Interrupted time-series analysis: studying trends in neurosurgery

Longitudinal evaluation of an electronic audit and feedback system for patient safety in a large tertiary hospital setting

Objective: This study sought to assess the impact of a novel electronic audit and feedback (e-A&F) system on patient outcomes. Methods: The e-A&F intervention was implemented in a tertiary hospital and involved near real-time feedback via web-based dashboards. We used a segmented regression analysis of interrupted time series. We modelled the pre-post change in outcomes for the (1) announcement of this priority list, and (2) implementation of the e-A&F intervention to have affected patient outcomes. Results: Across the study period there were 222,792 episodes of inpatient care, of which 13,904 episodes were found to contain one or more HACs, a risk of 6.24%. From the point of the first intervention until the end of the study the overall risk of a HAC reduced from 8.57% to 4.12% - a 51.93% reduction. Of this reduction the proportion attributed to each of these interventions was found to be 29.99% for the announcement of the priority list and 21.93% for the implementation of the e-A&F intervention. Discussion: Our findings lend evidence to a mechanism that the announcement of a measurement framework, at a national level, can lead to local strategies, such as e-A&F, that lead to significant continued improvements over time.

Impacts of bariatric surgery in health outcomes and health care costs in Brazil: Interrupted time series analysis of multi-panel data

Background

The increasing burden of obesity generates significant socioeconomic impacts for individuals, populations, and national health systems worldwide. The literature on impacts and cost-effectiveness of obesity-related interventions for prevention and treatment of moderate to severe obesity indicate that bariatric surgery presents high costs associated with high effectiveness in improving health status referring to certain outcomes; however, there is a lack of robust evidence at an individual-level estimation of its impacts on multiple health outcomes related to obesity comorbidities.

Methods

The study encompasses a single-centre retrospective longitudinal analysis of patient-level data using micro-costing technique to estimate direct health care costs with cost-effectiveness for multiple health outcomes pre-and post-bariatric surgery. Data from 114 patients who had bariatric surgery at the Hospital of Clinics of the University of Sao Paulo during 2018 were investigated through interrupted time-series analysis with generalised estimating equations and marginal effects, including information on patients' characteristics, lifestyle, anthropometric measures, hemodynamic measures, biochemical exams, and utilisation of health care resources during screening (180 days before) and follow-up (180 days after) of bariatric surgery.

Results

The preliminary statistical analysis showed that health outcomes presented improvement, except cholesterol and VLDL, and overall direct health care costs increased after the intervention. However, interrupted time series analysis showed that the rise in health care costs is attributable to the high cost of bariatric surgery, followed by a statistically significant decrease in post-intervention health care costs. Changes in health outcomes were also statistically significant in general, except in cholesterol and LDL, leading to significant improvements in patients' health status after the intervention.

Conclusions

Trends multiple health outcomes showed statistically significant improvements in patients' health status post-intervention compared to trends pre-intervention, resulting in reduced direct health care costs and the burden of obesity.

An evaluation of an electronic audit and feedback system for patient safety in a tertiary hospital setting: A study protocol

An electronic audit and feedback (e-A&F) system was developed to support healthcare providers' awareness of their own performance, improve delivery of care and ultimately the safety of patients while in hospital. The point-of-care e-A&F system provides healthcare providers, from a 600-bed tertiary hospital in Western Australia, with near real-time feedback via web-based dashboards. The aim of this evaluation is to determine the implications of e-A&F across multiple dimensions and domains of care in a tertiary hospital setting. The study also aims to address the paucity in the literature by validating hypothesised design and implementation mechanisms on its effectiveness. Key datasets to be examined include those related to patient outcomes, staff behaviour and costs. Quantitative methods, such as interrupted time series analysis and multiple logistic regression analysis, amongst other methods, will be employed to achieve these aims.

Demand for Essential Nonambulatory Neurosurgical Care Decreased While Acuity of Care Increased During the Coronavirus Disease 2019 (COVID-19) Surge

Background

In times of health resource reallocation, capacities must remain able to meet a continued demand for essential, nonambulatory neurosurgical acute care. This study sought to characterize the demand for and provision of neurosurgical acute care during the coronavirus disease 2019 (COVID-19) pandemic.

Methods

This single-center cross-sectional observational analysis compared nonambulatory neurosurgical consult encounters during the peri-surge period (March 9 to May 31, 2020) with those during an analogous period in 2019. Outcomes included consult volume, distribution of problem types, disease severity, and rate of acute operative intervention.

Results

A total of 1494 neurosurgical consults were analyzed. Amidst the pandemic surge, 583 consults were seen, which was 6.4 standard deviations below the mean among analogous 2016–2019 periods (mean 873; standard deviation 45, P = 0.001). Between 2019 and 2020, the proportion of degenerative spine consults decreased in favor of spinal trauma (25.6% vs. 34% and 51.9% vs. 41.4%, P = 0.088). Among aneurysmal subarachnoid hemorrhage cases, poor-grade (Hunt and Hess grades 4–5) presentations were more common (30% vs. 14.8%, P = 0.086). A greater proportion of pandemic era consults resulted in acute operative management, with an unchanged absolute frequency of acutely operative consults (123/583 [21.1%] vs. 120/911 [13.2%], P < 0.001).

Conclusions

Neurosurgical consult volume during the pandemic surge hit a 5-year institutional low. Amidst vast reallocation of health care resources, demand for high-acuity nonambulatory neurosurgical care continued and proportionally increased for greater-acuity pathologies. In our continued current pandemic as well as any future situations of mass health resource reallocation, neurosurgical acute care capacities must be preserved.

Effectiveness of employer financial incentives in reducing time to report worker injury: an interrupted time series study of two Australian workers' compensation jurisdictions

Background

Early intervention following occupational injury can improve health outcomes and reduce the duration and cost of workers’ compensation claims. Financial early reporting incentives (ERIs) for employers may shorten the time between injury and access to compensation benefits and services. We examined ERI effect on time spent in the claim lodgement process in two Australian states: South Australia (SA), which introduced them in January 2009, and Tasmania (TAS), which introduced them in July 2010.

Methods

Using administrative records of 1.47 million claims lodged between July 2006 and June 2012, we conducted an interrupted time series study of ERI impact on monthly median days in the claim lodgement process. Time periods included claim reporting, insurer decision, and total time. The 18-month gap in implementation between the states allowed for a multiple baseline design. In SA, we analysed periods within claim reporting: worker and employer reporting times (similar data were not available in TAS). To account for external threats to validity, we examined impact in reference to a comparator of other Australian workers’ compensation jurisdictions.

Results

Total time in the process did not immediately change, though trend significantly decreased in both jurisdictions (SA: −0.36 days per month, 95% CI −0.63 to −0.09; TAS: 0.35, −0.50 to −0.20). Claim reporting time also decreased in both (SA: −1.6 days, −2.4 to −0.8; TAS: -5.4, −7.4 to −3.3). In TAS, there was a significant increase in insurer decision time (4.6, 3.9 to 5.4) and a similar but non-significant pattern in SA. In SA, worker reporting time significantly decreased (−4.7, −5.8 to −3.5), but employer reporting time did not (−0.3, −0.8 to 0.2).

Conclusions

The results suggest that ERIs reduced claim lodgement time and, in the long-term, reduced total time in the claim lodgement process. However, only worker reporting time significantly decreased in SA, indicating that ERIs may not have shortened the process through the intended target of employer reporting time. Lack of similar data in Tasmania limited our ability to determine whether this was a result of ERIs or another component of the legislative changes. Further, increases in insurer decision time highlight possible unintended negative effects.

Electronic supplementary material

The online version of this article (10.1186/s12889-017-4998-9) contains supplementary material, which is available to authorized users.

Quality Programs in Neurosurgery: The Memorial Hermann/University of Texas Experience

The importance of outcome measures is steadily increasing due to the rise of "pay for performance" and the advent of population health. In 2007, a quality initiative was started due to poor performance on rankings such as the University Health Consortium (UHC) report card. Inherent to all such efforts are common challenges: how to engage the providers; how to gather and ensure the accuracy of the data; how to attribute results to individuals; how to ensure permanent improvements. After analysis, a strategy was developed that included an initial focus on 3 metrics (mortality, infection rates, and complications), leadership from practicing neurosurgeons, protocol development and adherence, and subspecialization. In addition, it was decided that the metrics would initially apply to attending physicians only, but that the entire team would need to be involved. Once the fundamental elements were established, the process could be extended to other measures and providers. To support this effort, special information system tools were developed and a support team formed. As the program matured, measured outcomes improved and more metrics were added (to a current total of 48). For example, UHC mortality ratios (observed over expected) decreased by 75%. Infection rates decreased 80%. The program now involves all trainee physicians, advanced practice providers, nurses, and other staff. This paper describes the design, implementation, and results of this effort, and provides a practical guide that may be useful to other groups undertaking similar initiatives.