Solutions affaires int gr es et ing nierie

Document Sample
Solutions affaires int gr es et ing nierie Powered By Docstoc
					Home Telemonitoring for Chronic
Patients: A Systematic Review
Guy Paré, Ph.D.
Professor of IT
Canada Research Chair in IT in Health Care
HEC Montréal

Rob Kling Center for Social Informatics Speaker Series
Indiana University
February 26, 2010
Who am I?
   Ph.D. – Florida International University (1995)
       Area: Management Information Systems
       Topic: Dynamics of IT implementation
       Committee: J. Elam, D. Robey, R. Sabherwal, P. Hart
   Professor of IT at HEC Montréal
       Teaching:
           Undergrad: Systems analysis and design, business process
            reengineering and management
           Grad: IT implementation, IT & business transformation, research
            methods in IS
       IT area coordinator, Ph.D. program (12 students)
   Canada Research Chair in IT in Health Care
       Chair granted by the Social Sciences and Humanities Research
        Council (SSHRC) of Canada in June 2004
       My research agenda in health care
3                                 IT Project Lifecycle
         Initiation               Pre-deployment                  Post-deployment
           phase                       phase                           phase

    Organizational         Project risk management            IT impacts (evaluative studies)
    adoption of IT           - Delphi                            - RCT studies (field experiments)
    (OS, DI, EMR)            - Multiple case studies             - Pre-post cohort studies
                             - Instrument development            - Surveys (DeLone & McLean, 2003)
    Organizational                                               - Systematic reviews
    readiness for IT-      Individual adoption of IT
    based change             - Psychological ownership

                   Decision to                      Go-live
                  invest or not
   This study was conducted at the Agence d’évaluation
    des technologies et des modes d’intervention en santé
    (AETMIS)1 in 2007-2008
   Collaborators:
     Khalil Moqadem, M.D.
     Gilles Pineau, M.D.

     Carole St-Hilaire, Ph.D.

   This systematic review was requested by the Quebec
    Ministry of Health
    1 AETMIS is the Québec government agency responsible for health services and
    technology assessment.
The problem
The health care systems in many developed
countries, including Canada and U.S., are under
Population aging is enduring
                    Proportion of population 60 years and over: 1950-2050





                           1950                2009                 2050

                                   Source: United Nations (2009)
Population aging is also…
   Pervasive
     Affects   nearly all countries of the world
   Profound
     Has major consequences and implications for all facets
      of human life
       Economic  sphere: economic growth, savings, investments,
        consumption, labor markets, pensions, taxation, potential
        support ratio, etc.
       Social sphere: family composition and living arrangements,
        housing demand, migration trends, etc.
       Healthcare sphere: prevalence of chronic diseases and, in
        turn, the demand for healthcare services
Potential support ratio (PSR)

                 Potential support ratio: 1950-2050







                 1950               2009                2050

                        Source: United Nations (2009)
On the demand side
   With an aging population and increased life expectancy, chronic
    diseases are on the rise
       73% of Canadians aged 65 and up report having at least one of 7
        common chronic health conditions (Health Council of Canada, 2008)
   Chronic diseases have become major causes of death in almost all
       7 out of 10 deaths among U.S. citizens each year are from chronic
        diseases (Kung et al. 2008)
   Chronic patients have poor adherence to medication
           Varies between 11% and 42% (Thier et al. 2008)
           This is a massive, world-wide problem (WHO, 2003)
   Not surprisingly, chronic patients use medical, hospital and emergency
    services more often than non-chronic patients
       70% of all hospital admissions are associated with chronic diseases
        (Health Council of Canada, 2007)
     Prevalence of chronic diseases

                  How many Canadians have chronic diseases?
               No chronic condition   I chronic condition    2 chronic conditions   3+ chronic conditions



                                                      25                                           24

               18-44 years                           45-64 years                           65+ years

                       Source: Health Council of Canada (2008)
Economic burden of chronic diseases

   In the U.S.
     In 2008, the government spent $600 billions on chronic
      diseases which impacted 45% of the population
     75 cents of every $ spent on health in the U.S. goes to
      treating people with chronic diseases
   In Canada
     The  government spending on health care per person
      rises with age:
       Age 1 : about $1,000 per person
       Age 64 : about $3,800 per person
       Ages 70-74 : about $7,700 per person

           Sources: 2009 Almanac of Chronic Disease & CIHI (2009)
On the supply side
   The health care team is shrinking…
     Acutenursing shortage exists in many countries including
      Canada, U.S., UK, and Australia
       In Québec, there was a shortage of 3,500 RNs in 2008-
        2009 and this shortage is projected to grow to 23,000 RNs
        by 2022 (Quebec Ministry of Health, 2009)
       In the U.S., the RN vacancy rate was established at 8.1% in
        2007, representing 135,000 RNs. The shortage is projected
        to grow to 260,000 RNs by 2025 (American Health Care
        Association, 2008)
     Substantial  physician shortage also exists in various
      countries, including Canada and the U.S., and the
      shortage is more acute in rural regions
A fundamental change is required
How will we afford the explosion of costs as
older people become a larger share of the
We need to identify patient management
approaches that will ensure appropriate
monitoring and treatment of chronic patients
while reducing the costs involved in the process
A solution: Patient self-management
   For patients, self-management means having the skills
    and opportunity to be effective partners in their own
    health (Ontario Patient Self-Management Network)
   Patients with self-management skills can recognize when
    they have a problem and have the confidence to take
    appropriate action (Bodenheimer et al. 2002)
   The real issue is NOT whether patients with chronic
    conditions manage their health, BUT how well they
    manage it
   IT applications provide an opportunity for supporting
    self-management programs and changing the face of
    healthcare. Home telemonitoring represents one such
  How does home telemonitoring work?

                           clinical advices                           Direct phone
  RN (case manager)                                                     contacts
   Detects problems,
  proposes palliative
solutions & contacts MD          Phone line or
   whenever needed         Internet connection
                                                                Patients at home

                                      Patient’s clinical data
                                      (e.g., peak flow rate,
                                     symptoms, medication)
Home telemonitoring market
   Europe and U.S. market
     $3 billions in 2008
     Projected to be $7.8 billions by 2012

   Examples of major players
     Intel Digital Health
     GE Healthcare

     HomMed

     TELUS


Source: (2010)
Research Objectives & Context
   Objectives
     To determine the effects associated with home
      telemonitoring across a variety of chronic diseases
     To explore the conditions for success for this health care
      delivery approach
   Context
     Chronic   diseases:
         Diabetes
         Pulmonary  (asthma & chronic obstructive pulmonary disease)
         Hypertension
         Heart failure
A priori hypotheses
   Positive and significant effects are expected because
    home telemonitoring (HT) allows for a more frequent
    follow-up on each patient and, as a result, early
    detection of warning signs that the patient’s state of
    health is decreasing
   Hence, we hypothesize that:
     H1: HT will maintain or improve patients’ health status
     H2: HT will increase patients’ feeling of security and
     H3: HT will allow just-in-time interventions and hence,
      decrease the demand for health care services (e.g., ER visits,
      hospital stays)
     H4: HT will be more cost-effective than the traditional model
      (regular home visits)
Method: Systematic review
   Definition
       A review of a clearly formulated question that uses
        systematic and explicit methods to identify, select, and
        critically appraise relevant research, and to collect and
        analyze data from the studies that are included in the
        review (Moher et al., 2009)
   Systematic reviews in the IS field
       A search in ABI Inform using the term « systematic review »
        generated a total of 322 references. Among these:
         104 appeared in health journals
         34 appeared in computer science journals
         4 appeared in information systems journals
Search Strategy
    Databases consulted:
      Medline(PubMed)
      Cochrane Library

      INAHTA database1

    Keywords:
      Telemonitoring,telehomecare, telehealth and telecare
      Diabetes, hypertension, asthma, COPD & heart failure

    Period:
      All   studies published prior to January 2009
1   International Network of Agencies for Health Technology Assessment
Inclusion and exclusion criteria
   Inclusion criteria:
     Studies presenting results on telemonitoring effects
     Studies published in peer-reviewed journals or conference
     Studies published in English, French, German or Spanish
   Exclusion criteria:
     Studies examining multi-pathology groups of patients
      without separating patients with different conditions in
      different groups
     Studies that included regular phone calls by care providers
      without specialized telemonitoring equipment
     Studies that only involved in-home consultation sessions
      delivered via video visits (teleconsultation visits)
Selection of studies
        968 citations identified from
         initial electronic searches
                                         378 citations excluded based on
          590 potentially relevant
        articles retrieved for further
           scrutiny (based on full           461 references excluded:
                   articles)             • Other forms of home
                                         telehealth interventions (243)
                                         • No experiment or no clinical
                                         effects reported (165)
                                         •Studies involving patients with
                                         other chronic diseases (26)
                                         • Experimental studies involving
            129 included studies         multipathology patients (15)
                                         • Editorials or essays (12)
Coding of articles
   A codification scheme which contains 6 sections was
    1.   General profile (year, country, type of publication)
    2.   Type and quality of research design (randomization, equivalent
         control group, sample size, inclusion and exclusion criteria, etc.)
    3.   Participants’ characteristics (age, gender, primary diagnostic,
    4.   Characteristics of the interventions (length, data transmission
         frequency, types of data transmitted, health professionals
    5.   Characteristics of the technologies used by patients (type of
         equipment, type of network, video component, etc.)
    6.   Home telemonitoring effects reported (patients’ health status,
         patient satisfaction, home visits, ER visits, hospital stays, etc.)
Validation of the coding scheme
   10 studies were randomly selected and coded
    independently by two researchers
   Agreement rate = 94%
   All minor differences were resolved by consensus
   A few minor adjustments were made to the coding
   Once all 129 studies were coded, they were
    entered in the SPSS software
Profile of the 129 studies
Year of publication
         22%          78%
Where the studies were conducted




       32%                    Océanie
Variety of technological devices
Representativeness of diseases

           Diseases       Number of   Percentage
Heart failure                48         37%
Diabetes                     30         23%
Pulmonary diseases           20         16%
Hypertension                 18         14%
Mix of chronic diseases      13         10%
TOTAL                       129         100%
Variety of research designs

                 Design                Percentage
Large RCTs (each sample size >= 100)      7%
Small RCTs (each sample size < 100)       36%

Quasi-experiments                         15%

Cohort studies                            38%

Descriptive (case) studies                5%
   Profile of the interventions
                         Diabetes   Pulmonary    HF and      Mix of   All studies
                                                Hypertens. conditions
Sample      RCT –          76          60          72          30         72
            Exp. Gr
            RCT –          81          51          70          29         72
            Ctrl Gr.
            Non-           146         42          81          88         71
Length of intervention      8          6            7          7           7
Frequency of data         Daily       Daily       Daily       Daily      Daily
Vital signs entered       20%         33%         31%          8%         26%
Video consultations        3%         11%          3%          0%         4%
PC station + Internet     10%         28%         14%          0%         13%
Main Results
NOTE: More weight was given to findings of
large and small RCTs compared to non-
randomized and cohort studies
    Types of effects reported

                      Diabetes   Pulmonary   Cardiovascular    Mix of    Total
                                  diseases      diseases       chronic
Number of               30          20            66            13       129
Clinical effects        24          11            27             3         65
Attitudinal effects     16          16            55             9         96
Structural effects      11          12            23             5         51
Economical               6          5             13             0         24
effects                                                                  (19%)
H1: Clinical effects
Positive results
   Diabetes
       The 2 large RCTs in addition to 10 small RCTs and the 3 non-randomized
        studies reported a significant decline in HbA1c and increase in blood glucose
       9 small RCTs concluded that HT was as effective in glycemic control as the
        traditional approach (i.e. regular home visits)
           Smaller sample sizes
   Asthma
       5 out of 7 RCTs (1 large, 4 small) showed a significant improvement in peak
        expiratory flow (PEF), a significant reduction in asthma symptoms and large
        improvement in quality of life (QOL)
       Two small RCTs did not find significant results
           Less frequent data transmission
           Patients with low/moderate level of asthma severity
   Hypertension
       4 out of 5 RCTs and 6 out of 7 non-randomized studies reported a
        significant reduction in systolic and diastolic blood pressure
H1: Clinical effects (cont’d)
Limited empirical evidence
   COPD
       3 small RCTs showed a significant decrease in terms of number of
        exacerbations in the experimental group

Mix results
   Heart failure
       Most studies, including 5 large and 6 small RCTs, revealed positive
        effects in terms of QOL and better control of the symptoms
       However, mix results were observed in 6 studies with regard to the
        effects of HT on patient mortality rates
H2: Attitudinal effects
   In virtually all 129 studies, chronic patients showed a
    very high or high level of satisfaction
   Patients’ feeling of security increases due to more
    frequent follow-ups
   Patients’ feeling of empowerment was investigated in
    only a few studies (not all HT devices allow this)
       In those studies, however, HT has allowed patients to actively
        participate in the process of care, improved their awareness
        and understanding of warning signs, and ultimately led to
        greater feeling of control
H3: Structural effects
   HT leads to a significant reduction in the demand
    for health care (ER visits, hospital readmission rates,
    and length of stay) in patients with heart failure
    (85% of studies) and COPD (100% of all studies)

   No significant effects were observed in cohorts of
    patients with diabetes and asthma
   Structural effects were not investigated in most
    hypertension studies
H4: Economic viability
   Scarce evidence across all diseases (n=24)
       Studies are mainly conducted by clinicians (RNs and MDs)
   Among the few studies which investigated the cost-
    effectiveness of HT, we noticed:
     Sources of savings not identified or properly described
     Standard deviations up to 4 times larger than means

     Small sample sizes (e.g., Paré et al. 2006)

     Insufficient data provided

   Given the paucity of evidence and the ambiguity of the
    results obtained so far, no firm conclusions can be
    drawn regarding economical effects
In summary
   H1 (clinical effects)
       Partially supported
           Supported for diabetes, asthma and hypertension
           Limited support for COPD (limited number of studies)
           Not supported for heart failure
   H2 (attitudinal effects)
       Fully supported for all chronic conditions
   H3 (structural effects)
       Partially supported
           Supported for COPD and heart failure
           Not supported for asthma and diabetes
           Lack of empirical evidence for hypertension
   H4 (financial effects)
       Not supported
           Lack of empirical evidence
Health policy implications
   Given the demographic changes, the prevalence of chronic
    diseases and the acute shortage of health care
    professionals, our overall recommendation to the Quebec
    Ministry of Health was to pursue investments in the gradual
    deployment of home telemonitoring programs for the chronic
    conditions considered in this review
   Following our recommendation, budgets were allocated to
    two major home telemonitoring interventions in the Montreal
       Main chronic conditions : heart failure and pulmonary illnesses
       Length of observation = pre (12 mo) and post (12 mo)
       Sample size = >100 patients in each group
       Economic analysis will be performed
Critical success factors
Is home telemonitoring for everyone?

   Patients excluded for obvious reasons
     Patients with moderate or serious cognitive, visual or
      physical disability
     Patients whose life expectancy is measured in months

   Some patients seem to benefit from HT more than
     Patients whose state of health is severe
     Those who want to play an active role in the management of
      their health condition
     Those who are interested in using this type of technology
What about the technology?
   User-friendliness of the device and its non-
    intrusiveness in the lives of patients appear to be
    important acceptance criteria
   Given that all chronic patients do not have the same
    level of technological skills, the same level of
    education, the same professional constraints or even
    the same lifestyle, it would be preferable for
    providers to ensure that patients have the device
    best suited to their specific needs
Cautious interpretation of results
 Majority of studies (57%) were non-randomized
  trials without any control group
 Larger samples of patients evaluated over
  longer periods of time are needed
 Very few observations were made in relation to
  the effects of telemonitoring on healthcare
  providers, their acceptance of this approach and
  their concerns about it, their workload, the
  organization of work, etc. Future studies should
  investigate these important issues
For more details

Revue systématique des effets de la télésurveillance à
domicile dans le contexte du diabète, des maladies
pulmonaires et des maladies cardiovasculaires. AETMIS,
2009, 5(3):1-75.

Available at the following address:

Shared By: