Adherence to osteoporosis regimens among men and analysis of risk factors of poor compliance: a 2-year analytical review

The prevalence of osteoporosis and osteoporotic fractures has increased worldwide during the past few decades[1]. Osteoporosis and related fractures are a common cause of morbidity and mortality in the elderly globally[2‐4]. There are significant differences in skeletal size and structure between men and women that account for differences in fracture incidence, location and outcomes[5]. While women sought medical attention for the prevention of bone loss, men were referred because of the presence of signs and symptoms indicating a more severe disease[6]. In Taiwan, the prevalence of osteoporosis between 1999 and 2001 among subjects aged 50 years or older was 1.63% for men and 11.35% for women[7]. Although the prevalence of osteoporosis is lower in men than in postmenopausal women globally, the mortality and morbidity of osteoporosis among men are higher than among women[5, 8‐10].

A variety of regimens including bisphosphonates, estrogen, selective estrogen receptor modulators, calcitonin, strontium ranelate, and teriparatide are currently available in Taiwan for the prevention and treatment of osteoporosis, and have demonstrated anti-fracture efficacy in postmenopausal women and in men[11, 12]. The efficacy of osteoporosis medications depends on adherence, which consists of both compliance and persistence. Despite the availability of several treatment options for osteoporosis, adherence to anti-osteoporotic therapy in real-world practice is suboptimal and has generally been lower than in clinical trials. In a meta-analysis of 24 studies, 40-70% of osteoporotic patients were reported to be non-adherent[13]. Although male subjects at an advanced age often have osteoporosis and osteoporotic fractures, little is known about their exact adherence rates because researchers rarely include men in osteoporosis adherence studies,[14, 15] or seldom report adherence rates separately or exclusively for men. Non-adherence to osteoporosis regimens is clearly associated with poorer outcomes. Therefore, it is essential to improve drug adherence for both postmenopausal women and osteoporotic men, so as to enhance therapeutic outcomes.

Claims or pharmacy databases are usually used to measure adherence and to determine factors associated with non-adherence to osteoporosis medications. Non-adherence in women has been previously reported to be related to the drugs, the patients, and the doctors[16‐18]. Few studies in the literature have investigated the factors affecting adherence among males with osteoporosis[14, 15]. Therefore, further identification of risk factors for non-adherence among males with osteoporosis may help improve adherence and ultimately reduce fracture risk. In addition, many studies have demonstrated that realizing the anti-fracture efficacy of osteoporosis medications may require several years of therapy[19, 20], but most reports have investigated adherence to anti-osteoporotic therapy within one year only[13, 21, 22].

In this retrospective chart review study of male patients with osteoporosis treated at a single medical center in southern Taiwan, we aimed to investigate 1- and 2-year adherence (both compliance and persistence) to osteoporosis regimens (including bisphosphonates and calcitonin), and also explore the risk factors associated with poor adherence.

Osteoporosis in men has been documented as a serious health issue recently, as osteoporosis can cause higher morbidity and mortality in men than in women, though it affects males less frequently[5, 10]. Indeed, whereas numerous data related to the adherence of female osteoporosis patients are available, data are scarce for men. To our knowledge, the present study is one of few that have investigated the factors affecting medication compliance among males with osteoporosis. The findings were in accord with the results of our total cohort[26], of which more than 90% of subjects were female: adherence among male subjects was suboptimal. In contrast with previous investigations of female osteoporosis, our survey revealed the major risk factors of poor compliance at year 1 were index prescription by orthopedists and not having undergone baseline BMD measurement before pharmacologic treatment for osteoporosis was initiated. However, male subjects with RA were more likely to have good compliance.

Compliance with medications used to treat or prevent osteoporosis ranged from 34% to 94.6% at 1 year[27‐31] and 27% to 43% at 2 years[27, 32]. However, the results were subject to various definitions of compliance, and female gender, bisphosphonate users and different study populations were mainly focused on in each investigation[28, 30]. Our data examined adherence not only among male users of oral (daily and weekly) bisphosphonates, but also among male users of calcitonin. Hansen et al.[14] reported that adherence to alendronate in male veterans during the first 12 and 24 months of therapy was 59% and 54%, respectively. A small series study by Cevikoi A[15] demonstrated that full compliance rates for the 1- and 3-year periods were 49.4% and 17.9%, respectively. The results of our investigation of good compliance at the first year were similar to those of most studies on postmenopausal women[14, 32, 33]. However, at year 2, good compliance in our cohort was 37.5%, which was lower than that in Hansen’s study[14]. A recent study on compliance among male subjects reported that the 1-year good compliance (MPR ≥ 80%) and persistence rates were 91% and 39.7%, respectively[21]. They attributed the relatively high 1-year good compliance rate to (1) including both new and existing patients with at least 3 prescriptions; (2) the specific ethnic group (a Dutch population); and (3) reimbursement of all osteoporosis medications[21]. That was in contrast to our study population, which had fewer than 3 prescriptions and limited reimbursement for fracture subjects.

Previous investigations that were mainly focused on bisphosphonate users demonstrated 1-year PRs ranging from 24% to 74.8%[15, 21, 34‐36]. In our study, the persistence curve revealed an obvious drop (38.3%) in the first year, but a less prominent (19.5%) drop in the second year among those subjects persistent with their medication at the end of the first year. This was similar to the observation in several studies on adherence to anti-osteoporosis agents among postmenopausal women[37‐39]. The previous study had demonstrated that increasing age, pain, no use of diagnostic tests, and male showed a positive effect on the probability of quitting the anti-osteoporotic medication[40]. It also suggests that improving adherence in the first year is essential to achieving a better therapeutic response for either gender. Therefore, the factors related to inadequate compliance for each year were investigated in the current study.

Possible factors affecting medication adherence have been reviewed in the literature and include side effects, smoking behavior, surveying for bone density during therapy, educational status, social support, marital status, and income[14, 15, 41]. However, these factors were not analyzed annually, as done in our investigation. Data on several characteristics of male subjects, e.g., jobs, lifestyle, and educational levels, were not collected, which potentially may be related to compliance with therapy in the current study. Hence, our finding suggests that certain unidentified factors pertaining to male subjects may be related to poor compliance, as demonstrated by previous studies.

In our study, patients of orthopedists were more likely to discontinue osteoporosis medications at year 1 than those seeing rheumatologists. This is in agreement with a previous study showing that patients of orthopedists were more likely to discontinue their medications than the patients of rheumatologists and general practitioners, probably due to a lack of adequate doctor motivation[42]. Physician specialty was significantly associated with adherence to treatment guidelines[43]. Ideguchi et al. found that patients of gynecologists or rheumatologists were more likely to continue bisphosphonates than patients of orthopedists or other specialists[44]. BMD measurement was highest among patients treated with glucocorticoids prescribed by rheumatologists[45]; therefore, rheumatologists often had more opportunities to prescribe anti-osteoporotic medications to prevent or treat GIOP than other specialties. We speculated from our study results that having follow-up for each long-term prescription by a physician specializing in osteoporosis (especially an internist, for example, a rheumatologist) with feedback to the patients may improve the patient-physician relationship and adherence.

A lack of DXA at baseline (before therapy) was another factor of 1-year poor compliance in our study. Similar observations were also reported in Pickney’s[46] and Solomon’s studies[29]. In contrast, Hansen et al. showed that non-adherence to oral anti-osteoporosis medications among men was associated with a lack of measurements of bone mass during alendronate therapy, but not before therapy[14]. It has been demonstrated that a correct understanding of BMD results would lead to higher treatment rates and better adherence to treatment[46]. Adherers who have undergone bone mass measurements at baseline might be more aware of their disease course and of the risks of subsequent or recurrent fractures than non-adherers.

Richards et al. observed that non-adherence to bisphosphonates was common in a cohort of RA patients who were veterans in the United States[47]. In contrast, RA patients in our study were more likely to continue osteoporosis medications. We would expect good compliance with osteoporosis medications in patients with RA because of the awareness of physicians and patients about the negative effects of glucocorticoids on bone. Most patients with RA may need to receive steroid treatment in clinical practice, which puts them at high risk for GIOP. Rossini et al. demonstrated positive associations between glucocorticoid and anti-inflammatory treatment and compliance with osteoporosis medications[42]. Although steroid users did not have a higher compliance rate in the current study, other determining factors associated with adherence among osteoporotic RA male patients could be further explored in the future.

There are several limitations in our study. (1) It is unknown if patients actually took the dispensed drug. We assumed that patients who obtain prescription refills do take their medications based on chart review. In addition, in Taiwan, patients can obtain 2 to 3 prescriptions each visit for a stable disease. We were unable to ascertain whether patients obtained refills at outside pharmacies. As a result, adherence may be overestimated. (2) We determined adherence based on the number of refills. Therefore, whether the drug was delivered correctly at the correct time of day or in appropriate doses is unknown. (3) Due to the retrospective study design, some information obtained from chart reviews may be not complete and the data collection may be subjective. (4) We did not collect data on the side effects of anti-osteoporotic regimens from the chart reviews, which may actually have influenced adherence. (5) Our sample size was relatively small. Therefore, we may not have determined all factors associated with non-adherence among male patients. (6) Finally, patient characteristics linked to adherence from databases at a single medical center probably were not universally applicable to men treated by community-based practitioners.

However, our study has some advantages. (1) We reviewed medical charts to determine patient variables associated with adherence, information that is often not acquired in claims databases. (2) Most studies reported on the adherence mainly of postmenopausal women; we targeted men who are not normally enrolled or included in osteoporosis adherence studies exclusively. (3) We analyzed adherence over a period of 2 years and identified patient-specific factors of poor adherence on a year-by-year basis. This information can serve as a reminder to prescribers year by year to improve adherence.

Our study indicates that men with osteoporosis are not optimally treated in daily clinical practice in southern Taiwan. Starting prescriptions by orthopedists significantly contributed to non-compliance, and men with RA and baseline bone mass measurements were less likely to have poor compliance. Identifying risk factors associated with non-adherence may help healthcare professionals improve adherence.