Study design and procedures

A convergent parallel mixed-methods design was employed whereby qualitative and quantitative data were collected concurrently and analysed independently before being integrated to provide a comprehensive understanding of CSs' unmet needs. We developed a three-step framework to guide the selection of context-specific PROMs for CSs. First, survivors' unmet needs were identified through a combination of qualitative and quantitative approaches, including focus group meetings, individual semi-structured interviews and preparatory questionnaires that explored survivorship-related challenges. Second, all expressed needs were systematically linked to the ICF categories using validated linking rules [20, 27,28,29]. Third, existing PROMs (n = 14, identified through a previously published overview of reviews) were evaluated based on their coverage of the identified ICF categories, considering also their psychometric properties [14].

To demonstrate the feasibility of this framework, we applied it in a cross-sectional study conducted at a comprehensive cancer centre located in a large urban area in northern Italy (Azienda USL-IRCCS of Reggio Emilia). This setting, serving approximately 9000 cancer patients annually, provided a cohort of 35 CSs and seven caregivers.

The study protocol was registered on ClinicalTrials.gov (ID NCT06236373).

Participant selection

Adult CSs undergoing follow-up care between April 2023 and January 2024 were recruited from the cancer centre.

Inclusion criteria focused on survivors of any cancer types with both an incidence of ≥ 5% and a five-year survival rate of ≥ 65% in Italy, according to national epidemiological data, who had completed active treatment [30]. Based on these thresholds, the included tumour types were breast, prostate, colorectal, thyroid cancers, myeloma and lymphoma. Caregivers of CSs were also invited to participate to capture relational and contextual factors influencing CSs’ unmet needs and to explore the distinct challenges experienced by those providing informal support during the survivorship phase.

Participants were excluded if they had severe pre-existing cognitive impairments (e.g. dementia, severe intellectual disability, or other neurodegenerative conditions) that precluded providing informed consent or engaging meaningfully in study activities such as focus group meetings and interviews. These exclusions were distinct from cancer-related cognitive difficulties (such as chemotherapy-induced cognitive impairment or ‘chemo brain’), which were not exclusion criteria given their relevance to the survivorship experience. Cases of severe pre-existing cognitive impairment were identified through clinical evaluation by healthcare staff, with support from medical records when needed. Furthermore, participants undergoing concurrent cancer treatments were also excluded. Maintenance therapies were not considered active treatment in this context, provided that participants had completed curative or intensive oncology protocols.

Preparatory questionnaire

Before the qualitative sessions, participants completed a structured questionnaire designed to stimulate reflection on unmet needs related to survivorship. The questionnaire, derived from existing PROMs [14], investigated the frequency and perceived importance of 25 key needs using Likert scales. Participants rated both how often each need was experienced and its impact on their daily life in the previous three and 12 months. The preparatory questionnaire methodology and psychometric evaluation of source PROMs have been previously published [14].

The preparatory questionnaire is attached as Appendix 1.

Focus groups and individual interviews

One week after completing the preparatory questionnaire, participants joined focus groups or individual interviews, depending on availability. Discussions reflected on questionnaire topics but also allowed the emergence of additional unmet needs. Focus groups were organised by cancer type to foster open dialogue among CSs, while caregiver discussions were held separately. The recruitment process was guided by the principle of data saturation, ceasing when no new significant theme or insight emerged from the discussions [31, 32]. The qualitative methodology and full thematic analysis are reported in a separate qualitative study currently under review [33].

Data analyses

Sociodemographic data (age, sex, living situation, education level and employment status) and disease-related factors (cancer type, time since end of treatment and use of maintenance therapy) were analysed using descriptive statistics by a multidisciplinary research team.

Qualitative and quantitative data were analysed independently before being integrated to provide a comprehensive understanding of CSs' unmet needs. Qualitative data from focus group meetings and interviews were analysed thematically to identify the breadth and depth of unmet needs experienced by participants. Quantitative data from the preparatory materials, where participants rated the frequency and perceived importance of listed unmet needs, were analysed descriptively.

Data integration and triangulation

Data triangulation occurred at the interpretation phase, where qualitative themes were compared with quantitative rankings to identify areas of convergence and divergence [34]. This integration allowed for the identification of unmet needs that were both highly impactful on participants' lives (as evidenced through rich qualitative narratives) and perceived as frequent and important (as demonstrated through quantitative ratings). The convergence of findings from both data sources strengthened the validity of the results and informed the subsequent ICF mapping process, ensuring that the framework captured the most salient and impactful unmet needs of the local CS population.

For each unmet need, a Relative Importance Index (RII) was calculated to integrate both frequency and perceived importance, according to the following formula [35]:

$$\text=\frac=\frac_+4_+3_+2_+1_}$$

where w is the weighting assigned by each respondent (from 1 to 5), n is the number of respondents selecting each option, A is the highest weight (5 in this study) and N is the total number of respondents. Principal component analysis with optimal scaling (PRINQUAL procedure of the SAS/STAT module version 15.1, included in SAS System statistical analysis software version 9.4) [36, 37] was used to synthesise rankings based on RII scores for frequency and importance.

Linking process

The data for the analysis were derived from the unmet needs reported by CSs in our local setting. These were collected through their responses to the preparatory questionnaire and from the 'new' needs that emerged during the focus group meetings and individual interviews.

Regarding the data collected through the preparatory questionnaire, the unmet needs encompassed in its 25 items were linked to the ICF by our research group. The linking methodology and results are described in a recent publication [16].

The decision to adopt the ICF as a conceptual framework was based on its unique capacity to provide a comprehensive and standardised lens to understand the multidimensional nature of cancer survivors' unmet needs. Going beyond disease classification, the ICF includes physical, psychological, social, contextual and environmental components of health and functioning, aligning with the complex and often interrelated challenges faced by cancer survivors [38, 39]. This enables the systematic linking of patient-reported experiences with functioning domains, ensuring comparability, transparency and completeness in identifying and categorising unmet needs while facilitating interdisciplinary communication and supporting a biopsychosocial, person-centred approach with enhanced transferability across diverse clinical and cultural contexts [40].

Data collected through focus group meetings and interviews revealed new unmet needs, which yielded meaningful concepts. Two independent reviewers (AC and SC) extracted and linked these concepts to the ICF using the updated linking rules procedure by Cieza et al. [27,28,29]. Disagreements were resolved through discussion and consultation with a third reviewer (LG).

The new unmet needs were merged with those identified through the questionnaire analysis and linked to the most precise and appropriate ICF category (i.e. second or third level), which represented either body structures, body functions, activities, participation, or environmental factors.

Linking analysis and PROMs selection

Once the complete linkage between all the unmet needs identified by our participants and the ICF had been established, our objective was to determine the most comprehensive PROM(s) that covered the majority of these 'core' ICF categories, i.e. the ones best representing the most relevant unmet needs reported by the study participants. This was to determine whether one (or more) suitable PROM was readily available to capture the unmet needs of the local population appropriately. To achieve this objective, we constructed a two-way table comparing the 14 PROMs coverage of the 35 core ICF categories identified through our analysis. Coverage percentages were calculated, and tools were ranked to determine optimal alignment with CSs’ unmet needs.

Application of the framework and results

To test the feasibility and applicability of the proposed framework, a structured methodology was applied to inform the selection of the PROM(s) that optimally align with the unmet needs of local CSs using a cohort of 35 CSs and seven caregivers at a comprehensive cancer centre in Italy. Participants’ characteristics are reported in Table 1.

Table 1 Participants’ characteristics

The process is summarised in Fig. 1, which illustrates the application of the three-step framework and its output in terms of mapped needs and PROM selection.

Fig. 1

Three-step framework for PROM selection based on unmet needs in cancer survivors

Preparatory questionnaire analysis

All participants who completed the preparatory questionnaire (100% retention) subsequently took part in focus group meetings or individual interviews, ensuring continuity in data collection.

Table 2 presents unmet needs ranked by perceived frequency over the preceding three months. The three most frequent unmet needs were managing sexual concerns, treatment side effects and spiritual well-being. Over the preceding year, 40% of needs had decreased in frequency, 36% remained stable and 24% showed an increasing trend.

Table 2 Unmet needs in order of perceived frequency

When assessed by perceived importance, the ranking changed: managing memory and concentration difficulties, exercise-related issues and treatment side effects emerged as top concerns (Table 3). Needs related to sexuality and spirituality, initially highly ranked by frequency, dropped significantly in importance. Table 4 shows the unified ranking obtained by combining frequency and importance assessments through principal component analysis with optimal scaling [36, 37].

Table 3 Unmet needs in order of perceived importanceTable 4 Unmet needs in order of relevance

A full description of the preparatory questionnaire analysis is provided in Appendix 2.

Qualitative data analysis

Thematic analysis of focus groups and interviews identified 44 individual codes, grouped into 16 topic areas and four primary themes: dignity and respect, desire for normality, control over one’s life and existential frailty.

Eight additional unmet needs not captured by the questionnaire were identified:

Needs related to physical exercise

Need to manage sleep problems

Need to rationalise the fear of recurrence

Need for support from and relationships with healthcare professionals

Need to feel ‘normal’

Need for healthcare professionals who communicate more effectively with each other.

Need for services that include the entire family

Need for inpatient and outpatient services that support one’s overall well-being.

A detailed analysis is reported elsewhere [33].

ICF linking

The complete set of unmet needs was systematically linked to existing ICF categories, identifying 35 core ICF domains representing priority areas for survivorship assessment (Table 5).

Table 5 Results of the ICF linkingPROM selection

Fourteen PROMs were evaluated against the 35 identified ICF categories (Table 6). The Cancer Rehabilitation Evaluation System (CARES) [41, 42] demonstrated the highest coverage (94.3%), followed by the Comprehensive Needs Assessment Tool (CNAT) (77.1%) [43, 44].

Table 6 ICF-based cross-analysis of local survivors’ needs and 14 PROMs coverage

This application demonstrates that our framework can effectively align survivors' reported needs with existing measurement tools, ensuring context-specific, evidence-based PROM selection.