Study selection

Our search protocol yielded a total of 1229 articles including 1205 from database search and 24 via other methods. After duplicates were removed, a total of 901 were screened by titles and abstracts, and 96 articles remained for full-text review for eligibility. A final count of 9 articles met all criteria and were included in this review. A PRISMA 2020 flow diagram illustrating the process of selecting articles is shown in Fig. 1.

Fig. 1Characteristics of the included studies

Of the nine studies that were identified and met the criteria, six studies were conducted in Canada [31, 33, 35,36,37,38], one each in Australia [34], Belgium [32], and the US [39]. The study design consisted of one phase-2 randomised control trial (RCT) [31], three ‘before and after’ observational studies [32, 36, 38], one prospective cohort study [34], one retrospective observational study [33], one retrospective cohort study [39], and two surveys (clinicians [35] and patients [37]). Study data is summarised in Table 1 with more detailed information in Supplementary Table 2.

Table 1 A brief summary of the characteristics of the studiesThe i-PARIHS framework analysis

Table 2 summarises the findings using the iPARIHS framework with references to the studies aligned with each construct.

Table 2 Summary of the key findings using the i-PARIHS frameworkThe innovation constructSource of innovation (underlying knowledge)

All studies were based on published recommendations to reduce osteoporosis and fracture risk for PCa-ADT users. A health belief model was used by a Canadian research team to support the importance of patient education in delivering care [31, 36, 38]. They also adopted their intervention approach using strategies for women with osteoporosis [31], and findings from a previous systematic review of point-of-care reminders to increase adherence to recommendations [36]. A pathway framework was applied when a new pathway was developed by a Belgium team in referring patients for the comprehensive management of ADT-induced complications [32].

Types of innovationUsing an existing bone health care programme to proactively screen and treat patients

A study in the US proactively identified PCa-ADT patients within an existing osteoporosis prevention programme for members of a healthcare community [39]. Using the electronic medical record system and cancer registry, the authors found screening and treatment protocols were easily implemented. The patient population in the study was also large and diverse and had equal access to healthcare (The study contained 1482 PCa-ADT patients including 17% black men). The study reported a 70% reduction in hip fractures (from 18.1 to 5.1 per 1000 person-years) and an increased uptake of BPM (from 3.2% to 29.2%). For patients who sustained a hip fracture, the median interval from the first ADT treatment to hip fracture was also longer in the intervention group (801 vs 528 days) [39].

Development of new services with new care pathways or clinics incorporating existing services

New pathways or clinics were developed in three studies, all at a tertiary hospital level [32,33,34]. In a before–after study in Belgium, a multidisciplinary team consisting of a radiation oncologist, urologist, psychologist, dietician, oncology nurse, physical therapist, social worker, and pathway facilitator was created for pathway development, implementation, and evaluation according to a seven-phase pathway framework [32]. The multidisciplinary team analysed daily practices, deficits in the care process, and potential transferrable components of pathways. In addition to bone health, the pathway also assessed cardiometabolic side effects and provided advice for exercises, nutrition, and psychoeducation to patients. BMD screening increased from 10 to 58% after the pathway implementation [32]. The positive effects are attributable to standardisation of the care process [32].

Patients were also referred to a new dedicated clinic, for example, a Men’s Health Clinic by an Australian team where metabolic and bone complications were assessed and managed. After 2 years of attendance at the clinic, the proportion of men receiving BPM increased from 17 to 61%. Treatment maintained total hip BMD (+ 0.007 ± 0.239 g/cm2, p = 0.649) while a significant decline in hip BMD by 2.5% (− 0.026 ± 0.036 g/cm2, p < 0.0001) was reported in those without treatment [34].

A Canadian team referred patients to a bespoke osteoporosis clinic where they were seen by one doctor who specialised in ADT treatment-induced osteoporosis [33]. High-quality bone health care was achieved as defined by a valid fracture risk assessment tool used in all patients, and all patients with high fracture risk were recommended BPM [33].

Incorporating BMD tests into routine patient care

Researchers from Canada developed different ways to implement BMD tests for PCa-ADT patients. A healthy bone prescription tool entitled BoneRx that contained a pre-populated BMD request was provided to the PCa specialists to prompt them to order a BMD test for PCa-ADT patients. In a pilot prospective, before-and-after clinical trial, at 3-month follow-up, patients who received the study intervention were more likely to receive BMD testing (odds ratio 3.3) compared to the control group [38]. In a recent implementation study, BoneRx was provided to patients at the initiation of ADT. After 6 months of ADT treatment, significantly more patients received BMD tests (59.5%) compared with patients who did not receive BoneRx (34.7%) [36]. In another study, two strategies of requesting BMD tests were compared: BMD was ordered either by the patient’s family physicians or facilitated by a bone healthcare coordinator. Patients were also provided with a bone health pamphlet. Significant increases in receiving BMD tests within 6 months were seen in both groups (58% with family physicians, 78% with a coordinator) as compared with the usual care group (36%) [31].

The context construct

At an organisational level, either within a hospital or a community health care provider, all studies were driven by the gap between guidelines and poor real-world practice [31,32,33,34,35,36,37,38,39]. However, there was no direct information reported in any study related to local settings that influenced the implementation. A research group from the largest cancer centre in Canada conducted a number of research activities to investigate ADT-induced bone complications, indicating a high level of support and leadership within the organisation [31, 36,37,38]. No studies reported any information related to external contextual factors, e.g. policy or economic drivers.

The recipients constructSkills, knowledge, and resources for clinicians

Urologists and oncologists should prescribe ADT with the knowledge of its complications on bone loss [33]. However, they reported lack of time, supporting structure, and resources as major barriers [32, 35]. To address this, a multidisciplinary team developed a new pathway that incorporated transferrable existing services. A central pathway coordinator was created to facilitate the referral [32]. Lack of training and confidence were also reported in a survey from Canadian radiation oncologists and urologists [35]. Only 4.6% of survey respondents routinely used fracture risk assessment, and 37.3% had never even heard of them [35]. When osteoporosis was detected in PCa-ADT patients, referrals were commonly made to the patient’s primary care physician (47.4%), endocrinologist (23.1%), or an osteoporosis clinic (19.2%) [35]. Chahin et al. [33] also reported major barriers to optimising bone health for men on ADT are the lack of knowledge among PCa specialists and the perception that this issue would be looked after by the patient’s primary care clinicians [33]. It was suggested that referring patients to primary care clinicians or bone health specialists might be appropriate [33]. The research team subsequently conducted studies that involved family physicians or osteoporosis specialists’ referral. In a randomised phase 2 clinical trial to assess two education-based interventions to improve bone health care, one strategy was to refer patients to their family physicians who were provided with a hospital letter that contained recommendations for BMD screening and bone health care information [31]. The result showed that with the family physician’s input, BMD ordering was significantly increased compared with usual care (58% vs 36%, p = 0.047). In another study, patients were referred to a dedicated osteoporosis clinic which was specialised in ADT-induced bone loss. Patients received high-quality bone healthcare: all patients received fracture risk assessment and BPM was offered to all patients with a high risk of osteoporosis [33].

Support and collaboration at a multidisciplinary team level

The novel patient care pathways were predominantly multidisciplinary [31, 32, 34, 36, 38, 39]. However, perspectives from different specialists on the new services were usually not reported. In the above phase 2 clinical trial involving family physicians, despite an improvement in bone health care, family physicians reported a very low satisfaction rate (26%), which was the only feasible outcome that did not meet the target [31]. Problems included hospital letters that were neither timely, clearly written nor helpful. The authors recommended eliciting feedback from family physicians before embarking on a larger trial [31].

Patients as recipient

Patients as service users fit into a wide range of stakeholders within the recipient construct, and their views should be consulted [28]. Only one study described gathering feedback from patients as well as clinicians on intervention materials and workflow to fine-tune the implementation approach [36]. However, no detailed information, such as what patients’ input was or whether any changes were made from patients’ feedback, was reported [36].

A survey with 175 PCa-ADT patients in Canada exploring their knowledge and healthy bone behaviour (HBB, including calcium and vitamin D intake and exercises) showed that most patients lacked basic information and awareness of ADT-induced bone complications despite their relatively high health motivation [37]. The findings supported the application of the Health Belief Model in this population that increasing knowledge of bone health may increase compliance with HBB guidelines, and highlighted the importance of patient education [37]. The research team then incorporated patient education together with a healthy bone prescription tool, i.e. BoneRx, for the clinicians to prompt them to request BMD tests [31, 36, 38]. Different strategies were developed to provide information and education to patients. For example, a customised written booklet on bone health was created and given to patients at the initiation of ADT [36] or during ADT treatment [31, 38]. The booklet was created through a collaboration with specialists in osteoporosis and PCa, and the hospital patient education department [