Implementation and evaluation of a blood products traceability procedure in a District General Hospital

Aim: The study evaluated poor traceability compliance of blood products within a District General Hospital (DGH) and explored why the wards traceability compliance varied across the DGH. The study also explored risk factors that lead to poor traceability of blood products within the DGH and implemented a suitable model to improve traceability compliance. Method: A quantitative approach was adopted. Data were collected using questionnaires, observations, and audit where data were extracted from the laboratory information management systems. A questionnaire was designed, piloted, and sent to all ward managers within the DGH, a descriptive statistic of the survey data was analyzed using statistical package SPSS (version 19, SPSS Inc., Chicago, IL, USA). Results: The wards with high transfusion episodes were more compliant than wards with fewer transfusion episodes in this study. Moreover, the low usage group had a significant lower compliance rate than the medium and high usage groups (p < 0.001), also the study provided an insight into the variety of services delivered to the end-users of transfusion services. It highlighted a lack of training tools; poor procedures for the return of labels and the challenges faced with the portering service. Poor communication between the laboratory and end-users was identified as another issue.

iii The study resulted in the implementation of new approaches to the transfusion chain in order to improve practice. These included: the appointment of a medical laboratory assistant to assist with traceability compliance; the appointment of a clinical transfusion nurse specialist to assist with training and ensure safe practice on wards; the provision of trainers and clinical supervision on particular wards; the development of a competence programme to assess staff involved in blood collection and distribution; the development of clinical guidelines; and the administration of an annual skills checks for staff involved in the transfusion chain.
Most of the recommendations have been implemented and put into practice. A formal audit will be conducted in the future to evaluate their success but to date, 65% of staff members have passed the annual skills" assessment and the traceability compliance of the Trust has remained at 100%. iv

Introduction
Blood is one of the world"s vital substances. The Greeks referred to blood as a component of universal order. The Romans felt blood carried a person"s vital essence; the gladiators drank the blood of their fallen opponents. Doctors, from mediaeval to Victorian times assumed that blood had a fantastical power; draining blood removed evil humour, transfusing blood pacified the deranged (Starr, 2001, pp.119-120).
However, blood is fragile and expensive and has been a heavily traded product across the world. The global trade in blood and its products developed in the early 1970s. Initially, red cells and white cells stayed within national boundaries except for Swiss blood that was exported to Greece and the United States. Eventually, plasma derivatives became international commodities when the United States became the pioneer of plasmapheresis. A darker side of the industry emerged as plasma mills appeared in deprived areas of American cities with donation sessions from drug addicts and prisoners. Although Europe denounced this practice it continued to quietly purchase plasma and its derivatives from America. Other plasma mills, located in South and Central America were destroyed in 1978 (Farrugia, 2006).
The same, effective, system that was responsible for the collection of blood products was equally effective in distributing diseases. Transfusion Transmitted Infection became a significant and daily subject of alarm in western countries. Both clinicians and politicians were blamed for delivering disease to the population rather than preventing it (Feldman & Bayer, 1999, p.268).
Haemophilia patients became reliant on fractionated plasma in the early 1980s.
Because they did not need to go to a hospital for a plasma transfusion, some could keep fractionated products at home. Easy access to fraction blood products gave them a normal life (Feldman and Bayer, 1999, p. 269-270). Consequently, the stock of residual plasma products could not meet worldwide demand. However, more pooled plasma was needed worldwide than could be supplied from fresh blood supplies.
Blood and its products (derived from whole human blood or plasma) saves millions of lives every year but the entire process of collecting and distributing blood and its products depends on the trust and the goodwill of the public. Although some adverse events related to blood transfusions may occur immediately, transfusion transmitted diseases may not produce any illness for months, or even years. The ability to trace the path of the blood from the recipient to the original donor and vice versa is an important public health safeguard. In addition, accurate and complete record keeping is an essential part of professional practice.

Globalisation and blood safety
Globalisation may be defined as the, "transition from national and regional economies to global economies. This includes a nexus of an economic and social process where local market and culture are dominated by global market and cultures" . It is a consequence of the growing interdependence of countries worldwide through an increasing volume of cross-border trading in goods and services.
Data gathered by World Health Organization (World Health Organization, 2006) showed that the blood donation rates and the extent of viral testing are dependent on economic status (Mattar, 2004). In an ideal and globalised environment, one would expect that the challenges faced by the world in terms of the safety and supply of blood and its products would be addressed through the free movement of blood derivatives. However, the developed world"s capacity to deal with established and emerging blood safety threats, many of which are the results of globalisation is lacking (Snyder & Dodd, 2001, p.433). Malaria is the most common transfusiontransmitted infection worldwide and in non-malarious area is introduced into the blood supply via travellers returning from areas where it is endemic. In the United Kingdom deferral measures applied to donors returning from such countries are considered to provide adequate protection (Snyder & Dodd, 2001).
Unfortunately the continuing emergence of infectious agents is a feature of globalisation. It affects the safety of the blood supply and makes the cross-border movement of blood more difficult than in previous times. Nevertheless, the plasma industry, where blood products are traded across national borders is one example of the continuing global nature of the business. The reaction of humans to the pressures of globalisation is responsible for many of the current challenges posed by trade-related infections (Kimball, Arima & Hodges, 2005, p.2).
Higher levels of traffic in goods and people have affected both blood safety and supply in many ways. The increasing movement of plasma products has eroded some of the historical barriers between countries. Developed countries are strengthening the blood safety infrastructure through the development of new tests and pathogen elimination techniques (World Health Organisation, 2006). Many of these measures are the result of trends initiated by pharmaceutical manufacturers and in the healthcare sector although in general such processes have been restricted to, and focused on, rich countries (Kreil et al., 2003(Kreil et al., , pp.1023(Kreil et al., -1026 Italy and Portugal and one each in Japan Saudi-Arabia and Taiwan. vCJD is a new disease and the risk of transmission has been a major concern to blood services since its discovery. The precautionary measures taken since 1999 include the removal of white cells believed to contain much of the infectious agent-prions (McClelland et al., 1996).
The first incident of the contamination of blood by Hepatitis B was seen in 1982 in patients with haemophilia who had received blood fractions as therapy for their clotting disorder. These products were produced from the pooled plasma from a large number of donors, which was normal procedure at this time. Until then, the potential contamination of blood products with infectious diseases had not been seen as a risk. Since then, the response to Acquired Immune Deficiency Syndrome (AIDS) that affected government policies and the reputation of the medical profession has been called the blood scandal (Feldman,1999).
Canada, France, Japan, and the United States all contributed to the AIDS scare. All four countries were self-sufficient in whole blood and blood components from voluntary donors although France relied on prisoners as volunteers and Canada and Japan imported blood fractions for haemophiliacs. Haemophiliac patients organized themselves differently in different countries and this affected the pressure that they were able to exert on the political structure. They gained most recognition in Japan in the 1980s and received substantial compensation. In Canada and France they accepted compensation schemes offered by the government (Baldwin, 2005).
The French blood system was deeply involved in the scandal. Perhaps due to pride in the purity of French blood, the downfall of the system was its involvement in more than half of the blood-transmitted AIDS cases in Europe. In 1991, an internal government document revealed that the early success of heat treatment of factor VIII was ignored. The distribution of older, unheated, contaminated fractions continued from some blood centres in France and was justified by the cost. In 1992, the Director of the National Centre in Paris was sentenced and jailed, an event that is now widely believed to be a miscarriage of justice (Schmidt, 2006).
In Japan, as in France, the driving force behind the AIDS blood scandal was the effect of infected blood fractions on the nation's haemophiliac population. At this time commercial operators took control of plasmapheresis and Japan's corporations guaranteed plasma collections. It was the product of this unholy alliance that brought HIV to Japan. The problem set the stage for the most prominent health scandal in Japan"s history. Further evidence of failure was found in the continued use of unheated plasma long after the heat treatment of plasma was mandatory in the United States. The principals involved were charged with murder (Schmidt,2006). The Canadian blood system combined the problems seen in France of a loosely controlled and overly-trusted source of whole blood, and contaminated plasma from American blood. The government had left the regulation of the blood system to the Canadian Red Cross and the therapy for haemophiliacs was being met by plasma processed in the United States. The public became aware of the disaster as the untreated plasma spread among the haemophiliac population in Canada. The Red Cross"s response was slow and when the government finally decided to fund blood testing, the decision came seven months later than in the United States. The Canadian Red Cross later filed for bankruptcy and removed itself from all activities associated with the blood system. A new National Canadian Blood Service was created to do the work of the Red Cross. Consequently, Canada, France and Japan"s blood system have been constructed to protect the public from the previous mistakes (Gallo & Montagnier, 2003, pp. 2283-2284.
The United States carried out a major analysis of its blood programme in 1972 following the hepatitis problem. Although the United States was the country in which the new disease (that turned out to be HIV) was first seen in 1990s.The blood scandal resulted in lesser political and social upheaval than in Canada, France and Japan. The peculiarities of United States liability laws required patients to file for compensation in their local jurisdiction (Baldwin, 2005). Blood centres in the United States were protected by blood shield laws passed during the hepatitis era which absolved them of any responsibility if they had carried out the required testing. Although court rulings always found against the patients, many cases never reached the courts. It was not until 1995 that infected haemophiliacs were compensated by the federal government.
Studies by Glynn et al. (2003) and Ling (2010) claimed that blood is safer than ever. However, safety culture was not enforced during the hepatitis era. The Department of Health and Human Services in the United States have implemented a comprehensive safety vigilance system to address the threat from unknown and emerging infectious sources (Busch et al., 1999). Blood is not entirely safe, but neither is it our most dangerous drug. Like many good things, it comes with risks. The residual risk of viral infection can be estimated from data measuring the number and percentage of seropositive donations in previous years, the prevalence of seropositive donors, the length of the pre-seroconversion window period, the sensitivity of the screening test and the probability of blood components being contaminated. These models have made it possible to identify a number of preventative measures. The evaluation of blood donation safety can also be based on the follow-up of recipient cohorts by assessing residual risks in a given population, such as in multi-transfused patients (Debeir et al., 1999, p.81).

Haemovigilance
Haemovigilance is defined as, "a system of surveillance and alarm, ranging from the collection of blood products to the follow up of recipients, to gather and assess incidents resulting from the transfusion of the blood products". The main goal of haemovigilance is to prevent the recurrence of incidents by identifying their causes (McClelland et al., 1998). Haemovigilance contributes to the process of maintaining and improving transfusion safety. It grew out of government-driven reviews of the organization of transfusions following the HIV transmission scare in the early 1980s (Faber, 2002). The mandatory reporting at a national level of any undesirable events related to blood transfusion was introduced in the United States in 1975.
The concept was first implemented in Europe in 1988 by Belgium, Denmark, France, Luxembourg, the Netherlands, Portugal, Spain and Switzerland, with a simple objective of being able to share alerts (Herve, 2002, p.30). In 1995, the member states of the European Union indicated a need for countries to establish a haemovigilance system. This culminated in the European Blood Directive (Faber, 2004  In Europe, the French and British systems are different. The French haemovigilance system was established in 1994 (Andreu, Morel & Forestier, 2002), it is nationwide and there is a legal obligation to report any untoward effects related to blood transfusion (Andreu, Morel & Forestier, 2002). It must meet three objectives at local and national level: (1) identify risks and related factors and monitor these risks; (2) assess the relevance of medical indicators present in individuals that wish to donate blood and (3) at the transfusion level, estimate the incidence of transfusion-induced side-effects in recipients (de Vries et al., 2011).
In the UK, the Serious Hazards of Blood Transfusion (SHOT) was launched in 1996. Unlike France, only serious adverse reactions must be reported to SHOT on a voluntary basis (Williamson, 2002). The participation of UK hospitals in haemovigilance has improved since 2009 with an 85% increase in the number of reports submitted in 2010. At the same time, the number of non-reporting hospitals and trusts has fallen.

Blood transfusion in the UK
Over a 15 year period (1996-2011) more than 36 million blood components were provided by the United Kingdom Blood Services and 4,334 untoward incidents were analysed. Table 1-2 is a summary of all blood products issued 1999-2011 by regional UK transfusion centres. It shows a downward trend in blood products issued due to the postponement of blood donations during the influenza season and measures taken to exclude potential donors to reduce the incidence of vCJD.  Products Regulatory Agencythe regulatory body in the UK. However, many UK hospitals still do not submit reports despite legal requirements. Northern Ireland and Wales reported more incidence than Scotland and England which suggest that Northern Ireland and Wales have a high incidence detection level of reporting. This is a major concern from both professional and regulatory viewpoints (Serious Hazards of Transfusion, 2009).

12
The knowledge of ABO blood group system in the 18 th and 19 th century was minimal which led to a lot disasters. According to (Starr, 2001). Further blood transfusion procedures led to haemolytic episodes and deaths in the early part of the 18 th century and this eventually resulted in a ban for over a century. The discovery of ABO blood group systems in the early part of the 20 th century minimised haemolytic episodes but transfusion transmitted infections became a major concern due to global trade of blood products and inadequate screening .
Blood transfusion in the 20 th century had informed many of the decisions, initiatives and guidelines in the 21 st century and these are discussed fully in the following sections 1.5. Following the increased of transfusion transmitted hepatitis, the term haemovigilance was coined in the United States in the early 1970s to address the threat to patient safety and tighten regulations on the use of blood products. This led to the introduction of a robust quality management system for blood transfusion in order to protect the patient and the public. The risk of mistransfusion became greater in the 21 st century, see Page 34 "The risk of mistransfusion was found to be greater than the risk of transmission of HIV or HCV (Dzik, 2002(Dzik, , p.1190." SHOT, (2011,p.16) also emphasized of the incidence of incorrect blood component transfused in the U.K. Mistransfusion can also be due to poor control or knowledge of the collection, storage, distribution and traceability of blood products.
My research design was to explore and identify why traceability compliance was poor within the Trust and to implement a suitable model to improve traceability compliance within the Homerton University Hospital NHS Trust and to enhance patient"s safety within the transfusion chain.  (2005), also required the full traceability of blood and blood components. Blood establishments and hospitals are required to implement a system that permits the identification of each unit of blood component and its final destination. A system that has proved effective in the UK is the so-called "bag & tag" label system. When a unit of blood is prepared for a patient, a computerised laboratory system prints a tag. This includes information that identifies the patient and two traceability labels bearing the donation number.

The regulatory framework
It is important to understand the reasons behind the urgency of this current move to the traceability of blood and its products. The emergence of transfusion transmitted diseases has damaged trust in the blood establishment around the world and a survey of various haemovigilance systems indicated that errors in the blood transfusion chain, from the initial recipient identification to final blood administration occur in about 1 in 1000 events (Pagliaro & Rebulla, 2006). Specific tools have been developed to prevent the potentially fatal consequences of such mistakes.
These include barcoded patient identification bracelets and mechanical and electronic locks that can allow full audit trail of all transfused blood products.
Although a number of studies have demonstrated the effectiveness of these systems, they have not been widely adopted because of financial constraints.

Organisational and professional drivers of BSQR
A patient is entitled to be cared for by healthcare professionals with relevant and up-to-date skills and expertise (Kennedy, 2001, p.322). For the general public, this means that they can be also involved in the planning, organisation and delivery of healthcare.
1.6.1. Serious Hazards of Transfusion (SHOT) Serious Hazards of Transfusion is an independent, professionally-led haemovigilance scheme that was launched in 1996 and collects data on serious sequelae caused by the transfusion of blood components. The data collected contributes to improvements in the safety of the blood transfusion process, informs policy within the transfusion services in the UK and helps in the production of clinical guidelines for the use of blood components.

Clinical governance
Clinical governance consists of the processes healthcare organizations use to monitor and improve the quality of clinical services delivered within the healthcare system. It was introduced in 1998 as part of the UK government"s ten year plan to improve the overall standard of clinical care (Department of Health, 1998). The introduction of clinical governance was aimed at improving the quality of clinical care at all levels of provision. It was seen as a way of addressing concerns about the quality of health care following well-documented variations in standards.
It necessitates a more widespread adoption of the principles and methods of continuous quality improvement initially developed in the industrial sector and later applied to the healthcare sector (Berwick, 1989

Better Blood Transfusion
The Better Blood Transfusion circular (Department of Health, 1998a) was sent to all NHS Trusts in the UK that undertook blood transfusions. It asked them to participate in the SHOT scheme and fulfil MHRA requirements for the reporting of Serious Adverse Blood Reaction and Events (SABRE). It also emphasized the need for training for laboratory staff and giving them the power to review requests for blood transfusions in order to reduce the inappropriate use of blood. high standards of quality and safety in blood transfusion taking into account the fact that the organization and regulation of blood transfusion is very different in different European countries. The Directive aimed to harmonise these differences in order to improve the level of quality and safety in blood transfusion across the European Community (Faber, 2004).

19
The

UK Blood Safety and Quality Regulations
The Blood Safety and Quality Regulations 2005 require the full traceability of blood and blood components from the point of receipt of the blood or blood components by the hospital. The confirmed final destination of blood components received by the site must be retained in a readily-available format, recoverable for thirty years.
The Medicine and Health Products Regulatory Agency (MHRA) was designated as the competent authority in the UK under the legislation. The MHRA is a government agency whose role is to enhance and safeguard the public"s health by ensuring that medicine and medical devices meets acceptable standards of safety and they work. The MHRA aims to safeguard public health by:  Ensuring, through regulation, that medicine and devices have an acceptable balance of risks and benefits.
 Helping people to understand the benefits and risks of medicines and devices.
 Encouraging and helping the development of medicines and devices that will contribute to health.

20
The regulations required that the competent authority ensure compliance with the relevant regulations. To achieve this, the MHRA developed a compliance report which all hospitals with blood banks must complete annually. Hospital blood banks must provide details of their accreditation status, staffing levels, traceability and quality management system within the department. The report is reviewed and an on-site visit takes place if non-compliances are identified. The MHRA has also established a system of reporting and recording transfusion-related incidents known as Serious Adverse Blood Reactions and Events (SABRE).

Clinical audits and professional standards
The power of audits in the quality improvement process is well recognised, which is

Transfusion safety
Although the potential untoward effects of blood transfusion have been recognized almost since the practice began, no systematic data was collected for decades despite the publication of several reports (Herve et al., 2000). The major steps in the clinical transfusion process involve obtaining a patient sample, labelling it and submitting it to the transfusion laboratory for cross-matching (Figure 1.3). The intention is, "getting the right blood, to the right person, at the right place, at the Hospitals must be able to show that their blood transfusion practices are safe, clinically effective and efficient. This requires a paradigm shift in favour of a strongly founded belief that quality, safety and effectiveness must be built into the blood transfusion process, total process control and procedures that prevent errors.

Recipient recall programmes
Recipient recall programmes refer to the practice of notifying a large group of recipients who have been exposed to an infectious agent at the time of the transfusion (Busch, 1991, p.655) and who have been identified as infected with a specific disease. It requires hospital blood transfusion services to trace transfusion records in order to identify the individuals that received the specific component.
The recipient is usually notified of their potential risk by their clinician. Further tests are carried out after counselling. Current regulations require the recall of donors who have been identified as infected with a blood-borne infection. The requirements specifying which recipients must be informed are complex because of the evolution of Hepatitis C screening and confirmatory tests (Goldman et al., 1996).

Methods to identify blood transfusion incidents
Human error is routinely blamed for accidents in healthcare. Humans tend to rush to judgement following an incident and, all too often, blame the person most associated with the accident. This prevents the discovery of the whole story (Cook, Wood & Miller, 1998). Table 1-8 shows the results of a study that examined a series of events and departures from safe practice which were influenced by the working environment and the wider organisational content (Vincent, Stanhope, & Crowley-Murphy, 1999). A number of methods have been identified to reduce the contribution of human errors to the risk of blood transfusion. Table 1-9 shows some common methods used for error identification and these are reviewed in the following paragraphs. Subject to underreporting because individual do not feel comfortable.

Source: Kaplan et al., 1998
One approach is direct observation by skilled observers in the actual operating environment. Schulman (2004, p.652) developed a multidisciplinary approach to quality assurance and improvement directed at reducing patient identification error.
The authors found in-service education to be effective in increasing compliance and reducing the risk of error. However, the observation itself may have altered the circumstances studied.
A second approach is accident analysis. This approach has been extensively used in aviation by the National Transportation Safety Board (Nagel, 1988). However, hindsight bias and incomplete data can lead to distortion of the facts. Despite this limitation, accident analysis has also been an important source of information in blood transfusion incidents. For example, Mummert and Tourault (1993) reviewed 150 transfusion-associated fatalities from 1990-1992. They concluded that nearly a third of fatalities could have been prevented by full adherence to standard operating procedures. Interestingly, a failure to follow procedures is also responsible for a third of major air carrier accidents Nagel (1988) who also stressed the need for models of human error to be used in conjunction with error data collection and classification. The management of error to limit adverse outcomes or unplanned effects is now recognised to be of fundamental importance in system design and training (Reason, 1990). Although the importance of this has been recognised in transfusion medicine it has not been fully implemented in all critical areas of operation.
A third approach is the study of error through laboratory simulation. This approach is useful as it enables the simplification and control of variables; although simplification may itself be an important shortcoming in understanding inherently complex situations (Nagel, 1988). Taswell et al. (1974, p. 491) modified the conditions of work by introducing known errors and provided positive feedback when they were found. The blood transfusion department involved in the study not only improved the detection rate of introduced errors, but also increased the detection of real, previously undetected errors, from four in the first three months of the study to 73 in the final three months.
A fourth approach to error identification is the review of records. This has been a traditional way to perform quality assurance checks and document patient outcomes. Patient records document actions performed and can highlight missing information. The audit of records against predetermined criteria can be a valuable way to identify errors or near misses. Classen et al. (1991, p.301) successfully used a sophisticated automated hospital information and record system to identify adverse drug events that would otherwise not have been reported.
A fifth approach is the event report, including self-reporting. In aviation, despite excellent record keeping, human error is identified as a casual factor in more than half of all airline accidents. More than 250,000 reports have been archived, analysed and made available for research and study by interested professionals and regulatory bodies (Nagel, 1988). This no-fault, confidential, voluntary, selfreporting system enables pilots and controllers to report all non-calamitous mistakes, including caught errors. Confidentiality and immunity from prosecution for non-criminal acts are other important features of this or any system intended to capture operational errors. The no-fault, confidential nature of the system has led to an increased frankness in reporting and the provision of invaluable data (Classen et al., 1991).

Haemovigilance
Haemovigilance provides information that is useful in the analysis of the various causes of the untoward effects of blood transfusion. It provides reminders of known protective measures and helps to identify new ones. Haemovigilance has also helped to identify major technical, organizational and human errors. Table 1-10 Summary of the types of clinical errorsTable 1-10 is a summary of the various types of clinical errors. In the first 18 months following the implementation of haemovigilance in France, seven fatalities related to the bacterial contamination of blood components were reported, while three fatalities related to ABO incompatibility were reported over the same period, leading to increased awareness of bacterial contamination as a major complication of blood transfusion (Engelfriet & Reesink, 2000, pp.59-62).

Benign and caught errors
The study of benign or caught errors in transfusion medicine could provide a rich database for improving the safety of the blood supply (Gambino & Mallon, 1991, p.14). The relationship between near misses and accidents has been compared to a pyramid or iceberg. Zapt and Reason (1994, p.427) indicated that error detection is the first step of error management. If an error is not detected, it cannot be managed. Errors that are not detected for a long time can have disastrous consequences.
The goal of error management should be to increase error detection and the error response rate. High reporting rates indicate a high detection sensitivity level (DSL).
Few reported events indicate a low DSL. A low DSL can be seen as an indicator of inadequate error detection and reporting. When the DSL is high, the severity level of the incidents should decrease over time as corrective actions are implemented (Kaplan, 2005(Kaplan, , pp.1071(Kaplan, -1081. In order to achieve a high DSL, an organisation must remove any impediments to reporting events. Confidential no-fault reporting is one of the best ways to encourage event reporting ).

Technology
The incorrect use or deficiencies in the technology used in blood transfusions continues to cause incidents (Tables 1.11 and 1.12). Despite barriers created to prevented mistransfusion, few technological errors have been reported by SHOT. Common errors were due to software failure and hardware failure during blood product issuing. This reiterates the importance of human in the transfusion process.  Reason (1990) identified two major categories of failures or error that occur in Most health professional operate in a skill-based behavioural mode for routine tasks, from blood collection to the transfusion process . These highly skilled activities become routine and can be executed without any thought.
Slips are unintended errors caused by either not doing part of the routine or being distracted by someone. Slips can be prevented by the redesign of equipment and procedures in ways that make it harder to make a slip. For example, feedback mechanisms can be designed into the process that alerts individuals.
Rule-based behaviour involves recognising a situation and selecting the proper routine or protocol. Mistakes happen under two conditions, either by selecting the wrong rule for a given situation, or selecting the correct rule but carrying it out incorrectly. Such failures can occur when someone carries out a procedure that they are not trained or qualified to perform. Another example is inadequately assessing the situation, for example failing to check the patient"s identity before a transfusion . Routine violations often occur when procedures are changed and individuals continue to apply the old procedures. In some cases, rule-based failures can be prevented by redesigning the task to force a procedure to occur. An example would be a blood lock that prevents a transfusion starting until the patient"s identity has been matched to the blood product.
Knowledge-based behaviour involves solving unique problems or selecting a plan of action in a new or unfamiliar setting. It often occurs with new employees who do not have the knowledge to operate in a highly skilled mode and cannot draw upon experience to select the appropriate rule or protocol to carry out the appropriate task or solve a problem.
This suggests that the expert and the novice are likely to make different types of errors. The expert is more likely to make a slip or an occasional rule-based error, while the novice is more susceptible to knowledge-based failures. This is a strong argument for assessing the performance of an individual when they move to a new job or assume new responsibilities.
We can never eliminate human or active errors, but we can eliminate latent technical or organisational aspects that may set people up for an active failure (Reason, 1990). Reason refers to these as latent errors as organisational pathogens, which wait to combine with the right active human failure to have an adverse consequence. It is difficult to recognise a latent error or identify how it might manifest in the future as there are so many possible outcomes of technical and management decisions. However, since an event represents a fixed outcome, one is often able to identify the causes of latent errors by reasoning backward from the event. Once latent failures have been recognised, they can be diagnosed and corrected before they combine with an active error to produce an unwanted outcome.
Technical aspects associated with latent failures include such things as equipment design, software and mobile and material facilities. Organisational aspects stem from normal management responsibilities, which include the structure of the organisation, planning and scheduling, forecasting, budgeting and allocating resources.

Incident reporting
Although health care professionals share a common focus, which is taking personal responsibility for applying their skills to solve the patient"s health problem (Curtain, 1997), Perper (1994) noted that there is substantial underreporting of medical misadventures.
This can be attributed to the tendency in the healthcare field to blame the individual or individuals associated with an active failure. This produces a climate where individuals are reluctant to report events in case there are adverse consequences.
In addition, in a litigious environment it is easiest to blame the person involved. Reason (1990) pointed out that blaming individuals leads to ineffective countermeasures: disciplinary action, exhortations to be more careful, retraining and writing new procedures. Instead, error management efforts should be directed at learning how the system actually operates, as opposed to how to management thinks it is operating. Additionally, it is important to separate event reporting from employee assessment. Figure 1.4 highlights some factors that contribute to errors in healthcare delivery. Blood transfusion safety can be improved if errors (that are an indication of a weak system) can be identified before they result in an adverse outcome for a donor or patient. It is also important to avoid assigning blame when an error is identified, and to instead find the root causes of the error.

Barriers and limitations to incident reporting
Under-reporting of incidents is a widespread problem in the healthcare sector (Murff et al., 2003). The key factor highlighted by many authors is that the fear of reporting prevents the organisation from learning from the events and improving practices. Research in the healthcare and other industries has indicated that unjustified negative consequence have discouraged members of staff from reporting incidents. The non-punitive approach taken by the aviation industry has been a significant contributory factor to their impressive safety record in recent years (Vincent, Stanhope, & Crowley-Murphy, 1999 reporting. Staff members need to be provided with opportunities to reflect on practices in a multidisciplinary environment in order to become less fearful and give management an insight into the origins of incidents. More importantly, event analysis should be carried out with the members of staff who were involved in the incident, supported by a facilitator (Murff et al., 2003). Regulatory bodies and the health service have developed various reporting systems, although it is doubtful whether these reporting systems have in fact been utilised for organisational learning or to improve practice.

Hazards of poor traceability compliance
There is a lack of appreciation of the complexity of the traceability pathway.
Traceability is poorly understood and inadequately controlled in many cases. The level of safety, efficacy and quality of blood products must be maintained and optimised on an on-going basis. The only real test of effectiveness tends to be in live situations when an adverse event has occurred and needs to be handled urgently. In such situations traceability has often found to be wanting with long delays in tracing products and identifying recipients (Ashford, 2006).
Traceability is defined by three distinct elements. These are: the initial prescription of blood components, the distribution sheet stating the name of the recipient and the return of this sheet to the transfusion centre when the procedure is completed.
The prescription, distribution and transfusion of blood components is a complex process within the transfusion chain, traceability compliance explores the quality management system of the process.  Murphy and Kay (2004) looked at the extent of errors in many aspects of healthcare, including blood transfusion. Their review addressed issues of procedural errors and patient identification. The transfusion of blood to the wrong patient has been identified as the most important hazard. The risk of mistransfusion was found to be greater than the risk of transmission of HIV or HCV (Dzik, 2002(Dzik, , p.1190. The mistransfusion of blood products typically results from errors made during bedside checks prior to the transfusion (Sazama, 1990, p.583); this study estimated mistransfusion to occur at a rate of 1 in 12,000 units. Similar results were reported by Robillard, Chan and Kleimann (2004, p. 95) in Quebec. Moreover, they estimated that the incidence of mistransfusion was probably even higher due to the failure to recognise many errors. Another review by Pagliaro and Rebulla (2006, p.98), noted that errors in the blood transfusion chain seem to occur at a rate of 1 in 1,000 events and the data probably underestimated the magnitude of failures, not least because only one-third of errors have clinically significant consequences (Steinbrook, 2002(Steinbrook, , p.1758).
Finally, a Belgian study (Baele et al., 1994, p.117) uncovered numerous instances of mistransfusion that went unrecognised. When active tracking was introduced, they estimated that 1 in 400 units were mistransfused and the true frequency of some form of error in bedside blood administration was 30 times higher than rates reported using passive systems.

Potential solutions
Such studies have prompted the development of simple and effective methodologies to ensure correct patient identification in transfusion processes (Goodnough & Spence, 2003, p.161). The first positive donor-recipient identification study was described by Chambers et al. (1973, p.34). Since then, the main strategies that have been developed include the appointment of a transfusion safety officer, regular training, skills assessments, system re-engineering and the introduction of standard operating procedures (Shulman, Saxena & Ramer, 1999, p.595).
In addition to these organisational strategies, technology derived from industry and commerce has been incorporated into the blood transfusion chain. The ultimate purpose of these systems is to make the right and wrong actions easy and difficult respectively (Kohn, Corrigan & Donaldson, 1999). The aim of the study by Ballard et al. (2002, p.127) was to establish whether a patient had received a particular unit recorded on the blood transfusion computer system. They reported that 486 blood products were issued over a two-month period. Of these, 409 units (84%) could be verified against patient notes, but 77 units could not. These results are slightly lower than a national audit which found 92% compatibility (Murphy, Lowe & Pearson, 2001). The study was not large; a random sample of 5% of patients transfused February-March 2001. It also highlighted inconsistencies in documentation. Although the blood bank kept records of the issue of blood components, they did not know the ultimate destination of the blood, unless it was returned to them. The only proof that a specific unit had been transfused to a patient lay in the patient"s notes. This study did not identify or describe the problem across different medical specialities.
Turner, Casbard and Murphy (2003) carried out a retrospective study. A total of 51 blood units were audited before and after the introduction of barcode patient identification. The baseline data showed that only 14% of patients were asked to verbally identify themselves before the administration of blood. Twelve per cent of patients were not wearing their identification wristband when identification checks were carried out. Non-compliance was shown to be particularly related to poor checking of special blood requirements; proper procedures were followed in only 41% of cases. Furthermore, the study highlighted that there was no audible communication between staff members and the patient in relation to special blood requirements. After the introduction of barcode technology patients were asked to state their surname and forename and 96% of patients with special requirements had these matched to information on the blood bag. However, the barcode technology did not guarantee that all the necessary steps had been carried out as individuals could indicate that they had carried out checks when in fact they had not. This study also did not identify the particular medical area. Whitehead et al. (2003) conducted a survey of 500 sets of patient notes. A total of 1,212 blood components were issued and 47 labels were found to be missing.
Nursing staff in the clinical area were asked to assess a new system that used peel-off labels to be stuck in patient notes. Almost all respondents preferred the new system and found the labels easy to use. Other initiatives to improve the safety of the blood transfusion process included an improved blood collection protocol and staff training by a transfusion nurse. Davies et al. (2006) carried out an extended evaluation of an electronic blood collection system based on barcode technology. Practices were evaluated before and after its introduction in cardiac surgery. The baseline audit showed that patient documentation was brought to the refrigerator in 84%, 40% and 10% of collections from the main blood bank, theatre and the cardiac rehabilitation unit refrigerators respectively. Following the introduction of the electronic process, this increased to 95%, 90% and 100% respectively. The study estimated that there were 23 steps in the collection of blood before the implementation of the electronic process; this was reduced to nine.
The baseline audit demonstrated that there was very poor documentation of the date and time of transfusion and the number of units transfused. Observation of practices before implementation of the new procedures revealed that almost all the key steps in the transfusion process were not carried out correctly. The design of the electronic process highlighted the complexity of the transfusion process, which was identified as a problem. Another problem was improved training to avoid overreliance on the technology for the patient identification. A further weakness was that nursing staff were not required to enter any observations during and after the transfusion.
The study highlighted that the main reason for introducing electronic checking was that checking procedures were not being carried out correctly. However, the authors pointed out that total reliance the electronic system was also undesirable as the system was likely to fail from time to time. The study concluded that technology should not be used to take over thinking and that it was vital not to underestimate the role of education, training, and continued support.
A study by Davies et al. (2006, p.361) observed that traceability compliance was poor in some clinical areas because of the complexity of the transfusion chain and procedures. There were many steps in the requesting, matching, delivering and transfusing of blood products that involved a number of different departments and staff of different grades.
A pilot study at San Raffaele Hospital in Milan by Dalton, Poncet and Rossini (2005) used radiofrequency identification (RFID). Blood donors were given a RFID wristband and a label was placed on their donor unit at the time of donation. Later, at the time of transfusion, the data on the wristband and the blood unit were compared at the bedside using a handheld RFID reader. The reader communicated with the hospital computer via a hospital-wide wireless network. Users of this device reported that they were 27% more productive and no errors were observed during the pilot. Although it was difficult to measure the impact of the technology as only data estimates were collected prior to the study, 70% of staff commented that the system had dramatically reduced their fear of making errors.
Another pilot study at Maidstone Hospital in Kent used a questionnaire to evaluate a new labelling system. Seventy-nine per cent of respondents agreed that the labels were easy to use, although the poor response rate made it impossible to draw robust statistical conclusions from the analysis (NPSA,2006). Salmi, Azanowsky and Perez (1997, pp. 964-974) highlighted that data gathered from alert systems must be carefully interpreted to minimise bias. While Battle et al. (1998) emphasized that the characteristics of a good vigilance system should include no-fault reporting. They recommended that the donation number for blood and blood components and batch products for pooled blood products should be specifically recorded on the transfusion record. They also recommended the establishment of national documentation standards.
Other solutions have been proposed to the problem of ensuring that the correct blood unit is given to the patient. They include paper-based systems (Lau et al., 2000), mechanical systems (AuBuchon & Littenberg, 1996) and electronic systems (Norfolk, 2000). Dzik (2003) discussed technological approaches. The study highlighted that reductions in transfusion-related, knowledge-based errors depended on increased knowledge, more informed decision making and enhanced feedback. It also noted that reductions in slip errors was a problem ideally suited to machine-readable identification systems. Consolidated methodologies include biometric technologies, such as fingerprint identification, facial and vocal recognition (Ashbourn, 2004). Some of these technologies are not readily applicable to the field of medicine and surgery.

Conclusions
Although considerable effort has been devoted to improving traceability compliance following the introduction of national regulations,guidelines and mechanical barriers to prevent mistransfusion Davies et al. (2006), the literature review did not reveal any clear procedure for achieving it. The studies described in the previous sections suggest that the traceability of blood products is a problem in all hospital transfusion laboratories. Dzik (2006) and Murphy, Stearn and Dzik (2004) suggest that training and education might improve matters. Bennardello et al. (2009) and Pagliaro and Turdo (2008) also agreed that mechanical intervention may help.
However, none of these approaches attempted to assess the impact of ward size, speciality, training, traceability methods or volume of transfusion. There were considerable variations in the population and sampling criteria used in the various studies, and weaknesses in the audit tools. The articles reviewed also varied in quality and relevance. Most studies did not describe how fractionated products were traced from the donor to the recipient and none of them evaluated the variables that may affect traceability compliance, such as blood product usage by different specialities.
The design of some the studies was not explained which may have affected the results obtained. Other studies highlighted the problem before the introduction of the new model or simply explained improvements in traceability compliance. (Murphy and colleagues 2003), However, identified in their study tha the barcode technology did not guarantee that all the necessary steps had been carried out and the system can be overridden.
These findings justified the need of further investigation into why traceability compliance was poor within the Trust and why some wards were more compliant than others and the associated risk factors. While patients may be the victims of human errors, members of staff can be the targets and casualties of systems that are highly error-prone this will mostly contradict the research conducted by (NPSA,2006) that demonstrated that Trusts in the UK with a high level of incident reporting are more likely to demonstrate a safety culture (National Patient Safety Agency, 2006).

Proposed programme of research
The literature review had revealed weakness in traceability procedures in the healthcare sector and looked at some reasons why compliance is poor is some areas. In some Trusts, considerable effort has been devoted to improving training, redesigning electronic systems and proper patient identification before and after transfusion. However, none of these approaches have fully addressed the complexity of the transfusion chain which involves different specialities and staff.

Mandatory requirements introduced by European legislation, clinical governance
and patient safety concerns made this project necessary. There is a need for further exploration and better understanding of the issues surrounding traceability. I have some identified some that are realistic within the scope of this study. The findings from this study will provide new knowledge and understanding of these challenging problems.

Research aim and objectives
The overall aims of the project were to establish why traceability compliance was poor in the Homerton University NHS Trust, to implement changes to improve traceability compliance and to ensure that hospital wards complied with traceability procedures. Compliance was also expected to identify blood distribution and administration within the Trust.

Objectives
The study"s objectives were as follows: i.
To undertake a six year cohort study to identify the risk factors of traceability compliance for blood products within all wards of the Homerton University Hospital NHS Trust and identify the processes that lead to poor compliance. ii.
To select an appropriate traceability model and labelling procedure for blood products within the Homerton University Hospital NHS Trust. iii.
To evaluate the subsequent intervention. iv.
To evaluate impact and direct cost and saving blood products. v.
To examine the transferability of this evaluation and reflect on the impact on practice.

Results
The compliance data was normally distributed from 2005-08, the average traceability compliance ranged from 50 -90% (see Table 2-1 below). The overall mean compliance for the entire period for all wards was 72% with a standard deviation of 10.4. reporting, this led to a revolt by some clinicians who definitely opted out and labels were incomplete labels were discarded in wards bins but retrieved for audit, this might have suggest a poor compliance in 2008 as shown in (see Table 2-2). An interesting finding was that as the number of units transfused increased there was a corresponding increase in traceability compliance (Figure 2.1).  In the period 2005-2008, low usage groups had an average compliance of 68.5%.

Correlation between % compliance and units transfused
Medium usage groups had an average compliance of 73.8 and high usage groups had an average compliance of 80.2% (see Table 2-4). Moreover, the low usage group had a significant lower compliance rate than the medium and high usage groups (p < 0.001) using the independent T test. shape perceptions of a problem or situation and were not used.
The questionnaire was initially piloted to ensure that respondents could understand the questions and the response format was suitable. It was divided into sections that covered various aspects of the transfusion chain, including user satisfaction and performance related to traceability compliance. The objective of the survey was to gather more information related to the following research areas:  Training related to blood product collection and administration.
 Knowledge of traceability procedures.
 How traceability compliance can be improved within the Trust.
 Laboratory staff professionalism.
 Ward staff responsibilities related to compliance.
 Ward staff opinion on traceability.
 General considerations.
A postal questionnaire was used as this mitigates the effect of social acquiescence particularly as some questions could have been considered sensitive (e.g. ward staff compliance with Trust policy). A self-addressed envelope was enclosed for return of the questionnaire. A reminder letter was sent approximately a month later to encourage replies from non-responders. The initial response was not satisfactory with a total of 10 out of 36 (28%) returns. Further telephone calls were made to ward managers and a further 20 questionnaires were returned with a final response rate of 30 out of 36 (83%). During the data analysis, the outstanding nonresponders returned their questionnaires, which were included in the final analysis shown in Table 2-5. A Likert-style rating scale was used, which asked the respondent to rate how strongly they agreed or disagreed with a statement or series of statements. The data collected was verified by a departmental biomedical scientist to ensure accurate data entry, eliminate errors and identify missing values.
The data collected from the questionnaire was entered into a spreadsheet and analysed using the SPSS statistical package (SPSS, 2010). All data was screened to ensure that it did not violate the assumptions of non-parametric testing.
Descriptive statistics for each item were calculated.
The Kruskal-Wallis test is a non-parametric test used to compare more than two independent groups. This test was used to compare the compliance rate between different groups (e.g. specialities). Spearman"s rank correlation coefficient was used to analyse the strength of the relationship between continuous variables and traceability compliance and the Mann-Whitney U test was used to compare differences between two independent groups when the dependent variable was either (a) ordinal or (b) interval but not normally distributed.

Results
The results of the questionnaire sent to ward managers are shown in table 2-5 and 2.6 below.  The data was not normally distributed; a Kruskal-Wallis test was used to compare the relationship between average compliance and the various specialties. The results showed that there were no significant differences in compliance rates between specialties (p > 0.05; median 0.21). These outcomes suggested that traceability compliance was not associated with speciality. However, as there were only six specialty groups the study may not have been in a position to detect a significant difference.
The Kruskal-Wallis test also was used to compare the relationship between number of beds and average traceability compliance. The number of beds on the wards did not influence compliance rates (p = 0.70; median 0.49).
The Mann-Whitney U test was used to compare the relationship between average compliance and the presence or absence of a trainer on the ward. A statistically significant difference was found; average traceability compliance was better with a trainer on the ward (p = 0.03; median 0.16). Sixty-one per cent of respondents reported that 76-100% of their staff had been trained within the past year. It was evident that as the number of trained staff increased, compliance improved across all wards. Spearman"s rank correlation coefficient showed a strong correlation between the proportion of trained staff and traceability compliance (r = 0.59; p = 0.002).
Another statistically significant difference was found between timing of transfusions and average traceability compliance (p = 0.04; median 0.59).
Traceability compliance was better during the day than at night. However, the compliance rate was not associated with the method of sending labels (p = 0.14; median 0.40). per cent of respondent did not agree that there were clear procedures for returning labels, although 56% of respondents agreed that they were informed about monthly traceability compliance. It was also evident that some respondents did not think that they should report damaged or destroyed labels.
The transfer of patients between wards could be considered another risk factor for traceability compliance. Fifty per cent of respondents agreed.

Phase two: Direct observation
Direct observation was used to supplement the information obtained through the questionnaire.

Method
A structured observation checklist was designed to evaluate blood collection and bedside checking procedures at the Trust. The checklist highlighted relevant information about how blood products should be collected and checked at the issue fridge and patient"s bedside respectively. The checklist covered the procedures and standards issued by National Patient Safety Agency and SHOT, and British Committee for Standard in Hematology guidelines for blood collection. It also included the procedures displayed at the blood collection point and the bedside checking procedure on the ward.
The drawback of observational data activities is that they are time-consuming and the volume of data generated is immense. Therefore the sample size was kept small. One of the threats to the validity and reliability of data collected through observation is that of the observation effect. Another is time error. It is essential that the time at which the observation is conducted does not provide data that is atypical of the total period of the study. Due to a lack of access to the target population, a sample of 58 different transfusion episodes was observed. This represented 10% of the average monthly transfusions on the wards. The study was entirely voluntary and personal information that could be used to identify participants was kept to a minimum. All wards involved in the collection and administration of blood products with more than two transfusion episodes per week were included. Participants included staff members who collected and administered blood products on the ward. The anonymity of the staff members observed was protected.

Results
Data from the checklist was entered on into a spreadsheet and analysed using SPSS software (SPSS, 2010). The data generated by the observation was coded to enable a numerical analysis. Coding took the form of: 1 = compliant, 2 = noncompliant and 3 = non-applicable. Overall, the results obtained from the blood collection point and bedside check suggested that most of the time ward staff did comply with checking procedures at the blood collection point and the patient"s bedside. However, full compliance was not achieved at the blood collection point. Seven per cent of observed participants did not bring the prescription sheet for checking at the blood collection point, leading to the possibility of collecting the wrong blood. The transport box was not used by 18% of participants for blood collection, which increases the incidence of storage and wastage error. There was also non-compliance with the bedside checking procedure by ward staff. Seventeen per cent of observed staff did not check the patient"s identification against the blood transfusion request form. The patient"s notes were not used in the bedside checking procedure by 14% of observed staff and the blood pack was not checked against the form by another 14%.
My results was comparable to the National Comparative Audit of blood collection conducted in 2009, which identified that 3.9% of staff did not bring documented patient identification during blood collection and 5.3% did not check documentation against the blood bag collected.

Discussion
The peak percentage traceability compliance of 82% was obtained in 2007. This was followed by a drop to 65% in 2008. This led to the conclusion that the existing approach to resolving poor compliance (reporting non-compliance using the Trust"s incident reporting mechanism and targeting staff) did not resolve the issue.
Moreover, reviewing the volume of incidents from wards was not sustainable. The lowest traceability compliance was 50%, in 2005. The overall average traceability compliance from 2005-2008 was 69%. This shows that there was some improvement but it was not good enough. The results justified the need for establishing an enhanced system of transfusion practice and a well-structured change management programme which was not given a full attention during the study. The presence and absence of a trainer on the ward had a positive impact on traceability compliance within the Trust, there was a significant difference, It was encouraging to note that most of the procedures in the observation checklists were being adhered to. It was also encouraging to note that there was a 57 high awareness of error reporting in the Trust. Safety in blood transfusion is enhanced if errors can be identified before they result into an adverse outcome for a donor or patient (AuBuchon & Litternberg, 1996). Sellu et al., (2012) also highlighted in their study that simulation training is one way in which staff within the blood transfusion chain can be assessed effectively. This form of training has been used in the aviation industry to decrease system failures and individual errors and to introduce this method in the transfusion process is not easy as thought.

Blood usage levels and compliance
The study demonstrated a positive, significant correlation between the number of units transfused and user compliance. Medium and high usage groups had higher average traceability compliance than low usage groups throughout the years that were investigated (2005)(2006)(2007)(2008). High and medium usage groups did not have significantly different means.

Staff training
Eighty-nine per cent of respondents reported that they did not have a trained transfusion trainer or nurse on their ward, while 11% said they had such a specialist in-house. When there was no in-house trainer, the results showed that training was still given importance, even if a trainer was not available. This outcome suggested that training is a key element in the Trust.
The majority of the wards were aware of the importance of staff training, and this was highlighted by Murphy et al. (2004) who found that training and education may improve traceability compliance. The employment of a ward trainer is not as critical as the provision of other modes or means of training. However, the provision of training is significantly and positively correlated with traceability compliance.

Blood labelling
The results of the survey suggested that staff members agreed on the importance of blood labels. Moreover, survey participants strongly agreed that labels should be returned immediately after transfusion and they understood the importance of traceability compliance. It was encouraging to note that all the statements in the survey related to this area were evaluated positively by respondents.

Blood and patient identification
Out of all the behaviours that were observed using the bedside point observation checklist, the ones which were least observed were the patient's identity being checked against forms and labels and the blood pack identification being checked against reports and labels. Of the blood collection point checklist behaviours that were observed, the ones that were least respected were using the right transport box, patient identification being brought to the collection point and patient identification being checked against reports and labels. While staff members take their responsibilities in the transfusion process seriously, acknowledge the importance of training and skills assessment and generally follow working procedures there may be a need for a more permanent, technological intervention to increase traceability compliance. This is consistent with the results from Novis et al. (2003) who assessed the frequency with which the basic elements of bedside checking procedures were performed, including patient identification, blood compatibility identification and expiry date information. Their study revealed a failure to match wristband identification with the compatibility label in 25% of blood transfusions. The outcomes of the current study showed persistent difficulties in observing these patient identification protocols.
Errors in the blood transfusion process frequently relate to patient identification, blood units and its components. These erroneous slips can be reduced with the help of different techniques (Linden et al., 2000). In the case of the current study, it was found that majority of the wards still employed a paper-based tracking system.
There are many alternative technological approaches that may help to improve traceability compliance, such as mechanical barriers, barcodes, computerisation, automatic interface systems, portable computers and radiofrequency devices, and automated blood stores.

Feedback mechanisms
Slips are unintended errors caused by either not doing part of the routine or being distracted by someone. Slips can be avoided by the redesign of equipment and procedures in ways that make it more difficult to make a slip. For instance, a feedback mechanism can be designed into the IT system that alerts the individual when a slip is made i.e. alert from system to request label sent back and time and question operator if odd decisions were made. This is one approach that could be taken by the Trust to enhance traceability compliance.

Technology
Tracking blood samples using RFID technology has been cited in a number of studies looking at blood products and may be considered by the Trust. However, carrier and reader consistency and reliability are problems. A detailed evaluation is necessary to establish the limits of such advanced technology.

Patient transfer periods
The results indicated that the patient transfer period was significantly and positively correlated with ward compliance. The patient transfer period was measured by the number of patients transferred to other wards. Transfers happen regularly on wards and it is common to find the wrong ward indicated on the compatibility form and label.
The transfer period may be classified as an active failure since it is the health professional who decides it. A significant and positive correlation with traceability compliance indicates that as transfers to different wards become more frequent and the identification of the wrong ward on compatibility forms and labels increases, the likelihood of compliance increases correspondingly. Possibly, the identification of such errors causes more stringent enforcement of procedures, leading to higher traceability compliance within the ward.

Latent factors
It may also be worth noting some latent errors behind non-compliant behaviours.
Technical aspects associated with latent failures include such things as equipment design, computer software and mobile and material facilities, the barcode on the transfusion labels were not water proof, this need to be transcribed onto the computer system manually when damaged by an medical laboratory assistant. One aspect of organisational failures stems from normal management issues. This includes the structure of the organisation, planning and scheduling, forecasting, budgeting and allocating resources.

Blood product tracking models
This chapter first reviews the literature on the effectiveness of commercial traceability products. It then describes a survey of blood transfusion managers in NHS Trusts in North East England that gathered information about the actual process of traceability from professionals in the healthcare sector. The data gathered was used to inform the choice of traceability processes implemented by the Trust and described in Chapter 4. The purpose was to evaluate the error detection level, robustness, sampling techniques and the secondary outcomes of the blood tracking systems available.

Introduction
The aim of the literature review was to determine the range of blood tracking systems available and to summarise the evidence on the effectiveness of these systems. The specific objectives were:  To compare traceability procedures with the previous system or no system.
 To assess performance in terms of percentage compliance, error detection levels and mistransfusion events.
 To assess secondary measures such as cost-effectiveness, robustness and acceptance by staff members.
The review included any studies that evaluated the outcome of a traceability process. Traceability processes could include electronic tracking systems or paperbased methods. I included both randomized controlled trials and non-random trials, but as it was unlikely there would be many studies of this type, I also reviewed post-use evaluation reports and uncontrolled before-and-after studies.

Method
The review sought to identify all papers related to tracking systems for blood transfusion. I used search engines such as Medline , EMBASE (1980EMBASE ( -2012 and CINAHL ® , the Cochrane Library and the transfusion evidence library, which contains the National Health Service Blood and Transplant systematic reviews and Health Business Elite texts. The search terms used were traceability (including tracking*, traceable*, tracking system or vigilance), blood (including haem*, transfusion, red cell, blood* or blood products*), haemovigilance (or vigilance*) and blood products (including red cells, blood* or blood products*). In addition, the search included the websites of medical companies that produce well-known tracking systems and a manual search of conference proceedings up to 2012. The references of full papers were identified electronically and also included in the review.
The data collected related to the number of beds, number of blood units transfused, the need to trace blood products leaving the hospital (for external or home use), the quality of the study, the date of implementation and description of the new system, the system being replaced and outcome data. Studies were graded according to quality i.e. designed and outcome. Controlled studies were considered to be better quality than uncontrolled. Before-and-after and post-evaluation were both considered as uncontrolled studies.

Results
The initial review identified 378 references. Of these, 106 were found to not be relevant to the research question and were removed. The remaining 272 references were screened for eligibility against predefined criteria. Twelve references were identified to be of particular interest. These are described in detail in   analysis. The emergency room failure rate was above 20% due to the malfunction of the reader and the loss of data. The neuroscience unit failure rate was 25%, although only a few transfusions were performed by the unit.
The transfer of data between hospitals and National Blood Service is emerging in the U.K blood services but in its early stages. This will provide a complete traceability of blood products in the future, lately we could assess patient"s serology reports from the NBS and check blood products issued to hospitals across the country, and this is a beginning of total traceability within the health sector.
Doughty and Hitchinson (2010)  The study showed 81.9% traceability. The study also showed that traceability compliance differed according to the clinical area and was worst in surgical units.
They recommended that robust, efficient change in the organization and information systems was needed.
Hammadi (2009)  System (DMS), one hospital discarded all traceability records, while a second did keep some but a lot were missing. The rates of missing data were 87% and 40% respectively. After the DMS installation both hospitals were provided with a list of products supplied/received and were requested to check individual traceability records. Traceability compliance was found to have increased to 98%. The DMS implementation had resulted in a significant decrease in the number of missing records in both hospitals. The study recommended that manual records should not be discarded, but should be archived for easy access and retrieval.
All methods of traceability system from this systematic review had improved traceability compliance to a varied degree. The introduction of interactive training and prompt built into this tracking system will be a useful innovation to improve compliance and also a barrier to minimise poor compliance in areas of low usage of blood product as identified by Bachs and colleagues. It is also important to identify failure rate of electronic tracking system, an incident was identified in the emergency room highlighted by Uriz et al., 2011. The use of electronic tracking had improved traceability compliance in most of studies, but the occasional failure rate requires a risk assessment and further exploration of the limitation of this systems. Askeland et.al, (2008), also identified incidents such as mis-scanning of blood products, skipped steps and wrong steps. This shows the vunerability of this system.

Commercial software products
As early as the 1970s, Sherer et al. (1977) (Aller, 2005). Progress in forward-looking hospitals is being stimulated by products that have been specifically developed for the healthcare sector (Table 3-2). Barcode tracking systems such as I-TRAC (and its modified versions Safe Track and I-TRAC Plus) consist of a barcoded wristband for patient identification and a handheld portable data terminal. The patient and the blood pack are identified using a scanner and the information is downloaded via a portable printer.

Such a system was installed at the Georgetown University Hospital in the United
States (Sandler, Langeberg & Dohnalek, 2005), while Chan et al. (2004) described a simplified bar-coded method that reduced error rates in a Hong Kong hospital.
They reported that although compliance averaged 90%, problems were encountered with battery failures of the handheld device. In the United States, the University of Iowa evaluated a similar system and found that it was ten times better at capturing errors than the manual process. Despite the advantages of barcode systems, some studies found that traceability compliance did not improve (Nichols et al., 2004) as the barcodes on wristbands can be blurred and difficult to read.
Barcode-based systems require that the wrist and blood unit be scanned each and every time; in urgent situations this can make the process very clumsy.
The Tracesoft blood tracking system was designed to scan blood packs to be delivered to wards. It was designed to search for blood packs and identify their location. A central server provides a web-based view of activity, controls access to blood fridges and bedside software. All tracking base and bedsides devices are linked to the central server in a real-time thereby providing an overview of all product movements. Each process has a full audit trail and the system can send both visual and audible alerts to a nominated workstation. Full access control is incorporated into the system. Every user has a login and password. It also provides fine-grained control over access rights.
In 1991 the Albert Einstein College of Medicine in the United States evaluated the BloodLoc system. In this system, an alphabetic code taken from the patient identification bracelet must be used to access the assigned blood unit. The same system was implemented at the Dartmouth-Hitchcock Medical Centre in New Hampshire (United States) and the Gaetano Pini Orthopaedic Institute in Milan, Italy (Mercurialli et al., 1996).

Radiofrequency tracking systems
Another method is radiofrequency identification (RFID) to link barcodes or microchips to palm computers. In the mid-1990s, little attention was paid to RFID but since then it has gained momentum as the cost of implementation has decreased. Most studies show that the technology is suitable for transfusion medicine but its implementation may present challenges. All blood bag trials have used a frequency of 13.56 MHz, because of the ability to take readings through fluids. Trials have tested the durability of tags under real conditions, such as centrifugation, temperature changes and irradiation. There appear to be no trials that investigated the entire blood chain, from blood bag manufacturers to donation, production, storage, distribution and transfusion. Consequently, no studies have been identified that determine the return on investment from RFID.
Radiofrequency tracking systems have several advantages over barcoding in terms of patient safety. RFID chips can hold more data than a barcode and can include more information such as patient allergies, blood needs and any special requirements. Secondly, unlike a barcode, RFID does not require the use of light beams and RFID chips are interrogated through simple proximity to an electric reader. Dzik (2005) and Sandler, Langeberg and Dohnalek (2005) described the radiofrequency (RFID) identification method using a frequency of 13.56MHz. Blood bag tags were integrated into the blood bank computer system. Tags and patient wristbands were read by a built-in radiofrequency reader.
There are two kinds of RFID technology. In Active RFID the chip is powered by a battery and emits energy that can read over a distance, this is commonly used to track assets within an establishment. Passive RFID technology is more suited for blood bags and patient identification as it emits no energy. Tags are read only when in close proximity to the reader. At Massachusetts General Hospital (United States) a pilot study was conducted using passive RFID. A bedside reader was used to identify the blood bag and to provide feedback on the match between the identity of the patient and the blood bag (Dalton, 2005).
RFID was used at Saarbrucken Hospital in Germany to track blood bags, record transfusions and ensure that patients received the right blood, this was a small study and the effect of durability of the tags, centrifugation, temperature changes, irradiation was not investigated (Wessel, 2006). Hospital staff attached selfadhesive labels to all blood bags arriving at the hospital. The labels contained passive RFID chips with two kilobytes memory for storing unique identification numbers. The Return of investment is a big issue with the use of RFID in transfusion medicine and this require full further evaluation.

Limitations and challenges of barcode and RFID technology
Although traceability should enable tracking of all phases of the transfusion process, it can also breach privacy. Patient"s data are stored on chips and transmitted at different frequency depending on the reader and carrier. The full security of this data have not been fully evaluated, even though firewalls can be created to prevent access.
Carriers and readers are an important aspect of tracking. Carriers come in many forms, such as barcodes, data matrices, RFID and smartcards and each carrier has its own standards. RFID, for example, uses three frequency ranges (125 kHz, 13.56 GHz. and 900 MHz), each with advantages and disadvantages. Carriers and readers must be standardised if the codes they carry are to be easily read and stored. Moreover, there are still reliability problems with carriers. As yet, there is no ideal carrier; the choice depends on physical and environmental constraints as well as robustness and cost. Despite a lack of reliable studies, RFID is a very promising technology and may have a significant impact in healthcare (Fanberg, 2004, p.43).
In the United States, the American Food and Drug Administration requested the national RFID consortium to subject blood products to radiofrequency energy in worst-case conditions to determine whether there was any in vitro adverse impact.
The aim of the study was to ensure that any impact of exposure to the radiofrequency reader on blood safety could be assumed to be acceptable. Testing included an evaluation of product heating and any key biochemical changes from exposure to an intense radiofrequency magnetic field for longer than 23-25 hours (Davis et al., 2009).
The use of handheld devices to capture barcoded information from the patient wristband improves patient safety and minimises mistransfusion. The handheld device of a typical commercially available system receives and transmits information about patients and blood units. This requires an interface between the handheld device and the hospital network or a dedicated wireless network. Privacy and security are concerns with the use of both RFID and wireless technology.
Other technological limitations include the implementation of firewalls to protect the RFID database, tag disposal and recycling and shielding RFID tag or tag reading areas with metal screens to prevent unauthorized access (Davis et al., 2009). The use of RFID on blood products is in its early stages, because of its return on investment cannot be ascertain, most studies have been carried out on red cells, the effect of radiation on plasma products requires further explorartion and effect of electromagnetic radiation on blood products is currently under investigation. The cost of installing a tracking system is becoming affordable and cost-effective since the inception of this study compared to RFID.

Survey of blood bank managers
This phase of the study aimed to provide information about the actual process of traceability and opinions from healthcare professionals. The data gathered was used to inform the choice of traceability processes implemented by the Trust and described in Chapter 4. Data was gathered through a self-administered questionnaire sent to blood bank managers working in NHS Trusts in North East of London. It aimed to answer the question, "What variables are associated with traceability compliance in NHS Trust hospitals in North East of London" (Appendix 6). Most of the NHS Trusts that participated in this study came from Brentwood user group.
The Homerton University Hospital NHS Trust clinical governance committee approved the study. Participants were informed in writing that their participation was voluntary and anonymous. Non-respondents were not followed up as it was felt that they should not have to justify their decision not to participate.

Method
Hospitals with no blood bank services were excluded from the study. Twenty-five questionnaires were sent to blood bank managers who were invited to participate on a voluntary basis. A covering letter was attached to the questionnaire explaining that all data collected would be treated anonymously and destroyed following analysis. The questions related to individual Trust traceability compliance and addressed issues such as bed capacity, the availability of a satellite fridge, whether there was a tracking system in place, its cost and training.
Questionnaires were posted to participants together with a self-addressed envelope for the return of the questionnaire. Reminders were sent a month later to maximize the response.
Completed questionnaires were numerically coded to facilitate the analysis and organise the results. Descriptive statistics were used to evaluate the sample. The data was not normally distributed; the Kruskal-Wallis test, Spearman"s rank correlation coefficient and the Mann-Whitney U test were used in the analysis.

Results
The response rate was 64% (16 out of 25 trusts) and table 3-3 shows the demographics of the Trusts surveyed.  Demographic data was not normally distributed; hence the Kruskal-Wallis test was used to compare the relationship between bed capacity and average traceability compliance. The number of beds on the wards did not influence compliance (p = 0.64; median 0.62).
All respondents confirmed that they transferred blood products to other sites or hospitals, however the transfer of blood products to satellite hospitals did not influence traceability compliance (p = 0.44).
Seventy-five per cent of respondents said they had a paper-based tracking system, while 25% said they were using electronic tracking. The analysis showed that there was no relationship between electronic tracking and traceability compliance (p = 0.52; median = 0.25). It was evident that although electronic tracking and paperbased methods improved traceability compliance, the cost of the two systems required further exploration.
Regarding the percentage of staff trained in the collection of blood products, 50% said the percentage was 51-75%, while the other half assessed it to be 76-100%. The Trusts with most trained staff had better traceability compliance as Spearman"s correlation showed there was a positive correlation between training and traceability compliance but not clinically significant (r = 0.48; p = 0.06).
Eighty-one per cent of respondents said that their hospital had transfused less than 5,000 red cell units in the past year. Twelve per cent reported transfusing more than 5,000 red cell units and 6% did not know the number. There was positive correlation between transfused units and traceability compliance (r = 0.54; p = 0.03).
Ninety-four per cent of respondents said that average traceability compliance was greater than 75%. At the same time, few Trusts were working towards improving their compliance which suggested that both tracking methods (paper and electronic) improved traceability compliance.

Discussion
The objective of this review of current practice was to identify and explore the effectiveness of different blood product tracking models to inform the choice of traceability processes implemented by the Trust and described in Chapter 4. It was evident that staff training within the transfusion chain remained a big issue in some of the Trusts surveyed, and not all staff had been adequately trained.
Overall, the results showed that audits, regular feedback, traceability methods and the percentage of trained staff positively influenced traceability compliance.
Nevertheless, there are limitations related to data collection and the transferability of the study. The sample size was small and the selection of participating Trusts was not representative, which might have skewed the data. Participant"s were assured that all respondents will not be identified, there is tendency that compliance might have been overstated if they felt that they might be identified or data might be used outside the purpose of this study. The NHS Trusts with good compliance will likely give more information how compliance was achieved. Error detection level within NHS Trusts remains a concern, hospital laboratories with high error reportable level tends to come under scrutiny by regulatory bodies, consequently this has led to underreporting of incidents as identified by SHOT, 2011. Further studies will be required on a larger scale to assess the transferability and problems with traceability compliance .

Literature review
Of the 12 studies identified in the literature review, nine included data about postimplementation traceability compliance. However, the quality of the evidence varied and was subject to limitations. First, some studies did not specify the medical speciality concerned. In other studies a before-and-after comparison was made, but the lack of concurrent controls meant that there could be alternative explanations for the change in the compliance rate. Moreover, before-and-after studies and non-blind studies are prone to a reporting bias. One study simply reported post-implementation compliance rates. There were inconsistencies in the descriptions of how the studies were conducted and outcomes were reported. In addition to the limitations of individual studies, there were several potential biases in the review process itself. Relevant data may not have been captured either because it was not provided or because it was not published. It was evident that both the tracking system and the paper-based method shown in Table 3-1 improved traceability compliance; therefore the cost-effectiveness of both systems needs further exploration. The tracking system also freed up staff time by sorting blood products, reducing wastage and increasing the use of old stock.

Tracking methods
Documentation and tracking methods are important as they many improve compliance. The review of records or audits has conventionally been a way of performing quality assurance checks and documenting patient outcomes. The audit of records against predetermined criteria can be an important method in identifying errors or near miss events. In the survey of blood bank managers, 75% said they had a paper-based tracking method, while 25% said they were using electronic tracking. While this may be an area for improvement in compliance, technology is not a panacea. Moreover, electronic tracking may be an expensive option, reflected in the fact that 44% of respondents said they did not know the cost of the current system.
RFID technology may be a valuable tool, if it is affordable. It captures the movement of all blood products using a reader and carrier. It keeps patient information safe and secure, reduces the turnaround time for blood products tracing and helps to track and sort blood products in time and date order, hence reducing wastage and improving stock management.
Self-assessment may be another way to improve practices that are likely to lead to increased compliance. Self-assessment must be carried out regularly and monitored so that incidents can be reported immediately. To achieve secure and traceable procedures concepts regarding quality, security and efficiency must be incorporated into the process itself. On the other hand, the members of staff that are involved in the transfusion process must initiate action. Practices and methods must be established to prevent errors. Furthermore, it is essential to have a robust policy, implement systems that support the policy, assess the skills of staff members against national standards and audit the process.
The cost benefit of RFID in light of its capability is enormous but requires NHS funding.

New system implementation (Method 2)
The Trust had initially attempted to meet the requirements of MHRA guidelines using a paper-based method, which failed to achieve 100% compliance. As the guidelines required all Trusts to achieve 100% compliance this had become a concern and the hospital management and transfusion teams were made aware of the situation.
The Trust was not able to explore the electronic tracking systems described in Chapter 3 because of lack of funding for this project. There were doubts about whether it would be able to resolve the non-compliance issues. Consequently, the Trust management asked the transfusion team to provide a business case for alternative methods in order to achieve full compliance. This chapter describes and evaluates the changes that were made to quality control processes to improve traceability compliance at the Trust.
The new system consisted of an air-tube that was primarily designed for sending samples from wards to the pathology department. In 2009 it was extended to various other wards in the Trust who were asked to use it to return the labels from  Descriptive statistics was used to answer the research question to determine the mean and standard deviation of the annual compliance rate for each ward and visually illustrating data using graphs. The variance analysis was used to investigate the predictors of compliance at user group level (i.e. low, medium and high usage groups) and the one-sample t-test compares the mean score of method 1 and method 2 average compliance.

Results
Over all wards and both years the mean compliance rose to 91% (standard deviation 8.7%); although in 2010 there was a sharp increase in compliance (from 82-100%).     Variance analysis showed that there was no significant difference in the compliance means of the low, medium and high usage groups during the period (F = 0.70; p = 0.51) and this relationship was maintained each year.

Further interventions to improve compliance
Two further interventions were made to improve traceability compliance. These labels. Subsequently, unreturned labels or missing blood labels were followed up by requesting the patient notes from the Trust"s medical records department, and checking the wards and operating theatres for disposed labels.

Transfusion nurse specialist
The training of staff involved in the transfusion chain was the responsibility of a part-time specialist transfusion nurse. The Trust management also agreed to convert this post to a full-time post.

Discussion
Although between 2005-08 traceability compliance improved in some areas of the  Shulman, Saxena and Ramer (1999, p.595) highlighted that the main elements of a quality approach for preventing identification errors include having a transfusion safety officer, regular training, competent staff, system re-engineering and standard operating procedures. Similarly, Doughty and Hitchinson (2010)  Simple, innovative methods were developed to alert staff in the transfusion chain. A barcode label was created to simplify data input into the hospital"s information system and minimise errors. This was attached to every blood product issued.

Change management
This project attempted to ensure the rapid adoption of traceability compliance by wards in the Trust and improve the attitude of stakeholders to patient safety. A major challenge of the project was to instigate a change in staff behaviour, encourage them to move away from ritualistic practice and demonstrate that change can be a good thing. It was difficult to introduce change in such a complex setting and I encountered difficulties in breaking old habits.
Change management has been an integral part of organisational theory and practice for a long time. Indeed, some suggest that the future survival of all organisations will depend on their ability to successfully manage change (Burnes, 1996;Peters, 1989;Toeffler, 1983). Organisational change is usually required when there are changes in the environment in which an organisation operates.
However, it can be also be brought about by environmental factors such as political, economic, sociological and technological variables (Jury, 1997, p.27). It was anticipated during the design of this study that managing change will not be a problem. Consequently, there was resistance from senior clinicians within the Trust and complaints from the wards. In my role as a BMS I had to change the approach of frontline staff and explain our intentions in plain language. The study also assisted getting through to core staff within the Trust to explain the changes that need to be met. Orlikowski and Hoffman (1997, pp.11-21) proposed an improvisational model for managing technological change. The model is based on the assumption that technological changes constitute an ongoing process and that changes associated with an ongoing process cannot be anticipated in advance. Evans (2000) highlighted that in the twenty-first century change leadership is not simple. He sees modern leadership as a balance between a track record of success and the ability to admit mistakes and handle failure well. He argues that leaders need to balance their efforts across three dimensions of organisational change, namely:  Outcomes: Developing and delivering outcomes.  People faced with a potential change do not actually take it in, but become emotionally numb and have a sense of disbelief. In a sense they shut down and avoid thinking about the news. In the Trust, the clinical staff felt that additional tasks had been added to their workload and became resistant to further information.
When people allow themselves to acknowledge what has happened, they enter a stage of anger and begin to ask themselves, why me? They focussed their anger and frustration externally. Anger is a way of displacing real feelings to a situation.
Within the Trust labels were misplaced and discarded.
When they have exhausted themselves by attacking others or themselves, people still try to wrest back some control of the situation or their fate. Kubler-Ross identified bargaining as the next stage. The person desperately looks for something to remedy the situation. In the Trust, it was made mandatory to account for all transfused labels; a clinical incident was raised whenever a label was not returned.
Staff become depressed when it becomes clear that no amount of bargaining is going to provide an escape route from the situation, Kubler-Ross observed that many people move out of depression into a fifth stage of acceptance. Within the Trust, clinical incidents raised were investigated and corrective action was initiated.  Satir et al. (1991) observed individuals and families experiencing a wide range of changes. The model not only has a number of stages but also highlights two key events that disturb or move an individual. These are the "foreign element" and the "transforming idea". The study describes the initial stage as the status quo. This changes when something new enters the system. Satir calls this a "foreign element" in the sense that a factor previously not present is introduced.

Force field analysis
Kurt Lewin (1951) developed the idea of organisational change. Lewin introduced the idea of force field analysis, which examines the driving and resisting forces in any change situation (Figure 4.6). The underlying principle is that driving forces must exceed resisting forces if change is to happen.

Funding and training
It is important to note that funding was a concern. Hiring additional staff to help to achieve compliance with the new regulations was a challenge for the Trust. The potential benefits of electronic tracking systems will be explored when funding becomes available and a cost-benefit analysis will have to be carried out.
An in-house training session was developed for all staff involved in the transfusion process and it is now a mandatory requirement that only trained staff members can collect and administer blood products.

Clinical implications
This study has shown that several aspects of blood product traceability require more rigorous research. For example, there are few studies that have looked at traceability procedures after a blood product leaves the blood fridge. The impact of temperature and electromagnetic radiation on blood products has not been fully investigated and bacteria contamination has only been investigated. Few studies have examined how blood products were traced offsite or when patients are in transit or at home.

Application to practice
To reiterate briefly, the aims of the project were to explore reasons why traceability compliance was poor within the Trust and identify why compliance varied between wards. The key objectives were:  To assess traceability compliance for blood products in all wards of the Trust and identify the processes that led to poor compliance.
 To select an appropriate traceability model and labelling procedure for blood products.
The results of the study showed that blood transfusion is a complex chain that involves many stages. Some key variables were identified that had immediate financial and business benefits i.e. there was a drop in wastage of blood within the Trust, staff communication with transfusion department improved, more staff phoned the department and were willing to get involved in the training and assessment programme.
It also showed that training varied between wards and few wards were equipped with a trainer. In fact, the majority of ward staff only had a formal introduction to blood transfusion during their induction when they were first employed by the Trust.
Consequently, ward staff were not sure how to return transfusion labels to the laboratory. Wards that carried out a large number of transfusions were more compliant that wards that only did them occasionally.
The study also showed that traceability processes vary from Trust to Trust and the most Trusts anticipated that they would request tenders for an electronic tracker system in the future.
The study"s objectives were as follows: To examine the transferability of this evaluation and reflect on the impact on practice.

Evidence of achievements
As a result of the project, the following objectives were met.
To undertake a six year cohort study to identify the risk factors of traceability compliance for blood products within all wards of the Homerton University Hospital NHS Trust and identify the processes that lead to poor compliance.
 Specialist transfusion nurse: It was evident from my report that, the wards with a trainer had better traceability compliance. The report of my findings was sent to the Hospital Transfusion Committee, which agreed that we could submit a business case for a specialist transfusion nurse because the frontline staff lacked the skills in blood transfusion. It took a further three years to get funding but eventually the Trust approved full funding for the position.
 Development of competency: Following the appointment of the specialist transfusion nurse, a skills assessment was developed to be taken by all staff involved in the collection and administration of blood products.
Although the proposal was not welcomed by clinicians it eventually became mandatory. Training has begun, but only 20% of staff members have been assessed.

 Provision of clinical supervision:
There is now a system in place to support all staff involved in blood transfusion, day and night. Academic training has been provided to improve practice. The training is cascaded by the specialist transfusion nurse and credits are awarded to participants to promote professional development.
Objective 2 To select an appropriate traceability model and labelling procedure for blood products within the Homerton University Hospital NHS Trust.
The Trust Chief Executive approved funding for the extension of the air-tube system to the various wards within the Trust. The evidence gathered from the auditing of patient"s note resulted to the submission of a business case which eventually led to the appointment of an MLA. This had impact on practice by improving the return rate of labels from 90% to 100% and also investigating what had happened to blood units with missing labels.  Air tube system: This was ordered by the pathology manager to improve the turnaround time of samples sent from the various wards. Ward managers were asked to use this traceability system to return labels to the transfusion laboratory as soon as the transfusion was completed.
 Redesign of information technology: Transfusion labels and forms were redesigned to include the date, time and signature of the staff member that checked the product at the bedside before the transfusion began. Currently there are two different systems in place which do not comply with the requirement for full traceability of blood products.
Objective 3 To assess the impact on traceability compliance at the Trust.
 Annual skills check: it has become mandatory for all staff to be reassessed annually to prove that they are fit to practice. In the event of noncompliance they are re-trained and assessed before being allowed to collect or administer blood products. Objective 4 To evaluate the cost effectiveness of the traceability model.
The cost effectiveness of the traceability model was not fully met in this study but the direct cost savings in wastage was evaluated in this study. I make a total of £67,790 saving over a 6 year period see Table 5-1.
The following sections provide an overview of the implementation of these recommendations, although it is not possible to provide complete details of the planning, development and implementation of each initiative.

Specialist transfusion nurse
The study demonstrated that the role of a specialist transfusion nurse is crucial in the delivery of an effective service. Results showed that wards with a trainer had better traceability compliance than wards with no trainers. It was also clear that supervision and training from a practitioner with skills in transfusion was an essential requirement. The final outcome of the research was initiatives designed to meet the need identified by the study. In particular:  Teach and support frontline staff throughout their induction and while working on the wards.
 Offer information and support to frontline staff on a daily basis.
 Promote a range of continuous professional development options for staff responsible for the collection and administration of blood and its products.
 Assist staff in their training and become more proactive in identifying their learning needs.
 Promote and offer clinical supervision for the collection and transfusion of blood products.
 Liaise with transfusion laboratory scientific staff on clinical staff learning.
 Plan, teach and support practice nurses throughout their induction training programme.
This finding enabled the development of a tool for all frontline staff to encourage them to maintain a personal development plan. This initiative was in keeping with the requirement of the General Medical Services contract (Department of Health, 2003). It also provided supporting evidence for progression through the knowledge and skills framework along within the Agenda for Change implementation (Department of Health, 2004).

Skills assessment
In response to concerns from the National Patient Safety Agency, the clinical governance committee suggested a training programme and competency assessment designed to meet personal and professional accountability requirements. This initiative prompted the development of an in-house weekly training programme followed by a competency assessment. This new initiative enabled frontline staff involved in the collection and administration of blood to understand the issues, share their concerns and challenge evidence, which led to improvements in the services provided by the Trust. In addition, a forum was organised to improve awareness of blood transfusion services within the Trust.

Redesign of information technology
It was apparent that frontline staff involved in the transfusion chain lacked the skills to handle evidence-based information. The hospital intranet was updated with more information on transfusion services and hand-outs were made available. Nurses were encouraged to enter all records of transfusion procedures into the information system for accurate traceability.

Medical laboratory assistant (MLA)
The data obtained from this research revealed that there were various methods for returning transfusion labels to the department. It was evident that although some labels were left on the wards and others were found in the disposal bin the majority were sent through the air-tube system. The appointment of an MLA had impact on practice by improving the return rate of labels and also investigating what had happened to blood units with missing labels. The appointment of an MLA met the need highlighted in the study by:  Improving traceability of labels after the completion of the transfusion.
 Encouraging frontline staff to call the laboratory when labels were misplaced or destroyed.
 Identifying areas of poor traceability compliance within the Trust as a stimulus for more training.
 Minimising potential problems in the surgical setting where there was previously low compliance.

Air tube system
The air tube system installed at the Trust alleviated the pressure of returning the labels from transfused products to the laboratory promptly. The system has also changed and improved the rate of return to the pathology laboratory. The return of the labels enables the transfusion department to track the destination of the unit and update the system within 4-8 hours following the transfusion. Annual skills check Clinical governance introduced systems for continuing professional development (Department of Health, 1997). These systems are monitored and linked to registration, peer assessment, evidence-based standards of care and individual appraisal (Barr, 2000, p.81). Clinical governance is concerned with the ways in which the National Health Service can improve and maintain the quality of care and services provided for patients (Scally & Donaldson, 1998, p.61). National Occupation Standards were developed to raise the standard of practice in a given sector and providing a benchmark against which the performance of both the individual and the organization could be assessed and measured. These standards provide a systematic approach to the establishment of good practice, supported by a clear framework of underpinning knowledge, standards and expected outcomes (National Institute for Mental Health in England, 2003, p.3, 5).

Cost-benefit analysis
There has been a drop in annual blood wastage within the Trust from 290 units in 2006 to 100 units in 2011. The benefit of these savings amounted to £67,790 over the five year period within the Trust. The cost of for the full implementation was not evaluated in this study. The average cost of a full tracking system in a medium size hospital was valued at £350,000 excluding training and staffing during our lunch time presentation by vendors. The main area of cost impact were the appointment of a transfusion practitioner, a medical laboratory assistant and installation of the air-tube system within Trust. Porters were not needed to carry out deliveries, the reduction in manpower was also a big savings to the Trust and there was an improved of turnaround times in labels deliveries.

Email questionnaire
The duration of the project prompted a follow-up questionnaire to identify any gaps and capture the current trends in traceability compliance. An email-based survey was chosen to reduce the time and resources involved in individual interviews and to eliminate travel requirements.

Method
Participants were purposively selected from NHS Trust hospitals in the North East of London. The twenty respondents were blood bank managers. Participants were asked closed questions on individual Trust traceability compliance. The responses to the various questions were coded. As before, data was entered into a spreadsheet and analysed using SPSS software (SPSS, 2010). The data generated was coded to enable a numerical analysis. The data was not normally distributed,

Results
Table 5-2 showed that out of 30 emails sent to blood bank managers, 20 responses (66%) were acknowledged. The results obtained were similar to my previous study on page 71. There was a positive correlation between transfusion and traceability compliance (r = 0.59; p = 0.01) and training but with no significant difference (r = 0.30; p = 0.38).

Summary
The findings from the study reiterated that the importance of continuous professional development and skills assessments cannot be overemphasized, effective training at ward level improved traceability compliance.
The findings from this study also reiterated the importance of a trained specialist transfusion nurse who can provide support and training to frontline personnel involved in the collection and transfusion of blood products. Their role is also to ensure that training and the delivery of healthcare meet acceptable standards. The research highlighted the importance of training and continuing professional development. Wards where there are trainers provide a higher standard of care than wards without a trainer.
I also found out that wards that carried out many transfusions performed better than those that only carried out a few. This information enabled enhanced training to be provided in the poorly performing units.
The project provided an insight into the variety of services delivered to the endusers of transfusion services. It highlighted a lack of training tools; poor procedures for the return of labels and the unreliability of porters. Poor communication between the laboratory and end-users was identified as another issue. The response to these problems led to the innovative use of the air tube system and greater incident reporting at every stage of the transfusion process.
The initial survey provided a baseline from the problems that can be encountered at the blood collection point and in the bedside checking procedure. The issues that were identified included incomplete documentation brought to the blood collection point and untrained personnel collecting blood products at the blood collection 100 point. The problem of blood product collection was addressed by providing security locks on the blood fridges to deter untrained staff collecting blood products, In greater number of cases this had been effective. It has become mandatory that two nurses must check blood products administered on the wards.
It became evident that traceability compliance varied from one Trust to the other and various strategies were being deployed to attain full compliance. Technology has been introduced by many Trusts, but in this case there was neither the time nor the money to investigate it fully. This may not be so much of a problem; the study by Murphy et al. (2004) observed that most tracking systems could be overridden and concluded that such systems could only assist in reducing errors. Arinsburg et al.,(2011) also concluded from their study that additional training in transfusion medicine would be beneficial by physicians in all specialities. This was based on their findings from the survey conducted which demonstrated lack of knowledge of transfusion medicine across all specialities and all training levels.
It was also evident that previous studies had only focused on blood product traceability and not on the direct or indirect variables that affected traceability e.g.
ward size, transfusion times, training, traceability methods, transfusion volume, emergency situations, training and errors in the transfusion chain. The full impact of traceability will only become clear when more rigorous, large-scale studies are conducted that take into consideration all possible variables.

Reflection
My enrolment onto the professional doctorate has been an attempt to advance the scope of my practice and become a competent professional. I have found the professional doctorate to be both challenging and rewarding. The course has improved how I think about different situations and how I function as an advanced practitioner.
According to Honey and Munford (1992) I was exposed to new experiences compared to my prior learning. The professional doctorate influenced me to adopt a new learning style as an activist and a reflective thinker. Active learning engages students in class activities was encouraged and cooperative learning (working in small group teams under conditions that hold all team members accountable for the learning objective associated with the assignment (Felder and Brent, 2003, pp.282-283).
The curriculum was well-structured and related to my employment. There were opportunities to ask questions and read extensively in order to confirm any unclear evidence which I have missed during my previous learning. Berhold et al. (2000, p.191) described two groups of students, one taught in a circle and the other taught with tradition methods. The course grade of the two groups were compared and indicated that the traditionally taught students were more likely to get lower grades than the students taught in a circle. They concluded that holistic instruction may help a more diverse group to succeed. According to Kolb (1993, pp.3-4), learning begins with a concrete experience; the effect is reflective observation, documenting what happened and making sense, then investigating or theorizing. This allows conclusions to be drawn and plans to be made in order to take further action based on the experience. Felder and Brent (2004, p.269) proposed five instructional conditions that provide a balance of challenge and need to promote intellectual growth:  Variety and choice of learning tasks.
 Explicit communication and explanation of expectation.
 Modelling, practice and constructive feedback.
 A centred instructional environment.
 Aware of current levels of development.  Kolb and Fry (1975, pp.35-36) argued that the learning cycle can begin at any of the four points and should be approached as a continuous spiral. In this context, I was able to reflect on my previous employment about wastage of resources within the transfusion department. Bloom (1956, p.6) ranked learning into categories of knowledge e.g. ability to recall from memory to the most sophisticated and complex evaluations of making critical judgement. It is clear to me that the learning methods employed within the professional doctorate were designed to guide me to an advance stage of acquiring and understanding. The theoretical application in the real world encouraged me to decide on an issue and make critical judgments when planning to undertake any research for the department. This process also reinforces the importance of Kolb learning cycle.
The professional doctorate at the University of Portsmouth comprises two parts.
Part one was the taught element that covered professional review and development, advance research techniques and publication and dissemination.
Part two consisted of the project proposal and the research thesis.
Abstract conceptualization (3) Active Experimentation Observation and reflection (2) Concrete Experience (1) I was motivated throughout the course to learn new skills in research methodology, which I have overlooked in the past. The qualitative analysis was a challenge that was completely new to me, which I later found useful. It became obvious that I needed to focus more to understand this aspect. Aspects were the publication and dissemination module, and research and development that I found more relevant in. I have put this process into practice in decision-making as well as the modification of the training programme for Biomedical Scientist undertaking professional portfolio examination in Biomedical Sciences.
Part two of the course entailed completing an innovative research project that can also help in providing data in support of evidence-based information in blood transfusion.
I was excited to start this thesis that aimed to implement and evaluate blood product traceability procedures at the Homerton University Hospital NHS Trust. In the initial stages, it was clear that coordination was a problem because of a lack of funding to implement a tracking system and the lack of support from management.
Most of the activities aimed at achieving good traceability compliance were carried out behind closed doors and a lot of phone calls were made in order to follow up on poor compliance within the Trust. This meant that there was a heavy workload for myself and junior staff in the department and a steep learning curve. The project entailed big changes within the Trust which I overlooked in the early stages. While not stopping to think before taking action can produce spectacular results, I can recall that on many occasions I made things worse by blaming frontline staff and raising incident reports. Although this is normal procedure it did not resolve the problem of poor compliance. This happened because I failed to stop and think before trying to changes things. Later, I realised that I had to take a broader, educational approach to deal with the situation and frontline staff.
The visit by the Medicines and Healthcare products Regulatory Agency (MHRA) at an early stage of my project resulted in a change in direction concerning how traceability was perceived by frontline staff. The visit reinforced the importance of achieving full compliance, particularly as the Trust"s management including the Chief Executive were present to hear the MHRA"s summary of the inspection. This showed that traceability compliance was poor and that there were no structures in place or indications that the Trust was addressing the situation. This gave my work fresh impetus and management then took a lot of interest in my area of study.
I realised that I must take time to review the situation and consider theoretical aspects prior to commencing any work. I was aware of the risk that I might lose focus when working on such a large project, but I have a good record of bringing tasks to completion. I believe that I have achieved success when I have unconsciously taken the approach of breaking the task down into smaller bits, and approaching the problem logically. I now appreciate that this approach includes reflection as it creates natural breaks where I have been able to reflect upon what has been done and plan what remains to be done; this has contributed to my success.
The Trust"s Chief Executive became involved when a message was sent to all staff members saying that the collection, handling and usage of blood products was an area of great concern. I realised at this stage that a few senior clinicians had opted out, but the majority of frontline staff members were happy to be trained and assessed.
In 2008, I was nominated by my professional body, the IBMS, and became an assessor for Specialist Portfolios in Haematology and Blood Transfusion. This means that I must provide reports against pre-defined standards, and made me pay further attention to quality issues in blood transfusion; I compared information from visits and developed a framework for my thesis as the study progressed.
Although my project has finally ended, the importance of clinical and non-clinical personnel in the transfusion chain cannot be overemphasized. It was difficult to coordinate and pass instructions to frontline staff at every stage of the study. These The research has been daunting. It has required me to rally a multidisciplinary task force within the Trust for data collection and analysis. I had to get the complete support of staff and the Trust for the study to succeed.
I found the data analysis very difficult, but interesting. It required coding and rechecking for missing figures. The module I took on data analysis using SPSS was useful and this was followed by months of writing up. I would like to commend the patience of staff members; although many had little time, they listened and participated in the study.
The literature review during the study, highlighted incidents in the early days of blood and the advancement in blood, problems with blood safety, innovation achieved in the collection, testing, distribution and traceability of blood products.
The human error in the handling and the use of blood and its products remains a challenge in the 21 st century. My thesis, identified that there was lack of knowledge and training of personnel in the handling and the use of blood within the Trust which led to .

Developing the extending role
Achieving traceability compliance within the Trust had been a painful and complex process. I had to engage with all staff in the Trust in order to highlight the importance of full traceability. The Trust management were very helpful in providing their time and expertise. That the project has succeeded clearly demonstrates that biomedical scientists can interact with other Trust professionals and extend their role beyond the laboratory using their knowledge to provide diverse health service models.

Future goals
The professional doctorate considerably contributed to my self-development and management skills. I am confident in discussing research with my colleagues. My role as the transfusion lead has led to increased exposure at a more senior level allowing further development of my career.
My goal is to use my rich experience in clinical research to teach, and develop protocols and guidelines within the profession. Specifically, to influence practice through meetings, special interest group and disseminating information. I am also keenly looking forward to helping others to develop and maximize their skills in laboratory medicine. The achievement of the doctorate will help me to progress towards the post of consultant biomedical scientist.

Dissemination
A requirement of the doctorate is to ensure dissemination of the data and the outcome of the project; this has been achieved through various means both locally and nationally.
Locally, I made audits presentations at lunchtime meetings to health personnel in the Trust. Information was disseminated to wards with poor compliance levels in writing during the implementation stages. Furthermore, I developed feedback on the clinical effectiveness of the model, this was needed in order to assess its success and determine any modifications required to improve full traceability compliance.