Summary

Objective: To audit the process of hormone replacement therapy (HRT) prescription in a government primary care clinic.

Design: A retrospective clinical audit was performed.

Subjects: All HRT users attending the clinic were recruited.

Main outcome measures: Evidence based audit criteria were made after extensive literature search on latest evidence including local and international guidelines. All HRT users who attended my clinic were recruited. Relevant data were collected and analyzed retrospectively. First phase evaluation was performed from August 2004 to November 2004 and several areas of deficiency were identified. Strategies for improvement were suggested and implemented with consensus from all staff in the clinic. After the period of implementation phase (April 2005 - July 2005), second phase evaluation was performed in late July 2005.

Results: There were 124 patients in phase 1 and 93 patients in phase 2. More than half of the HRT users were aged between 50 and 59 years old. 31 patients stopped HRT after the initiation of this audit. All the performances in the second phase reached the standard with significant test results (p<0.0001).

Conclusion: The impact of this audit on patient care is positive. The process of HRT prescription was improved so that the HRT users would be more certain about the current evidence and hence decide whether they should continue the HRT therapy. The benefit from this audit was not only limited to patient care. It provided opportunity to enhance communication among staff in the clinic. Clinic staff had better understanding on nature of audit and all realized that quality of care could be improved through this exercise. It also facilitated a positive culture among clinic staff so that they would try to identify deficiencies of the clinic and find practical methods to tackle them in a systematic way in the future.

Keywords: Hormone replacement therapy, primary care, evidence based audit

摘要

目的：審計一個政府基層醫療診所的激素替代療法(HRT)處方過程。

設計：回顧性臨床審計。

對象：所有到診所就醫的HRT使用者。

測量內容：就本地和國外包括HRT指南在內的最新資料進行了大量文獻檢索，制定了實証審核標準。所有到本診所就診的HRT使用者均被納入。回顧性地收集並分析了相關的資料。 2004年8-11月開展了第一期評估，發現多個不足之處後，提出了相應改善的措施，得到診所全體員工的同意後實施。在執行期（2005年4-7月）之後，於2005年7月下旬進行了第二期評估。

結果：第一期有124名HRT病人參加；第二期有93名HRT病人。半數以上的HRT使用者介於50-59歲之間。本次審核開始後，有31名病人終止了HRT治療。第二期的所有操作都達到標準，結果具有顯著意義（ p<0.0001）。

結論：本次審計對病人治療有積極影響，HRT處方過程得到改進。HRT使用者對目前的醫學證據有更肯定的瞭解，因而可以自行決定是否繼續治療。
本次審核不僅有益於病人的治療，還加強了診所工作人員之間的交流，使個人更好地瞭解了審計的性質，讓全體人員認識到通過審計可以提高醫療服務質量。而且審計有助於在診所員工中培養一種好的文化氛圍，努力發現診所工作的不足，並找出系統而實用的方法加以解決。

主要詞彙：激素替代療法，基層醫療，實證審核

---

Introduction

Audit is the process of critically and systematically assessing our own professional activities with a commitment towards improving personal performance and ultimately, the quality and/or cost-effectiveness of patient care.¹ Many studies have shown that audit with feedback has resulted in improved performance.² Audit can be viewed as a framework to implement quality into practice.³

The field of menopause management is changing rapidly. The pros and cons of all treatments, including lifestyle changes and medication must be explained to potential users in order to let them make an informed choice of treatment based on the best current available evidence.

The average age of menopause worldwide is 51 years and the average life expectancy after menopause is now 30 years. Vasomotor symptoms occur in 50-85% of Caucasian women at the time of menopause. In addition to the vasomotor symptoms, about 45% of women over the age of 60 have symptoms of urogenital atrophy.⁴

Endogenous oestrogen declines at menopause and there is an associated increase in osteoporosis, bone fracture and cardiovascular disease. Concerning some local figures for osteoporosis and bone fracture, between 1960 and 1985, the incidence of hip fracture in elderly Chinese was more than doubled from 153 to 353 per 100 000 population.⁵ In other words, the fracture rates became closer to those found in other developed countries. The estimated incidence for women aged over 80 years was 1 521/100 000 per year and was 32/100 000 in those aged 50-59 years.⁶ Similar to osteoporosis and bone fracture, deaths due to coronary artery disease (CAD) increased with advancing age. Over the 20 years period, the crude death rate due to CAD in Hong Kong was almost doubled, from 28.4/100 000 population in 1972 to 54.7/100 000 in 1992.⁷ This occurred in association with a longer life expectancy as well as the development of a more atherogenic lipid profile. The problem of cardiovascular disease appeared to be increasing, with the ratio of deaths from heart diseases compared with cancer rising from 0.65 in 1985 to 0.87 in 1994 in women and from 0.45 to 0.56 in men. In addition, women appeared to be more affected by the increased risk than men. The age-specific mortality rate from ischaemic heart disease increased from 402.30/100 000 women in the period 1979-1983 to 445.95/100 000 in the years 1989-1993.⁸

Despite the frequent symptoms and consequences encountered during and after menopause, Chinese women in Hong Kong have very little knowledge of the effects of menopause and the treatments available.⁹ The lack of awareness of the possible long-term complications of the treatment means that Chinese women in Hong Kong may also be less likely to seek advice about menopause and its treatment from a medical professional than Caucasian women.

On the other hand, family physicians (FP) should have a good understanding of the physiology of the menopause transition. A FP requires a supply of up-to-date, balanced and non-promotional literature, with the knowledge, willingness and skills to explain it. The current evidence should be understood and the situation should be explained to the patients. Moreover, FPs should be able to offer options for management, to understand the positive and negative implications of each and to counsel the patients through the choices put before the menopausal women.

The author's clinic is one of the primary care clinics of Department of Health which provides service to civil servants, their dependents and pensioners. It is also an accredited community based training centre in Family Medicine. There is a considerable number of patients receiving hormone replacement therapy (HRT). Before the audit, there were approximately 120 patients receiving regular HRT. Most of them started taking HRT as ordered by the hospital specialists. The patients were then referred to the primary care clinic for continuing the prescriptions. The duration of HRT use varied from a few months to more than 2 decades.

Due to changing evidences regarding the use of HRT, many patients receiving the HRT did not understand the updated information of the therapy; for example, the indications, contraindications, risks and benefits. According to a local study regarding the use of HRT, the 5-year compliance rate was 71.1%. "Doctors' advice" was the most important reason for maintenance of long-term compliance, accounting for 88.0% of all compliant cases. The commonest reason for non-compliance was the experience of side effects. Other important reasons were fear of cancer, fear of other side effects, and discouragement from other doctors. Overall, 16.0% of women expressed concern regarding potential side effects or risk of breast cancer. Younger age at menopause and at time of initial consultation, shorter duration of menopause at presentation and previous history of hysterectomy were associated with a higher degree of compliance at 5 years. Presence or absence of menopausal symptoms, on the other hand, was not a significant factor.¹⁰

As there was no evidence based guideline/protocol on the prescription of HRT in the clinic, it was a good opportunity to review literatures systematically and collaborate with other medical colleagues in the evidence based medicine group (EBM group) to develop a tailor-made practical guideline/protocol through the audit process. The EBM group is formed by a team of primary care colleagues working in the Department of Health who are interested in promoting EBM practice. Most of the members, who have attended relevant EBM workshops overseas, are experienced in EBM. With the help of EBM group, there was an extensive literature search on the subject of HRT in primary care. Many of them were high quality randomized controlled trials of interventions, systematic review and evidence based guidelines. In order to prepare for this audit project, several important national guidelines were searched using the text terms "hormone replacement therapy" and "menopause". The major ones included National Guideline Clearinghouse, NeLH guideline finders, and local guideline from the Hong Kong College of Obstetricians and Gynaecologists. Besides the guidelines, secondary pre-appraised literature such as Cochrane, Best evidence, Clinical evidence were searched. In order to ensure a comprehensive search, Medline was also browsed for high quality primary literatures using the keywords search and MeSH terms search.

Objectives

• To audit the process of HRT prescription in a primary care clinic.
• To define criteria and standards of care which are supported by the current best evidence.
• To identify areas of deficiencies in the process of HRT prescription.
• To identify strategies for improvement and implementation of changes.
• To evaluate performance after the changes.
• To provide a platform for further audit and study on process and outcome of HRT prescription in the clinic.

Method

1. Setting criteria and standard

Setting criteria

For audit of the process on prescription of HRT, the criteria were developed from literature review from international and local evidence based guidelines, secondary pre-appraised articles from Cochrane, Best evidence and Clinical evidence and primary literatures from Medline. All of them were classified as "must do" criteria due to the abundant supporting evidence and important impact on patient care.

The following criteria were identified:

For all women on HRT who have attended my clinic for HRT prescription within the recent 12 months.

Criterion 1

The records show whether a woman have an intact uterus.

Justification: Women with an intact uterus experience an increased risk of irregular bleeding, endometrial hyperplasia and endometrial cancer if they use unopposed oestrogens.¹¹ A meta-analysis of 30 observational studies demonstrated an overall relative risk of endometrial cancer of 2.3 for oestrogen users compared with non-users, rising to 9.5 after 10 years of use.¹²

Continuous or cyclical (for 12 days each month) progestogen used in conjunction with the oestrogens eliminated the increased risk.^13,14 Taking progestogen for only 10 days each cycle reduced but did not eliminate the increased risk.¹⁵ Oestradiol implants necessitated the use of long term progestogens.¹⁶

Criterion 2

The records show the indications for hormone replacement therapy:

Patients suffering from vasomotor symptoms at/after menopause could be offered oral HRT.

Justification: Oral HRT was highly effective in alleviating hot flushes and night sweats.¹⁷ HRT was effective for the relief of vasomotor symptoms, insomnia and urogenital atrophy associated with menopause.^18-20

Criterion 3

The records show that absolute contraindications are assessed, which include existing breast cancer, acute liver disease, venous thrombosis and existing endometrial cancer.

Justification: The absolute contraindications to the use of HRT included existing breast carcinoma, existing endometrial carcinoma, venous thrombosis and acute liver disease.²¹ This local guideline from the Hong Kong College of Obstetricians and Gynaecologists was written with reference to quality evidence and this was accepted by the EBM group as well as the clinic staff.

Criterion 4

The records show that women with osteoporosis should not be offered HRT as first line treatment.

Justification: Million Women Study showed an increase in fatal disease with a relative risk of 1.22 of death from breast cancer over controls. Current users of oestrogen only had a relative risk of breast cancer of 1.30, whereas the combined oestrogen-progestogen preparations had a relative risk of 2.00.²² The oestrogen-progestogen arm of the WHI study was stopped because the group reached the limit on excess risk of breast cancer, however the oestrogen alone arm of the trial had no significant increased risk of breast cancer and it was later stopped due to increased risk of stroke.²³ The Committee on Safety of Medicines concluded that HRT should no longer be recommended as first line therapy for preventing osteoporosis, due to the increased breast cancer risk on the long term treatment required to have a lasting effect on bone metabolism.²⁴

Given the balance of harms and benefits, the Canadian Task Force on Preventive Health Care recommended against the use of combined oestrogen-progestin therapy and oestrogen-only therapy for the primary prevention of chronic diseases such as osteoporosis or cardiovascular disease/cerebrovascular disease in menopausal women.²⁵

Criterion 5

The records show that patients on HRT should be advised on regular 3-yearly mammography screening.

Justification: International data based on 52 705 breast cancer patients and 108 411 controls concluded that the risk of breast cancer increased with duration of use of HRT.²¹ This excess risk reduced when therapy was withdrawn and disappeared 5 years after stopping HRT. Between the ages of 50 and 70 years the cumulative incidence of breast cancer was 45 per 1 000 in never-users of HRT. This risk was increased by 2 cases per 1 000 women after 5 years of use, by 6 cases per 1 000 after 10 years of use and by 12 cases per 1 000 after 15 years of use. Swedish studies confirmed an RR of between 1.4 and 2.43 after 10 years of HRT use.^27,28 Progestins did not reduce the risk and there was some evidence that they may actually increase it.²⁹

New data from the Million Women Study showed an increase in fatal disease with a relative risk of 1.22 of death from breast cancer over controls. Current users of oestrogen only had a relative risk of breast cancer of 1.30, whereas the combined oestrogen-progestogen preparations had a relative risk of 2.00.¹⁹ The oestrogen-progestogen arm of the WHI study was stopped because the group reached the limit on excess risk of breast cancer, however the oestrogen alone arm of the trial had no significant increased risk of breast cancer and it was later stopped due to increased risk of stroke.²⁰

Breast examination should be carried out if indicated by personal or family history. All women should be advised about breast awareness and be encouraged to participate in the breast cancer screening programme as appropriate for their age. Mammography had higher sensitivity and specificity than clinical examination.³⁰ 3-yearly mammogram screening was based on most updated screening recommendation by the National Screening Committee in the UK which was adopted by all staff in the clinic.

Criterion 6

The records show that women starting or continuing HRT should be counselled with regard to the perceived benefits and possible risks.

Justification: Early studies demonstrated accelerated blood clotting in women on conjugated equine oestrogen ^31,32 and an increased risk of thromboembolism in women currently taking HRT was found in 1996.^33-36 A relative risk of venous thromboembolism in women on HRT between 2.1 and 3.6 has been reported in case controlled studies.^33-38 The Committee on Safety of Medicines advised that the baseline risk of thromboembolism for non-users of HRT is 1 per 10 000 per year which increased to 3 per 10 000 per year for current users of HRT.³⁹

In the Women's Health Initiative study, the primary outcome measure was coronary heart disease. After 5.2 years of follow-up, the trial was stopped as there was an increased risk of coronary heart disease (hazard ratio 1.29, 95% CI 1.02-1.63), stroke (hazard ratio 1.41, 95%CI 1.07-1.85) and breast cancer (hazard ratio 1.26, 95%CI 1.0-1.59) as well as pulmonary embolism.²⁰

For women who wished to alleviate menopausal symptoms using HRT, a discussion between the woman and her physician about the potential benefits and risks of HRT was warranted.²⁵

All women should be reviewed at least annually, as the risks and benefits of HRT for each individual woman would alter with time, and needed to be discussed on an annual basis.⁴⁰

Standard setting

In view of no similar local or overseas study and audit being done before, no reference could be made to compare for standard setting. After discussing with my colleagues on the above issues, a clinic consensus was reached. The clinic consensus was based on the quality of the evidence found and the perception of the impact to the quality of care in HRT prescription. All agreed that 95% is a reasonable standard to be achieved as some patients may default at follow up. The only exception is the standard on criteria 3 in which the absolute contraindication must be documented in all patients, that is 100%. Table 1 summarizes the standard for each criterion:

 

2. Data collection and analysis

First phase evaluation

In the first data collection, all patients who attended the clinic for HRT were identified and recruited. 4 months interval (August 2004 - November 2004) was used because all HRT users would be followed up in the clinic every 3 months. To ensure all HRT patients were included in this audit, a full list of patients on HRT attending the clinic was obtained from the computer registry in the Pharmacy. As there was no HRT registry in the clinic, all patients who attended the clinic for HRT within the first data collection period were identified using the computer registry in the Pharmacy. The records were then retrieved manually in the record office or consultation rooms after everyday consultation during the 4 months period (August 2004 - November 2004) and these were reviewed by the HRT coordinator. There were 124 clients recruited in the first phase of evaluation. Information concerning the prescription practice of HRT was retrieved mainly from the front page summary and progress note of the medical records. The relevant data was entered into SPSS programme by the HRT coordinator.

Second phase evaluation

HRT registry was set up in November 2004 and all the relevant information of patients who attended the clinic was then recorded in the HRT registry. In the second data collection period (end of July 2005), all patients who attended the clinic for HRT were recruited again using the HRT registry. There would be no sampling bias as all the patients on HRT within the data collection period were documented and assessed. A list of HRT users who attended the clinic was retrieved from the HRT registry. There were 93 clients recruited in the second phase evaluation. Again, the records were then retrieved manually in the record office during the second phase data collection period (end of July 2005) and these were reviewed by the HRT coordinator. Information concerning the prescription practice of HRT was retrieved mainly from the annual checklist for HRT as shown in Appendix A. The relevant data was then entered into SPSS programme.

Statistical analysis

All the data were entered and analyzed by Statistical Package for the Social Sciences version 10.0 (SPSS Inc, Chicago [IL], United States). The results of first phase and second phase were compared for statistically significant difference.

 

3. Identification of deficiencies and corresponding changes

A formal clinic meeting was held in March 2005. The results of first phase evaluation were discussed and comments from all clinic staff were collected. A summary of areas of deficiencies and corresponding implemented strategies is shown in Table 2.

Results

Please see Table 3 and Table 4 below.

Characteristics of HRT users

More than half of the HRT users were aged between 50 years old and 59 years old. This is an expected finding as most of the women enter into menopause at around 50 years of age and after they are prescribed the HRT, they will be referred to the primary care clinic for further management under the current policy in the specialty clinics.

Discussion

Improvements in the standards of care

The results of process performance in the two phases were shown in Table 4. 31 patient stopped HRT after the initiation of this audit. All the performance in the second phase reached the standard. Statistical test was also performed to assess for statistical significance. All the results were statistically significant.

Impact on patient care

The quality of care in HRT prescription could be improved through direct and indirect effects of the audit.

a) Direct effect

Process of HRT prescription

This was an important aspect of this audit. There was definite improvement in the standard of the process of HRT prescription after the implementation of changes. The management of HRT users was standardized and HRT users had adequate opportunity to decide whether they should continue the HRT. Moreover, not only the HRT users had better understanding of the current evidence, but also the medical colleagues, nursing colleagues and other staff in the clinic got more comprehensive and updated knowledge.

From the patients' perspective, informal discussion showed more satisfactory use of HRT. Due to better understanding of the current evidence, patients would continue the medication only when indicated (for vasomotor symptoms) and were more compliant with the medication because they were less worried about the side effects. Moreover, after they understood the implication of menopause on their health, they would be more aware of the healthy lifestyles related to the reduction of osteoporosis, bone fracture and cardiovascular disease.

31 patients stopped taking HRT in this audit. The exact reasons need to be studied in further details but this may be due to the fear of the side effects of long term HRT use. Further research on the underlying reasons is required. However, informal interview revealed that the HRT quitters were most concerned about the risk of breast cancer before they were informed of the new evidence. After knowing the current evidence on HRT, some of them were even more anxious about the other side effects such as the increase in cerebrovascular and cardiovascular risk in the WHI study. In addition, other quitters chose alternative treatment like bisphosphonates for osteoporosis.

From the clinic staff's perspective, informal meeting showed that all staff realized the advantage of streamlined process in HRT prescription. Every member of staff was more well-equipped and confident in counselling the HRT users.

b) Indirect effect

Medical staff

Better team work

After identifying the deficiencies in HRT prescription in my clinic, namely, (i) lack of regular review to monitor the process of HRT prescription, (ii) lack of standard protocol/flowchart, (iii) lack of reminder/recall system, (iv) lack of awareness on referral sources for mammogram or osteoporosis, (v) lack of structured, standard material and lack of regular training and education, clear definition of the roles of each member of clinic staff, better understanding of updated evidence on HRT and regular clinic meetings there was better utilization and distribution of manpower in my clinic.

Improved education and training on current HRT evidences

Regular updated activities were held with the help by the EBM group, as a result, all the staff took a more active role in counselling on use of HRT after they were better equipped with updated knowledge and skills on HRT counselling.

Better use of resources

A detailed annual checklist, information leaflet, and user-friendly protocol on the process of HRT prescription saved a lot of time when performing counselling in the consultation. Regular health talks and poster exhibition were alternative means of delivering the current updated information about HRT to those in need.

Promotion of positive culture towards audit and research in the clinic

Active participation of clinic staff in this audit allows them to have better understanding of clinical audit and research. It will help to diminish the myth of interpreting audit as a form of blaming activity, which some of our staffs were worried about.

The clinic

Strategies like development of evidence based HRT audit criteria, implementation of audit protocol and annual checklist, use of computer registry, establishment of referral sources for mammogram or osteoporosis, regular health talks and literature search by the EBM group enhanced the quality of HRT prescription in my clinic. Eventually, patient care could be improved through the change of practice.

Conclusion

Clinical audit on process performance of HRT prescription was completed systematically. With the rapidly changing evidence on use of HRT, all staff in my clinic agreed to perform this audit in view of the huge impact on health care and potential for improvement.

In this audit, 6 major key elements of care of HRT prescription were identified. Evidence based audit criteria were made after extensive literature search on latest evidence and international guidelines. All HRT users who attended the clinic were recruited. Relevant data were collected and analyzed. First phase evaluation was performed from August 2004 to November 2004 and several areas of deficiency were noticed. Formal and informal discussions were held among all the clinic staff.

Strategies for improvement were suggested and implemented with clinic consensus. After the period of implementation phase (April 2005 - July 2005), second phase evaluation was performed in late July 2005. All the results of process performance showed statistically significant improvement and had met the standards. Further studies and audits were necessary to measure the continuity of the process performance and outcome of HRT prescription.

The impact of this audit on patient care is positive. The process of HRT prescription was improved so that the HRT users and providers would be more certain about the current evidence and hence decide whether she should continue the HRT therapy.

The benefit of this audit was not only limited to patient care. It provided opportunity to enhance communication among staff in the clinic. Clinic staff had better understanding of the nature of audit and all realized that quality of care could be improved through this exercise. It also facilitated a positive culture among clinic staff so that they would try to identify deficiencies of the clinic and find practical methods to tackle them in a systematic way.

In summary, the main aims and objectives of this audit are met and it is agreed that continuous audit cycles are necessary to safeguard the quality of care in my clinic.

Acknowledgement

I would like to give my sincere thank to Dr. Luke Tsang for his support in preparing this manuscript. I would also like to thank the clinic staff and the Evidence based medicine group of the Professional Development and Quality Assurance unit, Department of Health for continuous support throughout the audit cycle.

Key messages

1. The field of menopause management is changing rapidly. The pros and cons of all treatments, including lifestyle changes and medication must be explained to potential users
2. Family physician requires a supply of up-to-date, balanced and non-promotional literature, with the knowledge, willingness and skills to explain the current evidence on hormone replacement therapy.
3. In this audit, 6 major key elements of prescribing HRT were identified.
4. The process of HRT prescription was improved so that the HRT users and providers would be more certain about the current evidence and hence decide whether the patients should continue HRT therapy.

---

---

References