Korean Cancer Survivors Get a Validated Tool to Voice Their Care Priorities

When cancer treatment ends, many survivors describe a strange silence. The appointments thin out, the medical team that once coordinated every scan and prescription recedes, and patients are left to navigate follow-up care, lingering symptoms, and the fear of recurrence largely on their own. Health systems around the world have recognized this gap, and the concept of patient-centered survivorship care has emerged as a response: care organized around what survivors themselves actually prioritize, rather than around what institutions find convenient to deliver. But measuring whether care is truly patient-centered requires a validated instrument, and instruments do not travel across languages and cultures automatically. A new study published in BMC Health Services Research reports the cross-cultural validation of the Korean version of the Patient-Centered Survivorship Care Index, or K-PC-SCI, offering Korean clinicians and researchers a rigorously tested tool for capturing survivors’ own priorities.

The original Patient-Centered Survivorship Care Index was developed from the stated priorities of cancer survivors regarding the care they receive after treatment. It asks respondents to rate items across domains such as health monitoring, coordination of care, information and education, and support for daily living and emotional wellbeing, all framed from the survivor’s perspective. Translating such an instrument is not a matter of dictionary substitution. Health care concepts are embedded in cultural expectations: how patients in Korea relate to physicians, what they consider appropriate follow-up, and how they express unmet needs may differ substantially from the settings where the original tool was built. The research team, led by Ol Han of Ewha Womans University with colleagues from the National Cancer Center and the Dongnam Institute of Radiological and Medical Sciences, therefore undertook a full cross-cultural adaptation before any statistical testing.

The adaptation followed the canonical sequence used in modern instrument validation. First, the source questionnaire underwent forward translation into Korean, followed by expert review to resolve ambiguities and adapt wording to Korean health-system realities. The Korean version was then back-translated into the source language so that the original developers’ intent could be checked against the reverse rendering. A pilot group of survivors completed the draft under cognitive debriefing, a technique in which respondents think aloud and explain how they interpret each item, exposing phrasing that reads naturally to clinicians but confusingly to patients. Finally, a panel assessed content validity item by item. The outcome was striking: of the original 36 items, 34 were retained, and the scale-level content validity index averaged 0.98, a value indicating near-unanimous expert agreement that the items remain relevant and clearly worded in Korean.

With the adapted instrument in hand, the team moved to the psychometric core of the study. They recruited adult cancer survivors from three hospitals across South Korea using convenience sampling, distributing 220 questionnaires and analyzing 198 completed responses. The central question was structural validity: does the seven-factor architecture of the original index, in which each domain represents a distinct facet of patient-centered survivorship care, hold up in the Korean data? To answer it, the researchers used confirmatory factor analysis with robust maximum-likelihood estimation, a method chosen because survivor-reported data often deviate from perfect normality. The prespecified seven-factor simple-structure model served as the primary test, with two additional models examined as sensitivity analyses to probe how the data behaved when plausible refinements were allowed.

The results tell a nuanced story. The prespecified seven-factor model achieved moderate fit, with a robust comparative fit index of 0.913, a Tucker-Lewis index of 0.904, a root mean square error of approximation of 0.075, and a standardized root mean square residual of 0.060. In conventional interpretation, these values sit in the acceptable-to-borderline range: the model captures the broad architecture of the data but leaves room for local misfit. A sensitivity model informed by the original development study, which permitted specific residual covariances suggested by prior research, improved fit modestly, and a further data-informed model with four substantively interpretable within-factor residual covariances improved it again, reaching a comparative fit index of 0.937 and an RMSEA of 0.064. The authors are appropriately cautious here: this most parsimonious refined model fits well but requires independent replication before it can be treated as the definitive structure.

The item-level statistics were reassuring. Standardized factor loadings ranged from 0.70 to 0.94, meaning each item correlates strongly with its intended domain, and average variance extracted ranged from 0.64 to 0.81, indicating that each subscale explains the majority of the variance in its own items. Discriminant validity, assessed with the heterotrait-monotrait ratio, produced values between 0.675 and 0.896, below the thresholds that would suggest problematic overlap between domains, although the upper end of that range hints at partial domain overlap that the residual-covariance analyses also detected. Reliability was strong throughout: subscale Cronbach’s alpha ranged from 0.85 to 0.94, McDonald’s omega from 0.86 to 0.94, and the total scale alpha reached 0.98, an exceptionally high internal consistency for a 34-item multidimensional instrument.

Two findings deserve particular attention from a measurement-science standpoint. First, six items showed ceiling proportions above 40 percent, meaning that large shares of survivors gave the same top response. Ceiling effects compress variability and can obscure differences between patient groups or changes over time, so future users of the K-PC-SCI should interpret those items carefully, especially in samples where care quality is uniformly high. Second, the team examined convergent evidence by relating K-PC-SCI scores to the Korean Cancer Survivors’ Unmet Needs scale, a well-established instrument measuring what survivors feel is missing from their care. The results were illuminating rather than disappointing: K-PC-SCI scores showed essentially no association with the simple count of unmet needs, with a Spearman’s rho of 0.072 and a confidence interval spanning zero, but a small positive association with severity-weighted unmet-needs intensity, rho of 0.162 with a confidence interval excluding zero.

That dissociation is conceptually meaningful. Counting unmet needs treats every gap in care as equal, whereas the PC-SCI asks survivors to evaluate care against their own priorities, so a survivor with few but severe unmet needs may rate patient-centeredness poorly while one with many trivial gaps rates it well. The modest correlation with intensity-weighted unmet needs suggests the two instruments measure related but distinct constructs, which is precisely what one would want from a new tool rather than a redundant copy of an existing one. In other words, the K-PC-SCI appears to capture something the unmet-needs literature has been missing: the survivor’s own weighting of what matters, which is the definitional core of patient-centered care.

The authors are candid about the limits of their evidence. The sample was drawn from three hospitals by convenience sampling, which may not represent the full diversity of Korean cancer survivors, particularly those years past treatment who receive care in community settings rather than specialized cancer centers. The moderate fit of the primary model, the residual local dependence revealed in sensitivity analyses, and the partial overlap among some domains all point to the need for further validation in independent samples, especially post-treatment and community-based cohorts. The study also received no specific external funding, and the authors declare no competing interests, with ethical approval obtained from all three participating institutions and written informed consent from every participant in accordance with the Declaration of Helsinki.

Even with those caveats, the practical significance is considerable. South Korea has one of the world’s most rapidly aging cancer survivor populations, and its national cancer control program has invested heavily in survivorship services. Until now, evaluating whether those services reflect survivors’ own priorities required instruments validated elsewhere. The K-PC-SCI, with 34 items across seven domains, strong reliability, and preliminary structural validity, gives Korean researchers a foundation for national surveys, intervention trials, and quality-improvement programs grounded in what survivors themselves say matters. More broadly, the study is a model of methodological transparency: rather than reporting only the best-fitting model, the team presents the prespecified test, the theory-informed refinements, and the data-driven alternatives side by side, letting readers see exactly where the instrument is solid and where replication is needed. In a field where measurement tools often cross borders with little scrutiny, that kind of honesty is itself a form of patient-centeredness.

Subject of Research: Cross-cultural validation of the Korean Patient-Centered Survivorship Care Index among adult cancer survivors

Article Title: Cross-cultural validation of the Korean version of the patient-centered survivorship care index (K-PC-SCI) among adult cancer survivors

Article References: Han, O., Baek, J. B., Moon, C. S., & Kim, Y. E. (2026). Cross-cultural validation of the Korean version of the patient-centered survivorship care index (K-PC-SCI) among adult cancer survivors. BMC Health Services Research. https://doi.org/10.1186/s12913-026-15859-3

Image Credits: AI Generated

DOI: 10.1186/s12913-026-15859-3

Keywords: cancer survivors, survivorship care, patient-centered care, psychometric validation, cross-cultural adaptation, confirmatory factor analysis, Korea, health services research, unmet needs, questionnaire validation, reliability, content validity