Text table 2
Validation grid of POHEM-Neurological meta-model^{Note 1} Text table 2
Validation grid of POHEM-Neurological meta-model^{Note 1}

Text table 2
Validation grid of POHEM-Neurological meta-model
Table summary
This table displays the results of Validation grid of POHEM-Neurological meta-model. The information is grouped by Source of evidence (appearing as row headers), Concept, Score and Comments (appearing as column headers).
Source of evidence	Concept	Score^Note †	Comments
Model development	Conceptual models: Underlying theories	2	Based on committee of experts (several subprojects in NPHSNC project were devoted to aetiology of neurological conditions).
	Conceptual models: Definition of variables	2	Based on committee of experts. Variables adjusted according to available data; for example, Parkinson’s disease was expanded into Parkinsonism because prevalence (or case algorithm) and cost data included both categories and could not be disentangled.
	Conceptual models: Content and structure	2	The objectives of building the model were to define the path that correctly describes each NC, and use a common pathway for all seven NCs.
	Parameters	2	Parameter values obtained from experts, the literature, data analysis, or through calibration (last column of Text table 1).
	Computer implementation: Selection of model type	2	No formal process determined the type of model, but the value of using a microsimulation model was implicit. However, the choice was supported by the literature.^{Note 2}^{Note 3}
	Computer implementation: Simulation software	2	If microsimulation is considered the appropriate model, the software used (Modgen^{Note 4}) is justified: several models are built with this software.
	Computer implementation: Computer program	1	The code has been made available internally and to clients, through informal meetings. BioBrowser (a tool within Modgen that displays individual values) was used extensively for validation of person-level trajectories. However, scrutiny of verification of the code was not deemed essential.
Model performance	Output plausibility	2	Verification and validation performed to ensure prevalence (produced by the models) were similar to observed prevalence for each of the seven NCs. For other outcomes (costs, HUI3, informal care), continuous monitoring of results with developers and clients ensured that outcomes were plausible for the years for which data were available.
	Internal consistency	1	Internal consistency throughout a wide range of conditions, including extreme values, was not analyzed because the goal was to project assuming status quo conditions, not to use scenarios.
	Parameter sensitivity	1	No uncertainty was provided for the parameters. In some situations, standard errors were wide for a given parameter for some combinations of age and sex. It was not always possible to regroup age groups to lower the standard error. Sensitivity analysis was not performed.
	Between-model comparisons	NA	For each NC, only one microsimulation model was built.
	Comparisons with external data	2	Extensive, related to other NPHSNC projects.
Consequences of model-based decisions	Quality of decisions	1	Results are expected to be useful for decision-makers; quality of decisions will be evaluated in future work.
Consequences of model-based decisions	Model usefulness	1
Note † Validation scores: 0 = none, 1 = partial, 2 = complete, NA = not applicable Return to note † referrer

References for Text table 2

Footnote 1.

Kopec JA, Finès P, Manuel DG, et al. Validation of population-based disease simulation models: a review of concepts and methods. BMC Public Health 2010; 10: 710.

Return to note 1 referrer

Footnote 2.

Norton S, Matthews FE, Brayne C. A commentary on studies presenting projections of the future prevalence of dementia. BMC Public Health 2013; 13: 1.

Return to note 2 referrer

Footnote 3.

Zucchelli E, Jones AM, Rice N. The evaluation of health policies through microsimulation methods. HEDG Working Paper 10/03. Heslington, United Kingdom: The University of York, 2010. Available at: york.ac.uk/res/herc/hedgwp

Return to note 3 referrer

Footnote 4.

Modgen. Model Generator. Available at: http://www.statcan.gc.ca/eng/microsimulation/modgen/modgen

Return to note 4 referrer

Date modified:: 2017-07-12

Language selection

Search and menus

Search

Text table 2
Validation grid of POHEM-Neurological meta-model^{Note 1} Text table 2
Validation grid of POHEM-Neurological meta-model^{Note 1}

Text table 2 Validation grid of POHEM-Neurological meta-modelNote 1 Text table 2 Validation grid of POHEM-Neurological meta-modelNote 1

Note of appreciation

Standards of service to the public

Copyright

Text table 2
Validation grid of POHEM-Neurological meta-model^{Note 1} Text table 2
Validation grid of POHEM-Neurological meta-model^{Note 1}