Chapter 3: Mapping Connections: An understanding of neurological conditions in Canada – Scope (prevalence and incidence)

Figure 3-4: Prevalence of traumatic brain injury, by sex and age group, British Columbia, 2009/2010, BC Administrative Data Project [1]

NOTES:

Data were rounded to the nearest ten.

BC

British Columbia.

SOURCE: Population Health Surveillance and Epidemiology, British Columbia Ministry of Health (September 2011).

Figure 3-5: Prevalence of traumatic spinal cord injury, by sex and age group, British Columbia, 2009/2010, BC Administrative Data Project [1]

NOTES:

BC

British Columbia.

SOURCE: Population Health Surveillance and Epidemiology, British Columbia Ministry of Health (September 2011).

The age at onset of symptoms and diagnosis differ by neurological condition. The age at onset of symptoms and age at diagnosis differed by neurological condition (Table 3-3) [16]. However, based on survey data, differences were not statistically significant by sex except for the age at diagnosis for Alzheimer’s disease and other dementias.^{Footnote 40} A better analysis of sex differences in the age of onset of various neurological conditions could be addressed with a longitudinal cohort analysis or through the use of administrative data.

There are common comorbidities associated with neurological conditions. Men and women differed in their experiences of the physical and social aspects of neurological conditions. Chronic conditions such as neurological conditions may impact men and women at different stages in life, affecting their professional or familial roles. Because the onset of many of these neurological conditions occurs in later life, Canadians may be faced with various comorbidities, and women, who tend to live longer than men, may face additional burdens.

Certain chronic conditions, such as arthritis and high blood pressure, were common comorbidities for the overall population with a neurological condition (excluding migraine) (Table 3-4)[3].^{Footnote 41} The prevalence of asthma, arthritis, and bowel disease was higher among women than men, while heart disease was higher among men than women. The BC Administrative Data Project [1] also listed the eight most common ‘other diagnoses’ throughout 2005/2006 to 2009/2010 that were associated with each of the 13 neurological conditions addressed in this project, but these associations were a mix of causes and complications related to neurological conditions.

3.3 Building capacity for the surveillance of neurological conditions is well underway

One of the goals of the Study was to provide evidence that could inform the development of a surveillance system for neurological conditions in Canada. Progress towards this goal was advanced through four interdependent approaches: systematic reviews and meta-analyses [18], new developments at the Agency to add neurological conditions to the CCDSS [2], the estimates of prevalence and incidence based on different data sources by the projects and surveys of the Study [1][3][6][10][12][17], and the development of guidelines and toolkit for the creation and management of registries [4][15]. In general, none of the data sources examined in the Study was able to produce reliable prevalence estimates for certain targeted neurological conditions, specifically, those that were less prevalent. There were also varying levels of concordance among prevalence estimates for certain conditions, in part because of non-uniform case definitions applied by the different projects or because their coverage was limited to certain jurisdictions.

Systematic reviews of the literature provide evidence for data coding and sources. To determine the best sources of ascertainment (identification of cases) for 15 neurological conditions, the Systematic Reviews Project [18] identified 30 studies that validated ICD coding of case definitions in health administrative data. The project team also evaluated the sources of ascertainment from over 1,200 studies for prevalence and incidence of neurological conditions and, in consultation with 50 experts, identified the most frequently used sources of data for each condition. They also prepared an inventory of neurological registries.

The results of this investigation supported the development of surveillance for some neurological conditions. Findings from this project demonstrated that the accuracy of ICD coding had been validated for Alzheimer’s disease and other dementias, motor neuron disease,^{Footnote 42} epilepsy, multiple sclerosis, Parkinson's disease, spinal cord injury, and traumatic brain injury, but not for specific brain tumours, cerebral palsy, dystonia, Huntington's disease, hydrocephalus, muscular dystrophy, spina bifida, or Tourette syndrome.^{Footnote 43} The project team also concluded that the following sources of ascertainment were the most appropriate for the surveillance of certain neurological conditions:

• Multiple sources: Alzheimer’s disease and other dementias, ALS, cerebral palsy, Huntington's disease, hydrocephalus, multiple sclerosis, muscular dystrophy, Parkinson's disease, and traumatic brain injury.
• Registries: Brain tumours, cerebral palsy, spinal cord injury, and rare disorders.
• Administrative data or EMRs: Epilepsy.
• Population-based random sampling, door-to-door surveys,^{Footnote 44} or school-based studies: Dystonia and Tourette syndrome.
• Specific surveillance programs: Congenital anomaly surveillance programs for spina bifida.

The addition of neurological conditions to the CCDSS is progressing. The CCDSS is a collaborative network of provincial and territorial surveillance systems led by the Agency to collect data on chronic conditions (such as diabetes or hypertension). In conjunction with some of the other projects of the Study [1][10][12][18], the CCDSS Neurological Conditions Working Group conducted development work to add four neurological conditions to the CCDSS (Alzheimer’s disease and other dementias, epilepsy, multiple sclerosis, and parkinsonism) [2]. The inclusion of stroke in the CCDSS has been ongoing and is led by the CCDSS Stroke Working Group.

Studies to validate algorithms for these five conditions were performed by the ON Administrative Data Project [12]. This represented a major step toward the development of the CCDSS for the pan-Canadian surveillance of neurological conditions, as it allowed for the identification of the best performing algorithms. In 2013/2014, these five neurological conditions were included in the CCDSS national pilot, a process in which all provinces and territories test the case definitions in their jurisdiction. Careful analysis of national pilot data will ensure that the selected case definitions consistently apply across all Canadian provinces and territories. With the formal inclusion of these conditions in the CCDSS in the near future, this component of the Study will continue on an ongoing basis as part of the Agency’s national public health surveillance mandate.

The Canadian Primary Care Sentinel Surveillance Network (CPCSSN) has been expanded to include neurological conditions. The EMR Project [6] used the existing CPCSSN platform to develop case definitions for Alzheimer’s disease and other dementias, epilepsy, and parkinsonism. Findings from this project suggest that EMRs could be used for regular surveillance of these conditions in community-based primary care when ascertainment algorithms are appropriately validated.

Guidelines have been created for the development and maintenance of registries. For neurological conditions not well suited to surveillance through surveys and/or administrative data, the creation of a registry may be another approach. Disease or patient registries are collections of data related to patients with a specific diagnosis or condition. The Registry Guidelines Project [15] carried out extensive literature reviews, held three structured focus group meetings and two consensus conferences, and used a modified Delphi consultation process to produce comprehensive guidelines for the development, implementation, and maintenance of registries of neurological conditions in Canada. These guidelines have been published,^{Footnote 45} and an implementation toolkit will assist in the planning of future registries for neurological conditions. This project delineated the key elements of:

• Good registry design: Participant informed consent, transparency, an advisory council, clear data ownership, appropriate data security, data release procedures, multi-modal patient recruitment, patient follow-up, planned data linkage, standard operating procedures, a data management plan, and appropriate data collection.
• A high quality registry: A quality management plan; standardized methodologies to address data collection and source inconsistencies; the use of an iterative or pilot-tested data collection method; rigorous, consistent and documented processes for data cleaning and correction; trained personnel; and a defined audit system with initialization triggers.
• A registry with high impact: Advance planning, adequate human and financial resources, regular communication with participants and stakeholders, efficient data collection, and transparent operation.

Although Canadian patients were generally supportive of the concept of disease and patient registries and were willing to participate in such registries, certain considerations emerged from the structured focus groups:

• Before providing their consent to participate in a registry, patients wanted to know certain features of the registry: Its purpose; its capacity to facilitate ethical research with meaningful results; that it is well managed and sustainable; that participation will not be onerous; and that they may withdraw at any time.
• Patients also had difficulty understanding the need to share other information such as: Personal information including age, occupation, income, provincial health number, and marital status. They also did not wish to provide their Social Insurance Number to a registry, even if they knew this was the only national individual identifier.

The Cerebral Palsy Registry Project [4] also produced information about the key challenges involved in expanding an existing registry. Although the experiences of this project team were specific to cerebral palsy, lessons learned could be helpful in the development of registries for other neurological conditions.

3.4 Looking ahead: 2011 to 2031

Results from the Microsimulation Project [10], which are based on status quo assumptions,^{Footnote 46} project that between 2011 and 2031:

• The number of Canadians living with Alzheimer's disease and other dementias and Parkinson's disease/parkinsonism (based on prevalence numbers) will almost double (Table 3-5).

• In the population age 65 years and older, there will be at least twice as many, if not more, Canadians living with each of the seven modelled neurological conditions (based on prevalence numbers) (Table 3-6).

• The number of Canadians newly diagnosed (incidence count) with cerebral palsy, epilepsy, multiple sclerosis, and spinal cord injury will grow in line with the growth of the population, so that incidence rates in 2031 will remain close to those seen in 2011 (Table 3-7). For conditions whose risk increases markedly with age, the incidence will increase beyond the rate of population growth. In particular, for Alzheimer’s disease and other dementias and Parkinson’s disease/parkinsonism, the incidence rate will increase by almost 50%.
• For traumatic brain injury, the number of hospitalizations is expected to increase by 28% over the next 20 years unless better prevention strategies are implemented.

• The number of deaths among Canadians with a neurological condition will increase, with the greatest impact seen in those living with Parkinson’s disease/parkinsonism, where a 74% increase in the number of deaths is projected over time. The total number of deaths for those living with Alzheimer’s disease and other dementias, which is already the highest of the modelled conditions in 2011, will increase by a further 70% by 2031 (Figure 3-6).

Figure 3-6: Projected number of deaths, by select neurological condition, Canada, 2011, 2016, 2021, 2026, and 2031, Microsimulation Project [10]

NOTES:

* Traumatic brain and spinal cord injuries were based on hospitalized cases, and excluded injuries that did not present to hospital.

SOURCE: POHEM–Neurological (Statistics Canada and Public Health Agency of Canada).

3.5 Knowledge gaps

While several projects of this Study generated new findings that hold promise for the further determination of scope of neurological conditions in Canada, it was also demonstrated that for some conditions, different data sources produced incongruent results. This complexity calls for caution and careful consideration when selecting the most appropriate statistics to describe the burden of neurological conditions in Canada.

Regarding the scope of neurological conditions, the Study identified two main categories of gaps – gaps in research and gaps in infrastructure. Research gaps included the lack of data on:

• The identification of certain neurological conditions, such as multiple sclerosis or Parkinson's disease, at their earliest stages (noting that early symptoms are often non-specific, which makes it hard to reach a diagnosis, in addition to a tendency to use coding to describe symptoms rather than the diagnosis itself);
• Less prevalent conditions such as ALS, dystonia, and Huntington's disease, as well as other neurological conditions not targeted by the Study;
• Neurological conditions in populations typically excluded from participation in national population surveys;Footnote 47 and
• The extent of neurological conditions among children (including cerebral palsy, epilepsy, multiple sclerosis, brain injury, and stroke), and the persistence of these conditions into adulthood.

Gaps in infrastructure included the need for:

• Standardization of diagnostic codes (e.g. ICD-10-CA) used in physician billing claims and hospital data in all provinces and territories;
• Standardization of case definitions and algorithms, as well as their periodic review;
• Ongoing consideration and incorporation of newly available data (e.g. pharmaceutical or costing data); and
• Complete data capture and reporting of benign brain tumours in existing provincial and territorial cancer registries.

Findings from the Study were highly valuable in providing researchers and the Agency the tools and evidence to consider when developing national surveillance of neurological conditions in Canada. 

3.6 Key themes

In this chapter, various data sources were explored for their utility and suitability in examining the epidemiology of neurological conditions in Canada. It was noted that:

• The Study has produced many new estimates of the prevalence and incidence of neurological conditions in Canada which were not previously available. For many neurological conditions, estimates derived from administrative data, self-report survey, and EMRs conform to estimates based on meta-analyses of international studies. However, primarily for technical reasons, additional studies will be necessary to obtain reliable prevalence estimates for ALS, muscular dystrophy, spinal cord injury, and traumatic brain injury in Canada.
• When the estimates of prevalence and incidence from different sources differ significantly, they need to be interpreted with caution and the differences must be carefully balanced and critically appraised.
• Estimates of prevalence and incidence compiled periodically and systematically from health administrative data can be used to trace the course of neurological conditions over time, as well as across provinces and territories.
• The expansion of the evolving surveillance system for neurological conditions in Canada has been enhanced by the determination of the best sources of epidemiological data for certain neurological conditions and by the validation of algorithms for use with administrative data.
• Detailed guidelines and a toolkit for the development and maintenance of registries that can contribute to the surveillance of neurological conditions are now available.
• Neurological conditions are the primary diagnoses of more than half of Canadians in continuing care (home care programs and long-term care facilities). Therefore, to determine the prevalence of severely debilitating neurological conditions in Canada, data from both the community and long-term care facilities are necessary to obtain a more comprehensive picture.

Microsimulation modelling of seven neurological conditions projects that by 2031:

• Traumatic brain injury, Alzheimer's disease and other dementias, and epilepsy will continue to be the most prevalent of the modelled conditions.
• In a population age 65 years and older, the number of Canadians living with each of the modelled neurological conditions will more than double.
• The number of new cases of cerebral palsy, epilepsy, multiple sclerosis, and spinal cord injury will rise with the growth of the population.
• The number of Canadians receiving a new diagnosis of Alzheimer's disease and other dementias or Parkinson's disease/parkinsonism will close to double.
• The number of Canadians hospitalized with a brain injury will increase by 28%.