Author
- David Currow1
- Flinders University1
2014
Excerpt
The evidence base in hospice / palliative care needs to continue to expand to complement the rapid growth of clinical services over the last four decades. Quality evidence is required to refine further clinical care and service delivery, yet hospice / palliative care has, at times, compromised on the quantitative research methodologies used and, at other times failed to use high quality, existing methodologies.
Quantitative research is crucial to building the evidence base for hospice / palliative care, requiring researchers in the discipline to adopt existing methodologies wherever possible, and to adapt other well developed methodologies sensitively to the research needs of the discipline. Any adaptation should not add to any systematic bias or, if this is inevitable, the direction and magnitude of the bias should be known and minimised.
This process of adopting and adapting quantitative research methodologies in hospice / palliative care includes:
- codifying simply the populations and settings in hospice / palliative care research (because the populations seen by these services vary so widely around the world);
- ensuring randomised controlled trials do not under-estimate the benefits of interventions being evaluated because of withdrawals / deaths unrelated to the intervention;
- creating ‘communities of practice’ to evaluate critically current practices; and
- using large datasets to understand the population served and the real-world performance of those services.
With no accepted national nor international standards for referral to hospice / palliative care services, there is a need to continue to refine the key factors that must be used to describe, in a structured way, populations in hospice / palliative care research. Ensuring a small number of key patient and service descriptors will aid clinicians, researchers, planners and funders to conduct quantitative research in more structured and transparent way in order to aid uptake into hospice / palliative care practice, policy, and to facilitate further research.
The randomised, controlled trial remains the gold-standard design for evaluating interventions. This methodology must be used more widely in hospice / palliative care research. In adapting randomised controlled trials to hospice / palliative care research, there is a need to consider a specific sensitivity analysis in randomised controlled trials in hospice / palliative care – the palliative-modified intention-to-treat analysis. This is due to the frailty of the population entering controlled clinical trials in hospice / palliative care, some of whom will withdraw from the study due to disease progression or death: events which cannot be predicted at the time of randomisation and which are totally unrelated to the intervention being evaluated. The proposed palliative-modified intention-to-treat sensitivity analysis excludes people who withdraw from studies only because of disease progression or death, in order to minimise false negative results. This places a new analysis between an intention-to-treat analysis and a per protocol (‘valid case’) analysis.
As therapies are offered to a wider range of patients than those on whom the original evidence is based, post-marketing effectiveness data are crucial to good clinical care. An international pharmacovigilance ‘community of practice’ is building rapidly and can be adapted to range of interventions: medications (e.g. metoclopramide), a therapy (e.g. oxygen), a ‘palliative’ procedure (blood transfusion) or complex interventions (supportive counselling for families).
Quantitative methodological approaches that are novel to hospice / palliative care in this thesis include the use of:
- samples from the whole of the population - in order to learn more about the roles and burdens of caregiving at the end of life, the performance of the health system in supporting these caregivers and differential access to, and use of specialist hospice / palliative care services;
- populations - to understand the impact of symptoms including chronic breathlessness and chronic pain;
- geocoding - to assess physical access to services, patterns of service utilisation and referral patterns by socio-economic status.
Ultimately, people at the end of life deserve the same level of evidence to underpin the clinical care that is offered to them as any other group of patients. Likewise, decisions on health service expenditure need to be based on the best available evidence. Such an evidence base can be built without compromise to the gold-standard methodologies used in other clinical and health services research. The people served by hospice / palliative care will be the beneficiaries of improved evidence to underpin the clinical and service decisions that are made on a daily basis.
Quantitative research is crucial to building the evidence base for hospice / palliative care, requiring researchers in the discipline to adopt existing methodologies wherever possible, and to adapt other well developed methodologies sensitively to the research needs of the discipline. Any adaptation should not add to any systematic bias or, if this is inevitable, the direction and magnitude of the bias should be known and minimised.
This process of adopting and adapting quantitative research methodologies in hospice / palliative care includes:
- codifying simply the populations and settings in hospice / palliative care research (because the populations seen by these services vary so widely around the world);
- ensuring randomised controlled trials do not under-estimate the benefits of interventions being evaluated because of withdrawals / deaths unrelated to the intervention;
- creating ‘communities of practice’ to evaluate critically current practices; and
- using large datasets to understand the population served and the real-world performance of those services.
With no accepted national nor international standards for referral to hospice / palliative care services, there is a need to continue to refine the key factors that must be used to describe, in a structured way, populations in hospice / palliative care research. Ensuring a small number of key patient and service descriptors will aid clinicians, researchers, planners and funders to conduct quantitative research in more structured and transparent way in order to aid uptake into hospice / palliative care practice, policy, and to facilitate further research.
The randomised, controlled trial remains the gold-standard design for evaluating interventions. This methodology must be used more widely in hospice / palliative care research. In adapting randomised controlled trials to hospice / palliative care research, there is a need to consider a specific sensitivity analysis in randomised controlled trials in hospice / palliative care – the palliative-modified intention-to-treat analysis. This is due to the frailty of the population entering controlled clinical trials in hospice / palliative care, some of whom will withdraw from the study due to disease progression or death: events which cannot be predicted at the time of randomisation and which are totally unrelated to the intervention being evaluated. The proposed palliative-modified intention-to-treat sensitivity analysis excludes people who withdraw from studies only because of disease progression or death, in order to minimise false negative results. This places a new analysis between an intention-to-treat analysis and a per protocol (‘valid case’) analysis.
As therapies are offered to a wider range of patients than those on whom the original evidence is based, post-marketing effectiveness data are crucial to good clinical care. An international pharmacovigilance ‘community of practice’ is building rapidly and can be adapted to range of interventions: medications (e.g. metoclopramide), a therapy (e.g. oxygen), a ‘palliative’ procedure (blood transfusion) or complex interventions (supportive counselling for families).
Quantitative methodological approaches that are novel to hospice / palliative care in this thesis include the use of:
- samples from the whole of the population - in order to learn more about the roles and burdens of caregiving at the end of life, the performance of the health system in supporting these caregivers and differential access to, and use of specialist hospice / palliative care services;
- populations - to understand the impact of symptoms including chronic breathlessness and chronic pain;
- geocoding - to assess physical access to services, patterns of service utilisation and referral patterns by socio-economic status.
Ultimately, people at the end of life deserve the same level of evidence to underpin the clinical care that is offered to them as any other group of patients. Likewise, decisions on health service expenditure need to be based on the best available evidence. Such an evidence base can be built without compromise to the gold-standard methodologies used in other clinical and health services research. The people served by hospice / palliative care will be the beneficiaries of improved evidence to underpin the clinical and service decisions that are made on a daily basis.