Effect of a patient-support program on once-daily teriparatide adherence and persistence in the Japan Fracture Observational Study (JFOS)

Osteoporotic fractures are a major public health burden in Japan’s “super-aged” society; the percentage of the Japanese population aged ≥ 65 years reached 25% in 2013 and is anticipated to exceed 30% in 2025 [1]. Osteoporotic fractures mostly (≥ 85%) result from falls [1] and cause significant disability that necessitates care (e.g., medication, nursing, rehabilitation, surgery) at a substantial cost [2, 3]. In Japan, the primary reason elderly people require long-term care insurance is falls and fall-related fractures [4]; the number of people who require long-term care insurance has doubled since the service was introduced in 2000 [5].

Osteoporosis medication can reduce osteoporotic fracture risk [6]. However, low adherence and persistence with these medications are common, reducing their clinical effectiveness and cost-effectiveness [7]. Osteoporosis medication non-adherence (medication possession ratio [MPR] < 80%) and non-persistence (> 30 days’ treatment gap) have been found to increase the risk of osteoporotic fracture by 28 and 32%, respectively [6]. As a consequence, osteoporosis medication non-adherence has been estimated to double the cost per quality-adjusted life-year gained [8].

Teriparatide (recombinant human parathyroid hormone), an anabolic agent that stimulates osteoblastic bone formation, has been shown to improve skeletal health and outcomes when administered continuously for 24 months rather than for shorter periods [9]. Medical and pharmacy claims database studies of once-daily teriparatide use in the USA [10, 11], Japan [12], and Taiwan [13] have also indicated that higher rates of adherence and persistence were associated with reduced fracture risk and, in the Japan study, reduced hospital admissions and inpatient costs [12].

Strategies to help improve medication adherence and persistence include patient-support programs that consist of interventions such as individualized medication counseling, call center support, and/or virtual reminders to take medication. Patient-support programs to help patients administer once-daily teriparatide and provide follow-up support have been of benefit for European patients, with reported persistence rates considered high, ranging from 64.1% (N = 23,069) at 24 months [14] to 81.5% (N = 5413) over 15 months [15] and 85.6% (N = 382) at 18 months [16]. However, the effect of a patient-support program on once-daily teriparatide adherence and persistence for patients with osteoporosis in Japan has not been evaluated.

The Japan Fracture Observational Study (JFOS) was a prospective observational study that assessed the effectiveness of once-daily teriparatide in Japanese patients with osteoporosis treated in a real-world clinical setting. The baseline patient characteristics, preliminary 12-month findings [17], and final 24-month findings for JFOS have been published [18]. The main aim of this secondary analysis of the original study was to assess the effect of patient-support program enrollment on adherence to once-daily teriparatide. In addition, we reported 12-month and 24-month persistence with once-daily teriparatide for patients who did and did not enroll in the program.

Findings from this secondary analysis suggest that patient-support programs have beneficial effects on adherence to once-daily teriparatide. Although the overall percentage of enrolled patients with good to very good (> 75%) adherence in both groups was suboptimal, higher rates of good to very good (> 75%) adherence and a significantly greater likelihood of adherence were reported in enrolled patients compared with those who did not enroll (54.2 vs. 48.3%; OR = 1.44). In addition, it was observed that persistence at 12 months (77.2%) and 24 months (63.2%) was higher among enrolled patients compared with those who did not enroll (69.6 and 54.8% at 12 and 24 months, respectively) (Fig. 1). This observation may indicate that early patient interaction and communication via the program during the initiation period is key to motivating patients to maintain persistence. In turn, the availability of an osteoporosis liaison service may be one of the beneficial options for facilitating this early interaction at the initiation of a new treatment. The key finding of a decreased likelihood of subsequent fracture after 6 months of teriparatide treatment from the primary observational study upon which this secondary analysis is based reinforces the importance of persistence [18]. Patients who discontinue within 6 months would not expect to gain this benefit from teriparatide, and so support during treatment introduction is one of the critical factors to achieving effectiveness.

Benefits on adherence were not assessed or reported in other studies of patient-support programs involving teriparatide [14‐16]. Despite this, the proportion of patients with high adherence (MPR > 80%) was 48% at 24 months in a medical claims database study in the USA [11], which, despite differences in the definition of good vs. high adherence, supports our finding for patients who did not enroll in the program. In addition, our findings extend those reported in two separate studies of medical and pharmacy claims databases in Japan. The proportion of patients with high adherence (MPR > 80%) in these two studies was 61% at 12 months [21] and 56% at 18 months [12] of teriparatide treatment.

Our findings of a difference in persistence between patients who were enrolled and those who were not enrolled should take into account the descriptive nature of the analyses and that the two groups were not matched for baseline characteristics. However, benefits related to persistence have been reported in other studies of patient-support programs involving teriparatide treatment [14‐16]. Spanish patients (N = 23,069) with severe osteoporosis, who were enrolled in a program involving regular phone contact with trained nurses designed to help them administer teriparatide correctly, had a persistence rate of 64.1% at 24 months [14]. An Italian study found that the persistence rate at 18 months among those enrolled in a support program (85.6%, N = 382) was 8.2% higher than that among non-enrolled patients (persistence = 77.4%, N = 398) [16]. Finally, a large study of French patients (N = 5413) also reported high persistence (81.5% over 15 months) among those enrolled in a patient-support program [15]. The lower persistence rate noted among Japanese patients in our study compared with those noted in the Italian and French studies may relate to the intensity of follow-up and injection training offered, particularly at the start of the program. For example, in the French study, up to five phone calls could be made during the first month to train patients in self-injection and evaluate their ability, with further home-based training offered for patients experiencing difficulty [15]. Similarly, patients enrolled in the Italian study were phoned weekly during the first month [16]. In contrast, Japanese patients enrolled in this analysis of the support program were called at the patient’s request during the first 6 months, in addition to receiving a monthly calendar containing a daily injection checklist and certificates of recognition for treatment milestones. Our findings, however, were similar to those of a Canadian patient-support program among 1676 patients administering 6-monthly injections of denosumab [22], in which 59.1% of patients persisted with denosumab at 24 months. In this Canadian study, follow-up by phone or email continued at 6-monthly intervals over the course of therapy.

Apart from higher adherence, bivariate analysis found several differences in baseline demographic and clinical characteristics among patients with patient-support program enrollment. Specifically, there were statistically greater proportions of females, patients with a family history of fracture, prevalent non-vertebral fractures, less severe back pain, and previous use of bisphosphonates and active vitamin D₃ among those enrolled in the patient-support program. The fact that patients who enrolled in the program had the highest rate of prevalent non-vertebral fractures suggests that this was possibly a key motivator for enrollment. Alternatively, patients who did not enroll may have required less support given that their rate of good to very good adherence was higher than that with an unknown enrollment status (48.3 vs. 28.7%, respectively). Patients who did not enroll had significantly higher mean back pain scores at baseline (55.8 ± 30.9) compared with those who did enroll (46.9 ± 30.6). This may relate to the lower use of previous osteoporosis medication among patients who did not enroll and possible use of alternative methods to improve their back pain. Ultimately, however, it is difficult to make conclusions based on the differences between the enrollment groups in terms of back pain and the prevalence of non-vertebral fractures for two reasons: a clear pathophysiological mechanism for pain (e.g., from spinal cord compression) has not been confirmed in patients with fractures and the manifestations of back pain are variable in this population [18]. Our finding that a higher proportion of females tended to enroll in the patient-support program in Japan may relate to higher anxiety among females about their higher fracture risk and such programs being less attractive to men, although this is speculative. This finding also contrasts with that of Tamone et al. [16], who found that a lower proportion of females enrolled in an educational and phone call follow-up program (86.9% vs. did not enroll 97.7%) to help Italian patients with severe osteoporosis administer once-daily injections of teriparatide. However, this difference was found not to influence persistence rates [16]. Finally, from multivariate analysis using a backward elimination method, the independent variables in addition to the patient-support program associated with good to very good adherence among Japanese patients were non-smoking status and previous osteoporosis medication use. Other studies of patient-support programs with teriparatide did not assess these associations with treatment adherence [14‐16]. However, the Canadian study of denosumab found several factors significantly associated with higher 24-month denosumab persistence, including private medication insurance, age 75 or above, previous osteoporosis medication use, and previous fracture [22].

In conclusion, Japanese patients with osteoporosis who enrolled in a patient-support program were found to have better adherence to and observed better persistence with once-daily teriparatide in this secondary analysis of data conducted in a real-world clinical setting [17]. Our findings highlight the value of patient-support programs in helping patients adhere to and persist with their medication regimen, which has been shown to provide benefits in terms of clinical outcomes [10‐13]. Further work would help improve adherence outcomes. Treating physicians may also find that this program provides them assistance in encouraging adherence and persistence among their patients.