The Multinational Association for Supportive Care in Cancer (MASCC) Unmet Needs Assessment of Cancer-Related Cognitive Impairment Impact (MASCC COG-IMPACT) is a 55-item self-report measure developed by Haywood and colleagues (2025) under the auspices of MASCC. It is designed for adults aged 18 and over who have completed curative-intent cancer treatment, have no current evidence of disease, and who self-identify as experiencing cancer-related cognitive impairment (CRCI).
CRCI is the difficulties with thinking, memory, attention, and processing speed that many cancer survivors notice after treatment. The measure is endorsed by the Clinical Oncology Society of Australia (COSA) as the official MASCC tool for assessing CRCI impact (Haywood, Chan, Chan, Baughman, et al., 2025b). It does not measure the presence or severity of CRCI itself; rather, it characterises the personal impact of CRCI on the survivor’s life and the supportive care needs that arise from that impact, across eight domains of everyday functioning.
The instrument is intended for use in supportive care, oncology, and clinical psychology settings where the clinical question is not “To what extent is this person experiencing cognitive difficulties?” but “now that CRCI is present, what difficulties is CRCI causing in this person’s life and what level of support do they need for those difficulties?” Because what this tool assesses is impact and unmet need rather than cognitive impairment per se, the MASCC COG-IMPACT is best used alongside subjective and objective cognitive assessments such as the PROMIS Cognitive Function scale or the Functional Assessment of Cancer Therapy Cognitive Function scale, not as a substitute for them (Haywood, Chan, Chan, Baughman, et al., 2025b).
The MASCC COG-IMPACT assesses eight subscales spanning the practical, relational, vocational, psychological, communicative, social, and informational domains in which CRCI most commonly impacts cancer survivors. Each subscale is reported on two paired indices, one capturing the survivor’s CRCI-related Difficulties in that domain and the other capturing their Unmet Supportive Care Needs. The subscales are:
The MASCC COG-IMPACT was designed primarily to support tailored and person-centred supportive care planning and provision in cancer survivorship. Its central clinical contribution is the separation of CRCI-related Difficulties from CRCI-related Unmet Needs within each domain. This separation lets a clinician see not only where CRCI is impacting the survivor’s life, but also where support is currently inadequate. A subscale that is high on Difficulties but low on Unmet Needs suggests the survivor is experiencing impact but is already well supported in that domain. A subscale that is high on both Difficulties and Unmet Needs suggests a care provision and/or referral-worthy gap and is the strongest signal for action.
The measure is well suited to repeat administration for progress monitoring, to shared review with the survivor (where the per-domain profile makes the impact of CRCI concrete in session), and to formulation conversations across multidisciplinary supportive care teams. It also functions as a brief screen for CRCI impact and CRCI-related unmet supportive care needs, with the author-anchored Unmet Needs descriptor bands providing a natural threshold for identifying which survivors warrant a supportive care conversation.
The MASCC COG-IMPACT comprises 55 items grouped into eight subscales. Each item is presented to the respondent as two linked questions: a binary Difficulty question (“NO/YES”), and, only where the Difficulty is endorsed, an Unmet Needs follow-up (“If yes, how much additional support do you need?”).
The 55 items therefore appear as 110 questions in NovoPsych. Every subscale yields two scores: a Difficulties Subscale Score, the proportion of the subscale’s items endorsed as a difficulty, ranging from 0 to 1; and an Unmet Needs Subscale Score, the average of the subscale’s Unmet Needs values, ranging from 0 to 3. The eight subscales, with their item counts and item and form-question ranges, are:
The two Total scores share these ranges: the Difficulties Total Score (the average of the eight Difficulties Subscale Scores) ranges from 0 to 1, and the Unmet Needs Total Score (the average of the eight Unmet Needs Subscale Scores) ranges from 0 to 3.
The Difficulties index is reported as a raw proportion. as “endorsed X of Y”. A higher proportion indicates that the survivor is endorsing more CRCI-related difficulties in that domain.
The Unmet Needs index is reported with descriptor bands derived from the response-scale semantics in the Manual (MASCC, 2025).
Because the Unmet Needs response options carry consistent semantic meaning (from “no need for support” through to “high need”), the descriptor bands are anchored to the resulting subscale score range from 0 to 3. The same bands apply both to subscale Unmet Needs scores and to the Total Unmet Needs Score:
The clinical signal of the MASCC COG-IMPACT lies in the per-domain comparison of the Difficulties index against the Unmet Needs index. Higher Difficulties indicate that the survivor is endorsing more CRCI-related difficulties in that domain, and higher Unmet Needs indicate greater unmet supportive care needs in that domain. A subscale on which Difficulties are elevated but Unmet Needs are low suggests the survivor is currently well supported with respect to those difficulties. A subscale on which both indices are elevated represents a care provision and/or referral-worthy gap, indicating the survivor is experiencing difficulties in that domain and is not currently receiving the supportive care that would help. The eight-subscale profile lets the clinician identify which life domains carry this gap and direct conversation, intervention, or referral accordingly.
Where Unmet Needs reach the Low-to-Moderate band or above on a subscale (an Unmet Needs Score of 1 or more), item-level review can help identify the specific aspects of the domain the survivor is endorsing most strongly. This is particularly useful for shared review with the survivor, where the survivor’s own words (in the form of endorsed item content) anchor the conversation.
On first administration, the report produces two subscale bar charts side by side. The Difficulties subscale bar chart displays each of the eight subscale Difficulties scores on a 0 to 1 scale. The Unmet Needs subscale bar chart displays each of the eight subscale Unmet Needs scores on a 0 to 3 scale, with descriptor-band shading marking the Low-to-Moderate, Moderate-to-High, and High bands (boundaries at 1.0, 2.0, and 3.0, including a narrow High band at the top of the scale). Each chart also leads with a stacked Total bar: its segments are the eight subscale contributions to the Total for that index, coloured to match the subscales and proportional to the subscale scores so that they sum to the Total.
On repeat administration, the two subscale bar charts are replaced by a multi-administration line plot for each index. Each plot displays the subscale scores across all administrations as a visual trajectory, with the Unmet Needs plot containing the descriptor-band shading for context.
The MASCC COG-IMPACT was developed through an eight-step mixed-methods co-design process led by the Multinational Association of Supportive Care in Cancer (Haywood, Chan, Chan, Baughman, et al., 2025a). The process began with 32 qualitative interviews with cancer survivors who self-identified as experiencing CRCI, followed by 19 interviews with oncology health professionals to identify the most clinically relevant CRCI-related difficulties and unmet supportive care needs. An initial pool of 183 candidate items was developed from this qualitative work. A modified single-round Delphi consensus process with 29 health professionals refined the item pool, and 22 cognitive interviews with cancer survivors across two rounds (using both Reparative and Descriptive cognitive interviewing approaches) further refined the wording and format. The resulting 55-item measure was then administered to a large international validation sample, with structural, reliability, and validity analyses conducted on the final dataset.
The validation sample comprised 491 adult cancer survivors recruited through Prolific across 23 countries (Haywood, Chan, Chan, Baughman, et al., 2025a). The largest country contributions were the United States (35.6%), South Africa (24.2%), the United Kingdom (23.0%), Australia (3.1%), and Poland (2.0%). The sample had an average age of 44.39 years (spread 15.02; range 18 to 81), with 73.1% female and 26.9% male respondents. The most common cancer types were breast (34.4%), lymphoma (7.9%), and bowel or colorectal (6.5%) cancers. The most common treatments received (selected from multiple options) were chemotherapy (68.4%), radiation therapy (52.1%), hormone therapy (27.1%), surgery (22.4%), immunotherapy (15.9%), and targeted therapy (13.4%).
Structural validity was evaluated using principal components analysis with oblique (oblimin) rotation in the development sample (N = 491; Haywood, Chan, Chan, Baughman, et al., 2025a). Sampling adequacy was strong (Kaiser-Meyer-Olkin = 0.964) and Bartlett’s test of sphericity was significant (p < 0.001), with all individual-item measures of sampling adequacy greater than 0.9. Eight factors had eigenvalues greater than 1 and together accounted for 67.97% of the variance, supporting the eight-subscale structure. All 55 items loaded above 0.30 on at least one factor and all were retained. One item showed a small cross-loading (0.360 on Executing Regular Activities versus 0.307 on Relational Difficulties) and was retained on Executing Regular Activities following team consultation.
Internal consistency was strong across all eight subscales on both indices in the development sample (N = 491; Haywood, Chan, Chan, Baughman, et al., 2025a). For the Difficulties index, Cronbach’s alpha ranged from 0.742 (Verbal Communication Challenges) to 0.875 (Psychological Challenges), described by the authors as “good” or “very good” across subscales. For the Unmet Needs index, Cronbach’s alpha ranged from 0.879 (Verbal Communication Challenges) to 0.943 (Psychological Challenges), described by the authors as “very good” or “excellent” across subscales.
Test-retest reliability was evaluated in a subsample of 119 respondents who completed the measure twice with a two-week interval. Intra-class correlations for the Difficulties index ranged from 0.724 to 0.869 (average 0.814), and for the Unmet Needs index from 0.691 to 0.856 (average 0.785). The authors describe these as “good” or “excellent” across subscales (Haywood, Chan, Chan, Baughman, et al., 2025a).
Convergent validity was supported by significant positive correlations (p < 0.01) between the MASCC COG-IMPACT subscales and established measures of related constructs in the development sample (Haywood, Chan, Chan, Baughman, et al., 2025a). The strongest convergent associations were with the PROMIS Cognitive Function scale, which correlated with multiple Difficulties subscales in the range r = 0.55 to 0.66 (for example, Psychological Challenges Difficulties r = 0.66 with PROMIS Cognitive Function). The Cancer Survivors Unmet Needs Scale Existential Survivorship subscale correlated with Unmet Needs subscales in the range r = 0.50 to 0.62, and the Depression Anxiety Stress Scale Depression subscale correlated with Psychological Challenges and Social Functioning Difficulties in the range r = 0.55 to 0.64. Significant correlations were also found with the Cognitive Symptom Checklist – Work, the remaining four CaSUN subscales, and the Assessment of Quality of Life six-dimension instrument, in the expected directions.
Discriminant validity was supported by weak correlations between the MASCC COG-IMPACT and a construct-distant comparison: the AQoL six-dimension Senses subscale. Across all 16 MASCC COG-IMPACT subscales (eight subscales by two indices), correlations with AQoL-Senses ranged from r = 0.206 to 0.303, all falling in the small effect range (below 0.4 per Hair, Page, & Brunsveld, 2019), as expected given the construct distance (Haywood, Chan, Chan, Baughman, et al., 2025a).
Additional validity evidence comes from two application studies that drew on subsets of the same development dataset. In a study of CRCI-related unmet needs predictors (N = 456 employed and unemployed subset), cognitive impairment severity (measured by PROMIS Cognitive Function) and psychological distress (measured by the Depression Anxiety Stress Scale) were the most pertinent predictors of CRCI unmet needs across all eight subscales, with hierarchical regression models accounting for between 24.5% and 47.4% of variance in Unmet Needs subscale scores (Haywood, Chan, Chan, Dauer, et al., 2025). In a second study focused on the Occupational/Vocational Functioning subscale (N = 358 currently employed subset), level of education accounted for a small but significant amount of variance in Occupational/Vocational Unmet Needs beyond perceived cognitive functioning and employment level (R-squared change = 0.011, p = 0.030) (Haywood, Rossell, et al., 2026).
Acceptability, appropriateness, and feasibility were assessed using the Weiner et al. Acceptability of Intervention Measure, Intervention Appropriateness Measure, and Feasibility of Intervention Measure (each rated on a five-point scale) in the development sample (Haywood, Chan, Chan, Baughman, et al., 2025a). Average ratings were high on all three: Acceptability M = 4.03 (SD = 0.629), Appropriateness M = 4.12 (SD = 0.619), and Feasibility M = 4.12 (SD = 0.604). All 55 items were endorsed as a “difficulty” by more than 10% of participants, supporting the relevance of every item to the target population.
The actionable signal is the pairing of the two indices within a subscale, not either index alone. Subscales elevated on both Difficulties and Unmet Needs are the gaps where impact is present and support is currently insufficient, and ranking those by Unmet Needs Score indicates where a supportive care conversation or referral is most warranted first. Domains high on Difficulties but low on Unmet Needs are better monitored than acted on, since the respondent is already adequately supported there. Item-level content within a flagged domain then sharpens the focus onto the specific difficulties the respondent has endorsed most strongly.
Because the measure separates eight domains, the one with the highest unmet need can orient the direction of support. When Occupational/Vocational Functioning is most elevated, support might focus on vocational rehabilitation, workplace adjustment conversations, and strategies for managing cognitive load at work; when Psychological Challenges predominates, it might shift toward adjustment-oriented psychological support around identity, anxiety, and coping. A high Informational Needs domain points toward targeted psychoeducation and signposting to credible resources, while elevated Relational Difficulties may indicate a role for couple or family-inclusive support. These are directions for clinical reasoning rather than prescriptions, with the clinician bringing their own judgement and the respondent’s preferences to the plan.
The MASCC COG-IMPACT characterises the impact of cognitive change and the supportive care needs that follow, which is a different question from how much cognitive change is present or how severe it is. It is best paired with a measure of perceived cognitive function and, where objective deficits are in question, formal neuropsychological testing, so the workup captures both the cognitive picture and its lived consequences. Two respondents with similar perceived or measured cognition can differ markedly in impact and unmet need depending on their roles, supports, and demands, and that is the gap this measure is designed to fill.
It was developed for adults who have completed curative-intent treatment, have no current evidence of disease, and already identify as experiencing cancer-related cognitive impairment, so it is most informative once a survivor has raised cognitive concerns in the post-treatment period rather than as an unprompted screen during active treatment. Responses reflect the respondent’s current experience of impact and need, which makes the measure well suited to repeat administration at review points to track how needs are shifting. It is often completed shortly before a survivorship or supportive care appointment so the profile is current for that discussion.
Movement on the two indices does not always track together, and a rise in unmet need can reflect a change in circumstance rather than worsening cognition. Support structures that surround active treatment often fall away once treatment ends, demands such as returning to work can increase, and growing awareness of cognitive change can itself surface needs that were not previously articulated. For this reason, change between administrations is best read in the context of the respondent’s situation, and the report describes that movement in plain language without applying a statistical change-score label.
Haywood, D., Chan, A., Chan, R. J., Dauer, E., Dhillon, H. M., Henneghan, A. M., Lustberg, M. B., O’Connor, M., Vardy, J. L., Rossell, S. L., & Hart, N. H. (2025). Accounting for unmet needs resulting from cancer-related cognitive impairment. Journal of Cancer Survivorship. https://doi.org/10.1007/s11764-025-01769-6
Haywood, D., Chan, A., Chan, R. J., Baughman, F. D., Dauer, E., Dhillon, H. M., Henneghan, A. M., Lawrence, B. J., Lustberg, M. B., O’Connor, M., Vardy, J. L., Rossell, S. L., & Hart, N. H. (2025a). The MASCC COG-IMPACT: An unmet needs assessment for cancer-related cognitive impairment impact developed by the Multinational Association of Supportive Care in Cancer. Supportive Care in Cancer, 33(2), 120. https://doi.org/10.1007/s00520-025-09149-7
Haywood, D., Chan, A., Chan, R. J., Dauer, E., Dhillon, H. M., Henneghan, A. M., Lustberg, M. B., O’Connor, M., Vardy, J. L., Rossell, S. L., & Hart, N. H. (2025). Accounting for unmet needs resulting from cancer-related cognitive impairment. Journal of Cancer Survivorship. https://doi.org/10.1007/s11764-025-01769-6
Haywood, D., Rossell, S. L., Henneghan, A., Baughman, F. D., Haywood, J., Dauer, E., Hegde, A., Moustafa, A. A., & Hart, N. H. (2026). Work challenges and cancer-related cognitive impairment: Level of education accounts for unmet needs beyond cognitive impairment severity. Supportive Care in Cancer, 34(3), 188. https://doi.org/10.1007/s00520-026-10420-8
Haywood, D., Chan, A., Chan, R. J., Baughman, F. D., Dauer, E., Dhillon, H. M., Henneghan, A. M., Lawrence, B. J., Lustberg, M. B., O’Connor, M., Vardy, J. L., Rossell, S. L., & Hart, N. H. (2025b). The MASCC COG-IMPACT: The COSA endorsement of a MASCC developed unmet needs assessment tool for cancer-related cognitive impairment impact. Asia-Pacific Journal of Clinical Oncology, 22(1), 3–7. https://doi.org/10.1111/ajco.70002
Multinational Association of Supportive Care in Cancer. (2025). MASCC COG-IMPACT Manual and Scoring Procedure (v4). https://osf.io/5zc3a/
