Joint replacement surgery is a very effective treatment. Hip replacement, for example, brings pain relief and improved function to 70000 UK patients with hip arthritis each year.
No surgery is without risk and the risks during and after surgery are increased in elderly patients and those with other health conditions. Weighing up risks and benefits, and making the decision whether or not to have an operation is more difficult when there are other health problems. There is, however, no strong evidence to best advise patients. As a consequence we don't know if the 'right' decision is made.
The proposed research will use a large general practice database of patients identified as having arthritis, including both 'wear-and-tear' arthritis and autoimmune arthritis. Using the database the researchers will investigate levels of other existing health problems in the patients and analyse how they informed the decision for any subsequent joint surgery. Those who received surgery will be followed up using both general practice and hospital data sources to determine if their pre-existing other health conditions actually did affect outcomes from surgery.
The goal of the research is to provide information to better support decision making for patients and those who advise them.
Changes to the UK population demographic have led to a rise in the number of total hip replacement (THR) and total knee replacement (TKR) procedures performed on elderly patients. These patients have an increasing level of comorbidity, which can constrain the benefits of surgery and increase intra-operative and post-operative risks. At present there is limited evidence on the effect of pre-existing comorbidites on benefits and risks from THR and TKR surgery, making decision-making on whether to proceed with surgery difficult. This project aims to investigate the relationship between comorbidity and the likelihood of a patient with arthritis being referred for arthroplasty, and, if surgery is undertaken, the short and medium term outcomes.
Using the Clinical Practice Research Datalink and Hospital Episode Statistics databases, we will interrogate the records of over 100,000 patients aged >65 with a coded GP diagnosis of osteoarthritis or rheumatoid arthritis to assess their baseline levels of comorbidity using a range of metrics, the time taken between diagnosis and surgery, and key outcomes of surgery. Statistical techniques, including survival modelling and multiple regressions, will be used to describe the associations between the outcomes. We hope our results will prove useful to patients and clinicians.
First objective:
The primary outcome is validity of the CPRD electronic diagnostic code for hip osteoarthritis.
- To determine whether the diagnostic code is valid, we will seek further information from a sub-group of 100 patients, who will be randomly selected by CPRD from the larger cohort. A questionnaire has been designed, which will be sent to the GPs of the 100 selected patients. The GPs will be requested to complete the questions based on information available in each patient's electronic primary care notes. The questions will relate to the NICE diagnostic criteria for hip osteoarthritis, with questions focusing on symptoms of pain, results of X-rays and referral to secondary care. Depending on the information gained from the questionnaire, each patient's diagnosis will be deemed valid or invalid. If the positive predictive value of the diagnostic codes is >75% (i.e. over 75% of patients have an accurate diagnosis), then we will conclude the diagnostic code is valid and continue with objectives two and three.
Second objective:
The primary outcomes are
(i) Dichotomous outcome of referred or not referred to secondary care (to secondary care for orthopaedic opinion of their diseased joint)
(ii) Time from GP diagnosis of osteoarthritis to referral to secondary care for orthopaedic opinion
(iii) Time from referral to secondary care to receipt of surgery (THR/TKR)
(iv) Time from GP diagnosis of osteoarthritis to receipt of surgery (THR/TKR)
Data on these outcomes will be extracted from CPRD
Third objective:
The primary outcome is complication rate in the 90 days following THR or TKR surgery. Included complications are:
- Death
- Venous thromboembolism
- Myocardial infarction
- Stroke
- Anaemia
- Respiratory, urinary tract and wound infections
The secondary outcomes are:
- PROMs (Oxford hip score, Oxford knee score, EQ-5D Index and EQ-VAS) (raw score and improvement from pre-op score at 24 months post-operatively)
- Length of hospital stay (for admission for THR/TKR)
- Early (90-day) re-admission rate
- Mortality rate
- Re-operation rate
- Revision surgery rate
Daniel Prieto-Alhambra - Chief Investigator - University of Oxford
Rory Ferguson - Corresponding Applicant - University of Oxford
Alan Silman - Collaborator - University of Oxford
Andrew Judge - Collaborator - University of Oxford
Christine Walker - Collaborator - Keele University
Christopher Edwards - Collaborator - St Woolos Hospital
Cyrus Cooper - Collaborator - University of Southampton
Dahai Yu - Collaborator - Keele University
Edward Burn - Collaborator - University of Oxford
George Peat - Collaborator - Keele University
John Griffiths - Collaborator - Oxford University Hospitals
Jose M Valderas - Collaborator - University of Exeter
Kelvin Jordan - Collaborator - Keele University
Nigel Arden - Collaborator - University of Oxford
Rafael Pinedo-Villanueva - Collaborator - University of Oxford
Rory Ferguson - Collaborator - University of Oxford
Samuel Hawley - Collaborator - University of Bristol
Sion Glyn-Jones - Collaborator - University of Oxford
HES Admitted Patient Care;HES PROMS (Patient Reported Outcomes Measure);ONS Death Registration Data;Patient Level Index of Multiple Deprivation