Clinical pathways before total knee replacement in England: a population-based retrospective cohort study using sequence analysis and multilinked electronic health records

Background

The UK National Institute for Health and Care Excellence recommends a structured care pathway for osteoarthritis; however, real-world practice often diverges from these guidelines, and the patient journey through primary care before knee replacement is not well described. We aimed to characterise real-world care pathways before total knee replacement and to identify patient characteristics associated with different patterns of care using sequence analysis and linked electronic health records.

Methods

In this population-based retrospective cohort study, we identified patients aged 45 years or older who underwent primary total knee replacement between Jan 1, 2007, and Dec 31, 2021, from the Clinical Practice Research Datalink Aurum database, linked to Hospital Episode Statistics data in England, UK. Sequence analysis was applied to classify health-care patterns in the 10 years preceding total knee replacement, based on care strategies such as consultations, analgesic prescriptions, referrals, and knee surgeries. Multinomial logistic regression was used to determine the associations between patients’ characteristics and care pathways.

Findings

We included data for 129 470 patients in the analysis (mean age 69·7 years [SD 9·4]; 73 212 [56·5%] were female, 56 258 [43·5%] were male). Sequence analysis identified five distinct care pathways: low health-care use (48 922 [37·8%] of 129 470 patients), moderate health-care use (30 947 [23·9%] patients), opioids predominant (25 983 [20·1%] patients), gabapentinoids and antidepressants predominant (5910 [4·6%] patients), and non-steroidal anti-inflammatory drugs predominant (17 708 [13·7%] patients). Patients from the most deprived areas had higher odds of being in the opioids predominant (odds ratio 3·86, 95% CI 3·57–4·18) and gabapentinoids and antidepressants predominant (3·50, 3·07–3·99) clusters, relative to the low health-care use cluster. In addition, female sex, age above 75 years, BMI of 25·0 kg/m2 or higher, multiple comorbidities, and being Black or South Asian ethnicity (vs White ethnicity) were significantly associated with the odds of membership in all other clusters relative to the low health-care use cluster.

Interpretation

The identification of distinct and non-guideline-concordant pathways highlights persistent gaps in primary care management of patients with osteoarthritis. These findings underscore the need for targeted primary care interventions to support timely, evidence-based, non-pharmacological management and to promote more consistent and equitable care before knee replacement.

Funding

National Natural Science Foundation of China, Henan Key Research and Development Special Project, and National Institute for Health and Care Research Applied Research Collaboration West Midlands.