The primary objective of this study was to examine potential disparities in the management of hip fractures between teaching hospitals and non-teaching hospitals. Additionally, the study aimed to assess the mediating effect of the teaching hospital setting in comparison to public non-teaching hospitals. By investigating these factors, the study sought to gain insight into the impact of hospital teaching status on hip fracture management and explore any potential differences that may exist.
Regarding the outcomes of interest identified in the study, patients receiving treatment in teaching hospitals demonstrated shorter time intervals to surgery, higher rates of postoperative ICU admissions, increased requirements for blood transfusions, and elevated revision rates. However, perioperative hemoglobin levels, occurrences of venous thromboembolic events (VTE), readmission rates, and mortality did not exhibit significant differences compared to patients treated in non-teaching hospitals.
When comparing our study findings to international standards such as those established by the National Hip Fracture Database and the National Falls and Fragility Fracture Adult Programme (FFFAP) 202324, notable some emerge.
Firstly, the mean length of hospital stays in our study groups averaged around 7 days, considerably shorter than the approximately 16 days reported in the FFFAP data. This disparity can be attributed in part to the provision of postoperative rehabilitation in the FFFAP groups prior to discharge, contrasting with the more limited empirical rehabilitation provided in our study due to the absence of dedicated rehabilitation facilities. Secondly, analysis of anesthesia types revealed differences between our study groups and the FFFAP data. In teaching hospitals, 66.9% of patients received spinal anesthesia, compared to 54.0% in non-teaching hospitals and 53.5% in the FFFAP data. Conversely, the use of general anesthesia was more prevalent in non-teaching hospitals (46.0%) compared to teaching hospitals (33.1%) and the FFFAP data (48.1%). Finally, regarding revision rates, our study observed a higher revision rate in teaching hospitals (4.4%) compared to non-teaching hospitals (1.7%) and the FFFAP data (1.1%). This disparity may be attributed to surgeries performed by residents still in training in teaching hospitals, potentially leading to a higher reoperation rate. However, further research is warranted to delve deeper into this issue and explore potential contributing factors.
Hip fracture patients often exhibit frailty and multiple medical comorbidities, as evidenced by previous research. Boddaert et al. observed that 95% of elderly hip fracture patients present with at least one major comorbidity25. Härstedt et al. identified common postoperative comorbidities including hypertension, cognitive disorders, and ischemic heart disease26. Henderson et al. reported hypertension, osteoporosis, and ischemic heart disease as prevalent comorbidities in geriatric hip fracture patients27. In our study, encompassing both teaching and non-teaching hospital groups, hypertension and diabetes were the most prevalent comorbidities, while neurocognitive disorders were less common.
The literature concerning outcomes of hip fracture management in teaching hospitals is sparse, with limited evaluation of these outcomes. However, McGuire et al. found that patients treated in teaching hospitals exhibited a 1.4% decrease in 6-month mortality compared to those in non-teaching hospitals28. In our study, we observed a 0.4% lower mortality rate at 6 months in teaching hospitals compared to non-teaching hospitals. Similarly, Konda et al. reported associations between hip fracture management in teaching hospitals, shorter lengths of stay, and reduced in-hospital mortality29, findings that align with our study results.
In our study, the 1-year mortality rate was 3.7% for both teaching and non-teaching hospitals. This rate is significantly lower than the rates reported in previous studies on geriatric hip fracture management. Several observations can explain this discrepancy.
Firstly, the average age of patients in our cohort was 76 years for the teaching hospital group and 74 years for the non-teaching hospital group. These averages are lower than those reported in previous studies. For instance, Moyet et al. found the mean age for elderly patients with hip fractures to be 82.6 ± 7.4 years in a systematic review of the orthogeriatric care model30. Similarly, Basques et al. reported a mean age of 83.8 years in a study of 8434 patients31. Other studies have also reported higher mean ages, including Chen et al. (80.8 years)32, Krishnan et al. (81 years) 33, and Gleich et al. (85 years)34.
Secondly, a significant proportion of our patient cohort were non-smokers: three-quarters in the teaching hospital group and two-thirds in the non-teaching hospital group. Additionally, medical conditions linked to increased mortality, such as cardiovascular disease35, cerebrovascular disease36, renal impairment37, and neurocognitive diseases including dementia38, were less common in our cohort, affecting at most one-third of the patients.
Thirdly, we assessed the mortality rate using comprehensive data from both the Ministry of Health’s electronic health records and those of the teaching hospitals. This method allowed for accurate detection of mortality dates relative to the date of surgery. However, this number may still be underestimated. Some patients might seek care at hospitals outside these two major health sectors due to financial reasons or health insurance coverage issues, a phenomenon reported by Tewari et al. 39. Consequently, our study’s one-year mortality rate might not capture all deaths.
Therefore, combining the lower mean ages of our cohort, the lower prevalence of serious medical comorbidities, and the smaller number of smokers, along with considering that some patients may seek care outside the Ministry of Health and Teaching hospital health networks at the time of death, all contribute to the lower one-year mortality rates observed. However, a future follow-up study is in our scope to track mortality rates over longer time periods.
In our study, we observed that patients treated in teaching hospitals experienced significantly shorter times from hospital admission to surgery compared to those in non-teaching hospitals. Specifically, the time from admission to surgery was 2.2 days in teaching hospitals, while it was 3.65 days in non-teaching hospitals. This difference underscores the potential benefits of receiving care in teaching hospitals, where there may be greater resources and capabilities to promptly stabilize and optimize patients’ medical conditions prior to surgery. This efficient preoperative management likely contributes to the shorter time to surgery observed in teaching hospital settings.
In our study, hip fracture patients treated in teaching hospital settings were more likely to receive postoperative blood transfusions. This higher rate of transfusions in teaching hospitals suggests a lower threshold for administering blood products, possibly indicating a tendency towards more aggressive postoperative care practices.
Furthermore, an investigation and comparison of postoperative ICU admissions were conducted. Notably, there were no strict policy standards across the participating hospitals, with the decision to admit patients to the ICU depending upon various factors, including clinical status deterioration, laboratory and imaging assessments, evaluation by the ICU and medical teams, as well as consultations with relevant consultants. The observed higher rate of postoperative ICU admissions in teaching hospitals, while potentially associated with increased complication rates, also suggests a propensity for ICU utilization, indicating a heightened level of vigilance and attentiveness to patient needs within these settings. Collectively, these findings underscore the importance of comprehensive care and heightened vigilance characteristic of teaching hospitals. Contrary to concerns about increased hospital costs associated with heightened patient care in teaching hospitals, Konda et al. found no such association in the context of hip fractures29.
Previous research has examined the complications associated with hip fracture management in teaching hospitals, yielding conflicting results. Weller et al. reported a decreased risk of in-hospital mortality for hip fracture patients in teaching hospitals40. In contrast, Koval et al. reported higher in-hospital mortality rates and increased overall complications in hospital settings41. Similarly, Anderson et al. found a 3.6% increase in mortality among hip fracture patients in teaching hospitals compared to non-teaching hospitals42. Contrasting these findings, our study revealed that hip fracture patients treated in teaching hospitals experienced slightly higher rates of venous thromboembolic (VTE) events, comparable readmission rates, and similar mortality rates. These findings contribute to the existing literature by providing additional insights into the outcomes of hip fracture management in teaching hospitals, highlighting the need for further investigation and potential variability across different healthcare settings.
One limitation of this study pertains to its retrospective design and the relatively modest sample size of patients included. Despite these limitations, the study was able to draw meaningful conclusions and identify significant differences in findings. Another limitation is the absence of cost analysis, which was challenging to incorporate due to the complexity of healthcare insurance systems and variations in coverage within Jordan. However, it is worth noting that to the best of the authors’ knowledge, this study represents the first investigation into the management of hip fractures in teaching hospitals in Jordan and the Middle East.
link