Methodological issues in the non pharmacological treatment of BPSD in nursing home — the TNM study —

Abstract

Behavioural and psychological symptoms of dementia (BPSD) such as agitation, aggressiveness, oppositional behaviour and psychotic symptoms are often reported by caregivers and staff members in institutions for the elderly. Taking into account the limited efficacy and the importance of side effects observed with psychotropic agents, the majority of existing guideline underlines the importance of non-pharmacological strategies. The aim of this article is to describe the major methodological problems in non-pharmacological treatments studies and to present the design of the TNM study, a nursing home randomized controlled trial conducted in 16 French nursing homes.

Behavioral and Psychological Symptoms are major and frequent manifestations of Alzheimer’s disease (AD) and other dementia. They are responsible for a large share of the suffering of patients and caregivers, and they strongly determine the patient’s lifestyle and management. Most patients, over the course of the illness, develop behavioural and psychological symptoms of dementia (BPSD) (1) which are associated with patient distress, impairment in functionality (2).

BPSD are highly common (50-80%) among people with dementia (3, 4), their prevalence increase as the disease progresses (5) and they are associated with earlier institutionalisation (6). Taking into account the limited efficacy and the importance of side effects observed with psychotropic agents, the majority of existing guideline underlines the importance of non pharmacological strategies.

Non pharmacological treatment have distinct aspects and goals (reality orientation, cognitive rehabilitation, cognitive stimulation, reminiscence therapy) (7, 8, 9, 10). Those strategies are largely used in memory clinics, daily care centres, rehabilitation department, day clinic of geriatrics departments, by speech therapists, special care units, nursing or residential homes for elderly people, but there is little evidence of their efficiency at short or middle term.

Several systematic literature reviews have been done in order to rate the effectiveness of non-pharmacological treatment for outpatient and patient living in nursing home. Livingston et al. (11) identified a total of 1,632 studies and only 162 satisfied the inclusion criteria for the review. Specific types of caregiver and residential care staff education and possibly cognitive stimulation appear to have lasting effectiveness for the management of BPSD. The authors indicated that lack of evidence regarding other therapies is not evidence of lack of efficacy. Conclusions are limited because of the paucity of high-quality research (12, 13).

Figure 1 summarize the most important methodological problems suggested by the literature review (12).

Figure 1.

Methodological weakness in non pharmacological therapeutics studies

Randomisation is one of the major difficulties because investigators want to use the same model as in pharmacological studies even if non-pharmacological therapies are not of the same nature. In fact, randomisation of patient within the same nursing home is not very pertinent taking into account that intervention group effects could be generalized and then work indirectly on the residents of the control group.

Diagnostic is another point to be underlined. In France it is difficult to be sure of the diagnosis which has been done before the arrival in the institution. One option is to systematically verify the diagnosis at the inclusion in the study. Criteria of efficacy are different than in cognitively oriented pharmacological studies. In nursing home BPSD are the most important criteria and BPSD are present in various type of dementia. This is in favour of transnological design, in other word, to include in the studies not only AD, but also patients with other dementia.

One other problem comes from the lack of standardisation in non-pharmacological strategies. This limits the understanding of the results and furthermore limit the replication of the technique in real life after the study.

The TNM study design

In France, Alzheimer’s disease (AD) was selected as a National Cause in 2007 and a National Plan has recently been developed. However, the number of clinical trials devoted to BPSD in nursing homes is rather limited.

The TNM in EHPAD study (non pharmacological treatment study for BPSD in nursing home) was designed at the initiative of the French ministry of health. The present document will developed the design of the study.

The aim of this study was to evaluate the effectiveness of staff education as a non-pharmacological intervention to manage BPSD in older people with a diagnosis of dementia. The training programme was designed both to provide advice on how to manage specific BPSD and to encourage nursing home staff to propose non-pharmacological solutions. It was hypothesised that staff education and training would have a positive impact on residents’ behaviours when compared to a control group.

Population

The study was conducted in 16 nursing homes in two French regions (Alpes Maritimes and Gironde). In each region nursing homes were randomly selected from the list of all the known nursing homes and divided according to the type of administrative organisation (private, public and associative). A majority of residents in each unit experienced dementia. The staff/resident ratio ranged from 0.09 to 0.45. A member of the study approached the Directors of the selected nursing homes and all agreed to take part in the study. An open questionnaire was then mailed to the physician in charge of each nursing home. Each physician, in consultation with staff members, was asked to select residents they considered suitable for inclusion in the study. The questionnaire included demographic information, the patient’s diagnosis of dementia, the Mini-Mental State Examination (MMSE) score and the occurrence of some of listed BPSD. Criteria for inclusion were patients with a diagnosis of dementia according to the ICD 10 criteria (14), an MMSE score ≤ 24 and presenting at least one of the following BPSD at least once a week: opposition, denial of care, aberrant motor behaviour, agitation, delusions, hallucinations or screaming. The project received institutional approval from the Ethics Committee of Nice. As the study was classified as a routine clinical procedure, no signed informed consent was required from patients or caregivers.

Design

Nursing homes were randomly assigned to the control group or to the intervention group. This design was chosen in order to avoid contamination of the intervention. None of the nursing homes shared facilities or staff.

The nursing homes assigned to the control group were informed to take care of the patients as usual with their own practice and care processes. It was only mentioned that the purpose of the study was to assess behavioural disturbances regularly with an independent rater.

Staff training programme

Two independent professionals with extensive experience of working with demented patients conducted the training programme. In each nursing home the programme began with a teaching session on dementia, BPSD and the use of ‘how to’ instruction cards (Staff Instruction Cards). There were four instructions cards, summarising practical advice on how to deal with BPSD. They were designed in order to be small and resistant enough to be easily carried by staff members. The first card gave general guidelines on what to do and what to avoid when faced with opposition, denial of care, aberrant motor activity, agitation, aggression, delusions, hallucinations or screaming. The second card explained how to act during the day to avoid or to decrease the emergence of BPSD, such as what to do at the patient’s bedtime or during meals. The other two cards provided recommendations on non-pharmacological interventions, giving examples and ideas for mini interventions designed to deal with individual instances of BPSD.

The remaining part of the training programme consisted of individual and interactive sessions in which trainers provided constructive feedback on how staff members dealt with BPSD. They also emphasised the importance of using the instruction cards in daily practice. The trainers were at each staff member’s disposal, rather like a coach, for two hours twice a week during the first month and then once a week during the second month, thus providing an opportunity for more personalised training, advice and feedback.

Measurement

Assessments were performed by four psychologists blind to the intervention condition and previously trained in the use of the assessment tools. None of them participated in the staff-training programme. Each psychologist had in charge the same nursing homes during the study. Data on residents were collected from nursing staff and they were asked not to talk about the intervention.

Before assessments the first step at baseline was to confirm screening data and to collect demographic, clinical and therapeutic information.

For outcome measures, data were collected at baseline (W0), week eight (W8) corresponding to the end of the training programme and at week twenty (W20), namely 3 months after the end of the training programme. Assessment tools were the Neuropsychiatric Inventory (NPI) (15) which was particularly interesting for the screening. Two other instrument were chosen for their possible better sensitivity to change; the Cohen-Mansfield Agitation Inventory (CMAI) (16), the Observation scale (OS) and the Quality of Life –Alzheimer’s Disease scale (QoL-AD) (17).

The NPI nursing home version is an interview-based instrument designed to elicit information from an informal caregiver to evaluate behavioural disturbances. The 12 NPI items were divided into four subgroups: Psychotic, Hyperactivity, Apathy and Affective subgroups. In the present study, targets were the Psychotic subgroup, including NPI items hallucinations and delusions, and the Hyperactivity subgroup, including NPI items agitation, euphoria, disinhibition, irritability, and aberrant motor behaviour. The scores for each subgroup were the sum of the frequency x severity for each of the NPI items.

The CMAI is an interview-based instrument designed to measure the frequency of 29 behaviours as observed by the caregiver over the previous two weeks. In addition to the global score, it is also possible to consider four subscale scores for physically aggressive and non-aggressive behaviour, verbally aggressive and non-aggressive behaviour.

The Observation scale (OS) is a scale derived from the Agitated Behaviour Mapping Instrument (18) and was specifically developed for the study in order to assess behavioural disturbances directly through patient observation. Clinical raters observed the patient for 3 minutes. The OS comprises 25 items describing positive BPSD, especially agitated behaviours (e.g. screaming, hitting, tearing things, making verbal sexual advances, biting): the higher the score is the more severe the patient’s behavioural disturbances are. For this scale the assessment was done at baseline, W8 and W20.

Lastly, the QoL-AD is a brief measure designed specifically to obtain a rating of their quality of life from the patients themselves. For this scale the assessment was done at baseline, W8 and W20.

Statistical analysis

The study is a 2 (intervention conditions) x 4 (assessments) nested partial crossover control group design. Statistical analysis was carried out using the SPSS program. An independant statistician performed the analyses.

First result

The study was conducted in all nursing homes between October 15 and December 15 2007, with the 3 months follow-up being performed in March 2008. The study was planned so as to avoid assessments during the holiday period at the end of the year.

Through the inclusion process 306 residents were selected for the study from among the 1369 residents living in the 16 nursing homes, that is 22%.

In order to check the frequency of patients presenting the most severe symptoms a second selection was done using as cut of a score higher than 7 on the frequency x severity NPI score for items belonging to the hyperactivity or psychotic symptoms group. 222 patients representing 39,9 % of the patient included in the study and 9% of the total number of residents living in the nursing in the selected nursing homes responded to this criteria. Hyperactivity symptoms were more frequent than psychotic symptoms.

Table 1 shows relationships between the NPI, CMAI and OS scores. As expected the correlation were highly significant between the scales using the caregiver assessment point of view (NPI frequency x severity score, NPI distress score, CMAI total score). The correlation between the caregiver assessment scales and the observation scale using an independent clinical raters observation are lower. It is reasonable to think that the different points of view are true and complementary. This underlines the importance to have a multiple types of assessment in behavioural oriented studies.

Table 1.

Correlation between BPSD assessments at base line

	CMAI Total score	NPI Hyperactivity Frequency x Severity	NPI Hyperactivity Distress score
Observation scale	.245 **	.201**	.118*
CMAI
Total score		.655**	.632**
NPI Hyperactivity
Frequency x Severity			.901**

Pearson correlation coefficient: * p < .05 ** p < . 01

Non-pharmacological studies are important in France, where knowledge about BPSD in nursing home and proof of the efficacy of non-pharmacological methods are lacking, while at the same time professional guidelines state, “For patients living in an institution, management of BPSD is part of the institutional plan. Care teams have a capital role in the evaluation, management and follow-up of BPSD. Non-pharmacological management and support of the patient should be preferred and should be undertaken in usual practice” (6).

We hope that TNM study demonstrated the feasibility of conducting such non-pharmacological randomised studies in nursing homes in France.

Acknowledgements: The “TNM en EHPAD” (Traitement Non Pharmacologique en Etablissement pour Personnes Agés Dépendantes) study is a national multicentre study supported by a grant from the French Ministry of Health (Direction Génerale de la Santé) and the Fondation Méderic Alzheimer.

Financial disclosure: None of the authors had any financial interest or support for this paper.