Table 1

Description of ‘dimensions of complexity’ in the adapted study as per MRC's guidelines10

DimensionReason for complexity
Number of and interactions between components within the experimental interventionsSince the intervention will be delivered by the companion and we anticipate that no two companions or participants will have the same skills, interests or deficits, there may be a degree of variation in each therapy session. The therapy is based on this assumption and is designed to maximise outcomes by adopting a person-centred approach. Participants choose the activities they are most interested in from a library of over 60 cognitively stimulating topics. Companions subsequently personalise the activity through verbal, visual, tactile, auditory, gustatory or olfactory cues. Thus, the therapy subscribes to a structured, yet tailored, approach, delivering an intervention that is standardised with respect to dose and delivery but variable with respect to content.
Number and difficulty of behaviours required by those delivering or receiving the interventionStandardised companion training is critical to minimise heterogeneity. All companions will be trained to criterion and their skills monitored over time. This element of treatment fidelity will reduce the likelihood of non-significant results at the end of the study being attributed to poor training rather than an ineffective intervention. Companions will receive a minimum of 2 hours training and must have sufficient cognitive function (by not meeting clinical criteria for dementia) to be able to deliver the therapy. Companions’ ability to deliver the therapy will be self-assessed and researcher rated at the end of the training and monitored throughout the study.
Number and variability of outcomesAlthough the trial is powered to assess cognitive functioning as the main outcome, it is likely that there may be effects on different outcomes such as behaviour, companion variables, relationship features and health economic issues.
A good theoretical understanding is needed of how the intervention causes change, so that weak links in the causal chain can be identified and strengthenedThe background literature of evidence of potential mechanisms, as well as clinical experience, has suggested that each function of the intervention can be linked to identifiable intermediate impacts and final outcomes and can be outlined in a logic model32 33 which will be outlined elsewhere.
  • MRC, Medical Research Council.