You can respond to this article at http://www.anaesthesiacorrespondence.com
Charting change on the labour ward
Version of Record online: 11 MAR 2013
Anaesthesia © 2013 The Association of Anaesthetists of Great Britain and Ireland
Volume 68, Issue 4, pages 338–342, April 2013
How to Cite
McGlennan, A. P. and Sherratt, K. (2013), Charting change on the labour ward. Anaesthesia, 68: 338–342. doi: 10.1111/anae.12189
- Issue online: 11 MAR 2013
- Version of Record online: 11 MAR 2013
In this issue of Anaesthesia, Carle et al.  have created a statistically based template that they have used to create an internally validated warning score for the obstetric setting. This marks the belated start to a process that may produce a scientifically derived universal obstetric early warning score (EWS). If successful, it will bring the obstetric patient up to speed with the general adult population.
I read the NEWS today
In 2007, the Royal College of Physicians (RCP) issued the report Acute Medical Care: the Right Person, in the Right Setting – First Time . This noted that a lack of standardisation resulted in a variation of methodology, approach and familiarity with EWS, leading to a lack of consistency in the response to acute illness. In reaction, the RCP commissioned the National Early Warning Score Development and Implementation Group (NEWSDIG) in 2009. This body met to review the available evidence, consider the existing models and devise an observation chart, scoring system and escalation pathway. This has now been completed and in 2012, the National Early Warning Score NEWS  was implemented.
There has been no consensus regarding the utility of EWS. Indeed, a 2009 Cochrane review  on outreach and EWS found only two reliable randomised controlled trials worldwide. One showed an improvement for in-hospital mortality but the other reported no statistical change at all. The review concluded that “the lack of evidence on outreach requires further multi-site RCTs to determine potential effectiveness”. Given that there is no overwhelming evidence for their efficacy, one would expect a degree of reluctance from the medical fraternity. However the uptake of EWS continued, perhaps as a sticking plaster for the reduction in nursing contact with patients. As a result, the use of EWS has been inconsistently adopted throughout developed healthcare systems. Consequently, there are also many different models of the scoring system, using many different observations and many different parameters. This makes it problematic to ascertain exactly which aspects of the scores are valuable. The fact that NEWSDIG has homogenised EWS charts to produce this system is credit-worthy in itself.
NEWSDIG reviewed all existing published literature and elected to use an aggregate-weighted system. These systems assign each observation with a score (usually 0_3). The sum of all scores is taken and the magnitude of this determines the strength of the intervention required. NEWSDIG devised its score by utilising a review paper by Smith et al. , that examined 33 different EWS systems already in use. In addition, NEWSDIG suggested that an extreme variation in a single parameter should also trigger a response and has thus incorporated this into its system. The variables chosen were common, easy-to-measure physiological observations that came from the National Institute for Health and Clinical Excellence (NICE) guideline Acutely Ill Patients in Hospital. Recognition of and Response to Acute Illness in Adults in Hospital . The weightings assigned were not steeped in science, rather mainly based on NEWSDIG's members’ clinical opinions and their views of existing EWS. The weightings led to a score that stratifies patients into low-, medium- or high-risk categories – this then leads to escalating levels of monitoring and medical review. Perhaps this shows a large amount of pragmatism – acknowledging that some evidence just does not exist. Alternatively, it can be taken to suggest that what was a scientific endeavour has been jeopardised by opinion. Nonetheless, it is difficult to see how the chart would have come into existence otherwise.
NEWSDIG rightly agonised that there is no agreed-upon validation of EWS. As its report states, what is the outcome measurement that is used? What is the gold standard to measure against? Instead, it has left ‘validation’ to others and to further research, to be achieved after the successful introduction of a national system. In its place, NEWSDIG attempted its own evaluation of the system. Again, this can be seen as either pragmatism or ducking a difficult question. Nevertheless, evaluation of the NEWS chart was performed by Professors Smith and Prytherch from NEWSDIG . They analysed patient data in differing clinical settings (both medical and surgical) over a 12-month period. The percentage trigger rate for different aggregate scores was reviewed and compared with a ‘typical’ EWS already in use. They felt that the NEWS aggregate score was a more sensitive trigger than most other EWS systems and thus a better system. However, a recent editorial in the British Medical Journal  stressed that rigorous validation of the trigger thresholds is vital to frontline NHS staff, to ensure that the system is effectively standardised and benefits the whole NHS.
One and one is two
NEWSDIG stresses that NEWS should not be used in obstetrics “because the physiological response to acute illness can be modified by pregnancy” . Other EWS also usually exclude obstetric patients, as their normal resting physiology is different from the non-pregnant state. This begs the question, what is the state of EWS in the pregnant population?
The 2003–2005 Report on Confidential Enquiries into Maternal Deaths in the UK stated that “There is an urgent need for the routine use of a national obstetric early warning chart, similar to those in use in other areas of clinical practice which can be used for all obstetric women, which will help in the more timely recognition, treatment and referral of women who have, or are developing, a critical illness” . Has this happened? And if so, with what chart? Anecdotal conversation indicates that most units are using an obstetric specific EWS but there seems to be no uniform choice of system. This may be because the report cited a single variable ‘track-and-trigger’ (essentially a system where one abnormal observation elicits escalation, as opposed to aggregate scoring) system developed at Stirling Royal Infirmary. No details of how the system and the parameters were devised were given. A nationwide survey conducted by the Obstetric Anaesthetists’ Association is underway to ascertain exactly what charts are being used across the country's maternity units.
At first glance, the uptake of obstetric-specific EWS is a simple and useful conception. A more cynical view is that maternity units have only produced EWS charts for two reasons: first, in response to its inclusion in the Clinical Negligence Scheme for Trusts grading requirements; and second, because of the parallel decrease in midwives’ ability to perform nursing duties. Since the introduction of these charts, one could argue that the maternity community has only paid lip service to the idea, with relatively little progress. There has been no nationally unified chart, no convention as to the model used, no agreement on the parameters, no body of evidence exhibiting their efficacy, and no decisions made regarding an escalation pathway. No one has taken the bull by the horns, and no organisation has shown real leadership. In short, this has been an opportunity spurned – effectively a ‘failure to progress’.
We have attempted to validate the use of the recommended Stirling EWS (the Modified Early Obstetric Warning System (MEOWS)) to attempt to show the utility of such a system. We prospectively studied two months of consecutive obstetric admissions . Our analysis showed a high sensitivity and specificity (89% and 79%, respectively) in detecting morbidity, but a relatively low positive predictive value of 39%, suggesting that further refinement of trigger thresholds would be beneficial. It was, however, only a small study of 676 patients and used a very liberal definition of morbidity. Apart from this study, literature searches for MEOWS or EWS in obstetrics mainly reveal audits and abstracts. This highlights a concerning lack of formal research into obstetric EWS. Recent abstracts presented to the OAA have mainly focused on implementation of charts rather than the actual parameters or trigger levels. Thus, EWS improve the recording of 4-hourly observations after a caesarean section , may aid early recognition of pre-eclampsia or sepsis , and may lead to a fall in the late detection rate of maternal illness . Further, although most healthcare professionals using the charts think it is a positive step , improvement is required to increase accurate recording of observations and to ensure that the correct action is taken once a patient triggers. This can be achieved by providing training and education to labour ward staff on the use of EWS charts.
Early Warning Scores may well have the potential to be life-saving within the obstetric population as much as in the non-obstetric group. We have a golden opportunity to improve the early detection of sick parturients and to reduce the consequent morbidity, but searches show a paucity of quality evidence in this area. Why is this? A recent editorial  reports a relative decrease in obstetric research over the years, whilst the number of case studies and audits submitted has increased dramatically. This may be because initiating trials in obstetrics is difficult, especially if there are thorny ethical issues. Also, trainees are often not in one hospital long enough to see a trial through to conclusion, which makes it less appealing to instigate. Added to this, despite initially seeming an easy concept, developing an obstetric EWS from scratch will involve accounting for a large number of variables, with relatively little evidence on which to base a scoring system.
It is refreshing to see that one group has attempted to buck this trend and to create and validate an obstetric EWS. Carle et al.'s study  is the only national randomised study attempting to evaluate the best variables to use for such a scoring system. The investigators have sensibly utilised data from all primary obstetric admissions from the Intensive Care National Audit and Research Centre (ICNARC) over a 13-year period in an attempt to create a validated, obstetric-specific, EWS. These data recorded the most extreme physiological measurement in the first day after admission to the intensive care unit (ICU). Significant variables were analysed with respect to mortality using a multiple logistic regression model, and then regression coefficients were used to weight each covariate. The impact of each variable was then assessed using the area under receiver operating characteristic curves. When analysing and choosing parameters for the model, Carle et al. recognised that as well as being statistically significant, a national system must also encompass the NICE minimum standards for monitoring, and recent recommendations from the Confidential Enquiries into Maternal Deaths . These data were then used to formulate a new, statistically-based, obstetric EWS.
Carle et al.'s proposed model has similarities with the NEWS model, in that it is an aggregate score from weighted variables, presented very simply on a single colour-coded chart to allow for ease of use. All of the variables are straightforward to measure and the aggregate score is then divided into three groups of low, medium and high risk. The trigger thresholds have been chosen to correspond with those used in the NEWS system. Maintaining similarities between the two charts should allow increased familiarity with the system, encouraging accurate data collection from ward staff.
There are limitations to the study, though. The dataset, while large, is from admission to ICU. Surely, this is after multiple medical interventions have already taken place; can these be reliable data on which to base an obstetric EWS? Given that the ability to collect pre-admission data nationally is not possible, Carle et al. suggest that the ICNARC data represent the best current option in the sick obstetric population. A further drawback to the obstetric EWS produced is over the inclusion of data. Whilst some of the variables have been included on a statistically validated basis (respiratory rate, systolic blood pressure), Carle et al.'s model also uses parameters that were not shown to be statistically relevant (diastolic blood pressure). As these variables are still valuable to the clinician, some have been weighted to be included in the final aggregate scores. Whilst this may be useful from a documentation perspective, it could be argued that the inclusion of such variables will skew results and alter the new model's sensitivity and specificity. Lastly, there is a large assumption when taking mortality data from patients admitted to ICU and applying them to a ward-based system to detect clinical deterioration.
We can work it out
Nationally we need to have a dialogue: just what do we want our warning scores to warn us of? Mortality? – too rare. The need for ICU admission? – possibly, but still quite rare at 260 per 100 000 maternities . Serious morbidity? – maybe, but the definitions are woolly and it will be harder to validate the system. That is why it is understandable that Carle et al. used data on mortality but produced a system to predict serious morbidity. Acknowledging this, Carle et al. have taken the approach that, despite its flaws, their new model will act as a starting point for development of a national obstetric EWS. Once in place, this can be further evaluated and refined. Whilst the new parameters or weightings may not be ideal, this is the largest and best study we currently have looking at observations from the sick obstetric population.
We suggest that the subspecialty is at a junction. There is a nascent movement focusing on maternal critical care . While the infrastructure is shifting and patient care is changing, we need to harness the opportunity to link obstetric EWS with it. By which we are saying that the outcome measure for the system should be the ability to spot the deteriorating patient. By focussing lower down the severity pathway we may well prevent serious morbidity, avoid admission to ICU and reduce obstetric death rates.
As with NEWS, the implementation of a single chart nationwide may be the biggest step in the prevention of maternal morbidity. As a secondary outcome, it will also make future data collection and potential refinement of the EWS a much more accessible option. Arguably, the best way to do this is also to learn from the NEWS implementation group – to create a multidisciplinary panel, incorporating anaesthetists, obstetricians, intensivists, nurses and midwives, to create a nationally agreed EWS chart for obstetrics. Like NEWSDIG, this group would be able to investigate existing systems (especially this work by Carle et al.), research which methods to use, and, through a Delphi process, achieve consensus on a nationally approved chart throughout the NHS. In our opinion, this is certainly overdue. To do this, however, we need leadership from the OAA, the Royal College of Anaesthetists and the Royal College of Obstetricians and Gynaecologists.
The resulting EWS may not be perfect and not everyone may agree on the inclusion or exclusion of certain variables, but as a priority we all need to be united in working together and having a nationally standardised tool to detect and prevent obstetric morbidity. If no further action is taken on the improvement of these scores, then the recommendation within the Confidential Reports all those years ago may effectively be stillborn.
No external funding and no competing interests declared.
- 2Royal College of Physicians. Acute Medical Care. The Right Person, in the Right Setting – First Time. Report of the Acute Medicine Task Force. London: RCP, 2007.
- 3Royal College of Physicians. National Early Warning Score (NEWS): Standardising the Assessment of Acute Illness Severity in the NHS. Report of a Working Party. London: RCP, 2012.
- 4Outreach and Early Warning Systems (EWS) for the prevention of Intensive Care admission and death of critically ill adult patients on general hospital wards. Cochrane Database of Systematic Reviews 2007; 3: CD005529., , , , , .
- 6National Institute for Health and Clinical Excellence. NICE Clinical Guideline 50: Acutely Ill Patients in Hospital. Recognition of and Response to Acute Illness in Adults in Hospital. London: NICE, 2007.
- 8The Confidential Enquiry into Maternal and Child Health (CEMACH). Saving Mothers’ Lives: Reviewing Maternal Deaths to Make Motherhood Safer – 2003–2005. The Seventh Report on Confidential Enquiries into Maternal Deaths in the United Kingdom. London: CEMACH, 2007., ed.
- 10Impact of modified early obstetric warning score systems on monitoring of basic physiological parameters on maternity wards. International Journal of Obstetric Anesthesia 2010; 19: S1–S54., .
- 11Severe obstetric morbidity and the value of modified early obstetric warning scores. International Journal of Obstetric Anesthesia 2008; 17: S1–S58., , , .
- 12The impact of MEOWS charts on clinical incident reporting. International Journal of Obstetric Anesthesia 2010; 19: S1–S54., .
- 13Obstetric early warning score charts: a midwife's view. International Journal of Obstetric Anesthesia 2009; 18: S1–S63., , .
- 15The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. Saving Mothers’ Lives: Reviewing maternal deaths to make motherhood safer: 2006–2008. British Journal of Obstetrics and Gynaecology. 2011; 118: 1–203.
- 16Intensive Care National Audit and Research Centre. Female Admissions (Aged 16-50 Years) to Adult, General Critical Care Units in England, Wales and Northern Ireland, Reported as “Currently Pregnant” or “Recently Pregnant”. London: ICNARC, 2009.