Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Home visits versus comprehensive geriatric assessment in a clinic
- George A Heckman (29 September 2001)
-
Meta-analysis a step too far
- Stephen Conaty (10 October 2001)
-
Preventive home visits for mental health
- Kallur Suresh (22 October 2001)
-
Primary care assessments in general and not only home visits reduce mortality among older people
- Ulrike Junius (27 October 2001)
-
What the box adds
- Lindsay Forbes (7 November 2001)
|
|
|||
|
George A Heckman, Clinical scholar MacMaster University
Send response to journal:
|
My response is actually meant to be a question to the authors of the meta-analysis. I apologize if this is not the appropriate forum. As a geriatrician, I perform comprehensive geriatric assessments in a clinic setting. A cognitively intact caregiver is always present. In the context of the lack of geriatric specialists in Canada, clinics have the advantage of maximizing the number of patients that can be seen by a specialist. Clearly, some patients cannot attend a clinic without great difficulty, in which case the assessment takes place in their home. In the studies that were included in the meta-analysis by Elkan et al, what types of interventions did the control groups undergo? Did they receive "usual care" (whatever that means) or did they undergo a comprehensive geriatric assessment by a geriatric specialist? In other words, does the meta-analysis suggest that all comprehensive geriatric assessments by a geriatrician be done in a patient's home, or do the results simply reflect the benefits of a proper and comprehensive geriatric assessment compared to usual primary care? George A. Heckman |
|||
|
|
|||
|
Stephen Conaty, Lecturer in Public Health Department of Primary Care and Population Science
Send response to journal:
|
EDITOR - Clarke in editorial comment (1) on the meta-analysis by Elkan and colleagues (2) (Sept 29) doesn't get it quite right: preventive home visits may usefully be evaluated by randomised controlled trials but meta-analysis is a step too far. Unless we have a reasonably discrete intervention, measurable and comparable across studies, and a plausible pathway through which our intervention influences our endpoint of interest then meta-analyses should not be attempted. This is because, although we may find an effect (such as reduction in mortality), we don't know the relevant features of the intervention that is producing the effect we observe and generalisation becomes hazardous. To adapt Clarke's analogy when it comes to preventive home visits we are trying to measure the effect of giving 10 different drugs, at dosages that we are failing to measure, on global endpoints. We are left with more questions than answers: Can you do the same in the clinic or is a home visit essential? What happens if you leave out this part of the home assessment? Does it work if you do it once a year? Does it have to be the same person doing it? Eggar (3) says as much in the commentary accompanying the meta- analysis of Elkan and colleagues, however, the hope that better studies and meta-regression will tease out the relevant features is pious. Studies (RCTs) tomorrow may well find no effect because other factors emerge such as changes in family support, or changes in attitudes to seeking health advice. Sources of heterogeneity such as these are numerous and difficult to measure. Instead of looking for the signal above the noise in different studies conducted across different health systems at different times, we should be looking closely at what works in our own context. Preventive home visits are just one way of applying a range of interventions to reduce disability in the elderly. The merits of this approach should be judged synthetically using available evidence of the utility of discrete interventions backed up by good quality observational studies. This is a reductionist approach but it frees us to be creative about service development. The alternative is accepting one way (home visits) as necessary because of some mysterious property that we cannot quite identify. We need to be creative about prevention in the elderly because, ironically, as Clark points out, health visitors, under increasing work pressure, mostly withdrew from this kind of activity a decade ago. Stephen Conaty
Steve Iliffe
Department of Primary Care and Population Science, Royal Free and University College Medical School, Royal Free Campus, London NW3 2PF 1. Clark J. Preventive home visits to elderly people [editorial]. BMJ 2001; 323: 708. 2. Elkan R, Kendrick D, Dewey M, Hewitt M, Robinson J, Blair M, et al. Effectiveness of home-based support for elderly people: systematic review and meta-analysis. BMJ 2001; 323: 719-24. 3. Eggar M. Commentary: When, where and why do preventive home visits work. BMJ 2001; 323: 724-5. |
|||
|
|
|||
|
Kallur Suresh, Specialist Registrar in Psychiatry of the Elderly Royal London Hospital (Mile End), London E1 4DG
Send response to journal:
|
The concept of preventive home visits to the elderly in order to reduce their long-term morbidity is interesting, but raises the important question of affordability of such visits. I refer here to the mental health of older people rather than their physical well-being, although each is significantly influenced by the other.
Given the fact that there are hardly any specific preventive measures that are known to work in mental health, it is difficult to envisage the effectiveness of such an intervention. Severe resource constraints mean that community teams can hardly cope with existing case-loads dealing with current service users, let alone planning a preventative visit in the presymptomatic stages. There may be several areas for targeted interventions such as education, information dissemination, provision of self-help and support groups, modification of life-styles to reduce risk factors for specific illnesses, and focusing on positive physical and mental health rather than 'absence of disease'. It would be a welcome development if such preventive visits can be undertaken for mental health problems but at present it only remains a distant dream. |
|||
|
|
|||
|
Ulrike Junius, Lecturer Department of General Practice, Medical School Hannover
Send response to journal:
|
Preventive primary care assessments in general and not only home visits reduce mortality among older people Junius U, Breull A, Jones D Elkan et al. conclude in their review that preventive home visits reduce mortality and institutionalisation in older people (1). However, home visits are just one method of assessing older peoples` needs in primary care. In practice, preventive “health checks” can be conducted in various ways. Using a more general approach we performed a meta-analysis and added two more types of geriatric team assessments in outpatient departments and assessments performed in general practices. Preventive primary care assessments in general were found to have a significant positive effect on mortality. Unlike Elkan et al. we observed that the pooled mortality data for targeted frail older people did not show a significant survival benefit, whereas the reduction of mortality in the larger group of unselected people was significant. Subgroup analyses of three assessment strategies also showed positive - but non significant - survival effects. The practice assessment method (strategy 4) revealed a non-significant increase in mortality (see. fig. 1). In total, we reviewed 29 RCTs and categorized them according to recruitment strategies, nature and content of assessments and interventions. Principally all primary care assessment RCTs that recruited older people living at home were included. Excluded were studies with participants enrolled during a hospital stay or those with non/one measure of health outcome or non/one specific intervention. 11 trials selected frail older people (N = 2.757) whereas 18 studies recruited unselected older subjects (N = 17.915). Primary care assessments could be classified into four strategies: Strategy 1 features a comprehensive assessment approach for targeted frail patients conducted by a geriatric team in outpatient departments (OPD) with the same team being responsible for interventions (Burns, Toseland, Engelhardt, Rubin). Strategy 2 is similar; but it has a solely consultative approach with referral to general practitioners for subsequent management (Reuben, Silverman, Schrijnemaekers, Fordyce, Epstein). Strategy 3 (Dalby, Stuck, Jitapunkul, Stuck, v.Rossum, Hall, Vetter a) und b), Hendriksen) comprises home visit schemes with nurses or health visitors undertaking functional and preventive assessments with follow-up visits. Strategy 4 (Wasson, Hay, Moore, Wagner, Tulloch) involves mainly unselected people for a brief “screen” in general practices. A self-rating questionnaire is used for problem identification and neither care plans nor follow-ups are part of this strategy. Unlike Elkan et al. we excluded home visit trials like Fabacher`s and McEwan`s studies because of the lack of follow-up visits. Carpenter used volunteers and Pathy`s description of the nature of assessment was too vague to permit classification. German`s study in general practices with the large sample of over 4000 people did not fit into strategy 4 as both assessments and follow-ups were performed by general practitioners according to a care plan. Even with pooled data, subgroups are often too small to show significant results for survival. At present, the challenge of identifying components that are responsible for positive health effects in primary care geriatric assessment (2) remains. Other health outcomes, such as self -rated health, well-being and functional abilities will need more attention in future. (1) Elkan R, Kendrick D, Dewey M et al. Effectiveness of home based support for older people: systematic review and meta-analysis. BMJ 2001; 323:719-24 (2) Egger M. Commentary: When, where, and why do preventive home visits work? BMJ 2001; 323: 724-5 Fig 1: mortality data from primary care preventive assessment trials: log odds ratios |
|||
|
|
|||
|
Lindsay Forbes, SpR in Public Health Medicine Bexley, Bromley and Greenwich Health Authority
Send response to journal:
|
Dear Editor What is already known on this topic -- The BMJ’s publication of a box in each article with the titles What is already known on this topic and What this study adds is potentially very useful for information-overloaded readers -- The content of these boxes is written by authors of the article What this letter adds -- The box in a recently published systematic review of effectiveness of home based support for older people gave incomplete information1 -- The box reported only the positive findings of better outcomes on mortality and admission to institutional care and did not include the finding that meta-analysis of other outcomes (hospital admissions, health status, functional ability) gave important negative and inconsistent results -- This casts doubt on the effectiveness of home based support -- The contents of the box in articles accepted by the BMJ should be subject to particularly rigorous peer review --The box should include information on any difficulties with the study that mean that the conclusions should be interpreted with caution Yours faithfully Lindsay Forbes, SpR in Public Health Medicine
Peter Buck, Public Health Practitioner Jacquetta Goy, Clinical Governance Co-ordinator Hilary Guite, Consultant in Public Health Medicine Priscilla Ibekwe, SpR in Public Health Medicine Ruth Lawson, SpR in Public Health Medicine Alide Petri, SpR in Public Health Medicine Andy Scott Clark, Pharmaceutical Adviser Addisalem Taye, SHO in Public Health Medicine Bexley, Bromley and Greenwich Health Authority, Marlowe House, 109 Station Road, DA15 7EU 1. Elkan R, Kendrick D, Dewey, M et al. Effectiveness of home based support for older people: systematic review and meta-analysis. BMJ 2001;323:719-725 |
|||