The study, published in Supportive Care in Cancer, enrolled 267 patients who underwent bladder perfusion at Shanghai General Hospital between August 2021 and January 2024. Bladder perfusion, in which therapeutic agents such as chemotherapy drugs are instilled directly into the bladder, is a mainstay of treatment for non-muscle-invasive bladder cancer and related conditions. Although the procedure avoids many of the systemic side effects of intravenous chemotherapy, it imposes a demanding schedule of hospital visits, catheterizations, and symptom management, and readmission remains a common and costly part of the patient journey.

To capture how patients were really doing, the team used the SF-36 questionnaire, one of the most widely validated instruments in outcomes research. The SF-36 measures health across ten domains, including physical functioning, role limitations due to physical problems, bodily pain, general health perceptions, vitality, social functioning, role limitations due to emotional problems, and mental health. These domains can be aggregated into two summary scores: the Physical Component Summary, or PCS, and the Mental Component Summary, or MCS. Patients completed the survey one week after the induction of bladder perfusion, a timepoint the researchers labeled T0, and again eight weeks later, labeled T1.

The researchers then tracked readmissions over a follow-up window of fourteen months. The results were striking: 131 of the 267 patients were readmitted, an incidence rate of 49.1 percent, with a median time to readmission of 7.2 months, and an interquartile range spanning 4.2 to 10.9 months. In other words, nearly half of the cohort returned to the hospital within roughly a year of starting treatment, underscoring the substantial burden that bladder perfusion patients continue to carry after their initial therapy.

With the readmission data in hand, the team constructed Cox proportional hazards regression models, the standard statistical framework for analyzing time-to-event outcomes. Three separate models were built: one using the T0 quality of life scores, one using the T1 scores, and one using the change scores between the two timepoints. This longitudinal design is what distinguishes the study from much of the existing literature, which typically relies on a single snapshot of quality of life. By modeling both the level and the trajectory of patient-reported health, the researchers could ask not just whether patients felt good at one moment, but whether changes in how they felt carried prognostic information of their own.

The answer, in short, was yes, and the details were revealing. In the T0 model, four of the ten SF-36 domains showed statistically significant associations with time to readmission: the Physical Component Summary, physical functioning, role-physical, and general health, all at P values below 0.05. In the T1 model, the picture broadened considerably, with eight of the ten domains reaching significance, everything except bodily pain and mental health. In the change-score model, six domains were significant: the Physical Component Summary, physical functioning, role-physical, general health, role-emotional, and social functioning. The pattern across the three models was consistent: post-treatment quality of life at eight weeks showed the strongest associations with readmission timing, followed by the change between the two assessments, while the pre-treatment baseline showed the fewest significant links.

Perhaps the most consequential finding concerned the divide between physical and psychological indicators. Across all three models, the physiological dimensions, anchored by the Physical Component Summary and its component domains, were significantly more strongly associated with time to readmission than the psychological dimensions captured by the Mental Component Summary. The differences were statistically significant in each model, with P values of 0.023, less than 0.001, and 0.017 for the T0, T1, and change models respectively. This suggests that in this population, the body’s report on its own condition, the ability to walk, to work, to perform daily tasks, and to perceive one’s general health, carries more prognostic weight than mood or emotional functioning alone.

The authors are careful to frame these findings as hypothesis-generating rather than definitive. Observational designs cannot establish that improving quality of life would directly delay readmission, and the study did not incorporate the full range of clinical variables, such as tumor characteristics, treatment intensity, and comorbidity burden, that would be needed to test whether quality of life adds predictive value beyond established clinical risk factors. The researchers explicitly call for further investigation in studies with more comprehensive clinical data. Still, the consistency of the associations across three separate modeling approaches, and the coherent pattern of physical dominance, lends weight to the central claim that longitudinal quality of life measurement carries real prognostic signal.

The practical implications are nonetheless significant. If quality of life scores collected during bladder perfusion can flag patients at risk of early readmission, clinicians could target supportive interventions, from symptom management and physical rehabilitation to closer follow-up scheduling, at those who need them most. The finding that post-treatment scores at eight weeks are the most informative suggests a concrete operational window: a routine SF-36 assessment at that timepoint could serve as a triage tool, identifying patients whose declining physical function warrants intensified monitoring. This aligns with a broader movement in oncology toward systematic patient-reported outcome monitoring, exemplified by trials showing that electronic symptom tracking during cancer treatment can improve outcomes.

The study also fills a gap in a literature that has largely focused on survival rather than readmission. Previous work has established that health-related quality of life predicts survival in cancer populations, including patients with lung and bladder cancers, and that bladder cancer and its treatments impose lasting burdens on physical function and daily life. By extending this logic to time to readmission, a metric that matters to patients, families, and health systems alike, the Shanghai team connects the subjective experience of treatment with a hard operational outcome. For a therapy that requires patients to return repeatedly for instillations, the distance between scheduled visits and unscheduled readmissions is where much of the disease burden lives, and the new results suggest that listening carefully to how patients describe their own health, especially their physical health, may help clinicians extend that distance.

Subject of Research: Longitudinal health-related quality of life as a predictor of time to hospital readmission in bladder perfusion patients

Article Title: Longitudinal data on health-related quality of life and its association with time to readmission in bladder perfusion patients

Article References: Zhang, M., Chen, Z., Chen, Y., & Zhou, Y. (2026). Longitudinal data on health-related quality of life and its association with time to readmission in bladder perfusion patients. Supportive Care in Cancer, 34(10), Article 1032. https://doi.org/10.1007/s00520-026-11243-3

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11243-3

Keywords: bladder perfusion, health-related quality of life, readmission, SF-36, non-muscle-invasive bladder cancer, Cox proportional hazards, patient-reported outcomes, supportive care, longitudinal study, prognosis, intravesical therapy, nursing research