|
|
|
|
|
|
Living Better: A personal
health research journey - the essential contributions of applied technology
Professor John C Beck MD Twenty Fifth Annual Oration: Living Better: A personal health research journey - the essential contributions of applied technology
Preamble
Preventing Loss of
Independence and Disability The aging of Australia’s population, which became evident in the 1970s can be attributed to the sustained decline in fertility which followed the post-war baby boom. Since this trend continues, Australia’s population is projected to continue aging into the next century. As the large baby boom generation ages, the number of older people is projected to increase rapidly with the youngest of the surviving baby boomers reaching 65 years of age in 2031. The population aged 65 and older is projected to reach 5.4 million (more than double the number in 1999) and will represent 22 percent of the total population (compared with 12 percent in 1999). As the youngest baby boomers reach 85 years of age in 2051, the population aged 85 years and over is projected to reach 1.3 million (more than five times the number in 1999) and represent 5 percent of the total population.1 The increasing prevalence of risk factors and chronic disease that precipitate loss of independence and disability associated with an aging population, is increasingly challenging to many developed countries, including Australia, to elders themselves, to healthcare providers, the broader healthcare community, and to society in general. The policies of the developing world are currently oriented toward systems of care for dependent elders, although as I will show, there are methods for delaying or preventing dependence which appear to be more effective as well as cost-effective. In relating my personal research journey directed towards preventing functional loss and disability and maintaining independence, I make two assumptions. The first is that, just as the present elderly cohort in the western world is more concerned about remaining independent rather than increasing life expectancy, the baby boomer generation will feel this equally strongly, and secondly, that individuals of all age groups, and particularly older persons, need to take more responsibility for maintenance of their health and independence. When I first directed my interests into the field of aging in the middle to late 1970s, I became aware of an international ‘giant’ in the field (a Scotsman named Williamson) who incidentally, indirectly has shaped some of your aging services in Australia. In 1964, he and his colleagues found that many of the multiple health problems and disabilities present in elderly people living at home in Scotland, were not known to their general practitioner. Williamson suggested that health visitors could undertake periodic visiting of old people and carry out screening and that medical and social measures, applied soon enough might prevent functional impairment.2 The first finding has been and continues to be true whenever and wherever it is examined. In our own experience, it holds for the USA, the United Kingdom, Denmark, Switzerland and Germany as recently as 2004. This observation in Scotland led my colleagues and I to ask an important research question in 1980. Could we find or develop evidence that once functional impairment or disability had occurred, could the process be reversed? We became aware of a longitudinal sample of non-institutionalised elderly in the state of Massachusetts in which data on function had been collected raising the possibility of answering the question. In 1983, we reported our findings and in so doing, suggested the concept of Active or Functional Life Expectancy in contrast to the traditional view of using mortality as an outcome. We were able, in this data set, to show that the duration of impairment was shorter for men than women (that is, from loss of independence to death) and that Active Life Expectancy showed a decrease with increasing chronological age. Furthermore, we were able to demonstrate that 24 percent of the population under study returned to independence over a 15 month interval.3 We could not identify why this latter intriguing phenomenon occurred. It however, raised the question of whether there was an ‘at risk’ population which could be identified for which preventive health and appropriate medical care could compress the duration of disability. Simultaneously (1980) I was aware that a Danish colleague (Hendriksen) had available to him a sample of community elderly with an excellent municipal data system for the possible measurement of outcomes. He and his colleagues were convinced to begin a three-year randomised control trial (RCT) of the impact of a home-based interview at three monthly intervals, identifying whether scheduled medical and social preventive services were being developed and facilitate their delivery. The outcome measures were general practitioner (GP) use, hospital and nursing home admissions and mortality. They demonstrated a reduction in emergency visits, hospital and nursing home admissions and mortality with no change in the number of GP visits.4 Assuming that functional decline and disability could be delayed or prevented, we next posed the question of whether it might reduce costs for medical and social services. My younger colleagues developed an epidemiological cost model to estimate the impact of interventions to decrease the incidence of functional decline. Three common causes for decline — stroke, hip fractures, and incontinence — were selected. As an example of the results, they showed that treating just half of older adults aged 65-74 with untreated hypertension would reduce the incidence of stroke by three percent (1,500 fewer cases annually) with major cost savings (1993).5 Comprehensive Geriatric Assessment (CGA) is a hands-on process aimed at detecting biologic, psychological, social, and environmental risk factors for disability, in common use by geriatricians and some other physicians worldwide. In 1987, we embarked on a sizable three-year RCT of annual CGA conducted in the home by specially trained nurse practitioners to detect risk factors for decline. The sample was population based and 75 years and older in age living in the city of Santa Monica, California. The assessment included functional status, oral health, mental status (presence or absence of depression and cognitive impairment), gait and balance, medications, body weight, vision, hearing, social network effectiveness, quality of social support, home safety, and ease of access to the environment. Risk factors were identified, prioritised and recommendations for solution given to the participants. Three-monthly follow-up visits by the nurse practitioners took place to reinforce the recommendations, facilitate compliance with them, and detect new problems. All the participants were encouraged to take an active role in their care and improve their ability to discuss problems with their physician. In 1995, we reported the results of this trial. These included convincing evidence of either the prevention or the delay in the development of disability and a significant reduction in nursing home stays among elderly people living in the community. The imputed costs of the intervention for each year of disability free life gained was about 6,000 US (1995) dollars.6 Additional analyses of this extensive data set also revealed that on average two new medical and one new social problem threatening loss of independence occurred per year in this population sample. There was a suggestion as well, that individuals with earlier impairments (lower risks) responded more favorably to the intervention than those with more severe impairments (high risk subjects). In addition, there was a suggestion that some of the nurse intervention staff were more effective than others. The Santa Monica trial raised new questions which we decided were best answered in Switzerland where a representative sample was available to us for study and which included accurate health care use and expenditure data, as well as detailed income information. The major new questions included: collecting additional information in a much larger sample of subjects as to whether the intervention was more effective in low versus high risk populations; an examination of differences in nurse performance; obtaining additional health and social care expenditures for a cost-effectiveness analysis; and finally to determine whether special educational efforts directed toward the physician care-givers and increased nurse-physician communication would further enhance the intervention. The experimental design was conceptually similar to our Santa Monica study, modified to address the questions which had emerged from that study. The Swiss study began in 1993, and in 2000, we reported that after three years the intervention reduced functional impairment and disability among elderly people at low risk, but not among those at high risk for functional impairment. The intervention effects were related to the effectiveness of the specially trained visiting nurses. In two out of the three nurses’ subjects involved in this study, there were major effects on all measures of functional status and a profound reduction in nursing home admissions, with a net cost savings of 1,403 US (1999) dollars per person per year. The third nurse had no favorable affects of the intervention and in fact, her population of patients showed a marked increase in nursing home admissions.7 The studies which I have shared with you have been replicated in a variety of nations and have come to be knows as Preventive Home Visitation Programs in the Elderly. They have become part of national policy in several countries, including Denmark (1985), the United Kingdom (1999), and Australia (1999). The rationale is to delay or prevent functional impairment and subsequent nursing home admissions by primary prevention (eg, immunisation and exercise), secondary prevention (eg, detection of untreated problems), and tertiary prevention (eg, improvement of medication use). However, the value of these home visitation programs has remained controversial. Although individual trials and earlier meta-analyses have suggested that some programs are effective, there is uncertainty regarding whether they can prevent functional status decline, which program components are effective and which populations are most likely to benefit. In an earlier meta-analysis of hospital-based comprehensive geriatric assessment programs, we found that programs including extended ambulatory follow-up were more effective than other schemes. A sub-group analysis of the trial of a home visitation program which I shared with you, suggested that older people with relatively good functional status at baseline were more likely to benefit. This was confirmed as I have related to you in the planned analysis of the Swiss trial, which showed favorable effects among individuals with low risk, but not among those at high risk for nursing home admission. Finally, evaluation in the home setting results in a high yield of undetected problems. Therefore, it seemed likely that successful programs should include multi-dimensional geriatric assessment as a basis for in-home prevention. We performed a systematic literature review and a new meta-regression analysis to evaluate the effect of preventive home visits on functional status, nursing home admission and mortality, and to test the hypothesis that they are beneficial if they are based on multi-dimensional geriatric assessment and frequent follow-up visits and conducted in individuals at low risk of functional decline at baseline. We reported in 2002, the results of this effort in which we identified 18 trials, including 13,447 individuals, aged 65 and older. We were able in this meta-analysis to show that preventive home visitation programs are effective, provided the interventions are based on multi-dimensional geriatric assessment and include multiple follow-up home visits and target persons at lower risk for death. Benefits on survival were seen in young-old rather than old-old populations.8
Development of a Health Risk
Appraisal System for Older Persons Thus, in the early 1990s, to lay the groundwork for devising, improving and implementing strategies to prevent or delay the onset of disability in the elderly in higher numbers, including the development of the Health Risk Appraisal (HRA) system, we conducted a systematic literature review of all longitudinal studies published between 1985 and 1997, that reported statistical associations between individual base-line risk factors and subsequent functional status in community-living older persons. We were able to identify 14 risk factors domains and to grade the strength of evidence for the contribution of a given domain to the risk for functional status decline. The highest strength of evidence consisted of cognitive impairment and depression as two examples, with the lowest strength of evidence being nutrition (not body mass index or body weight) and the physical environment. This major study was reported in 1998.9 HRA in younger persons has been a promising approach for the maintenance of health. It begins with the identification of specific factors in individuals that increase the risks of impairments, disability, and premature mortality and then develops some recommended strategies for minimising their impact. This is being used extensively in large corporations with younger work forces and has been proven to be highly cost-effective. Traditionally, HRA is a process that entails:
In translating this concept into a possible less costly method for the identification of risk factors as persons age, it seemed more relevant to focus on risk factors for functional impairment and disability rather than on mortality, as has been the custom in younger populations. A group of senior faculty at UCLA began developing a health risk appraisal instrument for the elderly with an aim to reducing the development of functional impairment in the mid-1990s. The first phase in this project was the development of a prototype Health Risk Assessment (HRA) questionnaire and feedback statements for preliminary testing. We sought to examine risk factors that could impact function, such as vision, hearing, memory, chronic disease and their consequences and medications. Risk factors were selected based upon the magnitude of the effect and its potential impact on functional impairment; feasibility of assessment; generalisability of assessment; generalisability of these assessments results; and the potential for risk reduction. This process, aided by the major systematic review which I have just described, led to a selection of the domains which were in the first version of the questionnaire or survey instrument. Algorithms and feedback statements regarding the results of the assessment of each risk factor and important risk factor combinations were developed to provide to the participants a personalised response based on the current evidence-based literature. The algorithms covered all possible combinations of questionnaire responses. Software using these algorithms were developed and alpha and beta tested. Thus, each person completing a questionnaire received a computer-generated personalised HRA feedback report, with a series of statements in each domain, including a generic statement about the problems identified, individually tailored suggestions for actions and the rationale for them, sources for further information and assistance, and a health care summary report, including a section dealing with how more effectively to communicate with physicians or other providers. In addition, a computer-generated summary was developed that could be sent to the health provider. The next phase of the project included a large field trial of the feasibility of using the questionnaire survey instrument and personal reports in three very separate populations, including a national sample, which reflected the range of settings in which it was envisaged that the HRA could be beneficially employed. Findings from this study of almost 2,000 respondents, indicated that there were a large number of unrecognised risk factors and that nearly all participants found the questionnaire easy to complete and were pleased with its overall length. In addition, most participants read the reports and many reported planning to take action based on the report’s recommendations.10 Our next step was thus to evaluate the impact of an HRA on health related behaviors, medical care utilisation, and functional decline in a RCT. The most suitable environment to implement such a study at the time was in Europe, and we were able to convince three groups of colleagues, one in the UK, the second in Germany and the third in Switzerland, to join in this effort. The first step was to scientifically update and culturally adapt the HRA instrument to English (UK), German, and Swiss-German language versions of the HRA. This updated version was then feasibility tested in three separate populations in London, Hamburg, and Bern, Switzerland. This version of the questionnaire contained the following domains: administrative information, self-reported chronic conditions, preventive care use, medication use, signs and symptoms, self-perceived health, physical activity, nutrition, injury prevention, tobacco use, alcohol use, vision, hearing, depressive symptoms, self-reported memory, social network, social support, functional status indicators, socio-economic information, and health measurements such as weight, height, blood pressure and cholesterol. The sample size in the three countries in the feasibility study ranged from 149-348 participants. The results of this study was comparable to the American feasibility study where the acceptance of the questionnaire and the computer-generated individually based reports for older people and their general practitioners was high. It identified a high number of potentially modifiable risk factors for functional decline and set the stage for multi-site randomised control study in Hamburg, London and Switzerland.11 The goal of this large RCT was to identify risk factors for functional status decline in older persons with the longer-term aim of preventing disability and minimising unnecessary service utilisation. The three RCTs recruited older community dwelling persons in Hamburg, Germany (N=2,580), a community near Bern, Swizterland (N=2,284), and London, UK (N=2,503). The intervention was based on the survey instrument which identified modifiable risk factors for disability and generated feedback statements to older persons and their healthcare providers. At each site, a specific intervention program was developed and adapted to the local healthcare system, including reinforcement in primary care practices, group sessions, and preventive home visits. The intervention consisted of training sessions for the participating primary care physicians, the HRA feedback to older persons and their primary care providers (largely general practitioners), and personal reinforcement either based on telephone follow-up and home visits dependent on the site. Primary outcomes were health behavior and use of preventive care based on self-report and chart abstraction. The London group practices were selected based on their having electronic medical records and the physician data describing risks were entered into the electronic medical record with the assumption that the participating physician would respond to these either alone or with the help of the general practice nurse, which is available to small groups of general practitioners in the UK. In Hamburg, the intervention targeted at the older participants was of two kinds. The first were group sessions where participants, after receiving their reports, joined in groups with participating physicians to reinforce the recommendations that had been made. The second group relied on the trained physicians reinforcing the risk factor findings and the recommendations in the office setting at subsequent visits by the participant to their general practitioners. In the Swiss trial, the personal reinforcement was based on telephone follow-up and home visits. Primary outcomes were health behavior and use of preventive care at two-year follow-up based on self-report and chart abstraction. Baseline data at the three sites indicate high prevalence rates of potentially modifiable problems and under use of recommended preventive care measures (as was found in the feasibility study) such as are seen in Table 1.
These base-line data reveal significant differences in prevalence rates of problems (risks) between the sites, and the high potential for intervention effects at all three sites. We have at the moment follow-up data at one year from the London and Hamburg sites, and Bern community data at two years. The one-year follow-up data includes information on self-reported health behavior, preventive care use, and functional status outcomes. In summary, there was no effect of the intervention in the London study. In the Hamburg study, beneficial effects were seen in the participants who elected the group sessions for reinforcement of the recommendations with no significant change in the group in which physicians were responsible for providing reinforcement reducing the presence of risks. The main results at two-year follow-up in the Swiss site revealed the following (all differences with a significance of at least P=<0.05): see Table 2.
Sub-group analyses reveal that the favorable intervention effects were explained by the combination of the HRA with home visits. HRA alone, just as in London, had no effects on health behavior and weak effects only on use of preventive care. The results at the Swiss site suggest that this innovative approach is a promising method for modifying adverse risk factors for functional status decline and for optimising health care use among a population of older adults.
Conclusion
In summary, as we begin the journey through the 21st Century with the rapid increases in the number of older persons living in industrialised communities, and with this shift raising alarm at all levels of government and among service providers, programs to prevent or delay the onset of functional impairment must be viewed as an alternative to institutional, community and home-based long-term care strategies — dependency care. Moving health care policy from dependency services to promoting independence reflects a major culture shift and a change in the traditional treatment paradigm. The potential contributions of applied science and technology to the solution of the dependency care problem is enormous. The twin goals of a disability prevention program should be to empower older persons to maintain and improve their health and enable health care providers and health care organisations to provide safer and more evidence-based care. I believe becaue there is no single clinician provider or healthcare entity that has possession of all the health information belonging to a single individual, and that it is important to give individuals the opportunity to have more access and control over their personal health information. Moreover, I believe that giving individuals access to, and control over, their personal health information will bring benefits to them and to the entire society which include (Table 4). This is one of the strategies for implementing ‘independence care maintenance’.
It is our responsibility to ensure that the extraordinary potential of information and other technologies to improve the health and healthcare of each older adult, and thus their independence is translated into every day use as quickly and efficiently as possible. The result will be nothing less than a quantum leap in addressing a major emerging societal problem. I wish to acknowledge my colleagues over the last decades who were integral contributors to the data I have reported.
Alessi, C A *Professor of Medicine, University of California, Los Angeles, USA
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
