Fractures are frequent events; in 2017, a total of 143 350 osteoporotic femoral fractures occurred in Germany (Möckel et al, 2020). In the same year in the UK, approximately 520 000 osteoporotic fractures were diagnosed, of which 19% were hip fractures (International Osteoporosis Foundation (IOF), 2018a).
Femoral neck (hip) fractures place a major burden on patients, relatives and the healthcare system. One year after a hip fracture, the loss in patients' quality of life is comparable to that after an amputation in patients with diabetes (Vokó et al, 2017). In addition, 1-year mortality after first hip fracture is 14.1% (Lee et al, 2016) and up to 370 hours of care per 1000 persons with osteoporotic hip fracture have to be provided by relatives (IOF, 2018b). The annual costs related to hip fracture are estimated to be almost €6 billion (£5.2 billion) in Germany and more than £2.5 billion in the UK (IOF, 2018a; 2018b).
However, to the best of the authors' knowledge, little is known about the acute medical condition of patients immediately after femoral neck fracture. The majority of the literature focuses on the long-term consequences.
Paramedics are usually the first health professionals to examine emergency patients. Therefore, their examination and the information they report on patients' medical condition take place at the earliest possible time point after fracture.
In the German federal state of Hesse, paramedics use the 3-digit response indication number (RMI; Rückmeldeindikation) 273 (RMI273) to register patients with a diagnosis of suspected femoral neck fracture (Wetterau Ambulance Service, 2018).
In addition to the RMI273, three more digits are transferred to the control room and/or admitting hospital, showing patient age and hospital admission priority. The patient's admission priority is classified from low to high according to: no transport to hospital; outpatient care (<24 hours in hospital); inpatient care (>24 hours in hospital); and immediate intervention needed (e.g. resuscitation) (Wetterau Ambulance Service, 2018).
Simultaneously, the medical condition of patients with RMI273 can be transmitted using the six-digit response code (RMC; Rückmeldecode). Each digit within this code describes patients' characteristics, with the first digit showing consciousness, the second breathing, the third the circulatory system, the fourth injuries, the fifth neurology and the sixth pain. For each medical characteristic, the severity is transmitted using a scale from 1, which represents inconspicuous/none, to 5, representing the most severe function impairment (Wetterau Ambulance Service, 2013). More details on the RMC system are shown in Table 1.
| Number | Consciousness | Breathing | Circulatory system | Injuries | Neurology | Pain** |
|---|---|---|---|---|---|---|
| 1 | Inconspicuous | Inconspicuous | Inconspicuous | None | Inconspicuous | None: VAS/NRS: 0 |
| 2 | Somnolent: Slowed or disoriented, but can be woken up | Slightly dysfunctional: Condition as stated under no 4, but without acute threatening impression SpO2>90% | Slightly dysfunctional: |
Mild: Injury, outpatient treatment (case of doubt, digit 4) | Known dysfunction: Paralysis, speech/perception/orientation disorder independent from current incident | Mild: VAS/NRS:1–3 |
| 3 | Impending severe dysfunction: Emergency situation may lead to disturbed consciousness, e.g.: |
Impending severe dysfunction: (still) stable breathing, but e.g. |
Impending severe dysfunction: (still) stable circulation, but e.g. |
Severe injury conceivable: No signs of injury, severe injury due to accident conceivable, e.g. |
Impending acute dysfunction: Emergency situation may lead to severe dysfunction, e.g. |
Moderate: VAS/NRS: 4–6 |
| 4 | Reduced consciousness with reaction to pain: Comatose grade I–III | Severely dysfunctional: e.g. dyspnoea, bradypnoea, cyanosis, pneumothorax SpO2≤90% | Severely dysfunctional: |
Severe: Injury, inpatient treatment needed | Acute dysfunction: Paralysis, speech/perceptual/orientation disorder (mild or severe) resulting from the current incident | Severe: VAS/NRS: 7–9 |
| 5 | Reduced consciousness without pain reaction: Comatose grade IV | Respiratory arrest | Without pulse | Polytrauma | Progressive dysfunction: Increase of symptoms in course of time | Extreme: VAS/NRS 10 |
min: minute; BP: blood pressure; HR: heart rate; SpO2: blood oxygen saturation levels
The full 12-digit coding is used to allocate every patient to an appropriate hospital and to inform the hospital about the patient's arrival and their general condition. Only in special circumstances (e.g. in cases of resuscitation, stroke and ST elevation myocardial infarction) is an additional pre-alert phone call needed.
Since most studies report only on the long-term effects of femoral neck fractures and RMIs/RMCs describe the medical condition of patients at first possible contact with a health professional, the study had the following objectives:
Methods
Study design and data sources
This study was a retrospective analysis of RMI273 cases and corresponding RMCs of five counties and cities (Figure 1), representing urban and rural areas in the federal state of Hesse. The data were obtained from the five control rooms covering the counties and cities over 1 January 2017 to 31 December 2019. The control centres received the RMI273 codes including patients' ages and RMCs directly from the paramedics who provided the first care to patients suspected of having a femoral neck fracture.
To be able to calculate fracture rates by age group, the number of inhabitants in the five counties/cities in 2017–2019 were taken from reports by the Hessian State Statistics Office (2018; 2019; 2020).
Ethics and privacy
This article involved a retrospective analysis of anonymised data from emergency medical services control rooms, which made the identification of patients impossible. The data were handled in accordance with the data protection legislation.
Statistical analysis
The age of patients is presented as mean and standard deviation (SD). The authors recorded the number of patients record in each of the following age groups: <50 years; 50–64 years; 65–79 years; and ≥80 years. Based on the RMI273 cases, the age-specific incidence rates per 100 000 inhabitants and the corresponding 95% confidence intervals (CI) were calculated for each calendar year by age group. For hospital admission priority frequencies, corresponding 95% CIs were calculated.
The Spearman correlation (correlation factor rho) was used to identify relationships between different medical conditions, age and admission priority. A correlation was considered meaningful and significant if rho was ≥0.10 with a P≤0.05. Effect sizes for correlations are based on Cohen (1992), indicating a small (rho≥0.10), medium (rho≥0.30) or large effect size (rho≥0.50). Statistical analysis was performed using the JASP software package.
Results
Incidence of femoral neck fractures
Between 1 January 2017 and 31 December 2019, a total of 5954 suspected diagnoses of femoral neck fractures based on RMI273 were reported for the included counties/cities (Table 2). The mean patient age was 81.76 (SD 10.87) years and comparable over the 3 years: 2017: 81.76 (SD 11.27); 2018: 81.57 (SD 10.56); and 2019: 81.95 (SD 10.77).
| 2017 | 2018 | 2019 | |
|---|---|---|---|
| Number of RMI273 codes | 2004 | 1953 | 1997 |
| Mean age (SD) in years | 81.76 (11.27) | 81.57 (10.56) | 81.95 (10.77) |
| Patients per age group | |||
| <50 years | 29 (1.45%) | 28 (1.43%) | 20 (1.00%) |
| 50–64 years | 99 (4.94%) | 101 (5.17%) | 115 (5.76%) |
| 65–79 years | 530 (26.45%) | 532 (27.24%) | 514 (25.74%) |
| ≥80 years | 1346 (67.17%) | 1292 (66.16%) | 1348 (67.50%) |
| Patients with any RMC | 2000 (99.8%) | 1607 (82.3%) | 1094 (54.8%) |
The largest proportion (two-thirds) of patients with RMI273 were aged ≥80 years (2017: 67.17%; 2018: 66.16%; 2019: 67.50%). This was followed by patients aged 65–79 years (2017: 26.45%; 2018: 27.24%; 2019: 25.74%).
Even though all patients should receive an RMC code that describes their medical condition, only 82.3% in 2018 and 54.8% in 2019 received any condition-describing code (Table 2).
The incidence of femoral neck fractures per 100 000 inhabitants was comparable over the three years in each age group (Figure 2). The highest incidence was in those aged ≥80 years, ranging from 1678.31 (95% CI 1589.90–1770.35) per 100 000 inhabitants in 2019 to 1837.21 (95% CI 1740.63–1938.34) in 2017. The 65–79 years age group had incidence rates between 287.05 (95% CI 262.77–312.97) in 2019 and 295.46 (95% CI 270.89–321.67) per 100 000 inhabitants in 2018. Incidence rates per 100 000 inhabitants in age groups below 50 years and 50–64 years were considered low at 2.77 (95% CI 1.69–4.27) in 2019 to 3.98 (95% CI 2.66–5.71) in 2017) and 35.72 (95% CI 29.04–43.49) (2017) to 40.32 (95% CI 33.29; 48.40) in 2019 respectively.
In all 3 years, most patients were assigned a hospital admission priority for inpatient care (94.57% (CI (90.30%–98.98%)) to 95.49% (95% CI (91.20%–99.83%)) (Figure 3). Only a small number of patients were admitted to hospital with a priority of immediate intervention (0.25% (95% CI 0.08%–0.59%) to 0.60% (95% CI 0.31%–1.05%)). A total of 3.40% (95% CI 2.64%–4.30%) to 4.41% (95% CI 3.52%; 5.44%)) were handled in outpatient care settings and an additional (0.45% (85% CI 0.21%–0.86%) to 0.60% (95% CI 0.31%–1.05%)) were not transported to any health facility.
In terms of medical status regarding consciousness, breathing, the circulatory system and neurology, most patients indicated not a slight dysfunction (≥95%). The state of injuries and pain indicated a more heterogenous distribution with more moderate and severe cases (≥56%) compared to the above mentioned organ systems. Details on the percentage of patients by medical condition is shown in Table 3.
| Medical condition | <50 years (n=65) | 50–64 years (n=264) | 65–79 years (n=1258) | ≥80 years (n=3114) |
|---|---|---|---|---|
| Consciousness | ||||
| Inconspicuous | 98.46% | 96.97% | 97.53% | 96.76% |
| Somnolent | 1.54% | 3.03% | 2.07% | 2.99% |
| Impending severe dysfunction | 0.0% | 0.0% | 0.24% | 0.19% |
| Comatose grade I–III | 0.0% | 0.0% | 0.16% | 0.06% |
| Comatose grade IV* | - | - | - | - |
| Breathing | ||||
| Inconspicuous | 98.46% | 93.56% | 93.80% | 93.42% |
| Slightly dysfunctional | 1.54% | 5.68% | 5.72% | 6.26% |
| Impending severe dysfunction | 0.0% | 0.0% | 0.32% | 0.26% |
| Severely dysfunctional | 0.0% | 0.76% | 0.16% | 0.06% |
| Respiratory arrest* | - | - | - | - |
| Circulatory system | ||||
| Inconspicuous | 70.77% | 59.09% | 50.00% | 48.39% |
| Slightly dysfunctional | 26.15% | 36.74% | 45.79% | 46.43% |
| Impending severe dysfunction | 3.08% | 4.17% | 3.34% | 3.73 |
| Severely dysfunctional | 0.0% | 0.0% | 0.87% | 1.45% |
| Without pulse* | - | - | - | - |
| Injuries | ||||
| None | 7.69% | 4.92% | 5.09% | 4.08% |
| Mild | 21.54% | 13.26% | 11.53% | 12.72% |
| Severe injury conceivable | 27.69% | 35.99% | 39.66% | 40.75% |
| Severe | 43.08% | 45.83% | 43.72% | 42.39% |
| Polytrauma | 0.0% | 0.0% | 0.0% | 0.06% |
| Neurology | ||||
| Inconspicuous | 89.23% | 85.60% | 82.19% | 75.00% |
| Known dysfunction | 9.23% | 12.88% | 16.22% | 24.02% |
| Impending acute dysfunction | 0.0% | 0.76% | 0.80% | 0.67% |
| Acute dysfunction | 1.54% | 0.76% | 0.72% | 0.32% |
| Progressive dysfunction | 0.0% | 0.0% | 0.08% | 0.0% |
| Pain** | ||||
| None (VAS/NRS 0) | 7.81% | 2.65% | 2.71% | 2.32% |
| Mild (VAS/NRS 1–3) | 28.12% | 32.96% | 35.99% | 40.78% |
| Moderate (VAS/NRS 4–6) | 42.19% | 46.59% | 44.66% | 43.49% |
| Severe (VAS/NRS 7–9) | 18.75% | 17.80% | 15.84% | 12.80% |
| Extreme (VAS/NRS 10) | 3.13% | 0.0% | 0.80% | 0.61% |
Correlations between patient medical conditions, age and admission priority
The age of patients with femoral neck fracture was significantly correlated with neurological condition (rho 0.10 (95% CI (0.07–0.13); P≤0.001; n=4701), indicating a rise in neurological impairment with increasing age.
Hospital admission priority was significantly correlated with circulatory system status (rho 0.11 (95% CI (0.08–0.13); P≤0.001; n=4,698), severity of injuries (rho 0.20 (95% CI (0.18–0.23); P≤0.001; n=4,698) and pain (rho 0.15 (95% CI (0.13–0.18); P≤0.001; n=4690). This indicates that patients with a more impaired circulatory system, greater injury and/or pain were more likely to receive a higher hospital admission priority and therefore a higher likelihood of being admitted to hospital for immediate intervention needed. In contrast, patients with less impairment of the circulatory system or severe injuries and/or pain were more likely to be assigned a priority of outpatient care or no transport.
The status of the circulatory system after femoral neck fracture was also significantly correlated (Table 4) with breathing (rho 0.12 (95% CI (0.09–0.14); P≤0.001; n=4701), injuries (rho 0.10 (95% CI (0.06–0.13); P≤0.001; n=4701) and pain (rho 0.13 (95% CI (0.11–0.16); P≤0.001; n=4693). These data indicate that more severe injuries, more significant breathing problems or more severe pain are correlated with greater circulatory system dysfunction.
| Correlating factors | Spearman rho (95% CI) | P value |
|---|---|---|
| Age – neurology (n=4701) | 0.10 (0.07–0.13) | P≤0.001 |
| Admission priority – circulatory system (n=4698) | 0.11 (0.08–0.13) | P≤0.001 |
| Admission priority – injury (n=4698) | 0.20 (0.18–0.23) | P≤0.001 |
| Admission priority – pain (n=4690) | 0.15 (0.13–0.18) | P≤0.001 |
| Circulatory system – breathing (n=4701) | 0.12 (0.09–0.14) | P≤0.001 |
| Circulatory system – injuries (n=4701) | 0.10 (0.06–0.13) | P≤0.001 |
| Circulatory system – pain (n=4693) | 0.13 (0.11–0.16) | P≤0.001 |
| Neurology – consciousness (n=4701) | 0.18 (0.15–0.21) | P≤0.001 |
| Neurology – pain (n=4693) | −0.13 (–0.10 to –0.16) | P≤0.001 |
| Injuries – pain (n=4693) | 0.30 (0.27–0.33) | P≤0.001 |
Medical condition coded by RMC. Pooled analysis using data from 2017–2019
Neurological status was significantly correlated with consciousness (rho 0.18 (95% CI (0.15–0.21); P≤0.001; n=4,701) and inversely correlated with pain severity (rho: −0.13 (95% CI (−0.10 to −0.16); P≤0.001; n=4,693).
On the basis of these data, patients with more severe neurological dysfunction have more breathing dysfunction after femoral neck fracture but less pain. The strongest correlation, with a medium effect size, was detected between the severity of pain and the severity of injury (rho 0.30 (95% CI (0.27–0.33); P≤0.001; n=4,693). Other correlation analysis did not reveal a rho ≥0.10 and/or a P value ≤0.05.
Discussion
This study identified a high incidence of femoral neck fractures based on the RMI273. It also found correlations in patients with femoral neck fracture: between hospital admission priority and an acute state of the circulatory system; and between the severities of injuries and pain. In addition, greater age was significantly correlated with poorer neurological functioning, and the state of several organ systems correlated with each other.
Regarding age distribution, more than 90% of femoral neck fractures occurred in the 65–79 and ≥80 years age groups. The authors assume that most fractures were related to osteoporosis. In addition, a recent publication identified age- and sex-specific incidence rates for osteoporotic femoral fractures in Germany in the 65–74 years and ≥75 years age groups of 135 and 757 per 100 000 men and 223 and 1637 per 100 000 women respectively (Möckel et al, 2020), which are in the range of the incidence rates in the present analysis. This finding indicates that the suspected diagnosis of femoral neck fractures by paramedics are often subsequently confirmed as such. Therefore, EMS data are a potential source for future epidemiological studies.
Möckel et al (2020) calculated the number of osteoporotic femoral fractures based on the fractures reported in the GBE database (Information System of the Federal Health Monitoring; www.gbe-bund.de), which is comparable to the National Hip Fracture Database (NHFD; www.nhfd.co.uk) in the UK. However, the GBE database is less comprehensive than the NHFD. Furthermore, paramedic data is not included in the GBE database. Therefore, the authors recommend setting up fracture databases that capture patients' data from first examination by paramedics to inpatient treatment and following rehabilitation; this would allow the path of fracture patients through the healthcare system to be analysed with the aim of improving care and reducing risks of subsequent fracture and early mortality.
As expected, most patients with suspected femoral neck fracture were admitted to inpatient care. Nevertheless, around 4%–5% of patients were treated in outpatient settings or not even transported to a health facility. Unfortunately, the analysed data do not provide any information on why patients were not admitted for inpatient treatment. Therefore, the authors can only assume that these patients presented with isolated hip pain without clinical signs of a hip fracture and were referred to outpatient settings or general practitioners. Furthermore, a small number of patients may refuse to be conveyed to the hospital, even when this is recommended by the paramedics.
Admission priority was significantly correlated with severity of circulatory system impairments, injuries and pain. This indicates that decisions on hospital admission priority could be driven by these three characteristics. Therefore, patients with more severe injuries, pain or circulatory system issues are more likely to be admitted with an immediate intervention need, such as an emergency room admission. The authors could identify only one other study supporting these findings. This study reported on a criteria tool used to identify patients for direct inpatient hospital admission, and patient experience of trauma with injury (which has to be treated) was part of the tool (Jacobsson et al, 2020).
Based on the authors' analysis, poor functioning of the circulatory system was correlated with severity of breathing dysfunctions, injuries and pain. This indicates that patients in a great deal of pain are more likely to show severe circulatory system dysfunction. The relationship between the circulatory system and chronic pain in particular has already been described in the literature (Bruehl et al, 2005; Saccò et al, 2013). The data from Bruehl et al (2005) also indicate that chronic pain may be associated with hypertension.
A significant correlation was also detected between neurological functioning and age but indicated a small effect size of rho 0.10. Nevertheless, older patients are more likely to have neurological problems, including paralysis, disorientation and perceptual disorders. Since motor and cognitive impairment increase the risk of fracture (Logan et al, 2008; Stolee et al, 2009; Bonafede et al, 2016), neurological dysfunction may contribute to the higher incidence of femoral neck fracture in older age groups.
On the other hand, neurological disorders might contribute to less severe pain since neurological status was inversely correlated with pain (Table 4). The question of why neurological status is inversely correlated with pain severity remains open. The authors hypothesise that this might also be related to age and older patients tend to indicate higher rates of neurological impairment.
A relationship between age and pain perception was examined in a study with 41 postoperative patients aged ≥65 years and 249 patients aged <65 years. This revealed no differences in pain severity in both age groups, but analgesics were less needed in the older age group (Oberle et al, 1990). A more recent publication from 2020 with patients in these age groups with rib fractures showed older people were more likely to report no pain and had lower pain scores than the younger population (Choi et al, 2020).
Finally, consciousness and pain were both significantly correlated with neurological impairment. The authors suspect this might lead to an underdiagnosis and undertreatment of pain in patients with consciousness and/or neurological dysfunctions. Maintaining the balance between undertreatment and overtreatment in patients with disorder of consciousness is a well-known issue (Schnakers et al, 2012). Nevertheless, pain treatment is important in the prehospital setting.
Following the German guidelines and local protocols, the treatment options for patients with hip fracture available to paramedics include placement within a vacuum mattress with the knee elevated for splint-like reduction of movement and, if necessary, analgosedation with esketamine and midazolam (Stürmer et al, 2015; Hessian Ministry for Social Affairs and Integration, 2018).
Limitations
This study has several limitations, mainly related to the data available.
First, RMI273 and the corresponding RMCs did not provide any information on patient sex.
Second, there was no data available on any subsequent hospital diagnoses and interventions including surgeries, long-term comorbidities or outcomes of hospital interventions.
Third, as there was no information on comorbidities or on the causes of the femoral neck fractures, the present study cannot distinguish between osteoporotic and non-osteoporotic femoral neck fractures.
Fourth, the youngest age group (<50 years) must be considered small so the results for this group might be less reliable than those for the older age groups.
Conclusion
This analysis, based on the earliest possible examination of patients by health professionals, shows that most patients with femoral neck fracture have no or only mild comorbidities (with the exception of pain and injury).
As expected, most patients had to be admitted to inpatient care. Nevertheless, a small proportion of patients were treated in outpatient settings or not even transported and the reasons for this are unclear.
Decisions on hospital admission priority seem to be driven by the severity of injuries, pain and circulatory system impairment.
The authors identified that neurological impairments are significantly correlated with age, perceived pain and consciousness so more tailored analgesic medications might be needed for patients with suspected femoral neck fracture.
As mentioned above, the authors recommend setting up more comprehensive fracture databases that will link data from first paramedic examination to discharge from rehabilitation. The authors are hopeful that these kinds of data, including information gathered by paramedics, become available one day and might help to identify which patients are at a higher risk of mortality or of greater impairment in health-related quality of life after femoral fracture.
In addition, this kind of database would make it easier to trace patients who are treated in outpatient settings or not transported to a health facility and to find out why this happened.