A critical analysis of discovery health's claims-based risk adjustment of mortality rates in South African private sector hospitals

In 2019, Discovery Health published a risk adjustment model to determine standardised mortality rates across South African private hospital systems, with the aim of contributing towards quality improvement in the private healthcare sector. However, the model suffers from limitations due to its design and its reliance on administrative data. The publication's aim of facilitating transparency is unfortunately undermined by shortcomings in reporting. When designing a risk prediction model, patient-proximate variables with a sound theoretical or proven association with the outcome of interest should be used. The addition of key condition-specific clinical data points at the time of hospital admission will dramatically improve model performance. Performance could be further improved by using summary risk prediction scores such as the EUROSCORE II for coronary artery bypass graft surgery or the GRACE risk score for acute coronary syndrome. In general, model reporting should conform to published reporting standards, and attempts should be made to test model validity by using sensitivity analyses. In particular, the limitations of machine learning prediction models should be understood, and these models should be appropriately developed, evaluated and reported.

Hypotension during spinal anaesthesia for Caesarean section in a resource limited setting: towards a consensus definition

Background: Intraoperative hypotension following spinal anaesthesia for Caesarean section is associated with maternal morbidity and mortality. Because of inconsistent definitions the reported incidence of hypotension varies between 7% and 74%, making it almost impossible to set standard targets. Developing and adopting a clinically significant threshold for intraoperative hypotension will allow for benchmarking, comparison between studies, and consistency in guidelines and recommendations. Methods: Common definitions for spinal hypotension were first identified from a recent systematic review of the literature and a consensus statement on spinal hypotension. These definitions were applied to haemodynamic data taken from a prospective interventional obstetric spinal hypotension study conducted at Edendale Hospital, to determine the incidence of hypotension when applying these different thresholds. Finally, a definition was proposed based on these incidences and a review of the relevant literature. Results: Fifteen different definitions were identified. These were then applied to the study population with a resultant incidence of hypotension ranging from 15.8% to 91.4%. Based on a literature review of obstetric and other relevant perioperative and critical care medicine, targeting a mean arterial pressure > 70 mmHg and systolic blood pressure > 100 mmHg is recommended, and it is proposed that the lowest absolute values at which vasopressor therapy should be initiated are a mean arterial blood pressure < 65 mmHg or systolic blood pressure < 90 mmHg. Optimally, practitioners should maintain systolic blood pressure at greater than 90% of the baseline pre-spinal anaesthesia value. Conclusion: This study confirmed a wide variation in the incidence of obstetric spinal hypotension in a South African setting, depending on the definition used. An absolute threshold for intervention with vasopressor and an optimal target relative to baseline blood pressure are suggested. Further work is required to establish the effect of the adherence to these recommendations on important maternal and foetal outcomes

Sample-size determination and adherence in randomised controlled trials published in anaesthetic journals

Background: Sample-size calculations are critical to ensure that randomised control trials return robust and reliable results. The estimated treatment effects used in these calculations is often significantly different from the actual treatment effect and can dramatically impact trial validity. Methods: This study examined sample-size calculations in randomised controlled trials designed to show superiority between two-arm parallel groups with a single primary outcome that were published in the top five anaesthetic journals for 2014 (as per Thomson Reuters impact factors). In particular, it sought to determine treatment effect estimations used in a priori sample-size calculations and compare them with actual treatment effects. Results: A PubMed search identified 209 possible articles; 52 were drawn for full text review; and 28 were included in the final analysis. The relative difference between expected and actual event rates was greater than 20% in 80% of trials and greater than 50% in 44% of trials. Conclusions: Unrealistic assumptions of treatment effects in randomised controlled trials published in anaesthesia journals are common. Trial sample sizes should be calculated thoughtfully and realistically and should be fully reported in both trial protocols and publications. Researchers should be aware of the opportunity cost as well as the possible dangers to patients when unrealistic assumptions are made. Where possible researchers should collaborate to achieve meaningful trial sample sizes to ensure robust clinical findings

Development and validation of the isiZulu quality of recovery score

Background: Recovery from anaesthesia and surgery is an important marker of the quality of perioperative care. One extensively validated score in assessing this is the Quality of Recovery­15 items (QoR-15) score. This study aimed to translate the QoR-15 score into isiZulu and validate both the original and translated version on an isiZulu speaking population. Methodology: A randomised quantitative observational study was performed testing the original and the translated version of the QoR-15 score. In a crossover format, patients were asked to complete both questionnaires with 40 minutes allowed between each questionnaire. A 100 mm visual analogue score (VAS) was completed by each participant as a comparative tool for overall quality of recovery. Results: There was good correlation between the English and isiZulu score 0.91 (p < 0.001) and substantial agreement between the scores (mean weighted kappa: 0.69). There was a negative correlation between duration of surgery and total QoR-15 scores for both the English (­0.3; p < 0.001) and isiZulu (­0.29; p < 0.001) questionnaires, and a positive correlation between VAS scores and total QoR-15 scores for both the English (0.38; p < 0.001) and isiZulu (0.38; p < 0.001) questionnaires. Conclusion: This study demonstrates that the QoR-15 score is suitable to use in an isiZulu speaking patient population. The translated isiZulu version is comparable to the English QoR-15 score and should be used to assess the QoR to improve patient care

A meta-analysis of the efficacy of preoperative surgical safety checklists to improve perioperative outcomes

BACKGROUND:Meta-analyses of the implementation of a surgical safety checklist (SSC) in observational studies have shown a significant decrease in mortality and surgical complications.OBJECTIVE:To determine the efficacy of the SSC using data from randomised controlled trials (RCTs). METHODS:This meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and was registered with PROSPERO (CRD42015017546). A comprehensive search of six databases was conducted using the OvidSP search engine.RESULTS:Four hundred and sixty-four citations revealed three eligible trials conducted in tertiary hospitals and a community hospital; with a total of 6 060 patients. All trials had allocation concealment bias and a lack of blinding of participants and personnel. A single trial that contributed 5 295 of the 6 060 patients to the meta-analysis had no detection; attrition or reporting biases. The SSC was associated with significantly decreased mortality (risk ratio (RR) 0.59; 95% confidence interval (CI) 0.42 - 0.85; p=0.0004; I2=0%) and surgical complications (RR 0.64; 95% CI 0.57 - 0.71; petlt;0.00001; I2=0%). The efficacy of the SSC on specific surgical complications was as follows: respiratory complications RR 0.59; 95% CI 0.21 - 1.70; p=0.33; cardiac complications RR 0.74; 95% CI 0.28 - 1.95; p=0.54; infectious complications RR 0.61; 95% CI 0.29 - 1.27; p=0.18; and perioperative bleeding RR 0.36; 95% CI 0.23 - 0.56; petlt;0.00001.CONCLUSIONS:There is sufficient RCT evidence to suggest that SSCs decrease hospital mortality and surgical outcomes in tertiary and community hospitals. However; randomised evidence of the efficacy of the SSC at rural hospital level is absent