Psychometric properties of cancer survivors' unmet needs survey.

PURPOSE: This study aims to develop a psychometrically rigorous instrument to measure the unmet needs of adult cancer survivors who are 1 to 5 years post-cancer diagnosis. "Unmet needs" distinguishes between problems which survivors experience and problems which they desire help in managing. METHODS: The survey was developed from a comprehensive literature review, qualitative analysis of the six most important unmet needs of 71 cancer survivors, review of the domains and items by survivors and experts, cognitive interviews and a pilot test of 100 survivors. A stratified random sample of 550 cancer survivors, selected from a population-based Cancer Registry, completed a mailed survey to establish reliability and validity. RESULTS: The final 89-item Survivors Unmet Needs Survey (SUNS) has high acceptability, item test-retest reliability and internal consistency (Chronbach's alpha 0.990), face, content and construct validity. Five subscales measure Emotional Health needs (33 items, 19.4% of variance), Access and Continuity of Care (22 items, 15.1%), Relationships (15 items, 12.1%), Financial Concerns (11 items, 10.3%) and Information needs (eight items, 8.1% of the variance). CONCLUSIONS: This instrument has strong psychometric properties and is useful for determining the prevalence and predictors of cancer survivors' unmet needs across types of cancer, length of survivorship and socio-demographic characteristics. Use of the SUNS will enable more effective targeting of programmes and services and guide policy and health planning decisions. RELEVANCE: This study is an important step toward evidence-based planning and management of problems which the growing survivor population requires assistance in managing.

The cancer support person's unmet needs survey: psychometric properties.

BACKGROUND: A rigorous psychometric methodology was used to develop a measure of unmet needs for cancer survivors' principal support persons. Principal support person was defined as "someone you can count on and who helps you with your needs." METHODS: Development of the domains and the items followed an extensive literature review, iterative input from support persons, and consultation with health professionals and front-line staff working with cancer survivors and their supports. Cognitive interviews helped clarify item wording, and the draft questionnaire was reappraised by a group of support persons. The questionnaire was reduced to 90 items and sent to a stratified, random sample of cancer survivors selected from a provincial population-based cancer registry. They were asked to give the survey to their support person. RESULTS: The resulting 78-item Support Person Unmet Needs Survey has high acceptability, item test-retest reliability, internal consistency (Chronbach alpha = .990), and face, content, and construct validity. It captures 6 domains of unmet needs and accounts for 73.5% of total variance: Information and Relationship Needs (27 items, 22.1% of variance), Emotional Needs (16 items, 15.2%), Personal Needs (14 items, 14.0%), Work and Finance (8 items, 8.8%), Health Care Access and Continuity (9 items, 8.6%), and Worries About Future (4 items, 4.8%). CONCLUSIONS: This instrument will be of use where there is an interest in examining the impact of cancer not only on cancer survivors but also on their identified principal support persons.

Exploration of cluster structure-activity relationship analysis in efficient high-throughput screening.

Sequential screening has become increasingly popular in drug discovery. It iteratively builds quantitative structure-activity relationship (QSAR) models from successive high-throughput screens, making screening more effective and efficient. We compare cluster structure-activity relationship analysis (CSARA) as a QSAR method with recursive partitioning (RP), by designing three strategies for sequential collection and analysis of screening data. Various descriptor sets are used in the QSAR models to characterize chemical structure, including high-dimensional sets and some that by design have many variables not related to activity. The results show that CSARA outperforms RP. We also extend the CSARA method to deal with a continuous assay measurement.