Climacteric
ISSN: 1369-7137 (Print) 1473-0804 (Online) Journal homepage: http://www.tandfonline.com/loi/icmt20
Discontinuation rates of menopausal hormone therapy among postmenopausal women in the post-WHI study era I. Kyvernitakis, K. Kostev, O. Hars, U-s. Albert & P. Hadji To cite this article: I. Kyvernitakis, K. Kostev, O. Hars, U-s. Albert & P. Hadji (2015) Discontinuation rates of menopausal hormone therapy among postmenopausal women in the post-WHI study era, Climacteric, 18:5, 737-742, DOI: 10.3109/13697137.2015.1037267 To link to this article: http://dx.doi.org/10.3109/13697137.2015.1037267
View supplementary material
Accepted author version posted online: 07 Apr 2015. Published online: 15 Jul 2015. Submit your article to this journal
Article views: 61
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=icmt20 Download by: [NUS National University of Singapore]
Date: 06 November 2015, At: 00:14
CLIMACTERIC 2015;18:737–742
Discontinuation rates of menopausal hormone therapy among postmenopausal women in the post-WHI study era I. Kyvernitakis, K. Kostev*,†, O. Hars‡, U-S. Albert** and P. Hadji††
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Department of Gynecological Endocrinology, Reproductive Medicine and Osteoporosis, Philipps-University of Marburg; *IMS Health, Frankfurt; †Fresenius University of Applied Sciences, Health & Social Faculty, Idstein; ‡Statistical Institute, Berlin; **Department of Gynecology and Obstetrics, Nordwest-Hospital, Frankfurt; ††Department of Bone Oncology, Endocrinology and Reproductive Medicine, Nordwest-Hospital, Frankfurt, Germany Key words: PERSISTENCY, DISCONTINUATION, MENOPAUSAL HORMONE THERAPY, WOMEN’S HEALTH INITIATIVE
ABSTRACT Objectives Many women are reluctant to take menopausal hormone therapy (MHT) and discontinue the treatment within 12 months. The aim of this study was to investigate the persistence rates of combined MHT in the last decade, reflecting changes in the post-Women’s Health Initiative era. Methods We analyzed 17 020 patients receiving combined MHT from 2004 to 2013 using the Disease Analyzer database. Results After 12 months of follow-up, 44.6% and 33.5% of patients receiving 1 mg and 2 mg, respectively, of oral combined MHT were still on treatment (p ⬍ 0.0001). The persistence rate of patients receiving ⬍ 50 μg of transdermal MHT was 39.1% after 1 year of treatment and presented no differences compared to patients receiving ⫺ ⬎ 50 μg of transdermal MHT with a persistence rate of 38.2%. MHT start in the years 2007–2009 was associated with higher discontinuation rates (hazard ratio 1.04, p ⫽ 0.0709) than MHT start in the years 2010–2013 (hazard ratio 0.90, p ⫽ 0.0001). Conclusions Our results indicate that patients beginning their treatments in the years 2010–2013 were more treatment-persistent than patients beginning with MHT in the early years after publication of the Women’s Health Initiative study (2004–2009). Administration of low-dose oral MHT and transdermal MHT is associated with increased persistency compared to higher doses of oral MHT.
INTRODUCTION More than a decade ago, the large media coverage of the first results of the Women’s Health Initiative (WHI)1 induced a reluctance to, or rather anxiety towards, the use of menopausal hormone therapy (MHT) in the population. Today, after numerous more detailed analyses have been reported, a more rational and evidence-based approach predominates2. The International Menopause Society (IMS) issued a global consensus on the use of MHT and underlined the importance of the age at initiation and a generally rather positive safety profile of MHT in women younger than 60 years2,3. MHT is primarily used to alleviate menopausal symptoms, such as hot flushes, night sweats, and vaginal dryness4,5.
Additionally, MHT is also associated with secondary advantageous effects with regard to cardioprotection6, prevention of osteoporosis7 and colon cancer8,9. Regarding cardiovascular protection, data strongly suggest a coronary benefit with estrogen alone, and possibly with estrogen and progesterone in young women at the onset of the menopause2. Despite these numerous positive associations, many women refuse to start MHT or discontinue MHT early because of concerns about the potential risks related to the ambivalent knowledge in the early post-WHI era10. Younger and older women may discontinue for different reasons, but little is known about the age- and year-dependent discontinuation rates. Data from the U.S. physician survey confirmed the decline in use of total MHT from 2000 to 2009, which reflects clinical recommendations
Correspondence: Dr I. Kyvernitakis, Department of Gynecological Endocrinology, Reproductive Medicine and Osteoporosis, Philipps-University of Marburg, Balndingerstrasse 1, 35041 Marburg, Germany; E-mail: [email protected] ORIGINAL ARTICLE © 2015 International Menopause Society DOI: 10.3109/13697137.2015.1037267
Received 24-01-2015 Revised 20-03-2015 Accepted 31-03-2015
Persistency and menopausal hormone therapy
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
indicating the use of lower doses as well as vaginal formulations for local treatment11. In parallel to the US, the swinging pendulum of confusion with regard to MHT was also apparent in Europe. The proportion of gynecologists in the UK prescribing MHT increased after 1993, but persistency rates decreased from 2003 onwards, reflecting the changes in the post-WHI era12. The aim of the present study was to determine the changes in the rates of persistency in women receiving MHT from 2004 to 2013 in a large population-based analysis in Germany. We assessed changes in accordance with the form of application, MHT dose, age at treatment start and physician specialty.
METHODS The study was conducted according to the German law and the Declaration of Helsinki.
Disease Analyzer database The Disease Analyzer database (IMS Health Inc.) compiles drug prescriptions, diagnoses, and basic medical and demographic data directly obtained from the computer systems of the practices of general practitioners and specialists throughout Germany13. Diagnoses (ICD-10), prescriptions (Anatomical Therapeutic Chemical (ATC) Classification System), and the quality of reported data are continuously monitored by IMS Health based on a number of quality criteria (e.g. completeness of documentation, linkage of diagnoses and prescriptions). In the present analysis, we analyzed persistency according to discontinuation (considering a medication-free gap of more than 90 days) and overall persistency from treatment start to treatment end after 12 months. It is common in Germany to receive prescriptions either for 1 or 3 months. Therefore, we considered the 90-day medication-free gap as a standard for our analyses. Accordingly, we indirectly measured compliance from the existing registry of the prescriptions filled. The data were derived directly from the computers of the physicians’ practices via standardized interfaces, and they provide daily routine information on patients’ diseases and therapies. Practices transmit patient data stored in the physicians’ computers to IMS Health on a monthly basis. Before transmission, the data are encrypted for data protection. The validity of the Disease Analyzer data was previously evaluated and described elsewhere13. This approach has been the basis of a number of studies and peer-reviewed scientific publications in the fields of epidemiology and in osteoporosis research14–16.
Study population First-time combined MHT prescriptions from January 2004 to December 2013 were defined as the index dates; the latest
738
Kyvernitakis et al. follow-up date was identified as December 2013. Patients with a follow-up time of less than 365 days prior to the index date were excluded. This exclusion criterion was necessary for the per protocol identification of the first treatment initiation time point. Out of the original dataset of 15 million patients, we were able to include data from 17 020 women for the persistence analyses. All patients were treated by gynecologists or general practitioners.
Study outcome The primary endpoint of our study was to investigate the combined MHT treatment discontinuation rates within 12 months after the index date. We defined treatment discontinuation as 90 days without MHT therapy during this period with at least one visit after these 90 days. Furthermore, for comparison reasons, we also investigated the changes in persistence rates in patients with a refill gap of 180 days (at least one visit after 180 days). A longitudinal dataset of medication supply was compiled for each individual patient, and nonpersistence with MHT was calculated. Here, the number of days of drug supply was calculated from the quantity and dosage information associated with each prescription record. To better investigate the dynamic changes of persistency since 2004, we assessed Kaplan–Meier curves of each year separately.
Covariates Demographic data included age, health insurance type (private or statutory), general practitioner (GP) and gynecologist care, as well as the practice region (East versus West Germany). Statutory insurance does not imply co-payment for medications; patients with private insurance pay for their medication beforehand and receive the total amount reimbursed. Codiagnoses were determined based on primary-care diagnoses for diabetes, osteoporosis, menopausal symptoms, thrombosis, breast cancer and heart disease. Furthermore, co-diagnoses within 12 months of the index date, as well as number of different co-medications taken during MHT have been assessed.
Statistical analysis The cumulative non-persistence rate with initial treatments was estimated using a Kaplan–Meier analysis. A Cox proportional hazards regression model was used to estimate the relationship between non-persistence and the demographic and clinical variables previously described, for a maximum follow-up period of 12 months. A stepwise selection procedure with an entry criterion of p ⬍ 0.1 was used to select the final optimal model. All variables in Table 1 have been considered for the analyses. After the selection procedure and considering statistically significant variables from Table 1 with a p value
Climacteric
Persistency and menopausal hormone therapy
Kyvernitakis et al.
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Table 1 Baseline characteristics of women taking combination menopausal hormone therapy. Data are given as percentages Transdermal patch ⫺ ⬎ 50 g (n ⫽ 1322)
Variable
Oral 2 mg (n ⫽ 7047)
Age (years) ⬍ ⫺ 40 41–50 51–60 ⬎ 60
3.8 38.4 48.1 9.8
16.4 43.9 31.7 8.0
1.3 20.7 56.4 21.6
2.5 37.8 49.8 11.0
⬍ 0.0001 ⬍ 0.0001 ⬍ 0.0001 ⬍ 0.0001
Gynecologist Private health insurance West Germany
71.7 9.9 85.6
71.4 9.0 84.9
69.4 9.0 77.3
84.5 10.1 89.0
⬍ 0.0001 0.1738 ⬍ 0.0001
Therapy start 2004–2006 2007–2009 2010–2013
16.7 39.2 44.1
23.3 37.1 39.6
23.5 36.6 40.0
6.9 38.6 54.5
⬍ 0.0001 0.0439 ⬍ 0.0001
Co-diagnoses (within 12 months prior to index date) Osteoporosis Diabetes Menopausal symptoms Thrombosis Breast cancer Depression Heart disease
3.4 2.8 74.3 1.2 0.7 16.6 2.7
2.5 3.1 58.2 1.5 0.4 14.8 3.3
5.0 4.2 81.1 1.8 0.7 20.1 4.6
4.1 3.0 83.4 1.4 0.8 16.1 1.8
⬍ 0.0001 0.0292 ⬍ 0.0001 0.1041 0.0677 ⬍ 0.0001 ⬍ 0.0001
1.4
1.5
1.5
1.3
⬍ 0.0001
Number of different co-medications taken at the same time
of ⬍ 0.05, we included the significant co-variables in the model shown in Table 2. The adjusted hazard ratios (HR) and 95% confidence intervals (CI) are presented in Table 2 for the independent variables. The proportional hazards assumption was assessed and upheld for all analyses. Furthermore, potential confounders (age, gynecologists, private health insurance, and practice located in West Germany), co-diagnoses, and co-medication were included as independent variables. Two-sided tests were used, and a p value ⬍ 0.05 was considered to be statistically significant. All analyses were performed using SAS 9.3. (SAS Institute, Cary, USA). Best-practice methods for retrospective database studies were considered17.
RESULTS Baseline characteristics Altogether, 17 020 patients with first-time prescriptions of combined MHT were identified. Their overall and age-specific characteristics as well as the baseline characteristics in accordance with treatment start are summarized in Table 1.
Climacteric
Transdermal patch ⬍ 50 g (n ⫽ 1752)
Oral 1 mg (n ⫽ 6899)
p Value
Kaplan–Meier analyses After 12 months of follow-up, 44.6% and 33.5% of patients receiving 1 mg and 2 mg, respectively, of oral combined MHT were still on treatment (refill gap of 90 days; p ⬍ 0.0001) (Figure 1). Similarly, the persistence rates of patients receiving ⬍ 50 μg of transdermal combined MHT were 39.1% after 1 year of treatment, and presented no differences compared to patients receiving ⫺ ⬎ 50 μg of transdermal MHT with a persistence rate of 38.2% (Figure 2). We further investigated whether the persistency rates varied with regard to the timing of treatment start (2005–2006, 2007–2008, 2009–2010 and 2011–2012) (see Supplementary Figures 1 and 2 to be found at online http://informahealthcare.com/doi/abs/10.3109/ 13697137.2015.1037267). With regard to 1 mg of oral MHT, we found persistency rates of 41.3%, 43.8%, 44.0% and 46.4% for the year categories 2005–2006, 2007–2008, 2009–2010 and 2011–2012, respectively. With regard to 2 mg of oral MHT, persistency ranged from 35.3%, 29.5%, 32.5% and 35% for the year categories 2005–2006, 2007–2008, 2009–2010 and 2011–2012, respectively. Considering transdermal MHT, the persistency rates increased from 37.1% in 2005–2006 to 39.2% in 2011–2012 for patches ⬍ 50 μg,
739
Persistency and menopausal hormone therapy
Kyvernitakis et al.
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Table 2 Cox regression analyses of association of combination menopausal hormone therapy with discontinuation within 12 months of defined outcomes
Outcome variables
Hazard ratio (95% confidence interval)
Oral 2 mg vs. oral 1 mg Patch ⬍ 50 μg vs. oral 1 mg Patch ⫺ ⬎ 50 μg vs. oral 1 mg Patch ⬍ 50 μg vs. oral 2 mg Patch ⫺ ⬎ 50 μg vs. oral 2 mg Age ⬍ ⫺ 40 vs. age ⬎ 60 years Age 41–50 vs. age ⬎ 60 years Age 51–60 vs. age ⬎ 60 years Diabetes Menopausal symptoms Gynecologist Private health insurance West Germany Osteoporosis Thrombosis/embolism Breast cancer Depression Heart disease Therapy start in 2007–2009 Therapy start in 2010–013 Number of different co-medications
1.26 1.19 0.19 0.95 0.95 1.64 0.99 0.93 0.92 0.80 0.88 0.96 1.00 1.05 0.97 1.07 1.00 1.07 1.04 0.90 1.04
(1.21–1.31) (1.12–1.27) (1.11–1.28) (0.89–1.01) (0.88–1.02) (1.51–1.79) (0.93–1.06) (0.88–0.99) (0.84–1.02) (0.77–0.84) (0.84–0.92) (0.90–1.02) (0.95–1.05) (0.95–1.15) (0.84–1.12) (0.86–1.33) (0.96–1.05) (0.97–1.18) (1.00–1.09) (0.85–0.94) (1.02–1.05)
p Value ⬍ 0.0001 ⬍ 0.0001 ⬍ 0.0001 0.0746 0.1426 ⬍ 0.0001 0.8192 0.0230 0.1187 ⬍ 0.0001 ⬍ 0.0001 0.1631 0.9685 0.3282 0.6461 0.5522 0.8542 0.1947 0.0709 0.0001 ⬍ 0.0001
while the persistency rates for patches ⫺ ⬎ 50 μg increased from 38.4% in 2005–2006 to 44.5% in the years 2011–2012. To better scrutinize the dynamics of persistence within the study population, we assessed the persistence rates with a longer refill gap of 180 days. As such, the total persistency of oral MHT was 55.7% for patients receiving 1 mg and 44.5% for patients receiving 2 mg. Moreover, the total persistency of patch MHT was 51.1% for patients receiving less than 50 μg and 50.6% for patients receiving a dosage over 50 μg (data not shown).
Figure 2 Kaplan – Meier curves for persistency over 12 months for women taking transdermal combination menopausal hormone therapy (refi ll gap of 90 days)
Cox regression analyses The multivariate hazard ratios of the Cox regression models are presented in Table 2. Hazard ratios for the 12-month risk of MHT discontinuation were adjusted for demographic and clinical variables. A significantly increased risk for treatment discontinuation was found for patients 60 years of age or older (HR 1.64; 95% CI 1.15–1.79; p ⬍ 0.0001). Hereby, patients between 51 and 60 years old were more treatmentpersistent than patients older than 60 years (HR 0.93; 95% CI 0.88–0.99; p ⫽ 0.0230). The presence of menopausal symptoms was associated with a decreased risk of discontinuation (HR 0.80; 95% CI 0.77–0.84; p ⬍ 0.0001). Similarly, patients treated by gynecologists were more persistent than patients treated by general practitioners (HR 0.88; 95% CI 0.84–0.92; p ⬍ 0.0001). Finally, MHT started in the years 2007–2009 was associated with higher discontinuation rates (HR 1.04; 95% CI 1.00–1.09; p ⫽ 0.0709) than MHT start in the years 2010–2013 (HR 0.90; 95% CI 0.85–0.94; p ⫽ 0.0001), which resulted in lower discontinuation rates. Number of co-medications was an additional variable of MHT treatment discontinuation (HR 1.04; 95% CI 1.02– 1.05; p ⬍ 0.0001).
DISCUSSION
Figure 1 Kaplan – Meier curves for persistency over 12 months for women taking oral combination menopausal hormone therapy (refi ll gap of 90 days)
740
This study investigated real-world persistence with combined MHT for 12 months from 2004 to 2014 in Germany. Overall, our results indicate that, after 12 months of follow-up, 44.6% of patients receiving 1 mg oral MHT and 33.5% of patients receiving 2 mg oral MHT were persistent (gap longer than 90 days). Furthermore, after 12 months of follow-up, 39.1% and 38.2% of patients using ⬍ 50 μg and ⫺ ⬎ 50 μg of transdermal MHT, respectively, were still on treatment. We further showed that patients over 60 years presented the highest discontinuation rates. Moreover, patients with menopausal symptoms and patients treated by gynecologists were more
Climacteric
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Persistency and menopausal hormone therapy treatment-persistent. Most important, patients starting their MHT in the years 2010–2013 were more treatment-persistent than patients starting their therapy in the years before. Our results provide additional valuable information on persistence to MHT in the post-WHI era. Guay and colleagues demonstrated that clinical practice responded quickly to the results of ‘this largely mediatized study’18. They underlined that, 1 year after the first results of the WHI study were published, significant changes in the trends of use, women’s characteristics and in the doses of estrogen prescribed were apparent. The authors reported a significant decrease in persistence from 59% in the pre-WHI cohort to 45% in the post-WHI cohort. Interestingly, the proportion of new users presenting cardiovascular risk factors was not changed18. The decline in the total number of MHT prescriptions after 2002 was closely related to the WHI study’s negative reports. These findings are in line with the prescription trends in the Canadian (32%) and the US populations (37%)19,20. An additional study reported discontinuation rates of up to 38% in the first 6 months after release of the WHI study results through the lay press or medical literature21. The authors emphasized the negative impact of published negative results and their influence on pharmacotherapy decisions. Further literature from the Danish population stated that the majority of women experiencing a myocardial infarction during ongoing MHT continued the treatment after discharge and this pattern of MHT use did not change markedly after 200222. The most recent publication of persistency rates in the US population from 2000 to 2009 showed a stable decline and an increase in the use of lower-dose and vaginal MHT11. Buick and colleagues identified the reasons for use or discontinuation of MHT in a systematic literature research from 1980 to 200223. As such, the authors reported that the perceptions of benefits are often countered by concerns over potential adverse effects. Short-term symptom reliefs against long-term benefits were the two conflicting poles of decisionmaking with regard to MHT. As a result of this, in order to maintain a high MHT persistency rate, Vestergaard and colleagues suggested pragmatic approaches including offering of alternative MHT formulations to women not tolerating the first-line treatment24. Oren identified further motives for MHT discontinuation and stated high age, low education level, unemployment, treatment recommendation by a general practitioner and attitude towards MHT usage as factors associated with discontinuation25. A further study prospectively investigated factors associated with a low persistency towards MHT in the year 2003 – close to the publication of the WHI study – and supported that fear of breast cancer was the most common reason for early discontinuation26. Our results are congruent with these findings since we could confirm the low persistency rates in the early years after the WHI publication (lowest rate 29.5% in the year 2007) and the highest in recent years (highest rate 46.4% in the year 2013). In this context, Panay and colleagues scrutinized the risks and benefits of MHT in the most recent recommendations of the British Menopause Society and Women’s Health Concern in relation to breast cancer27. As such, the WHI estrogen and
Climacteric
Kyvernitakis et al. progestin study showed a small increase of 0.1% per annum in the risk of breast cancer after 5 years of MHT27. However, the WHI estrogen-alone arm (hysterectomized women only) showed a statistically significant decrease in breast cancer risk. Moreover, the Million Women Study (MWS) raised concerns over the long-term safety of MHT with regard to breast cancer. Overall, a recent critique on the WHI and the MWS has delivered a number of key weaknesses that limit the ability of trials to establish a causal association between MHT and breast cancer27. These approaches have been also considered in the recent recommendations of the International Menopause Society2. There are certain limitations to consider for the present analysis. First, the retrospective character of the study does not allow us to examine the dynamic changes within age groups and their co-morbidities. Consequently, it is not possible to prove that lower persistence in the early years of the post-WHI era was associated with the negative impact of the media. Second, no detailed documentation on the sideeffects of MHT was available in the database. However, we assigned the most common co-diagnoses to each therapy group. Finally, of particular concern is the lack of data with regard to the individual indication of MHT. It can be assumed that patients with severe menopausal symptoms comprise a higher level of ailment compared to women who use MHT for long-term preventative reasons; this could not be addressed, as the identification of these variables in the database used was not possible. Despite these limitations, our study has several strengths. This analysis used real-life data from gynecologists and GPs, indicating the realistic impact of access to medical care on treatment persistence. The present analysis investigated age and health-care provider-specific differences in patients receiving MHT for the first time. Additionally, for the first time in this regard, we stated that changes of persistence depended on the time of start of therapy over the years since the publication of the WHI study. In conclusion, this study is the first in which the yeardependent changes of discontinuation rates in the post-WHI times have been assessed. We showed that patients starting their treatments in recent years (2010–2013) were more treatment-persistent than patients beginning MHT in the early years after publication of the WHI study results (2004–2009). Altogether, our results indicate rather low persistency rates between 33.5% and 44.6% for different MHT dosages after 12 months of treatment. This underlines the need to improve long-term persistence in MHT in order to achieve the full benefit of the treatment. Our results support that administration of low-dose oral MHT and transdermal MHT is associated with higher persistency than higher doses of oral MHT.
Conflict of interest The authors report no confl ict of interest. The authors alone are responsible for the content and writing of this paper.
Source of funding
Nil.
741
Persistency and menopausal hormone therapy
Kyvernitakis et al.
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
References 1. Rossouw JE, Anderson GL, Prentice RL, et al. 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 2. de Villiers TJ, Pines A, Panay N, et al. Updated 2013 International Menopause Society recommendations on menopausal hormone therapy and preventive strategies for midlife health. Climacteric 2013;16:316–37 3. Sturdee DW, Pines A, Archer DF; International Menopause Society Writing Group. Updated IMS recommendations on postmenopausal hormone therapy and preventive strategies for midlife health. Climacteric 2011;14:302–20 4. Jalava-Broman J, Makinen J, Ojanlatva A, Jokinen K, Sillanmaki L, Rautava P. Change in the frequency of HRT use from 2000 to 2005 and reasons to discontinue: follow-up of a normal cohort in Finland. Acta Obstet Gynecol Scand 2011;90:351–7 5. Kyvernitakis I, Ziller V, Hars O, Bauer M, Kalder M, Hadji P. Prevalence of menopausal symptoms and their influence on adherence in women with breast cancer. Climacteric 2014;17: 252–9 6. Grodstein F, Stampfer MJ, Colditz GA, et al. Postmenopausal hormone therapy and mortality. N Engl J Med 1997;336: 1769–75 7. Schneider DL, Barrett-Connor EL, Morton DJ. Timing of postmenopausal estrogen for optimal bone mineral density. The Rancho Bernardo Study. JAMA 1997;277:543–7 8. Jacobs DM, Tang MX, Stern Y, et al. Cognitive function in nondemented older women who took estrogen after menopause. Neurology 1998;50:368–73 9. Nanda K, Bastian LA, Hasselblad V, Simel DL. Hormone replacement therapy and the risk of colorectal cancer: a meta-analysis. Obstet Gynecol 1999;93:880–8 10. Regan MM, Emond SK, Attardo MJ, Parker RA, Greenspan SL. Why do older women discontinue hormone replacement therapy? J Women’s Health Gend Based Med 2001;10:343–50 11. Tsai SA, Stefanick ML, Stafford RS. Trends in menopausal hormone therapy use of US office-based physicians, 2000–2009. Menopause 2011;18:385–92 12. Isaacs AJ, Drew SV, McPherson K. UK women doctors’ use of hormone replacement therapy: 10-year follow up. Climacteric 2005;8:154–61 13. Becher H, Kostev K, Schroder-Bernhardi D. Validity and representativeness of the “Disease Analyzer” patient database for use in pharmacoepidemiological and pharmacoeconomic studies. Int J Clin Pharmacol Therapeut 2009;47:617–26 14. Hadji P, Claus V, Ziller V, Intorcia M, Kostev K, Steinle T. GRAND: the German retrospective cohort analysis on compliance and persistence and the associated risk of fractures in osteoporotic
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
women treated with oral bisphosphonates. Osteoporos Int 2012;23:223–31 Potocki M, Ziller R, Mueller C. Mid-regional pro-adrenomedullin in acute heart failure: a better biomarker or just another biomarker? Curr Heart Fail Rep 2012;9:244–51 Kyvernitakis I, Kostev K, Kurth A, Albert US, Hadji P. Differences in persistency with teriparatide in patients with osteoporosis according to gender and health care provider. Osteoporos Int 2014;25:2721–8 Motheral B, Brooks J, Clark MA, et al. A checklist for retrospective database studies – report of the ISPOR Task Force on Retrospective Databases. Value Health 2003;6:90–7 Guay MP, Dragomir A, Pilon D, Moride Y, Perreault S. Changes in pattern of use, clinical characteristics and persistence rate of hormone replacement therapy among postmenopausal women after the WHI publication. Pharmacoepidemiol Drug Safety 2007;16:17–27 Austin PC, Mamdani MM, Tu K, Jaakkimainen L. Prescriptions for estrogen replacement therapy in Ontario before and after publication of the Women’s Health Initiative Study. JAMA 2003;289:3241–2 Gavin NI, Thorp JM, Ohsfeldt RL. Determinants of hormone replacement therapy duration among postmenopausal women with intact uteri. Menopause 2001;8:377–83 Bestul MB, McCollum M, Hansen LB, Saseen JJ. Impact of the Women’s Health Initiative trial results on hormone replacement therapy. Pharmacotherapy 2004;24:495–9 Bretler DM, Hansen PR, Abildstrom SZ, et al. Use and discontinuation of hormone replacement therapy in women with myocardial infarction: a nationwide study. Br J Clin Pharmacol 2011;71:105–15 Buick DL, Crook D, Horne R. Women’s perceptions of hormone replacement therapy: risks and benefits (1980–2002). A literature review. Climacteric 2005;8:24–35 Vestergaard P, Hermann AP, Gram J, et al. Improving compliance with hormonal replacement therapy in primary osteoporosis prevention. Maturitas 1997;28:137–45 Oren A. Motives for initiation, temporary discontinuation, and permanent discontinuation of hormone replacement therapy use among Norwegian women. Maturitas 2009;64:33–7 Corrado F, D’Anna R, Caputo F, Cannata ML, Zoccali MG, Cancellieri F. Compliance with hormone replacement therapy in postmenopausal Sicilian women. Eur J Obstet Gynecol Reprod Biol 2005;118:225–8 Panay N, Hamoda H, Arya R, Savvas M; British Menopause Society, Women’s Health Concern. The 2013 British Menopause Society & Women’s Health Concern recommendations on hormone replacement therapy. Menopause Int 2013;19:59–68
Supplementary Data Supplementary Figure 1a,b,c,d and Figure 2a,b,c,d.
742
Climacteric
ISSN: 1369-7137 (Print) 1473-0804 (Online) Journal homepage: http://www.tandfonline.com/loi/icmt20
Discontinuation rates of menopausal hormone therapy among postmenopausal women in the post-WHI study era I. Kyvernitakis, K. Kostev, O. Hars, U-s. Albert & P. Hadji To cite this article: I. Kyvernitakis, K. Kostev, O. Hars, U-s. Albert & P. Hadji (2015) Discontinuation rates of menopausal hormone therapy among postmenopausal women in the post-WHI study era, Climacteric, 18:5, 737-742, DOI: 10.3109/13697137.2015.1037267 To link to this article: http://dx.doi.org/10.3109/13697137.2015.1037267
View supplementary material
Accepted author version posted online: 07 Apr 2015. Published online: 15 Jul 2015. Submit your article to this journal
Article views: 61
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=icmt20 Download by: [NUS National University of Singapore]
Date: 06 November 2015, At: 00:14
CLIMACTERIC 2015;18:737–742
Discontinuation rates of menopausal hormone therapy among postmenopausal women in the post-WHI study era I. Kyvernitakis, K. Kostev*,†, O. Hars‡, U-S. Albert** and P. Hadji††
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Department of Gynecological Endocrinology, Reproductive Medicine and Osteoporosis, Philipps-University of Marburg; *IMS Health, Frankfurt; †Fresenius University of Applied Sciences, Health & Social Faculty, Idstein; ‡Statistical Institute, Berlin; **Department of Gynecology and Obstetrics, Nordwest-Hospital, Frankfurt; ††Department of Bone Oncology, Endocrinology and Reproductive Medicine, Nordwest-Hospital, Frankfurt, Germany Key words: PERSISTENCY, DISCONTINUATION, MENOPAUSAL HORMONE THERAPY, WOMEN’S HEALTH INITIATIVE
ABSTRACT Objectives Many women are reluctant to take menopausal hormone therapy (MHT) and discontinue the treatment within 12 months. The aim of this study was to investigate the persistence rates of combined MHT in the last decade, reflecting changes in the post-Women’s Health Initiative era. Methods We analyzed 17 020 patients receiving combined MHT from 2004 to 2013 using the Disease Analyzer database. Results After 12 months of follow-up, 44.6% and 33.5% of patients receiving 1 mg and 2 mg, respectively, of oral combined MHT were still on treatment (p ⬍ 0.0001). The persistence rate of patients receiving ⬍ 50 μg of transdermal MHT was 39.1% after 1 year of treatment and presented no differences compared to patients receiving ⫺ ⬎ 50 μg of transdermal MHT with a persistence rate of 38.2%. MHT start in the years 2007–2009 was associated with higher discontinuation rates (hazard ratio 1.04, p ⫽ 0.0709) than MHT start in the years 2010–2013 (hazard ratio 0.90, p ⫽ 0.0001). Conclusions Our results indicate that patients beginning their treatments in the years 2010–2013 were more treatment-persistent than patients beginning with MHT in the early years after publication of the Women’s Health Initiative study (2004–2009). Administration of low-dose oral MHT and transdermal MHT is associated with increased persistency compared to higher doses of oral MHT.
INTRODUCTION More than a decade ago, the large media coverage of the first results of the Women’s Health Initiative (WHI)1 induced a reluctance to, or rather anxiety towards, the use of menopausal hormone therapy (MHT) in the population. Today, after numerous more detailed analyses have been reported, a more rational and evidence-based approach predominates2. The International Menopause Society (IMS) issued a global consensus on the use of MHT and underlined the importance of the age at initiation and a generally rather positive safety profile of MHT in women younger than 60 years2,3. MHT is primarily used to alleviate menopausal symptoms, such as hot flushes, night sweats, and vaginal dryness4,5.
Additionally, MHT is also associated with secondary advantageous effects with regard to cardioprotection6, prevention of osteoporosis7 and colon cancer8,9. Regarding cardiovascular protection, data strongly suggest a coronary benefit with estrogen alone, and possibly with estrogen and progesterone in young women at the onset of the menopause2. Despite these numerous positive associations, many women refuse to start MHT or discontinue MHT early because of concerns about the potential risks related to the ambivalent knowledge in the early post-WHI era10. Younger and older women may discontinue for different reasons, but little is known about the age- and year-dependent discontinuation rates. Data from the U.S. physician survey confirmed the decline in use of total MHT from 2000 to 2009, which reflects clinical recommendations
Correspondence: Dr I. Kyvernitakis, Department of Gynecological Endocrinology, Reproductive Medicine and Osteoporosis, Philipps-University of Marburg, Balndingerstrasse 1, 35041 Marburg, Germany; E-mail: [email protected] ORIGINAL ARTICLE © 2015 International Menopause Society DOI: 10.3109/13697137.2015.1037267
Received 24-01-2015 Revised 20-03-2015 Accepted 31-03-2015
Persistency and menopausal hormone therapy
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
indicating the use of lower doses as well as vaginal formulations for local treatment11. In parallel to the US, the swinging pendulum of confusion with regard to MHT was also apparent in Europe. The proportion of gynecologists in the UK prescribing MHT increased after 1993, but persistency rates decreased from 2003 onwards, reflecting the changes in the post-WHI era12. The aim of the present study was to determine the changes in the rates of persistency in women receiving MHT from 2004 to 2013 in a large population-based analysis in Germany. We assessed changes in accordance with the form of application, MHT dose, age at treatment start and physician specialty.
METHODS The study was conducted according to the German law and the Declaration of Helsinki.
Disease Analyzer database The Disease Analyzer database (IMS Health Inc.) compiles drug prescriptions, diagnoses, and basic medical and demographic data directly obtained from the computer systems of the practices of general practitioners and specialists throughout Germany13. Diagnoses (ICD-10), prescriptions (Anatomical Therapeutic Chemical (ATC) Classification System), and the quality of reported data are continuously monitored by IMS Health based on a number of quality criteria (e.g. completeness of documentation, linkage of diagnoses and prescriptions). In the present analysis, we analyzed persistency according to discontinuation (considering a medication-free gap of more than 90 days) and overall persistency from treatment start to treatment end after 12 months. It is common in Germany to receive prescriptions either for 1 or 3 months. Therefore, we considered the 90-day medication-free gap as a standard for our analyses. Accordingly, we indirectly measured compliance from the existing registry of the prescriptions filled. The data were derived directly from the computers of the physicians’ practices via standardized interfaces, and they provide daily routine information on patients’ diseases and therapies. Practices transmit patient data stored in the physicians’ computers to IMS Health on a monthly basis. Before transmission, the data are encrypted for data protection. The validity of the Disease Analyzer data was previously evaluated and described elsewhere13. This approach has been the basis of a number of studies and peer-reviewed scientific publications in the fields of epidemiology and in osteoporosis research14–16.
Study population First-time combined MHT prescriptions from January 2004 to December 2013 were defined as the index dates; the latest
738
Kyvernitakis et al. follow-up date was identified as December 2013. Patients with a follow-up time of less than 365 days prior to the index date were excluded. This exclusion criterion was necessary for the per protocol identification of the first treatment initiation time point. Out of the original dataset of 15 million patients, we were able to include data from 17 020 women for the persistence analyses. All patients were treated by gynecologists or general practitioners.
Study outcome The primary endpoint of our study was to investigate the combined MHT treatment discontinuation rates within 12 months after the index date. We defined treatment discontinuation as 90 days without MHT therapy during this period with at least one visit after these 90 days. Furthermore, for comparison reasons, we also investigated the changes in persistence rates in patients with a refill gap of 180 days (at least one visit after 180 days). A longitudinal dataset of medication supply was compiled for each individual patient, and nonpersistence with MHT was calculated. Here, the number of days of drug supply was calculated from the quantity and dosage information associated with each prescription record. To better investigate the dynamic changes of persistency since 2004, we assessed Kaplan–Meier curves of each year separately.
Covariates Demographic data included age, health insurance type (private or statutory), general practitioner (GP) and gynecologist care, as well as the practice region (East versus West Germany). Statutory insurance does not imply co-payment for medications; patients with private insurance pay for their medication beforehand and receive the total amount reimbursed. Codiagnoses were determined based on primary-care diagnoses for diabetes, osteoporosis, menopausal symptoms, thrombosis, breast cancer and heart disease. Furthermore, co-diagnoses within 12 months of the index date, as well as number of different co-medications taken during MHT have been assessed.
Statistical analysis The cumulative non-persistence rate with initial treatments was estimated using a Kaplan–Meier analysis. A Cox proportional hazards regression model was used to estimate the relationship between non-persistence and the demographic and clinical variables previously described, for a maximum follow-up period of 12 months. A stepwise selection procedure with an entry criterion of p ⬍ 0.1 was used to select the final optimal model. All variables in Table 1 have been considered for the analyses. After the selection procedure and considering statistically significant variables from Table 1 with a p value
Climacteric
Persistency and menopausal hormone therapy
Kyvernitakis et al.
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Table 1 Baseline characteristics of women taking combination menopausal hormone therapy. Data are given as percentages Transdermal patch ⫺ ⬎ 50 g (n ⫽ 1322)
Variable
Oral 2 mg (n ⫽ 7047)
Age (years) ⬍ ⫺ 40 41–50 51–60 ⬎ 60
3.8 38.4 48.1 9.8
16.4 43.9 31.7 8.0
1.3 20.7 56.4 21.6
2.5 37.8 49.8 11.0
⬍ 0.0001 ⬍ 0.0001 ⬍ 0.0001 ⬍ 0.0001
Gynecologist Private health insurance West Germany
71.7 9.9 85.6
71.4 9.0 84.9
69.4 9.0 77.3
84.5 10.1 89.0
⬍ 0.0001 0.1738 ⬍ 0.0001
Therapy start 2004–2006 2007–2009 2010–2013
16.7 39.2 44.1
23.3 37.1 39.6
23.5 36.6 40.0
6.9 38.6 54.5
⬍ 0.0001 0.0439 ⬍ 0.0001
Co-diagnoses (within 12 months prior to index date) Osteoporosis Diabetes Menopausal symptoms Thrombosis Breast cancer Depression Heart disease
3.4 2.8 74.3 1.2 0.7 16.6 2.7
2.5 3.1 58.2 1.5 0.4 14.8 3.3
5.0 4.2 81.1 1.8 0.7 20.1 4.6
4.1 3.0 83.4 1.4 0.8 16.1 1.8
⬍ 0.0001 0.0292 ⬍ 0.0001 0.1041 0.0677 ⬍ 0.0001 ⬍ 0.0001
1.4
1.5
1.5
1.3
⬍ 0.0001
Number of different co-medications taken at the same time
of ⬍ 0.05, we included the significant co-variables in the model shown in Table 2. The adjusted hazard ratios (HR) and 95% confidence intervals (CI) are presented in Table 2 for the independent variables. The proportional hazards assumption was assessed and upheld for all analyses. Furthermore, potential confounders (age, gynecologists, private health insurance, and practice located in West Germany), co-diagnoses, and co-medication were included as independent variables. Two-sided tests were used, and a p value ⬍ 0.05 was considered to be statistically significant. All analyses were performed using SAS 9.3. (SAS Institute, Cary, USA). Best-practice methods for retrospective database studies were considered17.
RESULTS Baseline characteristics Altogether, 17 020 patients with first-time prescriptions of combined MHT were identified. Their overall and age-specific characteristics as well as the baseline characteristics in accordance with treatment start are summarized in Table 1.
Climacteric
Transdermal patch ⬍ 50 g (n ⫽ 1752)
Oral 1 mg (n ⫽ 6899)
p Value
Kaplan–Meier analyses After 12 months of follow-up, 44.6% and 33.5% of patients receiving 1 mg and 2 mg, respectively, of oral combined MHT were still on treatment (refill gap of 90 days; p ⬍ 0.0001) (Figure 1). Similarly, the persistence rates of patients receiving ⬍ 50 μg of transdermal combined MHT were 39.1% after 1 year of treatment, and presented no differences compared to patients receiving ⫺ ⬎ 50 μg of transdermal MHT with a persistence rate of 38.2% (Figure 2). We further investigated whether the persistency rates varied with regard to the timing of treatment start (2005–2006, 2007–2008, 2009–2010 and 2011–2012) (see Supplementary Figures 1 and 2 to be found at online http://informahealthcare.com/doi/abs/10.3109/ 13697137.2015.1037267). With regard to 1 mg of oral MHT, we found persistency rates of 41.3%, 43.8%, 44.0% and 46.4% for the year categories 2005–2006, 2007–2008, 2009–2010 and 2011–2012, respectively. With regard to 2 mg of oral MHT, persistency ranged from 35.3%, 29.5%, 32.5% and 35% for the year categories 2005–2006, 2007–2008, 2009–2010 and 2011–2012, respectively. Considering transdermal MHT, the persistency rates increased from 37.1% in 2005–2006 to 39.2% in 2011–2012 for patches ⬍ 50 μg,
739
Persistency and menopausal hormone therapy
Kyvernitakis et al.
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Table 2 Cox regression analyses of association of combination menopausal hormone therapy with discontinuation within 12 months of defined outcomes
Outcome variables
Hazard ratio (95% confidence interval)
Oral 2 mg vs. oral 1 mg Patch ⬍ 50 μg vs. oral 1 mg Patch ⫺ ⬎ 50 μg vs. oral 1 mg Patch ⬍ 50 μg vs. oral 2 mg Patch ⫺ ⬎ 50 μg vs. oral 2 mg Age ⬍ ⫺ 40 vs. age ⬎ 60 years Age 41–50 vs. age ⬎ 60 years Age 51–60 vs. age ⬎ 60 years Diabetes Menopausal symptoms Gynecologist Private health insurance West Germany Osteoporosis Thrombosis/embolism Breast cancer Depression Heart disease Therapy start in 2007–2009 Therapy start in 2010–013 Number of different co-medications
1.26 1.19 0.19 0.95 0.95 1.64 0.99 0.93 0.92 0.80 0.88 0.96 1.00 1.05 0.97 1.07 1.00 1.07 1.04 0.90 1.04
(1.21–1.31) (1.12–1.27) (1.11–1.28) (0.89–1.01) (0.88–1.02) (1.51–1.79) (0.93–1.06) (0.88–0.99) (0.84–1.02) (0.77–0.84) (0.84–0.92) (0.90–1.02) (0.95–1.05) (0.95–1.15) (0.84–1.12) (0.86–1.33) (0.96–1.05) (0.97–1.18) (1.00–1.09) (0.85–0.94) (1.02–1.05)
p Value ⬍ 0.0001 ⬍ 0.0001 ⬍ 0.0001 0.0746 0.1426 ⬍ 0.0001 0.8192 0.0230 0.1187 ⬍ 0.0001 ⬍ 0.0001 0.1631 0.9685 0.3282 0.6461 0.5522 0.8542 0.1947 0.0709 0.0001 ⬍ 0.0001
while the persistency rates for patches ⫺ ⬎ 50 μg increased from 38.4% in 2005–2006 to 44.5% in the years 2011–2012. To better scrutinize the dynamics of persistence within the study population, we assessed the persistence rates with a longer refill gap of 180 days. As such, the total persistency of oral MHT was 55.7% for patients receiving 1 mg and 44.5% for patients receiving 2 mg. Moreover, the total persistency of patch MHT was 51.1% for patients receiving less than 50 μg and 50.6% for patients receiving a dosage over 50 μg (data not shown).
Figure 2 Kaplan – Meier curves for persistency over 12 months for women taking transdermal combination menopausal hormone therapy (refi ll gap of 90 days)
Cox regression analyses The multivariate hazard ratios of the Cox regression models are presented in Table 2. Hazard ratios for the 12-month risk of MHT discontinuation were adjusted for demographic and clinical variables. A significantly increased risk for treatment discontinuation was found for patients 60 years of age or older (HR 1.64; 95% CI 1.15–1.79; p ⬍ 0.0001). Hereby, patients between 51 and 60 years old were more treatmentpersistent than patients older than 60 years (HR 0.93; 95% CI 0.88–0.99; p ⫽ 0.0230). The presence of menopausal symptoms was associated with a decreased risk of discontinuation (HR 0.80; 95% CI 0.77–0.84; p ⬍ 0.0001). Similarly, patients treated by gynecologists were more persistent than patients treated by general practitioners (HR 0.88; 95% CI 0.84–0.92; p ⬍ 0.0001). Finally, MHT started in the years 2007–2009 was associated with higher discontinuation rates (HR 1.04; 95% CI 1.00–1.09; p ⫽ 0.0709) than MHT start in the years 2010–2013 (HR 0.90; 95% CI 0.85–0.94; p ⫽ 0.0001), which resulted in lower discontinuation rates. Number of co-medications was an additional variable of MHT treatment discontinuation (HR 1.04; 95% CI 1.02– 1.05; p ⬍ 0.0001).
DISCUSSION
Figure 1 Kaplan – Meier curves for persistency over 12 months for women taking oral combination menopausal hormone therapy (refi ll gap of 90 days)
740
This study investigated real-world persistence with combined MHT for 12 months from 2004 to 2014 in Germany. Overall, our results indicate that, after 12 months of follow-up, 44.6% of patients receiving 1 mg oral MHT and 33.5% of patients receiving 2 mg oral MHT were persistent (gap longer than 90 days). Furthermore, after 12 months of follow-up, 39.1% and 38.2% of patients using ⬍ 50 μg and ⫺ ⬎ 50 μg of transdermal MHT, respectively, were still on treatment. We further showed that patients over 60 years presented the highest discontinuation rates. Moreover, patients with menopausal symptoms and patients treated by gynecologists were more
Climacteric
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
Persistency and menopausal hormone therapy treatment-persistent. Most important, patients starting their MHT in the years 2010–2013 were more treatment-persistent than patients starting their therapy in the years before. Our results provide additional valuable information on persistence to MHT in the post-WHI era. Guay and colleagues demonstrated that clinical practice responded quickly to the results of ‘this largely mediatized study’18. They underlined that, 1 year after the first results of the WHI study were published, significant changes in the trends of use, women’s characteristics and in the doses of estrogen prescribed were apparent. The authors reported a significant decrease in persistence from 59% in the pre-WHI cohort to 45% in the post-WHI cohort. Interestingly, the proportion of new users presenting cardiovascular risk factors was not changed18. The decline in the total number of MHT prescriptions after 2002 was closely related to the WHI study’s negative reports. These findings are in line with the prescription trends in the Canadian (32%) and the US populations (37%)19,20. An additional study reported discontinuation rates of up to 38% in the first 6 months after release of the WHI study results through the lay press or medical literature21. The authors emphasized the negative impact of published negative results and their influence on pharmacotherapy decisions. Further literature from the Danish population stated that the majority of women experiencing a myocardial infarction during ongoing MHT continued the treatment after discharge and this pattern of MHT use did not change markedly after 200222. The most recent publication of persistency rates in the US population from 2000 to 2009 showed a stable decline and an increase in the use of lower-dose and vaginal MHT11. Buick and colleagues identified the reasons for use or discontinuation of MHT in a systematic literature research from 1980 to 200223. As such, the authors reported that the perceptions of benefits are often countered by concerns over potential adverse effects. Short-term symptom reliefs against long-term benefits were the two conflicting poles of decisionmaking with regard to MHT. As a result of this, in order to maintain a high MHT persistency rate, Vestergaard and colleagues suggested pragmatic approaches including offering of alternative MHT formulations to women not tolerating the first-line treatment24. Oren identified further motives for MHT discontinuation and stated high age, low education level, unemployment, treatment recommendation by a general practitioner and attitude towards MHT usage as factors associated with discontinuation25. A further study prospectively investigated factors associated with a low persistency towards MHT in the year 2003 – close to the publication of the WHI study – and supported that fear of breast cancer was the most common reason for early discontinuation26. Our results are congruent with these findings since we could confirm the low persistency rates in the early years after the WHI publication (lowest rate 29.5% in the year 2007) and the highest in recent years (highest rate 46.4% in the year 2013). In this context, Panay and colleagues scrutinized the risks and benefits of MHT in the most recent recommendations of the British Menopause Society and Women’s Health Concern in relation to breast cancer27. As such, the WHI estrogen and
Climacteric
Kyvernitakis et al. progestin study showed a small increase of 0.1% per annum in the risk of breast cancer after 5 years of MHT27. However, the WHI estrogen-alone arm (hysterectomized women only) showed a statistically significant decrease in breast cancer risk. Moreover, the Million Women Study (MWS) raised concerns over the long-term safety of MHT with regard to breast cancer. Overall, a recent critique on the WHI and the MWS has delivered a number of key weaknesses that limit the ability of trials to establish a causal association between MHT and breast cancer27. These approaches have been also considered in the recent recommendations of the International Menopause Society2. There are certain limitations to consider for the present analysis. First, the retrospective character of the study does not allow us to examine the dynamic changes within age groups and their co-morbidities. Consequently, it is not possible to prove that lower persistence in the early years of the post-WHI era was associated with the negative impact of the media. Second, no detailed documentation on the sideeffects of MHT was available in the database. However, we assigned the most common co-diagnoses to each therapy group. Finally, of particular concern is the lack of data with regard to the individual indication of MHT. It can be assumed that patients with severe menopausal symptoms comprise a higher level of ailment compared to women who use MHT for long-term preventative reasons; this could not be addressed, as the identification of these variables in the database used was not possible. Despite these limitations, our study has several strengths. This analysis used real-life data from gynecologists and GPs, indicating the realistic impact of access to medical care on treatment persistence. The present analysis investigated age and health-care provider-specific differences in patients receiving MHT for the first time. Additionally, for the first time in this regard, we stated that changes of persistence depended on the time of start of therapy over the years since the publication of the WHI study. In conclusion, this study is the first in which the yeardependent changes of discontinuation rates in the post-WHI times have been assessed. We showed that patients starting their treatments in recent years (2010–2013) were more treatment-persistent than patients beginning MHT in the early years after publication of the WHI study results (2004–2009). Altogether, our results indicate rather low persistency rates between 33.5% and 44.6% for different MHT dosages after 12 months of treatment. This underlines the need to improve long-term persistence in MHT in order to achieve the full benefit of the treatment. Our results support that administration of low-dose oral MHT and transdermal MHT is associated with higher persistency than higher doses of oral MHT.
Conflict of interest The authors report no confl ict of interest. The authors alone are responsible for the content and writing of this paper.
Source of funding
Nil.
741
Persistency and menopausal hormone therapy
Kyvernitakis et al.
Downloaded by [NUS National University of Singapore] at 00:14 06 November 2015
References 1. Rossouw JE, Anderson GL, Prentice RL, et al. 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 2. de Villiers TJ, Pines A, Panay N, et al. Updated 2013 International Menopause Society recommendations on menopausal hormone therapy and preventive strategies for midlife health. Climacteric 2013;16:316–37 3. Sturdee DW, Pines A, Archer DF; International Menopause Society Writing Group. Updated IMS recommendations on postmenopausal hormone therapy and preventive strategies for midlife health. Climacteric 2011;14:302–20 4. Jalava-Broman J, Makinen J, Ojanlatva A, Jokinen K, Sillanmaki L, Rautava P. Change in the frequency of HRT use from 2000 to 2005 and reasons to discontinue: follow-up of a normal cohort in Finland. Acta Obstet Gynecol Scand 2011;90:351–7 5. Kyvernitakis I, Ziller V, Hars O, Bauer M, Kalder M, Hadji P. Prevalence of menopausal symptoms and their influence on adherence in women with breast cancer. Climacteric 2014;17: 252–9 6. Grodstein F, Stampfer MJ, Colditz GA, et al. Postmenopausal hormone therapy and mortality. N Engl J Med 1997;336: 1769–75 7. Schneider DL, Barrett-Connor EL, Morton DJ. Timing of postmenopausal estrogen for optimal bone mineral density. The Rancho Bernardo Study. JAMA 1997;277:543–7 8. Jacobs DM, Tang MX, Stern Y, et al. Cognitive function in nondemented older women who took estrogen after menopause. Neurology 1998;50:368–73 9. Nanda K, Bastian LA, Hasselblad V, Simel DL. Hormone replacement therapy and the risk of colorectal cancer: a meta-analysis. Obstet Gynecol 1999;93:880–8 10. Regan MM, Emond SK, Attardo MJ, Parker RA, Greenspan SL. Why do older women discontinue hormone replacement therapy? J Women’s Health Gend Based Med 2001;10:343–50 11. Tsai SA, Stefanick ML, Stafford RS. Trends in menopausal hormone therapy use of US office-based physicians, 2000–2009. Menopause 2011;18:385–92 12. Isaacs AJ, Drew SV, McPherson K. UK women doctors’ use of hormone replacement therapy: 10-year follow up. Climacteric 2005;8:154–61 13. Becher H, Kostev K, Schroder-Bernhardi D. Validity and representativeness of the “Disease Analyzer” patient database for use in pharmacoepidemiological and pharmacoeconomic studies. Int J Clin Pharmacol Therapeut 2009;47:617–26 14. Hadji P, Claus V, Ziller V, Intorcia M, Kostev K, Steinle T. GRAND: the German retrospective cohort analysis on compliance and persistence and the associated risk of fractures in osteoporotic
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
women treated with oral bisphosphonates. Osteoporos Int 2012;23:223–31 Potocki M, Ziller R, Mueller C. Mid-regional pro-adrenomedullin in acute heart failure: a better biomarker or just another biomarker? Curr Heart Fail Rep 2012;9:244–51 Kyvernitakis I, Kostev K, Kurth A, Albert US, Hadji P. Differences in persistency with teriparatide in patients with osteoporosis according to gender and health care provider. Osteoporos Int 2014;25:2721–8 Motheral B, Brooks J, Clark MA, et al. A checklist for retrospective database studies – report of the ISPOR Task Force on Retrospective Databases. Value Health 2003;6:90–7 Guay MP, Dragomir A, Pilon D, Moride Y, Perreault S. Changes in pattern of use, clinical characteristics and persistence rate of hormone replacement therapy among postmenopausal women after the WHI publication. Pharmacoepidemiol Drug Safety 2007;16:17–27 Austin PC, Mamdani MM, Tu K, Jaakkimainen L. Prescriptions for estrogen replacement therapy in Ontario before and after publication of the Women’s Health Initiative Study. JAMA 2003;289:3241–2 Gavin NI, Thorp JM, Ohsfeldt RL. Determinants of hormone replacement therapy duration among postmenopausal women with intact uteri. Menopause 2001;8:377–83 Bestul MB, McCollum M, Hansen LB, Saseen JJ. Impact of the Women’s Health Initiative trial results on hormone replacement therapy. Pharmacotherapy 2004;24:495–9 Bretler DM, Hansen PR, Abildstrom SZ, et al. Use and discontinuation of hormone replacement therapy in women with myocardial infarction: a nationwide study. Br J Clin Pharmacol 2011;71:105–15 Buick DL, Crook D, Horne R. Women’s perceptions of hormone replacement therapy: risks and benefits (1980–2002). A literature review. Climacteric 2005;8:24–35 Vestergaard P, Hermann AP, Gram J, et al. Improving compliance with hormonal replacement therapy in primary osteoporosis prevention. Maturitas 1997;28:137–45 Oren A. Motives for initiation, temporary discontinuation, and permanent discontinuation of hormone replacement therapy use among Norwegian women. Maturitas 2009;64:33–7 Corrado F, D’Anna R, Caputo F, Cannata ML, Zoccali MG, Cancellieri F. Compliance with hormone replacement therapy in postmenopausal Sicilian women. Eur J Obstet Gynecol Reprod Biol 2005;118:225–8 Panay N, Hamoda H, Arya R, Savvas M; British Menopause Society, Women’s Health Concern. The 2013 British Menopause Society & Women’s Health Concern recommendations on hormone replacement therapy. Menopause Int 2013;19:59–68
Supplementary Data Supplementary Figure 1a,b,c,d and Figure 2a,b,c,d.
742
Climacteric
