An international, multicentre, African cluster randomised trial.

Primary objective :

To determine whether increased postoperative surveillance in high-risk adult surgical patients reduces overall in-hospital mortality in surgical patients aged 18 years and over in Africa.

Primary outcome measure:

In-hospital mortality, censored at 30 days if the patient is still alive and in-hospital.

Secondary objective:

To determine whether increased postoperative surveillance in high-risk adult surgical patients reduces the overall incidence of the composite of severe in-hospital complications and mortality in adult surgical patients aged 18 years and over in Africa.

Secondary outcome measure:

Composite of severe in-hospital complications and mortality, censored at 30 days if the patient is still alive and in-hospital.

Study design:

ASOS-2 is an African, international, multicentre, cluster randomised trial.

Inclusion criteria:

• Patients: All consecutive adult patients aged 18 years and over admitted to participating centres undergoing elective and non-elective surgery
• Participating surgical centres: Randomised according to a stratification based upon the level of the surgical facility and the surgical case load. Recruitment will start in May 2019.

Exclusion criteria:

*Prior participation in ASOS-2

Recruitment and screening:

This is a pragmatic trial. It is an African, international cluster randomised controlled trial in several African countries. Participating surgical sites will be randomised to either increased postoperative surveillance or usual postoperative care. The investigators expect all consecutive adult patients aged 18 years and over admitted to participating centres undergoing elective and non-elective surgery to be included in the trial. 'Broadcasting' through appropriate hospital notices and signage will inform the patients and the public that the hospital is participating in the cluster randomised trial.

Ethics approval:

The requirement for patient consent is expected to vary according to regulations of the participating nations. The national leaders will ensure ethics approval is obtained from their respective countries and centres. Centres will not be permitted to record data unless ethics approval or an equivalent waiver is in place.

Randomisation:

Participating sites will be randomised to normal postoperative care, or increased postoperative surveillance. Randomisation will be stratified according to the level of the surgical facility and the expected weekly surgical case-load.The investigators plan to randomise 664 hospitals to either increased postoperative surveillance or standard care for high-risk adult (≥18 years) surgical patients. The follow up is in-hospital.

Trial intervention:

The intervention arm to which each participating site is randomised will be offered to all eligible surgical patients for the duration of the trial.

Intervention arm:

Participating sites which have been randomised to increased surveillance will need to provide increased surveillance to surgical patients with a predicted increased postoperative risk as determined by the ASOS risk stratification tool. Increased postoperative surveillance can include either of the following; i) admission to a higher care ward than had been planned at the time of surgery, ii) an increase in the frequency of nursing observations in the postoperative period, iii) ensuring that the patient is placed in view of the nursing station, and not in a remote location in the postoperative ward, or iv) allowing family members to stay with the patient in the ward in the postoperative period. The nature of the offered increased postoperative surveillance will be left to the discretion of the healthcare workers and the participating sites. However, all sites will be encouraged to include more than one of the increased postoperative surveillance intervention. The healthcare providers will also receive information on the leading causes of postoperative mortality in African surgical patients as documented in ASOS; surgical site infections, bloodstream infection and acute respiratory distress syndrome, pneumonia, acute kidney injury, postoperative bleeding, and cardiac arrest. This will be known as the 'Postoperative surveillance bedside guide' and will be placed at the bedside of every patient flagged as high-risk by the ASOS Surgical Risk Calculator. This should be placed in a visible position at the patient's bedside e.g. posted on the wall above the patient's bed.

Control arm:

Participating sites randomised to the control arm will provide usual postoperative care to patients. The care will be left to the discretion of the healthcare providers.

Data collection and collation:

Centre specific data will be collected once for each hospital including: university or non-university hospital, number of hospital beds, number of operating rooms, number and level of critical care beds and details about the reimbursement status of the hospital. An ASOS-2 case record form (CRF) will be completed for every eligible patient who undergoes surgery during the trial. Patients will be followed up until hospital discharge. This will be censored at thirty days i.e. patients will be followed up until discharge or for thirty days whichever is the shorter period.

Follow up procedures:

Follow-up data will be collected by a site trial investigator. Investigators will review a participant's in-hospital medical records (paper or electronic) up to hospital discharge.

Sample size calculation:

This is a cluster randomised trial of hospitals in Africa. The investigators will match hospitals on expected surgical volume in a week of surgery. This varied tremendously across the ASOS group; with a median number of surgical procedures per hospital for the study week in ASOS of 29 (IQR 10-71). The variability of the individual patient outcomes explained by the cluster (or surgical site) is taken into account in these sample size calculations. The inter-cluster correlation coefficient (ICC) in ASOS was 0.01. For the sample size calculation, the investigators have therefore used a conservative ICC of 0.015.

Based on ASOS, the investigators estimate the coefficient of variance for a 4-week recruitment period to be 0.63. A 4-week recruitment period is defined as the following; i) hospitals stop recruiting at the end of the week in which they have exceeded 100 enrolled patients, and ii) if a site had not reached 100 enrolled patients after 4 weeks of recruiting, it would stop recruiting.

The incidence of mortality in ASOS was 2.1%. The investigators expect a 25% relative risk reduction in mortality through increased surveillance of postoperative surgical patients at high-risk of severe complications or in-hospital mortality. Based on the intra-cluster correlation coefficient (ICC) for the composite of severe complications and mortality in ASOS-2 of 1.5%, a coefficient of variance of 0.63, and stratification for the level of the surgical facility, and the volume of procedures per week, a trial for efficacy of increased postoperative surveillance would require 64,200 patients, from 642 surgical centres across Africa.

Statistical analysis:

Outcomes will be presented at a continental level. All institutional level data will be anonymised prior to publication. Categorical variables will be described as proportions and will be compared using chi-square tests. Continuous variables will be described as mean and standard deviation if normally distributed or median and inter-quartile range if not normally distributed. Comparisons of continuous variables between groups will be performed using t-tests, one-way ANOVA or equivalent non-parametric tests as appropriate.

The primary outcome will be in hospital mortality. Overall differences in in-hospital mortality will be compared between the intervention and control clusters. All analyses will account for clusters. A list of baseline risk factors (the risk factors in the ASOS Surgical Risk Calculator) will be included in the analysis. The investigators will use logistic regression model to estimate the effect of increased postoperative surveillance, on the primary and secondary outcomes. The investigators will calculate the odds ratios and their associated 95% confidence intervals. The investigators will infer statistical significance if the computed 2-sided p-value is < 0.05. A single final analysis is planned at the end of the study.

Subgroup analyses:

An a priori decision has been taken to conduct the following subgroup analyses; i) between high- and low-outcome hospitals, ii) between district hospitals and higher-level hospitals, iii) between low- and high-human resourced surgical hospitals, iv) between low-income countries and middle-income countries, and v) between the different types of postoperative surveillance.

Secondary studies: The use of the ASOS-2 Trial data for secondary studies will be encouraged.

Monitoring and auditing:

The Sponsor will have oversight of the trial conduct at each site. The trial team will take day-to-day responsibility for ensuring compliance with the requirements of GCP in terms of quality control and quality assurance of the data collected as well as safety reporting. The ASOS-2 Trial Management Group will communicate closely with individual sites and the Sponsor's representatives to ensure these processes are effective.

Inclusion Criteria:

• All consecutive adult patients aged 18 years and over admitted to participating centres undergoing elective and non-elective surgery

Exclusion Criteria:

• Prior participation in ASOS-2