Trends in incidence rate, health care use, and costs due to rib fractures in the Netherlands

Thoracic wall injury after blunt chest trauma is common and rib fractures are diagnosed in 10% of patients after trauma and 30% of patients after blunt thoracic trauma [1‐3]. In younger patients, most rib fractures are caused by a high-energy trauma (HET), whereas more than half of the patients aged 65 years or older sustain rib fractures following low-energy trauma (LET) [4, 5]. Rib fractures can occur as single or multiple simple rib fractures or as a flail segment in which three or more rib fractures are fractured in two or more places [6]. Patient and injury characteristics influence the outcome after rib fractures. Increased age, increased number of rib fractures, and presence of concomitant thoracic injuries are associated with poorer outcome including higher pneumonia risk, increased hospital and Intensive Care Unit length of stay (HLOS and ICLOS, respectively), and increased mortality [5, 7‐12]. While negatively affecting in-hospital outcome, rib fractures are also associated with long-term disability, chronic pain, and reduction of quality of life [13‐16]. At 2 years post-injury, almost one-third of patients has not yet returned to their pre-injury work level [17]. While the prevalence of rib fractures is known, the incidence rate of rib fractures has only been studied in hospitalized patients or the elderly [18‐20]. Although the disabling effect of rib fractures on short- and long-term outcomes is indisputable, the economic effect on health care use, work absence, and associated detailed evaluations of direct and indirect costs has hardly been studied. Insight into the occurrence and economic impact of rib fractures can both assist in daily allocation of health care services and provide a projection for the future.

Therefore, the aim of this nationwide study was to examine population-based trends in the incidence rate of rib fractures for a 4-year period (2015–2018) and to give a detailed overview of the health care consumption and work absence with associated costs in these patients.

This epidemiologic nationwide study is the first to describe health care use and costs in admitted and non-admitted patients with rib fractures. Patients registered with rib fractures were more prevalent over time and the incidence rate increased with age. Almost two-thirds of patients with rib fractures required hospital admission with a mean stay of about 8 days with associated significant direct and indirect costs, and over 2 months of work absenteeism. The direct and indirect costs, duration of work absence, and years lived with disability remained stable during the study period and appeared unrelated to having sustained one or multiple (≥ 2) rib fractures but increased considerably with age and in admitted patients compared with non-admitted patients.

The incidence rate of 29 per 100,000 person years for admitted patients is similar to the incidence rate demonstrated by a recent study focusing only on admitted patients [18]. The incidence rate for the total cohort of 47 per 100,000 person years showed that rib fractures are common and not all patients require admission. To put things in perspective, the total incidence rate of patients with a humeral fracture is 40 per 100,000 person years [32]. The incidence rate increased sharply with age, from 34 to 107 per 100,000 person years in patients aged ≤ 65 years and ≥ 65 years, respectively. The incidence rate of 72 per 100,000 person years for admitted elderly patients is again in line with previous literature [18, 20]. Furthermore, the increase of 37% of the total number of patients with rib fractures between 2015 and 2018 demonstrates that rib fractures are a common and increasing problem, especially among the elderly [33]. Age is a known risk factor for mortality in patients with rib fractures [5, 9, 10, 12]. This highlights the need for improvement of preventive measures and a possibly more aggressive multimodal therapy in this type of patient.

While the cumulative HLOS of 48,737 days is not as high as for, e.g., patients with foot and ankle injuries (58,708 days), the mean HLOS of almost 8 days is higher than for patients with a humeral or tibia shaft fracture in the Netherlands [24, 32, 34]. The mean HLOS increased with severity of the injury (single or multiple rib fractures, or a flail chest), but without data on concomitant injuries, need for additional interventions (e.g., chest tube, surgical stabilization of rib fractures (SSRF), or video-assisted thoracoscopic surgery (VATS), or complications such as pneumonia, the true impact on the HLOS attributable to sustaining rib fractures, is not known. Nonetheless, it has been shown that almost half of the patients with rib fractures are polytraumatized (i.e., ISS > 15) patients and require Intensive Care Unit admission, demonstrating that rib fractures are a marker of severe injury [35].

The annual cumulative direct health costs were €54.5 million of which 90% was accounted for by admitted patients. Similar age-dependent mean HLOS for patients with one or multiple rib fractures suggests that the high direct health care costs might mostly be accounted for by out of hospital care. The age-dependent distribution of direct health costs showed a sharp increase after the age of 45 years. Thus, efforts to improve the preventive measures and both in and out of hospital care should not only focus on the elderly but possibly on patients as young as 45 years [11]. Again, as there was no insight into treatment parameters such as need for medicinal, radiologic, or operative interventions, the exact impact of rib fracture injuries on health care costs could not be distilled.

With a mean duration of work absence of 44 days and cumulative indirect costs for lost productivity of €49.5 million, rib fractures pose a serious societal health burden. While many possible confounding variables are not known for this outcome, these results are reinforced by previous studies indicating long-term disability and long work absenteeism after rib fractures [15‐17, 36]. The mean direct health care costs for a patient with rib fractures were €6785 which is higher than that of patients with ankle or foot injuries (€3461), but lower than for patients with a humeral fracture or a hip fracture (€8864 and €10,458, respectively) [32, 34, 37]. Comparing these mean costs per patient with previous literature is difficult as most studies have focused solely on in-hospital costs of patients with a flail chest, injury severity score (ISS) of ≥ 16 and compared operatively and nonoperatively treated patients [22, 38‐40]. With no information in this study on the received treatment modality, ISS, ICU admission, or costs in patients with flail chest, paralleling these results is not feasible. To our knowledge, the duration of work absenteeism and consequent costs for lost productivity in patients with rib fractures have not been studied before. However, rates of 33% to 42% of patients have been reported to be unable to work at their pre-injury capacity at three and even 12 months post-injury, respectively [41, 42]. These associated high costs with lengthy HLOS and work absenteeism, which increase with age, might consequently indicate the need for a different approach to the patient with rib fractures. Currently, most patients with rib fractures are treated nonoperatively, and SSRF is intended for the younger patient with a high ISS [42, 43]. Based on the age-dependent increase in HLOS and costs in admitted patients, the benefit of restoring the chest wall biomechanics through SSRF might actually be higher in the older population in which some physiological decompensation is present [44].

While this epidemiologic study is the first to evaluate health care use and costs in patients with rib fractures in the Netherlands, interpretation of the results should be done in the light of several limitations. First, miscoding and incomplete data are inherent to using nationwide or large registries such as the DISS and HDR. As rib fractures are often accompanied by other severe injuries, rib fractures might not have been registered, resulting in an underestimation of the incidence rate of rib fractures. Currently, different national registries are used for registration of trauma data and consequent research. The DISS database only records data from Dutch emergency departments, the HDR and the Dutch National Trauma Registry from admitted patients. The different registries provide complementary useful data, but currently cannot be linked. Thus, to provide adequate and complete data on for example the incidence rate, health care use, and costs, one national registry which includes complete short- and long-term data on both admitted and non-admitted patients is urgently needed.

Second, patients from the DISS were registered as having sustained one or multiple rib fractures where multiple was defined as 2 or more rib fractures. In current literature, multiple rib fractures are defined as having three or more rib fractures, regardless of side or adjacency [42, 45, 46]. Only for the HLOS, which was covered by the HDR, a third subgroup, patients with a flail chest, was available. This discrepancy in defining subgroups of patients with rib fractures hinders generalizability and providing practicable conclusions. In addition, rib fractures were diagnosed by either chest CT or radiograph, or both. The use of chest CT is more sensitive in the delineation of rib fractures as it finds two to three additional rib fractures compared with radiograph [47, 48]. Therefore, patients who were registered as having sustained one rib fracture, diagnosed through chest radiograph, might have been registered as a patient with multiple rib fractures if a chest CT had been made. In this light, this might have introduced selection bias in diagnosing a patient with or without rib fractures, but also in the number of rib fractures. The diagnosis of these additional rib fractures on chest CT might result in increased admission rates and subsequently increased total health care costs, but the effect of the diagnostic modality on in-hospital outcome such as complication rate and mortality remains unclear [47, 49, 50]. This limitation does, however, reflect daily practice as a large number of hospitals and EDs still use chest radiograph as the primary diagnostic modality in patients with rib fractures.

Third, the distinction between patients with rib fractures as primary diagnosis or secondary diagnosis was not possible with the available data. Thus, patients were admitted with rib fractures and not because of rib fractures. This distinction in rib fracture as primary or secondary diagnosis could have helped in providing insight in the individual impact of rib fractures on the outcome measures. Another covariate which was not available, was treatment modality. Outcomes such as costs for health care, lost productivity, and YLD might be influenced by whether a patient underwent SSRF or nonoperative treatment. To date, studies have shown significant in-hospital differences in health care costs between treatment modalities in patients with rib fractures, but data on the effect of treatment modality on outcomes such as lost productivity or YLD remains limited [22, 39, 40]. Also, comorbidity, concomitant injury characteristics, trauma mechanism, the ISS, and abbreviated injury scale scores were not known for these patients. Therefore, the outcome measures could not be corrected for possible confounding variables. This might explain the relatively similar direct and indirect costs as well as years lived with disability in patients with either one or multiple rib fractures with admitted patients contributing most to these outcomes. While this stresses the need for improved nationwide registries, these results show that sustaining even a single rib fracture is a marker of severe injury as it might be accompanied by a long duration of health care use, work absence, and high costs.

In conclusion, this epidemiologic study shows that the number of patients registered with rib fractures has been increasing over time and the incidence rate increases with age. Although it was not possible with the current available data to prove causality between outcomes and rib fractures specifically, it does demonstrate that sustaining rib fractures indicates severe injury and may be associated with lengthy HLOS and work absenteeism as well as high direct and indirect costs. The duration of work absence and associated costs increase with age and are considerably higher in admitted patients than in non-admitted patients. The outcomes appear to be similar in patients with one or multiple rib fractures. However, due to the non-standard definitions of rib fractures, the lack of additional individual patient data, and impossibility to combine data of the different national registries, the generalizability and results of this study should be interpreted with caution.