Parallel Session P14 - Screening, diagnosing and treating dementia
Detailed Programme and abstracts
Saturday, 2 October – 14.00-15.30 (Blue Room)
P14.1. Computational intelligence methods for automatic diagnosis of dementia
P. García Báez1, C. Fernández Viadero2, S. Rodríguez Moreno3, C.P. Suárez Araujo4
1 Dep.de Estadística, I.O. y Computación, Universidad de La Laguna. Spain
2 Unidad de Atención a la Dependencia de Santander, Gobierno de Cantabria, Spain
3Hospital Univ. de Gran Canaria Dr. Negrin, Spain, firstname.lastname@example.org
4Instituto Universitario de Ciencias y Tecnologías Cibernéticas, Universidad de Las Palmas de Gran Canaria, Spain, email@example.com
The clear tendency of population towards aging, elderly people will reach 2500 million, on a global scale, in 2050, produces an increase of its associated diseases. The dementia is one of ones with higher prevalence. In a clinical context, dementia refers to a syndrome of acquired cognitive deterioration that can be associated with various potential stages of the illness. The two most common variations of this illness are Alzheimer type dementia (AD) with around a 50% of the cases, and Vascular type dementia (VD) between 20%-30%, an 10%-20% are mixed dementia and the rest for other type of dementia. All of these forms can be associated with different patterns of anatomical affectation, different risk factors, multiple diagnostic characteristics and multiple profiles of neuropsychological tests. This consideration, together with the fact that still there is no clear consensus on what is meant by an AD cognitive profile as opposed to a VD profile or other profiles, clearly complicates the clinical diagnosis of AD, VD and/or other dementias. Another important issue about theses neuropathologies, yet unsolved, is the absence of therapeutic tools that manage or stop its progression and symptoms, in a constant and supported way. All these handicaps highlight the need to develop new and alternative methods of diagnosis, placing special emphasis in differential diagnosis of dementias, and the diagnosis of the severity level of dementia.
We propose new computational intelligent-based diagnostic tools using a data fusion scheme and neural ensemble approach. These tools offer important advantages referring to other computational solutions based on artificial neural networks. They are as well as capable to perform diagnostic process with categorical character, which is a very important aspect in medical diagnosis. We have designed two HUMANN-S based systems, with capacity of processing missing data, the HUMANN-S ensemble system with simple and weighed majority voting as combination strategies. In this paper we explore the ability of these systems for differential diagnosis of dementia and for establishing a severity level of the dementia, using a battery of cognitive and functional/instrumental neuropsychological tests no reported.
Finally we carried out a comparative study between these systems for both diagnostic processes. The computational system for diagnosing severity level of dementia has a 95.88% of accuracy and for differential diagnosis of dementia (DDD) reached an accuracy of 88.76%, These results indicate the insufficiency of the use of independent neuropsychological tests and the need to use clinical multi-criteria for an effective and reliable DDD and the suitable of these battery of test to determine the severity level of dementia. Finally, we also carried out a comparative study between the intelligent systems for DDD and an expert clinic, which showed that a high level of performance was achieved using our proposal, with results that were 47% better than those from the specialist. Our methods can be seen to be very promising and will be applied towards building an effective system to aid in the diagnosis of dementia both in specialized care as well as in primary care.
P14.2. A survey of physicians’ perceptions of adherence to oral cholinesterase inhibitor treatment in patients with Alzheimer’s disease
Manfredi Ventura1, Murielle Esquenet2, Stefaan Vancayzeele3
1 Grand Hôpital de Charleroi, Belgium, Manfredi.Ventura@ghdc.be
2 N.V. Novartis Pharma S.A., Belgium, firstname.lastname@example.org
3 N.V. Novartis Pharma S.A., Belgium, email@example.com
Background: Alzheimer’s disease (AD) is characterized by a progressive loss of memory and cognitive functions, resulting in an impaired ability to perform activities of daily living. Cholinesterase inhibitor (ChEI) treatment of persons with mild to moderate AD provides benefits on cognitive, functional, and behavioural aspects of the disease. Continuous and prolonged use of ChEIs is important to sustain clinical benefits, but suboptimal drug adherence and lack of persistency are a major problem in the treatment of AD. Non-compliance may be due to forgetfulness and difficulty with multiple medication regimens, or discontinuation due to a perceived lack of efficacy or cholinergic gastrointestinal side effects. Transdermal administration of ChEIs (e.g. the rivastigmine patch) may increase compliance: it reduces the number of pills the patient has to take, offers the caregiver a visual reassurance that the medication is being taken, and has demonstrated a similar efficacy to oral ChEIs but with reduced incidence of nausea and vomiting.
Methods: This was a descriptive correlational survey of physician perceptions of adherence to oral ChEIs. Over a period of 90 days, willing physicians implemented a self-observational exercise of ten persons treated with oral ChEIs that were representative of those with mild to moderately severe AD seen in their practice. Physicians provided summary data related to ChEI adherence, treatment decision in the case of non-adherence, and treatment preferences (i.e. oral versus patch). At baseline and at the end of the study, the physicians completed a questionnaire concerning their practice patterns and perceptions with regard to the management and treatment of patients with mild to moderately severe AD.
Results: A total of 408 physicians participated in this study. Of these, 318 were GPs and 90 were specialists (58% of which were neurologists). The physicians estimated a similar number of patients (± SE) with good compliance both before (67.1% ± 1.15 for GPs; 69.0% ± 1.86 for specialists) and after (65.8% ± 1.28 for GPs; 71.2% ± 1.82 for specialists) the observation period. These values were in accordance with the mean number of patients per physician with good compliance reported during the observation period. Before the observation period, both GPs and specialists had the impression that most (87.4% for GPs; 92.2% for specialists) of their patients stay on oral ChEI treatment for more than 1 year. While their perception decreased slightly following the observation period (79.5% for GPs; 82.3% for specialists), their perceptions were still greater than the observed findings (< 60% of patients). During the observation period GPs and specialists reported that 65.0% and 67.7% of their patients received an optimal therapeutic dose of oral ChEI. The physicians overestimated the number of patients who were receiving an optimal therapeutic dose both before (67.0% ± 1.05 for GPs; 71.9% ± 1.86 for specialists), and to an even greater extent, after (69.8% ± 1.04 for GPs; 73.8% ± 1.54 for specialists) the observational period. The most preferred intervention when the physicians observed non-compliance was to ‘change the ChEI to transdermal ChEI’.
Conclusion: Physicians’ assumptions regarding the treatment of persons with AD may not always be accurate. As one third of patients may be at risk of suboptimal compliance and/or may not reach the target dose of oral ChEI physicians should explain the benefits of reaching optimal therapeutic doses, encourage caregivers, and consider ways to increase compliance.
P14.3. Memantine in everyday clinical practice: Comparisons of data from observational studies in Greece and in Germany
Dr. Stamouli Sofia, "Eginiteion" Psychiatric Hospital, Athens
Prof. Dr. Hans Förstl, Klinik und Poliklinik für Psychiatrie und Psychotherapie der TU München, Ismaninger Str. 22, 81675 München, D
Dr. Wolfgang Janetzky, Lundbeck GmbH, Karnapp 25, 21079 Hamburg, D
Dr. Karageorgiou Clementine , General Hospital of Athens "G. Gennimatas", Athens
Dr. Tzanakaki Maria, General Hospital of Chania "Agios Georgios", Crete
Dr. Papalexi Eugenia, Lundbeck Hellas SA, Athens
Dr. Galanopoulos Andreas, Lundbeck Hellas SA, Athens
Introduction: Observational studies were performed both in Germany and Greece to evaluate the effectiveness and safety profile of memantine as a treatment in people with Alzheimer’s Disease (AD) in everyday clinical practice. A pooled analysis of the two studies could possibly add further insights in the everyday clinical use of memantine in the two countries.
Methods: Effectiveness of memantine was evaluated using the Mini Mental State Examination (MMSE) in both studies and in order to assess functional performance, the 8-item Instrumental Activities of Daily Living (IADL) scale as well as the Instrumental Activities subscale of NOSGER was used in the Greek and the German study respectively, at baseline and after 3 and 6 months. IADL scale score was converted into percentage with higher scores indicating greater functional impairment. Safety was evaluated by collecting spontaneously reported adverse events (AEs). Statistical analyses have been performed for the Intent-To-Treat (ITT) dataset that includes people who had the baseline and at least one post-baseline measurement of the relevant scale and the Per Protocol Set (PPS) that includes persons who had had measurements at all 3 visits. An analysis finding was considered reliable only if it was demonstrated in both sets.
Results: Overall, 4305 individuals participated in both studies, 56.7% were males, mean age was 75.5±7.5 years and the mean MMSE score at baseline was 17.1±5.5. People from the German study were significantly older and at a more severe stage of AD compared to the Greek sample (p<0,001). Significant differences (p<0,001) were also found in the number of persons receiving previously treated with AChEIs. (34,2% of Greek vs 16.9% of German population).
Effectiveness: Study participants improved from baseline (17,1) to 3rd visit (19,1- MMSE ITT and PPS), and at six months of treatment 82,5% of them either improved or remained stable (MMSE). Concerning functioning (IADL and NOSGER), the results were similar with overall 4% points average improvement between baseline and 3rd visit. No differences were identified in the magnitude of change in either the Greek or the German study.
Safety: 14.8% of the total sample discontinued prematurely from the study. Premature discontinuation of treatment was significantly higher in the Greek (18,9%) compared to the German population (8.3%) and was attributed to visit non-attendance. Out of 3910 persons in total, 245 (6,3%) reported an adverse drug reaction (ADR). The percentage of ADRs recorded in the Greek sample was more than twice of the corresponding percentage in the German sample (8,4% vs. 3,6%).
Conclusion: Memantine displayed similar effectiveness in both a German and a Greek observational study of people with Alzheimer’s Dementia. In addition, differences were detected in premature discontinuation and reported ADR.
P14.4. MemoVie: Prospective evaluation of neuropsychological and biological characteristics of mild cognitive impairment and of associated subclinical health problems
Magali Perquin, CRP Santé, Luxembourg, Magali.Perquin@crp-sante.lu
The multidisciplinary MemoVie project, approved by the National Committee of Ethics in Research (CNER) and authorised by the National Commission for Data Protection (CNPD), aimed to investigate the conditions under which the normal cognitive capacities of the senior population in Luxembourg (aged ≥65 years) progress (1) to mild cognitive impairment (MCI) and (2) to Alzheimer’s disease-type dementia (ADTD).
Improving the knowledge on MCI features is a crucial challenge, because this stage should offer the chance for prevention strategies to be the most efficient in delaying or even preventing further cognitive decline, leading to relevant public health interventions.
Medical profiles, biological and genetic factors, education, multilingualism and socio-cultural parameters were examined in association with MCI and possible evolution to AD.
At baseline a cross-sectional analysis within the MemoVie prospective cohort study was conducted among 1,377 people, randomly selected from the General Inspectorate of Social Security (IGSS) register, according to the criteria defined by the protocol.
Because this sample has been calculated to represent the global population of people aged over 64, if the participation rate of these 1,377 invited persons is consistent with the feasibility study, the prevalence of MCI and/or ADTD in Luxembourg should be subsequently deduced from this first step survey. Once the MemoVie cohort is set up, it will be continuously supplied by new volunteers, and a yearly follow-up of the subjects will consequently allow improving the knowledge of the preclinical stage of MCI.
P14.5. Increasing the quality of care of behavioural disorders in Alzheimer’s disease and reducting the neuroleptics’ use for a better quality of life
Armelle Desplanques-Leperre1, Nathalie Riolacci-Dhoyen2, Benoit Lavallart3, and the national clinical platform « Prescribing in the elderly »
1 French national Authority for Health (HAS), 2 avenue du Stade de France, 93218 Saint Denis la Plaine, France - firstname.lastname@example.org
2 French national Authority for Health (HAS), 2 avenue du Stade de France, 93218 Saint Denis la Plaine, France - email@example.com
3 Mission pilotage Plan Alzheimer, Ministry of Health, 14 avenue Duquesne, 75350 Paris, France – Benoit.LAVALLART@sante.gouv.fr
Behavioural disorders – aggressiveness, shouting, agitation, perambulation - are very common in Alzheimer’s disease (AD) and impact negatively the quality of life of people and their caregivers. Neuroleptics (NL) are often prescribed in AD, in the objective of reducing and/or preventing these behavioural disorders. In 2007 the exposure to NL in AD was five times higher (16.95%) than in the general population of the same age (2.9%). But today we know that these drugs (i) are non effective neither for treatment nor for prevention (ii) have frequent and severe iatrogenic effects such as excessive sedation, gait disorders, strokes and additional deaths. To solve this abuse of NL and its deleterious consequences in this complex clinical situation the French Alzheimer’s public health plan has set up a specific programme called Alert and Mastering of the Iatrogenicity of neuroleptics in AD “AMI-Alzheimer”
The AMI-AD was developed by a national clinical platform including physicians, nurses, representatives of people with dementia, and institutions. Because of the need for a shift in care process, the objective of the programme was simultaneously to set up (i) tools for clinical teams: new guidelines about the non drug care techniques in behavioural disorders, professional education programmes for mastering the NL prescription (ii) information measurements of NL exposure in AD and of quality of care both in health structures and in ambulatory setting.
A large campaign informing the general public, caregivers, prescribers and managers is in process with a gradual dissemination of tools. Tools for clinical teams includes (1) identification of people with AD exposed to NL (local warning signal = percentage of people with AD taking NL). This indicator is easy to measure in all care sectors, (2) new guidelines on behaviour disorders in people with AD that describe proper use of NL and other psychotropic drugs as well as new nondrug techniques of care, (3) review of prescriptions for identified people with AD in order to correct inappropriate prescriptions (local indicator for risk containment = percentage of prescriptions reviewed for people with AD). Information measurements include (1) indicators about NL exposure of people with AD both in the hospital and nursing home accreditation and in the whole French elderly population (2) indicators of the quality of care with good practice indicators ‘absence of daytime sedation’ and monitoring of prescriptions for other psychotropic drugs to prevent drug switching.
The first results of the survey show a slight decrease of the exposure to 16.1% (equivalent to 3500 fewer cases of exposure to NL and their consequences). A formal launching will be set up in September 2010 with the publishing and the promotion and dissemination of a common booklet featuring all key messages and resources by all sanitary and professional organisations. This integrated programme aims to improve the quality of care thus reducing the behavioural disorders and the NL abuse leading to a better quality of life for people with AD.
Last Updated: Wednesday 03 November 2010