Educational information to enhance physician–patient communication can take the form of graphs, brochures, videos, models, and other types of handouts, collectively called “decision aids.” These can reinforce and expand on what was said during an office or hospital visit, potentially facilitating clinical decisions. Decision aids can be helpful with “close call” decisions, where the balance of treatment benefits and risks is uncertain or the clinical circumstances are complex.

The effectiveness of decision aids were evaluated in a meta-analysis of 55 randomized controlled clinical trials [40]. Decision aids performed better than usual care in providing patients with a greater understanding of the treatment options. There were fewer decisional conflicts due to feeling uninformed, with more patients taking an active role in decision-making and fewer patients remaining undecided about treatment. Decision aids using risk probabilities resulted in a greater proportion of patients having accurate risk perceptions. The use of decision aids did not alter patient satisfaction with decision-making, anxiety, or health outcomes. The authors of the meta-analysis concluded that decision aids increased patient involvement in decision-making, leading to informed values-based decisions. Decision aids seemed to be most useful when more than one reasonable treatment option was available, with no clear advantage of one over the other, and each having potential benefits and harms.

In order for decision aids to be effective, the information provided must be accurate and unbiased. The quality of patient education material can be variable and is sometimes incorrect or biased. The quality of 165 printed consumer brochures about osteoporosis was evaluated according to criteria addressing evidenced-based content, risk communication, transparency of the development process, layout, and design [41]. The authors concluded that quality was “utterly inadequate,” with failure to provide evidence-based data on diagnosis and treatment, regardless of the source of the brochure. In another study evaluating osteoporosis websites, a wide range of quality was observed [42]. Overall quality scores were significantly lower for websites with a uniform resource locator (URL) suffix of “.com” compared to those with “.gov,” “.edu,” and “.org.”

Many decision aids present complex numerical information in a graphic format. The design features of graphs and the data scale may influence their effectiveness. A systematic review assessed the findings in 24 studies of graphs depicting probabilities, frequencies, or chances of occurrence of health-related events [43]. The best design for a graph depended on the purpose of the risk communication (e.g., understanding risk vs. changing behavior) and the demographics of the recipients (e.g., educational level, literacy). For expressing probability data, “part-to-whole icon array graphs” appeared to be more useful than providing percentages or proportions. This type of graph displays icons (symbols or figures) to illustrate the population at risk and highlighted icons showing those experiencing an event of some kind. Bar graphs are often perceived as analytical and difficult to understand [44]. Other ways of depicting risk include risk tables, ladders, scales, and survival and mortality curves. Each of these may be effective in portraying risk, depending on the type of risk and the target population.

A multicenter randomized controlled trial in primary care practices evaluated the effectiveness of a decision aid to improve osteoporosis treatment decisions [45]. The intervention in this study was an icon array graph displaying each patient’s FRAX 10-year probability of major osteoporotic fracture with and without bisphosphonate treatment , as well as listing possible side effects of treatment and out-of-pocket costs. Patients randomized to the control group received usual care and received an osteoporosis patient education brochure produced by the US National Osteoporosis Foundation. Cognitive, behavioral, and affective endpoints were assessed, with compliance and persistence measured at 6 months. It was concluded that the decision aid improved the quality of clinical decisions about bisphosphonate therapy by improving knowledge transfer and patient involvement. The decision aid did not alter medication start rates but may have improved adherence to therapy.

Methods for making treatment decisions have been classified in three forms [46, 47]. The first is “paternalism ,” where the physician has all the relevant information and is the sole decision-maker. The second is “independent choice,” where the physician provides the patient with relevant information and the patient makes all decisions. The third is “shared treatment decision-making,” where the physician and patient share information, discuss treatment options, and reach a collaborative decision . With shared treatment decision-making, it is appropriate and usually expected that the physician offers a recommendation. The patient is encouraged to respond to the recommendation. The physician should be receptive to verbal and nonverbal responses. A patient who expresses disagreement or discomfort with the recommendation is unlikely to be adherent to therapy, even if it is not rejected outright. The physician must be willing to offer an alternative recommendation if the first one is not acceptable and should be willing, as well, to consider the patient’s proposal for a treatment plan. Shared treatment decision-making is often a negotiation between the physician and the patient, with the goal of developing a plan of action that is medically reasonable for the physician and acceptable for the patient.

Shared treatment decision-making has been shown to improve outcomes with some medical conditions [48, 49] and may be helpful in the care of osteoporosis [50]. A Cochrane review evaluated 43 randomized studies involving “patient-centered care” [51], which includes studies using models of shared decision-making. It was found that healthcare providers could be successfully trained to improve their ability to share control with patients about topics and decisions addressed during a consultation. The results were mixed on whether patients were more satisfied when providers applied these skills. Beneficial effects on health behavior and health status were seen with interventions that combined provider training with decision aids, although the conclusions were tentative due to the heterogeneity of outcomes in the studies.

CPGs can be defined as “systematically developed statements to assist practitioner and patient decisions about appropriate healthcare for specific clinical circumstances” [52]. The purpose of CPGs is “to make explicit recommendations with a definite intent to influence what clinicians do” [53]. CPGs are commonly developed by a group of experts after evaluation of the best available medical evidence, often with consideration of healthcare policy and costs [54] using systematic approaches to achieve consensus [55]. Many professional societies and organizations have released CPGs for the treatment of osteoporosis, with updates subsequently needed to include new data, advances in diagnostic tools, and newly available treatments. These CPGs may improve health outcomes (i.e., fewer fractures in patients with osteoporosis) by advising physicians on when to measure BMD, how to assess fracture risk, what tests to order in evaluating for secondary causes of osteoporosis, when to start pharmacological therapy to reduce fracture risk, and sometimes what drug or drug class to consider for initiating therapy.

Despite the obvious benefits of CPGs, especially for practitioners who are unfamiliar with management of osteoporosis, there are limitations as well [56]. CPGs and the evidence from which they are derived are helpful for never sufficient for making clinical decisions with individual patients [57]. Data on efficacy and safety of osteoporosis treatments, for example, are applicable to groups of patients with specific demographics and allowable comorbidities, given the limitations of the study design and duration. Every patient seen in clinical practice is unique, each having his or her own cultural beliefs, biases, experiences, confounding health issues, concomitant medications, and concerns that must be considered alongside the CPGs and medical evidence.

Despite abundant evidence in clinical trials for efficacy and safety of medications approved for the prevention and treatment of osteoporosis, it remains uncertain whether each individual patient treated with one of these drugs in the clinical practice setting will achieve that same level of benefit and whether the balance of benefit and risk with long-term therapy is the same as reported for the relatively short duration of registration trials. For these reasons, patients treated for osteoporosis are monitored to provide some evidence that the treatment is effective and that adverse effects of therapy have not developed. Monitoring therapy and regular contact with a healthcare professional provide opportunities to assess the balance of benefit and risk, reassure the patient that taking the medication is worth the bother and the cost, and may improve adherence to therapy [58].

The most common measurements used to monitor therapy are BMD testing by DXA and bone turnover markers (BTMs). Stability or an increase in BMD, a decrease in BTMs with antiresorptive therapy, or an increase in BTMs with osteoanabolic therapy is generally considered to represent a favorable response to the therapy [59], with the caveat that the tests are conducted at facilities that follow quality standards and that the LSC is known.

Pharmacological therapy for osteoporosis must be taken correctly and for a sufficient length of time for patients to achieve the expected reduction in fracture risk. There are many studies showing that adherence and persistence with osteoporosis is suboptimal, with many or most patients discontinuing treatment within 1 year after a prescription is written [60] and poor adherence being associated with poor clinical outcomes [61]. A recent systematic review evaluated 20 studies of interventions intended to improve adherence and persistence in adult users of osteoporosis medications [62]. It was concluded that simplification of dosing regimens, electronic prescriptions, decision aids, and patient education may improve adherence and persistence, noting many limitations of the studies. There was wide variation in the quality of the studies, with differences in study design, inconsistent definitions for measurement of adherence and persistence, limited reporting of relevant information reported, lack of data on clinical outcomes, and short duration of follow-up. More vigorous investigation of likely interventions in large randomized controlled trials was recommended.

Until more definitive data are available, it is prudent for healthcare practitioners to customize approaches to improving adherence and persistence with consideration and understanding of issues of greatest importance to each individual patient. Potentially useful clinical strategies include patient education, effective risk communication, shared decision-making, monitoring therapy , and periodic reevaluation of the balance of benefit and risk with treatment.

A good response to therapy (stability or an increase in BMD or an appropriate change in BTMs) does not necessarily represent achievement of an acceptable level of fracture risk. For example, in a patient with a very high pretreatment fracture risk, the expected fracture risk reduction with the medication chosen for initial therapy may still leave the patient with a higher than desirable risk. Recognition of this concept has led to consideration of developing a treat-to-target strategy for osteoporosis [63, 64], as has been effectively used for other chronic asymptomatic disorders, such as hypertension and diabetes mellitus. An osteoporosis treatment target (e.g., a T-score or BTM value) might help physicians in the selection of an initial agent that is most likely to reach that target. If the treatment target is reached with bisphosphonate therapy, then a drug holiday may be considered. If there is a failure to reach the treatment target with initial therapy, a change in treatment might be indicated. At this time, treat-to-target for osteoporosis is being investigated, with no consensus and no guidelines on the use of treatment targets in clinical practice.

Few medical issues in the field of skeletal health have generated as much controversy, discussion, and confusion as the duration of treatment of osteoporosis. The question of “how long to treat?” is not often raised in association with other chronic asymptomatic disorders , such as hypertension and hypercholesterolemia. However, with osteoporosis, rare but disturbing bone-related occurrences (e.g., osteonecrosis of the jaw, atypical femur fractures) associated with long-term therapy have received a great deal of attention in scientific journals and news media, resulting in some patients stopping medication of their own accord and some being told to stop by their physicians. The concept of bisphosphonate holidays has emerged because this class of drugs has a long skeletal half-life, with discontinuation after years of therapy followed by persistence of antiresorptive effect for an undetermined period of time. The anti-fracture benefit may also persist, at least in low-risk patients, for a period of time, while the risk of some adverse events associated with long-term therapy, such as AFF, may rapidly diminish.