How often do we ask about pregnancy in the diabetes review clinic?

INTRODUCTION

The relationship between optimized management of glycaemic control in patients with diabetes who are pregnant or who are planning to become pregnant and better outcomes is well recognized. Reports such as the Confidential enquiry into maternal and child health (CEMACH) highlight how important it is to achieve and maintain good control of diabetes both in terms of maternal morbidity and mortality for the mother as well as for the fetus. ¹

In an era of increasing targets and performance management diabetologists, diabetes specialist nurses, obstetric physicians, obstetricians and midwives should all be involved and aware of the potential harm that poorly managed diabetes can cause and have effective systems in place to ensure that such patients have regular, frequent, joined-up review throughout pregnancy, as well as good preconception management. ²

The focus of this article forms part of the ‘prevention is better than cure’ argument, in that women of child-bearing age who are already known to diabetes services, whether in hospital or in the community, should be regularly reviewed and/or screened with regard to their plans to become pregnant. This allows two major benefits. Firstly the optimization of diabetic control in those who are actively planning to become pregnant and secondly the consideration and recommendation of counselling and the use of contraception in fertile women with very poorly controlled diabetes in whom an unplanned or unwanted pregnancy would be significant in terms of risk to the mother and the unborn child. ³

While this is increasingly the case in many diabetes services, it is not a standardized part of practice and can lead to health discrepancies especially in areas of unmet need. What is required is a robust, organized approach to ensure that women with diabetes are adequately monitored and managed to avoid the multitude of complications that can arise as a consequence of poorly controlled diabetes in pregnancy.

This study follows on from a previous piece of work which looked at variations in practice with regard to screening for erectile dysfunction (ED) in men attending the diabetes review clinic. ⁴ This is a yearly National Institute for Health and Clinical Excellence requirement but our research has shown that the number of patients who were actually asked about neuropathy and ED symptoms was very low indeed. It became apparent that there was a large discrepancy between the espoused ideals of diabetes review clinic practice and the concrete reality of what we were actually being able to check on a regular basis. This is an important issue, especially when it comes to maternal medicine and the knowledge that we can actively intervene to make diabetes control and hence outcomes better.

This may feel like another ‘tick-box exercise’ in the increasingly long list of things that we need to measure and monitor in diabetes care but as Atul Gawande asks, would you like to fly on an aeroplane in which the pilot hadn't bothered to check the destination or that all the engines were working? ⁵

The corollary is that setting out on a pregnancy is equally dangerous if the important and relevantly modifiable risk factors are not adequately addressed in advance.

The goal of this project was therefore to analyse diabetes review clinic performance around one key area, a very basic one: how often are we asking about pregnancy plans in the diabetes review clinic? If we wish to reduce the rates of complications in diabetic pregnancy this is arguably where we should begin. Are we enquiring about and documenting the intentions of young women in the fertile age range on a regular basis about family planning? This is a key question and ideally the answer should be that we are doing it 100% of the time.

METHOD

We analysed the electronic patient records of a target population of young women of child-bearing age (16–45) with both type 1 and type 2 diabetes who were attending the diabetes review clinic. Data were analysed from a three-year period (2007–2010). The Diabeta 3 computer program (Diabeta 3 (VECTOR) Health information systems (UK) Ltd, London, UK) which is commonly used in clinical practice has a ‘query builder’ function which allows a comprehensive analysis of all the recorded variables entered when a patient attends for a diabetes consultation with members of the diabetes multidisciplinary team. ⁶ Along with the standard measurements required on each diabetes outpatient attendance such as HbA1c, retinal screening check, renal function, etc. there are also a ‘pregnancy planning’, ‘pregnancy status’, ‘preconception counselling’ and ‘contraception education’ options to select as part of the larger diabetes care assessment.

By using these search terms and mining the data, we were able to simply and comprehensively explore how many women with diabetes had been seen over a given time period and how many had had it recorded on their electronic documentation that they had been asked about their pregnancy status. Faults with this approach are that the pregnancy issue may actually be raised in a consultation but not recorded for whatever reason, but this is important too, especially when one considers the more modern multidisciplinary approach to diabetes care and that patients may not always see the same health-care professional. It may also be the case that patients with diabetes are seen in the community either by diabetologists, general practitioners or practice nurses – who have separate notes and no access to the shared Diabeta 3 records – but given that all the patients in this study were under the review of a secondary care diabetes centre, then one could assume that this would be their main point of diabetes-oriented care.

RESULTS

There were 1808 female patients with diabetes mellitus who were under the care of the diabetes centre and within the fertile age range of 16–45 years. They attended for an average of 2.6 visits over the study period. With regard to the recorded data on Diabeta 3, this revealed that 88 patients were documented as being actively pregnant, although analysis of the expected date of delivery (EDD) revealed that this would only have been true in 83% of cases. Forty-four patients had pregnancy planning (discussed on the latest visit) and 54 had not had it discussed (and recorded) at all. Nineteen patients had contraception education discussed and 22 patients were documented as being offered preconception counselling (see Table 1).

DISCUSSION

These results are interesting and suggest several relevant points. Firstly it would appear that the pregnancy prevalence in this study population is lower than that which one might expect. It is clear that the active pregnancy status is not being updated. Of the population as a whole, only a very small percentage seem to be actively considering or trying to become pregnant and it would be useful to see the status of their glycaemic control as of this subgroup only 22/44 (50%) subsequently went ahead and had preconception counselling.

A concern here is of the relatively small number of patients overall who were taking up preconception or contraceptive education. It is often the poorly controlled diabetic patients with multiple risk factors who have the worst pregnancy outcomes so it appears disparate that there is no clear system of identifying these patients and trying our best to optimize their care. A recent article highlighted the fact that patients with a large body mass index often tend to be the ones not taking contraception and these are the patients who run in to more complications during pregnancy. ⁷

It is apparent in the diabetes review clinic setting that we are not identifying, enquiring and/or documenting pregnancy-related intentions in women of child-bearing age. This means that we are not able to provide adequate preconception care, contraception or monitoring of care, which is one of the mainstays of preventive obstetric medicine. Is it simply that we are not asking the right questions often enough?

For computer database programs such as Diabeta 3, which are locally adaptable, there are advantages and disadvantages of using forced criteria check boxes. We are all used to providing the ‘essential’ criteria on webpages, which will not let us proceed until a box is completed, but how appropriate is this for clinical care where such forced questions may not always be relevant? The lessons from CEMACH do demonstrate however that preparing the patient with diabetes for pregnancy is a crucial part of the diabetes service.

For more traditional, non-computer-based diabetes clinic visits, then this type of performance monitoring and measurement becomes more difficult. How do we ensure that we are checking for all the important variables – not just pregnancy plans? Several clinics tend to use paper-based forms that contain all the relevant criteria from visual acuity, to smoking status, to gum health. Other diabetes review clinic assessments may just start with a blank piece of paper and rely on the clinician's ability to remember to enquire about pregnancy. A previous article entitled ‘The perfect diabetes review’ ⁸ highlighted this problem and suggested that it is possible to develop a comprehensive, exhaustive list of all the things that need to be addressed at a diabetes review but admitted that it is often unwieldy and impractical in a busy clinical setting.

CONCLUSIONS

It is clear that in this study population, pregnancy plans and planning were not documented effectively and this means that when young women with diabetes do eventually become pregnant they are at more risk of complications as they will have missed out on the opportunity for preconception counselling.

The goal of care should be for robust clinical assessments to be in place, which mean that pregnancy intentions are automatically recorded and subsequently acted on as appropriate. One simple approach could be the use of preconsultation screening questionnaires which patients can complete in the waiting room and then hand in to the diabetes health-care professional so that they have the information complete and ready to hand. This would be useful for not only pregnancy but also neuropathy screening which as mentioned is something else that we are poor at recording and acting upon. ^4,8 The other goal of care is that female patients should be aware of the importance of good preconception glycaemic control a priori and of what the diabetes service can provide to aid this. We should be aiming for patients to approach us with their pregnancy intentions so that we can work together to achieve better pregnancy outcomes, rather than the current lacklustre situation in which we fail to ask either through poor memory or poor systems.

DECLARATIONS

The authors have no conflicts of interest to declare.