Experiences with an integrated screening programme targeted at women who had a hypertensive disorder or diabetes in pregnancy in the Netherlands: a qualitative study

Abstract

Objectives

Pregnancy is a predictor of women’s health later in life. The risk of eventually developing cardiovascular disease (CVD) and/or type 2 diabetes (T2D) increases three to seven times, after pre-eclampsia, hypertension gravidarum or gestational diabetes. The Heart for Women in The Hague project was designed to offer targeted screening to this high-risk population. This research aimed to gather insight regarding the initial experiences of healthcare providers and women with pre-eclampsia, hypertension gravidarum or gestational diabetes.

Design and setting

This study applied a qualitative semistructured interview design using an interview guide based on ‘The Consolidated Framework for Implementation Research (CFIR).’ Interviews were recorded, transcribed and coded in accordance with the five main CFIR themes. The setting of the study was primary and secondary care in The Hague, the Netherlands. Interviews were carried out from December 2021 until February 2022. The language was Dutch or English.

Participants

Participants (n=13) were women of the target population (n=7) and relevant healthcare professionals (n=6).

Results

Healthcare providers and women were generally positive concerning screening. Healthcare providers agreed that this forgotten group deserves attention and felt that the regional transmural agreement (RTA) provided a clear policy. Women stated that they would like screening to continue and were open-minded regarding lifestyle guidance. Reported barriers included current lack of an easy way of contacting the population after a year, lack of knowledge among the women concerning increased risk, unfamiliarity of new colleagues with the RTA and lack of evidence that screening actually prevents disease.

Conclusion

Implementation of screening programmes to prevent or delay the development of CVD and/or T2D after complicated pregnancies will likely improve awareness in both patients and healthcare providers. Healthcare providers considered the RTA important because it concerns a forgotten high-risk population. Future research should focus on the improved effectiveness of tailored interventions to delay or prevent CVD.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The women interviewed in this study were highly diverse in terms of socioeconomic status, ethnic background and pregnancy complications, therefore our conclusions can be applied in neighbourhoods with a diverse population.

• By following a semistructured style of interviewing, the respondents could express their own views and opinions while the interviewer maintained an overview.

• Results are based on a small sample size; a wide range of participants should be included in future studies.

• Purposive sampling was used to recruit participants to the study: in this way healthcare providers were already invested in the project. This also added to the potential to optimise healthcare for the women in (and out) of this study.

• By collaborating with patient-run support groups, we have strengthened our research.

Introduction

About 5%–10% of all pregnant women experience a hypertensive disorder during pregnancy (HDP), often including pre-eclampsia and gestational hypertension. The risk of chronic hypertension in later life following an HDP episode is 3–4 times higher compared with women with normal blood pressure during pregnancy.^1–3 Previous studies reported that up to 42% of women with severe pre-eclampsia still showed elevated blood pressure 1–10 years later.^{4 5} A history of HDP is also an individual risk factor for lifetime onset of cardiovascular disease (CVD).² In addition, 5%–9% of pregnant women experience gestational diabetes mellitus (GDM).⁶ Women with a history of GDM have a seven times higher risk of developing type 2 diabetes (T2D) later in life compared with women who maintain normal glucose levels during pregnancy.⁷ In addition to the increased risk of T2D, women with GDM also have a twofold increased risk of developing CVD later in life compared with women without GDM.⁸ If a pregnant woman experiences both HDP and GDM, the risk of developing hypertension and/or T2D is even higher with an adjusted HR of, respectively, 16.2 (95% CI, 13.2 to 19.9) and 16.8 (95% CI, 11.8 to 24.1).⁹ Thus, a woman’s physiological state during pregnancy predicts her risk of developing CVD and T2D later in life.¹⁰

Surprisingly, these important risk factors are not yet systematically included in CVD and T2D screening.¹¹ Several countries have incorporated pregnancy-related risk factors in their guidelines for CVD prevention, identifying a number of pregnancy-associated conditions with an excess risk of CVD.^{12 13} Whether sex-specific risk factors can be used to assess high-risk women beyond traditional risk-scoring algorithms is still not conclusive.¹⁴ A recent Dutch study concluded that addition of pregnancy complications to traditional eligibility criteria for CVD risk screening does not improve the prediction of CVD risk factors. But in the absence of screening, pregnancy complications may function as an eminent warning.¹⁵

The fact that 27% of women in the Netherlands will eventually die due to CVD shows that screening is desperately needed.¹⁶ CVD has also been linked to several other risk factors, including lifestyle-related (and thus modifiable) risk factors, such as smoking and physical activity.¹⁷ As lifestyle interventions can help reduce cardiovascular risk^18–20 and T2D,²¹ introducing periodic check-ups will allow lifestyle changes to be discussed and symptoms of disease to be addressed earlier.²²

Previous studies have mainly explored the possibility and attitudes towards the screening of women in this group but have not yet evaluated the implementation of a screening programme.^23–26 These studies recommended optimising a screening programme for this target population, concerning both monitoring and educating these women and getting these women into focus and keeping them in focus by their general practitioner (GP).

The Heart for Women in The Hague project was set up to detect an increased risk of CVD and/or T2D at an early stage in women who experienced HDP and/or GDM. Based on existing literature, these women had an acknowledged elevated risk; however, they were not regularly monitored prior to this project.

The aim of this study was to better understand the experiences of women and healthcare providers regarding an integrated multidisciplinary care path for women with pregnancy-related hypertensive disorders or GDM.

Materials and methods

Setting

The implementation of Heart for Women in The Hague began in March 2021. This involved setting up an integrated multidisciplinary care path that included obstetricians, gynaecologists, cardiologists, vascular internists, GPs and practice nurses (see online supplemental appendix 1). The first phase of the project consisted of gathering additional information from the gynaecologist’s/obstetrician’s referral letter to the GP regarding HDP and/or GDM. If the pregnancy occurred several years previously, women were usually not informed about, or had already forgotten about their higher risk for CVD and T2D. Therefore, in the second phase of the project an algorithm-based strategy was piloted in general practices to help identify women with a history of HDP and/or GDM. Women identified using these two approaches were offered appropriate screening and prevention in the form of periodic invitations for check-ups with the GP or practice nurse. An evaluation of these periodic screening check-ups is now needed to better map the needs, wishes and experiences of both patients and care providers, which will help further improve the project.

Blood pressure, heart rate, weight (body mass index), glucose levels, kidney function and cholesterol levels were checked during annual screening, together with other risk factors such as lifestyle, diet and family history for cardiometabolic diseases. In the first 5 years after the last pregnancy, with HDP and/or GDM, the annual screenings were performed each year. After the first 5 years after pregnancy, the screenings will be performed once every 3 years. This frequency was chosen using a combination of national and international guidelines, together with expert-based evidence.^{27 28} This manuscript was written in accordance with the consolidated criteria for reporting qualitative studies, which can be viewed as online supplemental file 1. The original study protocol is also available as online supplemental file 2.

Data collection of qualitative measures

Semistructured interviews lasting 10–30 min were carried out from the beginning of December 2021 until the end of February 2022. The interviews were conducted by the same researcher (SS) using either video calls (Microsoft TEAMS or ZOOM), or face-to-face in a healthcare setting, in a private room with no interruption, where each person was seated in a chair at a table. Field notes were made during the interviews by a student, who was introduced by the researcher to the participant after consent. The student was sitting next to the interviewer at the table, facing the patient. All researchers and the student were female.

First, a topic list was established using the Consolidated Framework for Implementation Research (CFIR). The CFIR is a compilation of 39 implementation-related constructs divided across five domains: (1) characteristics of the intervention; (2) outer setting; (3) inner setting; (4) characteristics of the individuals involved and (5) the process of implementation. According to Damschroder et al, researchers may select the CFIR constructs that are most relevant to their study setting.²⁹ Table 1 shows the subcategories included in the topic list for this evaluation. Process evaluation in this project included the experiences and opinions of both healthcare providers and women in the target population, with the topic lists differing according to the target population (table 1). An interview guideline was then prepared for use during interviews (see online supplemental appendix 1), and there was time during interviews for discussion of additional topics when these arose. The interviewee always ended with an open question concerning possible topics or remarks that the respondent might like to share.

Table 1.

Topic-list interviews

CFIR	Subcategories	Which respondent
Intervention characteristics	Experiences Materials	HP and W HP and W
Outer setting	Communication Patients’ needs External policies	HP and W W HP
Inner setting	Implementation climate	HP
Individual characteristics	Beliefs of evidence Views on RTA	HP HP
Process	Training Workload Reach Champion	HP HP HP and W HP

CFIR, Consolidated Framework for Implementation Research; HP, healthcare provider l; RTA, regional transmural agreement; W, women in target population.

Once the interview guide was developed, it was first assessed by the patient expert committee prior to the interviews. Next, during the interviewing phase, the introduction of questions was adjusted if necessary. When no new themes were raised, it was concluded that saturation had been reached.

Participants

Healthcare providers were selected on the basis of their profession. This allowed the specific experiences of each profession to be investigated. Healthcare providers were eligible if they met the following inclusion criteria: they had referred a patient with HDP and/or GDM to the GP and/or had invited a patient with HDP and/or GDM for screening check-ups.

Women with the appropriate medical history were recruited via a GP practice that had implemented the algorithm. Women eligible for periodic screening check-ups based on their medical history were first identified using the algorithm and then invited by their GP. After a periodic screening check-up, they were asked if they wished to be included in the research study by means of purposeful sampling. Women were eligible if they met the following inclusion criteria: a medical history of HDP and/or GDM, an invitation from their GP/practice nurse for a periodic screening check-up, able to speak Dutch or English and aged 18 years or older. After informed consent, no participant dropped out.

In total, 13 interviews were conducted (healthcare providers n=6, women in target population n=7). The professions of healthcare providers were diversely distributed (table 2). Characteristics noted for women in the target population included age range, pregnancy complications, years after birth the screening took place and the language in which the interview was conducted (table 3). The sample was diverse. The women represented different ethnicities with two respondents from the continent of Africa, one from southern Europe and four from northern Europe. The average length of interviews with healthcare providers was 23 min. One interview was not recorded due to technical problems, but the interview was written down afterwards and checked with the respondent. The average length of interviews with women in the target population was 13 min.

Table 2.

Respondents’ healthcare providers

Respondent	Function	Stage in healthcare
1	Practice nurse	CVRM/DM checks
2	GP	CVRM/DM checks
3	Gynaecologist	Postpartum care
4	Obstetrician	Postpartum care
5	Gynaecologist	Postpartum care
6	Practice nurse	CVRM/DM checks

CVRM, cardiovascular risk management; GP, general practitioner.

Table 3.

Respondent women in target population

Respondent	Age range	Pregnancy complication	Years after birth	Language
1	45–50	Gestational hypertension	15 years	Dutch/English
2	35–40	Hemolysis Elevated Liver enzymes Low Platelets (HELPP)	1 year	English
3	35–40	Pre-eclampsia	1 year	Dutch
4	30–35	Gestational hypertension and gestational diabetes	1.5 years	Dutch
5	40–45	Gestational diabetes	4 years	Dutch
6	30–34	Gestational diabetes	1 year	Dutch
7	35–40	HELPP	4 years	Dutch

The information folder and informed consent form were approved by the Medical Research Ethics Committee of Leiden, The Hague and Delft (G18.070). As participants are not subjected to actions other than screening based on a regional transmural agreement (RTA) between primary care and secondary care and rules of conduct are not imposed on them, the study itself was exempted under Dutch law (WMO). Compensation was offered in the form of a €20 gift voucher.

Data analysis qualitative measures

The interviews were recorded and transcribed by SS, and the transcriptions returned to the interviewees to ensure accuracy. Field notes were made during the interviews by a student. The transcripts were independently analysed by two researchers (SS and HV) in an open coding phase using Microsoft Excel version 2102. Fragments of the transcript important to answering a question were coded during this phase. By discussing the codes, intercoder agreement and consensus was reached between the two researchers. Next, the relationship between codes was investigated during an axial coding phase, followed by the grouping of codes into categories that were then divided into themes based on the CFIR. When a code did not fit within one of the five domains, new codes or themes were established. These themes allowed experiences to be grouped and the barriers/facilitators of the project defined. Finally, adjustments were proposed, and feedback provided to the project group’s principal investigators.

Patient and public involvement statement

A patient expert panel was involved in the first stage of research. The patient expert panel consisted of three women who have had HDP or T2D and are involved in different patient-run support groups. They helped with the design of the patient information folder, the invitation letter, and the way the screening was set-up (eg, if bloodwork would be included). The interview guide and the burden presented by the interview were also reviewed and discussed with them. Vrouwenhart, a patient-run support group, has already discussed some aspects of the study in their videos and they plan to disseminate the study results within their patient network. One of the members of Vrouwenhart was also a member of the patient expert panel.

Results

The results will be discussed following the CFIR domains in the following text, where possible with a quote to clarify. A table containing the categories fitting with the CFIR domains and their supporting quotes can be found in online supplemental appendices 2,3.

Intervention characteristics

Gynaecologists and obstetricians stated that the intervention itself did not materially differ from their regular activities. The only additional requirement was the use of a template referral letter. The postpartum check-up at 6 weeks was already standard care, as was informing women about the risks after a complicated pregnancy. The GP and practice nurses stated that the intervention resembled their current approach to treating T2D and cardiovascular risk management (CVRM). The algorithm was a useful tool for creating an overview of women eligible for screening. There is still no easy way in the system to keep track of women eligible for an annual invitation, but it was a necessary issue to tackle. The letter from the gynaecologist or obstetrician advising to invite women to the GP practice was appreciated, even by healthcare providers who were not yet using the algorithm. The RTA was felt to be a clear, accessible guideline and ensured good cooperation between involved healthcare providers.

The participating women reported a positive experience with the check-ups. The invitation letter from the GP practice was clear and portrayed the check-up as fast and easy. They were content with the fact that their GP invited them because they could not recall that the gynaecologist or obstetrician had discussed their increased risk after delivery, which they thought might be due to the circumstances around giving birth. This is demonstrated in the following quote:

I can't remember [the explanation about increased risk] but I was panicking, so I may not have registered it, but I can't remember—Respondent 4

Some participants stated that they would have liked to have received a folder or a website link, so that they could read more about their increased risk.

Outer setting

The gynaecologists and obstetricians reported no complaints regarding referral to the GP practice or the internal medicine department.

The GPs and practice nurses reported that they would like to see a part of the RTA included in the guidelines of the NHG (Nederlands Huisartsen Genootschap,i.e. the Dutch College of General Practitioners). This would make it easier to determine a GP’s responsibilities.

Women in the target population were asked if their needs were met during the check-ups. They appreciated the annual screening by their GP/practice nurse and that they found it reassuring. Respondent 2 mentioned the following to express their appreciation:

And like I said, you naturally just start looking after the baby and you don’t actually check up on yourself again, so that’s why this is just so important to do, you know. So, I’m very grateful for that.—Respondent 2

One of the women commented that she didn’t need it, since she was already included in an integrated care programme for other chronic conditions. The out-of-pocket cost for the laboratory analysis was not reported as a barrier by the women in the target population. The invitation letter met their needs, and all the information was clear. A translation was not needed. Nonetheless, three respondents who did not speak/read Dutch fluently mentioned that an email would be useful because it is easier to translate. Most of the women also reported that they did not object to the lifestyle questions and were open to personalised advice regarding lifestyle changes.

Inner setting

The healthcare providers involved in the project reported that their colleagues had a positive reaction, had no questions concerning the RTA and reported that they accepted it. They liked the fact that it was a single clear policy that promoted cooperation among different specialists. The departmental knowledge of the gynaecologists and obstetricians was present. While one respondent reported that this knowledge wasn’t present in the GP practice, this could be explained by the fact that this particular GP practice did not participate in the (retrospective) algorithm aspect of the project.

Individual characteristics

The healthcare providers had a positive view of the RTA and felt that it targeted a population often forgotten in standard care and guidelines. They considered the RTA useful as it regards prevention of a later life health burden in these women. A gynaecologist commented the following:

Yes, I think it’s good, there’s definitely a lot of benefit to be had, I think it’s good, yes. It is such a shame to find this out at a later stage and only then start playing catch-up, while if you had identified those people earlier you could have already done something, so I think it is a very valuable guideline.—Gynaecologist

Nevertheless, some participants indicated that more evidence is needed concerning CVD and T2D screening and its cost-effectiveness.

Process

The gynaecologists and obstetricians reported no change in workload due to the RTA, as it was only one extra ‘mouse click’. One of the gynaecologists explained this as follows:

Yes, because it is not really different care compared to what we already provided before the project started. It is not more work; it is actually just an extra mouse click to ensure that the correct text is sent to the GP.—Gynaecologist

Nevertheless, the GP and practice nurses expected their workload to increase, as the RTA requires the invitation of an extra population. RTA training was provided to GPs and practice nurses in The Hague region, at gynaecology departments and at a regional conference for gynaecologists and obstetricians. This training was in the form of a detailed presentation about the RTA. One reported vulnerability was that people new to a department didn’t always know about the RTA. The interviewed healthcare providers reported that a project contact person was available or that they were the contact person themselves. They stated that this aided referral to the various specialisms and helped with existing questions.

Several barriers were reported regarding the invitation of the women. The healthcare providers indicated that improving the women’s knowledge concerning their elevated risk would be useful as it might increase the likelihood of them approaching their GP practice unprompted. However, the women reported that they preferred that the initiative for annual screening remained with the GP, as otherwise they might forget. Opinions varied among respondents regarding where responsibility should lie. One possible solution mentioned was inclusion of the women in the Dutch integrated care programme. This would mean that the women would be invited for annual screening in the same way as people with a known (and treated) risk for a cardiometabolic disease. It was also suggested that diversity should be handled by each GP individually, with the GP deciding how to best address the population in their neighbourhood.

Discussion

This study aimed to provide an overview of initial experiences with an integrated care and prevention programme for women who experienced a hypertensive disorder or diabetes during pregnancy. The healthcare providers and the women in the target population were positive about the programme. Healthcare providers reiterated that these women represent a forgotten high-risk population that deserves more attention and that the project provided clear guidance to all involved. The women themselves preferred continuation of the screening appointments. Several barriers were also reported. From the healthcare provider’s perspective, the most important barriers were the difficulty in ensuring notification to invite a woman for annual screening, the implementation of the RTA by new colleagues and solid evidence that screening prevents disease when it is accompanied by follow-up. The most important barrier from the women’s perspective was a lack of knowledge concerning their increased risk and the need for more personalised lifestyle guidance.

Comparison with other studies

Previous studies focused mainly on views concerning diabetes screening after GDM, with only a few studies devoted to studying increased risk after an HDP.²⁶ The most commonly mentioned barrier concerning these interventions was the lack of a clear policy or guideline for healthcare providers. These studies also recommended better coordination between primary and specialist care.^{30 31} Parallel to the present study’s findings, healthcare providers in an earlier study mentioned that an ideal solution would be to include follow-up of these women in the GP’s CVRM programme.²⁶ The same study recommended that a clear referral letter should be sent to the GP, and that the gynaecologist and obstetrician should have a signalling function. These recommendations were met in the present intervention, with a single integrated guideline for healthcare providers and good coordination between specialists.

The present study’s finding of a preference that the initiative for an annual screening invitation should remain with healthcare providers is in agreement with a systematic review and a study about women’s experiences with diabetes prevention intervention.^{24 25} A qualitative study including women with pre-eclampsia came to the same conclusion. The women preferred regular check-ups, especially when offered by a healthcare provider.²⁶ This shows the importance of healthcare provider initiatives in inviting women for screening. The importance of this issue was underscored by the finding that women often did not recall being told about their higher risk, even when healthcare professionals clearly stated that they were informed regarding this risk. Therefore, it is also important that multiple healthcare professionals in the chain around pregnancy and childbirth provide this information to women. To facilitate annual screening, a system ensuring annual invitation of this population is needed. An ideal solution would be to integrate follow-up with the Dutch CVRM programme, as stated above.²⁶ An earlier study including women with a history of GDM found that they would have liked more personalised advice regarding lifestyle changes, supporting our findings that women prefer personalised lifestyle advice.²⁴ This study in conjunction with the reported literature, therefore, recommends that healthcare providers share a more personalised advice with their patients and show more initiative in promoting a healthy lifestyle.

Limitations

The Heart for Women in The Hague project was launched in March 2021. As this starting date overlapped with the COVID-19 pandemic, few GPs had the opportunity to invite women for annual screening, restricting selection options. Due to COVID restrictions and the high pressure on care, little extra time was available for innovation. Therefore, most of the interviewed healthcare providers were already involved in the project and thus could be expected to be familiar with and enthusiastic about the project. This may have led to selection bias. To counter this, healthcare providers were allowed to include their co-worker’s reactions, as stated in the interview guide. To obtain a broader view in future studies, healthcare providers not involved in the study should be interviewed. Although evenly distributed between high and low socioeconomic status, of different ethnic backgrounds and with different pregnancy complications, the interviewed women were all recruited from a single GP practice in The Hague. The CFIR is commonly used to interview healthcare providers but is not generally used to interview both healthcare providers and patients.

Conclusion

Implementation of screening programmes after pregnancies with a history of HDP and/or GDM may help delay or prevent the development of CVD and/or T2D and might help to improve awareness in both patients and healthcare providers. Healthcare providers and women in the target population saw the annual screenings as important and necessary and would like it to be continued. A discrepancy was noticed between healthcare providers wishing to give the women autonomy and the women’s wish to receive an invitation to help remember screening. The clinical implication of this research is that it may encourage healthcare providers to recognise and act on signals at an earlier point in time.

It can be concluded that there is sufficient evidence to justify continuing this project but future success in the prevention of CVD and T2D will depend on additional research on the effectiveness of the chosen approach. In addition, it is strongly recommended that follow-up protocols should be integrated with the Dutch CVRM programme.

Future recommendations

Based on the findings, a recommendation is to include the annual screening of women who have had a high-risk pregnancy in the guidelines for GPs. However, an evaluation and more evidence supporting the (cost-)effectiveness of CVD and T2D screening in delaying or preventing disease development is needed. This is also a requirement of health insurance companies, because only those patients covered by the Dutch integrated care programme will be eligible for reimbursement of recurrent screening costs. In future projects, it is recommended to interview other stakeholders, such as health insurance companies and policy makers, to add to the discussion regarding implementation and reimbursement.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involved human participants. Ethical approval for this study was obtained from the METC Leiden The Hague and Delft (Reference number/ethics ID is G18.070). Informed written consent was obtained from the participants, which included their agreement to be interviewed, that the interview may be audio-recorded and that the researchers may use anonymous quotes in publication. Participants gave informed consent to participate in the study before taking part.