, N (%)  
Primary care midwife 104  (75.4)
Obstetrician in secondary/tertiary care 28  (20.3)
Other/unknown 6  (4.3)

Table 2: Baseline characteristics of the 138 respondents.

The dominant question- in which the most favorable outcome was related to the least interventions, -treatment and complication rates was answered as expected by 132 women (97%), only five women did not provide the expected answer. The opt-out / no treatment option was chosen in 21% of the choice sets. Ten women opted for the ‘no treatment’ choice at all 16 questions, i.e. these women did not want to trade-off their choice. These 10 women did neither opt for cervical length measurement nor for progesterone administration. The questionnaire was difficult to answer for 41 (30%) women, who scored the difficulty as 5 or lower on a scale of 1 to 10. Still, there was no difference between the choices these women made and those of the women who found the questionnaire not difficult. Seventy six women (56%) scored the questionnaire as easy with a score of 7 or higher. Five respondents did not answer this question.

Table 3 shows the results of the regression model, which contains the main effects of the attributes. All coefficients were statistically significant in all cases on the choice-making of respondents. The mean coefficient indicates the relative likelihood of choosing a treatment alternative with a given attribute-level combination holding all other factors constant. A larger value indicates a greater likelihood of choosing a treatment alternative with the specific feature. No cervical length measurement was preferred above a cervical length measurement by abdominal ultrasound examination (mean coefficient -0.77 vs 1.69; p<0.01) or vaginal ultrasound (mean coefficient -0.87 vs 1.69; p<0.01).

Attributes and levels Mean coefficient (95%CI) Standard deviation (95% CI)
Constant# 3.65 (1.41 to 5.89)* 1.56 (0.34 to 2.78)*
Cervix length measurement    
         No measurement (omitted) 1.64 (1.10 to 2.18)* 1.69 (0.29 to 3.19)*
         Abdominal -0.77 (-1.03 to -0.51)* 0.38 (-0.03 to 0.79)
Vaginal -0.87 (-1.39 to -0.35)* 0.63 (0.05 to 1.21)*
Progesterone administration    
           No (omitted) 0.71 (0.63 to 0.99) 0.21 (0.03 to 0.39)*
           Oral -0.21 (-0.26 to -0.16)* 0.05 (0.00 to 0.09)
           Vaginal -0.49 (-0.70 to -0.28)* 0.12 (0.01 to 0.23)*
Short-term complication rate (per 1%) -0.16 (-0.24 to -0.08)* 0.04 (-0.06 to 0.15)
Long-term complication rate (per 1%) -0.37 (-0.55 to -0.19)* 0.12 (-0.05 to 0.29)
Number of responses 6624  
Number of respondents 138  
Log likelihood -874  
AIC 91  
BIC 97  

Table 3: Four attributes were used for discrete choice experiment to assess women’s preferences for. The negative sign of the coefficient reflects a negative effect on utility, the value indicates the relative importance of the attribute to total relative utility. A statistically significant coefficient indicates that respondents considered that attribute important.

No progesterone administration was preferred above oral (mean coefficient -0.21 vs 0.71; p<0.01) and vaginal progesterone (mean coefficient -0.49 vs 0.71; p<0.01). Oral progesterone was preferred above vaginal progesterone (mean coefficient -0.21 vs -0.49; p<0.05).

The standard deviation of the mean coefficient describes the degree to which respondent preferences were heterogeneous. As can be seen from Table 3 most estimated standard deviations were significant. Larger values indicate more heterogeneity across respondents. For example, from the results it can be inferred that no cervical length measurement had the greatest preference heterogeneity with an estimated standard deviation of 1.69. This indicates that although most respondents preferred no cervical length measurement (mean preference = 1.64) a part of the participant actually did prefer a cervical length measurement.

The sensitivity analysis, excluding the five women that did not correctly answer the dominant question, did not influence the main effects of the model.

Willingness to trade preferences

Most respondents were willing to make trade-offs between attributes. No cervical length measurement would be traded for a vaginal cervical length measurement in exchange for a 6.5% (95% CI 2.6 to 10.4) decrease in long-term complication rate. Similarly, no progesterone administration would be traded for vaginal progesterone administration in exchange for a 7.3% (95% CI 5.1 to 9.6) decrease in short-term complications risk.

Effect of baseline parameters

In a secondary analysis we evaluated the effect of baseline parameters on the choices of the participating women. Of the baseline parameters only women who previously had a new born child that needed extra care was a significant interaction term. Women who previously had a new born child that needed extra care made different choices; these women preferred the use of a vaginal cervical length measurement while means of progesterone application was not an important attribute for this group of women. As only 12 women had a previous newborn that had needed extra care (for instance because of asphyxia, hypothermia, hypoglycaemia, small- or large for gestational age, jaundice, infection etc.) the power of this sub analysis was very low. The sub analysis stratified for women who did and who did not previously had a new born child that needed extra care is shown in Table 4.

Attributes and levels Previous delivery requiring extra neonatal care  No extra neonatal care previous delivery
  Mean coefficient (95% CI) SD (95% CI) Mean coefficient (95% CI) SD (95% CI)
Constant 2.87 (0.65 to 5.09)* 0.35 (0.01 to 0.69)   0.05 (-0.11 to 0.22)* 1.03 (0.04 to 2.02)*
Cervix length measurement        
        No measurement (omitted) -1.77 (-2.85 to -0.65)* 0.81 (-0.07 to 1.69)   1.57 (1.05 to 2.09)* 1.14 (1.01 to 1.27)*
        Abdominal 0.65 (-0.13 to 1.43) 0.45 (0.01 to 0.89)*  -0.68 (-0.93 to -0.43)* 0.25 (0.06 to 34)*
        Vaginal 1.12 (0.02 to 2.22)* 0.58 (-0.16 to 1.32)  -0.89 (-1.31 to -0.47)* 0.72 (0.02 to 1.42)*
Progesterone administration        
        No (omitted) 0.02 (-0.01 to 0.05) 0.13 (-0.09 to 0.35)   1.17 (0.98 to 1.36) 0.71 (0.17 to 1.24)
        Oral −0.21 (-0.43 to 0.01) 0.25 (-0.02 to 0.52)  -0.52 (-0.67 to -0.37) 0.09 (-0.03 to 0.12)
        Vaginal 0.22 (0.03 to 0.41)* 0.20 (-.11 to 0.51)  -0.65 (-0.84 to -0.46) 0.52 (0.03 to 1.01)*
Short-term complication rate (1%) -0.16 (-0.31 to -0.01) 0.02 (-0.03 to 0.07)   -0.17 (-0.31 to -0.03) 0.07 (-0.09 to 0.23)
Long-term complication rate (1%) -0.43 (-0.73 to -0.13) 0.04 (-0.07 to 0.15)   0.37 (-0.58 to -0.16) 0.09 (-0.11 to 0.39)
Log likelihood   -564      

Table 4: Results of the sub analysis stratified for women who did and who did not previously had a new born child that needed extra care.


Mean findings

We evaluated women’s preference for cervical length measurement and progesterone administration in relation to health problems of their new born child due to prematurity. The participating low risk women generally expressed a preference for least interventions and least side effects but were willing to make trade-offs between attributes when this resulted in better health outcomes for their child. A cervical length measurement and progesterone administration were not preferred. However a transvaginal ultrasound cervical length measurement was accepted in exchange for a 6.5% decrease in long-term neonatal complication rate.

Opposite to the general population, the subgroup of women who experienced adverse neonatal outcome did have a preference for transvaginal ultrasound cervical length measurement. Due to their experience, these women are more likely to be fully aware of the risk of preterm birth. It must be stressed however that the power for this sub analysis was limited due to the low number of respondents (8.8%, 12/137) who experience adverse neonatal outcomes. Although the numbers in the subgroup analysis were small, this confirms clinical experience that women with a previous adverse outcome, will try to avoid a repetition of that adverse outcome.

Strengths and limitations

Our study provides insight into the relative weight women place on risk selection and health outcome of preterm birth and trade-offs they make. As far as we know this is the first discrete choice experiment in relation to risk selection and health outcome due to prematurity.

A limitation of our study may be that the participating women did not fully understand the questions and/or presented risks. It is known from literature that individuals have difficulty in understanding risk assessment [29,30]. Risk communication in complication rates may be hard to understand and therefore respondents may have had different perceptions towards risk problems due to prematurity. In an effort not to make the choice sets unnecessarily complicated, we opted to describe the risk assessment as a result of prematurity as short and long term complications (with additional information in the introduction section of the questionnaire). So, we decided not to make a further distinction according to gestational age at birth. Nevertheless, although the majority of the participating women reported that the questionnaire was easy to understand, 30% of the women reported the questionnaire to be difficult (score 5 or lower). Still, the choices these women made did not seem different from the choices the women made that judged the questionnaire to be easy. Furthermore, the dominant control questions were answered correctly by 97% of the women. It seems therefore that the difficulty of the questionnaire did not have a major effect on the validity of the results.

Several models are available to analyze discrete choice data [14]. A mixed logit model or a latent class model were both good alternatives to analyse the choice observations in the present discrete choice experiment. Testing the basic model using a latent class model with three knots resulted in a comparable AIC, making it unlikely that a latent class model would have resulted in different estimates.


The best hope for reducing the incidence of preterm birth at present seems to be screening followed by treatment of women at risk. Therefore it is important to have as much information as possible about women’s attitudes regarding this approach. Ongoing the triple P study, it was noticed that considerably less pregnant women were willing to measure their cervical length than expected. Moreover, even if the cervical length was measured ≤ 30 mm, quite a few women refused the second measurement or did not consent to randomization. It seemed that women do not want to think about future pregnancy risks or, if they do, seem to believe that it will not happen to them. Another explanation could be a lack of awareness of the complications as a result of (even late) preterm birth in pregnant women. This seems to be confirmed by the study of Goldenberg et al. [31,32], who reported that half of 650 surveyed women believed it is safe to deliver before 37 weeks of gestation. Nevertheless, the present study shows that most women were willing to trade-off their first choice. Moreover, women with previous adverse neonatal outcomes make different choices and preferred the use of a transvaginal cervical length measurement, thus confirming our hypothesis that the given information about the consequences and the prevention of preterm birth are essential to make a trade-off.

Future counseling

The Dutch health care system has been reformed in 2006 to make it more patient-oriented and demand-driven. The Dutch governmental healthcare internet portal for patients hosts at least 16 patient decision aids. The information about screenings test (such as first trimester screening and 20 weeks anomaly scan) that has to be provided to pregnant women is well documented and includes a counseling interview, nationwide brochure and access to information on the Internet, under supervision of the government (Ministry of Health, Welfare and Sports). Furthermore, counselors are all registered and meet national criteria.

With the advent of screening tests, there is more focus on informed choice, the way in which information is transmitted and monitoring whether the information is understood. In obstetric care in the Netherlands information is generally communicated orally with each patient individually, often supported by brochures and with references to reliable internet sites. Increasingly more information is available in several languages and is supported with pictures, images and/or video clips. All needs to be documented in the patient records.

There is free access to health care information on the website of The Dutch Society of Obstetrics and Gynecology (NVOG) as well as of the Royal Dutch Society of Midwives (KNOV), although quality standards and guidelines have been shielded. Nowadays, the Internet is a commonly used source for (medical) information. The disadvantage is that it is difficult for patients to assess the reliability of the information. Besides that, it is only possible to search for information if one is aware of specified personal health risks and applicable tests or -treatments. In this perspective, our study clearly shows that it is important to fully inform patients prior to any examination or treatment, in particular when the patient is not familiar with these examination or treatment (yet). This is of value in clinical practice regarding medical research or implementation of new medical insights.


This study shows that women at low risk for preterm birth generally expressed a preference for least interventions but were willing to make trade-offs between attributes when this resulted in better health outcomes for their child. The results of this study can be used to improve the counseling for the prevention of preterm birth in pregnant women and to achieve an enhanced participation in screening and treatment programs to prevent preterm birth.


We thank all midwives, sonographers, residents and research nurses of the participating midwifery practices, ultrasound centers and hospitals for their help with patient recruitment. We are grateful to all participating pregnant women for their willingness to fill in the questionnaire. We also thank the members of the DCE-project group of the Academic Medical Center for their advice in developing the questionnaires and the design of this study.


None of the authors have a conflict of interest.

Ethic Approval

The study was exempted from IRB approval.


Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Relevant Topics

Article Usage

  • Total views: 11588
  • [From(publication date):
    June-2015 - Nov 19, 2017]
  • Breakdown by view type
  • HTML page views : 7818
  • PDF downloads : 3770

Post your comment

captcha   Reload  Can't read the image? click here to refresh

Peer Reviewed Journals
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals
International Conferences 2017-18
Meet Inspiring Speakers and Experts at our 3000+ Global Annual Meetings

Contact Us

Agri & Aquaculture Journals

Dr. Krish

[email protected]

1-702-714-7001Extn: 9040

Biochemistry Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Business & Management Journals


[email protected]

1-702-714-7001Extn: 9042

Chemistry Journals

Gabriel Shaw

[email protected]

1-702-714-7001Extn: 9040

Clinical Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Engineering Journals

James Franklin

[email protected]

1-702-714-7001Extn: 9042

Food & Nutrition Journals

Katie Wilson

[email protected]

1-702-714-7001Extn: 9042

General Science

Andrea Jason

[email protected]

1-702-714-7001Extn: 9043

Genetics & Molecular Biology Journals

Anna Melissa

[email protected]

1-702-714-7001Extn: 9006

Immunology & Microbiology Journals

David Gorantl

[email protected]

1-702-714-7001Extn: 9014

Materials Science Journals

Rachle Green

[email protected]

1-702-714-7001Extn: 9039

Nursing & Health Care Journals

Stephanie Skinner

[email protected]

1-702-714-7001Extn: 9039

Medical Journals

Nimmi Anna

[email protected]

1-702-714-7001Extn: 9038

Neuroscience & Psychology Journals

Nathan T

[email protected]

1-702-714-7001Extn: 9041

Pharmaceutical Sciences Journals

Ann Jose

[email protected]

1-702-714-7001Extn: 9007

Social & Political Science Journals

Steve Harry

[email protected]

1-702-714-7001Extn: 9042

© 2008- 2017 OMICS International - Open Access Publisher. Best viewed in Mozilla Firefox | Google Chrome | Above IE 7.0 version