Acta Oncologica, 2015; 54: 704–711
ORIGINAL ARTICLE
New and chronic use of hypnotics after diagnosis with early breast cancer. A retrospective cohort study
LÆRKE TOFTEGÅRD ANDERSEN1, NIS PALM SUPPLI2, SUSANNE OKSBJERG DALTON2, NIELS KROMAN3,4, JACOB ROSENBERG1 & ISMAIL GÖGENUR1 1Herlev
Hospital, University of Copenhagen, Department of Surgery, Herlev, Denmark, 2Danish Cancer Society Research Center, Unit of Survivorship, Copenhagen, Denmark, 3Department of Breast Surgery, Rigshospitalet, Copenhagen, Denmark and 4Danish Breast Cancer Cooperative Group, Rigshospitalet Department 2501, Copenhagen, Denmark ABSTRACT
Background. To determine use and investigate factors associated with use of hypnotics the first year after a diagnosis with breast cancer. Material and methods. A retrospective registry based cohort study linking clinical data from the Danish Breast Cancer Group with the National Prescription Drug Database and other health and administrative registries. We included 26 082 women diagnosed with early breast cancer as first time primary cancer during 1996–2006. Use of hypnotics was measured as redeemed prescriptions in the first year after diagnosis of early breast cancer. Prior use of hypnotics was defined as one or more prescriptions of hypnotics 13 months to 1 month before diagnosis, and chronic use was defined as four or more prescriptions. Hazard ratios (HRs) for clinical variables, treatment-related factors and sociodemographic factors were calculated. Results. Among women with no prior history of hypnotic use, 17% became new users with 4% on a chronic level. Among prior users, 82% continued their use with one or more prescriptions of hypnotics and 15% increased their use to a chronic level. Significantly increased hazard ratios for use of hypnotics were seen with increasing age [age 50–59 years: HR 1.43 (95% CI 1.31–1.57), age 60–69: HR 1.57 (95% CI 1.44–1.71)], increasing number of tumor positive lymph nodes [1–3 lymph nodes: HR 1.12 (95% CI 1.05–1.09), 4–9 lymph nodes: HR 1.11 (95% CI 1.02–1.29)], chemotherapy [HR 1.25 (95% CI 1.12–1.39)], somatic comorbidity [HR 1.31 (95% CI 1.21–1.42)] and use of antidepressants the year before breast cancer diagnosis [HR 1.97 (95% CI 1.85–2.10)]. Conclusion. This study detected a group of patients at great risk for initiating and increasing use of hypnotics and preventive and prophylactic mechanism should be investigated and initiated when this group of patients is seen in the clinical setting.
Breast cancer is globally the most prevalent type of cancer among women [1]. The incidence is increasing and combined with improving survival, focus is on the impact of long-term side effects of breast cancer diagnosis and treatment on rehabilitation and survivorship [2–4]. Sleep disturbances have been reported as an important concern during breast cancer treatment and rehabilitation [5]. Previous studies have reported a high incidence of sleep disturbances among women with breast cancer with up to 70% of the patients having sleep
disturbances at some point after their diagnosis [6–8]. Besides the negative effect on quality of life, sleep disturbances in patients with breast cancer is associated with development of depression, cognitive dysfunction, disturbed immune function and increased pain sensation [9,10]. Sleep disturbances are widely treated with hypnotics. However, because of the well known side effects and tolerance [11], health authorities recommend that pharmacological treatment with hypnotics should be limited to 1–4 weeks [12].
ECRS manuscript, from the European Cancer Rehabilitation & Survivorship Symposium (ECRS 2014) in Copenhagen in September 2014. Correspondence: L. T. Andersen, Sønder Boulevard 53, 5.4, 1720 København V., Denmark. Tel: ⫹ 45 28186323. E-mail: [email protected] (Received 27 October 2014 ; accepted 3 December 2014 ) ISSN 0284-186X print/ISSN 1651-226X online © 2015 Informa Healthcare DOI: 10.3109/0284186X.2014.998276
New and chronic use of hypnotics after diagnosis with early breast cancer Despite the increased focus on sleep disturbances in women with breast cancer, a limited number of studies have investigated the incidence of sleep disturbances and the use of hypnotics in this group. The few existing studies have reported that around one third of women diagnosed with breast cancer use hypnotics at some point after the diagnosis [9,13– 15]. However, these studies were limited by small sample sizes, self-report of use of hypnotics and heterogeneity in the definitions of hypnotics. Our aim was to provide the largest and first nationwide investigation of use of hypnotics and especially the rate of new users in women diagnosed with primary breast cancer. Furthermore, we aimed to evaluate whether disease-specific and patient-related factors were associated with the use of hypnotics.
705
excluded from the ATC main group of hypnotics N05C. Hypnotics dispensed during hospital stay are not registered in the National Prescription Drug Database. We used the 10-digit personal identification number that is assigned to all residents of Denmark [18] to obtain information on the number of prescriptions that were redeemed from the pharmacy by each woman in the cohort for the period 13 months to 1 month before and for the year after the breast cancer diagnosis. Number of redeemed prescriptions was evaluated as one, two, three, and four or more prescriptions. The cohort was divided in two groups according to use of hypnotics prior to diagnosis of breast cancer. Women, who had redeemed one or more prescriptions of hypnotics 13 months to 1 month before diagnosis of breast cancer, were considered to be prior users. We interpreted the need for four prescriptions or more as chronic use of hypnotics.
Material and methods Study population Since 1978 the Danish Breast Cancer Group (DBCG) has registered breast cancers diagnosed in Denmark [16]. In women below the age of 70 years and diagnosed with breast cancer, 97% of all breast cancers are registered and the database holds information on histopathological characteristics, treatment and relapse [16]. Through the DBCG we had access to date of diagnosis (defined as date of surgery), clinical variables (tumor size, tumorpositive axillary lymph nodes, hormone sensitivity, tumor grade and menopausal status), treatment modalities (type of surgery and adjuvant treatment; chemotherapy, radiation and endocrine therapy) for all women born 1925–1973 and diagnosed with breast cancer as first diagnosis of cancer between 1996 and 2006 (Table I). Information on use of hypnotics In Denmark, hypnotics are available only on prescription. The National Prescription Drug Database contains information on prescriptions for drugs dispensed in all pharmacies in Denmark from 1995 and onwards [17]. Each prescription record comprises valid information about the drug and date of dispensing. All drugs are described according to the Anatomical Therapeutic Chemical (ATC) classification system. We obtained information on all prescriptions for hypnotics in the ATC groups N05CD and N05CF dispensed at all Danish pharmacies from 1995 through 2006. Thus, hypnotics used primarily to treat anxiety and epilepsy (N05BA and N03AE01) were not defined as hypnotics in this study and further, drugs based on melatonin (N05CH) were
Information on sociodemographic and treatment-related factors Individual information on education and cohabitation status two years before diagnosis of breast cancer was retrieved by linking the study population to the population-based social registers administered by Statistics Denmark since 1980 [18,19]. Cohabitation was defined as, in the absence of marriage, two persons of the opposite sex, over the age of 16 years, with a maximum age difference of 15 years living at the same address with no other adult in residence. This definition therefore did not allow us to include homosexual unmarried couples and couples with more than 15 years in difference in age. The Danish National Patient Registry [20] provided information about comorbidity. Since 1977, administrative data and information about treatment and diagnosis for all patients admitted to hospitals in Denmark have been recorded here. On the basis of all discharge diagnoses since 1978, a Charlson Comorbidity Index score (CCI) [21] was computed for each woman at date of diagnosis. On the basis of the cumulated sum of scores, comorbidity was assessed in three groups (0, 1, and ⱖ 2). The index provides an overall score for comorbidity and is based on a weighted index of comorbidity consisting of 19 selected conditions scored from 1 to 3 or 6. Score 1: myocardial infarct, congestive heart failure, peripheral vascular disease, connective tissue disorder, ulcer disease, mild liver disease, diabetes types I and 2, cerebrovascular disease, dementia, chronic pulmonary disease. Score 2: hemiplegia, diabetes with endorgan damage type 1 or type 2, moderate to severe renal disease. Score 3: Moderate or severe liver disease. Score 6: AIDS.
706
L. T. Andersen et al.
Table I. Descriptive characteristics of a cohort of 26 082 Danish women diagnosed with breast cancer as first diagnosis of cancer between 1 February 1996 and 31 December 2006. Not prior users of hypnotics Characteristic Clinical variable Tumor size 95% percentile Median 5% percentile Tumor positive axillary lymph nodes 0 1–3 4–9 ⱖ 10 Tumor grade I II III Non-ductal carcinoma Missing Estrogen and progesterone receptor statusa Positive Negative Treatment-related factors Type of surgeryb Lumpectomy Mastectomy Adjuvant treatment No adjuvant treatment Chemotherapy Other adjuvant treatment Missing Sociodemographic factors Age and menopausal statusc 30–39 premenopausal 40–49 premenopausal 40–49 postmenopausal 50–59 premenopausal 50–59 postmenopausal 60–69 postmenopausal ⱖ 70 postmenopausal Cohabitation status Living with a partner Living alone Level of education Higher (⬎ 13 years) Vocational (10–12 years) Basic school (7–12 years) Unknown Charlson Comorbidity Index 0 1 ⱖ2 Use of antidepressants before BCd No Yes Year of diagnosis 1996–1998 1999–2002 2003–2006
Prior users of hypnotics
n ⫽ 22 537
(% )
n ⫽ 3545
(%)
5.0 1.9 0.7
– – –
4.5 1.9 0.7
– – –
11 657 6707 2549 1624
(52) (30) (11) (7)
1922 1008 367 248
(54) (28) (10) (7)
5453 7959 4683 4034 408
(24) (35) (21) (18) (2)
908 1198 641 736 62
(26) (34) (18) (21) (2)
17 895 4642
(79) (21)
2876 669
(81) (19)
8879 13 652
(39) (61)
1264 2278
(36) (64)
2706 6805 10 537 2489
(12) (30) (47) (11)
551 644 1845 505
(16) (18) (52) (14)
(100) (94) (6) (24) (76) (100) (100)
48 317 33 187 801 1375 784
(100) (91) (9) (19) (81) (100) (100)
16 743 5794
(74) (26)
2315 1230
(65) (35)
5192 7558 9454 333
(23) (34) (42) (1)
783 1076 1619 67
(22) (30) (46) (2)
19 795 1758 984
(88) (8) (4)
2726 498 321
(77) (14) (9)
20 670 1867
(92) (8)
2637 908
(74) (26)
4898 8378 9261
(22) (37) (41)
747 1282 1516
(21) (36) (43)
1209 4340 267 1790 5643 6471 2806
a ⱖ 1% tumor cells positive for estrogen and/or progesterone receptors; bBiopsy was taken in 9 women. These were included in the analysis, but results are not shown; cResults reported for combined variable. Women aged 30–39, 60–69 and ⱖ 70 had same menopausal status; dOne or more redeemed prescriptions of a drug in the ATC group N06A 13 months to one month before the diagnosis with early breast cancer.
New and chronic use of hypnotics after diagnosis with early breast cancer From the National Prescription Drug Database [17] we obtained information on prescription of antidepressant medication (ATC N06A) for the period 13 months to 1 month prior to diagnosis of breast cancer, as a proxy for depression. In the Danish Civil Registration System [18] we obtained vital status (death, emigration or disappearances) in 1996–2006. Statistical analysis Cox proportional hazards model was used to estimate relative risks, i.e. hazard ratios (HRs) and 95% confidence intervals (CIs), to study the association between clinical variables, treatment-related factors and sociodemographic factors and use of hypnotics after diagnosis of breast cancer. Time since diagnosis was used as underlying time scale. All analyses were conducted for the total cohort and separately for not prior users and prior users of hypnotics, respectively. The covariates included in the model were: clinical variables (tumor size, tumor-positive axillary lymph nodes, tumor grade, hormone receptor status), treatment-related variables (type of surgery, adjuvant treatment), sociodemographic factors (age, cohabitation status, level of education), the Charlson comorbidity index, prior use of antidepressants, and year of diagnosis. The proportional hazard assumption was evaluated for all variables. This was done by visually evaluating the graphs of the survival function versus the survival time and the graphs of the log [-log (survival)] versus log of survival time. The continuous variables; age, year of surgery, tumor size, and number of tumor positive axillary lymph nodes were tested for linearity by including higher order polynomials of the same variable in the model and testing for model reduction. According to these results the variable tumor size was included as a continuous variable in the analysis of both not prior and prior users. We investigated the association between variables hormone receptor status and chemotherapy, and age and menopausal status in order to assess whether individual effect of each variable could be identified from information in the present data. Only moderate association was seen between hormone receptor status and chemotherapy allowing estimation of both effects. On the contrary, menopausal status was a direct effect of age. To handle this we created a new variable that combined age and menopausal and results are reported for this combined variable. All tests and analyses used for the Cox models were done in the PHREG procedure of SAS 9.3 (SAS Institute Inc, Cary, NC, USA). All analyses were complete case analyses.
707
Results Study population We identified a cohort of 29 157 women diagnosed with primary early breast cancer as first cancer diagnosis between 1 February 1996, and 31 December 2006. We excluded 1158 women for whom information on socio-economic variables was not available and 1915 women for whom no information on tumor size or lymph node status or hormone sensitivity existed. Two women were excluded as the registered dates of emigration were prior to the date of breast cancer diagnosis, leaving a final study population of 26 082 women with the majority of the study population below the age of 70 years (Table I). Based on redeemed prescriptions in the 13 months to 1 month before diagnosis of primary breast cancer, 86% (n ⫽ 22 537) were categorized as not prior users, whereas 14% (n ⫽ 3545) of the cohort redeemed one or more prescriptions for hypnotics and were categorized as prior users. Use of hypnotics Of the 26 082 women in the cohort, 26% (n ⫽ 6700) redeemed one or more prescriptions of hypnotics the first year after the diagnosis with no regards to prior use. Among not prior users 17% (n ⫽ 3784) redeemed one or more prescriptions and 4% (n ⫽ 864) redeemed four or more prescriptions the first year after the diagnosis (Table II). Among prior users of hypnotics, 82% (n ⫽ 2916) redeemed one prescription or more the first year after the diagnosis and 45% (n ⫽ 1605) redeemed four or more prescriptions. Among prior
Table II. Number of Danish women with breast cancer that redeemed 1, 2, 3 or ⱖ 4 prescriptions of hypnotics the first year after the diagnosis. Prescriptions redeemed 0 1 2 3 ⱖ4 ⱖ1
Not prior users of hypnotics (%) N ⫽ 22 537 18 753 1814 722 384 864 3784
(83) (8) (3) (2) (4) (17)
Prior users of hypnotics (%) N ⫽ 3 545 629 (18) 436 (12) 477(14) 398 (11) 1605 (45)a 2916 (82)
0: Zero prescriptions redeemed of a drug with the ATC code N05CD and N05CF the first year after the diagnosis of early breast cancer. 1: One prescription redeemed; 2: Two prescriptions redeemed; 3: Three prescription redeemed; ⱖ 4: Four or more prescriptions redeemed. a1083 women also redeemed four or more prescriptions the year before the diagnosis. Thus, n ⫽ 522 (15% of the total group of prior users) increased their number of redeemed prescriptions to four or more during the first year after diagnosis of early breast cancer.
708
L. T. Andersen et al.
users, 15% (n ⫽ 522) increased their use of hypnotics from less than four prescription in the year prior to diagnosis to four or more prescriptions the first year after the diagnosis. For not prior users more than 70% of use was initiated within the first three months after the breast cancer diagnosis and for prior users this proportion was 79% (Figure 1). Factors associated with use of hypnotics Increasing number of tumor positive axillary lymph nodes were significantly associated with use of hypnotics among not prior users. Among prior users association for use was seen at 10 ⫹ positive lymph nodes (Table III). Negative receptor status was associated with an increased use of hypnotics among not prior users [HR 1.15 (95% CI 1.04–1.27)], but not among prior users [HR 0.87 (95% CI 0.77–1.00)]. Mastectomy compared to lumpectomy and chemotherapy compared to no adjuvant treatment were associated with use of hypnotics among not prior users [HR 1.10 (95% CI 1.01–1.19) and HR 1.48 (95% CI 1.27–1.71), respectively]. This was not the case for women who were prior users of hypnotics (Table III). Among both not prior and prior users we found a general pattern of increased risk of use of hypnotics in older women. With premenopausal women aged 40–49 as reference group we observed a HR of 0.65 [95% CI 0.54–0.77] among not prior users and a HR of 0.64 [95% CI 0.43–0.96] among prior users aged 30–39. In the group of not prior users significantly increased risk of use was observed in postmenopausal women aged 50–59 and 60–69 year old women. Postmenopausal women aged 50–59, women
Figure 1. Kaplan-Meier plot of the first redeemed prescription of hypnotics among Danish women the first year after the diagnosis with breast cancer among not prior and prior users of hypnotics.
aged 60–69, and women aged 70 or older were found at increased risk of use among prior users. Basic school education was associated with lower risk of use of hypnotics among not prior users [HR 0.88 (95% CI 0.81–0.96)], though associated with use among prior users [HR 1.10 (95% CI 1.00–1.21)]. Somatic comorbidity of 1 was associated with the use of hypnotics among both not prior and prior users [HR 1.18 (95% CI 1.05–1.32) and HR 1.17 (95% CI 1.05–1.30), respectively]. Somatic comorbidity of 2 ⫹ was associated with use among prior users, though associated with lower risk among not prior users [HR 0.83 (95% CI 0.70–0.99)]. Use of antidepressants in the year before diagnosis of breast cancer was significantly associated with use of hypnotics in both not prior and prior users [HR1.37 (95% CI 1.23–1.52) and HR1.34 (95% CI 1.23–1.45), respectively]. Diagnosis of breast cancer in the later period (2003–2006) compared to the earlier period (1996– 1998) were associated with lower risk of use of hypnotics among both not prior [HR 0.90 (95% CI 0.83–0.99)] and prior users of hypnotics [HR 0.89 (95% CI 0.80–0.98)]. Discussion We showed that 26% of women with breast cancer were treated with hypnotics during the first year after the diagnosis. Among not prior users, 17% became first time users and 4% developed a chronic use (four or more prescriptions of hypnotics). Among prior users 15% increased their use of hypnotics to chronic use after the diagnosis. Use of hypnotics on a chronic basis is associated with impairment of cognitive and psychomotor function [11] and daytime sedation, and is associated with risk for tolerance, dependency and mortality [22,23]. International guidelines recommend 2–4 weeks of use [12] even though side effects may also occur on a short-term basis [22]. As a result of this limited usability of hypnotics other evidence-based treatments should be considered [24]. In addition to being a symptom of clinical depression, sleep disturbances are often comorbid with other psychiatric disorders [25]. Previous studies have concluded that psychological factors were predictors of sleep disturbances, and that cancer-related factors were secondary in the developing and maintaining of sleep disturbances [6,7,26,27]. Our study detected that prior use of antidepressants, as an indicator of prior history of clinical depression, almost doubled the risk for use of hypnotics compared to women with no prior history of use of antidepressants. In addition, we found that cancer-related
New and chronic use of hypnotics after diagnosis with early breast cancer
709
Table III. Adjusted hazard ratios for use of hypnotics among Danish women the first year after the diagnosis with breast cancer. Total cohort n ⫽ 26 082
Clinical variable Tumor size Per cm Tumor positive axillary lymph nodes 0 1–3 4–9 ⱖ 10 Tumor grade I II III Non-ductal carcinoma Missing Hormone sensitivea Positive Negative Treatment-related factors Type of surgeryb Lumpectomy Mastectomy Adjuvant treatment No adjuvant treatment Chemotherapy Other adjuvant treatment Missing Sociodemographic factors Age and menopausal statusc 30–39 premenopausal 40–49 premenopausal 40–49 postmenopausal 50–59 premenopausal 50–59 postmenopausal 60–69 postmenopausal ⱖ 70 postmenopausal Cohabitation status Living with a partner Living alone Level of education Higher (⬎ 13 years) Vocational (10–12 years) Basic school (7–12 years) Charlson Comorbidity Index 0 1 ⱖ2 Use of antidepressants before BCd No Yes Year of diagnosis 1996–1998 1999–2002 2003–2006
Hazard ratio
95% CI
0.99
0.97–1.01
1 1.12 1.11 1.25
Not prior users of hypnotics n ⫽ 22 537 Hazard ratio
Prior users of hypnotics n ⫽ 3545
95% CI
Hazard ratio
95% CI
1.01
0.98–1.03
1.01
0.98–1.04
– 1.05–1.19 1.02–1.21 1.14–1.38
1 1.18 1.18 1.33
– 1.09–1.28 1.06–1.32 1.17–1.51
1 1.01 1.02 1.15
– 0.92–1.11 0.89–1.16 0.98–1.34
1 0.97 0.99 1.01 0.94
– 0.90–1.03 0.91–1.08 0.93–1.09 0.78–1.14
1 0.97 1.00 0.98 0.90
– 0.89–1.06 0.90–1.11 0.88–1.09 0.70–1.17
1 0.93 0.99 0.91 0.95
– 0.84–1.03 0.87–1.12 0.82–1.02 0.71–1.27
1 1.06
– 0.98–1.15
1 1.15
– 1.04–1.26
1 0.87
– 0.77–0.99
1 1.10
– 1.03–1.16
1 1.10
– 1.01–1.19
1 1.01
– 0.92–1.10
1 1.25 0.98 1.15
– 1.12–1.39 0.89–1.07 1.04–1.28
1 1.48 1.04 1.29
– 1.27–1.72 0.91–1.19 1.09–1.48
1 1.06 0.97 0.99
– 0.89–1.25 0.85–1.11 0.85–1.15
0.62 1 1.07 1.18 1.43 1.57 1.59
0.53–0.73 – 0.83–1.40 1.06–1.32 1.31–1.57 1.44–1.71 1.43–1.77
0.65 1 0.92 1.11 1.30 1.22 0.99
0.54–0.78 – 0.66–1.28 0.98–1.27 1.17–1.45 1.10–1.36 0.86–1.15
0.64 1 1.22 1.12 1.31 1.34 1.44
0.43–0.96 – 0.80–1.87 0.91–1.38 1.11–1.54 1.15–1.58 1.22–1.71
1 1.04
– 0.98–1.09
1 0.94
– 0.87–1.02
1 1.02
– 0.95–1.11
1 0.88 0.87
– 0.83–0.94 0.82–0.93
1 0.92 0.88
– 0.84–1.00 0.81–0.96
1 0.99 1.10
– 0.89–1.10 1.00–1.21
1 1.31 1.20
– 1.21–1.42 1.08–1.32
1 1.18 0.83
– 1.05–1.32 0.70–0.99
1 1.17 1.24
– 1.05–1.30 1.09–1.40
1 1.97
– 1.85–2.10
1 1.37
– 1.23–1.52
1 1.34
– 1.23–1.45
1 0.99 0.89
– 0.92–1.06 0.83–0.95
1 1.00 0.90
– 0.91–1.09 0.82–0.99
1 1.04 0.89
– 0.94–1.15 0.80–0.98
All three models included all shown variables. Tumor size was evaluated as a continuous variable. aIn the years 1996–2006 defined as ⱖ 10% tumor cells positive for estrogen and/or progesterone receptors; bBiopsy was taken in 9 women. These were included in the analysis, but results are not shown; c Results reported for combined variable. Women aged 30–39, 60–69 and ⱖ 70 had same menopausal status; dOne or more redeemed prescriptions of a drug in the ATC group N06A 13 months to 1 month before the diagnosis with early breast cancer.
710
L. T. Andersen et al.
factors, such as having tumor positive axillary lymph nodes and receiving chemotherapy, were significantly associated with use of hypnotics among not prior users. Hormones receptor status affects treatment choices and women with HR negative breast cancer are mostly receiving chemotherapy. In this study, no correlation was seen between HR negative status and chemotherapy, so this is unlikely to have biased our findings. Interestingly, among prior users breast cancer-related factors seemed to have almost no association with use of hypnotics. It has previously been reported that after surgery, being younger may be a risk factor for development of somatic comorbidity with persistent pain and arm lymph-edema [28,29]. Furthermore, a recent review reported that younger women with breast cancer compared to older women with breast cancer, have poorer quality of life [30]. One may argue that these younger women would be at risk for use of hypnotics but our study in fact indicated that being ⱕ 40 years of age were associated with reduced risk for use of hypnotics among not prior and prior users of hypnotics. This is of interest, as our study found that somatic comorbidity increased the risk for using hypnotics and it seems that increasing age, no matter prior history of use of hypnotics, and having comorbidity independently contributed to the use of hypnotics in these women. We showed that the majority of the first redeemed prescriptions took place the first three months after surgery. This supports the findings from previous small studies reporting that 19–60% of women with breast cancer subjectively reported sleep disturbances the first two [6,31] to four months [7] after surgery. Our findings have implications for clinical practice, as we did separate analyses on not prior and prior users of hypnotics, thereby making it easier in the clinical setting to recognize women at risk for developing sleep disturbance. At this moment studies regarding preventive and prophylactic treatments of sleep disturbance in women with breast cancer, are being investigated [32]. We showed a difference in use of hypnotics and factors associated with use of hypnotics between not prior and prior users. We observed differences in associations in both factors related to the disease severity and to treatment, especially chemotherapy. Furthermore, basic education seemed to be associated with more use among prior users whereas it was associated with reduced use among not prior users, indicating that having a breast cancer and associated sleeping problems may be handled differently by socioeconomic position.
Our study has some limitations. Firstly, we did not have a control group and could not address if there was a possible additive hypnotic use in postmenopausal women. Furthermore, we did not compare our group with patients with other serious diseases. However, we showed that around 258/1000 women with breast cancers were prescribed a hypnotic in the ATC groups N05DC and N05CF. In the Danish female population, 58/1000 women and 17/1000 women were prescribed a N05DC and N05CF in 2005, respectively [33]. Thus, our data indicate that there is a marked increased use of hypnotics in women with breast cancer, which is confirmed in our findings of 17% new users. Second, we did not know if patients had used hypnotics during the hospital stay or if they were discharged with hypnotics from the hospital. If anything, this might underestimate the earliest use of hypnotics after breast cancer diagnosis. However, this is unlikely to bias our findings, since Danish women with breast cancer in the period of investigation were hospitalized on average three days [34]. Third, we only investigated drugs in the ATC groups N05CD and N05CF. We were only interested in hypnotics prescribed for sleep and not mainly prescribed for anxiety or epilepsy. We were not interested in the use of over the counter hypnotics which in some countries include the hypnotic melatonin. In Denmark, melatonin is only available on prescription and was first marketed in 2007 and therefore could not be investigated in this study. Fourth, we did not investigate the effect of sentinel node technique on use of hypnotics, however, only a limited number of patients have had sentinel node surgery in our cohort. In conclusion, this study found that one in four women used hypnotics the first year after the diagnosis and 17% became new users of hypnotic. A diagnosis with breast cancer was associated with the initiation of chronic hypnotic use in 4% of not prior users and 15% of prior users of hypnotics. A range of clinical, treatment and sociodemographic factors were associated with the use of hypnotics among not prior users whereas these associations were not as strong among prior users of hypnotics. These findings suggest the need for further focus on women with breast cancer and alternative prophylactic treatment modalities are warranted to prevent the development of new and potentially chronic users of hypnotics. Declaration of interest: The corresponding author affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been
New and chronic use of hypnotics after diagnosis with early breast cancer omitted; and that any discrepancies from the study as planned have been explained. The authors declare that they have no commercial or other associations that might pose a conflict of interest in connection with this article.
[17] [18] [19]
References [1] Bray F, Ren JS, Masuyer A, Ferlay J. Global estimates of cancer prevalence in 27 sites in the adult population in 2008. Int J Cancer 2013;132:1133–45. [2] Autier P, Boniol M, La Vecchia C, Vatten L, Gavin A, Héry C, et al. Disparities in breast cancer mortality trends between 30 European countries: Retrospective trend analysis of WHO mortality database. Br Med J 2010;341:c3620. [3] Rowland JH, Hewitt M, Ganz PA. Cancer survivorship: A new challenge in delivering quality cancer care. J Clin Oncol 2006;24:5101–4. [4] Dalton SO, Bidstrup PE, Johansen C. Rehabilitation of cancer patients: Needed, but how? Acta Oncol 2011;50:163–6. [5] Ewertz M, Jensen AB. Late effects of breast cancer treatment and potentials for rehabilitation. Acta Oncol 2011;50: 187–93. [6] Savard J, Ivers H, Villa J, Caplette-Gingras A, Moris CM. Natural course of insomnia comorbid with cancer: An 18-month longitudinal study. J Clin Oncol 2011;29: 3580–6. [7] Colagiuri B, Christensen S, Jensen AB, Price MA, Butow PN, Zachariae R. Prevalence and predictors of sleep difficulty in a national cohort of women with primary breast cancer three to four months postsurgery. J Pain Symptom Manage 2011;42:710–20. [8] Savard J, Simard S, Blanchet J, Ivers H, Morin CM. Prevalence, clinical characteristics, and risk factors for insomnia in the context of breast cancer. Sleep 2001;24:583–90. [9] Fortner BV, Stepanski EJ, Wang SC, Kasprowicz S, Durrence HH . Sleep and quality of life in breast cancer patients. J Pain Symptom Manage 2002;24:471–80. [10] Redwine L, Hauger RL, Gillin JC, Irwin M. Effects of sleep and sleep deprivation on interleukin-6, growth hormone, cortisol, and melatonin levels in humans. J Clin Endocrinol Metab 2000;85:3597–603. [11] Glass J, Lanctôt KL, Herrmann N, Sproule BA, Busto UE. Sedative hypnotics in older people with insomnia: Metaanalysis of risks and benefits. Br Med J 2005;331:1169. [12] National Institutes of Health State of the Science Conference statement on Manifestations and Management of Chronic Insomnia in Adults, June 13–15, 2005. Sleep 2005;28:1049–57. [13] de Bock GH, Musters RF, Bos HJ, Schröder CP, Mourits MJ, de Jong-van den Berg LT. Psychotropic medication during endocrine treatment for breast cancer. Support Care Cancer 2012;20:1533–40. [14] Moore TA, Berger AM, Dizona P. Sleep aid use during and following breast cancer adjuvant chemotherapy. Psychooncology 2011;20:321–5. [15] Costantini C, Ale-Ali A, Helsten T. Sleep aid prescribing practices during neoadjuvant or adjuvant chemotherapy for breast cancer. J Palliat Med 2011;14:563–6. [16] Møller S, Jensen M-B, Ejlertsen B, Bjerre KD, Larsen M, Hansen HB, et al.The clinical database and the treatment guidelines of the Danish Breast Cancer Cooperative Group
[20] [21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33] [34]
711
(DBCG); its 30-years experience and future promise. Acta Oncol 2008;47:506–24. Gaist D, Sørensen HT, Hallas J. The Danish prescription registries. Dan Med Bull 1997;44:445–8. Pedersen CB. The Danish Civil Registration System. Scand J Public Health 2011;39:22–5. Jensen VM, Rasmussen AW. Danish Education Registers. Scand J Public Health 2011;39:91–4. Lynge E, Sandegaard JL, Rebolj M. The Danish National Patient Register. Scand J Public Health 2011;39:30–3. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 1987;40:373–83. Lader M. Benzodiazepines revisited – will we ever learn? Addiction 2011;106:2086–109. Wei YJ, Simoni-Wastila L, Zuckerman IH, Huang TY, Brandt N, Moyo P, et al. Quality of psychopharmacological medication prescribing and mortality in Medicare beneficiaries in nursing homes. J Am Geriatr Soc 2014; 62:1490–504. Andersen SR, Würtzen H, Steding-Jessen M et al. Effect of mindfulness-based stress reduction on sleep quality: Results of a randomized trial among Danish breast cancer patients. Acta Oncol 2013;52:336–44. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. Washington, DC: American Psychiatric Association; 2000. [cited 2103 Jun 3]. Available from: http://www.psychiatry.org/practice/clinical-practiceguidelines. Bardwell WA, Profant J, Casden DR, Dimsdale JE, Ancoli-Israel S, Natarajan L, et al. The relative importance of specific risk factors for insomnia in women treated for early-stage breast cancer. Psychooncology 2008;17:9–18. Davidson JR, MacLean AW, Brundage MD, Schlze K. Sleep disturbance in cancer patients. Soc Sci Med 2002; 54:1309–21. Gartner R, Jensen MB, Nielsen J, Ewertz M, Kroman N, Kehlet H. Prevalence of and factors associated with persistent pain following breast cancer surgery. JAMA 2009; 302:1985–92. Gartner R, Jensen MB, Kronborg L, Ewertz M, Kehlet H, Kroman N. Self-reported arm-lymphedema and functional impairment after breast cancer treatment – a nationwide study of prevalence and associated factors. Breast 2010;19:506–15. Howard-Anderson J, Ganz PA, Bower JE, Stanton AL. Quality of life, fertility concerns, and behavioral health outcomes in younger breast cancer survivors: A systematic review. J Natl Cancer Inst 2012;104:386–405. Savard J, Villa J, Ivers H, Simard S, Morin CM. Prevalence, natural course, and risk factors of insomnia comorbid with cancer over a 2-month period. J Clin Oncol 2009;27:5233–9. Hansen MV, Madsen MT, Hageman I, Rasmussen LS, Bokmand S, Rosenberg J, et al. The effect of MELatOnin on Depression, anxietY, cognitive function and sleep disturbances in patients with breast cancer. The MELODY trial: Protocol for a randomised, placebo-controlled, doubleblinded trial. 2012;2:e000647. Medstat.dk. [cited 2013 Jun 3]. Available from: http://www. medstat.dk. Mertz BG, Kroman N, Williams H, Kehlet H. Fast-track surgery for breast cancer is possible. Dan Med J 2013;60:A4615.
Copyright of Acta Oncologica is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
New and chronic use of hypnotics after diagnosis with early breast cancer. A retrospective cohort study
LÆRKE TOFTEGÅRD ANDERSEN1, NIS PALM SUPPLI2, SUSANNE OKSBJERG DALTON2, NIELS KROMAN3,4, JACOB ROSENBERG1 & ISMAIL GÖGENUR1 1Herlev
Hospital, University of Copenhagen, Department of Surgery, Herlev, Denmark, 2Danish Cancer Society Research Center, Unit of Survivorship, Copenhagen, Denmark, 3Department of Breast Surgery, Rigshospitalet, Copenhagen, Denmark and 4Danish Breast Cancer Cooperative Group, Rigshospitalet Department 2501, Copenhagen, Denmark ABSTRACT
Background. To determine use and investigate factors associated with use of hypnotics the first year after a diagnosis with breast cancer. Material and methods. A retrospective registry based cohort study linking clinical data from the Danish Breast Cancer Group with the National Prescription Drug Database and other health and administrative registries. We included 26 082 women diagnosed with early breast cancer as first time primary cancer during 1996–2006. Use of hypnotics was measured as redeemed prescriptions in the first year after diagnosis of early breast cancer. Prior use of hypnotics was defined as one or more prescriptions of hypnotics 13 months to 1 month before diagnosis, and chronic use was defined as four or more prescriptions. Hazard ratios (HRs) for clinical variables, treatment-related factors and sociodemographic factors were calculated. Results. Among women with no prior history of hypnotic use, 17% became new users with 4% on a chronic level. Among prior users, 82% continued their use with one or more prescriptions of hypnotics and 15% increased their use to a chronic level. Significantly increased hazard ratios for use of hypnotics were seen with increasing age [age 50–59 years: HR 1.43 (95% CI 1.31–1.57), age 60–69: HR 1.57 (95% CI 1.44–1.71)], increasing number of tumor positive lymph nodes [1–3 lymph nodes: HR 1.12 (95% CI 1.05–1.09), 4–9 lymph nodes: HR 1.11 (95% CI 1.02–1.29)], chemotherapy [HR 1.25 (95% CI 1.12–1.39)], somatic comorbidity [HR 1.31 (95% CI 1.21–1.42)] and use of antidepressants the year before breast cancer diagnosis [HR 1.97 (95% CI 1.85–2.10)]. Conclusion. This study detected a group of patients at great risk for initiating and increasing use of hypnotics and preventive and prophylactic mechanism should be investigated and initiated when this group of patients is seen in the clinical setting.
Breast cancer is globally the most prevalent type of cancer among women [1]. The incidence is increasing and combined with improving survival, focus is on the impact of long-term side effects of breast cancer diagnosis and treatment on rehabilitation and survivorship [2–4]. Sleep disturbances have been reported as an important concern during breast cancer treatment and rehabilitation [5]. Previous studies have reported a high incidence of sleep disturbances among women with breast cancer with up to 70% of the patients having sleep
disturbances at some point after their diagnosis [6–8]. Besides the negative effect on quality of life, sleep disturbances in patients with breast cancer is associated with development of depression, cognitive dysfunction, disturbed immune function and increased pain sensation [9,10]. Sleep disturbances are widely treated with hypnotics. However, because of the well known side effects and tolerance [11], health authorities recommend that pharmacological treatment with hypnotics should be limited to 1–4 weeks [12].
ECRS manuscript, from the European Cancer Rehabilitation & Survivorship Symposium (ECRS 2014) in Copenhagen in September 2014. Correspondence: L. T. Andersen, Sønder Boulevard 53, 5.4, 1720 København V., Denmark. Tel: ⫹ 45 28186323. E-mail: [email protected] (Received 27 October 2014 ; accepted 3 December 2014 ) ISSN 0284-186X print/ISSN 1651-226X online © 2015 Informa Healthcare DOI: 10.3109/0284186X.2014.998276
New and chronic use of hypnotics after diagnosis with early breast cancer Despite the increased focus on sleep disturbances in women with breast cancer, a limited number of studies have investigated the incidence of sleep disturbances and the use of hypnotics in this group. The few existing studies have reported that around one third of women diagnosed with breast cancer use hypnotics at some point after the diagnosis [9,13– 15]. However, these studies were limited by small sample sizes, self-report of use of hypnotics and heterogeneity in the definitions of hypnotics. Our aim was to provide the largest and first nationwide investigation of use of hypnotics and especially the rate of new users in women diagnosed with primary breast cancer. Furthermore, we aimed to evaluate whether disease-specific and patient-related factors were associated with the use of hypnotics.
705
excluded from the ATC main group of hypnotics N05C. Hypnotics dispensed during hospital stay are not registered in the National Prescription Drug Database. We used the 10-digit personal identification number that is assigned to all residents of Denmark [18] to obtain information on the number of prescriptions that were redeemed from the pharmacy by each woman in the cohort for the period 13 months to 1 month before and for the year after the breast cancer diagnosis. Number of redeemed prescriptions was evaluated as one, two, three, and four or more prescriptions. The cohort was divided in two groups according to use of hypnotics prior to diagnosis of breast cancer. Women, who had redeemed one or more prescriptions of hypnotics 13 months to 1 month before diagnosis of breast cancer, were considered to be prior users. We interpreted the need for four prescriptions or more as chronic use of hypnotics.
Material and methods Study population Since 1978 the Danish Breast Cancer Group (DBCG) has registered breast cancers diagnosed in Denmark [16]. In women below the age of 70 years and diagnosed with breast cancer, 97% of all breast cancers are registered and the database holds information on histopathological characteristics, treatment and relapse [16]. Through the DBCG we had access to date of diagnosis (defined as date of surgery), clinical variables (tumor size, tumorpositive axillary lymph nodes, hormone sensitivity, tumor grade and menopausal status), treatment modalities (type of surgery and adjuvant treatment; chemotherapy, radiation and endocrine therapy) for all women born 1925–1973 and diagnosed with breast cancer as first diagnosis of cancer between 1996 and 2006 (Table I). Information on use of hypnotics In Denmark, hypnotics are available only on prescription. The National Prescription Drug Database contains information on prescriptions for drugs dispensed in all pharmacies in Denmark from 1995 and onwards [17]. Each prescription record comprises valid information about the drug and date of dispensing. All drugs are described according to the Anatomical Therapeutic Chemical (ATC) classification system. We obtained information on all prescriptions for hypnotics in the ATC groups N05CD and N05CF dispensed at all Danish pharmacies from 1995 through 2006. Thus, hypnotics used primarily to treat anxiety and epilepsy (N05BA and N03AE01) were not defined as hypnotics in this study and further, drugs based on melatonin (N05CH) were
Information on sociodemographic and treatment-related factors Individual information on education and cohabitation status two years before diagnosis of breast cancer was retrieved by linking the study population to the population-based social registers administered by Statistics Denmark since 1980 [18,19]. Cohabitation was defined as, in the absence of marriage, two persons of the opposite sex, over the age of 16 years, with a maximum age difference of 15 years living at the same address with no other adult in residence. This definition therefore did not allow us to include homosexual unmarried couples and couples with more than 15 years in difference in age. The Danish National Patient Registry [20] provided information about comorbidity. Since 1977, administrative data and information about treatment and diagnosis for all patients admitted to hospitals in Denmark have been recorded here. On the basis of all discharge diagnoses since 1978, a Charlson Comorbidity Index score (CCI) [21] was computed for each woman at date of diagnosis. On the basis of the cumulated sum of scores, comorbidity was assessed in three groups (0, 1, and ⱖ 2). The index provides an overall score for comorbidity and is based on a weighted index of comorbidity consisting of 19 selected conditions scored from 1 to 3 or 6. Score 1: myocardial infarct, congestive heart failure, peripheral vascular disease, connective tissue disorder, ulcer disease, mild liver disease, diabetes types I and 2, cerebrovascular disease, dementia, chronic pulmonary disease. Score 2: hemiplegia, diabetes with endorgan damage type 1 or type 2, moderate to severe renal disease. Score 3: Moderate or severe liver disease. Score 6: AIDS.
706
L. T. Andersen et al.
Table I. Descriptive characteristics of a cohort of 26 082 Danish women diagnosed with breast cancer as first diagnosis of cancer between 1 February 1996 and 31 December 2006. Not prior users of hypnotics Characteristic Clinical variable Tumor size 95% percentile Median 5% percentile Tumor positive axillary lymph nodes 0 1–3 4–9 ⱖ 10 Tumor grade I II III Non-ductal carcinoma Missing Estrogen and progesterone receptor statusa Positive Negative Treatment-related factors Type of surgeryb Lumpectomy Mastectomy Adjuvant treatment No adjuvant treatment Chemotherapy Other adjuvant treatment Missing Sociodemographic factors Age and menopausal statusc 30–39 premenopausal 40–49 premenopausal 40–49 postmenopausal 50–59 premenopausal 50–59 postmenopausal 60–69 postmenopausal ⱖ 70 postmenopausal Cohabitation status Living with a partner Living alone Level of education Higher (⬎ 13 years) Vocational (10–12 years) Basic school (7–12 years) Unknown Charlson Comorbidity Index 0 1 ⱖ2 Use of antidepressants before BCd No Yes Year of diagnosis 1996–1998 1999–2002 2003–2006
Prior users of hypnotics
n ⫽ 22 537
(% )
n ⫽ 3545
(%)
5.0 1.9 0.7
– – –
4.5 1.9 0.7
– – –
11 657 6707 2549 1624
(52) (30) (11) (7)
1922 1008 367 248
(54) (28) (10) (7)
5453 7959 4683 4034 408
(24) (35) (21) (18) (2)
908 1198 641 736 62
(26) (34) (18) (21) (2)
17 895 4642
(79) (21)
2876 669
(81) (19)
8879 13 652
(39) (61)
1264 2278
(36) (64)
2706 6805 10 537 2489
(12) (30) (47) (11)
551 644 1845 505
(16) (18) (52) (14)
(100) (94) (6) (24) (76) (100) (100)
48 317 33 187 801 1375 784
(100) (91) (9) (19) (81) (100) (100)
16 743 5794
(74) (26)
2315 1230
(65) (35)
5192 7558 9454 333
(23) (34) (42) (1)
783 1076 1619 67
(22) (30) (46) (2)
19 795 1758 984
(88) (8) (4)
2726 498 321
(77) (14) (9)
20 670 1867
(92) (8)
2637 908
(74) (26)
4898 8378 9261
(22) (37) (41)
747 1282 1516
(21) (36) (43)
1209 4340 267 1790 5643 6471 2806
a ⱖ 1% tumor cells positive for estrogen and/or progesterone receptors; bBiopsy was taken in 9 women. These were included in the analysis, but results are not shown; cResults reported for combined variable. Women aged 30–39, 60–69 and ⱖ 70 had same menopausal status; dOne or more redeemed prescriptions of a drug in the ATC group N06A 13 months to one month before the diagnosis with early breast cancer.
New and chronic use of hypnotics after diagnosis with early breast cancer From the National Prescription Drug Database [17] we obtained information on prescription of antidepressant medication (ATC N06A) for the period 13 months to 1 month prior to diagnosis of breast cancer, as a proxy for depression. In the Danish Civil Registration System [18] we obtained vital status (death, emigration or disappearances) in 1996–2006. Statistical analysis Cox proportional hazards model was used to estimate relative risks, i.e. hazard ratios (HRs) and 95% confidence intervals (CIs), to study the association between clinical variables, treatment-related factors and sociodemographic factors and use of hypnotics after diagnosis of breast cancer. Time since diagnosis was used as underlying time scale. All analyses were conducted for the total cohort and separately for not prior users and prior users of hypnotics, respectively. The covariates included in the model were: clinical variables (tumor size, tumor-positive axillary lymph nodes, tumor grade, hormone receptor status), treatment-related variables (type of surgery, adjuvant treatment), sociodemographic factors (age, cohabitation status, level of education), the Charlson comorbidity index, prior use of antidepressants, and year of diagnosis. The proportional hazard assumption was evaluated for all variables. This was done by visually evaluating the graphs of the survival function versus the survival time and the graphs of the log [-log (survival)] versus log of survival time. The continuous variables; age, year of surgery, tumor size, and number of tumor positive axillary lymph nodes were tested for linearity by including higher order polynomials of the same variable in the model and testing for model reduction. According to these results the variable tumor size was included as a continuous variable in the analysis of both not prior and prior users. We investigated the association between variables hormone receptor status and chemotherapy, and age and menopausal status in order to assess whether individual effect of each variable could be identified from information in the present data. Only moderate association was seen between hormone receptor status and chemotherapy allowing estimation of both effects. On the contrary, menopausal status was a direct effect of age. To handle this we created a new variable that combined age and menopausal and results are reported for this combined variable. All tests and analyses used for the Cox models were done in the PHREG procedure of SAS 9.3 (SAS Institute Inc, Cary, NC, USA). All analyses were complete case analyses.
707
Results Study population We identified a cohort of 29 157 women diagnosed with primary early breast cancer as first cancer diagnosis between 1 February 1996, and 31 December 2006. We excluded 1158 women for whom information on socio-economic variables was not available and 1915 women for whom no information on tumor size or lymph node status or hormone sensitivity existed. Two women were excluded as the registered dates of emigration were prior to the date of breast cancer diagnosis, leaving a final study population of 26 082 women with the majority of the study population below the age of 70 years (Table I). Based on redeemed prescriptions in the 13 months to 1 month before diagnosis of primary breast cancer, 86% (n ⫽ 22 537) were categorized as not prior users, whereas 14% (n ⫽ 3545) of the cohort redeemed one or more prescriptions for hypnotics and were categorized as prior users. Use of hypnotics Of the 26 082 women in the cohort, 26% (n ⫽ 6700) redeemed one or more prescriptions of hypnotics the first year after the diagnosis with no regards to prior use. Among not prior users 17% (n ⫽ 3784) redeemed one or more prescriptions and 4% (n ⫽ 864) redeemed four or more prescriptions the first year after the diagnosis (Table II). Among prior users of hypnotics, 82% (n ⫽ 2916) redeemed one prescription or more the first year after the diagnosis and 45% (n ⫽ 1605) redeemed four or more prescriptions. Among prior
Table II. Number of Danish women with breast cancer that redeemed 1, 2, 3 or ⱖ 4 prescriptions of hypnotics the first year after the diagnosis. Prescriptions redeemed 0 1 2 3 ⱖ4 ⱖ1
Not prior users of hypnotics (%) N ⫽ 22 537 18 753 1814 722 384 864 3784
(83) (8) (3) (2) (4) (17)
Prior users of hypnotics (%) N ⫽ 3 545 629 (18) 436 (12) 477(14) 398 (11) 1605 (45)a 2916 (82)
0: Zero prescriptions redeemed of a drug with the ATC code N05CD and N05CF the first year after the diagnosis of early breast cancer. 1: One prescription redeemed; 2: Two prescriptions redeemed; 3: Three prescription redeemed; ⱖ 4: Four or more prescriptions redeemed. a1083 women also redeemed four or more prescriptions the year before the diagnosis. Thus, n ⫽ 522 (15% of the total group of prior users) increased their number of redeemed prescriptions to four or more during the first year after diagnosis of early breast cancer.
708
L. T. Andersen et al.
users, 15% (n ⫽ 522) increased their use of hypnotics from less than four prescription in the year prior to diagnosis to four or more prescriptions the first year after the diagnosis. For not prior users more than 70% of use was initiated within the first three months after the breast cancer diagnosis and for prior users this proportion was 79% (Figure 1). Factors associated with use of hypnotics Increasing number of tumor positive axillary lymph nodes were significantly associated with use of hypnotics among not prior users. Among prior users association for use was seen at 10 ⫹ positive lymph nodes (Table III). Negative receptor status was associated with an increased use of hypnotics among not prior users [HR 1.15 (95% CI 1.04–1.27)], but not among prior users [HR 0.87 (95% CI 0.77–1.00)]. Mastectomy compared to lumpectomy and chemotherapy compared to no adjuvant treatment were associated with use of hypnotics among not prior users [HR 1.10 (95% CI 1.01–1.19) and HR 1.48 (95% CI 1.27–1.71), respectively]. This was not the case for women who were prior users of hypnotics (Table III). Among both not prior and prior users we found a general pattern of increased risk of use of hypnotics in older women. With premenopausal women aged 40–49 as reference group we observed a HR of 0.65 [95% CI 0.54–0.77] among not prior users and a HR of 0.64 [95% CI 0.43–0.96] among prior users aged 30–39. In the group of not prior users significantly increased risk of use was observed in postmenopausal women aged 50–59 and 60–69 year old women. Postmenopausal women aged 50–59, women
Figure 1. Kaplan-Meier plot of the first redeemed prescription of hypnotics among Danish women the first year after the diagnosis with breast cancer among not prior and prior users of hypnotics.
aged 60–69, and women aged 70 or older were found at increased risk of use among prior users. Basic school education was associated with lower risk of use of hypnotics among not prior users [HR 0.88 (95% CI 0.81–0.96)], though associated with use among prior users [HR 1.10 (95% CI 1.00–1.21)]. Somatic comorbidity of 1 was associated with the use of hypnotics among both not prior and prior users [HR 1.18 (95% CI 1.05–1.32) and HR 1.17 (95% CI 1.05–1.30), respectively]. Somatic comorbidity of 2 ⫹ was associated with use among prior users, though associated with lower risk among not prior users [HR 0.83 (95% CI 0.70–0.99)]. Use of antidepressants in the year before diagnosis of breast cancer was significantly associated with use of hypnotics in both not prior and prior users [HR1.37 (95% CI 1.23–1.52) and HR1.34 (95% CI 1.23–1.45), respectively]. Diagnosis of breast cancer in the later period (2003–2006) compared to the earlier period (1996– 1998) were associated with lower risk of use of hypnotics among both not prior [HR 0.90 (95% CI 0.83–0.99)] and prior users of hypnotics [HR 0.89 (95% CI 0.80–0.98)]. Discussion We showed that 26% of women with breast cancer were treated with hypnotics during the first year after the diagnosis. Among not prior users, 17% became first time users and 4% developed a chronic use (four or more prescriptions of hypnotics). Among prior users 15% increased their use of hypnotics to chronic use after the diagnosis. Use of hypnotics on a chronic basis is associated with impairment of cognitive and psychomotor function [11] and daytime sedation, and is associated with risk for tolerance, dependency and mortality [22,23]. International guidelines recommend 2–4 weeks of use [12] even though side effects may also occur on a short-term basis [22]. As a result of this limited usability of hypnotics other evidence-based treatments should be considered [24]. In addition to being a symptom of clinical depression, sleep disturbances are often comorbid with other psychiatric disorders [25]. Previous studies have concluded that psychological factors were predictors of sleep disturbances, and that cancer-related factors were secondary in the developing and maintaining of sleep disturbances [6,7,26,27]. Our study detected that prior use of antidepressants, as an indicator of prior history of clinical depression, almost doubled the risk for use of hypnotics compared to women with no prior history of use of antidepressants. In addition, we found that cancer-related
New and chronic use of hypnotics after diagnosis with early breast cancer
709
Table III. Adjusted hazard ratios for use of hypnotics among Danish women the first year after the diagnosis with breast cancer. Total cohort n ⫽ 26 082
Clinical variable Tumor size Per cm Tumor positive axillary lymph nodes 0 1–3 4–9 ⱖ 10 Tumor grade I II III Non-ductal carcinoma Missing Hormone sensitivea Positive Negative Treatment-related factors Type of surgeryb Lumpectomy Mastectomy Adjuvant treatment No adjuvant treatment Chemotherapy Other adjuvant treatment Missing Sociodemographic factors Age and menopausal statusc 30–39 premenopausal 40–49 premenopausal 40–49 postmenopausal 50–59 premenopausal 50–59 postmenopausal 60–69 postmenopausal ⱖ 70 postmenopausal Cohabitation status Living with a partner Living alone Level of education Higher (⬎ 13 years) Vocational (10–12 years) Basic school (7–12 years) Charlson Comorbidity Index 0 1 ⱖ2 Use of antidepressants before BCd No Yes Year of diagnosis 1996–1998 1999–2002 2003–2006
Hazard ratio
95% CI
0.99
0.97–1.01
1 1.12 1.11 1.25
Not prior users of hypnotics n ⫽ 22 537 Hazard ratio
Prior users of hypnotics n ⫽ 3545
95% CI
Hazard ratio
95% CI
1.01
0.98–1.03
1.01
0.98–1.04
– 1.05–1.19 1.02–1.21 1.14–1.38
1 1.18 1.18 1.33
– 1.09–1.28 1.06–1.32 1.17–1.51
1 1.01 1.02 1.15
– 0.92–1.11 0.89–1.16 0.98–1.34
1 0.97 0.99 1.01 0.94
– 0.90–1.03 0.91–1.08 0.93–1.09 0.78–1.14
1 0.97 1.00 0.98 0.90
– 0.89–1.06 0.90–1.11 0.88–1.09 0.70–1.17
1 0.93 0.99 0.91 0.95
– 0.84–1.03 0.87–1.12 0.82–1.02 0.71–1.27
1 1.06
– 0.98–1.15
1 1.15
– 1.04–1.26
1 0.87
– 0.77–0.99
1 1.10
– 1.03–1.16
1 1.10
– 1.01–1.19
1 1.01
– 0.92–1.10
1 1.25 0.98 1.15
– 1.12–1.39 0.89–1.07 1.04–1.28
1 1.48 1.04 1.29
– 1.27–1.72 0.91–1.19 1.09–1.48
1 1.06 0.97 0.99
– 0.89–1.25 0.85–1.11 0.85–1.15
0.62 1 1.07 1.18 1.43 1.57 1.59
0.53–0.73 – 0.83–1.40 1.06–1.32 1.31–1.57 1.44–1.71 1.43–1.77
0.65 1 0.92 1.11 1.30 1.22 0.99
0.54–0.78 – 0.66–1.28 0.98–1.27 1.17–1.45 1.10–1.36 0.86–1.15
0.64 1 1.22 1.12 1.31 1.34 1.44
0.43–0.96 – 0.80–1.87 0.91–1.38 1.11–1.54 1.15–1.58 1.22–1.71
1 1.04
– 0.98–1.09
1 0.94
– 0.87–1.02
1 1.02
– 0.95–1.11
1 0.88 0.87
– 0.83–0.94 0.82–0.93
1 0.92 0.88
– 0.84–1.00 0.81–0.96
1 0.99 1.10
– 0.89–1.10 1.00–1.21
1 1.31 1.20
– 1.21–1.42 1.08–1.32
1 1.18 0.83
– 1.05–1.32 0.70–0.99
1 1.17 1.24
– 1.05–1.30 1.09–1.40
1 1.97
– 1.85–2.10
1 1.37
– 1.23–1.52
1 1.34
– 1.23–1.45
1 0.99 0.89
– 0.92–1.06 0.83–0.95
1 1.00 0.90
– 0.91–1.09 0.82–0.99
1 1.04 0.89
– 0.94–1.15 0.80–0.98
All three models included all shown variables. Tumor size was evaluated as a continuous variable. aIn the years 1996–2006 defined as ⱖ 10% tumor cells positive for estrogen and/or progesterone receptors; bBiopsy was taken in 9 women. These were included in the analysis, but results are not shown; c Results reported for combined variable. Women aged 30–39, 60–69 and ⱖ 70 had same menopausal status; dOne or more redeemed prescriptions of a drug in the ATC group N06A 13 months to 1 month before the diagnosis with early breast cancer.
710
L. T. Andersen et al.
factors, such as having tumor positive axillary lymph nodes and receiving chemotherapy, were significantly associated with use of hypnotics among not prior users. Hormones receptor status affects treatment choices and women with HR negative breast cancer are mostly receiving chemotherapy. In this study, no correlation was seen between HR negative status and chemotherapy, so this is unlikely to have biased our findings. Interestingly, among prior users breast cancer-related factors seemed to have almost no association with use of hypnotics. It has previously been reported that after surgery, being younger may be a risk factor for development of somatic comorbidity with persistent pain and arm lymph-edema [28,29]. Furthermore, a recent review reported that younger women with breast cancer compared to older women with breast cancer, have poorer quality of life [30]. One may argue that these younger women would be at risk for use of hypnotics but our study in fact indicated that being ⱕ 40 years of age were associated with reduced risk for use of hypnotics among not prior and prior users of hypnotics. This is of interest, as our study found that somatic comorbidity increased the risk for using hypnotics and it seems that increasing age, no matter prior history of use of hypnotics, and having comorbidity independently contributed to the use of hypnotics in these women. We showed that the majority of the first redeemed prescriptions took place the first three months after surgery. This supports the findings from previous small studies reporting that 19–60% of women with breast cancer subjectively reported sleep disturbances the first two [6,31] to four months [7] after surgery. Our findings have implications for clinical practice, as we did separate analyses on not prior and prior users of hypnotics, thereby making it easier in the clinical setting to recognize women at risk for developing sleep disturbance. At this moment studies regarding preventive and prophylactic treatments of sleep disturbance in women with breast cancer, are being investigated [32]. We showed a difference in use of hypnotics and factors associated with use of hypnotics between not prior and prior users. We observed differences in associations in both factors related to the disease severity and to treatment, especially chemotherapy. Furthermore, basic education seemed to be associated with more use among prior users whereas it was associated with reduced use among not prior users, indicating that having a breast cancer and associated sleeping problems may be handled differently by socioeconomic position.
Our study has some limitations. Firstly, we did not have a control group and could not address if there was a possible additive hypnotic use in postmenopausal women. Furthermore, we did not compare our group with patients with other serious diseases. However, we showed that around 258/1000 women with breast cancers were prescribed a hypnotic in the ATC groups N05DC and N05CF. In the Danish female population, 58/1000 women and 17/1000 women were prescribed a N05DC and N05CF in 2005, respectively [33]. Thus, our data indicate that there is a marked increased use of hypnotics in women with breast cancer, which is confirmed in our findings of 17% new users. Second, we did not know if patients had used hypnotics during the hospital stay or if they were discharged with hypnotics from the hospital. If anything, this might underestimate the earliest use of hypnotics after breast cancer diagnosis. However, this is unlikely to bias our findings, since Danish women with breast cancer in the period of investigation were hospitalized on average three days [34]. Third, we only investigated drugs in the ATC groups N05CD and N05CF. We were only interested in hypnotics prescribed for sleep and not mainly prescribed for anxiety or epilepsy. We were not interested in the use of over the counter hypnotics which in some countries include the hypnotic melatonin. In Denmark, melatonin is only available on prescription and was first marketed in 2007 and therefore could not be investigated in this study. Fourth, we did not investigate the effect of sentinel node technique on use of hypnotics, however, only a limited number of patients have had sentinel node surgery in our cohort. In conclusion, this study found that one in four women used hypnotics the first year after the diagnosis and 17% became new users of hypnotic. A diagnosis with breast cancer was associated with the initiation of chronic hypnotic use in 4% of not prior users and 15% of prior users of hypnotics. A range of clinical, treatment and sociodemographic factors were associated with the use of hypnotics among not prior users whereas these associations were not as strong among prior users of hypnotics. These findings suggest the need for further focus on women with breast cancer and alternative prophylactic treatment modalities are warranted to prevent the development of new and potentially chronic users of hypnotics. Declaration of interest: The corresponding author affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been
New and chronic use of hypnotics after diagnosis with early breast cancer omitted; and that any discrepancies from the study as planned have been explained. The authors declare that they have no commercial or other associations that might pose a conflict of interest in connection with this article.
[17] [18] [19]
References [1] Bray F, Ren JS, Masuyer A, Ferlay J. Global estimates of cancer prevalence in 27 sites in the adult population in 2008. Int J Cancer 2013;132:1133–45. [2] Autier P, Boniol M, La Vecchia C, Vatten L, Gavin A, Héry C, et al. Disparities in breast cancer mortality trends between 30 European countries: Retrospective trend analysis of WHO mortality database. Br Med J 2010;341:c3620. [3] Rowland JH, Hewitt M, Ganz PA. Cancer survivorship: A new challenge in delivering quality cancer care. J Clin Oncol 2006;24:5101–4. [4] Dalton SO, Bidstrup PE, Johansen C. Rehabilitation of cancer patients: Needed, but how? Acta Oncol 2011;50:163–6. [5] Ewertz M, Jensen AB. Late effects of breast cancer treatment and potentials for rehabilitation. Acta Oncol 2011;50: 187–93. [6] Savard J, Ivers H, Villa J, Caplette-Gingras A, Moris CM. Natural course of insomnia comorbid with cancer: An 18-month longitudinal study. J Clin Oncol 2011;29: 3580–6. [7] Colagiuri B, Christensen S, Jensen AB, Price MA, Butow PN, Zachariae R. Prevalence and predictors of sleep difficulty in a national cohort of women with primary breast cancer three to four months postsurgery. J Pain Symptom Manage 2011;42:710–20. [8] Savard J, Simard S, Blanchet J, Ivers H, Morin CM. Prevalence, clinical characteristics, and risk factors for insomnia in the context of breast cancer. Sleep 2001;24:583–90. [9] Fortner BV, Stepanski EJ, Wang SC, Kasprowicz S, Durrence HH . Sleep and quality of life in breast cancer patients. J Pain Symptom Manage 2002;24:471–80. [10] Redwine L, Hauger RL, Gillin JC, Irwin M. Effects of sleep and sleep deprivation on interleukin-6, growth hormone, cortisol, and melatonin levels in humans. J Clin Endocrinol Metab 2000;85:3597–603. [11] Glass J, Lanctôt KL, Herrmann N, Sproule BA, Busto UE. Sedative hypnotics in older people with insomnia: Metaanalysis of risks and benefits. Br Med J 2005;331:1169. [12] National Institutes of Health State of the Science Conference statement on Manifestations and Management of Chronic Insomnia in Adults, June 13–15, 2005. Sleep 2005;28:1049–57. [13] de Bock GH, Musters RF, Bos HJ, Schröder CP, Mourits MJ, de Jong-van den Berg LT. Psychotropic medication during endocrine treatment for breast cancer. Support Care Cancer 2012;20:1533–40. [14] Moore TA, Berger AM, Dizona P. Sleep aid use during and following breast cancer adjuvant chemotherapy. Psychooncology 2011;20:321–5. [15] Costantini C, Ale-Ali A, Helsten T. Sleep aid prescribing practices during neoadjuvant or adjuvant chemotherapy for breast cancer. J Palliat Med 2011;14:563–6. [16] Møller S, Jensen M-B, Ejlertsen B, Bjerre KD, Larsen M, Hansen HB, et al.The clinical database and the treatment guidelines of the Danish Breast Cancer Cooperative Group
[20] [21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33] [34]
711
(DBCG); its 30-years experience and future promise. Acta Oncol 2008;47:506–24. Gaist D, Sørensen HT, Hallas J. The Danish prescription registries. Dan Med Bull 1997;44:445–8. Pedersen CB. The Danish Civil Registration System. Scand J Public Health 2011;39:22–5. Jensen VM, Rasmussen AW. Danish Education Registers. Scand J Public Health 2011;39:91–4. Lynge E, Sandegaard JL, Rebolj M. The Danish National Patient Register. Scand J Public Health 2011;39:30–3. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 1987;40:373–83. Lader M. Benzodiazepines revisited – will we ever learn? Addiction 2011;106:2086–109. Wei YJ, Simoni-Wastila L, Zuckerman IH, Huang TY, Brandt N, Moyo P, et al. Quality of psychopharmacological medication prescribing and mortality in Medicare beneficiaries in nursing homes. J Am Geriatr Soc 2014; 62:1490–504. Andersen SR, Würtzen H, Steding-Jessen M et al. Effect of mindfulness-based stress reduction on sleep quality: Results of a randomized trial among Danish breast cancer patients. Acta Oncol 2013;52:336–44. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. Washington, DC: American Psychiatric Association; 2000. [cited 2103 Jun 3]. Available from: http://www.psychiatry.org/practice/clinical-practiceguidelines. Bardwell WA, Profant J, Casden DR, Dimsdale JE, Ancoli-Israel S, Natarajan L, et al. The relative importance of specific risk factors for insomnia in women treated for early-stage breast cancer. Psychooncology 2008;17:9–18. Davidson JR, MacLean AW, Brundage MD, Schlze K. Sleep disturbance in cancer patients. Soc Sci Med 2002; 54:1309–21. Gartner R, Jensen MB, Nielsen J, Ewertz M, Kroman N, Kehlet H. Prevalence of and factors associated with persistent pain following breast cancer surgery. JAMA 2009; 302:1985–92. Gartner R, Jensen MB, Kronborg L, Ewertz M, Kehlet H, Kroman N. Self-reported arm-lymphedema and functional impairment after breast cancer treatment – a nationwide study of prevalence and associated factors. Breast 2010;19:506–15. Howard-Anderson J, Ganz PA, Bower JE, Stanton AL. Quality of life, fertility concerns, and behavioral health outcomes in younger breast cancer survivors: A systematic review. J Natl Cancer Inst 2012;104:386–405. Savard J, Villa J, Ivers H, Simard S, Morin CM. Prevalence, natural course, and risk factors of insomnia comorbid with cancer over a 2-month period. J Clin Oncol 2009;27:5233–9. Hansen MV, Madsen MT, Hageman I, Rasmussen LS, Bokmand S, Rosenberg J, et al. The effect of MELatOnin on Depression, anxietY, cognitive function and sleep disturbances in patients with breast cancer. The MELODY trial: Protocol for a randomised, placebo-controlled, doubleblinded trial. 2012;2:e000647. Medstat.dk. [cited 2013 Jun 3]. Available from: http://www. medstat.dk. Mertz BG, Kroman N, Williams H, Kehlet H. Fast-track surgery for breast cancer is possible. Dan Med J 2013;60:A4615.
Copyright of Acta Oncologica is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
