Development and validation of a dual language needs assessment tool for people living with colorectal cancer (NeAT-CC)

TL;DR

This study created and validated a bilingual tool to assess the needs of colorectal cancer patients in multi-ethnic, low- and middle-income settings.

Contribution

The NeAT-CC is a new, validated, dual-language needs assessment tool tailored for colorectal cancer patients in diverse socio-cultural contexts.

Findings

The 48-item NeAT-CC covers six domains of needs with good internal consistency and validity.

The tool demonstrated strong criterion validity when correlated with quality of life measures.

NeAT-CC is user-friendly, taking 15–20 minutes to complete and suitable for self-administration.

Abstract

Needs assessment tools may guide optimization of clinical services to be more patient-centered. As needs of patients living with and beyond colorectal cancer (CRC) may also be influenced by socio-cultural backgrounds and healthcare ecosystems, we developed and validated a needs assessment questionnaire for CRC in a multi-ethnic, low-and middle-income setting.

The study methodology was guided by the COSMIN checklist. Items generation was based on findings from independent qualitative inquiries with patients, input from cancer stakeholders, and literature review. Following translation into Malay language, content and face validation were undertaken. The tool was administered to 300 individuals living with and beyond CRC. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were performed. Criterion validity was assessed using EORTC QLQ-C30 and QLQ-CR29 questionnaires.

The 48-item bilingual needs assessment tool for colorectal cancer (NeAT-CC) encompassed six domains of needs, namely (i) diagnosis, (ii) psychosocial and information, (iii) healthcare, (iv) practical and living with cancer, (v) financial and (vi) employment. Cronbach’s alpha was above 0.70 for all domains, indicating good internal consistency. CFA also demonstrated acceptable convergent and divergent validity with composite reliability >0.70 and Heterotrait–Monotrait index <0.90 for all constructs. Criterion validity was established given the significant correlation with quality of life. The NeAT-CC was easily understandable, took 15–20 minutes for completion and may be self-administered.

Utilization of NeAT-CC may facilitate optimization of supportive and survivorship care services following CRC in local settings. The tool has wider potential for adaptation in other multi-ethnic and/or low and middle-income settings.

Figures

Confirmatory factor analysis model of the needs assessment tool for colorectal cancer (NeAT-CC) using partial least square-structural equation modelling.

Full text

1. Introduction

Recent global estimates suggest that survival following colorectal cancer (CRC) has been increasing in many parts of the world, including in some low and middle-income countries such as Malaysia [1]. The rise in number of patients who are surviving colorectal cancer (CRC) for extended periods of time has led to increased recognition of the importance of understanding and assessing their specific care needs [2]. A driving impetus here is the growing global movement for delivery of patient-centered care, which emphasizes that the needs, values, and preferences of patients are consistently taken into the planning of care delivery across the cancer continuum [3].

Notably, prior evidence suggests that patients whose care needs are met tend to exhibit higher self-efficacy and self-management, greater adherence to health promotion approaches and experience less suffering and lower symptom burden, leading to improved overall wellbeing [4,5]. From the health system’s perspective, detailed knowledge about the issues and concerns of people living with and beyond CRC, their care needs and the extent to which these are met by current services are critical in guiding the allocation of scarce healthcare resources, and planning the delivery of high-quality supportive care and survivorship care.

Commonly used tools such as the Supportive Care Need Survey (SCNS-SF) [6], Cancer Survivors’ Unmet Needs Measure (CaSUN) [7], Survivors’ Unmet Needs Survey (SUNS) [8], and the Comprehensive Needs Assessment Tool in Cancer (CNAT) developed in Korea [9] were reviewed. However, most of the currently available needs assessment tools for cancer were developed in affluent settings [9–11] with more advanced healthcare systems that may not necessarily be applicable to resource-limited settings [9]. Apart from clinical factors and healthcare ecosystems, care needs may also be influenced by the patient’s sociocultural background, particularly in multi-ethnic settings [9,12,13]. Prior reviews also suggest that many of the existing tools do not comprehensively cover all domains of needs following cancer such as social needs, or do not demonstrate adequate psychometric properties [14]. While certain needs following CRC have been reported to be similar with other cancers, there are some unique challenges, such as dealing with stoma that necessitates development of a disease-specific tool [14,15]. Needs assessment tools tailored to these unique challenges are essential to ensure that survivors’ concerns are effectively identified and addressed [14].

In sum, there are compelling justifications for development of a more comprehensive tool to measure needs following CRC in diverse populations. We sought to develop a dual language (English and Malay) needs assessment questionnaire for people living with and beyond CRC in a multi-ethnic, low and middle-income setting.

2. Methods

In the current study, need was defined as the level of service or support that an individual perceived as necessary to achieve optimal well‐being following CRC [16]. People living with and beyond CRC in the context of present study included individuals diagnosed at least one-month prior to study recruitment who were either waiting for cancer treatment or under active treatment, or individuals who had completed treatment (irrespective of time since diagnosis) [17]. Recruitment was carried out from 12 April 2019 to 28 November 2020.

Study methodology was guided by the COnsensus-based Standards for the selection of health status Measurement INstruments (COSMIN) checklist [18], comprising three major phases; a) item development, b) scale development and c) scale evaluation [19]. Ethical approvals were obtained from the Medical Research and Ethics Committee [NMRR-18-3687-44522], University Malaya Medical Centre Medical Ethics Committee [20181016-6757] and Subang Jaya Medical Centre [201907.1]. Written and informed consent was obtained from all the study participants.

2.1. Phase I: Item development

A co-design approach was adopted, where a wide range of cancer stakeholders were actively involved in developing the items for the needs assessment tool.

2.1.1. Item generation and domain identification.

Generation of items in the initial draft of the questionnaire was guided by findings from independent qualitative inquiries that were previously undertaken by the study team in the local setting among people living with and beyond cancer. Details can be found elsewhere [20–25]. Concurrently, a literature review of the existing needs assessment tools for cancer was also conducted to screen for items [6,7,9–11,26]. Given that prior qualitative inquiries and needs assessment tools were from the pre COVID-19 era, all the items were reviewed by a diverse group of cancer stakeholders including individuals living with and beyond CRC, caregivers, cancer advocates and healthcare professionals (oncologists, surgeons, nurses, psychologists, counsellors, public health physicians) to ensure that contemporary issues were not missed. Shortlisted items were categorized according to subconstructs of needs (domains) based on consensus from the stakeholders, and in accordance with the supportive care framework proposed by Fitch 2008 [27].

2.1.2. Content validation.

Expert validation and face validation were conducted to establish that the generated items were relevant to the needs of people living with and beyond CRC and that key items have not been missed. A panel of seven content experts comprising healthcare professionals, a cancer researcher, a patient advocate and a patient living with CRC were convened to provide feedback on the generated items. The panel members were asked to rate the items using a four-point scale (1: not relevant, 2: not important, 3: relevant and 4: important), where a rating of three or four indicated agreement. Two rounds of content validation were conducted. In the first round, the acceptable cut-off for the item content validity index (I-CVI) was >= 0.83, while in the second round, the cut-off was >= 0.86 [28]. Kappa was also computed for individual items where a value above 0.74 was considered excellent [29]. Scale content validity index (S-CVI) was computed by calculating the average I-CVI of each domain, in which a cutoff value ≥ 0.90 was deemed acceptable [29].

2.1.3. Readability and translation.

The Flesch–Kincaid reading grade (FK), which was calculated using an online tool was used to determine the readability of the (English) items in the questionnaire [30]. All items were forward translated to the Malay language by two bilingual members of the study team. The dual language items were then reviewed by patients and cancer advocates who were proficient in both languages. Their feedback was used to revise the Malay items to improve clarity and comprehensibility. In the next step, the Malay version was reverse translated into English by two other independent translators who were also proficient in both languages. The back-translated questionnaire in English was compared with the original questionnaire. After item harmonisation and consensus were achieved, the final version of the questionnaire covered dual languages (Malay and English). Grammar and spelling were checked prior to cognitive debriefing.

2.2. Phase II and III (Psychometric evaluation): Scale development and evaluation

2.2.1. Study population, data collection and study instruments.

Malaysians above the age of 18 years who were living with and beyond CRC, diagnosed with any stage of CRC at least one month prior to recruitment, with or without stoma, were recruited. Individuals who were clinically unfit, as well as those who were unable to converse in English or Malay were excluded. Malay is the national language and is widely spoken across all major ethnic groups in Malaysia, while English serves as the second language and is commonly used in healthcare and research settings.

Respondents were recruited from multiple study sites serving ethnically diverse catchment populations across Klang Valley, the largest urban agglomeration in Malaysia. These included a public university hospital, several Ministry of Health (MOH)-owned hospitals, a private hospital and a local non-governmental organization (NGO) for CRC. Our sampling strategy was expected to enable adequate representation of patients from diverse ethnic and socioeconomic backgrounds in Malaysia. Notably, Malaysia has a pluralistic health system, where healthcare is funded by various means in the public and private hospitals, with differences in the fee structures and package of subsidized care even between the public system leading to differences in the socioeconomic backgrounds of patients [31].

Eligible respondents were identified during their routine outpatient visits or day care attendance in the hospital settings, whereas in the NGO setting, they were identified by a liaison officer from within the organization. After identification, they were then randomly selected using an online random number generator. Several recruitment methods were employed, including face-to-face encounters, online recruitment by enumerators, and through an NGO representative, to ensure continued data collection during the COVID-19 pandemic. Face-to-face recruitment occurred when restrictions were lifted, while online methods were used during lockdowns to minimize contact with high-risk cancer survivors. The recruitment method was based on the patient’s location and preference, with informed consent obtained after participants were briefed and provided with a participant information sheet.

Given that for exploratory factor analysis (EFA), a minimum sample size of 100 [32], or three participants per item were recommended [33], 155 participants were deemed adequate. For confirmatory factor analysis (CFA), a sample size of 145 was considered appropriate based on sample size estimation using G-power software [34].

The NeAT-CC required study respondents to rate the magnitude of their needs based on a six-point Likert scale (ranging from zero, indicating no need, up to five, which denoted very high needs). The European Organisation for Research and Treatment of Cancer Quality of Life Core 30 (EORTC QLQ-C30) [35,36], as well as the colorectal cancer supplement (EORTC QLQ-C29) [37,38] quality of life questionnaires was also administered. Data on sociodemographic factors were collected using a brief data collection form. Clinical details were retrieved from the hospital, and/or patient-held records.

2.2.2. Cognitive pretest and test-retest.

Cognitive pre-testing was conducted with 20 people living with and beyond CRC. Purposive sampling was undertaken to recruit patients at various clinical stages of CRC (early, advanced), treatment status (active treatment, not on active treatment), ethnicity (Malay, Chinese, Indians), and type of hospital (public, private). Respondents were asked to complete the questionnaire, and their feedback was used to further revise the questionnaire to improve feasibility, acceptability, and item relevance.

Test-retest reliability was assessed among 30 participants who were conveniently sampled to complete the questionnaire at two time points (two weeks apart). Intra-rater reliability for each item was determined using weighted kappa with linear weightage via vassarstat.net online software. A cut-off value of>=0.4 was used [39].

2.2.3. Exploratory Factor Analysis (EFA).

Exploratory factor analysis was conducted via the Factor program, using Unweighted Least Squares (ULS) [40] and the oblique Promin rotation method. The Kaiser-Meyer-Olkin test (KMO; cutoff ≥ 0.60) and Bartlett’s test of sphericity (p value <0.05) [41] were conducted to determine sampling adequacy. Parallel analysis was performed to determine the appropriate number of domains to extract, and the pattern matrix was examined to assess the factor to which the items loaded. Items with a factor loading of < 0.32, as well as those with cross-loadings that did not differ by at least 0.15, were removed from the model [42,43].

2.2.4. Reliability.

The reliability of the questionnaire was determined via examination of Cronbach’s alpha, inter-item correlation, and corrected item-total correlation (CITC) using IBM Statistical Package for Social Science (SPSS). Values of at least 0.30 for the CITC and 0.70 for Cronbach’s alpha were deemed acceptable [19].

2.2.5. Confirmatory factor analysis (CFA).

Confirmatory factor analysis was conducted via the SmartPLS software 3.2.9 using the partial least squares structural equation modelling (PLS-SEM) method. This entailed measurement model assessment for a reflective construct, which included assessment of factor loadings (cut-off of ≥ 0.50) [42], Cronbach’s alpha and composite reliability (CR) (>0.70) [42,44]. To assess convergent validity, the average variance extracted (AVE) was examined. An AVE of above 0.4 was considered acceptable if the CR of the domain was above 0.60 [45]. Discriminant validity was determined by assessing the Heterotrait-Monotrait ratio (HTMT, < 0.90 [46], and cross-loading of indicators (< 0.10) [47]. During the CFA phase, we evaluated the factor structure established through EFA without contrasting alternative models. This decision was predicated on the robust theoretical framework and empirical precision of the EFA solution. The EFA demonstrated a clear and interpretable structure characterized by high item loadings and minimal cross-loadings, which informed the development of the CFA model.

2.2.6. Criterion validity.

Given the lack of a gold standard questionnaire for needs assessment in people living with and beyond cancer, predictive criterion validity was assessed by using health-related quality of life [14,48,49]. The English and Malay versions of the EORTC QLQ-C30 and the EORTC QLQ-CR29 questionnaires, which have been validated in local settings [36,37] were used. Here, higher functional scales (body image, role and emotional functioning) indicated higher quality of life while higher symptoms scale (sexual functioning and financial difficulties) indicated higher symptom burden [50]. Spearman’s rank correlations (r_s_) between selected domain or items in the quality-of-life questionnaire and comparable domains of NeAT-CC questionnaire.

2.2.7. Floor and ceiling effects.

The sum of all items in each domain was calculated. Domains where >15% of the respondents obtained the lowest or highest possible scores were considered indicative of floor and ceiling effects [51].

3. Results

3.1. Phase I: Item development

3.1.1. Item generation and domain identification.

The domains and items that were developed based on the findings from prior qualitative inquiries [20–25] formed the basic components (items) of the needs assessment tool for people living with colorectal cancer. These comprised 67 items, which were grouped into five domains, namely psychosocial, information, practical, financial and employment needs.

3.1.2. Content validation.

In the first round of content validation, eight items had an item-level content validity index (I-CVI) of less than 0.78, while nine items were considered redundant by the expert panel. After reviewing the comments and the CVI, eight items were dropped from the questionnaire, eight items were rephrased, while another nine items were combined with relevant items to prevent redundancy. The revised questionnaire containing 51 items subsequently underwent a second round of content validation. The overall CVI was 0.96 and the CVIs for each domain ranged from 0.90 to 1.00 as depicted in Table 1. Thus, after two rounds of content validation, the preliminary questionnaire included 51 items that were categorized under six domains namely diagnosis, psychosocial, information, practical, financial and employment.

Table 1: Content validation of the Needs Assessment Tool for Colorectal Cancer (NeAT-CC).

3.1.3. Readability.

The readability of the English items in the NeAT-CC was six, indicating that the items in the questionnaire were easy to be read and understood by students aged around 12 years.

3.2. Phase II and III: Scale development and evaluation (Psychometric assessment)

Close to half of the 300 participants who were recruited were Malays (48.3%), followed by the Chinese (40.0%) and Indians (10.3%) (Table 2). Both men and women were equally represented. Three-quarters of the participants were aged above 50 years old. A vast majority received their primary treatment at public hospitals (91.7%). Close to 60% were from low-income households. Approximately, 30% of study participants reported having private health insurance.

Table 2: Baseline characteristics of participants in the validation study of NeAT-CC.

A majority of participants have been living with CRC between one and five years (60.3%). Most of them were diagnosed with either stage III (47.7%) or stage IV (34.3%) CRC at initial presentation. Two-thirds of the study participants had undergone colorectal surgery (66.0%), while nearly half had stoma. A high proportion of participants had received cytotoxic chemotherapy.

Of the above, 155 respondents were randomly selected for EFA, while the remaining 145 respondents were included in the CFA.

3.2.1. Cognitive pre-test and test-retest.

Feedback and comments from cognitive pre-test in 20 patients with CRC indicated that the needs questionnaire was straightforward, easy to understand, comprehensive and relevant. Overall, participants took about 15–20 minutes to complete the questionnaire.

In the test-retest analysis, all items had a weighted kappa value of above 0.40, indicating moderate to substantial agreement (Table in S1 Table 1). The lowest kappa value was 0.42 for item B6, which states the need for religious or spiritual support to cope with emotions. The highest kappa value was 0.87 for item E40, that mentioned the need to understand insurance benefits and coverage.

3.2.2. EFA and reliability.

The three items in the employment domain were not included in the EFA as the items were optional questions for participants who were employed or had recently lost their jobs due to their cancer diagnoses. EFA was thus performed with the remaining 48 items. The pattern matrix revealed that items A1 to A3 in the diagnosis needs domain loaded together with the psychosocial needs domain. The diagnosis needs domain was to assess the needs of CRC patients at the point of diagnosis, while psychosocial needs were to assess the needs of CRC survivors over longer periods beyond diagnosis. Thus, items under the diagnosis needs domain were removed from the EFA. The remaining 42 items underwent EFA and results from parallel analysis with polychoric correlations suggested a four-domain model that accounted for 56.6% of the cumulative proportion of variance. Kaiser-Meyer-Olkin (KMO) value of 0.903 and a significant Bartlett’s test of sphericity (p-value less than 0.001) were obtained, indicating sampling adequacy. Items E38, C14 and C15 cross-loaded unto two factors, with a difference of less than 0.15 in the loadings and were subsequently dropped from the questionnaire. The domains were named a) psychosocial and information needs (containing 15 items), b) practical and living with cancer needs (8 items), c) healthcare needs (11 items) and d) financial needs (8 items). The final factor loadings from the EFA and the communalities are presented in Table 3. The overall Cronbach’s alpha was 0.937, while Cronbach’s alpha of the individual domains ranged from 0.763 to 0.891 and CITC ranged from 0.447 to 0.864, indicating excellent internal consistency.

Table 3: Exploratory factor analysis, Cronbach’s alpha and corrected item-total correlation (CITC) of NeAT-CC.

3.2.3. CFA.

The 42 items from EFA, three items from diagnosis needs, and three items from employment needs were included for CFA. Results from the CFA using PLS-SEM are shown in Fig 1. The factor loadings of all items exceeded 0.50 (Table in S2 Table 2), with three domains having an AVE of more than 0.50, while another three domains had an AVE of 0.40 (Table 4). The newly developed tool also demonstrated good internal consistency with a CR value of above 0.80 for all the reflective domains, and a Cronbach’s alpha that was larger than 0.70. The HTMT ratios were less than 0.90 for all the outer model domains (Table in S3 Table 3), with no cross-loadings between the items, demonstrating good discriminant validity.

Table 4: Measurement model assessment of NeAT-CC from confirmatory factor analysis.

Confirmatory factor analysis model of the needs assessment tool for colorectal cancer (NeAT-CC) using partial least square-structural equation modelling.

3.2.4. Criterion validity.

The correlation between scales in the quality of life questionnaire (EORTC QLQ-C30 and QLQ CR-29) and the related items in NeAT-CC are presented in Table 5. Significant negative correlations were observed between the total needs score, financial needs, and the practical and living with cancer needs domain of the NeAT-CC with the global health status of the quality of life questionnaire, with rs = −0.36 (p < 0.001), rs = −0.42 (p < 0.001), and rs = −0.30 (p < 0.001) respectively.

Table 5: Correlation (rs) between NeAT-CC domain and EORTC QLC C30 and EORTC QLQ CR29 domains.

A significant positive correlation between the financial domain of the need’s questionnaire and the financial difficulties scale from the quality of life questionnaire was also observed with r_s_ = 0.60 and p value <0.001. This indicated that high financial needs positively correlated with higher financial difficulties. Practical and living with cancer needs (with items related to sex and fertility from the NeAT-CC) also positively correlated with sexual symptoms scores for males with r_s_ = 0.30 (p < 0.001). Overall, the NeAT-CC demonstrated adequate criterion validity as depicted in Table 5.

3.2.5. Floor and ceiling effects.

The mean score of the domains ranged from 2.77 to 3.94. With the exception of the diagnosis needs domain with a ceiling effect of 21.3%, no floor and ceiling effects were observed for the other domains (Table 6).

Table 6: Response distribution for each domain of the NeAT-CC (floor and ceiling effect).

3.3. Comparison of NeAT-CC with other commonly used needs assessment tools

We undertook a post-hoc comparison of NeAT-CC against other commonly used needs assessment tools (Table 7). Besides being a tool, that specifically targeted individuals living with and beyond CRC, the NeAT-CC was noted to stand out in its comprehensiveness where the needs around the time of diagnosis could also be measured in newly diagnosed individuals. Compared to previous tools, it included items measuring financial as well as employment needs.

Table 7: Comparison of commonly used needs assessment tools in individuals with (colorectal) cancer.

4. Discussion

Our questionnaire development and validation exercise resulted in a 48-item bilingual Needs Assessment Tool for Colorectal Cancer (NeAT-CC) encompassing six domains, namely, diagnosis needs, psychosocial and information needs, healthcare needs, practical and living with cancer needs, financial needs, and employment needs. Study findings suggest that the NeAT-CC is a psychometrically sound tool with low respondent burden that can be used to assess the needs of people living with CRC and beyond.

Apart from being guided by findings of prior qualitative inquiries in local settings, the co-design approach that we had adopted ensured that the perspectives of a wide range of stakeholders including patients and caregivers were taken into account. Compared to prior questionnaires, more items related to financial and employment needs were included in the newly developed NeAT-CC, in line with expert recommendations [52,53]. The NeAT-CC questionnaire possesses distinct features that set it apart from commonly used questionnaires, including catering to patients who are newly diagnosed to assess their needs around the time of diagnosis, and covering a comprehensive array of needs following CRC along the cancer care continuum including financial needs, employment needs, needs related to traditional and complementary medicine, fertility needs, religious and spiritual support, needs related to diet, and needs for medical devices such as stoma bags.

The high proportion of younger individuals with colorectal cancer (< 50 years) [54], high risk of financial toxicity following cancer diagnosis [31] and high employment needs amid limited employment protection policies [21] in our settings, necessitated the inclusion of these domains. Moreover, the needs relating to traditional and complementary medicine that were covered in our tool reflect the cultural beliefs of our multi-ethnic population. It is worth noting that to the best of our knowledge, none of the currently available needs assessment tools were particularly developed for people living with and beyond CRC. Furthermore, only the Chinese version of SCNS-SF questionnaire was cross culturally validated among patients with CRC in Hong Kong and Taiwan [55]. All of the above strengthen our optimism that the newly developed NeAT-CC will be both meaningful and relevant to the needs of people living with and beyond CRC, not only in the local setting but also in other multi-ethnic settings.

The items in the NeAT-CC were finalized using multiple techniques, namely factor analysis, test-retest and confirmatory factor analysis. The test–retest estimate of reliability was determined by assessing the correlation between sets of scores at different time points [56] and was found to be particularly reliable for self-reported surveys [57]. Moderate to substantial reliability was observed in our study, surpassing previous tools with lower reliability [7] or no test–retest data [58].

Following EFA, three items that were removed were related to the need for written information (C14), information about all treatment options (C15), and information on the costs of treatment (E38). All three items cross-loaded with healthcare needs, suggesting they were likely highly correlated and complex [43]. The two domains that were excluded from EFA (diagnosis and employment) were retained in CFA due to their importance [20,21,23] and expert justification [43]. The diagnosis needs assess needs at diagnosis, highlighting the persistent emotional toll of colorectal cancer during treatment and long-term survivorship [22]. The employment needs domain, meanwhile, reflects the needs among the working group, which are crucial for understanding the challenges faced by the working class, including economic impacts and the ability to return to work post-treatment. It is also noteworthy that CFA demonstrated that these two domains achieved adequate convergent and discriminant validity to justify their retention in the NeAT-CC.

Quality of life assessment is commonly used to identify patients’ concerns and wellbeing. However, it only assesses the presence and severity of a concern in an individual and not whether they need assistance to address their concerns [14]. Comparatively, needs assessment identifies the range of concerns experienced by patients for which they require intervention or additional help [27]. Despite the differences, multiple domains of needs have been shown to be predictive of health-related quality of life measures, a finding that is also corroborated by prior research [44]. With no gold standard for needs assessment, criterion validity, which measures the relationship between a test score and a relevant measure, was established by correlating NeAT-CC with the EORTC QLQ-C30 and QLQ-CR29 quality of life questionnaires. While further validation with a gold standard is recommended [19], the significant correlation with quality of life demonstrates NeAT-CC’s effectiveness in capturing the relationship between needs and overall well-being.

The ceiling effects for the diagnosis needs domain indicated that most survivors were experiencing very high levels of needs. Although presence of ceiling effects may impair the responsiveness of the scale and may indicate limited content validity [51], our findings are in line with prior evidence where similar (floor and ceiling) effects were observed [45]. It was postulated that these observations may not necessarily be due to limitation of the scale but rather because people living with and beyond cancer were having either very high needs or very low care needs in some domains [59], such as diagnosis needs [20].

In sum, our newly developed NeAT-CC is not only psychometrically sound but also straightforward and easily understandable for patients. Given that the tool may be self-administered, it is expected that its use in routine clinical practice will pose minimal administrative burden on health systems. Utilization of the NeAT-CC in clinical and research settings will undoubtedly yield valuable knowledge and evidence on the needs of people living with and beyond CRC that can be leveraged to guide the development of new services that are patient-centered, or optimization of existing supportive and survivorship care services. Importantly, the tool may also be used by cancer civil societies to guide and evaluate their programs.

To the best of our knowledge, this is the first needs assessment tool that has been developed exclusively for CRC, which spans the supportive and survivorship care needs along the cancer care continuum. We undertook a rigorous methodology to develop and validate the NeAT-CC, in accordance with international standards [18]. Here, it is noteworthy that almost two-thirds of prior studies in this area did not employ CFA as part of their analyses [10]. Furthermore, the inclusion of patients from multiple settings allowed for diversity in ethnic and socioeconomic backgrounds, where the baseline characteristics of the study participants, including gender, ethnicity, cancer stage, and insurance status, are broadly consistent with national colorectal cancer (CRC) data and recent studies. A higher CRC burden among males and advanced-stage diagnoses was also observed in this study [60]. The predominance of Malay participants aligns with previously reported trends [54]. Additionally, 28.3% of participants reported having personal health insurance, which is comparable to national estimates [61].

The diversity informed the inclusion of additional items relevant to the health practices of multicultural societies, such as the need for traditional and complementary therapies. It is therefore believed that the NeAT-CC has potential for wider application and can be cross-culturally validated for use in other settings with diverse populations. However, in high-income countries with comprehensive social support systems, certain domains such as employment and access to treatment may require modification to ensure contextual relevance.

In the local context, future studies should validate the NeAT-CC among the indigenous populations, as well as Mandarin-and Tamil-speaking patients, to ensure that their voices are also represented. Future plans also include implementing a cohort study to assess supportive care needs longitudinally and to compare these needs between urban and rural populations.

5. Conclusion

The NeAT-CC is expected to enable accurate and effective needs assessment following CRC in Malaysia, with potential for adaptation in other multi-ethnic and/ or low- to middle-income settings. This, in turn, may facilitate the prioritization and development of supportive and survivorship care services that meet the needs of people living with and beyond CRC, as well as guide the allocation of scarce health resources.

Supporting information

S1 TableTest-retest reliability results of the Needs Assessment Tool for Colorectal Cancer (NeAT-CC).This table presents the test–retest reliability analysis for the Needs Assessment Tool for Colorectal Cancer (NeAT-CC), including Weighted Kappa for each item.(DOCX)

S2 Table2 Confirmatory factor analysis of the needs assessment tool for colorectal cancer (NeAT-CC).This table presents factor loadings for each item across the six domains.(DOCX)

S3 TableHeterotrait–monotrait (HTMT) ratios of the domains for discriminant validity.This table displays the HTMT ratios between domains of the NeAT-CC to demonstrate discriminant validity.(DOCX)