Development of the Health Awareness and Behaviour Tool (HABiT): reliability and suitability for a Canadian older adult population

Abstract

BackgroundDetermining the effectiveness of community-based health promotion and disease prevention programs requires an appropriate data collection tool. This study aimed to develop a comprehensive health questionnaire for older adults, called the HABiT, and evaluate its reliability, content validity, and face validity in assessing individual health-related items (e.g., health status, healthcare utilization) and five specific scales: knowledge, current health behaviors (risk factors), health-related quality of life (HRQoL), perceived risk and understanding, and self-efficacy.MethodsIterative survey development and evaluation of its psychometric properties in a convenience sample of 28 older adults (≥ 55 years old), half from a low-income population. Following item generation, the questionnaire was assessed for content validity (expert panel), face validity (participant feedback), internal consistency of each scale (Cronbach’s alpha), and test-retest reliability for each item and scale (Pearson’s r and phi correlations, as appropriate).ResultsQuestions were drawn from 15 sources, but primarily three surveys: Canadian Community Health Survey, Canadian Diabetes Risk Questionnaire (CANRISK), and a survey by the Canadian Hypertension Education Program. Expert consensus was attained for item inclusion and representation of the desired constructs. Participants completing the questionnaire deemed the questions to be clear and appropriate. Test-retest reliability for many individual items was moderate-to-high, with some exceptions for items that can reasonably change in a short period (e.g., perceived day-to-day stress). Of the five potential scales evaluated, two had acceptable internal consistency (Cronbach’s alpha ≥ 0.60) and a subset of one scale also had acceptable internal consistency. Test-retest reliability was high (correlation ≥ 0.80) for all scales and sub-scales.ConclusionsThe HABiT is a reliable and suitable comprehensive tool with content and face validity that can be used to evaluate health promotion and chronic disease prevention programs in older adults, including low-income older adults. Some noted limitations are discussed. Data collected using this tool also provides a diabetes risk score, health literacy score, and quality-adjusted life years (QALYs) for economic analysis.

Health knowledge and self-efficacy to make health behaviour changes: a survey of older adults living in Ontario social housing

“…All data was collected using the Health Awareness and Behaviour Tool (HABiT) [ 22 ]. The HABiT survey was developed as part of CP@clinic’s impact evaluation.…”

“…The HABiT survey was developed as part of CP@clinic’s impact evaluation. It is reliable, has internal consistency, and content and face validity in the target population (older adults living in social housing) [ 22 ]. It assesses individuals’ health indicators such as health status and healthcare utilization and contains scales for measuring health knowledge, health behaviours and risk factors, health-related quality of life (HRQoL), risk perception and understanding, and self-efficacy to make health behaviour change [ 22 ].…”

“…It is reliable, has internal consistency, and content and face validity in the target population (older adults living in social housing) [ 22 ]. It assesses individuals’ health indicators such as health status and healthcare utilization and contains scales for measuring health knowledge, health behaviours and risk factors, health-related quality of life (HRQoL), risk perception and understanding, and self-efficacy to make health behaviour change [ 22 ]. Questions were developed and included from 15 sources, but primarily three surveys: the Canadian Community Health Survey [ 23 ], Canadian Diabetes Risk Questionnaire (CANRISK) [ 24 ], and a survey by the Canadian Hypertension Education Program [ 25 ].…”

“…Where participants did not consent to waist circumference measurements, clothing/pant size was used as a proxy value. All other risk factors were self-reported by participants via the HABiT questionnaire [ 22 ]. All participants who self-reported any alcohol consumption or that they were an “Occasional” or “Daily” smoker were coded as ‘drinkers’ and ‘smokers,’ respectively.…”

“…The HABiT survey [ 22 ] included 19 knowledge questions in total, 11 regarding CVD and 8 regarding DM risk factors and symptoms. Each question was a statement, and the participant needed to indicate if they believed it was true or false using a 5-point Likert Scale (Definitely True, Maybe True, Not Sure or Don’t Know, Maybe False, Definitely False).…”