Establishing risk-adjusted quality indicators in surgery using administrative data-an example from neurosurgery

Benchmarking short-term postoperative mortality across neurosurgery units: is hospital administrative data good enough for risk-adjustment?

BACKGROUND

Surgical mortality indicators should be risk-adjusted when evaluating the performance of organisations. This study evaluated the performance of risk-adjustment models that used English hospital administrative data for 30-day mortality after neurosurgery.

METHODS

This retrospective cohort study used Hospital Episode Statistics (HES) data from 1 April 2013 to 31 March 2018. Organisational-level 30-day mortality was calculated for selected subspecialties (neuro-oncology, neurovascular and trauma neurosurgery) and the overall cohort. Risk adjustment models were developed using multivariable logistic regression and incorporated various patient variables: age, sex, admission method, social deprivation, comorbidity and frailty indices. Performance was assessed in terms of discrimination and calibration.

RESULTS

The cohort included 49,044 patients. Overall, 30-day mortality rate was 4.9%, with unadjusted organisational rates ranging from 3.2 to 9.3%. The variables in the best performing models varied for the subspecialties; for trauma neurosurgery, a model that included deprivation and frailty had the best calibration, while for neuro-oncology a model with these variables plus comorbidity performed best. For neurovascular surgery, a simple model of age, sex and admission method performed best. Levels of discrimination varied for the subspecialties (range: 0.583 for trauma and 0.740 for neurovascular). The models were generally well calibrated. Application of the models to the organisation figures produced an average (median) absolute change in mortality of 0.33% (interquartile range (IQR) 0.15-0.72) for the overall cohort model. Median changes for the subspecialty models were 0.29% (neuro-oncology, IQR 0.15-0.42), 0.40% (neurovascular, IQR 0.24-0.78) and 0.49% (trauma neurosurgery, IQR 0.23-1.68).

CONCLUSIONS

Reasonable risk-adjustment models for 30-day mortality after neurosurgery procedures were possible using variables from HES, although the models for trauma neurosurgery performed less well. Including a measure of frailty often improved model performance.

Quality indicators and early adverse in surgery for atypical meningiomas: A 16-year single centre study and systematic review of the literature

Introduction

Atypical meningiomas represent approximately 20% of all intracranial meningiomas and are characterized by distinct histopathological criteria and an increased risk of postoperative recurrence. Recently, quality indicators have been introduced to monitor quality of the delivered care.

Research question

Which quality indicators/outcome measures are being applied in patients being operated for atypical meningiomas? What are risk factors associated with poor outcome? How is the surgical outcome and which quality indicators are reported in the literature?

Material and methods

The primary outcomes of interest were 30-days readmission-, 30-day reoperation-, 30-day mortality-, 30-day nosocomial infection- and the 30-day surgical site infection (SSI) rate, CSF-leakage, new neurological deficit, medical complications, and lengths of stay. The secondary aim was the identification of prognostic factors for the mentioned primary outcomes. A systematic review of the literature was performed screening studies for the mentioned outcomes.

Results

We included 52 patients. 30-days outcomes in terms of unplanned reoperation were 0%, unplanned readmission 7.7%, mortality 0%, nosocomial infection 17.3%, and SSI 0%. Any adverse event occurred in 30.8%. Preoperative C-reactive protein over 5 ​mg/l was independent factor for the occurrence of any postoperative adverse event (OR: 17.2, p ​= ​0.003). A total of 22 studies were included into the review.

Discussion and conclusion

The 30-days outcomes at our department were comparable with reported outcomes in the literature. Currently applied quality indicators are helpful in determining the postoperative outcome but mainly report the indirect outcome after surgery and are influenced of patient, tumor and treatment related factors. Risk adjustment is vital.

Adverse events in neurosurgery: a comprehensive single-center analysis of a prospectively compiled database

PURPOSE

To prospectively identify and quantify neurosurgical adverse events (AEs) in a tertiary care hospital.

METHODS

From January 2021 to December 2021, all patients treated in our department received a peer-reviewed AE-evaluation form at discharge. An AE was defined as any event after surgery that resulted in an undesirable clinical outcome, which is not caused by the underlying disease, that prolonged patient stay, resulted in readmission, caused a new neurological deficit, required revision surgery or life-saving intervention, or contributed to death. We considered AEs occurring within 30 days after discharge. AEs were categorized in wound event, cerebrospinal fluid (CSF) event, CSF shunt malfunction, post-operative infection, malpositioning of implanted material, new neurological deficit, rebleeding, and surgical goal not achieved and non-neurosurgical AEs.

RESULTS

2874 patients were included. Most procedures were cranial (45.1%), followed by spinal (33.9%), subdural (7.7%), CSF (7.0%), neuromodulation (4.0%), and other (2.3%). In total, there were 621 AEs shared by 532 patients (18.5%). 80 (2.8%) patients had multiple AEs. Most AEs were non-neurosurgical (222; 8.1%). There were 172 (6%) revision surgeries. Patients receiving cranial interventions had the most AEs (19.1%) although revision surgery was only necessary in 3.1% of patients. Subdural interventions had the highest revision rate (12.6%). The majority of fatalities was admitted as an emergency (81/91 patients, 89%). Ten elective patients had lethal complications, six of them related to surgery (0.2%).

CONCLUSION

This study presents the one-year results of a prospectively compiled AE database. Neurosurgical AEs arose in one in five patients. Although the need for revision surgery was low, the rate of AEs highlights the importance of a systematic AE database to deliver continued high-quality in a high-volume center.

Can Currently Suggested Quality Indicators Be Transferred to Meningioma Surgery?-A Single-Center Pilot Study

BACKGROUND

 Risk stratification based on standardized quality measures has become crucial in neurosurgery. Contemporary quality indicators have often been developed for a wide range of neurosurgical procedures collectively. The accuracy of tumor-inherent characteristics of patients diagnosed with meningioma remains questionable. The objective of this study was the analysis of currently applied quality indicators in meningioma surgery and the identification of potential new measures.

METHODS

 Data of 133 patients who were operated on due to intracranial meningiomas were subjected to a retrospective analysis. The primary outcomes of interest were classical quality indicators such as the 30-day readmission, 30-day reoperation, 30-day mortality, 30-day nosocomial infection, and the 30-day surgical site infection rate. Uni- and multivariate analyses were performed. The occurrence of a new postoperative neurologic deficit was analyzed as a potential new quality indicator.

RESULTS

 The overall unplanned readmission rate was 3.8%; 13 patients were reoperated within 30 days (9.8%). The 30-day nosocomial infection and surgical site infection rates were 6.8 and 1.5%, respectively. A postoperative new neurologic deficit or neurologic deterioration as a currently assessed quality feature was observed in 12 patients (9.2%). The edema volume on preoperative scans proved to have a significant impact on the occurrence of a new postoperative neurologic deficit (p = 0.023).

CONCLUSIONS

 Classical quality indicators in neurosurgery have proved to correlate with considerable deterioration of the patient's health in meningioma surgery and thus should be taken into consideration for application in meningioma patients. The occurrence of a new postoperative neurologic deficit is common and procedure specific. Thus, this should be elucidated for application as a complementary quality indicator in meningioma surgery.

Mortality as an indicator of quality of neurosurgical care in England: a retrospective cohort study

OBJECTIVES

Postoperative mortality is a widely used quality indicator, but it may be unreliable when procedure numbers and/or mortality rates are low, due to insufficient statistical power. The objective was to investigate the statistical validity of postoperative 30-day mortality as a quality metric for neurosurgical practice across healthcare providers.

DESIGN

Retrospective cohort study.

SETTING

Hospital Episode Statistics data from all neurosurgical units in England.

PARTICIPANTS

Patients who underwent neurosurgical procedures between April 2013 and March 2018. Procedures were grouped using the National Neurosurgical Audit Programme classification.

OUTCOMES MEASURED

National 30-day postoperative mortality rates were calculated for elective and non-elective neurosurgical procedural groups. The study estimated the proportion of neurosurgeons and NHS trusts in England that performed sufficient procedures in 3-year and 5-year periods to detect unusual performance (defined as double the national rate of mortality). The actual difference in mortality rates that could be reliably detected based on procedure volumes of neurosurgeons and units over a 5-year period was modelled.

RESULTS

The 30-day mortality rates for all elective and non-elective procedures were 0.4% and 6.1%, respectively. Only one neurosurgeon in England achieved the minimum sample size (n=2402) of elective cases in 5 years needed to detect if their mortality rate was double the national average. All neurosurgical units achieved the minimum sample sizes for both elective (n=2402) and non-elective (n=149) procedures. In several neurosurgical subspecialties, approximately 80% of units (or more) achieved the minimum sample sizes needed to detect if their mortality rate was double the national rate, including elective neuro-oncology (baseline mortality rate=2.3%), non-elective neuro-oncology (rate=5.7%), neurovascular (rate=6.7%) and trauma (rate=11%).

CONCLUSION

Postoperative mortality lacks statistical power as a measure of individual neurosurgeon performance. Neurosurgical units in England performed sufficient procedure numbers overall and in several subspecialty areas to support the use of mortality as a quality indicator.

Postoperative surveillance in cranial and spinal tumor neurosurgery: when is this warranted?

OBJECTIVE

The outbreak of COVID-19 and the sudden increase in the number of patients requiring mechanical ventilation significantly affected the management of neurooncological patients. Hospitals were forced to reallocate already scarce human resources to maximize intensive care unit (ICU) capacities, resulting in a significant postponement of elective procedures for patients with brain and spinal tumors, who traditionally require elective postoperative surveillance on ICU or intermediate care wards. This study aimed to characterize those patients in whom postoperative monitoring is required by analyzing early postoperative complications and associated risk factors.

METHODS

All patients included in the analysis experienced benign or malignant cerebral or intradural tumors and underwent surgery between September 2017 and May 2019 at University Hospital Münster, Germany. Patient data were generated from a semiautomatic, prospectively designed database. The occurrence of adverse events within 24 hours and 30 days postoperatively-including unplanned reoperation, postoperative hemorrhage, CSF leakage, and pulmonary embolism-was chosen as the primary outcome measure. Furthermore, reasons and risk factors that led to a prolonged stay on the ICU were investigated. By performing multivariable logistic regression modeling, a risk score for early postoperative adverse events was calculated by assigning points based on beta coefficients.

RESULTS

Eight hundred eleven patients were included in the study. Eleven patients (1.4%) had an early adverse event within 24 hours, which was either an unplanned reoperation (0.9%, n = 7) or a pulmonary embolism (0.5%, n = 4) within 24 hours. To predict the incidence of early postoperative complications, a score was developed including the number of secondary diagnoses, BMI, and incision closure time, termed the SOS score. According to this score, 0.3% of the patients were at low risk, 2.5% at intermediate risk, and 12% at high risk (p < 0.001).

CONCLUSIONS

Postoperative surveillance in cranial and spinal tumor neurosurgery might only be required in a distinct patient collective. In this study, the authors present a new score allowing efficient prediction of the likelihood of early adverse events in patients undergoing neurooncological procedures, thus helping to stratify the necessity for ICU or intermediate care unit beds. Nevertheless, validation of the score in a multicenter prospective setting is needed.

Statistical Approaches for Quantifying the Quality of Neurosurgical Care

BACKGROUND

Quantifying quality of health care can provide valuable information to patients, providers, and policy makers. However, the observational nature of measuring quality complicates assessments.

METHODS

We describe a conceptual model for defining quality and its implications about the data collected, how to make inferences about quality, and the assumptions required to provide statistically valid estimates. Twenty-one binary or polytomous quality measures collected from 101,051 adult Medicaid beneficiaries aged 18-64 years with schizophrenia from 5 U.S. states show methodology. A categorical principal components analysis establishes dimensionality of quality, and item response theory models characterize the relationship between each quality measure and a unidimensional quality construct. Latent regression models estimate racial/ethnic and geographic quality disparities.

RESULTS

More than 90% of beneficiaries filled at least 1 antipsychotic prescription and 19% were hospitalized for schizophrenia during a 12-month observational period in our multistate cohort with approximately 2/3 nonwhite beneficiaries. Four quality constructs emerged: inpatient, emergency room, pharmacologic/ambulatory, and ambulatory only. Using a 2-parameter logistic model, pharmacologic/ambulatory care quality varied from -2.35 to 1.26 (higher = better quality). Black and Latinx beneficiaries had lower pharmacologic/ambulatory quality compared with whites. Race/ethnicity modified the association of state and pharmacologic/ambulatory care quality in latent regression modeling. Average quality ranged from -0.28 (95% confidence interval, -2.15 to 1.04) for blacks in New Jersey to 0.46 [95% confidence interval, -0.89 to 1.40] for whites in Michigan.

CONCLUSIONS

By combining multiple quality measures using item response theory models, a composite measure can be estimated that has more statistical power to detect differences among subjects than the observed mean per subject.

The Relation of Surgical Procedures and Diagnosis Groups to Unplanned Readmission in Spinal Neurosurgery: A Retrospective Single Center Study

Background: Unplanned readmission has gained increasing interest as a quality marker for inpatient care, as it is associated with patient mortality and higher economic costs. Spinal neurosurgery is characterized by a lack of epidemiologic readmission data. The aim of this study was to identify causes and predictors for unplanned readmissions related to index diagnoses and surgical procedures. Methods: In this study, from 2015 to 2017, spinal neurosurgical procedures were recorded for surgical and non-surgical treated patients. The main reasons for an unplanned readmission within 30 days following discharge were identified. Multivariate logarithmic regression revealed predictors of unplanned readmission. Results: A total of 1172 patient records were examined, of which 4.27% disclosed unplanned readmissions. Among the surgical patients, the readmission rate was 4.06%, mainly attributable to surgical site infections, while it was 5.06% for the non-surgical patients, attributable to uncontrolled pain. A night-time surgery presented as the independent predictive factor. Conclusion: In the heterogeneous group of spinal neurosurgical patients, stratification into diagnostic groups is necessary for statistical analysis. Degenerative lumbar spinal stenosis and spinal abscesses are mainly affected by unplanned readmission. The surgical procedure dorsal root ganglion stimulation is an independent predictor of unplanned re-hospitalizations, as is the timing of surgery.

Applicability of contemporary quality indicators in vestibular surgery-do they accurately measure tumor inherent postoperative complications of vestibular schwannomas?

Background

Due to rising costs in health care delivery, reimbursement decisions have progressively been based on quality measures. Such quality indicators have been developed for neurosurgical procedures, collectively. We aimed to evaluate their applicability in patients that underwent surgery for vestibular schwannoma and to identify potential new disease-specific quality indicators.

Methods

One hundred and three patients operated due to vestibular schwannoma were subject to analysis. The primary outcomes of interest were 30-day and 90-day reoperation, readmission, mortality, nosocomial infection and surgical site infection (SSI) rates, postoperative cerebral spinal fluid (CSF) leak, facial, and hearing function. The secondary aim was the identification of prognostic factors for the mentioned primary outcomes.

Results

Thirty-day (90-days) outcomes in terms of reoperation were 10.7% (14.6%), readmission 9.7% (13.6%), mortality 1% (1%), nosocomial infection 5.8%, and SSI 1% (1%). A 30- versus 90-day outcome in terms of CSF leak were 6.8% vs. 10.7%, new facial nerve palsy 16.5% vs. 6.1%. Hearing impairment from serviceable to non-serviceable hearing was 6.8% at both 30- and 90-day outcome. The degree of tumor extension has a significant impact on reoperation (p < 0.001), infection (p = 0.015), postoperative hemorrhage (p < 0.001), and postoperative hearing loss (p = 0.026).

Conclusions

Our data demonstrate the importance of entity-specific quality measurements being applied even after 30 days. We identified the occurrence of a CSF leak within 90 days postoperatively, new persistent facial nerve palsy still present 90 days postoperatively, and persisting postoperative hearing impairment to non-serviceable hearing as potential new quality measurement variables for patients undergoing surgery for vestibular schwannoma.

Neurosurgical Care during the COVID-19 Pandemic in Central Germany: A Retrospective Single Center Study of the Second Wave

The healthcare system has been placed under an enormous burden by the SARS-CoV-2 (COVID-19) pandemic. In addition to the challenge of providing sufficient care for COVID-19 patients, there is also a need to ensure adequate care for non-COVID-19 patients. We investigated neurosurgical care in a university hospital during the pandemic. We examined the second wave of the pandemic from 1 October 2020 to 15 March 2021 in this retrospective single-center study and compared it to a pre-pandemic period from 1 October 2019 to 15 March 2020. Any neurosurgical intervention, along with patient- and treatment-dependent factors, were recorded. We also examined perioperative complications and unplanned readmissions. A statistical comparison of the study groups was performed. We treated 535 patients with a total of 602 neurosurgical surgeries during the pandemic. This compares to 602 patients with 717 surgeries during the pre-pandemic period. There were 67 fewer patients (reduction to 88.87%) admitted and 115 fewer surgeries (reduction to 83.96%) performed, which were essentially highly elective procedures, such as cervical spinal stenosis, intracranial neurinomas, and peripheral nerve lesions. Regarding complication rates and unplanned readmissions, there was no significant difference between the COVID-19 pandemic and the non-pandemic patient group. Operative capacities were slightly reduced to 88% due to the pandemic. Nevertheless, comprehensive emergency and elective care was guaranteed in our university hospital. This speaks for the sufficient resources and high-quality processes that existed even before the pandemic.

Classical and disease-specific quality indicators in glioma surgery—Development of a quality checklist to improve treatment quality in glioma patients

Background

There is a pressing demand for more accurate, disease-specific quality measures in the field of neurosurgery. Aiming at most adequately measuring and reflecting the quality of glioma therapy, we developed a novel quality indicator bundle in form of a checklist for all patients that are treated operatively for glioma.

Methods

On the basis of possible glioma-specific quality indicators retrieved from the literature and quality guidelines, a multidisciplinary team developed a checklist containing 13 patient-need-specific outcome measures. Subsequently, the checklist was prospectively applied to a total of 78 patients compared with a control group consisting of 322 patients. A score was generated based on the maximum of quality measures achieved.

Results

Significant improvements in quality after prospectively introducing the checklist were achieved for supplemental physical and occupational therapy during inpatient stay (89.4% vs 100%, P = .002), consultation of a social worker during inpatient stay (64% vs 92.3%, P &lt; .001), psycho-oncological screening (14.3% vs 70.5%, P &lt; .001), psycho-oncological consultation (31.1% vs 82.1%, P &lt; .001), and consultation of the palliative care team (20% vs 40%, P = .031). Overall, after introduction of the checklist one-third (n = 23) of patients reached best-practice measures in all categories, and over half of the patients (n = 44) achieved above 90% with respect to the outcome measures.

Conclusions

Aiming at ensuring comprehensive, consistent, and timely care of glioma patients, the implementation of the checklist for routine use in glioma surgery represents an efficient, easily reproducible, and powerful tool for significant improvements.

The 5-factor modified frailty index: an effective predictor of mortality in brain tumor patients

OBJECTIVE

Health measures such as the Charlson Comorbidity Index (CCI) and the 11-factor modified frailty index (mFI-11) have been employed to predict general medical and surgical mortality, but their clinical utility is limited by the requirement for a large number of data points, some of which overlap or require data that may be unavailable in large datasets. A more streamlined 5-factor modified frailty index (mFI-5) was recently developed to overcome these barriers, but it has not been widely tested in neuro-oncology patient populations. The authors compared the utility of the mFI-5 to that of the CCI and the mFI-11 in predicting postoperative mortality in brain tumor patients.

METHODS

The authors retrospectively reviewed a cohort of adult patients from a single institution who underwent brain tumor surgery during the period from January 2017 to December 2018. Logistic regression models were used to quantify the associations between health measure scores and postoperative mortality after adjusting for patient age, race, ethnicity, sex, marital status, and diagnosis. Results were considered statistically significant at p values ≤ 0.05. Receiver operating characteristic (ROC) curves were used to examine the relationships between CCI, mFI-11, and mFI-5 and mortality, and DeLong's test was used to test for significant differences between c-statistics. Spearman's rho was used to quantify correlations between indices.

RESULTS

The study cohort included 1692 patients (mean age 55.5 years; mean CCI, mFI-11, and mFI-5 scores 2.49, 1.05, and 0.80, respectively). Each 1-point increase in mFI-11 (OR 4.19, p = 0.0043) and mFI-5 (OR 2.56, p = 0.018) scores independently predicted greater odds of 90-day postoperative mortality. Adjusted CCI, mFI-11, and mFI-5 ROC curves demonstrated c-statistics of 0.86 (CI 0.82-0.90), 0.87 (CI 0.83-0.91), and 0.87 (CI 0.83-0.91), respectively, and there was no significant difference between the c-statistics of the adjusted CCI and the adjusted mFI-5 models (p = 0.089) or between the adjusted mFI-11 and the adjusted mFI-5 models (p = 0.82). The 3 indices were well correlated (p < 0.01).

CONCLUSIONS

The adjusted mFI-5 model predicts 90-day postoperative mortality among brain tumor patients as well as our adjusted CCI and adjusted mFI-11 models. The simplified mFI-5 may be easily integrated into clinical workflows to predict brain tumor surgery outcomes in real time.

Development and validation of prediction scores for nosocomial infections, reoperations, and adverse events in the daily clinical setting of neurosurgical patients with cerebral and spinal tumors

OBJECTIVE

Various quality indicators are currently under investigation, aiming at measuring the quality of care in neurosurgery; however, the discipline currently lacks practical scoring systems for accurately assessing risk. The aim of this study was to develop three accurate, easy-to-use risk scoring systems for nosocomial infections, reoperations, and adverse events for patients with cerebral and spinal tumors.

METHODS

The authors developed a semiautomatic registry with administrative and clinical data and included all patients with spinal or cerebral tumors treated between September 2017 and May 2019. Patients were further divided into development and validation cohorts. Multivariable logistic regression models were used to develop risk scores by assigning points based on β coefficients, and internal validation of the scores was performed.

RESULTS

In total, 1000 patients were included. An unplanned 30-day reoperation was observed in 6.8% of patients. Nosocomial infections were documented in 7.4% of cases and any adverse event in 14.5%. The risk scores comprise variables such as emergency admission, nursing care level, ECOG performance status, and inflammatory markers on admission. Three scoring systems, NoInfECT for predicting the incidence of nosocomial infections (low risk, 1.8%; intermediate risk, 8.1%; and high risk, 26.0% [p < 0.001]), LEUCut for 30-day unplanned reoperations (low risk, 2.2%; intermediate risk, 6.8%; and high risk, 13.5% [p < 0.001]), and LINC for any adverse events (low risk, 7.6%; intermediate risk, 15.7%; and high risk, 49.5% [p < 0.001]), showed satisfactory discrimination between the different outcome groups in receiver operating characteristic curve analysis (AUC ≥ 0.7).

CONCLUSIONS

The proposed risk scores allow efficient prediction of the likelihood of adverse events, to compare quality of care between different providers, and further provide guidance to surgeons on how to allocate preoperative care.

Causes and Predictors of Unplanned Readmission in Cranial Neurosurgery

OBJECTIVE

A better understanding of the risks and reasons for unplanned readmission is an essential component in reducing costs in the health care system and in optimizing patient safety and satisfaction. The reasons for unplanned readmission vary between different disciplines and procedures. The aim of this study was to identify reasons for readmission in view of different diagnoses in cranial neurosurgery.

METHODS

In this single-center retrospective study, adult patients after neurosurgical treatment were analyzed and grouped according to the indication based on International Classification of Diseases and Related Health Problems, Tenth Revision, German Modification diagnosis codes. The main outcome measure was unplanned readmission within 30 days of discharge. Further logistic regression models were performed to identify factors associated with unplanned rehospitalization.

RESULTS

Of the 2474 patients analyzed, 183 underwent unplanned rehospitalization. Readmission rates differed between the diagnosis groups, with 9.19% in neoplasm, 8.26% in hydrocephalus, 5.76% in vascular, 6.13% after trauma, and 8.05% in the functional group. Several causes were considered to be preventable, such as wound healing disorders, seizures, or social reasons. Younger age, length of first stay, surgical treatment, and side diagnoses were predictors for unplanned readmission. Diagnoses with an increased risk of readmission were glioblastoma, traumatic subdural hematoma, or chronic subdural hematoma.

CONCLUSIONS

Reasons and predictors for an unplanned readmission differ considerably among the index diagnosis groups. In addition to well-known reasons for readmission, we identified social indication, meaning a lack of home care, which is particularly prevalent in oncologic and elderly patients. A transitional care program could benefit these vulnerable patients.

Early unplanned readmission of neurosurgical patients after treatment of intracranial lesions: a comparison between surgical and non-surgical intervention group

Background

Recent health care policy making has highlighted the necessity for understanding factors that influence readmission. To elucidate the rate, reason, and predictors of readmissions in neurosurgical patients, we analyzed unscheduled readmissions to our neurosurgical department after treatment for cranial or cerebral lesions.

Methods

From 2015 to 2017, all adult patients who had been discharged from our Department of Neurosurgery and were readmitted within 30 days were included into the study cohort. The patients were divided into a surgical and a non-surgical group. The main outcome measure was unplanned inpatient admission within 30 days of discharge.

Results

During the observation period, 183 (7.4%) of 2486 patients had to be readmitted unexpectedly within 30 days after discharge. The main readmission causes were surgical site infection (34.4 %) and seizure (16.4%) in the surgical group, compared to natural progression of the original diagnosis (38.2%) in the non-surgical group. Most important predictors for an unplanned readmission were younger age, presence of malignoma (OR: 2.44), and presence of cardiovascular side diagnoses in the surgical group. In the non-surgical group, predictors were length of stay (OR: 1.07) and the need for intensive care (OR: 5.79).

Conclusions

We demonstrated that reasons for readmission vary between operated and non-operated patients and are preventable in large numbers. In addition, we identified treatment-related partly modifiable factors as predictors of unplanned readmission in the non-surgical group, while unmodifiable patient-related factors predominated in the surgical group. Further patient-related risk adjustment models are needed to establish an individualized preventive strategy in order to reduce unplanned readmissions.

Electronic supplementary material

The online version of this article (10.1007/s00701-020-04521-4) contains supplementary material, which is available to authorized users.