Closing the gap in osteoporosis management: the critical role of primary care in bone health

Abstract Objective The World Health Organization issued a call to action for primary care to lead efforts in managing noncommunicable diseases, including osteoporosis. Although common, osteoporosis remains underdiagnosed and undertreated. Primary care practitioners (PCPs) are critical in identifying individuals at risk for osteoporosis and osteoporotic fractures; however, recent advances in assessment, diagnosis, and treatment of osteoporosis have not been incorporated into clinical practice in primary care due to numerous reasons including time constraints and insufficient knowledge. To close this gap in clinical practice, we believe PCPs need a practical strategy to facilitate osteoporosis assessment and management that is easy to implement. Methods In this article, we consolidate information from various global guidelines and highlight areas of agreement to create a streamlined osteoporosis management strategy for a global audience of PCPs. Results We present a systematic approach to facilitate osteoporosis assessment and management that includes four steps: (1) identifying patients at risk through proactive screening strategies, (2) investigating and diagnosing patients, (3) intervening with personalized treatment plans, and (4) implementing patient-centered strategies for long-term management and monitoring of patients. Conclusion Primary care has a central role in ensuring the incorporation of key elements of holistic care as outlined by the World Health Organization in managing noncommunicable diseases including osteoporosis; namely, a people-centered approach, incorporation of specialist services, and multidisciplinary care. This approach is designed to strengthen the health system’s response to the growing osteoporosis epidemic. PLAIN LANGUAGE SUMMARY Osteoporosis is a chronic condition associated with aging in which bones become “porous” and weak, and are more likely to break (i.e., fracture) even with minimal trauma such as tripping or falling from a standing height. A broken bone is a serious condition that not only affects daily activities, but can also lead to reduced quality of life, need for caregiver support, work loss, hospital and rehabilitation costs, nursing home costs, and increased mortality. Although osteoporosis is common, it is often undiagnosed or untreated, leaving many people at risk for experiencing broken bones. A broken bone increases the risk of more broken bones. Given the growing size of the aging global population, osteoporosis and the risk of broken bones represent an urgent problem and growing burden. We need ways to make it easier for primary care practitioners (PCPs), such as family physicians, internists, physician assistants, nurse practitioners, and nurses, to include osteoporosis care as part of routine clinical visits. In this article, we discuss the critical role of PCPs in early detection, diagnosis, and treatment of osteoporosis as they are often the first point of contact for at-risk patients. We present a simple, four-step approach to help PCPs and patients navigate the journey from osteoporosis diagnosis to a treatment plan. The four steps are to: (1) identify at-risk patients by screening for weak bones or osteoporosis, (2) perform necessary tests to diagnose patients, (3) develop a personalized treatment plan, and (4) determine long-term strategies for managing and monitoring bone health. Video Abstract Thoracic endovascular aortic repair of a large, multilobulated, calcified thoracic aneurysm. Read the transcript Watch the video on Vimeo

World Health Organization's call to action for primary care The World Health Organization (WHO) issued a call to action for primary care to lead efforts in screening, assessing, and managing noncommunicable diseases, which are major causes of death and disability worldwide 1 . Osteoporosis is one such disease, characterized by low bone mass, structural deterioration of bone tissue, and disruption of bone microarchitecture. Osteoporosis can lead to increased risk of osteoporotic fractures resulting from low or minimal trauma experienced upon falling from a standing height or lesser impact 2,3 . Osteoporotic fractures include fractures of the hip, spine [clinical], wrist, humerus, tibia, pelvis, and vertebral fractures with limited clinical expression and can be life-altering events 3 . Given the growing size of the aging population globally, osteoporosis represents an increasing societal and economic burden worthy of dedicated focus by primary care practitioners (PCPs).
As the primary contact with patients, PCPs are uniquely positioned to identify and manage individuals at risk for osteoporosis and osteoporotic fracture. Despite this critical role, more recent advances in assessment, diagnosis, and treatment of osteoporosis are not regularly incorporated into primary care clinical practice due to multiple factors, including time constraints, insufficient knowledge, doubts about effectiveness of treatments, fear of adverse events, and misinformation. In the past, several organizations and bonehealth expert groups had developed clinical guidelines for osteoporosis management; however, the guidance was not always consistent or aligned across organizations, or across countries and regions. Notably newer and/or updated guidelines from a number of organizations including the American Association of Clinical Endocrinologists/American College of Endocrinology (AACE/ACE) 4 , the Bone Health and Osteoporosis Foundation (BHOF; formerly the National Osteoporosis Foundation) 5 , the International Osteoporosis Foundation (IOF) together with the European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO) 6 , the Endocrine Society 7 , and the North American Menopause Society (NAMS) 8 are now more aligned than previous guidelines on a number of points including criteria for patient classification by fracture risk category and recommendations on treatment plans per the risk categories. Additionally, the IOF and BHOF developed the Radically Simple Tool 9,10 , which is a simple visual to aid PCPs initiate dialogue with their patients about osteoporosis and fracture risk during medical consultations. With the aligned guidelines and the Radically Simple Tool 9,10 now available to PCPs, this critical group of healthcare providers are well positioned to successfully take action to manage osteoporosis and lower the risk of patients suffering debilitating fractures. To this effect, we have developed a practical approach to osteoporosis assessment and management that is easy to implement in clinical practice and can guide clinical decisionmaking. We consolidated information from various global guidelines and highlighted areas of agreement to create a streamlined osteoporosis management strategy for a global audience of PCPs, which we have also summarized in a related video abstract (see Video Abstract). This approach incorporates the key elements outlined by the WHO; namely, a people-centered approach, incorporation of primary care and specialist services, and multidisciplinary care, and is designed to strengthen the health system's response to the growing osteoporosis epidemic.

Osteoporosis burden
In 2000, an estimated nine million new osteoporotic fractures occurred worldwide 11 . A more recent study that evaluated fractures at all sites and for all ages reported an estimated 178 million fractures across 204 countries and territories in 2019, with most of the fractures occurring in the elderly 12 .
Globally, one in three women and one in five men aged 50 years and above will suffer an osteoporotic fracture during their remaining lifetime 2,11,13 ; the corresponding rates in the United States (US) are one in two women and one in four men 14 . Osteoporotic fractures can lead to reduced quality of life, need for caregiver support, work loss, hospital and rehabilitation costs, nursing home costs, and increased mortality 2,[4][5][6][15][16][17][18][19] . An analysis of Medicare data commissioned by the BHOF and conducted by the independent actuarial firm Milliman 15 showed that patients who suffer an osteoporotic fracture are likely to experience several other negative and costly health consequences, including hospitalizations (40% within 1 week after the fracture), subsequent bone fractures (14% within the first year following the prevalent fracture), and institutionalization in nursing care facilities (3%) 15 . In the US, the number of hospitalizations for osteoporotic fractures (43%) exceeded those for heart attack (25%), stroke (26%), and breast cancer (6%) from 2000 to 2011 20 (Supplemental Figure 1 in Supplemental Material). In Europe, osteoporosis-related disability is comparable to or greater than disability due to high blood pressure-related heart disease, rheumatoid arthritis, and asthma 2,11 . One-year mortality following hip fracture has been reported to range from 20% to 35% 15,21 .
In the US, the total annual expenditure for all osteoporosis-related fractures, including direct and indirect costs, was estimated at US$57 billion in 2018 and is projected to increase to over $95 billion in 2040 22 . In Europe, the total annual expense for new osteoporotic fractures for five of the largest European Union (EU) countries plus Sweden (EU6) was estimated at e37.5 billion in 2017 and is projected to increase to e47.4 billion by 2030 23 . The recent SCOPE 2021 report of osteoporosis data from 27 countries in the EU plus Switzerland and the United Kingdom (EU þ 2) estimated the economic burden of fractures in 2019 to be e55.3 billion 24 . The United Nations projects a dramatic increase in the global old-age dependency ratio (i.e. ratio of the population aged 65 years and above to the population aged 15-64 years), thus increasing the proportion of the population at risk for osteoporotic fractures 25 . In the EU þ 2 countries, the population aged 50 years and above is projected to increase by 11.4% between 2019 and 2034 and the number of fractures in general is expected to rise by 25% 24,26 . Hip fractures, considered the most life altering and devastating of fractures, are projected to increase by 240% in women and 310% in men by 2050, globally, compared with 1990 rates 27,28 . These regional and global projected figures are likely to increase osteoporosis-related healthcare costs.
Despite the high associated morbidity and mortality, osteoporosis remains underdiagnosed and undertreated 22 . Of the estimated 200 million women with an osteoporosis diagnosis globally 2 , fewer than 20% receive an osteoporosis diagnosis 2,29 ; of the patients who receive an osteoporosis diagnosis, fewer than 35% receive treatment even after a fracture [30][31][32][33] ; and of the patients who receive treatment, fewer than 50% persist with treatment beyond 6 months 31,34,35 (Supplemental Figure 2 in Supplemental Material). The SCOPE 2021 report for the EU þ 2 countries 24 estimated that 25.5 million women and 6.5 million men had osteoporosis in 2019, with 4.3 million fractures sustained in the same year. A follow-up article on osteoporosis management in the same countries reported an average increase in the treatment gap from 55% in 2010 to 71% in 2019; 10.6 million women eligible for treatment went untreated in 2010 and this number rose to 14.0 million women in 2019 26 .
Osteoporosis risk factors include prior fracture, age (65 years and above), low bone mineral density (BMD), parent hip fracture history, low body mass index (BMI), calcium/ vitamin D deficiency, low protein intake, use of medications that may increase bone loss (e.g. glucocorticoids), inadequate physical activity, tobacco smoking, and excessive alcohol intake 2,4,5,36-39 . If osteoporosis is unrecognized and left untreated, the traditional osteoporosis patient journey can often include multiple fractures across decades 40 . Importantly, occurrence of a recent fracture is most predictive of a second or subsequent fracture 15,[41][42][43][44] . One study reported a five-fold higher risk of fracture in the first year following a prevalent fracture 44 . Another study reported a 10% risk of recurrent fracture at year 1, 18% by year 2, and 31% by year 5 41 . The Milliman analysis 15 reported that about 14% of Medicare beneficiaries who had a new osteoporotic fracture suffered one or more subsequent fractures within 12 months of the initial fracture. These findings suggest that early treatment with agents that rapidly reduce fracture risk could prevent secondary fractures in high-risk individuals 6,43 , especially in individuals aged 65 years and above who had experienced a recent hip or vertebral fracture 45,46 . In patients who have not experienced a prevalent fracture, the focus should be on proactive primary prevention, which requires screening and treating patients at risk to prevent occurrence of the first fracture.
Of note, although there is agreement within the healthcare community on the need for osteoporosis screening 4,5,8,47 , the specific guidelines for screening such as patient age, frequency, and risk factors remain a point of debate among opinion leaders and health economics specialists. Despite this debate, given the high individual and societal burden that results from underdiagnosed and undertreated osteoporosis and subsequent fractures, there is a clear need for bone health working groups within the PCP communities to provide leadership consistent with WHO's call for primary care to lead efforts in screening, assessing, and managing noncommunicable diseases.

The crucial role for primary care
Osteoporotic fractures typically present via the emergency department or orthopedic and neurosurgery services. Despite initial acute attention to the fracture, many patients fail to receive continued care beyond the fracture episode 5 . Moreover, the clear risk factors for osteoporosis go completely unrecognized 2,22,29 . As providers who generally have a long-term, trust-based relationship with patients, PCPs can and should play a critical role in postfracture osteoporosis assessment and management. Osteoporosis does not belong to any one specialty and, given the rising prevalence of osteoporosis and limited number of specialists in the field, primary care involvement is critical.
A growing trend is the establishment of postfracture care (PFC) or multidisciplinary osteoporosis care programs that consist of teams of healthcare providers including PCPs, specialists (orthopedic surgeons, endocrinologists, rheumatologists, geriatricians, and others), and allied healthcare workers (dietitians, nurses, nutritionists, pharmacists, and physical therapists) who work in a coordinated way to assess, diagnose, educate, and treat patients with fractures to avoid subsequent fractures. The most common PFC programs include fracture liaison services (FLSs) that focus on preventing subsequent fractures 48 and geriatric/orthogeriatric services (OGSs) that focus on improving overall outcomes (morbidity, mortality, and/or physical function) for inpatients hospitalized for a fracture 49 . PFC programs have been reported to improve patient follow-up, testing, and treatment rates 48,[50][51][52] ;with one study demonstrating increased BMD testing from 21% to 93%, increased vitamin D assessments from 25% to 84%, increased calcium/vitamin D prescriptions from 36% to 93%, and increased osteoporosis medication prescriptions from 20% to 54% 51 . Results from a study that explored the course of health state utility value over 3 years in patients who had experienced a recent fracture and were enrolled in an FLS suggested that although the overall change in health-related quality of life was not significant over the 3 years, significant improvements were observed at 6 and 12 months compared with baseline 53 .
The success of PFC programs hinges on adoption of best practices by PCPs with regards to referring and receiving patients in these programs. It is important to maintain appropriate communication with patients and other members of the PFC program, maintain patients' treatment plans, continue to conduct patients' bone-health assessments, and continue to evaluate patients' osteoporosis risk factors.
In areas where PFCs are not yet available, PCPs can take the initiative to establish one using the educational resources and counsel provided by a number of medical societies including the IOF's Capture the Fracture, the IOF PFC resource center, BHOF FLS resources, BHOF FLS coding guide, American Orthopaedic Association (AOA)'s Own the Bone, American Geriatrics Society (AGS)'s CoCare model, and the Fragility Fracture Network (FFN) Clinical and Policy Toolkits. An opportunity also exists within the broader healthcare community to recommend specific actions and frameworks for different models of PFCs in order to improve patient care following a fracture.
Another avenue for enhancing osteoporosis management is for bone health organizations (e.g. IOF, BHOF, ESCEO) to develop relationships with PCP leading organizations such as the World Organization of National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians (WONCA). One possibility is to establish bone health working groups on osteoporosis management under the WONCA umbrella, similar to currently existing working groups on quality, digital health, and equity. Developing multidisciplinary clinical leadership in bone health at the primary care level could also strengthen relationships with leading organizations on topics such as family medicine, nursing, and allied professions.
Four key steps to manage osteoporosis and prevent disability due to osteoporotic fractures In Figure 1, we present a schematic overview of a proposed algorithm to aid in decision-making when managing patients at risk for osteoporotic fractures. This approach includes four steps: (1) identifying patients at risk through proactive screening strategies; (2) investigating and diagnosing patients with osteoporosis based on age, fracture history, BMD and/or fracture risk score; (3) intervening with personalized treatment plans for patients diagnosed with osteoporosis or at high risk for fracture; and (4) implementing patient-centered plans for long-term monitoring and reevaluation of patients to minimize fracture risk (see Video Abstract). We discuss implementation of this algorithm within a clinical setting and also provide sample dialogues with patients to demonstrate effective communication strategies.
Step 1: identify patients at risk for osteoporotic fractures through proactive screening The strongest predictor of a future osteoporotic fracture is a prior osteoporotic fracture 4-6 . Occurrence of a fracture at one anatomic site is usually an indication of systemically compromised bone quality and increases the risk of future fracture at other sites [54][55][56][57] ; therefore, risk assessment, either in person or via virtual care (medical consultations through video calls, phone calls, emails, and text messaging) should start by eliciting a history of fracture ( Figure 2). According to guidance from the BHOF 5 , AACE/ACE 4 , IOF/ESCEO 6 , and a multi-stakeholder coalition assembled by the American Society for Bone and Mineral Research (ASBMR) 45 , patients with a recent hip or vertebral fracture are at very high risk for fracture and should immediately proceed to a personalized treatment plan.
Patients who do not report a recent hip or vertebral fracture should be asked follow-up questions designed to identify risk factors for osteoporosis or osteoporotic fractures (including fractures of the hip, spine, wrist, humerus, tibia, and pelvis) ( Figure 2). Patients who are identified as having any of the risk factors for osteoporosis or osteoporotic fractures, show signs and symptoms of a vertebral fracture and/ or the risk of falling, or women who are postmenopausal should be further investigated as described in Step 2 below. Engaging with patients using appropriate language is critical. A sample dialogue between a practitioner and a 60-year old postmenopausal woman who has an annual visit is presented in Supplemental Figure 3 in Supplemental Material, Clinical Scenario 1.  A. J. SINGER ET AL.
Step 2: investigate and diagnose patients with osteoporosis who are at risk for osteoporotic fractures Step 2 consists of investigating and diagnosing patients suspected to be at risk for fracture either due to age, the presence of signs and symptoms of a vertebral fracture and/or their risk of falling ( Figure 3). The investigation and diagnosis process includes referring patients for BMD measurement performed by dual energy x-ray absorptiometry (DXA) to determine the degree of bone loss and, when indicated, to proactively perform vertebral imaging by lateral spine radiographs or vertebral fracture assessment (VFA) to identify any previously undiagnosed fractures. The DXA scan provides a T-score that is derived by comparing a patient's BMD values (in g/cm 2 ) with BMD values from a uniform Caucasian female normative database; the International Society of Clinical Densitometry recommends the normative database be used for women of all ethnic groups and a Caucasian female reference group be used for men of all ethnic groups 58 . The patient's T-score can be used in fracture risk prediction and diagnosis. Osteoporosis is diagnosed with a BMD T-score À2.5 at any site (lumbar spine, total hip, femoral neck, or distal radius); patients should be given one diagnosis based on the lowest BMD score. Of note, a clinical diagnosis of osteoporosis can also be made in a person with a hip or vertebral fracture, regardless of bone density and in someone  with low bone mass (osteopenia; T-score between À1.0 and À2.5) and prior fracture at the humerus, pelvis, or distal radius) 4-6 (Supplemental Figure 4 in Supplemental Material).
Clinical guidelines in some regions recommend BMD testing for postmenopausal women aged 65 years and above and postmenopausal women aged 65 years and below who have risk factors [4][5][6]59 . There is no overall consensus on BMD testing for men, but some guidelines recommend BMD testing for men over 70 years of age as bone loss accelerates at age 70 years and above in men 60,61 . Guidelines in some regions do not currently recommend BMD testing for all patients; rather, some guidelines recommend BMD measurements using a case-finding strategy 62 . As best practice, we recommend implementing BMD testing as part of bone health screening per accepted clinical guidelines and as directed by country and local regulations. It may be important to establish a BMD baseline that can be followed over time with or without treatment. This would be analogous to preventive practices such as periodic examinations in adults to screen for diabetes (fasting glucose or hemoglobin A1C) 63 and heart disease (cholesterol) 64 , and mammography for women to screen for breast cancer 65 .
PCPs should also try to identify patients suspected of having vertebral fractures. This includes routinely asking patients if they have experienced prolonged or unusual back pain, which may signal vertebral fracture (Figure 3). PCPs should ask patients about signs that might indicate vertebral fractures not previously clinically recognized, such as loss of height by !2 cm, the inability to touch the back of the head to the wall when standing with back and heels against the wall (occiput-to-wall distance), or kyphosis (progressive spinal curvature). PCPs should also evaluate for factors that increase the risk of falling such as gait abnormalities, balance problems, and decreased ability to perform the timed "up-andgo" test used to assess balance and gait 4-6 ( Figure 3). Patients with vertebral fractures confirmed by lateral spine radiographs or densitometric VFA or those with bone loss on BMD measurements (T-score À2.5 or between À1.0 and À2.5 with prior fracture) (Supplemental Figure 4 in Supplemental Material) should proceed to a personalized treatment plan (Figure 1 and as described in Step 3).
For patients with signs and symptoms that could indicate possible vertebral fracture but no history of fractures and/or a BMD T-score >À2.5, further risk assessment should be performed. Over 80% of postmenopausal women with osteoporotic fractures have a T-score >À2.5 66 , suggesting that BMD is not the only factor contributing to fracture risk. It has now been established that BMD and other clinical factors taken together improve fracture risk prediction 4-6 .
Fracture risk prediction can be easily performed using online risk assessment tools such as the Fracture Risk Assessment Tool (FRAX), Garvan, American Bone Health (ABH) Fracture Risk Calculator TM , and others to help identify patients at increased risk for fracture. These tools appear to perform well and are moderately accurate at predicting fracture risk within a specified time frame 43,59 . FRAX is the most widely used prediction tool and incorporates risk factors, including age, sex, BMI, fracture history, and others 43 with or without BMD measurement, to predict absolute risk as a 10year probability of major osteoporotic fracture (fracture of the hip, spine [clinical], wrist, or humerus) and hip fracture 2,67 (Supplemental Figure 5 in Supplemental Material). Overall, FRAX and other online tools are meant to guide risk stratification, diagnosis, and treatment decisions; however,  68 Routes of administration by drug class: the bisphosphonates ALN, IBN, and RIS are administered as oral tablets; IBN is also available as intravenous injection in several countries; the bisphosphonate ZOL is administered by infusion; denosumab is administered by subcutaneous injection; PTH class of drugs ABA and TPTD are administered by subcutaneous injection; raloxifene is administered as an oral tablet, and romosozumab is administered by subcutaneous injection. Of note, bisphosphonate antiresportive agents (ALN, IBN, RIS, and ZOL) have a different mechanism of action from the other antiresorptive agent denosumab, and it is important that clinicians understand the differential bone effects between these two types of antiresorptive agents. a Regional and local guidelines may override some of these criteria based on differences in FRAX data and cost-effectiveness thresholds. b If FRAX is not available, major determinants of risk should include age, BMD, fractures, and medication harmful to bone. c Off-treatment period for consideration after treatment with bisphosphonates for 3-5 years due to BMD gains plateauing at $3 years. d Applicable if decision is made to discontinue denosumab. e ENDO requires both risk factors of multiple fractures and very low T-score < -3.0 to be met for very high risk categorization. f Very high risk category includes the imminent risk subcategory that refers to the relative risk of recurrent fracture that is highest in the first years following an index fracture in patients. ABA, abaloparatide; ALN, alendronate; BMD, bone mineral density; ENDO, Endocrine Society; FRAX, Fracture Risk Assessment Tool; IBN, ibandronate; MOF, major osteoporotic fracture; PTH, parathyroid hormone; RIS, risedronate; TPTD, teriparatide; ZOL, zoledronic acid. clinical judgement is important in determining whether or not to treat patients. Recent guideline updates, including the 2019 updated IOF/ESCEO guideline 6 also summarized in a 2020 position paper 43 , the 2020 updated AACE/ACE guideline 4 , and the the 2020 Endocrinology Society guideline 7,68 , provide criteria for categorizing patients into high risk and very high risk for fracture using FRAX assessment strategies in combination with prior fracture occurrence, age, and T-scores (Figure 4). Occurrence and recency of a prior fracture are critical in determining patients' risk for future fracture. In the US, patients are placed into the high risk category for fracture if they are postmenopausal and have experienced a prior fracture, have a T-score À2.5, or have a T-score between À1.0 and À2.5 (osteopenic) together with a FRAX probability of !20% for major osteoporotic fracture or !3% for hip fracture 4,6,7,68 . Patients are placed into the very high risk category for fracture if they have experienced a prior fracture within the past 12-24 months 4,6,7,43,68 due to the high likelihood of a subsequent fracture occurring following a recent fracture 15,[41][42][43][44] . Patients are also placed into the very high risk category if they have experienced multiple fractures, had fractures while on osteoporosis medications, have very low T-score (<-3.0), and/or have FRAX probability >30% for major osteoporotic fracture and >4.5% for hip fracture 4,6,7,43,68 .
Within the very high risk category is the subcategory of patients at imminent risk for fracture, which refers to the relative risk of recurrent fracture that is highest in the first year following an index fracture in patients 7,41,42,44,69 .
Other regions and countries may use thresholds for T-score and FRAX probability similar to those used in the US for categorizing patients into high risk and very high risk categories, for example Australia. However, these thresholds may vary in other regions and countries and so it is imperative for PCPs to follow regional and country guidelines.
In cases where fracture risk is determined to be high or very high by the criteria provided in the updated guidelines 4,6,7,43,68 (Figure 4), patients should be immediately moved to a personalized treatment plan (Figure 1 and as discussed in Step 3). All other patients, including those at low risk of fracture and/or who have not started treatment with an osteoporosis medication, should be counselled on the use of nonpharmacological measures to preserve bone health (as described in Step 3) and should continue to be monitored through physical examinations, updated history, BMD testing, and FRAX or other risk assessment tools [4][5][6] . For particularly complex cases (e.g. patients with multiple fractures while on treatment or patients with comorbidities and who are on concomitant medications), PCPs may want to refer the patient to a specialist, such as an endocrinologist, rheumatologist, or other clinician with bone-health expertise, or to a PFC program for further patient care.
A sample dialogue with a 60-year-old postmenopausal woman with back pain on a visit to discuss BMD and FRAX results is presented in Supplemental Figure 3 in Supplemental Material, Clinical Scenario 2.

Nonpharmacologic intervention
Nonpharmacologic strategies to preserve bone strength and reduce the risk of fracture, including fall risk assessment and reduction, should be incorporated into routine patient care and assessed at visits for all patients (Supplemental Figure 6 in Supplemental Material). These interventions include counselling patients on maintaining sufficient dietary protein and adequate calcium intake (800-1200 mg/day), preferably from dietary sources with use of calcium supplementation to make up any shortfall, and prescribing vitamin D (800-1000 IU/day or more as indicated) in patients at risk of, or showing evidence of, vitamin D deficiency 6 . It also includes recommending regular weight-bearing, muscle-strengthening, and balance exercises tailored to the needs and abilities of the individual patient, counselling on cessation of smoking and consumption of alcohol in moderation, and guidance on measures to prevent falls 4-6 . If warranted, patients can be referred to a dietitian, physiotherapist, or specialist nurse for patient-specific care. It is also important to follow-up with patients on nonpharmacologic intervention, generally 1-2 years after initial osteoporosis screening, to assess for any changes in clinical status or risk factors, and to perform or repeat BMD testing as appropriate. The time frame at which repeat BMD testing is recommended should be individualized based on initial BMD results and clinical risk factors.

Pharmacologic intervention
In patients identified to be at risk for fracture or in those in whom nonpharmacologic intervention is inadequate, personalized treatment plans that incorporate pharmacologic agents should be developed. It is also important to include or continue nonpharmacologic strategies as part of the personalized treatment plan.
Effective therapeutic options for osteoporosis include antiresorptives such as bisphosphonates (alendronate 70 , ibandronate 71 , risedronate 72 , and zoledronic acid 73 ), a RANK ligand inhibitor (denosumab) 74 ,a selective estrogen receptor modulator (raloxifene) 75 , and menopause hormone therapy 6 ; and anabolic agents such as a parathyroid hormone (PTH) agonist (teriparatide 76 and its biosimilars), a PTH receptor agonist (abaloparatide) 77 , and a sclerostin inhibitor (romosozumab) 78 (Supplemental Figure 6 in Supplemental Material). Osteoporosis is a chronic disease, and as such, requires lifetime management, which generally includes pharmacologic treatment and may include a sequence of several medications 79 . The choice of treatment options is a shared decision between the PCP and patient and takes into consideration several factors, including the patient's baseline level of risk, concomitant medical conditions, and patient preference, and clinical guidelines and local recommendations ( Figure  4) [4][5][6][7]43,68 . If needed, treatment decisions can be made in consultation with a specialist so that patient care on subsequent visits remains aligned among healthcare providers.
For treatment-naïve patients at high risk for fracture, guidelines recommend treatment with an antiresorptive agent as the preferred initial medication 4,6,7,43,68 ( Figure 4). However, for treatment-naïve patients at very high risk for fracture, guidelines recommend an anabolic agent as the preferred initial medication 4,6,7,43,68 (Figure 4), as treatment with an anabolic followed by antiresorptive therapy is likely to provide greater improvement in bone strength compared with treatment with antiresorptive therapy followed by an anabolic or with antiresorptive therapy alone 80 . An ESCEO expert working group concluded that current evidence supports starting patients at very high risk of fracture on an anabolic agent followed by maintenance therapy using an antiresorptive agent and recommend addressing subsequent need for therapy to manage osteoporosis over the lifetime of patients 81 Many patients who are already on antiresorptive therapies may be at very high risk for fracture due to incident fracture, declining BMD, or BMD that remains persistently in the osteoporotic range (T-score -2.5). These patients might also benefit from treatment with anabolic therapy that should be followed with antiresorptive therapy to preserve BMD gains obtained while on an anabolic 80 . Once patients are placed on the antiresorptive therapy of choice, reassessment is recommended every 1 to 2 years 4 or after a fracture 4,68 .
Bisphosphonates and denosumab, although both antiresorptives, have different mechanisms of action and differential bone effects. Treatment with bisphosphonates has been shown to result in BMD gains plateauing at $3 years 70-73 ; however, the half-life of bisphosphonates in bone is very long (1-10 years) and as such bisphosphonate effect on bone is not readily reversible on drug discontinuation. Thus, a temporary suspension of bisphosphonate therapy (i.e. a drug holiday) after 3-5 years has been recommended in appropriate individuals, as antifracture benefits may continue due to skeletal retention and thus implementing a drug holiday may reduce the risk of potential skeletal adverse events 4,5 . Conversely, the effect of denosumab on BMD gains is reversible, and while continuous BMD gains have been observed with up to 10 years of treatment 82 , its discontinuation can result in the loss of bone mass gained while on treatment, with reports of multiple vertebral fractures observed following denosumab discontinuation [83][84][85][86] . As such, discontinuation of denosumab treatment should be followed with an antiresorptive, specifically a bisphosphonate. Treatment with anabolics (teriparatide, abaloparatide, or romosozumab) should also be followed by antiresorptive treatment to maintain BMD gains 5,87,88 .
Like most therapeutic treatments, considerations for possible adverse events including risk/benefit should be incorporated in decisions for a treatment plan and discussed with the patient. Safety information for each drug, including side effects, precautions, and boxed warnings, can be found in the prescribing information and have also been extensively reviewed as well as summarized in multiple articles 5,68,89,90 .
Some guidelines and expert panels also address treatment goals; i.e., goal-directed therapy, which is a proactive treatment strategy with a clear goal and a commitment to change the therapy if the target is not achieved. The AACE/ ACE guideline cites an increase in BMD with no evidence of new fractures or vertebral fracture progression as a response to therapy 4 ; a European expert panel recommends the recovery of prefracture functional level and reduction of subsequent fracture risk as a treatment target 91 ; and the ASBMR-NOF Working Group on goal-directed treatment for osteoporosis recommends freedom from fracture, a T-score >-2.5, or attaining a lower estimated fracture risk level than at the time of treatment initiation 92 . Irrespective of the goal selected, it is important to retain the diagnosis of osteoporosis even after the patient achieves improvement, as osteoporosis is a chronic condition like diabetes and hypertension, and requires ongoing treatment and monitoring.
During the COVID-19 pandemic, guidance 93,94 had recommended virtual care, use of alternative methods for administering osteoporosis medications given by subcutaneous injection or intravenous delivery, and temporary transition from injectable to oral medications if necessary until original treatment could be resumed. As a result, various online tools became more widely used for patient care. Incorporation of virtual care and expanded/alternative access measures into routine practice should be considered beyond the COVID-19 pandemic, as this can benefit patients with mobility restrictions or those in rural settings.
A sample dialogue with a 72-year old postmenopausal woman to discuss the results of her annual BMD test is presented in Supplemental Figure 3 in Supplemental Material, Clinical Scenario 3.
Step 4: implement, monitor, and re-evaluate Monitoring can include ongoing bone-health assessments such as periodic BMD testing and reviewing risk factors as well as interval history of falls and fractures ( Figure 2). Patient adherence is critical in realizing the benefits of osteoporosis therapy, and motivating the patient to continue treatment should be part of ongoing patient-provider discussions 95,96 . As many as 30-50% of patients with osteoporosis are nonadherent for a variety of reasons 35,97 . Improving adherence to osteoporosis medication requires effective communication with patients about the benefits and risks of treatment, the risks of not treating and consequences of fracture, tailoring treatment options to patient preferences, and close monitoring for early identification of nonadherence. Studies have reported that patients who are persistent with osteoporosis medications have a 30-40% greater decline in fracture rates compared with those who are not 98,99 . Practices to encourage adherence include simplification of dosing regimens, use of decision aids, changes in route of drug administration, and use of technology to send reminders for refills and/or injection/infusion dates [100][101][102] . Additional support can be provided via ongoing patient education, including open discussion about any treatment sideeffects, the chronic nature of osteoporosis, patient support programs, patient-centered counselling, and encouraging positive lifestyle changes.
A sample dialogue with a 65-year old postmenopausal woman who was previously diagnosed with osteoporosis (Tscore À2.8) and has been on therapy for 1 year is presented in Supplemental Figure 3 in Supplemental Material, Clinical Scenario 4.

394
A. J. SINGER ET AL.

Summary
Due to the growing size of the aging population globally, osteoporosis represents an increasing societal and economic burden that warrants the attention of primary care. Osteoporotic fractures are associated with high rates of disability, loss of independence, reduced quality of life for patients and caregivers, and high costs to individuals and healthcare systems. Osteoporosis is currently underdiagnosed and undertreated. As the primary contact with patients, PCPs are uniquely positioned to play a critical role in improving osteoporosis management and closing the treatment gap. We recommend that PCPs adopt a systematic approach of proactively screening patients to identify those at risk for osteoporosis and implement individualized treatment plans and follow-up to reduce the likelihood of debilitating osteoporotic fractures. This requires PCPs to gain knowledge of osteoporosis risk factors and understand the role of BMD testing and application of risk assessment tools to best treat and serve their patients. PCPs may also need to work as part of multidisciplinary teams with other medical specialties, including working groups, to offer outreach, case finding, treatment, referral, and health education. The simple strategy presented in this article can be used to guide PCPs in making decisions regarding the course of action when managing patients at risk for osteoporosis and osteoporotic fracture within the primary care setting.

Declaration of funding
Amgen Inc. provided funding for the work and for the preparation of the manuscript in the form of writing assistance. Amgen reviewed the manuscript for medical and scientific accuracy as well as intellectual property considerations.
Declaration of financial/other relationships Andrea J. Singer has received research/grant support from Radius and UCB Pharma, paid to Georgetown University and MedStar; consulting/advisory board fees from AgNovos, Amgen, Radius, and UCB Pharma; speaker's bureau fees from Amgen and Radius, and has also served on the National Osteoporosis Foundation Board of Trustees (non-remunerative). Anita Sharma has received honorarium from Amgen. Cynthia Deignan is an employee and stockholder of Amgen. Liesbeth Borgermans has received research/grant support from Amgen. Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions
The authors meet the criteria for authorship as recommended by the International Committee of Medical Journal Editors (ICMJE). All authors take responsibility for the integrity of the work as a whole, from inception to published article. All authors contributed to the study concept, analysis of data, interpretation of the results, drafting and editing of the manuscript, and approved submission of the manuscript.