Bastiaanse LP, Vlasveld G, Penning C, Evenhuis HM. Intellectual Disability Medicine, Department of General Practice, Erasmus Medical Center, Rotterdam, The Netherlands. email@example.com. J Nutr Health Aging. 2012;16(9):759-62. doi: 10.1007/s12603-012-0050-z.
OBJECTIVE: Feasibility and reliability of the Mini Nutritional Assessment (MNA) in older adults with intellectual disabilities (ID).
DESIGN: Instrument development.
SETTING: Three care providers for people with ID.
PARTICIPANTS: 48 persons aged 50 years and over with borderline to profound ID and their professional caregivers.
MEASUREMENTS: The MNA was performed by means of interviews with participants (N = 12) and caregivers (N = 48) and physical assessments of participants (N = 47). Aspects of feasibility: completion of interview, difficulty of answering interview items, duration of interview and completion of physical assessment. Aspects of reliability: inter-observer reliability between caregivers and between participants and caregivers, test-retest reliability and internal consistency. For inter-observer and test-retest reliability, intraclass correlation coefficients (ICC) were calculated, and for internal consistency Chronbach’s alpha.
RESULTS: All participants and caregivers completed the interview part. For 7 out of 12 personally interviewed participants and none of the caregivers, at least 3 out of 15 questions were difficult to answer. Mean duration of the interview was 7 minutes in participants and 4 minutes in caregivers. Physical assessment was successfully performed in 40 participants (85.1%). In the remaining 7 participants (14.9%) missing values were retrieved from the medical records. ICCs (95% confidence interval) for test-retest and inter-observer reliability between caregivers were good, 0.85 (0.72 – 0.92) and 0.86 (0.74 – 0.92) respectively, but ICC for inter-observer reliability between caregivers and persons with ID was low, 0.03 (-0.51 – 0.59). Internal consistency was 0.61.
CONCLUSION: The MNA is feasible and reliable for older people with ID. Interview data can be reliably obtained through caregivers, but not through people with ID.