Developing a tool to assess mattress satisfaction: the Boston Mattress Satisfaction Questionnaire.

Study objectives: Adults are advised to spend approximately one third of their lives sleeping, yet there is a dearth of scientific research on mattresses, a common sleep surface. We develop and conduct initial validation of the Boston Mattress Satisfaction Questionnaire (BMSQ). Methods: The BMSQ was designed with sleep scientists and clinicians (n = 5) and mattress industry professionals (n = 2) to assess two broad domains: mattress satisfaction (MS) and mattress characteristics (MC), including mattress type, size, age, and extent to which the mattress is pain-inducing. MS is measured with questions assessing mattress comfort, firmness, temperature, and overall satisfaction on 10-point scales from 1 (least) to 10 (most satisfied). We administered the BMSQ to a large, population-based sample of US adults. We also asked demographic questions. We conducted exploratory factor analysis, then dichotomized BMSQ responses (low: ≤ 5; high ≥6) for multivariable logistic regression to explore the demographic characteristics associated with mattress satisfaction. Results: Among participants (n = 1,055), 47.7% were male and 52.2% female. Average age was 49.4 (s.d. = 17.5 years). The 4 BSMQ items demonstrated high inter-item correlation (≥0.8) and Cronbach's α of 0.95. BMSQ-MS variables were inversely correlated with perceptions of the mattress being pain-inducing (p < 0.001) and mattress (p < 0.001). BMSQ variables had a weak correlation with mattress size (p < 0.01). Regression revealed higher mattress satisfaction among those ≥75 years old (v. 18-24 years); Hispanic and Asian individuals (compared to White, non-Hispanic); those earning >$20,000 annually (compared to <$10,000); and those reporting foam, hybrid, air-filled chamber mattresses (compared to all-spring). Conclusions: Our findings suggest that the BMSQ may be useful for assessing mattress satisfaction.