Obstetrics & Gynecology Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Obstetrics & Gynecology 2000;95:507-512
© 2000 by The American College of Obstetricians and Gynecologists
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by TONKELAAR, I. D.
Right arrow Articles by ODDENS, B. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by TONKELAAR, I. D.
Right arrow Articles by ODDENS, B. J.

ORIGINAL RESEARCH

Determinants of Long-term Hormone Replacement Therapy and Reasons for Early Discontinuation

ISOLDE DEN TONKELAAR, PhD and BJÖRN J. ODDENS, MD, MSc, PhD

From the International Health Foundation, Utrecht, The Netherlands.

Address reprint requests to: Isolde den Tonkelaar, PhD, International Health Foundation, Europalaan 506, Utrecht 3526 KS, The Netherlands, E-mail: identonkelaar{at}ihf.nl


   Abstract
Top
Abstract
Methods
 Results
 Discussion
 References
 
Objective: To identify factors associated with long-term hormone replacement therapy (HRT) and reasons for early discontinuation of it.

Methods: A cross-sectional study was conducted in four United Kingdom group general practices. Six hundred fifteen past or present HRT users (representing a response rate of 66%) responded to questionnaires on HRT and potential determinants of long-term use. Main outcome measures were long-term HRT use (at least 6 years) as opposed to short-term use (at most 2 years) and self-reported reasons for discontinuation. Odds ratios (ORs) of long-term use were adjusted for age and other variables, in the same groups, calculated by logistic regression and 95% confidence intervals (CIs).

Results: Ovariectomy (OR 2.59, 95% CI 1.12, 5.97), hysterectomy (OR 2.28, 95% CI 1.37, 3.79), previous oral contraceptive use (OR 1.76, 95% CI 1.03, 3.01), HRT prescription to prevent osteoporosis (OR 1.81, 95% CI 1.04, 3.13), opinion that HRT prevents health problems (OR 3.22, 95% CI 1.57, 6.63), opinion that HRT is associated with health risks (OR 0.23, 95% CI 0.08, 0.65), and opinion that HRT has cosmetic benefits (OR 2.52, 95% CI 1.45, 4.40) were statistically significantly associated with long-term HRT. Women surveyed most often reported side effects and weight gain (each about 30%) as reasons for discontinuation, followed by possible health risks and dislike of menstrual bleeding or hormones (each about 15%).

Conclusion: Ovariectomy, hysterectomy, and opinions about benefits and disadvantages of HRT were the most important determinants of long-term use, whereas women themselves mentioned side effects and weight gain most frequently as reasons for discontinuing it.

Long-term hormone replacement therapy (HRT) has been advocated for treatment of osteoporosis, cardiovascular disease, and urogenital atrophy.1,2 However, several studies have indicated that HRT is often discontinued early.3,4 A recent Swedish study found that long-term users were more likely to have had hysterectomies or ovariectomies and low parity, older age at first birth, high educational level, and low body mass index (BMI).5 Previous studies6,7–13 suggested that dislike of withdrawal bleeding or irregular bleeding and concerns about health risks were the most common reasons why women discontinued HRT early, but those samples were generally small. The aim of the current cross-sectional study was to investigate which characteristics are associated with long-term use (at least 6 years), with special emphasis on women’s opinions about HRT, on bleeding induced by HRT, and on factors described as positive or negative determinants of HRT prescription, such as concern about developing osteoporosis and breast cancer.11 The reasons for discontinuation of HRT use women mentioned themselves also were examined to gain better understanding of the reasons for early discontinuation.


   Methods
Top
Abstract
 Methods
Results
 Discussion
 References
 
Four group general practices in the United Kingdom with computerized patient administration systems, based in Huntingdon, Godmanchester, Boston, and East Molesey, participated in the study. None of those practices ran menopause clinics or specialized menopause services. Patients of one individual general practitioner who had a research interest in menopause-related problems were excluded. Women who had used HRT (at least one prescription during the previous 3 years) were identified from patient records. The project started in the first practice in September 1995 and ended in the last practice in January 1997. Nine hundred thirty-eight women were sent a questionnaire with an accompanying letter from their doctor introducing this International Health Foundation study as a survey on women’s health and specific problems women might have around menopause and afterward. Questionnaires were returned by 645 women, an initial response rate of 68.7%.

Thirteen women pretested the questionnaire for comprehensibility. It took approximately 20 minutes to complete and included 36 questions, of which 27 (75%) were used in the current analysis. The complete questionnaire is available on request from the authors. The study was approved by the local ethics committees.

In the statistical analyses, reasons for early discontinuation were studied as proportions, whereas associations between factors and long-term HRT were studied by contrasting long-term users (6 years or more) with those who had discontinued use within 2 years of starting it. Statistical analyses were done using SPSS for Windows, release 8.0.0, 1997 (SPSS Inc., Chicago, IL). Multiple logistic regression analyses were done for clusters of variables (eg, demographic variables, lifestyle variables, variables related to health of persons in immediate circle) to calculate odds ratios (ORs) and 95% confidence intervals (CI). Odds ratios were adjusted for age and for other variables in the same cluster because those factors were the most likely confounders. A predictive logistic regression model was built by entering simultaneously all variables that were associated with long-term use in univariate analysis at a level of statistical significance of P < .10 and by deleting variables by the stepwise backwards method. Variation in long-term user status explained by the variables was expressed by Nagelkerke’s R2.


   Results
Top
Abstract
 Methods
 Results
Discussion
 References
 
The mean age of respondents was 53.2 years (standard deviation [SD] 6.67, median 53, range 30–75 years). Of 645 women included in the sample, 18 reported that they had never been prescribed HRT, 12 had never started the therapy, 439 were current users, and 176 were past users. The reasons for prescriptions are given in Table 1Go. Of 615 women who started HRT, four did not report duration of use and another four did not report their year of birth. Of 611 women with known duration of use, 79 (12.9%) had used it for less than 6 months, 85 (13.9%) for 6–12 months, 125 (20.5%) for 1–2 years, 174 (28.5%) for 3–5 years, 120 (19.6%) for 6–10 years, and 28 (4.6%) for more than 10 years.


View this table:
[in this window]
[in a new window]
  		Table 1. Reasons for Prescribing Hormone Replacement Therapy*
 
Table 2Go shows the frequency of possible determinants of long-term HRT in long-term users (at least 6 years) (n = 148) opposed to short-term users (at most 2 years and stopped) (n = 131) and the associated ORs adjusted for age and other variables in the same item group. Women who had hysterectomies or ovariectomies were significantly more likely to be long-term users (P = .002 and P = .03, respectively). Women who believed that HRT prevents osteoporosis or cardiovascular disease, and women who believed that HRT prevents skin aging and keeps hair and nails in good condition were also statistically significantly more likely to be long-term HRT users (P = .002 and P = .001, respectively), whereas women who believed that HRT causes cancer or cardiovascular disease were statistically significantly less likely to be long-term users (P = .006). Age, previous oral contraceptive (OC) use, and prescription for prevention of bone loss, heart disease, or other health problems were also significantly associated with long-term use (P = .002, P = .04, and P = .04, respectively). Short-term users were somewhat younger (mean age 53.3) than long-term users (mean age 55.9) (t test, P = .001). Additional adjustment of the ORs for clinic and OC use resulted in similar odds ratios.


View this table:
[in this window]
[in a new window]
  		Table 2. Possible Determinants of Long-term (>=6 years) Hormone Replacement Therapy (n = 279)
 
Women who had stopped menstruating spontaneously and in whom regular bleeding was induced by HRT were significantly more likely to be long-term users than women in whom HRT did not induce bleeding. However, that association appeared to be partially confounded by the year in which HRT was started. The proportion of women who reported no induced bleeding increased from 20% when HRT was started in 1976–1988 to 50% when HRT was started in 1993–1995. Regular bleeding decreased from 50% in 1976–1988 to 30% in 1993–1995. After adjustment for the year in which HRT was started, instead of for age, the OR for regular bleeding was smaller and no longer significant (OR 2.7; 95% CI 0.80, 9.07), whereas the negative association became stronger for irregular bleeding (OR 0.20; 95% CI 0.04, 1.06).

In the predictive model, only age group, nulliparity, bilateral ovariectomy, and opinions about prevention, fears, side effects, and cosmetic aspects remained in the model, whereas educational level, previous OC use, and prescription for prevention of osteoporosis were not retained. Hysterectomy was not included in the model because of overlap with ovariectomy. Together those variables explained about 30% of the variation in long-term user status. An alternative model including age group and all opinions on hormone treatment (prevention, relief of complaints, fears, side effects, and cosmetic aspects) explained 27% of the variation in long-term user status.

Table 3Go shows the reasons for discontinuation. One hundred thirty-one of 176 past users stopped within 2 years (74.4%), 33 (18.8%) had stopped after 3–5 years of use, and 12 (6.8%) after more than 5 years of use. Women who ever used OCs and women who never used OCs were equally likely to discontinue HRT because of dislike of hormones (15.2% and 15.7%, respectively).


View this table:
[in this window]
[in a new window]
  		Table 3. Reasons for Discontinuation of Hormone Replacement Therapy
 

   Discussion
Top
Abstract
 Methods
 Results
 Discussion
References
 
Most HRT users started because of menopausal symptoms, and one in three for osteoporosis prevention, which agrees with previous findings.14 The most important determinants of long-term HRT were ovariectomy or hysterectomy; opinions on HRT; prescription for prevention of osteoporosis, heart disease, or other health problems; previous OC use; and probably also nulliparity and high educational level. The finding that a history of hysterectomy or ovariectomy was strongly associated with long-term use, as in a previous study,5 was not surprising because women who have bilateral ovariectomy before age 40 have evident indications for long-term HRT, and they do not experience the negative side effect of withdrawal (or irregular) bleeding. Among opinions on advantages and disadvantages of HRT, opinions about cosmetic advantages were strongly associated with long-term use. HRT might prevent dry skin and skin wrinkling.15,16 Prescriptions for preventing osteoporosis, heart disease, or other health problems concerned mainly osteoporosis prevention (which accounted for 92% of such prescriptions). We did not expect to find that women who were more concerned about developing osteoporosis were not more likely to be long-term users. This finding might be explained by the fact that women might be less concerned about osteoporosis once they use HRT for a long time. The lack of significant association between diagnosis of low bone mass and long-term use was probably due to small numbers (only eight short-term users and 16 long-term users had ever had bone densitometry). The association of OC use with long-term HRT could not be explained by less dislike of hormones in women who ever used OCs compared with women who never used OCs. Our findings on higher educational level and nulliparity agree with those of a large Swedish study, although our results were not statistically significant.5 Women of high educational level might have been more aware of the risks of osteoporosis and were probably more likely to comply with their own decisions. The positive association between nulliparity and long-term use has not been explained. Physical exercise, which was related to continued HRT in some9,11,17 but not all5 previous studies, was not associated with long-term use in the present study.

Our finding that regular induced bleeding was associated with long-term use after adjustment for age in women who had spontaneously stopped menstruating before HRT use appeared to be confounded by year in which HRT was first prescribed. The proportion of women who reported no induced bleeding increased with year in which HRT had been prescribed, probably indicating increased popularity of regimens not associated with withdrawal bleeding. Women who recently started HRT (and had higher probability of no induced bleeding) had not yet had the chance to become long-term users, which might explain that regular bleeding was no longer associated with long-term use after adjustment for year in which HRT was started.

When we investigated the reasons that women gave for discontinuation of HRT, some additional factors emerged, ie, side effects (headaches, painful breasts, and sickness) and weight gain. However, in placebo-controlled studies, HRT did not increase body weight.18 Although in some smaller studies,7–9,19 and in a recent larger study in Turkey,13 unexpected bleeding was a major reason for discontinuation of HRT, only 15% of our respondents mentioned dislike of menstrual bleeding (regular or irregular) or painful menstrual bleeding as reasons for discontinuing HRT. Our results in that respect agreed with data from more recent and larger studies.10–12 Our finding that only one in five women took the initiative to discontinue HRT on the advice of their physicians agreed with studies in which physicians in general would have favored longer term HRT than was achieved.20

Women who discontinued HRT were preferentially selected to contrast short-term and long-term use and to analyze the reasons for discontinuation, so the proportion of those who stopped HRT was not representative of the generalizable population. That is no reason to suspect selection bias in the associations between the characteristics of those women and long-term use.

Monitoring of weight gain and side effects might prevent early HRT discontinuation. Counseling a positive attitude towards long-term HRT, in particular among women without hysterectomies or ovariectomies, might motivate users to continue HRT for longer.


   Footnotes
 
The authors thank Dr. M. W. Whitton from the Charles Hicks Centre, Huntingdon, and Roman Gate Surgery, Godmanchester (UK), Dr. M. R. Walling and Mr. L. Viner from Parkside Surgery, Boston (UK), and Dr. S. Britton and Dr. C. Brant from the Glenlyn Medical Centre, East Molesey (UK), for identification of eligible patients from their practice registration and for mailing and receipt of the questionnaires.

Supported financially by the International Health Foundation, Geneva, Switzerland.

PII S0029-7844(99)00586-4

Received May 3, 1999. Received in revised form September 28, 1999. Accepted October 1, 1999.


   References
Top
Abstract
 Methods
 Results
 Discussion
 References
 
1. Barrett-Connor E. Hormone replacement therapy. BMJ 1998;317: 457–61.[Free Full Text]

2. Calaf I, Alsina J. Benefits of hormone replacement therapy—overview and update. Int J Fertil 1997;42(Suppl 2):329–46.

3. Ahmed AIH, Ryan PJ, Snelling T, Blake GM, Rymer J, Fogelman I. Long term compliance with hormone replacement treatment following screening for postmenopausal osteoporosis by bone density measurements. J Obstet Gynaecol 1996;16:41–4.

4. Berman RS, Epstein RS, Lydick EG. Compliance of women in taking estrogen replacement therapy. J Womens Health 1996;5: 213–20.

5. Persson I, Bergkvist L, Lindgren C, Yuen J. Hormone replacement therapy and major risk factors for reproductive cancers, osteoporosis, and cardiovascular disease: Evidence of confounding by exposure characteristics. J Clin Epidemiol 1997;50:611–6.[Medline]

6. Køster A. Hormone replacement therapy: Use patterns in 51-year-old Danish women. Maturitas 1990;12:345–56.[Medline]

7. Wren BG, Brown L. Compliance with hormonal replacement therapy. Maturitas 1991;13:17–21.[Medline]

8. Ryan PJ, Harrison R, Blake GM, Fogelman I. Compliance with hormone replacement therapy (HRT) after screening for post menopausal osteoporosis. Br J Obstet Gynaecol 1992;99:325–8.[Medline]

9. Johannes CB, Crawford SL, Posner JG, McKinlay SM. Longitudinal patterns and correlates of hormone replacement therapy use in middle-aged women. Am J Epidemiol 1994;140:439–52.[Abstract/Free Full Text]

10. Collins A, Landgren BM. Psychosocial factors associated with the use of hormonal replacement therapy in a longitudinal follow-up of Swedish women. Maturitas 1997;28:1–9.[Medline]

11. Oddens BJ, Boulet MJ. Hormone replacement therapy among Danish women aged 45–65 years: Prevalence, determinants, and compliance. Obstet Gynecol 1997;90:269–77.[Abstract]

12. Stadberg E, Mattsson LÅ, Milsom I. Womens attitudes and knowledge about the climacteric period and its treatment. A Swedish population-based study. Maturitas 1997;27:109–16.[Medline]

13. Karakoç B, Erenus M. Compliance considerations with hormone replacement therapy. Menopause 1998;5:102–6.[Medline]

14. Newton KM, LaCroix AZ, Leveille SG, Rutter C, Keenan NL, Anderson LA. Women’s beliefs and decisions about hormone replacement therapy. J Womens Health 1997;6:459–65.[Medline]

15. Dunn LB, Damesyn M, Moore AA, Reuben DB, Greendale GA. Does estrogen prevent skin aging? Results from the First National Health and Nutrition Examination Survey (NHANES I) Arch Dermatol 1997;133:339–42.[Abstract]

16. Castelo-Branco C, Figueras F, Martinez de Osaba MJ, Vanrell JA. Facial wrinkling in postmenopausal women. Effects of smoking status and hormone replacement therapy. Maturitas 1998;29:75–86.[Medline]

17. Derby CA, Hume AL, MacFarland Barbour M, McPhillips JB, Lasater TM, Carleton RA. Correlates of postmenopausal estrogen use and trends through the 1980s in two southeastern New England communities. Am J Epidemiol 1993;137:1125–35.[Abstract/Free Full Text]

18. Espeland MA, Stefanick ML, Kritz-Silverstein D, Fineberg SE, Waclawiw MA, James MK, et al. Effect of postmenopausal hormone therapy on body weight and waist and hip girths. Postmenopausal Estrogen-Progestin Intervention Study Investigators. J Clin Endocrinol Metab 1997;82:1549–56.[Abstract/Free Full Text]

19. Nachtigall LE. Enhancing patient compliance with hormone replacement therapy at menopause. Obstet Gynecol 1990;75:77S–83S.

20. Topo P, Hemminki E, Uutela A. Women’s choice and physicians’ advice on use of menopausal and postmenopausal hormone therapy. Int J Health Sci 1993;4:101–9.




This article has been cited by other articles:


Home page
 Obstet GynecolHome page
J. A. Simon, C. Bouchard, A. Waldbaum, W. Utian, J. Zborowski, and M. C. Snabes
Low Dose of Transdermal Estradiol Gel for Treatment of Symptomatic Postmenopausal Women: A Randomized Controlled Trial
Obstet. Gynecol., March 1, 2007; 109(3): 588 - 596.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurol. Neurosurg. PsychiatryHome page
V W Henderson, K S Benke, R C Green, L A Cupples, L A Farrer, and for the MIRAGE Study Group
Postmenopausal hormone therapy and Alzheimer's disease risk: interaction with age
J. Neurol. Neurosurg. Psychiatry, January 1, 2005; 76(1): 103 - 105.
[Abstract] [Full Text] [PDF]

Home page
 CMAJHome page
S. A. Atkinson and W. E. Ward
Clinical nutrition: 2. The role of nutrition in the prevention and treatment of adult osteoporosis
Can. Med. Assoc. J., November 1, 2001; 165(11): 1511 - 1514.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by TONKELAAR, I. D.
Right arrow Articles by ODDENS, B. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by TONKELAAR, I. D.
Right arrow Articles by ODDENS, B. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS