The study, published as an open-access review in the journal Supportive Care in Cancer, evaluated the NUH Prehabilitation Service, a multimodal programme combining exercise, nutritional support and psychological preparation. Rather than offering the same package to everyone, the service follows a tiered, risk-stratified model endorsed by Macmillan Cancer Support, in which patients are allocated to one of three pathways according to clinical need. High-risk patients, identified using factors such as frailty, functional capacity, comorbidity and treatment complexity, were assigned to the Specialised pathway, attending supervised one-to-one or two-to-one exercise sessions with the hospital’s prehabilitation team up to three times a week. Lower-risk patients joined the Targeted pathway, attending community-based group classes led by an external provider that combined modified high-intensity interval training with strength exercises. A third group, the Universal pathway, received an individualised home exercise prescription with weekly check-ins for education and advice — an arms-length offer designed for patients who were already independently active or lived outside the immediate area.

The scale of the programme is striking. Of 1,961 patients referred over the 38-month period, 1,720 were included in the analysis, spanning seven surgical specialities including colorectal, gynaecological oncology, hepatopancreatobiliary, upper gastrointestinal, urology, head and neck, and thoracic surgery. That equates to an average of around 52 eligible referrals per month, a level of uptake the authors interpret as strong engagement from referring clinical teams. The cohort had a median age of 67 years, slightly more women than men, and nearly half were classified as obese. Colorectal surgery accounted for the largest share of participants at 38.5 per cent, followed by gynaecological oncology at 20.3 per cent.

To measure whether the programme worked, the team relied on validated functional and psychological instruments collected as part of routine clinical monitoring. Aerobic capacity was assessed with the Incremental Shuttle Walk Test, an externally paced protocol in which walking speed increases with each audio-cued level until the participant can no longer keep up. Lower-limb strength and endurance were measured with the 60-second sit-to-stand test, and grip strength was recorded with a hand-held dynamometer. Psychological well-being was tracked using two widely used self-report questionnaires: the Patient Health Questionnaire-9 for depression and the Generalised Anxiety Disorder-7 scale for anxiety. Self-reported weekly physical activity and the number of completed strength sessions rounded out the behavioural picture.

The results showed statistically significant improvements across every domain measured. On average, patients walked 57 metres further on the shuttle walk test after the programme, a gain with a moderate effect size that exceeds the minimal important differences of 35 to 53 metres typically reported for rehabilitation in long-term conditions. Sit-to-stand performance improved by six repetitions, roughly double or triple the threshold generally considered clinically meaningful. Depression scores fell by about two points, meeting the minimal clinically important difference established in UK primary care cohorts, while anxiety scores also decreased significantly. Perhaps most dramatically, self-reported weekly physical activity rose by 142 minutes — more than doubling — and the number of weekly strength sessions increased by 2.4, both with large effect sizes.

Importantly, these gains were not confined to the most intensive pathway. Although patients in the Specialised group showed greater improvements in depression scores compared with the Targeted group, and the Universal group increased strength sessions more than the Targeted group, most outcomes — including aerobic capacity, anxiety, grip strength and overall physical activity — showed no statistically significant differences between pathways. For the shuttle walk test, gains exceeded the minimal important difference in all three groups: 43 metres in Universal, 56 metres in Targeted and 64 metres in Specialised. The authors suggest this indicates that a tiered model can deliver meaningful benefit across all starting levels, and that even a low-intensity, arms-length offer may be a viable and scalable option for lower-risk patients when service capacity is constrained.

The findings align closely with other NHS prehabilitation programmes. The Greater Manchester Prehab4Cancer programme, one of the best-known community-based models, previously reported shuttle walk gains of around 50 metres and sit-to-stand improvements of 4.8 repetitions — figures comparable to those seen in Nottingham. The consistency across services strengthens the case that prehabilitation can be delivered at scale within publicly funded healthcare systems, not only in tightly controlled research trials. The Nottingham evaluation is also notable for reflecting routine clinical care rather than a research-driven intervention, which the authors argue enhances its real-world relevance and supports generalisation across a broad cancer population.

Not everything translated into clinically meaningful change. Although anxiety scores fell statistically significantly, from an average of about 6 to 4, the change fell short of the established minimal clinically important difference of four points on the GAD-7 scale, and baseline anxiety in this cohort was already low. The authors also note that research in oncology populations suggests the conventional GAD-7 threshold of 10 may underestimate anxiety in cancer patients, with better sensitivity at cut-offs of around 7 to 8, which calls for cautious interpretation of absolute scores. Depression, by contrast, showed a change that met the clinical significance threshold, and the greater improvement seen in the Specialised pathway hints that structured, individualised psychological support may add value over more generic delivery.

The evaluation is candid about its limitations. The analysis used a complete-case approach, excluding 241 patients with missing data, and assumes that data were missing completely at random — an assumption that, if wrong, could bias effect estimates. The observational design lacks randomisation, so causal inference is limited, and between-group comparisons were unadjusted for baseline differences and are therefore exploratory. Physical activity was self-reported, leaving results vulnerable to recall and social desirability bias. Selection bias may also operate upstream: referral depended on clinician discretion, tumour type and service capacity, and patients allocated to lower-intensity pathways may have been healthier or more motivated to begin with. Notably, the evaluation did not capture postoperative outcomes such as length of stay or complication rates, so it remains unknown whether the preoperative gains translated into downstream surgical benefits — a gap the authors flag as a priority for future data collection.

Engagement data reveal the practical friction points that any national rollout would need to address. A total of 8,250 service sessions were delivered, averaging 4.9 per patient, with median time in the service of 33.4 days — long enough, the authors note, to sit above the three-week threshold associated with improvements in postoperative complications, though benefits can begin to emerge within two weeks. Documented reasons for non-completion included short windows before surgery, non-attendance and discharge after assessment, often reflecting patient preference or a perception of already being fit enough. Reported barriers included travel distance, competing hospital appointments and caring responsibilities. Among 662 patients who completed a three-month postoperative questionnaire, 83.3 per cent agreed that prehabilitation helped them prepare for surgery and 61.1 per cent felt it supported their recovery — although only 31.2 per cent reported being more physically active afterwards, while 38.1 per cent felt less active, underscoring how difficult it is to sustain behaviour change through the surgical period itself.

The authors recommend several service improvements: routinely recording hospital length of stay to strengthen benchmarking and the health-economic case, and mitigating engagement barriers through hybrid delivery, digital follow-up and clearer clinician-led communication about the benefits of prehabilitation. They conclude that a multimodal, tiered prehabilitation service aligned with national guidance was linked to improvements in mood, strength and functional capacity, supporting patients’ surgical preparation and recovery, and that the findings justify integrating prehabilitation into routine cancer care. With 45 million cancer surgical procedures expected annually worldwide by 2030, and systematic reviews suggesting multimodal programmes may shorten hospital stays by around two days while proving cost-effective in a majority of economic evaluations, the Nottingham experience offers a template for how health systems might prepare a growing surgical population — not in the operating theatre, but in the crucial weeks before it.

Subject of Research: Multimodal prehabilitation before cancer surgery delivered through tiered NHS pathways

Article Title: Delivering prehabilitation in cancer surgery: a service evaluation at Nottingham University Hospitals

Article References: Alharbi, A., Haywood, A., & O’Connor, D. (2026). Delivering prehabilitation in cancer surgery: a service evaluation at Nottingham University Hospitals. Supportive Care in Cancer, 34(10), Article 973. https://doi.org/10.1007/s00520-026-11113-y

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11113-y

Keywords: prehabilitation, cancer surgery, NHS, exercise, risk stratification, functional capacity, depression, anxiety, physical activity, service evaluation, Nottingham University Hospitals, Supportive Care in Cancer