Nottingham Hip Fracture Score and Acute Physiology and Chronic Health Evaluation-II: Predicting 30-day mortality in elderly hip fracture

Abstract

BACKGROUND

Hip fractures are a major cause of morbidity and mortality among older individuals. Accurate prediction of 30-day mortality after hip fracture surgery is crucial for effective resource allocation and patient management.

AIM

To compare the Nottingham Hip Fracture Score (NHFS) and Acute Physiology and Chronic Health Evaluation (APACHE)-II score in predicting 30-day mortality among 182 geriatric patients (≥ 60 years) undergoing elective hip fracture surgery at Tata Main Hospital, Jamshedpur, India.

METHODS

NHFS and APACHE-II scores were calculated preoperatively. Patients were followed up for 30 days post-surgery to determine mortality outcomes. Statistical analysis, including receiver operating characteristic curve analysis, was performed to assess the predictive accuracy of both scores.

RESULTS

Results indicated that APACHE-II demonstrated superior predictive ability compared to NHFS. The study also assessed secondary outcomes such as ventilator and inotropic support, acute kidney injury (AKI), cardiac morbidity, pulmonary embolism, deep vein thrombosis, and length of hospital stay (LOS). APACHE-II showed superior performance in predicting the need for ventilator support and AKI.

CONCLUSION

Both scores showed significant discrimination for ventilator and inotropic support and LOS but not for AKI. The findings highlight the importance of comprehensive risk stratification for managing geriatric patients with hip fracture.

Key Words: Acute Physiology and Chronic Health Evaluation-II; Hip fractures; Scoring methods; Mortality; Geriatrics; Outcome assessment; Receiver operating characteristic curve

Core Tip: This prospective observational study at Tata Main Hospital in Jamshedpur, India, evaluated the Nottingham Hip Fracture Score (NHFS) and Acute Physiology and Chronic Health Evaluation (APACHE)-II score for predicting 30-day mortality in 182 geriatric patients with hip fracture (≥ 60 years). The study assessed the primary (mortality prediction) and secondary objectives, including the requirement for ventilator/ionotropic support and monitoring for acute kidney injury and other complications. Preoperatively, the NHFS and APACHE-II scores were calculated, and patients were followed up for a minimum of 30 days post-surgery. Statistical analyses included receiver operating characteristic curves, area under the curve (AUC), sensitivity, specificity, and P value using Statistica. Results indicated that NHFS showed moderate predictive accuracy (AUC = 0.709, P = 0.006), whereas APACHE-II demonstrated excellent predictive ability (AUC = 0.904, P < 0.001). Significant predictors of mortality included the APACHE-II score, NHFS, mean arterial pressure, pH, respiratory rate, and serum creatinine levels. Although age and gender showed notable trends, they were not statistically significant. The findings indicate that the APACHE-II score is a strong predictor of 30-day mortality following hip fracture surgery, highlighting the importance of incorporating multiple factors for effective prediction.

INTRODUCTION

Hip fractures are a major public health concern, particularly in older individuals[1]. The projected increase in hip fracture incidence driven by aging populations and the increasing prevalence of osteoporosis poses substantial challenges for healthcare systems, leading to increased morbidity and mortality[2,3]. The effects of hip fracture extend beyond the immediate injury, affecting mobility, quality of life, and cognitive function. Patients often experience declines in activities of daily living, increased dependence, and a higher risk of delirium and long-term cognitive impairment[4].

Postoperative mortality rates following hip fracture surgery can reach 5%–10% within 30 days and may rise to 33% within a year[5-7]. Delays in surgical intervention significantly increase these rates, highlighting the need for accurate risk assessment.

Various scoring systems, such as the Nottingham Hip Fracture Score (NHFS) and the Acute Physiology and Chronic Health Evaluation-II (APACHE-II) score, can help predict mortality risk[8,9]. The NHFS uses seven preoperative variables [age, sex, hemoglobin levels, Mini-Mental State Examination (MMSE) score, institutionalization status, comorbidities, and malignancy] for straightforward assessment of frailty among patients[10]. In contrast, the APACHE-II score integrates 12 physiological variables, offering a more comprehensive but complex assessment of illness severity[11,12].

Both scoring systems have shown efficacy in predicting 30-day mortality; however, comparative analyses of their effectiveness remain limited[9,12]. Addressing this knowledge gap is vital for identifying the most effective risk stratification tools, ultimately enhancing personalized care strategies and patient outcomes. The present study aimed to address this gap by providing essential insights into hip fracture management among geriatric patients. This study directly addresses this knowledge gap and is possibly the first such study to directly compare the predictive accuracy of the APACHE-II and NHFS, specifically for mortality.

Our research examined whether there is a 10% difference in NHFS and APACHE-II scores in predicting 30-day mortality among geriatric patients (aged ≥ 60 years) undergoing hip fracture surgery. This comparative analysis aimed to determine which scoring system more accurately predicts postoperative mortality and to support clinical decision-making for improved patient care. The results of this study can potentially influence resource allocation and treatment strategies for hip fracture management.

MATERIALS AND METHODS

Study design

This study employed a prospective observational design. The prospective approach allowed real-time data collection from the time of patient admission until the 30-day follow-up or the occurrence of death. The observational nature of the study ensured that data collection occurred without intervention in the standard clinical care pathway, minimizing bias that could arise from experimental manipulation. This study aimed to compare the NHFS and APACHE-II scores for predicting 30-day mortality among geriatric patients (≥ 60 years) undergoing elective hip fracture surgery. Interobserver bias in APACHE-II scoring was reduced by implementing strict guidelines and training programs. The NHFS minimizes interobserver bias by relying on objective, routinely available data at admission, including age, sex, hemoglobin levels, comorbidities, living status, and cognitive function assessment.

Patients

The study population comprised geriatric patients aged ≥ 60 years admitted to Tata Main Hospital, Jamshedpur, India, between June 20, 2022, and June 19, 2023, for elective hip fracture surgery. Inclusion criteria included patients aged ≥ 60 years, undergoing elective surgery for hip fracture, and who provided informed consent. Exclusion criteria were patients younger than 60 years, those with hip fractures not undergoing invasive surgery, those scheduled for implant removal of a previously healed hip fracture, and those undergoing re-exploration surgery during the same admission as the initial hip fracture surgery. Ethical approval was obtained from the Institutional Ethics Committee of Tata Main Hospital before study commencement, and written informed consent was obtained from all patients. This study was registered on ClinicalTrials.gov (NCT06598566). A total of 182 patients met the inclusion criteria.

Data collection

Data were collected prospectively from patients’ medical records. Preoperative data included demographic information (age and sex) and clinical measurements [temperature, mean arterial pressure (MAP), heart rate, respiratory rate, partial pressure of oxygen, arterial pH, serum sodium levels, serum potassium levels, serum creatinine levels, hematocrit, and total leukocyte count], which were collected within 2 days of admission. Preoperative assessments also included the NHFS and APACHE-II scores. For NHFS, data were obtained on age, sex, hemoglobin level on admission, MMSE score, institutional living status, number of comorbidities, and presence of malignancy, as outlined in the original scoring system[13]. APACHE-II score was calculated using the established scoring system, which includes physiological and chronic health variables. Postoperative data were collected for a minimum of 30 days after surgery to determine 30-day mortality outcomes. Secondary outcome measures included the need for postoperative ventilator support, the need for ionotropic support, acute kidney injury (AKI), cardiac morbidity, pulmonary embolism (PE) or deep vein thrombosis (DVT), and length of hospital stay (LOS). AKI was defined using Kidney Disease: Improving Global Outcomes criteria (2012 guidelines)[14].

Statistical analysis

The study data were analyzed using Statistica version 13 (TIBCO Software Inc.). The distribution of continuous variables was tested using the Shapiro–Wilk test. For normally distributed data, the independent samples t-test was used to compare means between survivors and non-survivors, whereas the Mann–Whitney U test was used for non-normally distributed data. Categorical variables were analyzed using the χ² test or Fisher’s exact test, as appropriate.

Receiver operating characteristic (ROC) curve analysis was used to evaluate differences in NHFS and APACHE-II scores between survivors and non-survivors, revealing statistically significant differences. Sensitivity, specificity, and optimal cutoff points were determined from this analysis. Secondary outcomes, such as postoperative ventilator support, ionotropic support, and AKI, were also assessed. The LOS was analyzed using Spearman’s rank correlation coefficient. There were no specific incidences of cardiac morbidity, PE, or DVT. A P value < 0.05 was considered statistically significant. Multivariate logistic regression was performed to address potential confounders.

RESULTS

Study demographics and baseline characteristics

This prospective observational study analyzed data from 182 geriatric patients aged ≥ 60 years undergoing elective hip fracture surgery at Tata Main Hospital between June 2022 and June 2023. The cohort had a diverse age distribution, with 31.9% aged 60–70 years, 44.0% aged 71–80 years, and 24.2% aged over 80 years. Sex distribution was relatively balanced (53.8% female and 46.1% male). Baseline characteristics revealed that 63.7% of the patients had more than two comorbidities, and 47.8% had hemoglobin levels below 10 g/dL. The American Society of Anesthesiologists (ASA) classification indicated that most patients had moderate–to-severe systemic disease, approximately 3.3% had a malignancy, and almost 54% had organ insufficiency. The mean MMSE score of the cohort was 6.07, with most patients diagnosed with an intertrochanteric fracture. The average LOS was < 7 days with nearly 16% of patients, and the majority stayed between 7 days and 15 days (72.5%). The overall 30-day mortality rate was 8.8%.

Of the patients, 65 hip fractures were treated with proximal femoral nailing, 57 underwent fixation with a dynamic hip screw, and 50 patients underwent hemiarthroplasty. Among the remaining 10 patients, 2 underwent total hip arthroplasty, 3 underwent bipolar hip arthroplasty, 3 underwent fixation with a dynamic condylar screw, 1 underwent open reduction and internal fixation, and 1 underwent fixation with cannulated cancellous screws. The demographic characteristics of the patients are summarized in Table 1.

Table 1 Distribution of different variables in studied cases.

Statistics (categorical variables)
Name of characteristics (n = 182)	Categories	Numbers	Percentage
Age in year	60-70 years	58	31.9
	71-80 years	80	44.0
	> 80 years	44	24.2
Sex	Female	98	53.8
	Male	84	46.1
Hemoglobin in gm/dL	> 10	95	52.2
	≤ 10	87	47.8
Length of stay in days	< 7	29	15.9
	7-15	132	72.5
	16-30	21	11.5
Malignancy	No	176	96.7
	Yes	6	3.3
Number of comorbidities	< 2	66	36.3
	≥ 2	116	63.7
Admission Mini Mental Test Score	> 6/10	100	54.9
	≤ 6/10	82	45.1
Diagnosis (fracture)	IT left femur	60	33.0
	IT right femur	65	35.7
	Left acetabulum	1	0.5
	NOF left	29	15.9
	NOF right	27	14.8
American Society of Anesthesiologists	II	92	50.5
	III	88	48.4
	IV	2	1.1
H/O organ insufficiency	No	84	46.2
	Yes	98	53.8
Renal failure	No	148	81.3
	Acute renal failure	13	7.1
	Chronic renal failure	21	11.5
Outcome	Survived	166	91.2
	Mortality	16	8.8

IT: Intertrochanteric; NOF: Neck of femur.

Key factors influencing mortality

Several variables demonstrated statistically significant associations with 30-day mortality. Higher NHFS and APACHE-II scores were significantly associated with increased mortality. Patients with hemoglobin levels below 10 g/dL had significantly higher mortality rates. The LOS was significantly associated with survival. A history of organ insufficiency and the presence of renal failure were also significantly associated with increased mortality. The comparison of variables between survivors and non-survivors is summarized in Table 2.

Table 2 Comparison of variables between survival and mortality cases.

Comparison of continuous variables with outcome
Name of characteristics at admission (n = 182)	Survived (n = 166)		Mortality (n = 16)		P value
	Mean	SD	Mean	SD
Weight	61.44	5.56	61.88	16.78	0.182
Nottingham Hip Fracture Score	5.23	1.10	5.94	0.68	0.004a
Temp (rectal) (°C)	36.64	0.35	36.74	0.36	0.163
Mean arterial pressure	100.47	16.49	110.21	18.71	0.028a
Heart rate/minute	84.01	14.71	93.31	28.29	0.262
Respiratory rate/minute	15.66	2.74	18.38	2.68	< 0.001a
Partial pressure of oxygen	92.99	29.00	96.48	30.85	0.837
PH	7.40	0.05	7.34	0.09	0.002a
Serum Na+ (mEq/L)	130.91	6.63	129.69	8.02	0.498
Serum K+ (mEq/L)	4.08	0.58	4.19	0.80	0.663
Serum creatinine (mg/dL)	1.12	0.62	1.42	0.46	0.003a
Hematocrit (%)	31.11	4.73	30.00	3.95	0.252
Total leukocyte count (total/mm3)	9486.14	3219.32	11243.75	4120.84	0.080
Glasgow Coma Score	14.90	0.45	14.13	1.02	< 0.001a
Acute Physiology and Chronic Health Evaluation-II score	10.11	3.79	16.94	3.26	< 0.001a

^aP < 0.005 is significant.

Predictive performance of NHFS and APACHE-II scores

Both the NHFS and APACHE-II scores showed predictive capability for 30-day mortality; however, the findings indicated that APACHE-II has better predictive capability. The area under the curve (AUC) for NHFS was 0.709 (95%CI: 0.60-0.81), with a sensitivity of 75.00%, specificity of 61.00%, and cutoff value of 6. The AUC for APACHE-II was significantly higher at 0.904 (95%CI: 0.83-0.97), with a sensitivity of 87.50%, specificity of 74.00%, and cutoff value of 13, indicating a more accurate prediction model. Although both models could be used to assess mortality risk, the statistical differences indicate that APACHE-II is a better predictor (Figure 1).

Figure 1 Receiver operating characteristic curve for Nottingham Hip Fracture Score and Acute Physiology and Chronic Health Evaluation II score. APACHE-II: Acute Physiology and Chronic Health Evaluation-II; NHFS: Nottingham Hip Fracture Score; ROC: Receiver operating characteristic.

NHFS: The ROC curve analysis for NHFS provides insights into its predictive ability for a particular outcome. The AUC for NHFS was calculated to be 0.709. The AUC value ranges from 0 to 1, where a higher value indicates better discriminatory power of the variable. In this case, an AUC of 0.709 indicates that NHFS has a moderate level of predictive accuracy.

The 95%CI for the AUC is 0.60-0.81. This interval indicates the range within which the true AUC value is likely to fall. The lower and upper bounds indicate the possible values the AUC may take with 95%CI.

The P value associated with NHFS was 0.006, which is used to test the null hypothesis that the AUC is equal to 0.05 (which represents no discriminatory power). In this case, the P value was less than the typical significance level of 0.05, indicating that the ability of NHFS to predict the outcome was statistically significant. The sensitivity of NHFS was calculated to be 75.00%, and the specificity was 61.00%. Sensitivity measures the proportion of true positives correctly identified by the model, whereas specificity measures the proportion of true negatives correctly identified. These values provide information about the balance between correctly identifying positive and negative cases.

The cutoff value for NHFS was 6. This value represents the threshold at which the model differentiates between positive and negative cases. In other words, if an individual’s NHFS score is > 6, the model is more likely to predict a specific outcome (Table 3).

Table 3 Depicting the sensitivity and specificity of Nottingham Hip Fracture Score and Acute Physiology and Chronic Health Evaluation-II score.

Name of characteristics	Area under the curve	95%CI	P value	Sensitivity	Specificity	Cut of value
Nottingham Hip Fracture Score	0.709	0.60-0.81	0.006a	75.00%	61.00%	6
Acute Physiology and Chronic Health Evaluation-II	0.904	0.83-0.97	< 0.001a	87.50%	74.00%	13

^aP value < 0.05 considered significant.

APACHE-II score: The ROC curve analysis for the APACHE-II score indicates its predictive performance. The AUC for APACHE-II was calculated to be 0.904. With an AUC value close to 1, the APACHE-II score demonstrated excellent discriminatory power in predicting the outcome. The 95%CI for the AUC ranged from 0.83 to 0.97, indicating a range within which the true AUC value is likely to fall. The narrow CI range further demonstrates the model’s accuracy. The P value associated with APACHE-II was < 0.001, which indicates that APACHE-II was highly significant in predicting the outcome. The small P value indicates that the observed relationship between APACHE-II and the outcome was unlikely to have occurred by chance. The sensitivity and specificity of APACHE-II were 87.50% and 74.00%, respectively. These values indicate that APACHE-II is effective for correctly identifying both positive and negative cases. The cutoff value for APACHE-II is 13, which represents the threshold beyond which the model predicts a specific outcome. If an individual’s APACHE-II score exceeds 13, the model is more likely to predict a particular outcome.

The comparison of the APACHE-II and NHFS scores revealed valuable insights into their predictive accuracy for the risk of mortality. The APACHE-II score was significantly associated with the risk of mortality (P < 0.001). The Hosmer–Lemeshow test indicates a good fit for the model, with a statistic of χ² = 5.85 and a P value of 0.66, indicating that the predictions are consistent with observed outcomes. In contrast, the NHFS score also demonstrated a significant association with the risk of mortality (P = 0.015), although its model fit was weaker than that of the APACHE-II score. The Hosmer–Lemeshow test yielded a statistic of χ² = 5.37 and a P value of 0.15, indicating an acceptable model fit, although with slightly lower reliability. Overall, the APACHE-II model showed superior predictive accuracy for mortality risk assessment compared to that of the NHFS model.

To determine the optimal cutoff thresholds for predicting mortality using the APACHE-II and NHFS scores, Youden’s index, which identifies the point on the ROC curve that maximizes the sum of sensitivity and specificity, was also evaluated. For APACHE-II, the optimal cutoff threshold was approximately 15.5. At this point, the sensitivity was 0.75 and specificity was approximately 0.92. For NHFS, the optimal cutoff threshold was approximately 5.5. At this point, the sensitivity was 0.75 and specificity was approximately 0.62. This indicates that patients with an APACHE-II score > 15.5 are at higher risk of mortality, as are those with an NHFS score > 5.5; however, the APACHE-II score demonstrates better specificity.

Secondary outcomes

The APACHE-II score demonstrates superior performance compared to the NHFS across various clinical outcomes. Specifically, for ventilatory support, APACHE achieved an AUC of 0.838, which was significantly higher than NHFS, which achieved 0.686. Similarly, in terms of inotropic support, the APACHE score (AUC = 0.751) outperformed NHFS (AUC = 0.600). When assessing the prediction of AKI, APACHE excelled with an AUC of 0.845 compared with NHFS’s 0.698. Additionally, for LOS, although both scores exhibited weak positive correlations, APACHE exhibited a slightly stronger correlation (P = 0.356) than NHFS (P = 0.207). Overall, the APACHE-II score consistently exhibited better discriminatory ability than NHFS, particularly in predicting AKI and the need for ventilator support. Both scoring systems showed statistically significant discrimination for ventilator support, inotropic support, and LOS (P < 0.05). For AKI, although both scores showed good discrimination (particularly APACHE with AUC = 0.845), the difference between the scores was not significant (P = 0.117). APACHE consistently demonstrated stronger discriminatory ability across all outcomes than NHFS. Table 4 presents the discriminatory abilities of APACHE-II and NHFS scores for various outcomes, along with the statistical significance of their performance in predicting secondary outcomes.

Table 4 Discriminatory abilities of Acute Physiology and Chronic Health Evaluation-II score and Nottingham Hip Fracture Score for secondary outcomes.

Outcome	APACHE-II	NHFS	APACHE-II (P value)	NHFS (P value)
Ventilatory support (AUC)	0.838	0.686	0.00264	0.00264
Inotropic support (AUC)	0.751	0.6	0.00063	0.00063
Acute kidney injury (AUC)	0.845	0.698	0.117	0.117
Length of stay (Spearman's P)	0.356	0.207	8.41E-07	0.00513

APACHE-II: Acute Physiology and Chronic Health Evaluation-II; AUC: Area under the curve; NHFS: Nottingham Hip Fracture Score.

This prospective study included 182 geriatric patients with hip fracture. Most patients had multiple comorbidities, moderate-to-severe systemic disease, and short LOS. The 30-day mortality rate was 8.8%. The NHFS and APACHE-II scores were predictive of mortality, with APACHE-II demonstrating better predictive capability. Lower hemoglobin levels, organ insufficiency, and renal failure were associated with increased mortality. APACHE-II also showed superior performance in predicting secondary outcomes such as the need for ventilator support and AKI compared to NHFS.

DISCUSSION

The analysis of factors affecting survival outcomes highlights the crucial role of age and specific clinical indicators in determining patient prognosis. Age appears to be a critical factor, with individuals aged 60–80 years showing higher survival rates than older patients, particularly those above 80 years who exhibit increased mortality risk. This trend indicates that advanced age can adversely affect survival outcomes. Gender appears to play a negligible role in this context, indicating that survival rates are more influenced by other clinical variables.

The NHFS has been independently validated in various studies, revealing its efficacy as a predictor of 30-day mortality. For instance, Rushton et al[15] analyzed 1079 hip fracture cases in the United Kingdom, with a mean age of 83 years and 26% male patients. Tilkeridis et al[16] conducted a similar validation in a Greek population, confirming the NHFS’s utility in predicting outcomes among patients over a 3-year period, with a mean age of 80.82 years. These studies affirm the NHFS’s relevance across diverse demographics and geographic locations.

Hemoglobin levels were also found to be a critical predictor of survival. Patients with hemoglobin levels > 10 g/dL demonstrated significantly better survival odds, indicating that maintaining adequate hemoglobin levels is crucial for overall health[17]. Although some studies have shown an association between hematocrit levels and all-cause mortality in geriatric patients with hip fractures[18], our study did not show significant differences between survivors and those who succumbed within 30 days. Similarly, prolonged hospital stays (16–30 days) are correlated with higher mortality rates, indicating that extended exposure to medical settings may increase the likelihood of complications that negatively affect patient outcomes[19]. This association underscores the importance of efficient hospital management and timely intervention.

Further insights were obtained by recognizing the strong influence of malignancy on mortality rates. Patients with malignancies had markedly higher mortality rates, emphasizing the need to consider underlying health conditions when evaluating patient prognosis. Additionally, the analysis of comorbidities revealed a direct relationship between an increased number of coexisting health problems and mortality rates, highlighting the cumulative effect of multiple health issues. The ASA classification system, which evaluates systemic disease severity, also proved valuable, as patients in ASA class II exhibited better survival rates than those in higher classifications. Our study showed that higher MAP, higher respiratory rates, higher serum creatinine levels, and lower Glasgow Coma Score (GCS) scores were significantly associated with 30-day mortality among geriatric patients with hip fractures. These findings are consistent with those of similar studies on the mortality risk in geriatric patients with hip fractures[20-23].

The predictive power of scoring systems such as NHFS and APACHE-II cannot be overstated. The NHFS exhibited moderate predictive accuracy for mortality, whereas APACHE-II exhibited excellent discrimination and sensitivity, making it a robust tool for assessing patient outcomes. Recent studies have compared these models, with NHFS emerging as the preferred choice in clinical settings because of its ease of implementation and reliable predictive capabilities[24-26]. Continuous variables such as MAP, serum creatinine levels, and GCS scores further delineate the difference in outcomes between survivors and non-survivors, emphasizing the multifaceted factors influencing mortality. The APACHE-II score is a valuable tool for predicting mortality in geriatric patients with hip fractures because of several reasons. It provides a comprehensive assessment of a patient’s physiological status, considering various clinical parameters, such as heart rate, blood pressure, and age, that are crucial for older adults with multiple comorbidities. It assigns a score based on various physiological measurements and past medical history. Previous studies have reported its reliability in predicting mortality after geriatric hip fractures[27-29].

Although various studies have validated NHFS or APACHE-II independently for risk prognostication in hip fractures, only one study has compared the APACHE-II score with NHFS for mortality up to 1 year after surgery[27]. It showed that APACHE-II and Portsmouth Physiological and Operative Severity Score for the enumeration of Mortality and Morbidity were significantly better than ASA, Charlson Comorbidity Index, and NHFS for predicting mortality and can be used to predict mortality in older individuals with hip fractures[27]. This is the first study to compare APACHE-II score with NHFS for mortality. However, our findings are consistent with those of Tian et al[27], who concluded that although NHFS and APACHE-II scores are both valuable tools for risk stratification, APACHE-II score exhibits enhanced predictive accuracy, which significantly improves the ability to identify high-risk patients and aids in clinical decision-making. The study is limited by its single-center design, potentially restricting the generalizability of the findings to other populations and healthcare settings. The sample size of 182 patients, although adequate, may not be large enough to detect subtle differences between the NHFS and APACHE-II scoring systems. The focus on 30-day mortality outcomes provides only a short-term perspective, and longer-term outcomes were not assessed. Further multicenter trials with larger sample sizes across diverse geographic locations and longer follow-up periods are needed to validate our findings.

Regarding secondary outcomes, both the NHFS and APACHE-II scores served as effective tools for risk stratification, with each demonstrating significant discrimination for ventilator support, inotropic support, and LOS (P < 0.05). Regarding AKI, both scores exhibited good discrimination, particularly APACHE-II with an AUC of 0.845; however, the difference between the two scores was not significant (P = 0.117). Notably, APACHE-II consistently exhibited stronger discriminatory capability across all outcomes than NHFS.

Implementing APACHE-II in lower-resource settings presents substantial barriers. The complexity of data collection, including the 12 physiological variables, requires consistent access to laboratory facilities and monitoring equipment, which may be limited or unavailable. Staff training on APACHE-II scoring and data interpretation is essential but is frequently lacking. Furthermore, the time required for comprehensive data collection can strain already burdened healthcare personnel, potentially affecting workflow and accuracy. Overcoming these challenges necessitates simplified data collection protocols, targeted training programs, and resource allocation to ensure that basic monitoring facilities are available. The finding that APACHE-II is a superior predictor of mortality after hip fracture surgery could prompt a shift in clinical practice toward its use in resource allocation and patient management. Hospitals may prioritize intensive care resources for patients identified as high-risk by APACHE-II, potentially improving outcomes. In settings where both scores are feasible, use of APACHE-II can lead to more informed decisions regarding surgical timing and aggressiveness of postoperative care. It also highlights the need for resource investment to support the collection and use of these scores in clinical practice.

CONCLUSION

In conclusion, the ROC curve results for NHFS and APACHE-II provide valuable information about their predictive abilities. NHFS demonstrated a moderate level of predictive accuracy, with statistically significant results, whereas APACHE-II demonstrated a high AUC, strong statistical significance, and excellent sensitivity and specificity values. These findings indicate that APACHE-II is a key determinant of outcome, making it a valuable tool for predicting outcomes.

This study highlights the complex interplay of multiple factors in predicting 30-day mortality among geriatric patients undergoing hip fracture surgery. Although the NHFS and APACHE-II scores are valuable tools for risk stratification, our findings indicate that the APACHE-II score demonstrates superior predictive accuracy, offering substantial potential for identifying high-risk patients and guiding clinical decision-making. The combination of both measures may prove more effective. Furthermore, modifiable factors such as hemoglobin levels and the clinical management of comorbidities and organ dysfunction play critical roles in improving patient outcomes.