EUnetHTA published assessment of screening for osteoporosis in the general population

In mid-September 2019, the report of the project OTCA19, titled “Screening for osteoporosis in the general population” was published on the website of the European Network for HTA (EUnetHTA).

The project was executed by the independent Institute for Quality and Efficiency in Health Care (IQWiG), Germany and Swiss Network for HTA (SNHTA), Switzerland.

The dedicated reviewers come from the Agency for Health Quality and Assessment of Catalonia (AQuAS), Spain, Austrian Public Health Institute (GÖG) and National School of Public Health, Management and Professional Development (NSPHMPD), Romania.

Patients with osteoporosis-related fracture risk can be identified either via organized screening or opportunistically by case finding. However, there is currently no agreed policy for osteoporosis screening in Europe. In the context of this assessment, population-based screening using a clinical risk assessment tool and/or bone mineral density (BMD) measurement (dual-energy x-ray absorptiometry [DXA], quantitative ultrasound [qUS], or quantitative computer-tomography [qCT]) was evaluated and then compared with usual care, i.e. no screening. Various tools for assessing the risk of fracture or osteoporosis have been developed, as BMD measurement alone fails to identify many patients at risk of fracture. These tools are used to make decisions regarding further diagnostic work-up of patients and/or to directly make a decision regarding their eligibility for treatment. Risk factors include sex, age, body mass index (BMI), and lifestyle. The aim of screening and the subsequent changes in drug and/or non-drug management is to avoid the significant morbidity and mortality associated with osteoporotic fractures.

In order to identify relevant primary studies a 2-step approach was performed: in a first step, comprehensive, high-quality and up-to-date systematic reviews (SRs) / health technology assessments (HTAs) were searched for in a focused search covering the period from 2013 onwards in MEDLINE, the Cochrane Database of Systematic Reviews, and the Health Technology Database. The websites of HTA agencies were also searched (National Institute for Health and Care Excellence [NICE]), Agency for Healthcare Research and Quality (AHRQ). The AHRQ report on screening for osteoporosis fulfilled the eligibility criteria and was assessed as comprehensive; it was therefore included for the purpose of identifying primary studies. In a second step, an updated search for primary studies was conducted in MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials for the period not covered by the AHRQ report.

Results - clinical effectiveness:

• Mortality – results were available from the SCOOP trial. There was no statistically significant difference between the 2 study arms with respect to this outcome. Mortality was also reported in 4 enrichment design RCTs. No statistically significant difference could be observed between the 2 study arms in either (sub-) study (HR: 1.05, [0.93; 1.19])
• Symptomatic fractures - the evidence on this outcome was based on data from all 3 strategy design RCTs included (SCOOP, ROSE, and COSHIBA). For all fracture (sub-) outcomes investigated, no advantage could be derived in favor of the intervention. Regarding hip fractures, the evidence was inconclusive. Additional evidence primarily came from the 2 sub-studies of the FIT trial (Clinical Fracture study, Cummings 1998, Vertebral Deformity study, Black 1996) and from Reid 2018. However, the evidence was too weak to influence the findings of the strategy design studies (HR: 0.94, [0.86; 1.03]; OR: 0.6, [0.35; 1.03])
• Back pain - No data from studies with marker-based strategy design were available. In the Vertebral Deformity study of the FIT trial (women with pre-existing vertebral fractures), the outcome was assessed using several different operationalizations. Both the reduction in the number of bed-rest days due to back pain and the reduction in the number of limited activity days due to back pain were statistically significant in favor of the intervention group (RR: 0.44, [0.30; 0.64] and RR: 0.85, [0.63; 1.15] correspondingly). However, no statistically significant effects were shown for other operationalizations
• Health-related quality of life (HrQoL) - Data on HrQoL were available from the SCOOP trial. The HrQoL was measured by means of the EQ-5D (p = 0.154) and SF-12 for mental health (p = 0.554) as well as the SF12 for physical health (p = 0.237) after 60 months of follow-up. For all 3 HrQoL instruments, no statistically significant difference between the 2 study arms was found
• Utilization of healthcare resources -no data from studies with marker-based strategy design was available for the outcome "utilization of health care resources." Data from enrichment design studies were additionally consulted. In the Vertebral Deformity study of the FIT trial (women with pre-existing vertebral fractures), the outcome was assessed using different operationalizations. A limitation of these operationalizations was that only fracture-related utilization of healthcare resources was reported. While there was no statistically significant difference for overall hospital stays or stays in nursing homes / rehab, the authors found a statistically significant reduction in favor of the intervention group in the fracture-related use of all resources and in the number of emergency room visits (OR: 0.73, [0.54; 0.98] and OR: 0.70, [0.52; 0.96] correspondingly)

Results – safety:

• Data on (serious) adverse events were not reported adequately in the strategy-design RCTs. Therefore, it was only possible to extract (serious) adverse events from studies with enrichment design. Both the 2 sub-studies of the FIT trial (Vertebral Deformity study, Clinical Fracture study), as well as Reid 2018 and Liang 2017, reported (serious) adverse events. However, 2 different drugs were examined in different dosage forms and with different durations of use. In addition, the quality of the evidence differed in these studies
• Data from Reid 2018 indicate that screening for osteoporosis in postmenopausal women may result in little or no difference in the rate of serious adverse events if zoledronic acid (administered intravenously in 18-month intervals) forms part of the screening-treatment strategy within the follow-up of 6 years (OR: 0.84, [0.70;1.00])
• Data on short-term adverse events were reported by Liang 2017. However, these were only assessed within a timeframe of 3 days after drug administration and only if they occurred in more than 5% of the participants. Although the observed effects were very large (OR: 0.729, [1.67; 31.38] for headache (influenza-like illness), OR: 11.77, [3.45; 39.19] for pyrexia, OR: 6.39, [2.19; 18.66] for myalgia, OR: 1.79, [0.85; 3.74] for arthralgia), the risk of bias was high and the ITT principle was not adequately implemented, which is why the extent of short-term adverse events is uncertain if zoledronic acid forms part of the screening-treatment strategy

Results – overdiagnosis:

• An estimation of the frequency or proportion of overdiagnoses caused by osteoporosis screening was not possible, as no data on this outcome were identified

Results – reimbursement:

• Screening for osteoporosis is generally not reimbursed in Europe. In some EU countries, opportunistic screening is offered to identify people at a high risk of fractures. However, Poland is currently planning a national screening program for osteoporosis, which is due to be launched in 2023

Conclusions:

• Screening for osteoporosis results in little or no difference in mortality, the incidence of symptomatic fractures compared with no screening. The baseline variables investigated (age, anthropometry, BMI, dietary calcium intake, baseline fracture status, recent falls history, BMD, and calculated fracture risk) may have little or no influence on the effectiveness of screening. This conclusion is based on low-quality evidence. It is uncertain whether screening for osteoporosis improves back pain and/or lowers fracture-related utilization of healthcare resources
• Since the available studies of moderate quality show no effect of screening on the incidence of symptomatic fractures, screening for osteoporosis in postmenopausal women probably has little or no benefit. These findings are mainly based on studies investigating a screening strategy using FRAX for risk assessment and DXA for BMD measurement. The studies included did not allow the evaluation of screening strategies based on other screening tools. As in any screening intervention, benefits and harms are affected by multiple factors such as the type and uptake of screening and treatment. No studies were found on osteoporosis screening in men or younger women

The full report in English can be found here.

Subscribe to our newsletter delivered every second week not to miss important reimbursement information.