Well this is an interesting one. The abstract of a recent review by Carrion et al concludes that cognitive psychosocial interventions such as reality orientation and skills training improve cognition (Carrion, 2013).

Woohoo, I thought; fantastic!  And then I read the review…

Methods

The authors searched the usual bibliographic databases for randomised and non-randomised controlled trials, including the Cochrane Register of Controlled Trials (CENTRAL).  The search was carried out in 2010, three years before the publication of the review.  They restricted the papers to those written in English or with an abstract in English.  There was no attempt to identify trials with unpublished results.

There were two types of intervention falling under the definition of cognitive psychosocial interventions:

1. Reality orientation interventions: typically, the repeated presentation of orientation information to help the patient understand their surroundings
2. Skills training interventions: short-term face-to-face or software-based training programmes including activities like categorising objects or performing basic daily activities aiming to improve cognitive functioning and slow cognitive decline, albeit temporarily

Results

The researchers found 17 randomised trials meeting the inclusion criteria; nine of reality orientation interventions and eight of skills training interventions.

Of the nine trials on reality orientation:

• All trials found better cognitive function in the intervention group,
• But only six of the nine showed a statistically significant results and
• Two trials were methodologically weak and therefore at high risk of bias.

What good is statistical significance to a dementia patient?  Surely, it’s clinical significance that counts!

So, onto the skills training interventions – perhaps there would be better news there.  Nope. Even worse:

• Positive effects were found in “most” of the trials; no actual number given,
• But only two “achieved” statistical significance and only one of those was of good methodological quality.

In general, the conclusions that can be drawn from this review about effectiveness are limited by the extreme variation in the characteristics of the interventions, the patient population in terms of age, stage of disease progression etc, duration of intervention, tools used to measure cognitive outcomes, and of course the methodological quality.  Quite rightly in my view the researchers judge that a meta-analysis is inappropriate.

Conclusions

But after all the discussion about these huge limitations and lack of clinical significance, the researchers conclude that:

Stimulation of cognitive functions, especially by means of reality orientation, improve overall cognitive function (measured by MMSE or ADAS-Cog) in patients suffering from dementia.

This is a slightly toned-down conclusion than is reported in the abstract, but not much. I really think they should have stuck a “may” in there before the word “improve”, but I guess I am just a miserable old Cochrane square.

I am biased (sacrilege, I know) and I prefer the more focussed approach of the Cochrane Review on Psychosocial Interventions (Richter, 2012) which looks at the specific (and patient-centred) outcome of reduction of antipsychotic medication.  This is less vague, and although it concludes as pretty much all reviews do that “more high-quality research is needed”, at least it draws attention to the fact that the one included study of high methodological quality showed the most positive effect.  So despite a very Cochrane-y cautious conclusion, to my mind, the Cochrane Review gives a much more constructive and positive message.

I would take the message of the Cochrane Review further and recommend that the high-quality trial (Fossey, 2006) should act as a blueprint for future researchers who may hopefully replicate the positive result and strengthen the evidence of effectiveness for what looks like a promising intervention for reducing the use of antipsychotic medication.

Links

Carrion C, Aymerich M, Baillés E, López-Bermejo A, Cognitive Psychosocial Intervention in Dementia: A Systematic Review. Dement Geriatr Cogn Disord 2013;36:363-375

Richter T, Meyer G, Möhler R, Köpke S. Psychosocial interventions for reducing antipsychotic medication in care home residents. Cochrane Database of Systematic Reviews 2012, Issue 12. Art. No.: CD008634. DOI: 10.1002/14651858.CD008634.pub2.

Fossey J, Ballard C, Juszczak E, James I, Alder N, Jacoby R, Howard R. Effect of enhanced psychosocial care on antipsychotic use in nursing home residents with severe dementia: cluster randomised trial. British Medical Journal 2006; 332(7544): 756–61.