The History Of Clinical Pharmacy Health And Social Care Essay

Historical Perspective

Clinical pharmacy has developed radically over the years; from the traditional apothecarial role to the current medication optimisation role. Hepler and Strand (1990) identified the importance of patient centred pharmaceutical care and analysed the impact of pharmacist interventions (1). They recognised that pharmacists can help reduce adverse drug reactions, optimise safety, reduce patients hospital stay, hence reducing the cost of care. Their paper ‘Opportunities and Responsibilities in Pharmaceutical Care’ emphasised that pharmacists have a social mandate in improving patient care and that a reprofessionalisation was required of Pharmacy (1). They found that a majority of adverse drug eents could have been prevented and identified that these occurred due to no inherent problem in the drug itself but errors in prescribing, dispensing or patient use of the drug(1). In 1990, Hepler and Strand proposed a mandate in which the role of the pharmacist included promoting patient welfare- a patient advocacy role, by preventing, identifying and resolving drug related problems and protecting patients from any harm due to medication; leading to an overall improvement in the patients quality of life(1).

Current trends

Hospital pharmacists are now fully engaged in medicine management and the risks involved with their use(2). Pharmacists contribute to integrated patient care; they may attend ward rounds with other members of the healthcare team and aid in delivering appropriate and rational drug therapy to patients (2,3).

Clinical pharmacy amalgamates the science of pharmacy with the skills of a practitioner to ensure safe, clinically appropriate as well as cost effective treaments (4). The culmination of the developments in clinical pharmacy comes with the changes in the government’s White paper ‘Equity and excellence: liberating the NHS’, supporting the clinical pharmacist’s role(5). The second principle from this White Paper focuses on health outcomes as opposed to targets (5). The framework indicates that clinical measures can be used to monitor progress and outcomes (5). Some of the proposed indicators include incidence of medication errors, number of controlled drug incidents and medicines reconciliation and compliance (5).

Medication errors

Medication errors are an important outcome indicator as 59 802 of them were reported to the National Reporting and Learning System between January 2005 and June 2006 by NHS staff(6). These were the second most reported errors after patient accidents, representing 8.3% of all incidents reported(6). Just over 80% of these occurred in acute, general and community hospitals(6). Although useful, information gleaned from thisvoluntary reporting of incidence is limited and subject to bias as it is dependent on individuals and others around thems willingness to report errors; many near misses are not reported as no significant harm actually came of them(7). This is especially true with prescribing or medication errors as doctors have a culture of not reporting hence these schemes have modest participation by them(7).

Medication errors, although unwelcome are an undeniable reality of any healthcare system due to its nature and complexity. There are many landmark reports which address the subject of medication errors namely, Building a Safer NHS for Patients: Improving Medication Safety(4), Pharmacy in England: building on strengths(8) and A Spoonful of Sugar: Medicines Management in NHS Hospitals(9).

Doctors’ swear an allegiance to the Hippocratic oath; to use treatments for ‘the benefit of the sick’ and to ensure patients are kept from "harm" (10). However as with any complex process the multiple steps in the medication chain leads to a high risk of errors (see figure 1).

Figure 1: medication supply chain

Medication errors can occur at any stage of the medication process from prescribing, transcribing, dispensing, administration and monitoring of the drug (11). A majority of these errors are seen to occur at the prescribing stage of the medication process (11).

Dean et al (2002) defined a prescribing error using the Delphi technique as a "prescribing decision or prescription writing process that results in an unintentional, significant reduction in the probability of treatment being timely and effective or increase in the risk of harm, when compared with generally accepted practice"(12).

The majority of prescribing errors found had a bias towards incorrect dosage selection (13-15). Dean et al conducted a pilot study to assess prescribing errors in patients in a London hospital over a four week period by pharmacists and their significance was rated(13). They found a prescribing error in 1.5% (95% confidence interval (CI) 1.4 to 1.6) of the 36200 medication orders written during the study. The majority of these errors; 54%, were associated with dose of the drug(13). Interestingly they noted changes in error rates in the period that patients were in hospital with a reduction in error rates in admission and discharge while there was a higher rate during the inpatients stay in hospital l(13).

Other errors include the wrong quantity prescribed, a mismatch between patient and medicine, wrong or omitted medicines or patient being allergic to medicines prescribed (6). Many errors originate in the process of writing the prescription as opposed to the actual prescribing decisions (16). The omission of vital medication such as insulin and anticonvulsants was highlighted as one of the major medicine incidents according to the NPSA (2007). With regards to high risk medicines see the heading ‘Never events’ below.

Studies show that the drugs most commonly associated with prescribing errors were antimicrobials and these errors were more common in adults (14). A reason for this may be due to anti-microbials being more one of the more frequently prescribed drugs. Other medications which are associated with errors include analgesics, calcium supplements, diuretics and thyroxine (16).

The majority of prescribing errors are made by junior medical staff (13,15). A confounding factor may be that they are the staff responsible for the majority of prescribing on the wards hence it is not possible to draw conclusions from this finding alone. In addition analysis of individual errors is of limited usefulness as bias can occur concealing the true source of the errors. Systematic and trend analysis gives a better indication of the scope the problem. This is an area this report aims to explore further in order to identify trends in errors and interventions to establish any patterns over time.

Systems analysis

Medication errors mostly cause negligible harm however some can produce fatalities; in order to prevent these errors it is important to analyse when and where they occur and assess the preventative systems in place.

Healthcare like many other industries for example the airline and engineering industry, has many areas of potential risk in which latent errors can occur due to the multiple processes, feedback loops and entry pathways(17). These industries give high priority to safety and have initiaves in place in order to minimise the risk of harm which the healthcare industry could learn from them. They, like the healthcare industry have many complex processes. However there are diffefferences berween them. In the healthcare system any damage nomarly occurs to a third party ie the patient, in addition there can be one single event which leads to harm or several cumulative events. Also there are many differnet modes of treatment namely surgery, counselling, medicines or investigations and hence the end outcomes differ too. In order to learn how to overcome errors first the systems in place need to be identified, then any weaknesses can be detected and corrected.

Actions do not occur in isolation; they tend to be part of a system of various entities; the hospital medication system is an example of this. In order to identify any weaknesses in the process; the type of system needs to be identified.

In engineering, the term coupling relates to interconnections between various components or processe s(17). Those systems which are loosely coupled have less interdependency, less co-ordination and less information flow (17). These systems can tolerate delays in processing ie time buffers are present so a problem in one component of the system will have limited impact on the subsequent process. Hence there is increased flexibility; re-ordering of sequences can occur and other resources can be used instead.

A tightly coupled system is one in which there is more dependency and more co-ordination and fewer buffers i.e. no time delay; events proceed quickly so an error in one component will lead to a significant impact subsequently (17).

This can be related to the healthcare system; initially entry into the system is a loosely coupled system as for example waiting lists are the time buffers, when the patient is in the system this is seen to be tightly coupled as things tend to move more quickly and there is little room for error. In this system one if the best ways to limit errors is to learn from previous failures and improve the system in order to prevent their repetition (18). This can be facilitated by error reporting; evidence suggests ‘safety cultures’, those in which reporting is encouraged show improved overall safety with little or no propogation of blame and proactively strives to identify latent risks (4). This contrasts to ‘blame cultures’ in which individual fear retributions leading to reluctance to report errors and may hinder the quality of care given(4). The report ‘Organisation with a memory’ (OWAM) emphasised the need for the NHS to implement an open culture in which reporting is encouraged to ensure that lessons can be learnt from failures/adverse events(4)

Why errors occur

http://patientsafetybydesign.files.wordpress.com/2009/12/swiss-cheese-model.jpg

Figure 2: Reason’s Swiss Cheese Model

There are various causes of medication occurs. In their report, An Organisation with a Memory, the Department of Health acknowledged that there must be individual accountability for errors however they emphasise that the cause of serious errors is due to latent defects that lie dormant in the system until they combine with human error, to cause a serious incident(4,5). The root causes of errors can be grouped according to the Reason’s model of accident causation or the ‘Swiss cheese model’ into active failures, error-provoking conditions, and latent conditions-see figure 2 (19).

The health care system can be likened to a stack of Swiss cheese slices. The holes in the cheese are potential weaknesses or areas that the process can fail in. Each cheese slice is a defensive layer in the process. As an error passes through one of the holes in the line of defence, it is normally blocked by another defence directly ahead hence preventing a serious error. However, if the Swiss cheese stacks line up such that all of the "holes" are aligned then the error may reach the patient –an active failure (See Figure 2) .

The commonly occurring active failures are due to errors/slips as a result of inadequate knowledge of the drug or the patient or lapses i.e. failure of memory or omissions. Error-producing conditions are those which affect performance, these include lack of training or experience, fatigue, stress, high workload for the prescriber and inadequate communication between the healthcare team (16). Latent conditions are organisational; areas of weaknesss that have the potential to cause errors for example are those that include reluctance to question senior colleagues and inadequate provision of training or inadequate resources i.e. no clinical pharmacy service at weekends. Defences are those factors which protect against hazards, in the case of pharmacy they include any checks or controls in place (15).

This model can be employed as a guide to root cause analysis. However its use is limited in that that the cause is often multifactorial due to several active failures and error-provoking conditions occurring simultaneously to cause medication errors.

As seen in the Swiss Cheese model; there are various factors which cause errors; Dean et all (2002) interviewed doctors who had made medication errors, the most frequent causes of errors cited were the work environment, workload, individual factors, and teamwork issues(16) Prescribing errors can occur due to inappropriate drug selection, wrong doses or strength or quantity, and contraindications and interactions (11).

Literature- prescribing errors

The EQUIP (Errors—Questioning Undergraduate Impact on Prescribing) study commissioned by the GMC investigated prescribing errors by foundation trainees (doctors in the first 2 years after leaving medical school) and compared them with errors made by consultants/seniors. The study consisted of three systematic reviews, a prevalence study, and qualitative data derived from interviews. For the prevalence study, 124 260 prescriptions in 19 hospitals in northwest England were examined over 7 days. 11 077 of these prescriptions contained errors—an overall error rate of 8·9%(15). If this is examined individually according to the level of training the doctors had, the error rate was 8·4% for foundation year 1 compared with 10·3% for foundation year 2 doctors, 8·3% for those in fixed-term specialty training posts, and 5·9% for consultants. Of all the errors found 1·7% of them were potentially lethal (15). Most of the errors were detected by other healthcare professionals i.e. pharmacists, nurses, or other doctors before the medication was actually dispensed (15).

Pharmacist interventions

Studies suggest that hospital pharmacists play a vital role in the defences against prescribing error, it has been shown that when pharmacists attend ward rounds they help prevent errors, and lower drug costs and reduce the occurrence of adverse drug events (2,20,21).

One study in the US found that in the intensive care unit having a pharmacist present reduced ADEs by 66% , in addition it was found pharmacist made 366 recommendations of which 99% were accepted by the doctors(20). A similar study based on general medical wards with 165 patients found that pharmacist made 150 interventions of which 98% of them were accepted by the doctors(2).

Having pharmacists on rounding teams has been proved tp be effective as pharmacists can play a more pro-active role using their expert knowledge advising doctors as opposed to a reactive role in which they just check prescriptions for any errors(2). However there may be issues with the human resources available; hospital trusts tend to employ relatively few pharmacist hence there may not be one available to attend every ward round.

Having reviewed the available literature above it is evident that there are studies available detailing the various types of intervention and their incidence rates however there is limited information on the patterns and trends of these interventions over time. This is something this report will address.

Adverse events

Adverse drug events (ADEs) have been defined as any injury which occurred as resukt of the medical management of a patient(22). This could be an omission or an oversight during the clinical care of the patient. It is important to note that adverse drug reactions alothough a subset of ADEs relate to inevitable reactions which could not have been predicted beforehand ie individual patient allergies. It has been estimate ADEs cost £750 million nationwide(6).

Error reduction

Never Events’

Never events are patient safety incidents which should not happen and preventative measures must be in place to ensure this. Primary care trusts are obliged to monitor and report any occurances to the NPSA on an annual basis (23).

The core list of Never Events for 2012-13 includes the following medication events(23):

Maladministration of Insulin

Insulin is a high risk medication and there were 3,881 patient safety incidents reported to the NPSA between 2004 and 2009(6). The ‘never events’ guidance states maladministration of insulin relates to employing abbreviations for the term ‘units’ when prescribing in addition when other syringes besides insulin syringes or insulin pens are used to administer; also if insulin is admistered at the incorrect time.

Overdose of midazolam during conscious sedation

Death or severe fatality due to a midazoloam overdose if a high dose (5mg/ml or 2mg/ml) is a never event and must be reported if this occurs(23).

Opioid overdose of an opioid-naïve patient

if an overdose of opiods is given to an opiod naïve patient and death or severe harms occurs this must be reported; or if the dose prescribed does not match with the dose agreed in the organisations formulary(23).

Inappropriate administration of daily oral methotrexate

This relates to prescribing, supplying or administering oral methotrexate daily to non cancer patients(23).

 Pharmacist Interventions

Pharmacists based on the ward routinely examine drug charts each weekday to check that all medication orders are clear, legal, and clinically appropriate to minimise the risk of treatment-induced adverse events(13). Pharmacists may annotate charts to ensure clarity for those administering i.e. nurses, in addition they may pick up on discrepancies in patients medications, or prompt the prescriber to change the drug choice if non formulary or clinically inappropriate. In New Cross Hospital; prescribers are required to complete an allergy box and if required, an antimicrobials sticker (Appendix 2); pharmacists tend to pick up on non=completions and advice doctors to complete these. Pharmacists on the ward must also ensure prescriptions are legal; ie the doctor has signed the prescription. Phrmacists are also involved in medicines reconciliation ensureing the patient is on all medicines they usually require. The recording of any interventions is a useful tool as it enables analysis of data to show any common areas of deficiency and errors.

Medicines reconciliation

On admission a drug history is taken for each patient; studies show this is the first step in which errors occur as omissions and errors in transcription are eviden(24)t. This may be due to a majority of patients entering into hospital on plyphaemacy and sometimes the sole source if information is the patient; patient may not recal the name of medications or the doses adequately, in addition non-compliance may not be disclosed to the doctor. These discrepancies in medication histories have been reported in 65% of patients(24).

Methodology

A retrospectively analysis was conducted; reviewing pharmacist intervention records, over 12 months from august 2011 to august 2012. In this report Pharmacist interventions are defined as any communication between the pharmacist and clinician that was recorded in the intervention booklet. Initially the intervention records of four pharmacists were selected; two randomly picked from the medicine wards and two randomly picked from the surgical wards. After review and recording of the two surgical pharmacists’ interventions it was decided to only use one from each ward as there was not a great difference in types of interventions recording between pharmacist and also consideration was given to time constraints (see Appendix 1). To facilitate analysis these interventions were risk rated using a standard risk assessment matrix (see Appendix 2) used by the Royal Wolverhampton hospital Trust with the help of the chief pharmacist and ward based pharmacits. This matrix is an adaptation of the National Patient Safety Agency (NPSA) matrix used to categorise incidents(25). The matrix assesses the likelihood of patient harm versus the potential consequences of the error/incident. A colour coded alert system was made for each intervention and allows comparison of the potential risk of harm of the emdicatin related issue. The colour coded system uses traffic light coding in order to assess the severity of the incident. Green indicates low risk and yellow, a moderate risk while red indicates a major risk. Other analysis that was conducted include; the common types of intervention and the most common medicines involved and if and how they changed with time and ward.

Interventions were classed into ----- categories related to the dose, dose frequency, dose timing, or formulation, omissions, non formulary, route queries, whether the allergy box was complete, or the antimicrobial sticker was used and filled in (see appendix 4), transcription errors, failure to communicate or obtain information ie with regards to the allergy status of the patient, and the use of drugs or doses inappropriate for the individual patient..

The Royal Wolverhampton NHS Trust

The Royal Wolverhampton NHS Trust serves a catchment population of around 330,000 (HES data 2006/07)(26). It is one of the largest teaching hospitals in the Black Country and has links with both University of Birmingham Medical School as well as The University of Wolverhampton training medics as well as allied professionals.

In addition to providing a range of specialist services the Trust provides acute care and has an operating budget of £255 million, employs nearly 5000 staff, contains more than 700 beds including 27 intensive care beds and 14 neonatal intensive care costs(26).

(include info on pharm staff numbers, items per year from dispensary; mi enquiries)

Limitations

One limitation of this study was the use of retrospective intervention data to assess pharmacists’ contributions

No way of truly knowing the outcoe of intervention, ie no way to follow up whether omission of drugs was intentional or not

Did not include any verbal communication between pharm n vlinician.

The PRACTICE study commissioned by the GMC on prescription errors in tgeneral practice referred to the following strategies to reduce risk: relating to personal prescriber strategies, organistion strategies and IT straaetgies(27).

The main strategies recommend reading the prescription aloud as a check, clarify any doses etc if they are unsure. Organisational or trust strategies include having a formuary in place guide prescribing to safer alternatives. Provide on-line access to clinical/medicines information resources, linking directly from clinical computer systems.

In addition in secondary care audit and error reporting which can raise awareness and lead to improvements.

Group training

Antimicrobial stickers

Complete allergy box

Pharmacists check/meds recon

A study has shown that two commercial electronic prescribing systems were associated with a statistically significant reduction in total prescribing error rates (66.1%, 57.5% and 60.5% in a total of three intervention wards in two hospitals). - See more at: http://www.npc.nhs.uk/rapidreview/?p=4998#sthash.tvSrEdyq.dpuf

http://www.npc.nhs.uk/rapidreview/?p=4998

http://www.health.org.uk/public/cms/75/76/313/3337/Reducing%20prescribing%20errors%20-%20evidence%20scan.pdf?realName=3bXWvp.pdf

http://www.nelm.nhs.uk/en/NeLM-Area/News/2012---May/02/GMC-report-Investigating-prescribing-errors-in-general-practice/