Determinants of Oral Medication Compliance in Osteoporosis: The Role of Medication Beliefs (original) (raw)

Chapter 6 Conclusions 129 References 141 reduce fracture risk are indicated in those with osteoporosis by bone density criteria, with a prevalent radiographic vertebral fracture, or a recent clinical fracture. Most prescriptions for fracture prevention therapy are medications in the bisphosphonate family. The two most commonly prescribed medications, alendronate and risedronate, are taken once weekly, and a less commonly used bisphosphonate, ibandronate, is taken once monthly. Hormone replacement therapy (HRT) is now used much less frequently for the primary purpose of fracture prevention, since the Women's Health Initiative study in 2002 documented increased risk of myocardial infarction, stroke, breast cancer, and thromboembolic disease associated with its use.(22) Raloxifene is a daily selective estrogen receptor modulator that is also FDA-approved for fracture prevention, but also is not generally used as a first line agent for fracture prevention because of its lack of efficacy with respect to preventing non-spine fractures.(16, 23) As is true with many other chronic diseases such as hypertension(24, 25) and hyperlipidemia,(26, 27) a substantial proportion of those prescribed medication to prevent osteoporosis stop medication prematurely, take it less frequently than prescribed, or take it inappropriately in some other manner. Assessing compliance and persistence with fracture prevention medication using pharmacy claims databases Recently, Kothawala and colleagues have performed a meta-analysis of observational studies that have assessed medication persistence and compliance to oral medication to prevent osteoporotic fracture.(28) The pooled persistence rate from the six studies that assessed persistence using pharmacy claims through 1 year from the index prescription was 50%. Thirteen studies assessed persistence by self-report, and among these the pooled estimated rate of persistence through 1 year of therapy was 80%. This meta-analysis included, however, studies of compliance with hormone replacement therapy and cyclical etidronate, an older bisphosphonate taken daily only for a two week period every 3 months that is no longer used for fracture prevention therapy. A systematic review limited to 14 studies of compliance and persistence with oral bisphosphonate therapy assessed with large pharmacy claims databases estimated that persistence 1 year after an index prescription for a daily alendronate or risedronate ranged from 26% to 56%, and after an index prescription for a weekly 7 and 54% of those prescribed alendronate, respectively, answered "no" to all items of the Morisky-Green scale, considered to be indicative of good compliance. The mean treatment duration was 324 days for raloxifene, and 291 days for alendronate. Participants prescribed raloxifene estimated that they remembered to take a mean 94% of all prescribed tablets, and those prescribed alendronate estimated that they took 90% of all prescribed tablets. A second set of investigators in Spain identified patients from 126 general practices who were being prescribed raloxifene, and randomized the 126 practices to either hand out a leaflet to study participants with additional information regarding raloxifene or to simply follow their usual practice.(52) Compliance was assessed by self-report with the Morisky-Green scale, with high compliance defined as all 4 items of the scale being answered "no". High self-reported compliance was no different between the two groups, being 47.5% and 52.5%, respectively, in the intervention and no intervention groups after 12 months of followup. Quality of life assessed by self-report with the visual analog scale part of the EQ-5D was negatively associated with self-reported compliance in both groups. Carnvale and colleagues surveyed over 2,000 Italian patients who suffered a hip fracture a mean 543 days before the survey date. Twenty one percent of those treated with a fracture prevention medication at the time of or immediately after the fracture self-reported stopping that medication by the time of the survey date. Performance of bone densitometry, younger age, and female sex were associated with fracture prevention medication persistence.(53) Five studies have used surveys to assess both self-reported compliance to and persistence with osteoporosis medication, and reasons for discontinuation. Tosteson and colleagues conducted telephone surveys with 956 women prescribed medication to prevent osteoporotic fracture a mean 7 months after their first prescription. Twenty six percent, 19%, and 19%, respectively, of those prescribed hormone replacement therapy (HRT), raloxifene, or alendronate had discontinued therapy.(54) Among those discontinuing HRT, raloxifene, or alendronate, respectively, 59%, 42% and 49% reported very bothersome side effects. With all three medications, persistence with medication was more likely among those with an accurate recollection of their bone density test results. Similarly, among 275 consecutive post-menopausal women prescribed risedronate (another oral bisphosphonate) 42 of 48 patients who by selfreport discontinued the medication cited one or more side effects as the reason for 8 discontinuation.(55) Two other studies, however, have suggested that side effects account for a lower percentage of premature discontinuation of osteoporosis medications. Of 310 post-menopausal women prescribed medication to prevent osteoporotic fracture following a bone density test, Pickney and colleagues reported that 150 self-reported discontinuation of the medication, and of those less than half cited side effects as the main reason.(56) Twenty six percent (26%) cited cost as the main reason for discontinuation. Side effects were also cited by a smaller proportion of patients as the main reason for discontinuation of osteoporosis medication prescribed to 9,851 post-menopausal women by several osteoporosis specialty centers across Italy.(57) Self-reported discontinuation rates over a mean follow-up time of 14 months (range 11-18 months) varied from just 6.9% for weekly alendronate to 23% for hormone replacement therapy, to 28.7% for intramuscular clodronate (administered once every 1-2 weeks). Persistence was positively associated with prior vertebral fractures, early menopause, and having had a bone density test showing low bone mass. Self-reported reasons for discontinuation were side effects in 24%, lack of motivation in 21%, safety concerns in 13%, and cost in 10%. Uniquely among these studies, Cline and colleagues assessed the association of self-reported use of hormone replacement therapy or oral bisphosphonate medication within the month prior to the survey date with latent variables postulated to be associated with medication use by the Health Belief Model, specifically perceived susceptibility to osteoporosis, perceived severity (health consequences) of osteoporosis, perceived effectiveness of medications to treat osteoporosis, and perceived barriers to use of medications to treat osteoporosis.(58) Notably, the target condition was framed as "osteoporosis" and not specifically as osteoporotic fractures. This study also assessed type of health insurance coverage and "cues to action" (events or health history that indicate risk of osteoporosis or related fractures) such as having had a diagnosis of osteoporosis or osteopenia, family history of fracture, and a personal history of fracture. Having had a bone density test and having Medicare HMO health care coverage were both strongly associated with self-reported recent use of an anti-resorptive agent. Perceived susceptibility to osteoporosis, perceived effectiveness of drug therapy, and perceived lack of barriers to medication use were all modestly associated with self-reported recent use of anti-resorptive drug therapy. 10 Consequences of non-compliance with fracture prevention medications Several studies now have documented that among those at high risk of fracture, suboptimal compliance with prescribed fracture prevention medication is associated with a higher risk of fracture, (34, 59-65) and higher net health care utilization and costs.(60, 62, 64) The precise nature of the association between noncompliance and fracture risk, however, remains unclear. Siris and colleagues, however, could find no fracture reduction benefit for those with an MPR less than 50%, and then an exponentially greater fracture reduction benefit as MPR increased from 50% to 100%.(63) A major difficulty with these studies based entirely on administrative pharmacy and health care claims, however, is that is not clear that those with low compliance are in fact as comparable risk of fracture as those with higher compliance. The study of Rabenda and colleagues from Belgium is helpful in that pharmacologic fracture prevention therapy in Belgium is restricted to those with osteoporosis by bone density criteria (femoral neck T-score ≤-2.5) and/or vertebral fractures, ensuring that all of those studied are indeed at high risk of fracture. In this study, there was a linear 33% reduction in hip fracture incidence as compliance improved from an MPR of 0% to 100%.(65) Patient-Provider Relationship Quality Medication Concerns Med Use Self-Efficacy Perceived Medication Cost Burden 109 Results: The patient-provider relationship quality was modestly but significantly associated with self-reported non-persistence, such that with all other predictors and covariates at their mean value, a one standard deviation increase in the patientprovider relationship quality would reduce the probability of non-persistence from 34.2% to 28.5%. Half of this effect appears to be mediated by perceived need for fracture prevention medication, with lesser proportions mediated by medication concerns, and medication use self-efficacy. Conclusion: Providers who have a better relationship with their patients at high risk of fracture may be better able to influence those patients' perceived need for...