Salva 2011.
| Study characteristics | ||
| Methods | Cluster RCT. Duration: 12 months. Location: Spain. |
|
| Participants |
Inclusion criteria: diagnosis of mild‐moderate dementia; MMSE < 26; living at home; ambulatory with identified caregiver. Exclusion criteria: MMSE > 26; residents in an institution; nasogastric feeding; terminal care; already participating in a nutrition intervention study. Diagnosis: dementia (SSM IV criteria). Number randomised: 946 (intervention group, n = 448; control group, n = 498). Attrition: fully described; intervention group 4/29 (14%), control group 4/31 (13%). Gender split: 302/646 (32%) male, 644/946 (68%) female. Age: mean (SD) years 79 (7.3). Nutritional status: MNA (intervention group 7.8% malnourished and 51.5% at risk; control group 2.8% malnourished and 34.5% at risk). |
|
| Interventions |
Intervention: participants received dietary advice in the form of standardised protocol for feeding and nutrition delivered as education to caregivers, participants and relatives; Control: participants received no dietary advice in the form of usual care (detail not described). |
|
| Outcomes | Activities of daily living; MNA; caregiver burden scale; nutritional status (weight, MAC, calf circumference); cognitive function (MMSE, Clinical Dementia Rating scale, Eating Behaviour Scale, neuropsychiatric inventory questionnaire, depression, instrumental activities of daily living, healthcare costs, caregiver burden (Zarit scale), mortality. | |
| Publication details |
Language: English. Funding: commercial funding ‐ Nestec Limited. Publication status: peer‐reviewed journal. |
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk |
Quote from paper: "The unit of randomisation was the medical centres..." Comment: insufficient detail of the method provided. |
| Allocation concealment (selection bias) | Unclear risk | Not described. |
| Blinding (performance bias and detection bias) Clinical outcomes | Low risk | Not described but judged to be low risk as assessment of clinical outcomes is unlikely to be influenced by lack of blinding. |
| Blinding (performance bias and detection bias) Functional outcomes | High risk | Not described but judged to be high risk as assessment of some functional outcomes is likely to be influenced by lack of blinding. |
| Blinding (performance bias and detection bias) Nutritional outcomes | Low risk | Not described but judged to be low risk as no nutritional intake outcomes assessed. |
| Blinding of participants and personnel (performance bias) All outcomes | High risk | Not described and likely that researchers and participants were aware of group allocation as this was a nutritional intervention. It is possible that assessment of some outcomes was influenced by lack of blinding. |
| Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described and likely that some outcomes might be influenced by lack of blinding of assessment. |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | Number of participants lost to follow‐up/deaths/dropouts fully described: intervention group 155/448(35%) (43 deaths,7 lost to follow‐up, dropouts 105); control group 135/498 (27%) (29 deaths,5 lost to follow‐up, dropouts 101). Attrition high but numbers and reasons similar in each group. |
| Selective reporting (reporting bias) | Low risk | Published protocol identified. All outcomes fully reported. |
| Other bias | High risk | Baseline characteristics were compared. The intervention group were frailer at baseline (MMSE score Clinical Dementia Rating score, NPI‐Q score and activities of daily living score and had a higher caregiver burden (Zarit scale). More participants in the intervention group were malnourished or at risk of being malnourished intervention group 7.8% malnourished and 51.5% at risk; control group 2.8% malnourished and 34.5% at risk. All of these factors have the potential to influence outcomes assessed. Assessment of risk of bias in cluster‐randomised trials (1) Recruitment bias: no (2) Baseline imbalance: frail status (3) Loss of clusters: no (4) Incorrect analysis: no (5) Comparability with individually randomised trials / different types of clusters: different types of clusters |