National Public Health Service for Wales Chronic Heart Failure

Treatment of Chronic Heart Failure
Clinical Audit Report
Author:Primary Care Quality and Information Service
Date: 29 June 2009 / Version:1
Status:Final
Intended audience: GP Practices, Local Health Boards, NPHS
Purpose and summary of document:
In December 2008, the NPHS Primary Care Quality Information Service (PCQIS) produced a Chronic Heart Failure Quality Improvement Toolkit to complement the November 2008 Welsh Medicines Resource Centre (WeMeReC) bulletin, Treatment of Chronic Heart Failure and its associated education module.
This Report is based on results of the PCQIS/WeMeRec Audit carried out by practices in the Wrexham LHB area who were asked to reflect on the results and consider lessons learnt from carrying out the audit
Publication / distribution:
  • Publication in NPHS document database (Primary Care Quality and Information)

Introduction

In December 2008, PCQIS produced a Chronic Heart Failure Quality Improvement Toolkit1to complement the November 2008 WeMeRec bulletin, Treatmentof Chronic Heart Failure2 and its associated education module. The PCQIS toolkit contained an audit designed to reinforce the main points of the bulletin and supplement the learning opportunities provided by case studies. Completion of the audit constitutes a suitable piece of continuing professional development that GPs can include within their appraisal process.

The following is a report on the results returned from 14 GP practices in Wrexham LHB area.

Method

Audit methodology was provided in the PCQIS toolkit,together withpatient data collection sheets, a sample size calculation matrix and recommended READ codes for searching on the practice system.1

Data collected was retrospective and could be taken both from the GP practice system and from the practice manual records.

Practices were asked to reflect on the results of the audit andto consider lessons learned from carrying out the audit, changes the practice might to implement as a result of the audit and any support needed for the practice to enhance the service it provides to patients.

Practice comments and reflection having a bearing on the dataare given below in their relevant context and a full list of comments is given in the appendix.heychanges to practice that youn against pneumococcusimmunisation recorded in the last 15 months
mpoatance of reporting changes

Results

1.Number of patients registered as having CHF

Practices were asked to compile a list of patients from the practice register with a recorded diagnosis of heart failure at least 12 months previously (Heart Failure G58.., Congestive Cardiac Failure G580. and Left Ventricular Failure G581.)and then select a patient sample with a diagnosis of heart failure for the audit. (Patients with a diagnosis of heart failure within the last 12 months were excluded).

The audit data collection form did not require practices to give their list sizes and total numbers of patients with CHF and so the following figures have been obtained using QOF data for 2007/08 (HF1)3.

Number of patients with CHF (all causes):

  • In all 23 Wrexham LHB practices:1189 (0.8% of total practice list for Wrexham LHB)
  • In the 14 practicesin the audit:708(0.8% of total list of practices taking part)
  • In the audit sample:564 (0.7% of total list of practices taking part)

Graph 1: Number of patients sampled as a total of QOF HF1 register

The bars show the number of patients sampled (Green) in relation to the total CHF register, each bar representing one practice with the aggregate for all the practices shown to the far right. The percentages should be read from the vertical axis, with the numbers in the bars relating to the actual numbers of patients. Two practices had in fact sampled more than the numbers on their QoF register for CHF. This may be an issue relating to the timing of the audit and when the data for the register was extracted, however this would need to be verified at the practice. Samples were on the whole high, with the aggregate rate at around 80 percent. Practice samples ranged from around 50% to over 90%

Practice comments – CHF register / audit sample:

  • The whole practice population was audited due to the small numbers of patients
  • There were 5 others labelled as HF but were either acute or right sided HF so were excluded.

Practice reflection

  • Re-coding is required to ensure patients with CHF can be easily identified / searched for futureaudits. Re code CHF patients
  • We may not be picking up all patients with CHF due to a low pick up on computer search
  • 1000 lives campaigners will provide patient held records

Note -as well as the points above:

  • Some practices may have selected patients with any type of Heart Failure (Left ventricular dysfunction (LVD) and CCF, Acute, or right sided HF)) others may have looked only at those with LVD.

2. Diagnosis

Diagnosing CHF should always include a detailed history and clinical examination (including precipitating or exacerbating factors and identification of concomitant disease, combined with diagnostic tests.1,2,4,5Electrocardiogram (ECG) and plasma B-type natriuretic peptide (BNP) measurement can be used to exclude heart failure.5If an ECG and/or BNP levels are abnormal, an echocardiogram should be performed as this is considered to be the most effective tool in confirming diagnosis.2,4,6CHF severity is commonly categorised using the New York Heart Association (NYHA) classification based on symptoms and exercise capacity. Other classifications exist, based on structural changes and symptoms.6,2,1

Audit criteria:

  • All patients with CHF should have had an ECG at the time of diagnosis
  • Diagnosis should be confirmed by echocardiography (ECHO)
  • All patients with CHF should have a NYHA, or similar classification recorded within the last 12 months (CHF Classification)

Practices were asked to record:

  • Number of patients with CHF who have had an ECG
  • Number of patients who have had an ECHO
  • Number of patients CHF classificationrecorded

Of all the patients in the audit with a recorded diagnosis of CHF across the 14 practices:

  • 81% had a record of an ECG
  • 73% had a record of an ECHO
  • Recording of CHF classification has not been analysed due to lack of data from participating practices

Graph 2: Patients with CHF - Recording of diagnostic tests

The chart above shows the percentage of patients having diagnostic tests, represented by the shapes (see chart legend). The percentages are read from the right hand vertical axis. The bars show the actual number of patient notes used for the audit. Each bar and set of shapes, in line with the bar, represents one practice. An indication of good recorded practice would be a cluster of shapes towards the top of the graph e.g. practice 12. Here fifty patients have been audited.

The aggregated figure shows 584 patients have been audited across the practices, showing;

  • 60 percent of these patients had heart rhythm tested;
  • Over 70 percent had an echo and
  • Around 80 percent had an ECG recorded

Practice comments - diagnosis:

  • The relatively high values for echocardiograms and ECGs represent our attempts to consolidate diagnosis in these patients. Our high monitoring figures for these is enhanced by coincident diagnoses.
  • ECG – no recorded ECGs in medical record at time of diagnosis
  • There are probably some problems with recording significant investigations in the notes
  • Some patient’s notes suggest they had an echo but not documented
  • Some people diagnosed before echocardiograms became routine or ECGs availablein surgeries
  • Current monitoring is good but historically diagnosed. Patients not referred for echocardiogram or ECG if well controlled
  • Some patients diagnosed years ago with no indication on letters as to whether they have had an echocardiogram or not
  • Two patients without a record of echocardiograms (one of whom did not have an ECG either) were diagnosed in early 1990’s
  • Need to READ code previous echocardiograms
  • CHF class – only one patient had this noted in a hospital letter. It would appear not to be so widely used by the hospital or primary care services in this area
  • Not all patients have CHF class as only added to the template this year. Will need to do an audit to get these updated

Practice reflection

  • ECG needs to be more often used as an initial diagnosis
  • Open access to echocardiograms to confirm diagnosis
  • I would like to see access to BNP blood tests for screening and ongoing monitoring
  • 8 patients were removed from the register as there was no confirmed evidence of CHF. In the past hospital didn’t always document in the patient’s discharge letter that they had done an echocardiogram. Recently we have received more reliable documentation but not exclusively.
  • More room space and Nurse / HCA hours to perform ECGs
  • Do we review the diagnosis of CHF with echocardiogram and ECG in those patients labelled historically? Easier access for echocardiograms and ECGs in patients with a historical diagnosis
  • Some patients have a diagnosis dated from many years ago and with no indication on the hospital letter whether they have had an echocardiogram or not
  • We will continue to search and audit to confirm accuracy of diagnosis, perhaps on a four monthly basis as well ad mark records to re-confirm accuracy of diagnosis, encouragement of correct and substantiated diagnoses with relevant removal of later disproved tags.

Note -as well as the points above:

  • ECG is used to exclude HF therefore one would not expect all patients with suspected CHF, and who have consequently had an ECG to have a confirmed diagnosis.
  • Purpose of ECHO is to confirm HF so all patients recorded as having CHF should have a record of having had an ECHO

3.Pharmacological Therapy

Pharmacological therapy is the mainstay of treatment for people with CHF.2

3.1Angiotensin Converting Enzyme (ACE) inhibitors and Angiotensin II receptor

blockers (ARB)

ACE inhibitors should be considered for all patients with CHF associated with left ventricular systolic dysfunction(LVD)2,4,5Angiotensin II receptor blockers (ARB) may be considered as an alternative for patients who are intolerant of ACE inhibitors.2,4,5

Audit criterion: All patients with CHF should be on an ACE Inhibitor (or ARB) unless contra-indicated

Practices were asked to record:

  • Number of patients with CHF taking an ACE Inhibitor (or ARB)
  • Number of patients with contraindications to ACE or ARB.

Of all the patients in the audit with a recorded diagnosis of CHFacross the 14 practices:

  • 76% were recorded as taking ACE inhibitors or ARB
  • 10% were recorded as having contraindications to both/either ACE or ARB
  • 14% were not recorded as being on ACE/ARB nor recorded as having contraindications

Graph 3: Patients with CHF recorded as taking ACE / ARB

Each bar relates to one practice, with the aggregated total to the far right. Green indicates patients on an ACE or contra indicated, with red those who could be prescribed an ACE and are not. The percentage of patients in each category is read from the vertical axis and the numbers in the bars showing the number of patients audited in each category. Good practice would be indicated by smaller areas of red. Most practices have 20 percent or fewer patients who could be on an ACE and are not, totalling 77 patients across the 14 practices. There are four practices indicating over 20 percent which account for just over 50 percent of patients who should be treated with an ACE.

3.2Beta-blockers

Beta-blockers should be considered for all patients as soon as their condition is stable, usually following the introduction of an ACE inhibitor (and a diuretic, if necessary)2

Audit criterion: All patients with CHF should be considered for a beta-blocker unless contra- indicated

Practices were asked to record:

  • Number of patients with CHF, recorded as taking Beta-blockers
  • Number of patients with CHF with Beta-blockers not indicated
  • Number of patients with CHF with a contraindication to Beta-blockers

Of all the patients in the audit with a recorded diagnosis of CHF across the 14 practices:

  • 52% were recorded as taking Beta-blockers
  • 5% were recorded as not indicated for Beta-blockers
  • 25% were recorded as having contraindications to Beta-blockers.

Graph 4: Patients with CHF recorded as taking beta blockers

Similarly to the previous chart, the green shaded areas show patients who are prescribed Beta Blockers, not indicated for or contra indicated to Beta Blockers. Good practice is represented by smaller areas of red. Consideration should be given to the numbers of patients shown bythe numbering each bar. This is particularly evident with practice 8 which shows over 75 percent of patients not on a Beta Blocker; although the numbers involved are small.

Practice comments – ACE / ARB and Beta blockers:

  • Only 28 of the CHF patients taking ACE /ARB. These patients require a more detailed search of their notes to clarify any CIs and to be reviewed with reference to NICE guidance.
  • Moderate level of CHF patients on ACE or labelled to exclude. Likely to be due to poor labelling or contraindication rather than inadequate treatment.
  • ACE /ARB has not been offered to three elderly patients
  • 6 patients not on ACE /ARB with no contraindications recorded
  • Lack of documentation in some notes as to why some patients are not on an ACE /ARB (not indicated or CI)
  • Some patients are not on a beta blocker with no documented explanation (Not indicated or CI)
  • 9 patients could potentially be on a beta blocker
  • Beta blockers use is at a low level. Likely to be due to poor labelling or contraindication rather than inadequate treatment.
  • We have not pushed beta blockers in long standing stable patients
  • Need to review beta blocker usage in patients overall

Practice reflection

  • Carefully review patient and try to start on ACE /ARB or beta blockers where appropriate. Titrating to target doses of beta blockers and ACE / ARB is also reviewed - not covered inaudit
  • The usage of beta blockers could possibly increase. Put message on screen to remind clinicians to increase dose of ACE / beta blockers to maximum tolerated dose and document once maximum dose achieved
  • Some patients may benefit from taking their medication to maximise doses. Unfortunately these are the ones that decline to attend the surgery
  • The data collection we were asked to perform shows fair performance in relevant prescriptions. Further study of these patient groups shows that our relative under- prescribing of ACE and beta blockers is better than these raw data suggest – an additional data collection point of hypotension reveals that full doses of these drugs are often elderly patients – perhaps this study could be age stratified?
  • Review need of beta blockade in patients without contraindications or other record of usage. Review ACE /ARB use for patients with not contraindications or other record

Note -as well as the points above:

  • ACE inhibitors / ARB are indicated for CHF patients with LVD. In the audit, practices were not asked to differentiate between those with CHF who had, or did not have LVD. Therefore the percentages of patients taking ACE/ARB are given out of the total number with CHF.
  • ARB is appropriate for patients who are contra-indicated to ACE. Consideration could be given for future audits to record contra-indications to medication types separately i.e. number with CI to ACE who are taking ARB.
  • The data collection required practices to indicate where there was a contra-indication to ACEand / orARB so this figure will relate to patients with CI to either one or both.
  1. Monitoring and Review

The frequency of monitoring should depend on the clinical status and stability of the patient. The monitoring interval should be short (days to 2 weeks) if the clinical condition or medication has changed, but is required at least 6 monthly for stable patients with proven heart failure. 4

ACE inhibitors should be started at low doses and titrated up to target doses or the highest tolerated dose. Blood chemistry (urea, creatinine, and electrolytes) should be checked at baseline, one to two weeks after initiating therapy, and after each dose titration.2

Treatment for Beta Blockers should be initiated at low doses and titrated slowly (at not less than two weekly intervals) to target doses. This can be undertaken in primary care by experienced practitioners. Heart rate, blood pressure, and clinical status should be assessed after each titration.2

Audit criteria:

  • All patients with CHF should have had U&E’s recorded in the last 12 months
  • All patients with CHF should have had Creatinine recorded in the last 12 months
  • All patients with CHF should have had BP recorded in the last 12 months
  • All patients with CHF should have had heart rhythm recorded in the last 12 months
  • All patients with CHF should have had a detailed medication review in the last 12 months

Practices were asked to record:

  • Number of patients with CHF with U&E’s recorded
  • Number of patients with CHF with creatinine level recorded
  • Number of patients with CHF having had their BP
  • Number of patients with CHF having had their heart rhythm recorded

Of all the patients in the audit with a recorded diagnosis of CHF across the 14 practices:

  • 95% had U&Es recorded
  • 95% had a creatinine level recorded
  • 97% had their BP recorded
  • 61% had their heart rhythm recorded
  • 87% had a medication review recorded

Graph 5: Patients with CHF – recording of monitoring and review

The chart above shows the percentage of patients having monitoring tests or a review, represented by the shapes (see chart legend). The percentages are read from the right hand vertical axis. The bars show the actual number of patient notes used for the audit. Each bar and set of shapes, in line with the bar, represents one practice. An indication of good recorded practice would be a cluster of shapes towards the top of the graph which is demonstrated by many Practices here, and is reflected in the overall aggregated coverage.