Summary of studies included in this analysis
| Authors, publication year | Comparison | Joint | Study type | Country |
| Enhanced recovery pathway | ||||
| Brunenberg et al, 200537 | Joint Recovery Programme (pre-assessment and intensive rehabilitation), vs conventional care | Hip and knee | Trial-based | The Netherlands |
| Larsen et al, 200938 | Accelerated perioperative care and rehabilitation, vs conventional care | Hip and knee | Trial-based | Denmark |
| Preoperative | ||||
| McLawhorn et al, 201639 | Bariatric surgery followed by TKA 2 years later, vs immediate TKA | Knee | Markov model | USA |
| Fernandes et al, 201740 | Supervised neuromuscular exercise and an educational package, vs educational package alone | Hip and knee | Trial-based | Denmark |
| Courville et al, 201241 | Preoperative nasal screening for Staphylococcus aureus colonisation followed by mupirocin treatment for patients with positive cultures, vs empirical treatment of all preoperative patients with mupirocin, vs standard infection prevention measures without S. aureus screening or mupirocin decolonisation | Hip and knee | Decision tree model | USA |
| Intraoperative | ||||
| Jackson et al, 200042 | Postoperative erythrocyte recovery and transfusion, vs usual transfusion practice | Hip and knee | Markov model | USA |
| Ramkumar et al, 201843 | Single-dose intravenous tranexamic acid, vs single-dose intravenous aminocaproic acid, vs no pharmacologic haemostatic agent | Hip and knee | Decision-tree model | USA |
| Sonnenberg, 200244 | Autologous blood donation and transfusion, vs usual practice without autologous donation | Hip | Markov model | USA |
| Marques et al, 201545 | Intraoperative local anaesthetic wound infiltration administered before wound closure in addition to standard anaesthesia, vs standard anaesthesia | Hip and knee | Trial-based | UK |
| Cummins et al, 200946 | Antibiotic-impregnated bone cement, vs conventional cement | Hip | Markov model | USA |
| Graves et al, 201647 | Nine arms, comparing combinations of prophylactic systemic antibiotics, antibiotic-impregnated cement, laminar airflow and body exhaust suits | Hip | Markov model | UK |
| Merollini et al, 201348 | No antibiotic prophylaxis, antibiotic prophylaxis and antibiotic-impregnated cement and antibiotic prophylaxis and laminar airflow, each compared with a baseline strategy of routine antibiotic prophylaxis | Hip | Markov model | Australia |
| Nherera et al, 201749 | Single-use negative pressure wound therapy dressings, vs usual care | Hip and knee | Decision-tree model | UK |
| Postoperative | ||||
| Fusco and Turchetti, 201650 | 10 face-to-face rehabilitation sessions plus 10 telesessions, vs 20 face-to-face rehabilitation sessions | Knee | Markov model | Italy |
| Kauppila et al, 201151 | Multidisciplinary biopsychosocial outpatient rehabilitation programme, vs conventional orthopaedic care | Knee | Trial-based | Finland |
| Smith et al, 201852 | Telephonic health coaching and financial incentives vs telephone calls conveying general health messages | Knee | Markov model | USA |
| Bolz et al, 201053 | 2-yearly routine follow-up vs follow-up at 3 months and 1 or 2 years, vs no follow-up | Hip | Markov model | Australia |
TKA, total knee arthroplasty.