Improving Services and Outcomes for Joint Replacement Patients

The primary aim of the study is to identify whether compliance with specific processes is associated with better outcomes.

The study is a longitudinal cohort study to examine outcomes after total hip and knee replacement surgeries. We hypothesise the site and patient level compliance is related to the prevalence of a subsequent complication (composite outcome comprising 30 day mortality, readmission, reoperation or surgical complication).

METHODS This study will select high volume (\>275 cases annually) public and private hospitals across Australia. By involving public and private sites across Australia, the generalisability of the findings is enhanced. Each site will identify a site coordinator to oversee and manage the project at that site. Site coordinators are identified by each site. After hospital discharge, patients are followed up by the investigators.

This is a prospective observational study with projected sample of c.2200 consenting patients from up to 20 high volume hospitals. The site coordinator will screen, approach and consent eligible patients during routine preadmission processes. At the time of consent, the patient will be required to: provide socio-demographic details about themselves, past medical history, complete 2 standard patient reported outcome measures. Contact with patients about the study will be made by the site coordinator and routine hospital staff involved in their acute care. Acute data collection will involve prospective audit of medical records to collect peri- and post- operative acute care process data, client characteristics and outcomes by staff at each site. Random and convenience sampling will be used to identify eligible hospital sites which will remain anonymous.

Patient follow up regarding longer term outcomes will be completed via telephone by study investigators at three additional time points (approximately 35 and 90 days and 1 year post surgery) to collect repeat data as provided by each patient at the time of consent.

Descriptive statistics will be used to profile site level and patient level compliance with key processes of care. Multivariable logistic regression (MLoR) will be used to identify the association between the binary composite outcome measure and patient level compliance. Site and individual patient compliance with evidence based guidelines (where available) or best practice recommendations on key processes and relevant outcomes will be examined (e.g. Venous thromboembolism (VTE) prevention). Sample size calculations have been completed to ensure the analyses have sufficient power. If a useful association between processes and outcomes is identified, cost analyses with specific foci will be undertaken to consider costs for different factors and stages (e.g. acute / post discharge). Secondary analysis will determine the predictors of patient -reported outcomes. Depending on the results of initial data analyses, additional secondary analyses will potentially include analyses at additional time points (beyond initial 35 days) for association between key processes and outcomes, and analysis looking at the association between specific complications and compliance with specific processes.

Interim analysis will be undertaken to assess the level of observed compliance with key processes.

Re-abstraction of acute care data will be undertaken to ensure accuracy and completeness of submitted date from the sites.

Site survey Prior to study commencement at each site, the site coordinators will complete a survey detailing each surgeons protocols for key processes including their VTE and antimicrobial prophylaxes. The extent surgeon protocols align with Australian practice guidelines will be ascertained.

Additional secondary analyses will include, but may not be limited to:

1. a comparison of outcomes between patients who go to inpatient rehabilitation and those who go directly home after surgery. Propensity based scoring and matching will be used to determine if outcomes are better in one of the two groups 2. determination of the predictors of the 'unhappy knee' 12 months post surgery. A case controlled matched design will be used. 3. comparison in satisfaction with the acute care experience between private and public patients in a subset 4. a comparison of outcomes between THR patients with an anterior surgical approach and a posterior approach 5. description of ambulation profiles immediately post surgery and variables associated with early ambulation 6. snapshot of rehabilitation received after TKA or THA