Designing for quality
Good practice in clinical governance requires that systems are continually being designed, monitored, and redesigned to ensure that QI is sustained. One approach to designing for safety is standardization of processes.
This is usually achieved through policies, protocols, and guidelines. Another approach is the technological approach; this aims to promote automation and eliminate or reduce the human element, thus promoting quality (in particular, the safety dimension of quality). Both approaches have their foundations in systems engineering. They bring about first-order change, but second-order change (which involves change in attitudes and behaviour) is essential for sustained QI in healthcare.Some examples of ‘designing for quality' are briefly discussed in the following sections.
Handover
Handover is the transfer of responsibility and/or accountability for patient care from one provider or team of providers to another. Along with the transfer of responsibility goes transfer of clinical information. A handover between shifts is an opportunity to assess the resources available (staff, beds, cots, etc.), to plan contingencies, and to maintain team situational awareness. Each team should design its own handover procedures, using the available good practice guidance.
Recognizing and responding to clinical deterioration in acute care
One of the major causes of mortality or serious morbidity in healthcare globally is the failure to recognize the deteriorating condition of a patient or, if recognized, failure to respond appropriately to clinical deterioration (sometimes referred to as ‘failure to rescue'). Failure to rescue is often due to loss of situational awareness.
To address the problem of recognition, scoring charts based on physiological parameters have been designed. A modified early obstetric warning score (MEOWS) chart was introduced for routine use in the care of all pregnant or postpartum women who become unwell.
The United Kingdom Confidential Enquiry report recommends that care providers should use the charted observations in order to ‘ensure confirmation of normality rather than presumption' (15). In other words, use of the chart should change behaviour such that clinicians constantly seek objective confirmation of normality, rather than presume normality and passively await any development that challenges this presumption. The report also states that ‘it is the response to the abnormal score that will affect outcome not simply its documentation'—another pointer to the importance of clinician behaviour in the management of risk.Prevention and management of sepsis
One aspect of clinical practice where an early warning score and/or other structured set of observations makes a big difference to clinical outcome is the prevention and management of sepsis. An initial evaluation showed that a scoring system designed specifically for an obstetric population can reliably identify women at high risk for admission to the intensive care unit (16).
Sepsis is a major cause of maternal death globally. In the United Kingdom, the National Institute for Health and Care Excellence (NICE) produced national guidelines on the management of sepsis in 2016, and these were updated in 2017 (17). The Royal College of Obstetricians and Gynaecologists has produced guidelines for the management of sepsis in pregnancy and a second edition is currently in development (18). National and local guidelines provide algorithms for the recognition and management of sepsis.
Critical to survival in cases of severe sepsis is the Sepsis Six bundle, which comprises three diagnostic and three therapeutic steps. This bundle—all of it—should be delivered within 1 hour of the initial diagnosis of sepsis. The components of this bundle are as follows:
• Deliver high-flow oxygen.
• Administer intravenous antibiotics.
• Commence intravenous fluids.
• Obtain blood cultures.
• Check serum lactate and perform a full blood count.
• Measure urine output.
The Sepsis Six could halve the number of deaths from sepsis but a national audit in the United Kingdom showed that its implementation was poor, with median implementation rates of 27-47% for individual components of the bundle (19). For implementation to be effective and sustainable, it has to draw from behavioural science (20). This provides another illustration of the value of a bionomic approach to quality in healthcare.
Safety in blood transfusion
Antenatal haemorrhage and postpartum haemorrhage are anticipated events in everyday obstetric practice. Obstetricians (and gynaecologists too) should be familiar with procedures and practices to minimize the risks associated with blood transfusion. These risks include infection, blood transfusion reactions, transfusion- associated circulatory overload, transfusion-related acute lung injury, and blood administration errors. Concern about these risks has led to the establishment of various mechanisms of haemovigilance at local, national, and international levels. Haemovigilance is the totality of surveillance procedures covering the whole transfusion chain from the donation of blood and its components to the followup of recipients. Information on adverse events is collected and used to improve the safety of blood transfusion.
The first step towards reducing complications of blood transfusion is to reduce the need for transfusion—for example, through optimization of antenatal and preoperative haemoglobin concentration and through the use of cell-salvage techniques. Decisions to transfuse should be made on the basis of the complete clinical picture and not just on the basis of the haemoglobin result. Transfusion should be avoided at night, unless there is a strong clinical indication for immediate transfusion. National and local protocols for massive haemorrhage and massive transfusion should be applied and a multidisciplinary team approach should be adopted (21-23).
The protocols should be supported by training and regular drills.Administration errors are not limited to red cell transfusion; the incidence of errors in administration of anti-D immunoglobulin is rising. Such errors include omission of anti-D and administration of anti-D immunoglobulin to a mother of a rhesus D-negative baby, to a rhesus D-positive mother, or to the wrong woman. These highlight the need for anti-D prescription and administration to be as stringent as those of blood transfusion.
Medical devices, including new technologies
Clinical governance leads should implement procedures for ensuring that staff are familiar with the medical devices that they are expected to use, and individual practitioners should ordinarily not use devices that they are unfamiliar with.
The implementation of new approaches to management of acute conditions (e.g. non-surgical, outpatient management of ectopic pregnancy) introduces new risks, and these have to be identified and contained. Procedures for the safe introduction of new technology should be implemented.
Clinical effectiveness
Clinicians have a professional responsibility to ensure that the treatment they provide is treatment that can achieve the optimal outcome for the patient. To discharge this responsibility they are obliged to follow recognized best practice, based on the best available evidence.
In the spheres of research and evidence-supported practice, it is customary to grade evidence according to levels. Various modifications, including subclassification, have been made to the original grading but all variants conform approximately to the following simple hierarchy:
Level 1: random allocation trials, more widely known as ‘randomized controlled trials', and systematic reviews of such trials.
Level 2: cohort studies and systematic reviews of cohort studies.
Level 3: case-control studies and systematic review of case-control studies.
Level 4:case-series.
Level 5: expert opinion.
NICE and the colleges of obstetrics and gynaecology in the United Kingdom, United States, Canada, and Australia produce evidencebased guidelines on a range of conditions and these are available on their websites. The guidelines will usually specify what level of evidence their recommendations are based on. The evidence from research is also used to produce local guidelines.
It is important to continually benchmark current practice against the guidelines in place. The mechanism for this is clinical audit (discussed in a later section).
Clinical effectiveness is not, however, just about plucking the evidence from the literature. The clinician has to determine how that evidence can be applied in the care of the particular woman he or she is about to treat, taking account of the woman's individual characteristics and preferences.