interventions to delay functional decline in dementia

A recent review was done which looked at 23 systematic reviews (ie, a systematic review of systematic reviews) of interventions to delay functional decline in patients with dementia, finding a few nonpharmacologic (exercise, dyadic interventions) and a few pharmacologic ones (especially donepezil and memantine) seemed to be effective (see dementia delay decline bmjopen2016​ in dropbox, or doi:10.1136/bmjopen-2015-010767). Details:

--background re: dementia

                --35.6 million people are affected by dementia worldwide, including 5-7% of those >60yo

                --one of the leading causes of morbidity and mortality

                --huge personal and social costs: reduced quality of life, increased care burdens on family, increased costs to society to provide supportive and medical care

--interventions:

                --most patients in the reviews had mild-to-moderate dementia, 65% with any type of dementia and rest with Alzheimer's

                --nonpharmacologic (11 reviews). 

                                --exercise (6 studies, 289 people, standardized mean difference (SMD) of those doing exercise was 0.68: the largest magnitude of effect of any intervention, though studies considered low quality because of risk of bias.

                                --dyadic interventions (interventions targeted to patient and caregiver to maximize the patient's quality of life and meaningful activities): 8 studies, 988 patients, SMD 0.37. somewhat mixed findings in the studies, and many with risk of bias, so considered low quality evidence

                                --those not reaching statistical significance included: cognitive training, cognitive stimulation therapy, light therapy, aromatherapy, nutritional supplements, validation therapy, psychological treatment, or music therapy

                --pharmacologic (reviews)

                                --acetylcholinesterase inhibitors and memantine (12 studies, 4661 patients, including a variety of meds and doses: eg donepezil 5 or 10mg, galantamine 24mg, and rivastigmine 12mg): a pretty small but significant effect, with SMD in the 0.15-0.20 range). Moderate quality of evidence.

                                --Selegiline also had small effect on Activites of Daily Living (ADL) function after 8-17 weeks (7 studies, 810 patients, SMD 0.27)

                                --other therapies (meds for sleep, latrepirdine, melatonin, statins, lecithin, nimodipine, acupuncture, vitamin B supplements) were ineffective. Gingko biloba and Huperzine A did have effect (which was really remarkable for Huperzine A, with SMD of 1.48), but these treatments had such low quality that they could not be recommended, and the latter one was based on 2 studies with a total of 70 people). The studies on Ginko Biloba were more extensive (7 studies with 2530 people, SMD 0.36), the quality was very low and the studies had mixed results.

COMMENTARY:               

--these types of reviews clearly are pretty large-scale data mining of studies of variable quality looking at different people (ages, types and severity of dementia), different interventions within the subgroups (different exercises, etc), different outcomes measured (memory changes, or perhaps the more clinically/socially-relevant changes in ADL), sponsors (drug company vs not), sites (community-dwelling vs in institution), and qualities of the studies (as measured by the AMSTAR tool, assessing protocol data extraction, comprehensiveness, data synthesis, biases, documentation of conflicts of interest,…). But that's all we got. though I should mention again that unfortunately there are fewer studies and fewer patients in the non-pharmacologic arena, likely because of less available funding (most studies being funded by drug companies), with about ten-fold more patients in the pharmacologic studies reviewed.

--perhaps not so surprisingly, the two standouts here are exercise and dyadic therapies.  In my limited experience, I have found that social day programs (these are adult day centers that I have many patients attend, with patients having social interactions, social activities, exercise, and some help with training family members in the care of patients) are really effective. I do need to lapse into one very recent anecdotal experience: a 78yo man who recently retired and after a short period of time, was forgetting how to find his house when either walking or driving (he soon stopped driving). I did start him on antidepressants, but the social day program was pretty immediately effective in both improving his mood dramatically and also improving his memory back to his baseline. Though I can't draw unequivocal conclusions, I do find that overall one of the most important clinical interventions I do for my somewhat isolated older patients is to get them into these types of programs, encourage them to do more exercise, and help/support /get services for their caregivers.

--so, bottom line to me: the most effective therapies for mild-to-moderate dementia (and perhaps those with the least potential for adverse effects) seem to be exercise and helping the patient/caregivers provide the most useful interventions to improve the ADLs, provide meaningful activities for the patient, etc