No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
An Exploratory Study of Oral and Dental Health in Canada
ResearchGate has not been able to resolve any citations for this publication.
The aim of this study was to analyse inequalities in the use of dental care services according to socioeconomic position (SEP) in individuals aged ≥50 years in European countries in 2006, to examine the association between the degree of public coverage of dental services and the extent of inequalities, and specifically to determine whether countries with higher public health coverage show lower inequalities.
We carried out a cross-sectional study of 12 364 men and 14 692 women aged ≥50 years from 11 European countries. Data were extracted from the second wave of the Survey of Health, Ageing and Retirement in Europe (SHARE 2006). The dependent variable was use of dental care services within the previous year, and the independent variables were education level as a measure of SEP, whether services were covered to some degree by the country's public health system, and chewing ability as a marker of individuals' need for dental services. Age-standardized prevalence of the use of dental care as a function of SEP was calculated, and age-adjusted indices of relative inequality (RII) were computed for each type of dental coverage, sex and chewing ability.
Socioeconomic inequalities in the use of dental care services were higher in countries where no public dental care cover was provided than in countries where there was some degree of public coverage. For example, men with chewing ability from countries with dental care coverage had a RII of 1.39 (95%CI: 1.29-1.51), while those from countries without coverage had a RII of 1.96 (95%CI: 1.72-2.23). Women without chewing ability from countries with dental care coverage had a RII of 2.15 (95%CI: 1.82-2.52), while those from countries without coverage had a RII of 3.02 (95%CI: 2.47-3.69).
Dental systems relying on public coverage seem to show lower inequalities in their use, thus confirming the potential benefits of such systems.
Interestingness measures play an important role in data mining, regardless of the kind of patterns being mined. These measures are intended for selecting and ranking patterns according to their potential interest to the user. Good measures also allow the time and space costs of the mining process to be reduced. This survey reviews the interestingness measures for rules and summaries, classifies them from several perspectives, compares their properties, identifies their roles in the data mining process, gives strategies for selecting appropriate measures for applications, and identifies opportunities for future research in this area.
Receiver Operator Characteristic (ROC) curves are commonly used to present results for binary decision problems in machine learning. However, when dealing with highly skewed datasets, Precision-Recall (PR) curves give a more informative picture of an algorithm's performance. We show that a deep connection exists between ROC space and PR space, such that a curve dominates in ROC space if and only if it dominates in PR space. A corollary is the notion of an achievable PR curve, which has properties much like the convex hull in ROC space; we show an efficient algorithm for computing this curve. Finally, we also note differences in the two types of curves are significant for algorithm design. For example, in PR space it is incorrect to linearly interpolate between points. Furthermore, algorithms that optimize the area under the ROC curve are not guaranteed to optimize the area under the PR curve.
A previously described coefficient of agreement for nominal scales, kappa, treats all disagreements equally. A generalization to weighted kappa (Kw) is presented. The Kw provides for the incorpation of ratio-scaled degrees of disagreement (or agreement) to each of the cells of the k * k table of joint nominal scale assignments such that disagreements of varying gravity (or agreements of varying degree) are weighted accordingly. Although providing for partial credit, Kw is fully chance corrected. Its sampling characteristics and procedures for hypothesis testing and setting confidence limits are given. Under certain conditions, Kw equals product-moment r. The use of unequal weights for symmetrical cells makes Kw suitable as a measure of validity.
This article estimates the prevalence of depression among employed Canadians aged 25 to 64, and examines its association with work impairment, as measured by reduced work activity, mental health/general disability days, and work absence.
Data are from the 2002 Canadian Community Health Survey: Mental Health and Well-being and the longitudinal household component of the National Population Health Survey (1994/1995 to 2002/2003).
Cross-tabulations were used to estimate and determine factors associated with the prevalence of depression among the employed population. Multiple logistic regression was used to examine associations between depression and work impairment while controlling for other variables. Longitudinal data for 1994/1995 to 2002/2003 were used to examine the temporal sequence of depression and work impairment.
In 2002, almost 4% of employed people aged 25 to 64 had had an episode of depression in the previous year. Crosssectional analysis indicates that these workers had high odds of reducing work activity because of a long-term health condition, having at least one mental health disability day in the past two weeks, and being absent from work in the past week. Longitudinally, depression was associated with reduced work activity and disability days two years later.
. We evaluate the power of decision tables as a hypothesis space for supervised learning algorithms. Decision tables are one of the simplest hypothesis spaces possible, and usually they are easy to understand. Experimental results show that on artificial and real-world domains containing only discrete features, IDTM, an algorithm inducing decision tables, can sometimes outperform state-of-the-art algorithms such as C4.5. Surprisingly, performance is quite good on some datasets with continuous features, indicating that many datasets used in machine learning either do not require these features, or that these features have few values. We also describe an incremental method for performing crossvalidation that is applicable to incremental learning algorithms including IDTM. Using incremental cross-validation, it is possible to cross-validate a given dataset and IDTM in time that is linear in the number of instances, the number of features, and the number of label values. The time for incre...
An interesting problem in statistics, and one that is generally difficult, is the estimation of a continuous function, such
as a probability density function or a nonlinear regression model. The statistical properties of an estimator of a function
are more complicated than statistical properties of an estimator of a single parameter or of a countable set of parameters.
In Chapter 4 we discussed ways of numerically approximating functions. In this brief chapter we will discuss ways of statistically estimating functions. Many of these methods are based on approximation methods such as orthogonal systems, splines, and kernels discussed
in Chapter 4. The PDF decomposition plays an important role in the estimation of functions.
S ummary
A broadly applicable algorithm for computing maximum likelihood estimates from incomplete data is presented at various levels of generality. Theory showing the monotone behaviour of the likelihood and convergence of the algorithm is derived. Many examples are sketched, including missing value situations, applications to grouped, censored or truncated data, finite mixture models, variance component estimation, hyperparameter estimation, iteratively reweighted least squares and factor analysis.
Many important problems involve clustering large datasets. Although naive implementations of clustering are computationally expensive, there are established efficient techniques for clustering when the dataset has either (1) a limited number of clusters, (2) a low feature dimensionality, or (3) a small number of data points. However, there has been much less work on methods of efficiently clustering datasets that are large in all three ways at once---for example, having millions of data points that exist in many thousands of dimensions representing many thousands of clusters. We present a new technique for clustering these large, high-dimensional datasets. The key idea involves using a cheap, approximate distance measure to efficiently divide the data into overlapping subsets we call canopies. Then clustering is performed by measuring exact distances only between points that occur in a common canopy. Using canopies, large clustering problems that were formerly impossible become practical. U...
This article examines socioeconomic differences in insurance for dental services among Canadians aged 15 or older and factors associated with the use of dental services.
The data on dental insurance coverage and use of dental services are from the cross-sectional file of Statistics Canada's 1996/97 National Population Health Survey. The sample size of respondents aged 15 or older was 70,884.
Logistic regression analysis was used to model variables related to dental insurance coverage and to dental visits in the past year. A weighted bootstrap resampling procedure was used to derive variance estimates.
In 1996/97, 53% of the population aged 15 or older reported having dental insurance, and 59% said they had visited a dentist in the past year. But even when they had insurance, individuals with low incomes and low educational attainment had much lower odds of visiting a dentist than those with higher incomes and more education.
This article, based on the Andersen model, describes patterns of consultation with general practitioners (GPs) and specialists among Canadians aged 18 or older. Associations with health status and other factors are examined.
Estimates are based on data from the 2005 Canadian Community Health Survey (CCHS), cycle 3.1.
Cross-tabulations were used to estimate the proportion of adult Canadians who had had a GP consultation, four or more GP consultations, or a specialist consultation in the previous year. Adjusted logistic regression models were used to examine factors associated with such consultations when the effects of health need were taken into account.
In 2005, 77% of Canadians aged 18 to 64 and 88% of seniors reported that they had consulted a GP in the previous year; 25% and 44%, respectively, had done so four or more times; and 27% and 34% had consulted a specialist. Individual health need, as measured by the presence of chronic conditions and self-reported general and mental health, was a strong determinant of service use. However, when need was taken into account, physician consultations were independently associated with age, sex, household income, race, language, urban/rural residence and having a regular family doctor. Seniors aged 75 or older and rural residents had low odds of specialist consultations, but high odds of four or more GP consultations. Visible minorities and Aboriginal people had lower odds of reporting specialist consultations than did Whites.
Many important problems involve clustering large datasets. Although naive implementations of clustering are computationally expensive, there are established efficient techniques for clustering when the dataset has either (1) a limited number of clusters, (2) a low feature dimensionality, or (3) a small number of data points. However, there has been much less work on methods of efficiently clustering datasets that are large in all three ways at once-for example, having millions of data points that exist in many thousands of dimensions representing many thousands of clusters. We present a new technique for clustering these large, high-dimensional datasets. The key idea involves using a cheap, approximate distance measure to efficiently divide the data into overlapping subsets we call canopies. Then clustering is performed by measuring exact distances only between points that occur in a common canopy. Using canopies, large clustering problems that were formerly impossible become practical. Under reasonable assumptions about the cheap distance metric, this reduction in computational cost comes without any loss in clustering accuracy. Canopies can be applied to many domains and used with a variety of clustering approaches, including Greedy Agglomerative Clustering, K-means and Expectation-Maximization. We present experimental results on grouping bibliographic citations from the reference sections of research papers. Here the canopy approach reduces computation time over a traditional clustering approach by more than an order of magnitude and decreases error in comparison to a previously used algorithm by 25%.
Weighted kappa: Nominal scale agreement with provision for scaled disagreement or partial credit
- H Gilmour
- B P Scott