A clinical audit on secondary preventive care in patients with ischaemic heart disease in a public primary care clinic in Hong Kong

Yin-mei Liu 廖燕媚, Chi-hang Lau 劉知行, Catherine XR Chen 陳曉瑞, Yim-chu Li 李艷珠

HK Pract 2022;44:106-115

Summary

Objective: To audit the secondary preventive care in patients with ischaemic heart disease (IHD) managed in a public primary care clinic in Hong Kong.
Method: Study design: Clinical audit with comparison of two samples from a case series at different time points. Setting: A public General Outpatient Clinic (GOPC) in the Hospital Authority of Hong Kong. Subject: All stable IHD patients who had been regularly followed-up at Robert Black GOPC during the audit cycle were recruited. Evidence-based audit criteria and performance standards were set after reviewing data from local and overseas audit studies and latest international guidelines. Phase 1 evaluation was performed from 1^st June 2017 to 31^st December 2017, and areas of deficiency were identified. Active interventions were implemented for 12 months and Phase 2 evaluation was carried out from 1^st January 2019 to 31^st July 2019. Chi-square test and student’s t test were used to detect statistically significant changes between Phase 1 and Phase 2.
Results: Phase 1 data showed pronounced deficiencies in the assessment and control of cardiovascular disease (CVD) risk factors. For CVD risk factor control, only 34% and 18.4% of patients can achieve the optimal blood pressure (BP) control target and lipid control target respectively. The glycaemic control rate was 49.3% among patients with diabetes mellitus (DM). Following active intervention, marked improvements in outcomes were observed. Phase 2 data showed that optimal BP control rate and lipid control rate were achieved in 58.5% and 58.4% of targeted patients respectively (P<0.001), while the optimal glycaemic control rate was improved to 70.5% (P=0.008).
Conclusions: Through the process of identifying deficiencies and implementing effective enhancement strategies in managing IHD patients in public primary care clinics, the control of CVD risk factors had been significantly improved. It is postulated that the mortality and morbidity of IHD patients would be reduced in the long run.

摘要

目的 ： 在香港醫院管理局轄下一所普通科診所對非急性缺 血性心臟病患者的二級預防進行循證審計。
方法 ： 所有患有缺血性心臟病而病情穩定的病人， 如於審計週期內有於柏立基普通科門診定期覆診並 符合研究標準，將被納入本審計。本審計所取用的 準則與標準均於參照了本地及國際的審計與最新的 臨床指引之後釐定。透過於2017年6月1日到2017年 12月31日進行的第一週期評估，識別了處理不足之 處。再進行了12個月的積極改進，而第二期週期評 估於2019年1月1日至2019年7月31日進行。卡方檢驗 和學生檢驗被用於檢測第一期與第二期評估之間的 統計學顯著變化。
結果 ： 第一期週期數據顯示了在心血管高危因素的評估和控制方面都存在明顯的不足。分別只有34% 及18.4%的病人能達到血壓及血脂的控制目標，而 49.3%的糖尿病患者能達致理想的血糖控制。經過 積極改進後，達到了明顯的改善。第二週期的數據 顯示分別有58.5%及58.4%的病人能達到血壓及血脂 的控制目標(p <0.001)，而理想血糖控制率則上升至 70.5% (P=0.008)。
結論 ： 通過於公營普通科門診識別對缺血性心臟病患 者二級預防的不足之處，再實施有效的管理策略，可 以顯著改善心血管高危因素的控制。據推測，此類心 臟病患者長遠之死亡率和發病率將降低。

Introduction

Ischaemic heart disease (IHD) is an important disease both in Hong Kong (HK) and worldwide. Its incidence is increasing and it is associated with significant morbidity and mortality. The World Health Organization ranked IHD as the top cause of global death in 2016.¹ It caused more than 0.36 million deaths in the United State in 2015² and contributed around 1.⁸ million deaths annually (20% of all death) in Europe.³ In Hong Kong, the number of in-patient discharges and deaths due to IHD had risen by 45% in the past decades.⁴ Local data in HK in 2017 showed there were 3,867 deaths (8.4% of all registered deaths) attributed to IHD.⁵

There is strong evidence supporting comprehensive risk factors modification via lifestyle changes and medical treatment can help decrease mortality, reduce risks of subsequent cardiac events, and improve quality of life.^6,7 However, quality of secondary preventive care of IHD patients was far from optimal. A clinical audit carried out in South Auckland in 2001 revealed that a large proportion of post-IHD patients were not receiving optimal care, with only 45% of dyslipidaemia, 39% of diabetic and 59% of hypertensive patients having achieved the desired treatment target. Only 34% of patients were prescribed Angiotensin-converting Enzyme Inhibitor (ACEI), and 29% patients were not prescribed statins even when they were indicated.⁸ Another study in the United State also revealed an underuse of statin in post-IHD patients, with statin prescribed in 58.4% of patients only.⁹

Previously, IHD management is mainly the responsibility of cardiologists in the specialist setting. With the increasing service demand in IHD care due to an aging population and in view of the robust development of family medicine in Hong Kong, more IHD patients are advised to have continued care by primary care doctors upon their discharge from hospitals. For example, according to the preliminary data from the Head Office of the Hospital Authority (HAHO), over 50, 000 IHD cases are now being looked after by family physicians in the General Outpatient Clinics (GOPCs) of HA. In addition, over 50 percent of attendances at the clinic where the author is working are due to chronic disease management, with 6 percent of chronic disease attendances being attributed to IHD. Therefore, it is of paramount importance that IHD care at the primary care clinics is regularly reviewed to ensure that effective preventive measures are taken. Having said so, local data on the performance of secondary preventive care of IHD patients managed in the public primary care settings is still lacking until now. In view of this, this study aims to audit the management of IHD cases from a primary care clinic of HA and to work out improvement strategies. We believe that by improving the standard of care of IHD patients managed in the community via the audit approach, the disease burden including the mortality and morbidities of CVD could be greatly reduced in the long run.

Objective
Implications for clinical practice or policy

This clinical audit demonstrated common problems encountered in secondary preventive care of patients with non-acute ischaemic heart disease. Through the process of clinical audit via a team approach targeting at clinic, doctor, nurse and patient levels, noticeable improvement in patient care was achieved.

Methods
Setting audit criteria and justification of audit standards

Guidelines on IHD management published in the recent 10 years were identified from the PubMed (https://pubmed.ncbi.nlm.nih.gov/). After thorough literature review, the following evidence-based audit criteria and performance standards were adopted for this IHD audit (Table 1).

The criteria were adopted from the following evidence-based international guidelines:

1. Dyslipidaemias 2016 (Management of) ESC Clinical Practice Guidelines” published by European Society of Cardiology.¹⁰
2. AHA/ACCF secondary prevention and risk reduction therapy for patients with coronary and other atherosclerotic vascular disease: 2011 updated" by American College of Cardiology Foundation and American Heart Association.¹¹

Criteria 1-8 were classified as ‘must do’ criteria with the standard set at 90%, as these are important for CVD risk factor or symptom identification for which subsequent effective secondary prevention strategies could be adopted. For criteria 12-14 (to assess outcome performance on blood pressure, glucose, and lipid control), the standard was set at 65% with reference to local audit protocols on management of HT and DM^12,13, local audit and cohort study on secondary prevention of stroke in primary care^14,15, and regular reviews of BP, lipid and glycaemic control in hypertensive and diabetic patients in the primary care performed by the HA. Table 1 summarised the criteria and standard of this audit.

All patients with stable non-acute IHD coded by International Classification of Primary Care (ICPC) K74 (IHD with angina) and K76 (IHD without angina) who have been regularly followed-up at our Robert Black GOPC during the audit cycle (phase 1 from 1 Jun 2017 to 31 Dec 2017 and phase 2 from 1 Jan 2019 to 31 July 2019) were included. Non-acute IHD refers to a reversible supply/demand mismatch related to ischemia, a history of myocardial infarction, or the presence of plaque documented by catheterisation or computed tomography angiography. Patients were considered stable if they were asymptomatic or their symptoms were controlled by medications or revascularisation.¹⁶ Exclusion criteria were patients followed-up by Medical Outpatient Clinic (MOPC), patients with sporadic consultation, wrongly diagnosed patients, acute IHD, or those certified dead.

Totally 499 stable IHD patients were identified in the first cycle. Assuming the confidence level being 95% and confidence interval being 5%, by using the formula:

a sample size of 217 was needed. To allow room for case exclusion, a total of 228 were selected. Therefore, 228 randomly chosen number from 1 to 499 were generated using an online computer programme “Research Randomizer” available at https://www.randomizer. org/#randomize.17 Among the 228 patients included, 11 patients were excluded according to the exclusion criteria, with 3 patients being followed-up in MOPC, 2 patients with wrong diagnosis, 2 patients with sporadic consultation and 4 patients certified dead. The remaining 217 cases were included in the data analysis.

  Implementing changes and intervention: 1^st January 2018 to - 31^st December 2018

    After reviewing areas of deficiencies, a clinic team approach with changes targeting at different levels were adopted. Educational seminar was arranged for doctors, with updated international guidelines and deficiencies in disease management reviewed. Audit criteria summary was affixed on the desk of all consultation rooms for easy reference. Specific management plan was inputted in the reminder in the CMS (Clinical Management System) to facilitate communication with doctors. Progress of intervention was reviewed via the monthly clinic meetings.

At the nurse level, nurses were reminded to review the patient’s lifestyle risk factors in particularly smoking status, drinking habit and exercise pattern. Patient counselling and education would be provided and were based on updated guidelines. Possible deficiencies and corresponding implementation strategies are summarised in Table 2.

   Second-phase data collection and analysis

   504 IHD patients in total were found to have regular follow-up in the clinic during the second cycle. A sample size of 218 was needed to achieve the same confidence level and interval as described in the first cycle. To allow room for case exclusion, 233 cases were randomly selected from the case cohort for review. Among them, 15 patients were excluded (4 patients followed-up in the MOPC, 3 patients with wrong diagnosis, 3 patients with sporadic consultation and 5 patients certified dead) and the remaining 218 cases were included in the data analysis.

   Determination of variables

The patient’s age, gender, smoking and drinking status, body mass index (BMI), blood pressure (BP), haemoglobin A1c (HbA1c) level if diabetic and lipid profile were retrieved from the Clinical Management System (CMS). The most recent clinic BP or blood test were used for analysis if more than one reading or test had been performed during the study period. The BMI was calculated as body weight/body height² (kg/m²).

   Data analysis

SPSS software (version 25) was then used for data analysis. The chi-square test was used to analyse the categorical variables and paired student’s test was used for continuous variables to identify the differences of the measurements between the two phases. P value <0.05 is considered as statistically significant.

Results

    Table 3 summarises the demographic characteristics of IHD patients recruited into the two phases. Their background demographic parameters were comparable. A comparison of the performance achieved in the two phases is summarised in Table 4. Proper assessment of CVD risk factor in phase 1 was far from satisfactory, with inadequate documentation of alcohol use (39.2%), smoking status (41.5%), exercise pattern (43.3%), BMI (46.1%), cardiovascular symptoms (42.9%), blood sugar (72.8%) and lipid profile (72.8%). Satisfactory BP, glycaemic and lipid control were attained only in 34.0% of the hypertensive, 49.3% of the diabetic and 18.4% of dyslipidaemia patients respectively. Annual flu vaccine was only given or advised in 29.0% of patients. Criteria that met standard set included recording of BP (90.8%) and anti-platelet given (94.0%).

After 12 months of active intervention and implementation of changes, marked improvement in most of these criteria was demonstrated in phase 2. All of the 11 process criteria in the second cycle met the desired standard. (criteria 1-11, all p value <0.001 except p value=0.054 concerning use of anti-platelet agents). Optimal BP and lipid control were achieved in 58.5% and 58.4% of the targeted patients respectively (P<0.001). 70.5% diabetic patients reached the set standard (P=0.008). Table 5 summaries the comparison of outcome indicators in 1^st and 2^nd cycle.

   Discussion

   This study was the first local clinical audit on secondary prevention of stable IHD in a public primary care setting and has provided important background information on IHD management in the community.

1.    Criteria that did not meet standard in phase 1, then met in phase 2:

Criteria 1-4 (assessment of drinking status, smoking status, exercise pattern and BMI)

Assessment of these CVD risk factors was far from optimal. Doctors’ lack of awareness on these assessment and limited consultation time were the main reasons. Following active intervention, doctors’ awareness was enhanced and significant changes were achieved. Similar standard was met in other local studies. A local stroke audit achieved a standard of >90% on evaluation of these risk factors¹⁵. A local cohort study on stroke prevention protocol also reached a standard of >90%, with the exception of the assessment rate of BMI which was 79.3%¹⁴.

Criterion 5 (assessmenton cardivoascular symptoms)

Only 93 patients (42.9%) were properly assessed about their CVS symptoms during the first cycle. Most doctors expressed that they would routinely discuss CVS symptoms with their patients during their consultations. The low performance index was likely due to poor documentation. Target standard was reached in phase 2 (90.8%, P<0.001).

Criteria 7 & 8 (FBG/ HbA1c and fasting lipid checked)

Although not meeting the desired standard, a relative good performance of 72.8% was found in phase 1. Doctors were aware of the importance of regular blood test but there was no consensus on the regularity of blood test, with some doctors checking it annually, some checking every two years. Some doctors were less motivated to offer blood test to the elderlies aged 90 or above, with 14.3% and 12.8% of IHD patients being 90 years old or above in phase 1 and phase 2 respectively. Some patients refused blood test due to busy schedule or lack of awareness on the importance of regular blood test.

After the concerted effect, the set standard was met in phase 2. A local stroke audit also achieved a standard of 91% in annual checking of glycaemic and lipid control in the second phase¹⁵. Internal data in HA showed 94.1% of diabetic patients had HbA1c hecked within one year. Another local study on the quality of diabetic care also showed it could achieve >90% annual monitoring of lipid.¹⁸

Criterion 10 (use of ACEI/ ARB in hypertensive patients)

Only 50.5% of hypertensive patients fulfilled this criterion in phase 1. This is consistent with findings from the Hong Kong Cardiovascular Task Force Risk Management Program showed that 63.5% of hypertensive patients under medical treatment were put on ARB.¹⁹ Some doctors were not aware of recommendations on first line anti-hypertensive agents indicated for IHD patients, therefore knowledge gaps exist. Some doctors were hesitant to switch current anti-hypertensive to ACEI/ ARB due to worrying about possible side-effects and the inconvenience to monitor the renal function after initiating ACEI or ARB. Furthermore, some patients refused to change medications as they felt well with their current regimen, or the BP control was already optimal even without ACEI or ARB. Doctors were more confident in initiating ACEI/ ARB following active staff education and engagement. The set standard was achieved in phase 2.

Criterion 11 (seasonal influenza vaccine (SIV) was given or advised)

Only 63 (29.0%) IHD patients received SIV in phase 1. This is partly explained by the fact that many patients were still reluctant to receive the SIV due to a fear of the side effects. In addition, patients aged below 65 years without social benefit were not eligible for the free SIV under the government vaccination program (GVP) before 2018/2019, which means that many indicated patients have to receive the SIV out of their own pocket money. Fortunately, GVP in 2018/19 extended its coverage to include patients aged 50 or above, with subsidy for injection in the private sector, and therefore much more patients could benefit from this new scheme. Doctors were also more aware of the importance of flu vaccination and proactively promote it to indicated patients. With all these facilitating factors, the SIV coverage was much improved in phase 2 and the set standard was reached. This is consistent with findings from an audit on secondary preventive care of IHD in primary care performed in the United Kingdom, where 67.0% of IHD patients got SIV after the intervention.²⁰

Criterion 13 (HbA1c <7% if diabetic)

49.3% of diabetic patients reached the target HbA1c in phase 1. A large proportion of patients were elderly with other co-morbidities such as chronic kidney disease, stroke or vision impairment, which might limit their choice of oral hypoglycaemic agents and also their ability to handle insulin injections. Doctors also tend to accept looser glycaemic control in the elderly in view of their higher risk of hypoglycaemia. After educational activities, doctors were more familiar with the recommended pharmacological agents, and treatment targets for HbA1c. They were more motivated to adjust medication to achieve better cardiovascular risks factor control. 55 (70.5%) patients achieved satisfactory glycaemic control in phase 2 following these interventions, which was slightly better than those achieved via other local data noted. A local stroke audit reached a standard of 62% in the second phase.¹⁵ The latest DM review in the HA showed 58.8% and 62.9% of DM patients in the HA overall and the author’s cluster achieved HbA1c <7% respectively during the same period.

Criteria 12 (BP < 130/80mmHg if hypertensive)

Only 66 (34.0%) hypertensive patients achieved the target BP control in phase 1, which was increased to 58.5% in phase 2 (P<0.001). Although the improvement is prominent, there is still room for further enhancement when compared with other local data. For example, hypertensive review in the HA has revealed a 79.5% achievement rate of BP below 140/90mmHg among HT patients during the same period. A local stroke audit aimed BP <140/90mmHg for hypertensive cases and a standard of 73% was achieved in the second phase. Target BP < 130/80mmHg was set for diabetic patients in the same audit and 61% of patients had attained this target.¹⁵ A hypertension audit in Malaysia reached 59% of standard in achieving BP < 140/90 mmHg and < 130/80 mmHg in DM/ renal impairment.²¹ Data in Canada showed BP control rate at 60.1% in DM patients.²²

The reasons accountable for this relative suboptimal performance is multifactorial. On the one hand, some doctors were not aware of the latest guidelines on BP management for IHD patients and therefore still took 140/90 mmHg as the target BP. In addition, some doctors were concerned about the risk of hypotension and fall in elderly patients so they were less stringent on this target. This is particularly a concern as a high proportion of IHD patients were of an advanced age with multiple comorbidities. Indeed, guideline from American Heart Association suggests that patients with frequent falls, advanced cognitive impairment and multiple comorbidities may be at risk of adverse outcomes with intensive BP lowering.²³ An observational study showed an increased risk of serious fall injury associated with the use of antihypertensives in older adults, especially those with a history of fall injury.²⁴ Some patients and their families also had similar concerns and they refused to step up anti-hypertensives. HT with whitecoat effect or whitecoat HT may also be one of the reasons of not achieving the set standard, however this was not assessed in this study.

Criterion 14 (LDL <1.8mmol/ L)

Only 40 (18.4%) IHD patients achieved target LDL control in phase 1. Some doctors were not aware of the treatment target of LDL level in IHD patients. Some doctors were not initiating or adjusting the dose of statin because of therapeutic inertia, with only 72.4% of patients being put on statin. Patients’ non-adherence to secondary prevention treatment may also contribute to the low LDL control rate. Indeed, several studies have demonstrated that high levels of non-adherence of cardiovascular medications exist among IHD patients.^25-27 For example, a meta-analysis revealed low adherence rate of 54% for statins and 59% for antihypertensives, in patients with CVD.²⁶ Another cross-sectional study in United Kingdom showed rate of non-adherence to at least one secondary prevention medicine in IHD patients was 43.5%.²⁷

Though the standard was not reached in phase 2, there was still significant improvement in the lipid control rate from 18.4% to 58.4%. Doctors were more knowledgeable on statin use and they initiated the use of statin earlier apart from offering advice on lifestyle modification. The prescription rate of statin increased from 72.4% in phase 1 to 83.5% in phase 2. However, doctors expressed concern over the side-effects related to high intensity statin use in older patients and they tend to accept looser control with their LDL level. What’s more, some doctors were less motivated to intensify treatment when patients’ LDL level was close to target. These doctors’ behavioural patterns were well demonstrated in some local studies. A local retrospective analysis in 2003 revealed significantly lower prescription rate of lipid lowering agents in patients aged 90 or above²⁸, while another study showed that therapeutic inertia was common in the management of patients with known CVD and a closer-to-normal LDL level.²⁹

Criteria 6 (BP recorded) and Criterion 9 (antiplatelet agents given)

More than 90% of patients had BP recorded in every routine visit. After review of consultation notes and interview with doctors involved, we found that some were missed due to patients’ refusal or busy consultation. These data demonstrated that almost all primary care doctors were well aware of the importance of regular BP monitoring and the use of antiplatelet therapy in the secondary prevention of patients with IHD.

Limitations of the audit

Firstly, as the study was conducted in public primary care clinics of the HA, selection bias might exist. The results from this study may not be applicable to the secondary care setting or the private sector. Nevertheless, these data should realistically represent the IHD care in public primary care settings and provide groundwork for future service enhancement. Secondly, as the subject inclusion relied highly on the proper ICPC coding and CMS documentation, patients not properly coded or miscoded would be missed. The result may not reveal the real clinical situation if the clinical assessment were poorly documented in the CMS. Thirdly, some factors that might affect patients’ cardiovascular outcome were not assessed in this audit. For instance, patients’ drug compliance, dietary habit and use of allied health service etc. were not included in the data analysis. Lastly, only short-term outcomes were assessed in this audit and long-term outcome parameters like IHD recurrence rate and mortality rate were not evaluated. In addition, the relatively short implementation phase (12 months) might not be long enough for some criteria to achieve the target standard.

Conclusion

Ischaemic heart disease is a common and important disease worldwide and it causes significant morbidity and mortality. Optimal control of CVD risk factors and use of anti-platelet agents, statins and ACEIs or ARBs are proven to be effective in reducing CVD related morbidity and mortality. This clinical audit showed significant improvement in both the process of care and outcome of secondary preventive care among patients with IHD managed in primary care. Future audit focusing on the long-term outcomes such as recurrent cardiac events and CVD related mortality should be performed for better evaluation on the quality of care provided to patients with IHD.

Acknowledgement

We would like to thank all medical and nursing staff of the Department of Family Medicine & GOPCs, Kowloon Central Cluster of the HA for their unfailing effort and support to this clinical audit.

References