Menopause: Scientific evidence changes prescribing practice—a comparison of the management of the climacteric and use of hormone replacement therapy among Swedish gynaecologists in 1996 and 2003

Authors


Dr L Thunell, Department of Obstetrics and Gynecology, University Hospital of Lund, S-221 85 Lund, Sweden.

Abstract

Objectives  To study changes in attitudes, knowledge and management strategies concerning hormone replacement therapy (HRT) among gynaecologists in Sweden.

Design  Comparative questionnaire study.

Setting  National survey.

Population  Practising gynaecologists.

Methods  In 1996, gynaecologists in Sweden (n= 1323) were invited to return a postal questionnaire concerning their attitudes, knowledge and management strategies concerning HRT. They were also asked about their own use of HRT. In 2003, a similar questionnaire was sent to practising gynaecologists (n= 1320) in Sweden.

Main outcome measures  Attitudes to and personal use of HRT.

Results  The response rate was 76% in 2003 when 11% of the gynaecologists thought that all women without contraindications should be offered HRT compared with 44% in 1996 and 89% found it difficult to evaluate pros and cons with HRT in a clinical situation (74% in 1996). More gynaecologists in 2003 believed that HRT increased the risk for breast cancer (95% vs 71%). Twenty-five percent in 2003 stated that risk factors for osteoporosis were absolute indications for HRT (60% in 1996). Current ischaemic heart disease was considered to be an indication for HRT by 7% in 2003 (60% in 1996). In 2003, current use of HRT was reported by 71% of female menopausal gynaecologists (88% in 1996).

Conclusions  Swedish gynaecologists were more cautious in their management strategies concerning HRT in 2003 compared with 1996, probably influenced by results from the Heart and Estrogen/Progestin Replacement Study (HERS) and Women's Health Initiative (WHI) studies. Current use of HRT was still high among female gynaecologists, although it had decreased since 1996.

Introduction

Climacteric medicine has been in focus during the last two decades and the advantages and disadvantages of hormone replacement therapy (HRT) have received considerable attention. The number of HRT users increased from the beginning of the eighties until the late 1990s.1–3 A large number of observational studies indicated that long-term use of oestrogens had prophylactic effects against coronary heart disease (CHD)4,5 and osteoporosis,6 which made HRT attractive to many women. Another possible explanation for the increased use of HRT may be an increased knowledge and interest in climacteric medicine among physicians and women.

Recently large randomised controlled trials such as the Women's Health Initiative trial (WHI)7 and the Heart and Estrogen/Progestin Replacement Study (HERS)8 could not find evidence for primary or secondary preventive effects of HRT on CHD. The collaborative study9 and the Million Women Study10 found an increased risk of breast cancer among HRT users. An increased risk of venous thromboembolism among HRT users was also reported.7,11 This information regarding side effects caused anxiety among women and the prevalence of HRT use has decreased during the last few years following the publication of these studies.12

Women with higher education and those who were current HRT users have been found to be better informed about risks and benefits of oestrogen therapy.13,14 In 1996, gynaecologists and general practitioners (GPs) in Sweden were asked in a postal questionnaire about their attitudes to and knowledge about HRT as well as their own use of HRT. In that study, 88% of the female, peri- or postmenopausal gynaecologists and 72% of female GPs were current users of HRT.15 The level of knowledge, regarding potential benefits and drawbacks of HRT, has been proposed to be of importance for the prevalence of HRT use. A high usage of HRT by women doctors has been reported from other countries as well.16–18

Large randomised controlled studies such as HERS8 and WHI7 have been published since the study of the Swedish gynaecologists attitudes in 1996 and data reported in those studies have probably influenced physicians in their management strategies with regard to HRT. The aim of this study was therefore to investigate changes in attitudes, knowledge and management strategies concerning HRT, as well as personal use of HRT among gynaecologists in Sweden, between 1996 and 2003.

Methods

In 1996 Swedish gynaecologists (n= 1323), both specialists and gynaecologists under training, were invited by letter to complete and return an enclosed questionnaire about their knowledge, attitudes and management strategies concerning HRT. The response rate was 80% (n= 1054). These results have been published.19 In 2003, a similar cross sectional study was performed among Swedish gynaecologists using the same design as in the study from 1996. Practising gynaecologists (n= 1320), both specialists and gynaecologists during their specialist training, were contacted and asked to answer similar questions as in the study from 1996. Addresses to the gynaecologists were obtained from the Swedish Society of Obstetrics and Gynecology (SFOG). The vast majority of gynaecologists were approached because almost every gynaecologists in Sweden, both specialists and gynaecologists during their specialist training, are members of the professional society, SFOG. The questionnaire consisted of 35 questions, including the following1: personal data such as age, sex, place of work, etc.,2 knowledge and attitudes about HRT3 management strategies concerning the climacteric period and4 when applicable the gynaecologists’ own or their partners’ use of HRT. If no reply was received, a reminder was sent within four months after the first questionnaire.

The accuracy of data entry was checked on an individual basis for each parameter in all subjects. Age was presented as mean [SD]. Differences between gynaecologists in 1996 and 2003 was analysed with a two-sample t test. A P value <0.05 was considered statistically significant. Spearman's rank correlation test was used when comparing gynaecologists older and younger than 50 years. The comparisons between male and female doctors were also made with Spearman's rank correlation test.

Results

The questionnaires were returned by 998 gynaecologists (76%) in 2003 and by 1054 gynaecologists (80%) in 1996. Characteristics of the responders in 1996 and in 2003 are given in 1Table 1. The mean age was 48.5 [8.4] in 1996 and 50.6 [10.6] in 2003. Fifty-six percent of the specialists were females in 2003 compared with 43% in 1996 (P < 0.001). The proportion of female non-specialists was numerically higher in 2003 compared with 1996, 80% versus 74% (NS).

Table 1.  Characteristics of the responding gynaecologists, 1996 (80%) and 2003 (76%)
 Gynaecologists 1996, n= 1054Gynaecologists 2003, n= 998
 n%n%
Gender
Male5805541041
Female4744558859
Specialist9729285586
Position
National health
 Hospital practice7377166968
 Outpatient clinic1051010711
Private practice2001920621

There were differences in the gynaecologists opinions in 2003 compared with 1996, concerning which women should be offered HRT. Eleven percent of the gynaecologists in 2003 thought that all women without contraindications should be offered HRT, compared with 44% in 1996 (P < 0.001). In 2003, 89% of the gynaecologists reported that they often or sometimes found it difficult to evaluate advantages and disadvantages of HRT when counselling, compared with 74% in 1996 (P < 0.001). More gynaecologists in 2003 compared with 1996, stated that the initiative to discuss whether to use HRT or not in the peri-and postmenopausal period came from the patients (48% vs 23%, P < 0.001). In 2003, the gynaecologists were asked if they informed about changes in lifestyle (smoking, exercise and diet) when they prescribed HRT. Older doctors (>50 years) informed to a higher extent about lifestyle factors compared with younger doctors (≤50 years, P < 0.001) (2Table 2).

Table 2.  The gynaecologists opinions in 2003 concerning information to women about lifestyle factors when prescribing HRT
 All gynaecologists, n= 993%Gynaecologists ≤50 years of age, n= 473%Gynaecologists >50 years of age, n= 484%
When you prescribe HRT, do you at the same time inform about lifestyle (diet, exercise and smoking)?
Yes, often575164
Sometimes363933
No, never7103

Different statements concerning HRT were made in the questionnaire and the gynaecologists could answer if they agreed, disagreed or did not know. More gynaecologists in 2003 compared with 1996 believed that HRT increased the risk of breast cancer (95% vs 71%, P < 0.001). Almost 100% of the gynaecologists on both occasions answered that HRT prevented osteoporosis. Fewer gynaecologists in 2003 compared with 1996 stated that HRT could prevent wrinkles (15% vs 32%, P < 0.001). Sixty-four percent in 2003 thought that HRT might be effective against depression compared with 82% in 1996 (P < 0.001). One percent of the gynaecologists in 2003 considered HRT to be a lifelong treatment compared with 8% in 1996 (P < 0.001).

The gynaecologists were asked how they handled different conditions in relation to prescription of HRT (3Table 3). Annoying hot flushes and night sweats were considered by 100% of the physicians to be an absolute or relative indication for HRT both in 2003 and 1996. Twenty-five percent of the gynaecologists in 2003 compared with 60% in 1996 (P < 0.001) answered that risk factors for osteoporosis were absolute indications for HRT. A history of venous thromboembolism was regarded to be an absolute or relative contraindication for HRT by 98% in 2003 and by 67% in 1996 (P < 0.001). Current ischaemic heart disease was considered an indication for HRT by 7% in 2003 compared with 60% in 1996 (P < 0.001). Ninety-four percent of the gynaecologists in 2003 answered that a history of radically treated breast cancer was a contraindication for HRT compared with 89% in 1996 (P < 0.001). Loss of libido was an absolute indication for HRT among 6% in 2003 compared with 16% in 1996 (P < 0.001).

Table 3.  The opinions of Swedish gynaecologists in 1996 and 2003 regarding indications and contraindications for HRT
 Absolute indication (%)Relative indication (%)Indifferent (%)Relative contraindication (%)Absolute contraindication (%)
 1996200319962003199620031996200319962003
Annoying hot flushes and night sweats78622238000000
Annoying changes in mood34216473260000
Loss of libido166717313210000
Risk factors for osteoporosis60254071040000
Prophylaxis against osteoporosis to women without risk factors63704923450211
History of venous thromboembolism1011121117505048
Current ischaemic heart disease15045716176551821
Risk factors for ischaemic heart disease2015910123415184
Prophylaxis against ischaemic heart disease to women without known risk factors70629297611015
History of radically treated breast cancer11228319567038

In 2003, differences between male and female gynaecologists with regard to prescribing HRT in specific clinical situations were analysed. More women doctors had the opinion that a history of radically treated breast cancer was a contraindication to HRT compared with male doctors (97% vs 90%, P < 0.001). There were no gender differences regarding the prescription of HRT to women with risk factors for osteoporosis, risk factors for ischaemic heart disease, previous ischaemic heart disease or current smoking (>15 cigarettes per day).

Most of the gynaecologists in 2003 (97%) prescribed oestrogen only to women who had had a hysterectomy performed, due to benign reasons, a figure that did not differ from 1996. Perimenopausal women with their last menstruation within two years were treated with oestrogens in combination with sequentially administered progestogens by 91% of the gynaecologists both in 2003 and 1996. Women with their menopause more than two years ago were recommended continuous combined HRT by 85% of the gynaecologists in 2003 compared with 75% in 1996 (P < 0.001).

In 2003, 61% of the gynaecologists recommended HRT for more than five years treatment of climacteric complaints (4Table 4). Doctors over 50 years of age recommended HRT for this reason for a longer time compared with doctors of ≤50 years (P < 0.001). Male gynaecologists recommended HRT for treatment of climacteric complaints for a longer time compared with female gynaecologists (P < 0.05). If HRT was prescribed as prophylaxis for osteoporosis, 88% of the gynaecologists in 2003 recommended that it should be taken for more than five years (Table 4). There were no differences between gynaecologists in the abovementioned age groups or between male or female gynaecologists concerning how long time HRT should be taken as prophylaxis for osteoporosis. Comparative figures were not available in 1996.

Table 4.  Opinions concerning duration of HRT expressed by Swedish gynaecologists in 2003
 <5 years (%)5–10 years (%)>10 years (%)Life long (%)
Which duration of HRT do you recommend if it is given for climacteric complaints?
All gynaecologists (n= 946)395533
Gynaecologists ≤50 years of age (n= 457)455221
Gynaecologists >50 years of age (n= 455)345844
Male gynaecologists (n= 391)365644
Female gynaecologists (n= 552)415522
Which duration of HRT do you recommend if it is given as prophylaxis against osteoporosis?
All gynaecologists (n = 868)12482614

Among the female gynaecologists in 2003, 24% answered that they suffered from climacteric complaints (20% in 1996) and 33% were postmenopausal (25% in 1996). In the group of female gynaecologists in 2003 who were either postmenopausal or had climacteric complaints, 71% were current users of HRT (88% in 1996, P < 0.001). Thirty-eight percent of partners to male gynaecologists had climacteric complaints and 53% were postmenopausal in 2003. Of those partners who were either postmenopausal or had climacteric complaints, 68% were current users of HRT (86% in 1996, P < 0.001) (1Fig. 1).

Figure 1.

Percentage use of HRT among female gynaecologists and partners to male gynaecologists, who were either postmenopausal and/or had climacteric complaints.

Discussion

Gynaecologists in Sweden were more cautious in their recommendations and management strategies concerning HRT in 2003 compared with 1996. Current use of HRT was still high among female gynaecologists and partners to male gynaecologists, but had decreased since 1996.15

Forty-four percent of the gynaecologists in 1996 thought that all women without contraindications should be offered HRT, compared with 11% in 2003, which showed a more hesitant attitude when counselling about HRT in 2003 com pared with 1996. Several studies have previously reported beneficial effects of HRT on the cardiovascular system4,5 and the skeleton.6 These observational reports might have encouraged women to use HRT 3,20 and they probably created positive attitudes towards such therapy among gynaecologists.19 Osteoporosis prophylaxis was found to be a reason for 20% of women to commence HRT while 31% chose to use HRT to increase wellbeing in a Swedish survey from 1998.20

In 2003, the gynaecologists also found it more difficult to evaluate advantages and disadvantages of HRT when counselling women compared with 1996. These changes may be due to studies recently published. The collaborative study reported an increased risk of breast cancer during HRT,9 and the risk for thromboembolic disorders was reported to increase among HRT users.7,11 Large randomised controlled studies such as the WHI7 and HERS8 could not find evidence for primary or secondary preventive effects on CHD in HRT users compared with non-users. However, interventional studies have shown beneficial effects of HRT on risk factors for CHD such as reduced levels of low density lipoprotein (LDL) and an increase in high density lipoprotein (HDL) concentrations.21 Favourable effects on glucose metabolism22 and reductions in the incidence of diabetes have been reported.23,24 In the WHI study, HRT reduced colorectal cancer and hip fractures.7 Accordingly, contradictory results have been published and the difficulties when counselling women in the climacteric period about HRT are obvious. It may also be difficult for each individual woman to decide whether or not to choose HRT.

Opinions about indications and contraindications for HRT use among physicians changed a lot between 1996 and 2003. In 1996, 60% considered risk factors for osteoporosis to be absolute indications for HRT, while 76% stated that prophylaxis for osteoporosis in women without risk factors was an indication. In 2003, these figures were dramatically reduced to 25% and 52%, respectively. Changes in attitudes may be due to the tentative side effects mentioned above and to the new guidelines for HRT, which were introduced in Europe in 2003.25 HRT should be used for relief of vasomotor symptoms but was no longer advocated first choice for osteoporosis prophylaxis or osteoporosis treatment. Current ischaemic heart disease was previously considered an indication for HRT by 60% of doctors but changed dramatically to 7% in 2003. These changes in attitudes are probably due to the results reported in the HERS8 and WHI7 studies. It is reasonable to believe that Swedish physicians are fairly well informed about the latest results published in the medical literature.

More physicians in 2003 (94%) compared with in 1996 (89%) had the opinion that radically treated breast cancer was a contraindication for HRT. However, in 2003 compared with 1996 only 38% versus 70% of the physicians considered it to be an absolute contraindication. These figures reflect a more liberal and individualised approach to the use of HRT, during the latter period, in women previously treated for breast cancer, suffering from severe climacteric complaints. Data in this study were collected before a randomised placebo-controlled trial investigating whether or not HRT following breast cancer treatment was safe was published (HABITS). That study was stopped due to a higher recurrence rate of breast cancer events in women given active therapy.26 Another Swedish study, the Stockholm Study, initiated at the same time as HABITS also investigated the recurrence rate of breast cancer during HRT. The Stockholm Study was closed because it was considered impossible to continue after the data from HABITS was published and not because of an increase in recurrence rate of breast cancer.27

In 2003, HRT use was still high among female gynaecologists (71%) and partners to male gynaecologists (68%), but compared with 1996 a significant decrease was recorded (88% and 86%, respectively).19 Attitudes and management strategies concerning recommendations with regard to patients had changed extensively but the gynaecologists personal use had changed only moderately. Previous studies in the USA and Europe have reported HRT use among physicians to vary between 41% and 78%.16–18 To our knowledge, there is only one study28 investigating gynaecologists attitudes, management strategies and personal use of HRT published after the publication of the HERS I and II29 and the first WHI report. In that study from Norway with a limited number of questions, a change to a more hesitant attitude concerning HRT prescription was found among Norwegian gynaecologists between 1997 and 2002. However, the figures for their personal use of HRT were almost unaltered in 2002 (74%) compared with 1997 (78%). HRT as secondary prevention to women with current ischaemic heart disease was considered an indication by 59% of the Norwegian gynaecologists in 1997 compared with 2% in 2002. These results were in accordance with our data.

The strength of this study was that the vast majority of gynaecologists were approached because almost every gynaecologist in Sweden is a member of the professional society, SFOG, which supported us with addresses. The high response rate of 76% made it possible to include a majority of practicing Swedish gynaecologists. This study also reflected the gynaecologists’ opinions about one year after publication of the WHI study. One limitation though was that some physicians are involved only in obstetrics on a daily basis, which might have influenced their knowledge about HRT. However, these circumstances were similar both in 1996 and 2003.

Conclusion

The results from the HERS and WHI studies have influenced attitudes to HRT among gynaecologists in Sweden. They were more cautious in their management strategies concerning HRT and their personal use of HRT had also decreased. This survey predated the publication of results from the Million Women Study. It might be of interest to study if the results from the Million Women Study have further affected management strategies among gynaecologists and influenced their willingness to use HRT.

Acknowledgements

We thank the Göteborg Medical Society for financial support and Björn Areskoug for his valuable assistance regarding statistical analysis. ▮

Ancillary