This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Clowes, J. A. Articles by Eastell, R. Search for Related Content PubMed PubMed Citation Articles by Clowes, J. A. Articles by Eastell, R.

The Impact of Monitoring on Adherence and Persistence with Antiresorptive Treatment for Postmenopausal Osteoporosis: A Randomized Controlled Trial

Bone Metabolism Group, University of Sheffield, Sheffield, United Kingdom S57 AU

Address all correspondence and requests for reprints to: Dr. Jackie Clowes, Division of Clinical Sciences (North), University of Sheffield, Northern General Hospital, Herries Road, Sheffield, United Kingdom, S5 7 AU. E-mail: j.a.clowes{at}sheffield.ac.uk.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Long-term adherence and persistence with any therapy are very poor (50%). Adherence to therapy is defined as the percentage of prescribed medication taken, and persistence is defined as continuing to take prescribed medication. We examined whether monitoring by nursing staff could enhance adherence and persistence with antiresorptive therapy and whether presenting information on response to therapy provided additional benefit. In addition we evaluated the impact of monitoring on treatment efficacy.

Seventy-five postmenopausal women with osteopenia were randomized to 1) no monitoring, 2) nurse-monitoring, or 3) marker-monitoring. All subjects were prescribed raloxifene. At 12, 24, and 36 wk, the nursing staff reviewed subjects in the monitored (nurse-monitoring or marker-monitoring) groups using a predefined protocol. The marker-monitored group were also presented a graph of response to therapy using percentage change in urinary N-telopeptide of type I collagen (uNTX), a bone resorption marker, at each visit. Biological response to therapy at 1 yr was determined using the percent change in bone mineral density (BMD) and uNTX. Treatment adherence and persistence were assessed using electronic monitoring devices.

Survival analysis showed that the monitored group increased cumulative adherence to therapy by 57% compared with no monitoring (P = 0.04). There was a trend for the monitored group to persist with therapy for 25% longer compared with no monitoring (P = 0.07). Marker measurements did not improve adherence or persistence to therapy compared with nurse-monitoring alone. Adherence at 1 yr was correlated with percent change in hip (BMD) (r = 0.28; P = 0.01) and percent change in uNTX (r = -0.36; P = 0.002). In conclusion, monitoring of patients increased adherence to therapy by 57% at 1 yr. Increased adherence to therapy increased the effectiveness of raloxifene therapy determined using surrogate end points.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
LONG-TERM ADHERENCE and persistence with any therapy are very poor and are not specific to the disease, disease severity, or treatment (1). Adherence is the extent a patient’s behavior coincides with medical advice and is defined as the percentage of prescribed medication taken. Persistence is defined as continuing to take prescribed medication. Therefore, stopping treatment permanently (nonpersistence) represents the most extreme form of nonadherence.

Osteoporosis is an asymptomatic disease that increases fracture risk and results in considerable morbidity, mortality, and health care cost (2). Antiresorptive agents for the prevention and treatment of postmenopausal osteoporosis decrease bone turnover, increase bone density, and decrease the risk of vertebral fracture by up to 50% (2). However, long-term adherence and persistence with antiresorptive therapy are poor (3, 4, 5, 6), limiting the effectiveness of treatment. Furthermore, low persistence with therapy cannot be explained solely by the influence of adverse events (5, 7, 8).

To improve the effectiveness of antiresorptive therapy and decrease the burden of osteoporotic fractures, simple techniques need to be developed to improve long-term adherence. Many interventions to improve long-term adherence in chronic disease are complex and multifaceted (e.g. combinations of counseling, reminders, and supervised self monitoring), with studies frequently subject to design limitations (1, 9, 10).

Monitoring is a simple clinical tool that provides patients with attention from a health professional and represents a resource that is readily available in the clinical setting. The one to one attention provided by health care professionals is usually combined with a functional or biological test, which is used to measure disease activity or treatment response. This assessment may enable identification of nonresponders with potential contributing factors (e.g. poor compliance or occult comorbidities) and the subsequent optimization of a treatment regimen. It has been suggested that monitoring may improve long-term adherence with therapy (10); however, the majority of studies using complex interventions failed to separate the impact of attention from health professionals from the intervention or test (1, 9).

A bone mineral density (BMD) measurement provides an assessment of fracture risk (2), appears to increase the initial uptake of antiresorptive therapy by patients (7, 11, 12), and improves lifestyle behavior (8). A baseline BMD did not, however, result in increased persistence with antiresorptive therapy (7, 13). Previous studies have suggested that a change in BMD or bone turnover can be used as a surrogate marker of response to therapy (14, 15, 16). Bone turnover markers demonstrate an early response to therapy, with a greater signal to noise ratio than BMD (17, 18, 19), and represent a surrogate marker of fracture risk reduction (20, 21). The role of bone markers in clinical practice has recently been reviewed (22, 23). It remains unclear, however, whether providing subjects with information on their response to therapy using bone turnover markers will increase patient adherence or persistence with therapy.

The aims of the study were to determine 1) whether monitoring (attention) by nursing staff could enhance adherence and persistence with an antiresorptive agent, and 2) whether demonstrating a treatment response to subjects (intervention) could provide additional benefit compared with usual care. Adherence is a process, and therefore the benefits of interventions designed to modify behavior should be judged by objective improvements in clinical outcome (1). Thus, we examined whether adherence to therapy could influence biological outcome determined using percent change in bone turnover and BMD as surrogate end points of the response to antiresorptive therapy.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Subjects

Seventy-five healthy postmenopausal women (aged 50–80 yr) were recruited from general practice surgeries or new patient referrals for BMD measurements. Subjects were eligible if they had osteopenia at either the spine or hip (T score, less than -1 SD and more than -2.5 SD) measured by dual energy x-ray absorptiometry and were either 1) more then 5 yr from their last menstrual period, or 2) after hysterectomy, under age 55 yr, and had an elevated FSH (>30 IU/liter).

Subjects were excluded if they had taken hormone replacement therapy or other antiresorptive treatments within the past 6 months, had a metabolic bone disease or other medical condition or treatment likely to affect bone metabolism, or had a history of venous thromboembolic disease, hepatic or renal impairment, unexplained uterine bleeding, or a previous history of malignancy. Subjects with significant degenerative disease or vertebral fractures present on the lumbar spine dual energy x-ray absorptiometry scan that prevented adequate analysis of the region L1–L4 were excluded. The study was conducted at the Osteoporosis Center, Northern General Hospital. Recruitment commenced in May 1999 and was completed in December 2000 with 1-yr follow up. The North Sheffield Research ethics committee approved the study, and written informed consent was obtained from each woman.

Study protocol

At screening, subjects completed a clinical assessment, osteoporosis risk factor questionnaire, screening investigations to exclude treatable underlying causes of osteoporosis, and duplicate BMD measurements of the lumbar spine and total hip (QDR 1000 W, Hologic, San Francisco, CA). Four fasting second morning void urine samples were collected on consecutive days between 0900–1100 h for measurement of bone turnover markers. The study was a randomized controlled open label study. Pharmacy used a random number program and block randomization to generate study codes independent of the study investigators and stored the codes in sealed envelopes. The study physician discussed the study, drugs, and results of baseline measurements with subjects before randomization.

At baseline, 75 postmenopausal subjects were randomized into one of three study arms involving 1) no monitoring (usual care), 2) nurse-monitoring (attention), or 3) marker-monitoring (attention plus information on response). At 0 and 24 wk, all subjects were prescribed a 26-wk supply of raloxifene (60 mg/d; Evista, Eli Lilly & Co., Indianapolis, IN) and 500 mg/d elemental calcium as calcium carbonate.

Follow-up visits

Subjects in the no monitoring group attended at 24 wk to collect a repeat prescription from clerical staff. No medical contact occurred during the visit.

Subjects in the nurse-monitoring and marker-monitoring groups (monitored group) attended for visits at 12, 24, and 36 wk. During each visit the nursing staff followed a predefined interview consisting of six open questions related to well-being, problems with medication, and adverse events. There was no assessment of compliance or any additional information provided on drugs, lifestyle, or osteoporosis. If a subject requested additional clarification (or medical advice) relating to these questions or raised any new questions with the nursing staff, then an appropriate response was provided.

Subjects in the marker-monitored group were also asked to collect urine samples on 4 d before each follow-up visit for measurement of urinary N-telopeptide of type I collagen (uNTX). The nursing staff followed an identical predefined interview and presented a graph displaying the response to therapy with a standard interpretation. All subjects were asked to return tablets and containers at each visit.

Biochemical response

uNTX (Ortho Clinical, Vitros ECi, Rochester, NY), a marker of bone resorption, was measured using an autoanalyzer. All urinary measurements were expressed as a ratio to creatinine excretion (dry slide method; Ortho-Clinical Diagnostics, Rochester, NY). A change that exceeds the least significant change (LSC) in individuals can be regarded as a statistically significant response (18, 24). For multiple measurements, a statistically significant change is calculated as the LSC divided by the square root of the number of measurements performed. Therefore, with four measurements, the LSC is calculated using the formula; LSC (%) = 1.39 x short-term CV (%) (P = 0.05). A good response to therapy was defined as a percentage decrease in uNTX of greater than 20%.

Outcome criteria

Data on adherence and persistence to raloxifene therapy were obtained using electronic monitoring devices (Aardex, Zug, Switzerland) that record the date and frequency a prescription bottle is opened. An electronic event is regarded as equivalent to tablet ingestion. Subjects were not informed of the electronic monitoring device. Nursing and medical staff were blind to the adherence data throughout the study. The study was conducted in accordance with ethical recommendations for monitoring adherence (25).

The primary outcome measure was cumulative adherence to raloxifene therapy, assessed using electronic monitoring devices. Cumulative adherence was the number of tablets actually taken since randomization divided by the number of tablets prescribed since randomization, expressed as a percentage. At randomization each subject had 100% cumulative adherence, and as medication was missed the cumulative adherence decreased. If cumulative adherence was above a threshold of 75% at 1 yr, the subject was defined as adherent. Conversely, subjects were nonadherent if cumulative adherence remained persistently below the 75% threshold before the 1-yr assessment. The date a subject became nonadherent was the last time point since randomization that cumulative adherence remained above 75%; any subsequent improvement in compliance was insufficient to raise the cumulative adherence above the 75% threshold before the 1-yr assessment.

The secondary outcome measure was persistence to therapy. Persistence was defined as continuing to take tablets for more than 7 of any 14 d immediately before the 1-yr visit. Conversely, a subject was defined as nonpersistent if less then 50% of tablets were taken in the 2 wk before the 1-yr visit, with the discontinuation date identified as the last occasion when more than 50% of the prescribed doses were taken in any 2-wk period.

Biological response to therapy

The biological response (percentage change) at 1 yr for duplicate BMD measurements and four uNTX measurements was calculated as the difference between the baseline and final measurements, expressed as a percentage of the mean of all measurements. All biochemical measurements and scan analyses were performed blind to adherence data and randomization.

Statistical analyses

There were no data available on adherence to osteoporosis therapy using electronic monitoring devices before the onset of this study. We therefore performed a retrospective power calculation for a sample size of 75 subjects using a difference in the proportion adhering to therapy (>75%) at 1 yr in the monitored compared with the nonmonitored group of 0.236. This demonstrated that we had 95% power to detect a difference in adherence to therapy between monitored (nurse-monitoring and marker-monitoring) and no monitoring at a significance level of 0.05.

Continuous variables were expressed as the mean ± SE and were compared between groups using one-way ANOVA. Categorical variables were expressed as absolute (number) and relative frequencies (percentage), with proportions compared between groups using a ² test.

Kaplan-Meier survival curves, which display the time to event, are presented for cumulative adherence, and persistence with significance calculated using the Wilcoxon (Gehan) test. For each survival analysis, the proportion of subjects adhering or persisting with therapy in the different groups was compared with the no monitoring group, and this ratio is expressed as a percentage. For cumulative adherence, an event occurred if a subject’s cumulative adherence was persistently less than the 75% threshold before the 1-yr assessment. The time to event was the number of days between randomization and the date cumulative adherence became persistently less than the 75% threshold. Conversely, adherent subjects had a cumulative adherence greater than 75% at 1 yr and were recorded as censored events.

For persistence, an event occurred if a subject had taken less than 50% of the prescribed doses in the 2 wk before the 1-yr assessment. The number of days between randomization and the date a subject last took more than 50% of prescribed doses in any 2-wk period was used to determine the time to event. Conversely, persistent subjects who took more than 50% of the prescribed doses in the 2 wk before the 1-yr assessment were recorded as censored events.

The predictor variables for cumulative adherence to therapy (>75%) were determined using univariate and multivariate backward stepwise (likelihood ratio) Cox regression analysis. The association between adherence (percentage) at 1 yr and surrogate end points for biological response (percent changes in BMD and bone turnover markers) were determined using Spearman rank correlation (two-tailed). Statistical analyses were performed using Statgraphics Plus version 4 (Manugistics, Inc., Rockville, MD) and SPSS 10.0 (SPSS, Inc., Chicago, IL).

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The baseline characteristics are summarized in Table 1. The mean age of the subjects was 62.4 ± 0.8 yr, with a mean age at the last menstrual period of 45.6 ± 0.8 yr. At baseline a greater number of subjects in the no monitoring group had a family history of hip fracture (P < 0.05; Table 1).

View larger version (21K): [in this window] [in a new window]   FIG. 1. Scatter plot of raloxifene manual tablet count vs. the electronic monitoring device. The correlation between the tablet count and the electronic monitoring device was r = 0.85 (P < 0.01). The diagonal line indicates that the techniques are equivalent. Any point lying above the diagonal line indicates that the tablet count overestimates intake and suggests pill dumping.

View larger version (15K): [in this window] [in a new window]   FIG. 2. The frequency that subjects in the study population (n = 75) became nonadherent (cumulative adherence <75% of tablets) or nonpersistent with therapy each month since starting therapy.

View larger version (11K): [in this window] [in a new window]   FIG. 3. The Kaplan-Meier survival curves for cumulative adherence to therapy (>75%) are shown for the monitored group (nurse-monitoring and marker-monitoring) compared to the no monitoring group (left), which compares the impact of attention with that of usual care. Monitoring increased cumulative adherence to therapy (>75%) by 57% compared with no monitoring (P = 0.04). The Kaplan-Meier survival curve for the nurse-monitoring, marker-monitoring, and no monitoring groups (right) compare the separate impacts of attention, information on response to therapy, and usual care. There was a trend for greater cumulative adherence to therapy in the nurse-monitoring and marker-monitoring groups (P= 0.05 and P= 0.15) compared to usual cure.

Nonadherence behavior may have preceded the delivery of a negative message. Thus, in the nonresponder group six subjects were nonadherent at 1 yr, and four of these subjects were identified as nonadherent by 12 wk. In general, subjects who were adherent at 12 wk tended to be adherent at 1 yr. Thus, in the nonresponder group, 81% of subjects who were adherent at 12 wk were adherent at 1 yr; in contrast, only 39% of subjects who were not adherent at 12 wk were adherent at 1 yr. This suggests that, at least in part, factors other then a negative message at 12 wk contribute to the small decrease in adherence. In addition it is interesting to note that all of the subjects defined as nonresponders had low bone turnover at baseline, with a uNTX result in the lower half of the premenopausal reference range.

The Kaplan-Meier survival curves for persistence with therapy are shown for the monitored vs. no monitoring groups (Fig. 4, left) and for the nurse-monitoring, marker-monitoring, and no monitoring groups (Fig. 4, right). The proportion persisting with therapy at 1 yr in the monitored group was 0.84 (95% CI, 0.74, 0.94) compared with 0.67 (95% CI, 0.54, 0.89) in the nonmonitored group, giving a difference in the proportion persisting with therapy at 1 yr of 0.163 (95% CI, -0.035, 0.361). There was a trend for monitored subjects to persist with therapy for 25% longer than no monitoring patients (P = 0.07). The proportion persisting with therapy at 1 yr in the nurse-monitored group was 0.88 (95% CI, 0.75, 1.01), and that in the marker-monitored group was 0.79 (95% CI, 0.63, 0.95). There was a trend for both nurse-monitoring (P = 0.06) and marker-monitoring (P = 0.26) subjects to persist with therapy for longer than no monitoring patients.

View larger version (11K): [in this window] [in a new window]   FIG. 4. Left graph, Kaplan-Meier survival curve (days) of persistence with raloxifene therapy at 12 months in the monitored group compared to the no monitoring group. Monitoring resulted in a trend for increased persistence by 25% compared to no monitoring [Wilcoxon (Gehan) test; P = 0.07]. Right graph, Kaplan-Meier survival curve (days) of persistence with raloxifene therapy at 12 months in the no monitoring compared with nurse-monitoring or marker-monitoring groups.

The biological response to therapy was determined at 12 months (n = 74) using the percent change in lumbar spine BMD, total hip BMD, and uNTX, a marker of bone resorption. The percent adherence to therapy at 1 yr was correlated to the percent change in hip BMD (r = 0.28; 95% CI, 0.051, 0.480; P = 0.01) and negatively correlated to the percent change in uNTX (r = -0.36; 95% CI, -0.549, -0.141; P = 0.002), but not to the percent change in lumbar spine BMD (P > 0.05).

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
In this study monitoring or attention by a health care professional increased adherence by 57% compared with no monitoring. This suggests that attention provided by a health professional in a one to one interaction with a patient improves adherence and supports a role for monitoring subjects in clinical practice. Marker measurements did not result in an additional improvement in adherence or persistence to therapy compared with nurse-monitoring alone. This may relate, in part, to the limitations of the small sample size. It is also possible that the inconvenience of collecting several urine samples in the marker-monitoring group may have had a negative impact on adherence; however, this was not supported by the results on participation or persistence with therapy.

In a post hoc analysis, subjects in the marker-monitoring group who were informed they had a good response to raloxifene were 92% more likely to adhere to therapy compared with usual care. In addition, subjects in the marker-monitored group who were given a positive message were 18% more likely to adhere then in the nurse-monitoring group. Importantly a negative message only had a minor impact on adherence to therapy. This was, however, a post hoc analysis using small study numbers and requires verification in a larger study.

Although the study was not adequately powered to examine persistence to therapy, we found a 25% increase in persistence in monitored subjects compared with no monitoring, which was borderline significant. Subjects participating in randomized controlled trials are likely to be more motivated, persisting with therapy for longer than subjects in the community (4, 26). Thus, in a randomized controlled trial, 90% of subjects persisted with hormone replacement therapy at 1 yr (27) compared with 46% of subjects in the general population (4). We found 83% of subjects persisted with raloxifene therapy at 1 yr. This suggests that implementing a monitoring protocol in the community, where persistence is low, may have even greater impact. In addition, 68% of subjects defined as nonadherent were categorized before the first monitoring visit at 12 wk. Therefore, selection of an earlier time point for monitoring may result in a greater increase in adherence with therapy.

We have also demonstrated an association between adherence to therapy at 1 yr and percent change in hip BMD and uNTX, a marker of bone resorption. It is important to demonstrate a biological response to increased adherence (28), because there are potential adverse consequences of interventions that modify behavior, including loss of privacy, loss of autonomy, increased adverse events, and potentially wasted resources (1, 26, 28). This study suggests that implementing monitoring as a tool to improve long-term adherence may result in an improved outcome for patients with postmenopausal osteoporosis.

We examined the factors that predicted adherence to therapy in a multivariate Cox proportional hazard analysis. Only monitoring was borderline significant. The baseline lumbar spine and hip BMD values did not predict adherence or persistence to therapy. In this study all subjects had a BMD scan before randomization; therefore, we were unable to confirm previous studies that had demonstrated an increased uptake of therapy after a BMD measurement (7, 11, 12). Raloxifene reduced the risk of breast cancer, but has not been shown to reduce the risk of hip fracture (29, 30). In this study a family history of hip fracture or breast cancer did not influence a patient’s decision to adhere or persist with therapy.

In clinical practice, 50% of subjects do not adhere to treatment protocols (1). Nonadherence with antiresorptive therapy is high, with only 46% of women continuing with hormone replacement therapy at 12 months, 35% of women continuing with alendronate treatment at 6 months, and 44% of women continuing with raloxifene at 24 months (3, 4, 6). An increase in adverse events is frequently stated as a reason for nonadherence to therapy. However, an increase in adverse events is reported in less then 50% of subjects who are nonadherent with therapy (5, 7, 8). In this study, 32 (43%) subjects reported adverse events, but only nine (12%) of these subjects subsequently stopped treatment. Other potential reasons for low adherence with antiresorptive therapy may include the failure of patients to obtain symptomatic benefit from treatment or patients perceiving a fracture as treatment failure.

Subjects participating in research may change their behavior (Hawthorne effect) (31) or if aware adherence is being monitored may adopt different patterns of behavior, including dumping of medication or increased adherence before a clinical assessment that has been described as a "white coat effect" (25, 32, 33). Therefore, during this study subjects were informed that tablets would be monitored; however, the precise mechanism of action of the electronic monitoring device was not disclosed. Electronic monitoring devices represent the gold standard for adherence monitoring (34, 35); they are able to determine time to event, which is important for survival analysis and, in contrast to tablet counts, have a high sensitivity to detect nonadherence (36, 37). In this study tablet counts consistently overestimated adherence, which is consistent with pill dumping.

We used a threshold for adequate adherence of 75%, which is consistent with the 75–80% adherence thresholds used in the majority of previous studies. There is, however, no consensus on how to analyze the more complex patterns of adherence observed using electronic monitoring devices (38). Drug holidays represent a pattern of nonadherence that may result in a period of nontherapeutic coverage and therefore reduced drug efficacy (28, 39). In this study adherence only accounted for 9% of the observed change in BMD and 13% of the observed change in bone markers. This may be explained at least in part by the relative importance of drug holidays on drug efficacy. The impact of drug holidays is related to the offset of biological response after drug withdrawal and the dose-response relationship and will vary for different drugs.

In conclusion the monitoring of subjects with osteopenia increased adherence to therapy by 57% compared with usual care. The monitoring assessment consisted of six open questions asked by nursing staff and could be applied to any clinical setting where long-term adherence is a problem. Marker measurements did not improve adherence or persistence to therapy compared with nurse-monitoring alone. However, there was a trend for increased adherence compared with usual care in subjects with a positive response to therapy using bone markers. Monitoring resulted in a borderline significant increase in persistence with therapy (25%) compared with usual care. Finally, increased adherence to therapy increased the effectiveness of raloxifene determined using surrogate markers of response.

	Acknowledgments

 
We thank the volunteers for generously participating in this study; Sister Debbie Swindell, Sheila Duffy, and Judy Finigan for their excellent nursing support; the radiographers and technicians in the Osteoporosis Center; and Jean Russell for statistical advice. J.A.C., R.E., and N.F.A.P. were involved in the study design, analysis, and interpretation of study data. The clinical study was conducted by J.A.C. and N.F.A.P. The manuscript was written by J.A.C. and reviewed by R.E. and N.F.A.P.

	Footnotes

 
This work was supported by Eli Lilly Pharmaceuticals UK (unrestricted grant toward the clinical study) and a fellowship from the National Health Service Executive, UK (to J.A.C.). The study sponsor was not involved in any aspect of the study design, conduct of the study, analysis or report writing, and there was no right of approval to publication.

This study was presented at the American Society for Bone and Mineral Research, Phoenix, AZ, September 2002.

Abbreviations: BMD, Bone mineral density; CI, confidence interval; LSC, least significant change; uNTX, urinary N-telopeptide of type I collagen.

Received March 21, 2003.

Accepted December 1, 2003.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Haynes RB, Montague P, Oliver T, McKibbon KA, Brouwers MC, Kanani R 2000 Interventions for helping patients to follow prescriptions for medications. Cochrane Database Syst Rev CD000011
2. London D, Barlow D, Cooper C, Kanis J, Whitehead M 1999 Osteoporosis: clinical guidelines for prevention and treatment. London: Royal College of Physicians; 1–86
3. Ettinger B, Pressman A, Schein J, Chan C, Silver P, Connolly N 1998 Alendronate use among 812 women: prevalence of gastrointestinal complaints, noncompliance with patient instructions, and discontinuation. J Managed Care Pharm 18:1051–1056
4. Faulkner DL, Young C, Hutchins D, McCollam JS 1998 Patient noncompliance with hormone replacement therapy: a nationwide estimate using a large prescription claims database. Menopause 5:226–229[Medline]
5. Davies GC, Huster WJ, Lu Y, Plouffe Jr L, Lakshmanan M 1999 Adverse events reported by postmenopausal women in controlled trials with raloxifene. Obstet Gynecol 93:558–565[Abstract/Free Full Text]
6. Kayser J, Ettinger B, Pressman A 2001 Postmenopausal hormonal support: discontinuation of raloxifene versus estrogen. Menopause 8:328–332[CrossRef][Medline]
7. Rubin SM, Cummings SR 1992 Results of bone densitometry affect women’s decisions about taking measures to prevent fractures. Ann Intern Med 116:990–995[Medline]
8. Jamal SA, Ridout R, Chase C, Fielding L, Rubin LA, Hawker GA 1999 Bone mineral density testing and osteoporosis education improve lifestyle behaviors in premenopausal women: a prospective study. J Bone Miner Res 14:2143–2149[CrossRef][Medline]
9. McDonald HP, Garg AX, Haynes RB 2002 Interventions to enhance patient adherence to medication prescriptions: scientific review. JAMA 288:2868–2879[Abstract/Free Full Text]
10. Sackett DL, Snow JC 1979 The magnitude of compliance and noncompliance. In: Sackett DL, Snow JC, eds. Compliance in health care. Baltimore: Johns Hopkins University Press; 11–22
11. Torgerson DJ, Thomas RE, Campbell MK, Reid DM 1997 Randomized trial of osteoporosis screening. Use of hormone replacement therapy and quality-of-life results. Arch Intern Med 157:2121–2125[Abstract]
12. Pressman A, Forsyth B, Ettinger B, Tosteson AN 2001 Initiation of osteoporosis treatment after bone mineral density testing. Osteoporos Int 12:337–342[CrossRef][Medline]
13. Marci CD, Viechnicki MB, Greenspan SL 2000 Bone mineral densitometry substantially influences health-related behaviors of postmenopausal women. Calcif Tissue Int 66:113–118[CrossRef][Medline]
14. Wasnich RD, Miller PD 2000 Antifracture efficacy of antiresorptive agents are related to changes in bone density. J Clin Endocrinol Metab 85:231–236[Abstract/Free Full Text]
15. Cummings SR, Karpf DB, Harris F, Genant HK, Ensrud K, LaCroix AZ, Black DM 2002 Improvement in spine bone density and reduction in risk of vertebral fractures during treatment with antiresorptive drugs. Am J Med 112:281–289[CrossRef][Medline]
16. Riggs BL, Melton III LJ 2002 Bone turnover matters: the raloxifene treatment paradox of dramatic decreases in vertebral fractures without commensurate increases in bone density. J Bone Miner Res 17:11–14[CrossRef][Medline]
17. Delmas PD, Bjarnason NH, Mitlak BH, Ravoux AC, Shah AS, Huster WJ, Draper M, Christiansen C 1997 Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. N Engl J Med 337:1641–1647[Abstract/Free Full Text]
18. Braga de Castro MA, Hannon R, Eastell R 1999 Monitoring alendronate therapy for osteoporosis. J Bone Miner Res 14:602–608[CrossRef][Medline]
19. Gluer CC 1999 Monitoring skeletal changes by radiological techniques. J Bone Miner Res 14:1952–1962[CrossRef][Medline]
20. Bjarnason NH, Sarkar S, Duong T, Mitlak B, Delmas PD, Christiansen C 2001 Six and twelve month changes in bone turnover are related to reduction in vertebral fracture risk during 3 years of raloxifene treatment in postmenopausal osteoporosis. Osteoporos Int 12:922–930[CrossRef][Medline]
21. Eastell R, Barton I, Hannon RA, Chines A, Garnero P, Delmas P. 2003 Relationship of early changes in bone resorption to the reduction in fracture risk with risedronate. J Bone Miner Res 18:1051–1056[CrossRef][Medline]
22. Delmas PD, Eastell R, Garnero P, Seibel MJ, Stepan J 2000 The use of biochemical markers of bone turnover in osteoporosis. Committee of Scientific Advisors of the International Osteoporosis Foundation. Osteoporos Int 11(Suppl 6):S2–S17
23. Delmas PD 2000 Markers of bone turnover for monitoring treatment of osteoporosis with antiresorptive drugs. Osteoporos Int 11(Suppl 6):S66–S76
24. Hannon R, Blumsohn A, Naylor K, Eastell R 1998 Response of biochemical markers of bone turnover to hormone replacement therapy: impact of biological variability. J Bone Miner Res 13:1124–1133[CrossRef][Medline]
25. Levine RJ 1994 Monitoring for adherence: ethical considerations. Am J Respir Crit Care Med 149:287–288[Medline]
26. Cleemput I, Kesteloot K, DeGeest S 2002 A review of the literature on the economics of noncompliance. Room for methodological improvement. Health Policy 59:65–94[CrossRef][Medline]
27. Writing Group for the PEPI 1996 Effects of hormone therapy on bone mineral density: results from the postmenopausal estrogen/progestin interventions (PEPI) trial. JAMA 276:1389–1396[Abstract]
28. Urquhart J, De Klerk E 1998 Contending paradigms for the interpretation of data on patient compliance with therapeutic drug regimens. Stat Med 17:251–267[CrossRef][Medline]
29. Ettinger B, Black DM, Mitlak BH, Knickerbocker RK, Nickelsen T, Genant HK, Christiansen C, Delmas PD, Zanchetta JR, Stakkestad J, Gluer CC, Krueger K, Cohen FJ, Eckert S, Ensrud KE, Avioli LV, Lips P, Cummings SR 1999 Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: results from a 3-year randomized clinical trial. Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators. JAMA 282:637–645[Abstract/Free Full Text]
30. Cummings SR, Eckert S, Krueger KA, Grady D, Powles TJ, Cauley JA, Norton L, Nickelsen T, Bjarnason NH, Morrow M, Lippman ME, Black D, Glusman JE, Costa A, Jordan VC 1999 The effect of raloxifene on risk of breast cancer in postmenopausal women: results from the MORE randomized trial. Multiple Outcomes of Raloxifene Evaluation. JAMA 281:2189–2197[Abstract/Free Full Text]
31. De Amici D, Klersy C, Ramajoli F, Brustia L, Politi P 2000 Impact of the Hawthorne effect in a longitudinal clinical study: the case of anesthesia. Control Clin Trials 21:103–114[CrossRef][Medline]
32. Rand CS, Sevick MA 2000 Ethics in adherence promotion and monitoring. Control Clin Trials 21:241S–247S[CrossRef][Medline]
33. Simmons MS, Nides MA, Rand CS, Wise RA, Tashkin DP 2000 Unpredictability of deception in compliance with physician-prescribed bronchodilator inhaler use in a clinical trial. Chest 118:290–295[Abstract/Free Full Text]
34. Kastrissios H, Blaschke TF 1997 Medication compliance as a feature in drug development. Annu Rev Pharmacol Toxicol 37:451–475[CrossRef][Medline]
35. Urquhart J 1994 Role of patient compliance in clinical pharmacokinetics. A review of recent research. Clin Pharmacokinet 27:202–215[Medline]
36. Pullar T, Kumar S, Tindall H, Feely M 1989 Time to stop counting the tablets? Clin Pharmacol Ther 46:163–168[Medline]
37. Rudd P, Byyny RL, Zachary V, LoVerde ME, Titus C, Mitchell WD, Marshall G 1989 The natural history of medication compliance in a drug trial: limitations of pill counts. Clin Pharmacol Ther 46:169–176[Medline]
38. Vrijens B, Goetghebeur E 1997 Comparing compliance patterns between randomized treatments. Controlled Clinical Trials 18:187–203[CrossRef][Medline]
39. Levy G 1993 A pharmacokinetic perspective on medicament noncompliance. Clin Pharmacol Ther 54:242–244[Medline]

This article has been cited by other articles:

				S. M. Cadarette, J. N. Katz, M. A. Brookhart, T. Sturmer, M. R. Stedman, and D. H. Solomon Relative Effectiveness of Osteoporosis Drugs for Preventing Nonvertebral Fracture Ann Intern Med, May 6, 2008; 148(9): 637 - 646. [Abstract] [Full Text] [PDF]

				L. Elaine, A. Papaioannou, L. Dolovich, J. Adachi, A. M. Sawka, S. Burns, K. Nair, and A. Pathak Patients' adherence to osteoporosis therapy: Exploring the perceptions of postmenopausal women Can Fam Physician, March 1, 2008; 54(3): 394 - 402. [Abstract] [Full Text] [PDF]

				M. E. Kraenzlin Biochemical Markers of Bone Turnover and Osteoporosis Management IBMS BoneKEy, July 1, 2007; 4(7): 191 - 203. [Abstract] [Full Text] [PDF]

				M. A. Brookhart, A. R. Patrick, S. Schneeweiss, J. Avorn, C. Dormuth, W. Shrank, B. L. G. van Wijk, S. M. Cadarette, C. F. Canning, and D. H. Solomon Physician Follow-up and Provider Continuity Are Associated With Long-term Medication Adherence: A Study of the Dynamics of Statin Use Arch Intern Med, April 23, 2007; 167(8): 847 - 852. [Abstract] [Full Text] [PDF]

				P. D. Delmas, B. Vrijens, R. Eastell, C. Roux, H. A. P. Pols, J. D. Ringe, A. Grauer, D. Cahall, N. B. Watts, and on behalf of the Improving Measurements of Persist Effect of Monitoring Bone Turnover Markers on Persistence with Risedronate Treatment of Postmenopausal Osteoporosis J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1296 - 1304. [Abstract] [Full Text] [PDF]

				J. Blouin, A. Dragomir, L.-G. Ste-Marie, J. C. Fernandes, and S. Perreault Discontinuation of Antiresorptive Therapies: A Comparison between 1998-2001 and 2002-2004 among Osteoporotic Women J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 887 - 894. [Abstract] [Full Text] [PDF]

				M. J. Valimaki and R. Tahtela Serum Tartrate-Resistant Acid Phosphatase 5b or Amino-Terminal Propeptide of Type I Procollagen for Monitoring Bisphosphonate Therapy in Postmenopausal Osteoporosis? Clin. Chem., December 1, 2005; 51(12): 2382 - 2385. [Full Text] [PDF]

				D. H. Solomon, J. Avorn, J. N. Katz, J. S. Finkelstein, M. Arnold, J. M. Polinski, and M. A. Brookhart Compliance With Osteoporosis Medications Arch Intern Med, November 14, 2005; 165(20): 2414 - 2419. [Abstract] [Full Text] [PDF]

				J.-Y. Reginster, K. M. Wilson, E. Dumont, B. Bonvoisin, and J. Barrett Monthly Oral Ibandronate Is Well Tolerated and Efficacious in Postmenopausal Women: Results from the Monthly Oral Pilot Study J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5018 - 5024. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Clowes, J. A. Articles by Eastell, R. Search for Related Content PubMed PubMed Citation Articles by Clowes, J. A. Articles by Eastell, R.