HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

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

Predictors of Difficulty When Discontinuing Postmenopausal Hormone Therapy

From the Department of Epidemiology and Biostatistics, University of California, San Francisco; Division of Research, Kaiser Permanente Medical Care Program, Oakland, California; and Dartmouth Medical School, Hanover, New Hampshire.

Address reprint requests to: Deborah Grady, MD, MPH, UCSF Women’s Health Clinical Research Center, 1635 Divisadero Street, Suite 600, San Francisco, CA 94115; E-mail: dgrady{at}itsa.ucsf.edu.

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
OBJECTIVE: To describe the experiences of postmenopausal women who try to stop hormone therapy and to identify characteristics associated with inability to stop.

METHODS: We conducted telephone interviews with 377 randomly selected female members of the Kaiser Foundation Health Plan, aged 50–69 years, who regularly used hormone therapy for at least 1 year before July 1, 2002 and had attempted to stop between July 2002 and March 2003.

RESULTS: Of the 377 women, 280 (74%) successfully stopped and 97 (26%) resumed taking hormone therapy. The major predictor of resuming hormone therapy use was the development of troublesome withdrawal symptoms (odds ratio 8.8; 95% confidence interval 4.9, 16.0). Report of hysterectomy, hormone therapy prescribed by a nongynecologist, and perception of high risk of hip or spine fracture were independently associated with a higher likelihood of unsuccessful stopping. Women with a hysterectomy who had used hormone therapy for 10 or more years and who started hormone therapy mainly for reasons other than health promotion were more likely (P < .001) to be unsuccessful in quitting (44%) compared with those with one or two (25%) or none (9%) of these three characteristics. Most successful stoppers (71%) stopped hormone therapy abruptly, but 29% tapered off hormone therapy; there was no difference in the incidence of troublesome withdrawal symptoms or successful quitting between these two groups.

CONCLUSION: Approximately one quarter of women who try to stop report that they are unable to discontinue postmenopausal hormone therapy, primarily because they develop troublesome withdrawal symptoms. Effective approaches to reducing hormone therapy withdrawal symptoms should be a priority for future research.

Large, randomized trials have recently shown that postmenopausal hormone therapy is associated with increased risk for coronary disease, stroke, breast cancer, venous thromboembolic events, dementia, and gallbladder disease.^1–3 These findings have led most experts to encourage postmenopausal women without vasomotor symptoms to stop hormone therapy and to limit use to short-term treatment for menopausal symptoms.^4,5

Reports indicate that sales of hormone therapy in the US decreased approximately 25–30% in the first few months after publication of the Women’s Health Initiative randomized trial primary findings in July 2002, which documented overall harm associated with hormone therapy.⁶ We recently interviewed hormone therapy users in a large health maintenance organization and found that 56% made the decision to stop in the 6–8 months after publication of the Women’s Health Initiative findings.⁷

To identify characteristics of women who were unable to stop using hormone therapy, we surveyed a random sample of Kaiser Foundation Health Plan members, aged 50–69 years, who had used hormone therapy regularly for at least 1 year. We hypothesized that having withdrawal symptoms on cessation of hormone therapy would force many women to resume taking it. We further reasoned that women tapering the estrogen dose would have a lower incidence of withdrawal problems and a higher probability of successful stopping than those stopping abruptly.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We used membership rolls and a computerized pharmacy database to identify all women aged 50–69 years who were members of Kaiser Permanente Medical Care Program of Northern California for at least 3 consecutive years before July 1, 2002.⁷ From the pharmacy database, we determined type of estrogen used (conjugated equine estrogens, micronized estradiol, or esterified estrogens), hormone regimen (estrogen or estrogen plus progestin), dose of estrogen (1.0 mg micronized estradiol and 0.625 mg esterified estrogens were considered equivalent to 0.625 mg conjugated equine estrogens), and the specialty of the hormone therapy prescriber (gynecologist, primary care, or other). We limited selection to women who had filled at least three prescriptions, each for 100 pills of estrogen, between July 1, 2001 and June 30, 2002. We excluded women using doses equivalent to more than 0.625 mg conjugated estrogens. We stratified this cohort of 35,275 eligible women by 5-year age groups (50–54, 55–59, 60–64, and 65–69 years) and randomly selected 250 from each stratum. We mailed these 1000 women a letter explaining the purpose of the study and a prepaid postcard that they could return to refuse further contact. If we did not receive a refusal card within 3 weeks, we attempted to contact the woman by telephone for an interview. The Kaiser Institutional Review Board approved the protocol and methods.

During the months of January, February, and March 2003, five trained interviewers contacted potential participants and carried out interviews lasting an average of 20 minutes. Information was obtained on demographics, height, weight, health habits and medical history. Perceived health status was obtained by asking women to rate their current health as excellent, very good, good, fair, or poor. Risk perceptions for being affected by breast cancer, colon cancer, coronary heart disease, stroke, venous thromboembolism, and hip or spine fracture (relative to an average woman’s risk) were assessed with a 5-point Likert scale (1 = much lower, 5 = much higher). The majority of these survey items have been validated and used in prior studies.^8,9

We also asked about sources of support for the decision to stop (spouse, children, friends, healthcare provider), approach to stopping (abruptly versus tapering), and symptoms that occurred after stopping. Among women who reported any symptoms associated with stopping hormone therapy, we elicited (without prompting) the main withdrawal symptom, the interval from stopping hormone therapy to onset of the main symptom, and the interval from onset to relief (complete relief or much improved) of this main symptom. We also asked specifically about the presence and severity of nine potentially menopause-related symptoms (hot flushes, excessive sweating, sleep disturbance, fatigue, mood swings or depression, headache, vaginal dryness, vaginal itching, and aches or pains). These nine symptoms were categorized as "vasomotor" (hot flushes and sweats) or "other." The degree of troublesomeness of withdrawal symptoms was assessed with a 5-point Likert scale (0 = none, 1 = a little, 2 = somewhat, 3 = very, 4 = extremely). Symptoms that were rated very or extremely troublesome were considered troublesome. Finally, we asked about the overall troublesomeness of all withdrawal symptoms taken together (global symptoms).

To assess each woman’s knowledge of the risks of hormone therapy, we summed correct responses to the following five true-or-false questions: 1) Using hormone therapy reduces risk of breast cancer (correct response, false), 2) Using hormone therapy reduces risk of heart attack (false), 3) Using hormone therapy reduces risk of stroke (false), 4) Using hormone therapy increases the risk of blood clots in the veins of the leg and lung (true), and 5) Overall, for the average woman, using hormone therapy improves health (false).

Women were categorized as "attempting to stop" if they reported that they had the intention to stop and either stopped or reduced the dose of hormone therapy for at least 1 week; those reducing the dose without the intention of stopping in the near future were not considered to be attempting to stop. Successful stopping was defined as not returning to the use of hormone therapy after stopping or not completely stopping after tapering.

We used ² or t tests, as appropriate, to compare the characteristics of those who successfully stopped with those who resumed therapy. When the expected number of subjects in any category was less than five, Fischer exact test was used. Variables found to be associated with unsuccessful stopping at P < .05 in univariate analyses were entered into a multivariable logistic regression model, with unsuccessful stopping as the outcome variable. To reflect the stratified random selection, we also included a variable for each age stratum in the multivariable model. Use of unopposed estrogen and hysterectomy were highly correlated; we elected to use only hysterectomy status in the multivariable model. We repeated the final multivariable model weighting each observation by the reciprocal of the sampling fractions. The weighted odds ratios and P values were very similar to the unweighted results. Thus, we present only unweighted findings.

To develop a predictive model to identify women who are likely to fail an attempt to stop hormone therapy, we used historical variables that were associated with unsuccessful stopping in univariate models. We reduced the number of variables in this model by removing variables that did not appreciably affect discrimination between successful and unsuccessful quitters. All statistical analyses were performed with SAS 8.2 (SAS Institute, Cary, NC).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Of the 1000 women randomly selected, 670 completed the interview.⁷ Among the 230 women who refused to be interviewed, we determined from the computerized database that mean age, hormone therapy regimen, and hormone therapy prescriber type were similar to women who agreed to be interviewed. Of the 670 women interviewed, 377 (56%) reported that they had attempted to stop using hormone therapy. Of these, 280 (74%) were not taking hormone therapy up to the time of telephone interview, a median of 5.7 months (interquartile range, 3.5–6.8 months) after they had stopped. The remaining 97 women (26%) had resumed taking hormone therapy and were taking it up to the time of interview.

The mean (± standard deviation) age of the 377 women who attempted to stop hormone therapy was 58.9 ± 5.1 years, 78% were white, and 42% were college graduates (Table 1). Hysterectomy was reported by 27% (Table 2).

In univariate analyses, predictors of resuming hormone therapy use were starting hormone therapy for reasons other than health promotion, troublesome withdrawal symptoms, hysterectomy, hormone therapy use for more than 10 years, receiving hormone therapy from a nongynecologist, lack of knowledge of hormone therapy risks, less than college education, and perceiving that risk of hip or spine fracture was higher than average (Table 4). In multivariable models, only troublesome withdrawal symptoms (odds ratio [OR] 8.8; 95% confidence interval [CI] 4.9, 16.0), hysterectomy (OR 1.9; 95% CI 1.1, 3.6), receiving hormone therapy from a nongynecologist (OR 2.2; 95% CI 1.2, 4.0), and perceiving higher-than-average risk of hip or spine fracture (OR 1.4; 95% CI 1.1, 1.8) were significantly associated with unsuccessful stopping. Development of either troublesome vasomotor symptoms (OR 2.2; 95% CI 1.2, 4.0) or other troublesome symptoms (OR 2.4; 95% CI 1.3, 4.6) was each independently associated with unsuccessful stopping.

View larger version (19K): [in this window] [in a new window]   Figure 1. Percentage of women unsuccessful in stopping hormone therapy by number of historical risk factors (hysterectomy, main reason for starting hormone therapy other than health promotion, duration of hormone therapy use greater than 10 years) and appearance of troublesome withdrawal symptoms. Dark bars = women with troublesome symptoms; hatched bars = women without troublesome symptoms after stopping hormone therapy. Grady. Discontinuing Hormone Therapy. Obstet Gynecol 2003.

After stopping hormone therapy, only 5.3% of women started taking other prescription or nonprescription agents to prevent or relieve withdrawal symptoms. Among women who stopped hormone therapy, 5.9% subsequently took a bisphosphonate; use of bisphosphonates was greater among women who perceived themselves at higher than average risk for hip or spine fracture (14.0%) versus those who considered themselves at average or lower risk (4.1%).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Among randomly selected women members of a large health maintenance organization who tried to stop hormone therapy, one out of four resumed hormone therapy. The strongest predictor of resuming hormone therapy use was development of troublesome withdrawal symptoms. Women who reported troublesome symptoms after quitting hormone therapy were nearly nine times more likely to resume using hormone therapy than women without such symptoms. Vasomotor and other symptoms were equally likely to result in resumption of hormone therapy use. Troublesome withdrawal symptoms were very common in women who were unable to stop hormone therapy but also occurred in approximately 20% of women who successfully quit. Most of the successful quitters who developed troublesome symptoms continued to have symptoms at the time they were interviewed, several months after stopping hormone therapy. It is unclear why these women tolerated symptoms that caused other women to resume using hormone therapy.

Hysterectomized women, long-term hormone therapy users, women receiving hormone therapy from a non-gynecologist health care provider, those with perceived low risk of fracture, and those who started hormone therapy for reasons other than health promotion were more likely to fail in their attempts to stop hormone therapy. It is unlikely that women with these characteristics were more knowledgeable about the risks of hormone therapy, because we adjusted for differences in education and knowledge of hormone therapy risks. In large surveys of hormone therapy use, women without a uterus are two to three times more likely to use hormone therapy than their peers with a uterus.¹⁰ Hysterectomized women who chose to stop hormone therapy might be more likely to return to its use because of lack of concerns about vaginal bleeding or endometrial cancer risk. Nearly all women with a hysterectomy were using unopposed estrogen, and it is possible that they are not convinced that there are risks associated with such use because the estrogen-alone arm of the Women’s Health Initiative randomized trial has not been stopped. However, given their generally poor performance on our hormone therapy quiz, we doubt that women in our study based their decision to resume hormone therapy use on a sophisticated understanding of potential differences in risks between use of estrogen alone and estrogen plus progestin.

The strength of commitment to stopping might be less among women who have used hormone therapy for many years without obvious harm. Their personal experiences might make them doubt clinical trial evidence that hormone therapy is harmful. Women who received their hormone therapy prescription from a gynecologist were two times more likely to successfully stop hormone therapy compared with women who received their prescription from a nongynecologist. Women who received hormone therapy from a gynecologist were only slightly better informed about hormone therapy risks and did not differ in duration of hormone therapy use or severity of withdrawal symptoms from women receiving care from a nongynecologist (data not shown). Gynecologists might provide better counseling or support for women making decisions about hormone therapy. However, the percentage of women in this cohort who reported that they received high-quality information from a health care provider was no greater among those receiving hormone therapy from a gynecologist than among those receiving hormone therapy from a nongynecologist.⁷

Women who perceive that their risk of osteoporotic fracture is high might be more likely to stop because they know that there are effective bone-specific drugs available for prevention of fractures. Consistent with this hypothesis, we found that women with a greater-than-average perceived risk of fracture were more likely to begin using bisphosphonates than those who perceived their risk to be low. Women who started hormone therapy for health promotion were highly likely to be successful quitters, probably because their main reason for using hormone therapy has been undermined by recent findings from randomized trials.

We hypothesized that women who have trouble stopping hormone therapy might be more successful if they were more knowledgeable about the risks of hormone therapy, tapered off hormone therapy slowly, or used another treatment for withdrawal symptoms during the first few months after quitting. Both higher education and better knowledge of the risks of hormone therapy were statistically associated with a higher likelihood of successful stopping in univariate but not in multivariable models. This might be because of lack of power to identify modest effects of education and knowledge, or because these effects are overwhelmed by the powerful effect of troublesome withdrawal symptoms. Although there was a trend to benefit, women tapering hormone therapy were neither less symptomatic nor less likely to resume hormone therapy than women who stopped abruptly. We did not obtain information on the duration of tapering, which might have been inadequate to promote successful stopping. We were unable to assess the value of adjunctive therapy to treat withdrawal symptoms because very few women used other medications at the time of quitting hormone therapy.

How should clinicians interpret and use the results of this study? Many women are afraid to stop hormone therapy because they believe they will develop severe symptoms. Based on our findings, clinicians can reassure women preparing to stop hormone therapy that approximately 70% will have either no or tolerable symptoms, even if they stop abruptly. However, it might be helpful to inform women that approximately 30% are likely to develop bothersome symptoms that might persist. Women with a hysterectomy who are long-term users and who started hormone therapy for reasons other than health promotion are especially likely to have difficulty stopping. It might be helpful to suggest that women who develop symptoms try behavioral measures, such as lowering ambient temperature, drinking cold liquids, and deep abdominal breathing.^11,12 Although we found that tapering hormone therapy was no more effective than stopping abruptly, the taper might not have been gradual enough to be helpful. It is not clear whether women for whom lifestyle measures are inadequate could successfully stop hormone therapy by using a prolonged taper or by adding other medications that have been shown to be effective for relief of hot flushes. Given the substantial proportion of women who have difficulty stopping hormone therapy, it is important to identify effective measures to help users who would like to stop but are unable because of withdrawal symptoms.

	Footnotes

 
This study was funded by Eli Lilly Co.

doi:10.1016/j.obstetgynecol.2003.09.025

Received July 9, 2003. Received in revised form August 27, 2003. Accepted September 11, 2003.

	REFERENCES

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Hulley S, Furberg C, Barrett-Connor E, Cauley J, Grady D, Haskell W, et al. Noncardiovascular disease outcomes during 6.8 years of hormone therapy. Heart and Estrogen/ progestin Replacement Study follow-up (HERS II). JAMA 2002;288:58–66.[Abstract/Free Full Text]

2. Writing Group for the Women’s Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321–33.[Abstract/Free Full Text]

3. Shumaker SA, Legault C, Rapp SR, Thal L, Wallace RB, Ockene JK, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: The Women’s Health Initiative Memory Study: A randomized controlled trial. JAMA 2003;289: 2651–62.[Abstract/Free Full Text]

4. American College of Obstetrics and Gynecology. Questions and answers on hormone therapy. Available at: http://www.acog.com/from_home/publications/press_releases/nr08-30-02.cfm. Accessed June 15, 2003.

5. North American Menopause Society. Amended report from the NAMS Advisory Panel on Postmenopausal Hormone Therapy. Menopause 2003;10:6–12.[Medline]

6. Colata G. Rush to fill void in menopause-drug market. New York Times. September 1, 2002, Section 1, page 1.

7. Ettinger B, Grady D, Tosteson AN, Pressman A, Macer JL. Effect of Women’s Health Initiative findings on women’s decisions to discontinue postmenopausal hormone therapy. Obstet Gynecol 2003;6:1225–32.

8. Tosteson AN, Grove MR, Hammond CS, Moncur MM, Roy GJ, Herbert GM. Early discontinuation of osteoporosis treatment among women with low bone mineral density. Am J Med. In press 2003.

9. National Committee for Quality Assurance. Management of menopause: Informed choices. Available at: http://www.ncqa.org/somc2001/menopause/somc_2001_mom.html. Accessed June 15, 2003.

10. Ettinger B, Li DK, Klein R. Continuation of postmenopausal hormone replacement therapy: Comparison of cyclic versus continuous combined schedules. Menopause 1996;3:185–9.

11. Stearns V, Ullmer L, Lopez JF, Smith Y, Isaacs C, Hayes DF. Hot flushes. Lancet 2002;360:1851–61.[Medline]

12. Freedman RR, Woodward S. Behavioral treatment of menopausal hot flushes: Evaluation by ambulatory monitoring. Am J Obstet Gynecol 1992;167:436–9.[Medline]

This article has been cited by other articles:

				R. Haimov-Kochman, E. Barak-Glantz, E. Ein-Mor, R. Arbel, A. Brzezinski, A. Milwidsky, and D. Hochner-Celnikier Duration not severity of the climacteric syndrome predicts resumption of hormone therapy after discontinuation: a prospective cohort study Hum. Reprod., September 1, 2006; 21(9): 2450 - 2454. [Abstract] [Full Text] [PDF]

				F. Wei, D. L. Miglioretti, M. T. Connelly, S. E. Andrade, K. M. Newton, C. L. Hartsfield, K. A. Chan, and D. S. M. Buist Changes in Women's Use of Hormones After the Women's Health Initiative Estrogen and Progestin Trial by Race, Education, and Income J Natl Cancer Inst Monographs, November 1, 2005; 2005(35): 106 - 112. [Abstract] [Full Text] [PDF]

				J. K. Ockene, D. H. Barad, B. B. Cochrane, J. C. Larson, M. Gass, S. Wassertheil-Smoller, J. E. Manson, V. M. Barnabei, D. S. Lane, R. G. Brzyski, et al. Symptom Experience After Discontinuing Use of Estrogen Plus Progestin JAMA, July 13, 2005; 294(2): 183 - 193. [Abstract] [Full Text] [PDF]

				D. S. M. Buist, K. M. Newton, D. L. Miglioretti, K. Beverly, M. T. Connelly, S. Andrade, C. L. Hartsfield, F. Wei, K A. Chan, and L. Kessler Hormone Therapy Prescribing Patterns in the United States Obstet. Gynecol., November 1, 2004; 104(5): 1042 - 1050. [Abstract] [Full Text] [PDF]

				B. Ettinger, K. E. Ensrud, R. Wallace, K. C. Johnson, S. R. Cummings, V. Yankov, E. Vittinghoff, and D. Grady Effects of Ultralow-Dose Transdermal Estradiol on Bone Mineral Density: A Randomized Clinical Trial Obstet. Gynecol., September 1, 2004; 104(3): 443 - 451. [Abstract] [Full Text] [PDF]

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