Quality of life after fragility fracture in the Russian Federation: results from the Russian arm of the International Cost and Utility Related to Osteoporotic Fractures Study (ICUROS)

Osteoporosis is a skeletal disorder characterized by compromised bone strength predisposing to an increased risk of fracture. Bone strength primarily reflects the integration of bone density and bone quality [1, 2]. Osteoporosis results in fractures that impose a considerable financial burden on health services due to reduced mobility, hospitalization, and nursing home requirements [3, 4]. Worldwide, osteoporosis causes more than 8.9 million fractures annually, equating with a new osteoporotic fracture every 3 s [5]. In the Western world, 1 in 3 women over age 50 years will experience osteoporotic fractures, as will 1 in 5 men aged over 50 years [6].

Loss of quality of life (QoL) after fragility fracture is significant, although, until recently, the numbers of studies and patients included were limited and the length of follow-up relatively short [7]. To overcome these issues and to estimate the costs and quality of life related to fractures in a number of countries across the world, the International Osteoporosis Foundation initiated the International Costs and Utilities Related to Osteoporotic Fractures Study (ICUROS) in 2007. Results of the ICUROS [8, 9] showed that fragility hip, vertebral, and distal forearm fractures resulted in substantial QoL loss directly after fracture. While QoL improved with time, 18 months after fracture, mean health state utility values (HSUVs) were lower than before the fracture in patients with hip fracture (0.66 vs. 0.77 p < 0.001) and vertebral fracture (0.70 vs.0.83 p < 0.001) [8].

Relatively little is known about the consequences of osteoporosis in Russia. The hip fracture rates lie in between the high rates reported in Scandinavia and the low rates reported in Latin America, and approximate those found in Australia and the Netherlands. Similarly, hip fracture probabilities are relatively low and similar to those reported for Hungary [10, 11]. By contrast, probabilities of a major fracture are much higher due to the unexpectedly high rates of forearm and humeral fractures in Russia [12]. There are few data on the consequences of fracture on QoL in Russia. Differences in accumulated QoL losses among countries have been reported but summary data only provided. Compared with Russia, the accumulated QoL losses over 18 months were consistently higher in Italy and Lithuania and lower in Austria for all three fracture types [8]. The objectives of the present study were to estimate QoL changes for patients with hip, vertebral, distal forearm, humeral, and ankle fracture in Russia and to identify predictors of recovery at 18 month after fracture.

This study presents data on the QoL impact of hip, vertebral, proximal humerus, distal forearm, and ankle fracture up to 18 months after fracture from the Russian arm of the ICUROS study. The information obtained in current study is of particular importance since existing data on QoL related to osteoporotic fractures in Russia are scarce.

The accumulated mean QoL loss after hip fracture was 0.5 over 18 months and significantly greater than after fracture of the distal forearm (0.13), spine (0.21), proximal humerus (0.26), and ankle (0.27). The very marked utility loss after hip fracture and partial recovery is very consistent with pattern identified in systematic reviews of studies from other countries that used EQ-5D for QoL assessment [7, 17], as well as in other ICUROS countries [9, 18, 19]. Patients with distal forearm fracture on average regained pre-fracture QoL at 18 months after fracture as reported elsewhere [7, 8], whereas patients with humeral, ankle, and vertebral fracture experienced sizable decrements in HSUVs of approximately 0.10 compared with before the fracture. Substantial impairment in self-care and usual activities immediately after fracture were important predictors of recovery across at all fracture sites and the effects generally remained after controlling for age, sex, and VAS prior to fracture. For example, among patients with a proximal humeral fracture, the absolute difference in recovery rates between the 67/192 (35%) of patients who reported severe impairment in usual activities immediately after fracture and the remaining 125/192 (65%) patients were 60%.

QoL in patients with humeral fragility fractures of the proximal humerus has been less well documented than that following forearm, spine, or hip fracture [20, 21]. In the present study, the pattern of response and cumulative disutility was similar to that after vertebral fracture. Interestingly, the data from the Australian ICUROS showed that the decline in HRQoL for humeral fractures was even greater in the immediate fracture period than for vertebral fractures returned to pre-fracture levels by 18 months [18].

Information on HSUVs following ankle fractures are scarce and reported previously only in the Australian arm of ICUROS [18]. The reason is that ankle fractures are not considered to be characteristic of osteoporosis [22]. Fractures of the ankle are inconsistently associated with low BMD in elderly women [23, 24]. There is, moreover, no age-related increase in risk from the age of 50 years in men or in women [22]. It is relevant that the risk factors for ankle fractures in women after the menopause differ from those for other osteoporotic fractures. For example, high body weight, but not early menopause are risk factors for ankle fractures, whereas low body weight and early menopause are risk factors for wrist fractures [25, 26]. The pattern of change in HSUVs after ankle fracture was similar to that of spine and humerus fracture as was the mean cumulative utility loss.

Despite qualitative similarities between studies, there are a number of quantitative differences between studies and countries. The disparities in short-term QoL multipliers may reflect time to interview after first fracture: longer time to interview may result in lower initial QoL decrement. In Russia, not all hip fracture patients are surgically managed and may contribute to the very high utility loss within 2 weeks after hip fracture. Over the longer term, there are also substantial differences. For example, compared with Russia, the accumulated QoL losses over 18 months were higher in Italy and Lithuania and lower in Austria for hip, humeral, and distal forearm fracture in the ICUROS study where data collection was standardized [4]. Despite standardization, the severity of fractures may differ between the studies depending on the participating centers. Some centers may be highly specialized and therefore treat more severe fractures or more frail patients than others. There will also be a consent bias, particularly in the case of hip fracture, in that many patients who sustain a hip fracture will also have dementia and would be excluded from study. The method of recruitment is of critical importance for vertebral fracture. In the Australian arm of ICUROS, the majority of patients with vertebral fracture were recruited through an emergency department [18]. In the Russian arm, participants were recruited mainly from Neurology or Radiology Departments, several days or weeks after the fracture event. Also, HRQoL loss varies according to the number and severity of vertebral fracture [27], which was not taken into account in ICUROS. The mean age of hip fracture in the present study (69 years) was less than in the other ICUROS countries (76 years) [8]. This is likely a reflection of the higher general population mortality in Russia but limits the interpretation to a younger population than in other ICUROS countries. For all these reasons, between-center and between-country differences in HSUVs are difficult to interpret. Notwithstanding, a persistent disutility at 18 months is a consistent finding for hip [28‐30] and spine fractures [8, 31]. Conversely, there is general agreement that the disutility following fracture of the distal forearm is transient [32, 33]. In ICUROS, recovery was on average incomplete before 18 months which contrasts with a much more rapid recovery pattern, albeit on fewer patients from the UK [33].

There are several additional limitations of ICUROS study [8, 18, 34]. HSUV prior to fracture was determined by recall with a potential for bias. However, it has been shown that patients can accurately recall their QoL up to 6 weeks [35], so that substantial recall bias is unlikely. This assumption is supported by the finding in patients with distal forearm fractures that HSUV 18 months after distal forearm fracture was similar to mean HSUV prior to fracture, consistent with modest and, on average, a transient natural history [32]. The exclusion criteria (long-term care prior to the fracture, cognitive impairment, and patients sustaining a new fracture during follow-up) are likely to bias the HSUVs and account also for the low mortality in this cohort. For this reason, the recovery rates may be better than average recovery rates in the general population.

In ICUROS, approximately 21% of patients were lost to follow-up. In the present study, the dropout rate was substantially less (8.3%). Rather than lessen bias, the low dropout rate, particularly for mortality (1.7%), reinforces the notion of preferential recruitment of a healthier segment of the population.

In conclusion, this study shows that fractures of the hip, vertebral, distal forearm, ankle, and proximal humerus incur substantial loss of QoL in Russia. Although there is marked variation in QoL losses between fracture sites, and with the exception of forearm fractures, QoL is markedly impaired for at least 18 months. Furthermore, severe impairment in self-care or usual activities after fracture indicate that patients may have suboptimal long-term recovery. In such studies, these results can help inform health technology assessment and early identification of patients with poor long-term prognosis, specifically for Russia.