Decision Tree for Prognostic Classification

Decision Tree for Prognostic Classification Decision Tree for Prognostic Classification of Multivariate Survival Data and Competing Risks 1. Introduction Decision tree (DT) is one way to represent rules underlying data. It is the most popular tool for exploring complex data structures. Besides that it has become one of the most flexible, intuitive and powerful data analytic tools for determining distinct prognostic subgroups with similar outcome within each subgroup but different outcomes between the subgroups (i.e., prognostic grouping of patients). It is hierarchical, sequential classification structures that recursively partition the set of observations. Prognostic groups are important in assessing disease heterogeneity and for design and stratification of future clinical trials. Because patterns of medical treatment are changing so rapidly, it is important that the results of the present analysis be applicable to contemporary patients. Due to their mathematical simplicity, linear regression for continuous data, logistic regression for binary data, proportional hazard regression for censored survival data, marginal and frailty regression for multivariate survival data, and proportional subdistribution hazard regression for competing risks data are among the most commonly used statistical methods. These parametric and semiparametric regression methods, however, may not lead to faithful data descriptions when the underlying assumptions are not satisfied. Sometimes, model interpretation can be problematic in the presence of high-order interactions among predictors. DT has evolved to relax or remove the restrictive assumptions. In many cases, DT is used to explore data structures and to derive parsimonious models. DT is selected to analyze the data rather than the traditional regression analysis for several reasons. Discovery of interactions is difficult using traditional regression, because the interactions must be specified a priori. In contrast, DT automatically detects important interactions. Furthermore, unlike traditional regression analysis, DT is useful in uncovering variables that may be largely operative within a specific patient subgroup but may have minimal effect or none in other patient subgroups. Also, DT provides a superior means for prognostic classification. Rather than fitting a model to the data, DT sequentially divides the patient group into two subgroups based on prognostic factor values (e.g., tumor size The landmark work of DT in statistical community is the Classification and Regression Trees (CART) methodology of Breiman et al. (1984). A different approach was C4.5 proposed by Quinlan (1992). Original DT method was used in classification and regression for categorical and continuous response variable, respectively. In a clinical setting, however, the outcome of primary interest is often duration of survival, time to event, or some other incomplete (that is, censored) outcome. Therefore, several authors have developed extensions of original DT in the setting of censored survival data (Banerjee Noone, 2008). In science and technology, interest often lies in studying processes which generate events repeatedly over time. Such processes are referred to as recurrent event processes and the data they provide are called recurrent event data which includes in multivariate survival data. Such data arise frequently in medical studies, where information is often available on many individuals, each of whom may experience transient clinical events repeatedly over a period of observation. Examples include the occurrence of asthma attacks in respirology trials, epileptic seizures in neurology studies, and fractures in osteoporosis studies. In business, examples include the filing of warranty claims on automobiles, or insurance claims for policy holders. Since multivariate survival times frequently arise when individuals under observation are naturally clustered or when each individual might experience multiple events, then further extensions of DT are developed for such kind of data. In some studies, patients may be simultaneously exposed to several events, each competing for their mortality or morbidity. For example, suppose that a group of patients diagnosed with heart disease is followed in order to observe a myocardial infarction (MI). If by the end of the study each patient was either observed to have MI or was alive and well, then the usual survival techniques can be applied. In real life, however, some patients may die from other causes before experiencing an MI. This is a competing risks situation because death from other causes prohibits the occurrence of MI. MI is considered the event of interest, while death from other causes is considered a competing risk. The group of patients dead of other causes cannot be considered censored, since their observations are not incomplete. The extension of DT can also be employed for competing risks survival time data. These extensions can make one apply the technique to clinical trial data to aid in the development of prognostic classifications for chronic diseases. This chapter will cover DT for multivariate and competing risks survival time data as well as their application in the development of medical prognosis. Two kinds of multivariate survival time regression model, i.e. marginal and frailty regression model, have their own DT extensions. Whereas, the extension of DT for competing risks has two types of tree. First, the â€Å"single event† DT is developed based on splitting function using one event only. Second, the â€Å"composite events† tree which use all the events jointly. 2. Decision Tree A DT is a tree-like structure used for classification, decision theory, clustering, and prediction functions. It depicts rules for dividing data into groups based on the regularities in the data. A DT can be used for categorical and continuous response variables. When the response variables are continuous, the DT is often referred to as a regression tree. If the response variables are categorical, it is called a classification tree. However, the same concepts apply to both types of trees. DTs are widely used in computer science for data structures, in medical sciences for diagnosis, in botany for classification, in psychology for decision theory, and in economic analysis for evaluating investment alternatives. DTs learn from data and generate models containing explicit rule-like relationships among the variables. DT algorithms begin with the entire set of data, split the data into two or more subsets by testing the value of a predictor variable, and then repeatedly split each subset into finer subsets until the split size reaches an appropriate level. The entire modeling process can be illustrated in a tree-like structure. A DT model consists of two parts: creating the tree and applying the tree to the data. To achieve this, DTs use several different algorithms. The most popular algorithm in the statistical community is Classification and Regression Trees (CART) (Breiman et al., 1984). This algorithm helps DTs gain credibility and acceptance in the statistics community. It creates binary splits on nominal or interval predictor variables for a nominal, ordinal, or interval response. The most widely-used algorithms by computer scientists are ID3, C4.5, and C5.0 (Quinlan, 1993). The first version of C4.5 and C5.0 were limited to categorical predictors; however, the most recent versions are similar to CART. Other algorithms include Chi-Square Automatic Interaction Detection (CHAID) for categorical response (Kass, 1980), CLS, AID, TREEDISC, Angoss KnowledgeSEEKER, CRUISE, GUIDE and QUEST (Loh, 2008). These algorithms use different approaches for splitting variables. CART, CRUISE, GUIDE and QUEST use the sta tistical approach, while CLS, ID3, and C4.5 use an approach in which the number of branches off an internal node is equal to the number of possible categories. Another common approach, used by AID, CHAID, and TREEDISC, is the one in which the number of nodes on an internal node varies from two to the maximum number of possible categories. Angoss KnowledgeSEEKER uses a combination of these approaches. Each algorithm employs different mathematical processes to determine how to group and rank variables. Let us illustrate the DT method in a simplified example of credit evaluation. Suppose a credit card issuer wants to develop a model that can be used for evaluating potential candidates based on its historical customer data. The companys main concern is the default of payment by a cardholder. Therefore, the model should be able to help the company classify a candidate as a possible defaulter or not. The database may contain millions of records and hundreds of fields. A fragment of such a database is shown in Table 1. The input variables include income, age, education, occupation, and many others, determined by some quantitative or qualitative methods. The model building process is illustrated in the tree structure in 1. The DT algorithm first selects a variable, income, to split the dataset into two subsets. This variable, and also the splitting value of $31,000, is selected by a splitting criterion of the algorithm. There exist many splitting criteria (Mingers, 1989). The basic principle of these criteria is that they all attempt to divide the data into clusters such that variations within each cluster are minimized and variations between the clusters are maximized. The follow- Name Age Income Education Occupation Default Andrew 42 45600 College Manager No Allison 26 29000 High School Self Owned Yes Sabrina 58 36800 High School Clerk No Andy 35 37300 College Engineer No †¦ Table 1. Partial records and fields of a database table for credit evaluation up splits are similar to the first one. The process continues until an appropriate tree size is reached. 1 shows a segment of the DT. Based on this tree model, a candidate with income at least $31,000 and at least college degree is unlikely to default the payment; but a self-employed candidate whose income is less than $31,000 and age is less than 28 is more likely to default. We begin with a discussion of the general structure of a popular DT algorithm in statistical community, i.e. CART model. A CART model describes the conditional distribution of y given X, where y is the response variable and X is a set of predictor variables (X = (X1,X2,†¦,Xp)). This model has two main components: a tree T with b terminal nodes, and a parameter Q = (q1,q2,†¦, qb) ÃÅ' Rk which associates the parameter values qm, with the mth terminal node. Thus a tree model is fully specified by the pair (T, Q). If X lies in the region corresponding to the mth terminal node then y|X has the distribution f(y|qm), where we use f to represent a conditional distribution indexed by qm. The model is called a regression tree or a classification tree according to whether the response y is quantitative or qualitative, respectively. 2.1 Splitting a tree The DT T subdivides the predictor variable space as follows. Each internal node has an associated splitting rule which uses a predictor to assign observations to either its left or right child node. The internal nodes are thus partitioned into two subsequent nodes using the splitting rule. For quantitative predictors, the splitting rule is based on a split rule c, and assigns observations for which {xi For a regression tree, conventional algorithm models the response in each region Rm as a constant qm. Thus the overall tree model can be expressed as (Hastie et al., 2001): (1) where Rm, m = 1, 2,†¦,b consist of a partition of the predictors space, and therefore representing the space of b terminal nodes. If we adopt the method of minimizing the sum of squares as our criterion to characterize the best split, it is easy to see that the best , is just the average of yi in region Rm: (2) where Nm is the number of observations falling in node m. The residual sum of squares is (3) which will serve as an impurity measure for regression trees. If the response is a factor taking outcomes 1,2, K, the impurity measure Qm(T), defined in (3) is not suitable. Instead, we represent a region Rm with Nm observations with (4) which is the proportion of class k(k ÃŽ {1, 2,†¦,K}) observations in node m. We classify the observations in node m to a class , the majority class in node m. Different measures Qm(T) of node impurity include the following (Hastie et al., 2001): Misclassification error: Gini index: Cross-entropy or deviance: (5) For binary outcomes, if p is the proportion of the second class, these three measures are 1 max(p, 1 p), 2p(1 p) and -p log p (1 p) log(1 p), respectively. All three definitions of impurity are concave, having minimums at p = 0 and p = 1 and a maximum at p = 0.5. Entropy and the Gini index are the most common, and generally give very similar results except when there are two response categories. 2.2 Pruning a tree To be consistent with conventional notations, lets define the impurity of a node h as I(h) ((3) for a regression tree, and any one in (5) for a classification tree). We then choose the split with maximal impurity reduction (6) where hL and hR are the left and right children nodes of h and p(h) is proportion of sample fall in node h. How large should we grow the tree then? Clearly a very large tree might overfit the data, while a small tree may not be able to capture the important structure. Tree size is a tuning parameter governing the models complexity, and the optimal tree size should be adaptively chosen from the data. One approach would be to continue the splitting procedures until the decrease on impurity due to the split exceeds some threshold. This strategy is too short-sighted, however, since a seeming worthless split might lead to a very good split below it. The preferred strategy is to grow a large tree T0, stopping the splitting process when some minimum number of observations in a terminal node (say 10) is reached. Then this large tree is pruned using pruning algorithm, such as cost-complexity or split complexity pruning algorithm. To prune large tree T0 by using cost-complexity algorithm, we define a subtree T T0 to be any tree that can be obtained by pruning T0, and define to be the set of terminal nodes of T. That is, collapsing any number of its terminal nodes. As before, we index terminal nodes by m, with node m representing region Rm. Let denotes the number of terminal nodes in T (= b). We use instead of b following the conventional notation and define the risk of trees and define cost of tree as Regression tree: , Classification tree: , (7) where r(h) measures the impurity of node h in a classification tree (can be any one in (5)). We define the cost complexity criterion (Breiman et al., 1984) (8) where a(> 0) is the complexity parameter. The idea is, for each a, find the subtree Ta T0 to minimize Ra(T). The tuning parameter a > 0 governs the tradeoff between tree size and its goodness of fit to the data (Hastie et al., 2001). Large values of a result in smaller tree Ta and conversely for smaller values of a. As the notation suggests, with a = 0 the solution is the full tree T0. To find Ta we use weakest link pruning: we successively collapse the internal node that produces the smallest per-node increase in R(T), and continue until we produce the single-node (root) tree. This gives a (finite) sequence of subtrees, and one can show this sequence must contains Ta. See Brieman et al. (1984) and Ripley (1996) for details. Estimation of a () is achieved by five- or ten-fold cross-validation. Our final tree is then denoted as . It follows that, in CART and related algorithms, classification and regression trees are produced from data in two stages. In the first stage, a large initial tree is produced by splitting one node at a time in an iterative, greedy fashion. In the second stage, a small subtree of the initial tree is selected, using the same data set. Whereas the splitting procedure proceeds in a top-down fashion, the second stage, known as pruning, proceeds from the bottom-up by successively removing nodes from the initial tree. Theorem 1 (Brieman et al., 1984, Section 3.3) For any value of the complexity parameter a, there is a unique smallest subtree of T0 that minimizes the cost-complexity. Theorem 2 (Zhang Singer, 1999, Section 4.2) If a2 > al, the optimal sub-tree corresponding to a2 is a subtree of the optimal subtree corresponding to al. More general, suppose we end up with m thresholds, 0 (9) where means that is a subtree of . These are called nested optimal subtrees. 3. Decision Tree for Censored Survival Data Survival analysis is the phrase used to describe the analysis of data that correspond to the time from a well-defined time origin until the occurrence of some particular events or end-points. It is important to state what the event is and when the period of observation starts and finish. In medical research, the time origin will often correspond to the recruitment of an individual into an experimental study, and the end-point is the death of the patient or the occurrence of some adverse events. Survival data are rarely normally distributed, but are skewed and comprise typically of many early events and relatively few late ones. It is these features of the data that necessitate the special method survival analysis. The specific difficulties relating to survival analysis arise largely from the fact that only some individuals have experienced the event and, subsequently, survival times will be unknown for a subset of the study group. This phenomenon is called censoring and it may arise in the following ways: (a) a patient has not (yet) experienced the relevant outcome, such as relapse or death, by the time the study has to end; (b) a patient is lost to follow-up during the study period; (c) a patient experiences a different event that makes further follow-up impossible. Generally, censoring times may vary from individual to individual. Such censored survival time underestimated the true (but unknown) time to event. Visualising the survival process of an individual as a time-line, the event (assuming it is to occur) is beyond the end of the follow-up period. This situation is often called right censoring. Most survival data include right censored observation. In many biomedical and reliability studies, interest focuses on relating the time to event to a set of covariates. Cox proportional hazard model (Cox, 1972) has been established as the major framework for analysis of such survival data over the past three decades. But, often in practices, one primary goal of survival analysis is to extract meaningful subgroups of patients determined by the prognostic factors such as patient characteristics that are related to the level of disease. Although proportional hazard model and its extensions are powerful in studying the association between covariates and survival times, usually they are problematic in prognostic classification. One approach for classification is to compute a risk score based on the estimated coefficients from regression methods (Machin et al., 2006). This approach, however, may be problematic for several reasons. First, the definition of risk groups is arbitrary. Secondly, the risk score depends on the correct specification of the model. It is difficult to check whether the model is correct when many covariates are involved. Thirdly, when there are many interaction terms and the model becomes complicated, the result becomes difficult to interpret for the purpose of prognostic classification. Finally, a more serious problem is that an invalid prognostic group may be produced if no patient is included in a covariate profile. In contrast, DT methods do not suffer from these problems. Owing to the development of fast computers, computer-intensive methods such as DT methods have become popular. Since these investigate the significance of all potential risk factors automatically and provide interpretable models, they offer distinct advantages to analysts. Recently a large amount of DT methods have been developed for the analysis of survival data, where the basic concepts for growing and pruning trees remain unchanged, but the choice of the splitting criterion has been modified to incorporate the censored survival data. The application of DT methods for survival data are described by a number of authors (Gordon Olshen, 1985; Ciampi et al., 1986; Segal, 1988; Davis Anderson, 1989; Therneau et al., 1990; LeBlanc Crowley, 1992; LeBlanc Crowley, 1993; Ahn Loh, 1994; Bacchetti Segal, 1995; Huang et al., 1998; KeleÃ…Å ¸ Segal, 2002; Jin et al., 2004; Cappelli Zhang, 2007; Cho Hong, 2008), including the text by Zhang Singer (1999). 4. Decision Tree for Multivariate Censored Survival Data Multivariate survival data frequently arise when we faced the complexity of studies involving multiple treatment centres, family members and measurements repeatedly made on the same individual. For example, in multi-centre clinical trials, the outcomes for groups of patients at several centres are examined. In some instances, patients in a centre might exhibit similar responses due to uniformity of surroundings and procedures within a centre. This would result in correlated outcomes at the level of the treatment centre. For the situation of studies of family members or litters, correlation in outcome is likely for genetic reasons. In this case, the outcomes would be correlated at the family or litter level. Finally, when one person or animal is measured repeatedly over time, correlation will most definitely exist in those responses. Within the context of correlated data, the observations which are correlated for a group of individuals (within a treatment centre or a family) or for on e individual (because of repeated sampling) are referred to as a cluster, so that from this point on, the responses within a cluster will be assumed to be correlated. Analysis of multivariate survival data is complex due to the presence of dependence among survival times and unknown marginal distributions. Multivariate survival times frequently arise when individuals under observation are naturally clustered or when each individual might experience multiple events. A successful treatment of correlated failure times was made by Clayton and Cuzik (1985) who modelled the dependence structure with a frailty term. Another approach is based on a proportional hazard formulation of the marginal hazard function, which has been studied by Wei et al. (1989) and Liang et al. (1993). Noticeably, Prentice et al. (1981) and Andersen Gill (1982) also suggested two alternative approaches to analyze multiple event times. Extension of tree techniques to multivariate censored data is motivated by the classification issue associated with multivariate survival data. For example, clinical investigators design studies to form prognostic rules. Credit risk analysts collect account information to build up credit scoring criteria. Frequently, in such studies the outcomes of ultimate interest are correlated times to event, such as relapses, late payments, or bankruptcies. Since DT methods recursively partition the predictor space, they are an alternative to conventional regression tools. This section is concerned with the generalization of DT models to multivariate survival data. In attempt to facilitate an extension of DT methods to multivariate survival data, more difficulties need to be circumvented. 4.1 Decision tree for multivariate survival data based on marginal model DT methods for multivariate survival data are not many. Almost all the multivariate DT methods have been based on between-node heterogeneity, with the exception of Molinaro et al. (2004) who proposed a general within-node homogeneity approach for both univariate and multivariate data. The multivariate methods proposed by Su Fan (2001, 2004) and Gao et al. (2004, 2006) concentrated on between-node heterogeneity and used the results of regression models. Specifically, for recurrent event data and clustered event data, Su Fan (2004) used likelihood-ratio tests while Gao et al. (2004) used robust Wald tests from a gamma frailty model to maximize the between-node heterogeneity. Su Fan (2001) and Fan et al. (2006) used a robust log-rank statistic while Gao et al. (2006) used a robust Wald test from the marginal failure-time model of Wei et al. (1989). The generalization of DT for multivariate survival data is developed by using goodness of split approach. DT by goodness of split is grown by maximizing a measure of between-node difference. Therefore, only internal nodes have associated two-sample statistics. The tree structure is different from CART because, for trees grown by minimizing within-node error, each node, either terminal or internal, has an associated impurity measure. This is why the CART pruning procedure is not directly applicable to such types of trees. However, the split-complexity pruning algorithm of LeBlanc Crowley (1993) has resulted in trees by goodness of split that has become well-developed tools. This modified tree technique not only provides a convenient way of handling survival data, but also enlarges the applied scope of DT methods in a more general sense. Especially for those situations where defining prediction error terms is relatively difficult, growing trees by a two-sample statistic, together with the split-complexity pruning, offers a feasible way of performing tree analysis. The DT procedure consists of three parts: a method to partition the data recursively into a large tree, a method to prune the large tree into a subtree sequence, and a method to determine the optimal tree size. In the multivariate survival trees, the between-node difference is measured by a robust Wald statistic, which is derived from a marginal approach to multivariate survival data that was developed by Wei et al. (1989). We used split-complexity pruning borrowed from LeBlanc Crowley (1993) and use test sample for determining the right tree size. 4.1.1 The splitting statistic We consider n independent subjects but each subject to have K potential types or number of failures. If there are an unequal number of failures within the subjects, then K is the maximum. We let Tik = min(Yik,Cik ) where Yik = time of the failure in the ith subject for the kth type of failure and Cik = potential censoring time of the ith subject for the kth type of failure with i = 1,†¦,n and k = 1,†¦,K. Then dik = I (Yik ≠¤ Cik) is the indicator for failure and the vector of covariates is denoted Zik = (Z1ik,†¦, Zpik)T. To partition the data, we consider the hazard model for the ith unit for the kth type of failure, using the distinguishable baseline hazard as described by Wei et al. (1989), namely where the indicator function I(Zik Parameter b is estimated by maximizing the partial likelihood. If the observations within the same unit are independent, the partial likelihood functions for b for the distinguishable baseline model (10) would be, (11) Since the observations within the same unit are not independent for multivariate failure time, we refer to the above functions as the pseudo-partial likelihood. The estimator can be obtained by maximizing the likelihood by solving . Wei et al. (1989) showed that is normally distributed with mean 0. However the usual estimate, a-1(b), for the variance of , where (12) is not valid. We refer to a-1(b) as the naà ¯ve estimator. Wei et al. (1989) showed that the correct estimated (robust) variance estimator of is (13) where b(b) is weight and d(b) is often referred to as the robust or sandwich variance estimator. Hence, the robust Wald statistic corresponding to the null hypothesis H0 : b = 0 is (14) 4.1.2 Tree growing To grow a tree, the robust Wald statistic is evaluated for every possible binary split of the predictor space Z. The split, s, could be of several forms: splits on a single covariate, splits on linear combinations of predictors, and boolean combination of splits. The simplest form of split relates to only one covariate, where the split depends on the type of covariate whether it is ordered or nominal covariate. The â€Å"best split† is defined to be the one corresponding to the maximum robust Wald statistic. Subsequently the data are divided into two groups according to the best split. Apply this splitting scheme recursively to the learning sample until the predictor space is partitioned into many regions. There will be no further partition to a node when any of the following occurs: The node contains less than, say 10 or 20, subjects, if the overall sample size is large enough to permit this. We suggest using a larger minimum node size than used in CART where the default value is 5; All the observed times in the subset are censored, which results in unavailability of the robust Wald statistic for any split; All the subjects have identical covariate vectors. Or the node has only complete observations with identical survival times. In these situations, the node is considered as pure. The whole procedure results in a large tree, which could be used for the purpose of data structure exploration. 4.1.3 Tree pruning Let T denote either a particular tree or the set of all its nodes. Let S and denote the set of internal nodes and terminal nodes of T, respectively. Therefore, . Also let |Ãâ€"| denote the number of nodes. Let G(h) represent the maximum robust Wald statistic on a particular (internal) node h. In order to measure the performance of a tree, a split-complexity measure Ga(T) is introduced as in LeBlanc and Crowley (1993). That is, (15) where the number of internal nodes, |S|, measures complexity; G(T) measures goodness of split in T; and the complexity parameter a acts as a penalty for each additional split. Start with the large tree T0 obtained from the splitting procedure. For any internal node h of T0, i.e. h ÃŽ S0, a function g(h) is defined as (16) where Th denotes the branch with h as its root and Sh is the set of all internal nodes of Th. Then the weakest link in T0 is the node such that   < Decision Tree for Prognostic Classification Decision Tree for Prognostic Classification Decision Tree for Prognostic Classification of Multivariate Survival Data and Competing Risks 1. Introduction Decision tree (DT) is one way to represent rules underlying data. It is the most popular tool for exploring complex data structures. Besides that it has become one of the most flexible, intuitive and powerful data analytic tools for determining distinct prognostic subgroups with similar outcome within each subgroup but different outcomes between the subgroups (i.e., prognostic grouping of patients). It is hierarchical, sequential classification structures that recursively partition the set of observations. Prognostic groups are important in assessing disease heterogeneity and for design and stratification of future clinical trials. Because patterns of medical treatment are changing so rapidly, it is important that the results of the present analysis be applicable to contemporary patients. Due to their mathematical simplicity, linear regression for continuous data, logistic regression for binary data, proportional hazard regression for censored survival data, marginal and frailty regression for multivariate survival data, and proportional subdistribution hazard regression for competing risks data are among the most commonly used statistical methods. These parametric and semiparametric regression methods, however, may not lead to faithful data descriptions when the underlying assumptions are not satisfied. Sometimes, model interpretation can be problematic in the presence of high-order interactions among predictors. DT has evolved to relax or remove the restrictive assumptions. In many cases, DT is used to explore data structures and to derive parsimonious models. DT is selected to analyze the data rather than the traditional regression analysis for several reasons. Discovery of interactions is difficult using traditional regression, because the interactions must be specified a priori. In contrast, DT automatically detects important interactions. Furthermore, unlike traditional regression analysis, DT is useful in uncovering variables that may be largely operative within a specific patient subgroup but may have minimal effect or none in other patient subgroups. Also, DT provides a superior means for prognostic classification. Rather than fitting a model to the data, DT sequentially divides the patient group into two subgroups based on prognostic factor values (e.g., tumor size The landmark work of DT in statistical community is the Classification and Regression Trees (CART) methodology of Breiman et al. (1984). A different approach was C4.5 proposed by Quinlan (1992). Original DT method was used in classification and regression for categorical and continuous response variable, respectively. In a clinical setting, however, the outcome of primary interest is often duration of survival, time to event, or some other incomplete (that is, censored) outcome. Therefore, several authors have developed extensions of original DT in the setting of censored survival data (Banerjee Noone, 2008). In science and technology, interest often lies in studying processes which generate events repeatedly over time. Such processes are referred to as recurrent event processes and the data they provide are called recurrent event data which includes in multivariate survival data. Such data arise frequently in medical studies, where information is often available on many individuals, each of whom may experience transient clinical events repeatedly over a period of observation. Examples include the occurrence of asthma attacks in respirology trials, epileptic seizures in neurology studies, and fractures in osteoporosis studies. In business, examples include the filing of warranty claims on automobiles, or insurance claims for policy holders. Since multivariate survival times frequently arise when individuals under observation are naturally clustered or when each individual might experience multiple events, then further extensions of DT are developed for such kind of data. In some studies, patients may be simultaneously exposed to several events, each competing for their mortality or morbidity. For example, suppose that a group of patients diagnosed with heart disease is followed in order to observe a myocardial infarction (MI). If by the end of the study each patient was either observed to have MI or was alive and well, then the usual survival techniques can be applied. In real life, however, some patients may die from other causes before experiencing an MI. This is a competing risks situation because death from other causes prohibits the occurrence of MI. MI is considered the event of interest, while death from other causes is considered a competing risk. The group of patients dead of other causes cannot be considered censored, since their observations are not incomplete. The extension of DT can also be employed for competing risks survival time data. These extensions can make one apply the technique to clinical trial data to aid in the development of prognostic classifications for chronic diseases. This chapter will cover DT for multivariate and competing risks survival time data as well as their application in the development of medical prognosis. Two kinds of multivariate survival time regression model, i.e. marginal and frailty regression model, have their own DT extensions. Whereas, the extension of DT for competing risks has two types of tree. First, the â€Å"single event† DT is developed based on splitting function using one event only. Second, the â€Å"composite events† tree which use all the events jointly. 2. Decision Tree A DT is a tree-like structure used for classification, decision theory, clustering, and prediction functions. It depicts rules for dividing data into groups based on the regularities in the data. A DT can be used for categorical and continuous response variables. When the response variables are continuous, the DT is often referred to as a regression tree. If the response variables are categorical, it is called a classification tree. However, the same concepts apply to both types of trees. DTs are widely used in computer science for data structures, in medical sciences for diagnosis, in botany for classification, in psychology for decision theory, and in economic analysis for evaluating investment alternatives. DTs learn from data and generate models containing explicit rule-like relationships among the variables. DT algorithms begin with the entire set of data, split the data into two or more subsets by testing the value of a predictor variable, and then repeatedly split each subset into finer subsets until the split size reaches an appropriate level. The entire modeling process can be illustrated in a tree-like structure. A DT model consists of two parts: creating the tree and applying the tree to the data. To achieve this, DTs use several different algorithms. The most popular algorithm in the statistical community is Classification and Regression Trees (CART) (Breiman et al., 1984). This algorithm helps DTs gain credibility and acceptance in the statistics community. It creates binary splits on nominal or interval predictor variables for a nominal, ordinal, or interval response. The most widely-used algorithms by computer scientists are ID3, C4.5, and C5.0 (Quinlan, 1993). The first version of C4.5 and C5.0 were limited to categorical predictors; however, the most recent versions are similar to CART. Other algorithms include Chi-Square Automatic Interaction Detection (CHAID) for categorical response (Kass, 1980), CLS, AID, TREEDISC, Angoss KnowledgeSEEKER, CRUISE, GUIDE and QUEST (Loh, 2008). These algorithms use different approaches for splitting variables. CART, CRUISE, GUIDE and QUEST use the sta tistical approach, while CLS, ID3, and C4.5 use an approach in which the number of branches off an internal node is equal to the number of possible categories. Another common approach, used by AID, CHAID, and TREEDISC, is the one in which the number of nodes on an internal node varies from two to the maximum number of possible categories. Angoss KnowledgeSEEKER uses a combination of these approaches. Each algorithm employs different mathematical processes to determine how to group and rank variables. Let us illustrate the DT method in a simplified example of credit evaluation. Suppose a credit card issuer wants to develop a model that can be used for evaluating potential candidates based on its historical customer data. The companys main concern is the default of payment by a cardholder. Therefore, the model should be able to help the company classify a candidate as a possible defaulter or not. The database may contain millions of records and hundreds of fields. A fragment of such a database is shown in Table 1. The input variables include income, age, education, occupation, and many others, determined by some quantitative or qualitative methods. The model building process is illustrated in the tree structure in 1. The DT algorithm first selects a variable, income, to split the dataset into two subsets. This variable, and also the splitting value of $31,000, is selected by a splitting criterion of the algorithm. There exist many splitting criteria (Mingers, 1989). The basic principle of these criteria is that they all attempt to divide the data into clusters such that variations within each cluster are minimized and variations between the clusters are maximized. The follow- Name Age Income Education Occupation Default Andrew 42 45600 College Manager No Allison 26 29000 High School Self Owned Yes Sabrina 58 36800 High School Clerk No Andy 35 37300 College Engineer No †¦ Table 1. Partial records and fields of a database table for credit evaluation up splits are similar to the first one. The process continues until an appropriate tree size is reached. 1 shows a segment of the DT. Based on this tree model, a candidate with income at least $31,000 and at least college degree is unlikely to default the payment; but a self-employed candidate whose income is less than $31,000 and age is less than 28 is more likely to default. We begin with a discussion of the general structure of a popular DT algorithm in statistical community, i.e. CART model. A CART model describes the conditional distribution of y given X, where y is the response variable and X is a set of predictor variables (X = (X1,X2,†¦,Xp)). This model has two main components: a tree T with b terminal nodes, and a parameter Q = (q1,q2,†¦, qb) ÃÅ' Rk which associates the parameter values qm, with the mth terminal node. Thus a tree model is fully specified by the pair (T, Q). If X lies in the region corresponding to the mth terminal node then y|X has the distribution f(y|qm), where we use f to represent a conditional distribution indexed by qm. The model is called a regression tree or a classification tree according to whether the response y is quantitative or qualitative, respectively. 2.1 Splitting a tree The DT T subdivides the predictor variable space as follows. Each internal node has an associated splitting rule which uses a predictor to assign observations to either its left or right child node. The internal nodes are thus partitioned into two subsequent nodes using the splitting rule. For quantitative predictors, the splitting rule is based on a split rule c, and assigns observations for which {xi For a regression tree, conventional algorithm models the response in each region Rm as a constant qm. Thus the overall tree model can be expressed as (Hastie et al., 2001): (1) where Rm, m = 1, 2,†¦,b consist of a partition of the predictors space, and therefore representing the space of b terminal nodes. If we adopt the method of minimizing the sum of squares as our criterion to characterize the best split, it is easy to see that the best , is just the average of yi in region Rm: (2) where Nm is the number of observations falling in node m. The residual sum of squares is (3) which will serve as an impurity measure for regression trees. If the response is a factor taking outcomes 1,2, K, the impurity measure Qm(T), defined in (3) is not suitable. Instead, we represent a region Rm with Nm observations with (4) which is the proportion of class k(k ÃŽ {1, 2,†¦,K}) observations in node m. We classify the observations in node m to a class , the majority class in node m. Different measures Qm(T) of node impurity include the following (Hastie et al., 2001): Misclassification error: Gini index: Cross-entropy or deviance: (5) For binary outcomes, if p is the proportion of the second class, these three measures are 1 max(p, 1 p), 2p(1 p) and -p log p (1 p) log(1 p), respectively. All three definitions of impurity are concave, having minimums at p = 0 and p = 1 and a maximum at p = 0.5. Entropy and the Gini index are the most common, and generally give very similar results except when there are two response categories. 2.2 Pruning a tree To be consistent with conventional notations, lets define the impurity of a node h as I(h) ((3) for a regression tree, and any one in (5) for a classification tree). We then choose the split with maximal impurity reduction (6) where hL and hR are the left and right children nodes of h and p(h) is proportion of sample fall in node h. How large should we grow the tree then? Clearly a very large tree might overfit the data, while a small tree may not be able to capture the important structure. Tree size is a tuning parameter governing the models complexity, and the optimal tree size should be adaptively chosen from the data. One approach would be to continue the splitting procedures until the decrease on impurity due to the split exceeds some threshold. This strategy is too short-sighted, however, since a seeming worthless split might lead to a very good split below it. The preferred strategy is to grow a large tree T0, stopping the splitting process when some minimum number of observations in a terminal node (say 10) is reached. Then this large tree is pruned using pruning algorithm, such as cost-complexity or split complexity pruning algorithm. To prune large tree T0 by using cost-complexity algorithm, we define a subtree T T0 to be any tree that can be obtained by pruning T0, and define to be the set of terminal nodes of T. That is, collapsing any number of its terminal nodes. As before, we index terminal nodes by m, with node m representing region Rm. Let denotes the number of terminal nodes in T (= b). We use instead of b following the conventional notation and define the risk of trees and define cost of tree as Regression tree: , Classification tree: , (7) where r(h) measures the impurity of node h in a classification tree (can be any one in (5)). We define the cost complexity criterion (Breiman et al., 1984) (8) where a(> 0) is the complexity parameter. The idea is, for each a, find the subtree Ta T0 to minimize Ra(T). The tuning parameter a > 0 governs the tradeoff between tree size and its goodness of fit to the data (Hastie et al., 2001). Large values of a result in smaller tree Ta and conversely for smaller values of a. As the notation suggests, with a = 0 the solution is the full tree T0. To find Ta we use weakest link pruning: we successively collapse the internal node that produces the smallest per-node increase in R(T), and continue until we produce the single-node (root) tree. This gives a (finite) sequence of subtrees, and one can show this sequence must contains Ta. See Brieman et al. (1984) and Ripley (1996) for details. Estimation of a () is achieved by five- or ten-fold cross-validation. Our final tree is then denoted as . It follows that, in CART and related algorithms, classification and regression trees are produced from data in two stages. In the first stage, a large initial tree is produced by splitting one node at a time in an iterative, greedy fashion. In the second stage, a small subtree of the initial tree is selected, using the same data set. Whereas the splitting procedure proceeds in a top-down fashion, the second stage, known as pruning, proceeds from the bottom-up by successively removing nodes from the initial tree. Theorem 1 (Brieman et al., 1984, Section 3.3) For any value of the complexity parameter a, there is a unique smallest subtree of T0 that minimizes the cost-complexity. Theorem 2 (Zhang Singer, 1999, Section 4.2) If a2 > al, the optimal sub-tree corresponding to a2 is a subtree of the optimal subtree corresponding to al. More general, suppose we end up with m thresholds, 0 (9) where means that is a subtree of . These are called nested optimal subtrees. 3. Decision Tree for Censored Survival Data Survival analysis is the phrase used to describe the analysis of data that correspond to the time from a well-defined time origin until the occurrence of some particular events or end-points. It is important to state what the event is and when the period of observation starts and finish. In medical research, the time origin will often correspond to the recruitment of an individual into an experimental study, and the end-point is the death of the patient or the occurrence of some adverse events. Survival data are rarely normally distributed, but are skewed and comprise typically of many early events and relatively few late ones. It is these features of the data that necessitate the special method survival analysis. The specific difficulties relating to survival analysis arise largely from the fact that only some individuals have experienced the event and, subsequently, survival times will be unknown for a subset of the study group. This phenomenon is called censoring and it may arise in the following ways: (a) a patient has not (yet) experienced the relevant outcome, such as relapse or death, by the time the study has to end; (b) a patient is lost to follow-up during the study period; (c) a patient experiences a different event that makes further follow-up impossible. Generally, censoring times may vary from individual to individual. Such censored survival time underestimated the true (but unknown) time to event. Visualising the survival process of an individual as a time-line, the event (assuming it is to occur) is beyond the end of the follow-up period. This situation is often called right censoring. Most survival data include right censored observation. In many biomedical and reliability studies, interest focuses on relating the time to event to a set of covariates. Cox proportional hazard model (Cox, 1972) has been established as the major framework for analysis of such survival data over the past three decades. But, often in practices, one primary goal of survival analysis is to extract meaningful subgroups of patients determined by the prognostic factors such as patient characteristics that are related to the level of disease. Although proportional hazard model and its extensions are powerful in studying the association between covariates and survival times, usually they are problematic in prognostic classification. One approach for classification is to compute a risk score based on the estimated coefficients from regression methods (Machin et al., 2006). This approach, however, may be problematic for several reasons. First, the definition of risk groups is arbitrary. Secondly, the risk score depends on the correct specification of the model. It is difficult to check whether the model is correct when many covariates are involved. Thirdly, when there are many interaction terms and the model becomes complicated, the result becomes difficult to interpret for the purpose of prognostic classification. Finally, a more serious problem is that an invalid prognostic group may be produced if no patient is included in a covariate profile. In contrast, DT methods do not suffer from these problems. Owing to the development of fast computers, computer-intensive methods such as DT methods have become popular. Since these investigate the significance of all potential risk factors automatically and provide interpretable models, they offer distinct advantages to analysts. Recently a large amount of DT methods have been developed for the analysis of survival data, where the basic concepts for growing and pruning trees remain unchanged, but the choice of the splitting criterion has been modified to incorporate the censored survival data. The application of DT methods for survival data are described by a number of authors (Gordon Olshen, 1985; Ciampi et al., 1986; Segal, 1988; Davis Anderson, 1989; Therneau et al., 1990; LeBlanc Crowley, 1992; LeBlanc Crowley, 1993; Ahn Loh, 1994; Bacchetti Segal, 1995; Huang et al., 1998; KeleÃ…Å ¸ Segal, 2002; Jin et al., 2004; Cappelli Zhang, 2007; Cho Hong, 2008), including the text by Zhang Singer (1999). 4. Decision Tree for Multivariate Censored Survival Data Multivariate survival data frequently arise when we faced the complexity of studies involving multiple treatment centres, family members and measurements repeatedly made on the same individual. For example, in multi-centre clinical trials, the outcomes for groups of patients at several centres are examined. In some instances, patients in a centre might exhibit similar responses due to uniformity of surroundings and procedures within a centre. This would result in correlated outcomes at the level of the treatment centre. For the situation of studies of family members or litters, correlation in outcome is likely for genetic reasons. In this case, the outcomes would be correlated at the family or litter level. Finally, when one person or animal is measured repeatedly over time, correlation will most definitely exist in those responses. Within the context of correlated data, the observations which are correlated for a group of individuals (within a treatment centre or a family) or for on e individual (because of repeated sampling) are referred to as a cluster, so that from this point on, the responses within a cluster will be assumed to be correlated. Analysis of multivariate survival data is complex due to the presence of dependence among survival times and unknown marginal distributions. Multivariate survival times frequently arise when individuals under observation are naturally clustered or when each individual might experience multiple events. A successful treatment of correlated failure times was made by Clayton and Cuzik (1985) who modelled the dependence structure with a frailty term. Another approach is based on a proportional hazard formulation of the marginal hazard function, which has been studied by Wei et al. (1989) and Liang et al. (1993). Noticeably, Prentice et al. (1981) and Andersen Gill (1982) also suggested two alternative approaches to analyze multiple event times. Extension of tree techniques to multivariate censored data is motivated by the classification issue associated with multivariate survival data. For example, clinical investigators design studies to form prognostic rules. Credit risk analysts collect account information to build up credit scoring criteria. Frequently, in such studies the outcomes of ultimate interest are correlated times to event, such as relapses, late payments, or bankruptcies. Since DT methods recursively partition the predictor space, they are an alternative to conventional regression tools. This section is concerned with the generalization of DT models to multivariate survival data. In attempt to facilitate an extension of DT methods to multivariate survival data, more difficulties need to be circumvented. 4.1 Decision tree for multivariate survival data based on marginal model DT methods for multivariate survival data are not many. Almost all the multivariate DT methods have been based on between-node heterogeneity, with the exception of Molinaro et al. (2004) who proposed a general within-node homogeneity approach for both univariate and multivariate data. The multivariate methods proposed by Su Fan (2001, 2004) and Gao et al. (2004, 2006) concentrated on between-node heterogeneity and used the results of regression models. Specifically, for recurrent event data and clustered event data, Su Fan (2004) used likelihood-ratio tests while Gao et al. (2004) used robust Wald tests from a gamma frailty model to maximize the between-node heterogeneity. Su Fan (2001) and Fan et al. (2006) used a robust log-rank statistic while Gao et al. (2006) used a robust Wald test from the marginal failure-time model of Wei et al. (1989). The generalization of DT for multivariate survival data is developed by using goodness of split approach. DT by goodness of split is grown by maximizing a measure of between-node difference. Therefore, only internal nodes have associated two-sample statistics. The tree structure is different from CART because, for trees grown by minimizing within-node error, each node, either terminal or internal, has an associated impurity measure. This is why the CART pruning procedure is not directly applicable to such types of trees. However, the split-complexity pruning algorithm of LeBlanc Crowley (1993) has resulted in trees by goodness of split that has become well-developed tools. This modified tree technique not only provides a convenient way of handling survival data, but also enlarges the applied scope of DT methods in a more general sense. Especially for those situations where defining prediction error terms is relatively difficult, growing trees by a two-sample statistic, together with the split-complexity pruning, offers a feasible way of performing tree analysis. The DT procedure consists of three parts: a method to partition the data recursively into a large tree, a method to prune the large tree into a subtree sequence, and a method to determine the optimal tree size. In the multivariate survival trees, the between-node difference is measured by a robust Wald statistic, which is derived from a marginal approach to multivariate survival data that was developed by Wei et al. (1989). We used split-complexity pruning borrowed from LeBlanc Crowley (1993) and use test sample for determining the right tree size. 4.1.1 The splitting statistic We consider n independent subjects but each subject to have K potential types or number of failures. If there are an unequal number of failures within the subjects, then K is the maximum. We let Tik = min(Yik,Cik ) where Yik = time of the failure in the ith subject for the kth type of failure and Cik = potential censoring time of the ith subject for the kth type of failure with i = 1,†¦,n and k = 1,†¦,K. Then dik = I (Yik ≠¤ Cik) is the indicator for failure and the vector of covariates is denoted Zik = (Z1ik,†¦, Zpik)T. To partition the data, we consider the hazard model for the ith unit for the kth type of failure, using the distinguishable baseline hazard as described by Wei et al. (1989), namely where the indicator function I(Zik Parameter b is estimated by maximizing the partial likelihood. If the observations within the same unit are independent, the partial likelihood functions for b for the distinguishable baseline model (10) would be, (11) Since the observations within the same unit are not independent for multivariate failure time, we refer to the above functions as the pseudo-partial likelihood. The estimator can be obtained by maximizing the likelihood by solving . Wei et al. (1989) showed that is normally distributed with mean 0. However the usual estimate, a-1(b), for the variance of , where (12) is not valid. We refer to a-1(b) as the naà ¯ve estimator. Wei et al. (1989) showed that the correct estimated (robust) variance estimator of is (13) where b(b) is weight and d(b) is often referred to as the robust or sandwich variance estimator. Hence, the robust Wald statistic corresponding to the null hypothesis H0 : b = 0 is (14) 4.1.2 Tree growing To grow a tree, the robust Wald statistic is evaluated for every possible binary split of the predictor space Z. The split, s, could be of several forms: splits on a single covariate, splits on linear combinations of predictors, and boolean combination of splits. The simplest form of split relates to only one covariate, where the split depends on the type of covariate whether it is ordered or nominal covariate. The â€Å"best split† is defined to be the one corresponding to the maximum robust Wald statistic. Subsequently the data are divided into two groups according to the best split. Apply this splitting scheme recursively to the learning sample until the predictor space is partitioned into many regions. There will be no further partition to a node when any of the following occurs: The node contains less than, say 10 or 20, subjects, if the overall sample size is large enough to permit this. We suggest using a larger minimum node size than used in CART where the default value is 5; All the observed times in the subset are censored, which results in unavailability of the robust Wald statistic for any split; All the subjects have identical covariate vectors. Or the node has only complete observations with identical survival times. In these situations, the node is considered as pure. The whole procedure results in a large tree, which could be used for the purpose of data structure exploration. 4.1.3 Tree pruning Let T denote either a particular tree or the set of all its nodes. Let S and denote the set of internal nodes and terminal nodes of T, respectively. Therefore, . Also let |Ãâ€"| denote the number of nodes. Let G(h) represent the maximum robust Wald statistic on a particular (internal) node h. In order to measure the performance of a tree, a split-complexity measure Ga(T) is introduced as in LeBlanc and Crowley (1993). That is, (15) where the number of internal nodes, |S|, measures complexity; G(T) measures goodness of split in T; and the complexity parameter a acts as a penalty for each additional split. Start with the large tree T0 obtained from the splitting procedure. For any internal node h of T0, i.e. h ÃŽ S0, a function g(h) is defined as (16) where Th denotes the branch with h as its root and Sh is the set of all internal nodes of Th. Then the weakest link in T0 is the node such that   <

Social Work Evidence Based Practice Strengths and Weaknesses

Social Work Evidence Based Practice Strengths and Weaknesses Critically analyse the strengths and weaknesses of using an evidence based approach to a specific area of social work practice, referring to recent research findings in this area. Specific area: Older adults and mental health Outlined below is an analysis of the strengths and weaknesses of using an evidence based approach to a specific area of social work practice, referring to recent findings with regards to older people and mental health in particular. Social work practice is split into various areas in order, to effectively assist with the problems of different individuals as well as social groups such as older adults with, or affected by mental health conditions and problems such as depression and dementia. Older adults with mental health conditions do and have benefited from well-focused social work practices. The use of the evidence based approach alone or alongside other factors has the capacity to shape social work practices designed to help and protect older adults with mental health conditions. Of course to maintain relevance as well as professional effectiveness social work practices should reflect the needs of the people being cared for not to mention altering to better practices as empirically demonstrated by the most recent research findings.[1] As mentioned below evidence based approaches to assisting older people with mental health conditions or issues have led to suggested methods to improve the social work practices that form the basis of the services provided for those vulnerable people that need to use them being made.[2] Social workers that are tasked with assisting older people with, or affected by mental health problems ideally should use the social work practices that have been proven to assist vulnerable older people the most. Therefore evidence-based approaches can help social workers determine the best working practices to actually use. To begin with it can be reasonably argued that using an evidence-based approach to the specific social work practices offers social workers working with older people linked with mental health issues strengths. The need for reliable and accurate research into older people and mental health for social work practices is undoubtedly becoming more important as the British population in common with many Western societies is ageing as people are generally living for longer. [3] Past and recent medical research has frequently shown that there is a strong connection between an ageing population taken as a whole and mental health issues becoming more common for older people. An ageing population raises issues and concerns for the National Health Service (NHS), and the Department for Work and Pensions (DWP), which includes the Pension Service as much as it does for social work practices.[4] Limited parts of medical research points to some mental health issues or conditions in older people as being preventable, and even in certain circumstances reversible. Social work practices ideally should be altered if it is possible to assist older people that can recover from temporary or reversible mental health problems to do so. However some mental health conditions and problems affecting older people such as vascular and senile dementia cannot be prevented and reversed, though medication can delay their full onset. In such circumstances NHS or private sector health care services not to mention social work practices have to be adjusted to cope with a higher demand for their use. When possible social work practices should be altered in order to prevent, tackle, or reverse the mental health conditions and that could possibly restore older people to full health. Social work practices should also help older people whose mental health conditions cannot be reversed.[5] When older adults are affected by mental health conditions whether upon a short – term basis or upon a long – term basis then social work practices arguably needs to be adaptable as well as practical enough to assist those older adults more effectively. Older adults with mental health  conditions may previously have been active normal people used to doing everything for them that suddenly find things much more difficult once their condition or illness becomes well established. Ongoing medical research is slowly finding new treatments that can in the right circumstances help older adults with mental health conditions and illnesses remain as healthy as possible for as long as possible. The basic assumption here is that when older adults with mental health conditions stay healthier for a longer period of time they could therefore be less reliant upon their carers, the NHS, social security benefits, as well as the social serves provided to them by social workers.[6] The reversing or the delaying of the worst or most debilitating of mental health conditions in older adults can help those people to lead normal and independent lives for as long as possible. Older adults that are able to resist or overcome the worst consequences of their mental health conditions will be more likely to retain their self-dignity as well as their independence. It is the leading of independent lives by the older adults affected to varying degrees by mental health conditions that can potentially receive the most beneficial assistance from their social workers as well as all relevant medical staff. Social work practice that enables older adults to remain in their homes and were possible within their own families gives such vulnerable people a sense of stability as well as helping to keep them in familiar surroundings. The strengths of using evidence based approach to guide and shape social work practices towards older adults with mental health conditions are therefore in many respects straightforward to understand. As in the majority of social work fields or areas, social workers make use of an evidence-based approach to increase the effectiveness of the social work services that they actually provide to the people who require their assistance the most. Using an evidence-based approach provides data or research that provided it is gathered efficiently and interpreted accurately provides information to social workers to point out the most effective social work practices. Using an evidence based approach has the strength of allowing the social workers who form social work practices to alter those practices to help larger numbers of older adults with mental health conditions to receive the social work services that should benefit them the most. It should also follow on logically that using an evidence-based approach would allow the resources and also the social services by the social work providers to be used most effectively. Older adults with mental health conditions can influence the type of social work provided to them by highlighting the best as well as the less effective social work practices.[7] At the end of the day social work practices are only there to help the most vulnerable older adults amongst other social groups and isolated individuals even though these social work practices might not always be popular with social workers themselves. The strength as well as the point of using the evidence-based approach is therefore that the best interests of the most vulnerable older adults with mental health conditions should always is taken into account.[8] Indeed the best interests of the most vulnerable older people as revealed via the use of the evidence-based approach are incorporated into both completely new as well as revised social work practices. In theory, and also to a very large extent in practice the use of the evidence based approach is that it has the strength of providing social workers with information and indications about, which older adults that need help due to their mental health conditions and issues.[9] The evidence based approach means social workers know where to concentrate help as well as guidance as to what kind of social work service would be most useful for these older adults.[10] There are actually as well as potentially sources of weakness when using the evidence based approach in forming and subsequently amending when necessary social work practices in relation to older adults with mental health conditions or issues. The evidence based approach to providing information with regard to older adults with mental health conditions is only useful in many respects when combined with other information or procedures that form the basis of social work practices.[11] To begin with social workers might have to make decisions with regard to the help or social services that individual older adults with mental health conditions or problems and whether or not to offer them help before the evidence based approach has provided enough information about these specific older adults.[12] It can and indeed does take time for the evidence based approach to be gathered and evaluated before it is strong enough to alter or perhaps even entirely replace all the relevant social work practices. Social workers need to have social work practices set in place all of the time in order for them to be always able to offer vulnerable older adults their help and advice with regard to mental health conditions. When or if existing social work practices are deemed to be highly effective in helping as well as protecting older adults with mental health conditions there would have to be convincing evidence. Without wide ranging proof that the findings and suggestions from the ev idence-based approach could offer profound improvements to the social work services they need not take place. It could improve the social services available for older adults that need or could eventually need to use them in the future.[13] One of the weaknesses of using the evidence based approach to assist in the formation and the subsequent shape of social work practices for older adults with mental health conditions is that such an approach does not take other factors or organisations into account.[14] An over reliance upon the use of the evidence based approach to drawing up social work practices to assist older adults with mental health conditions is that a narrowly focused concentration on such an approach could lead to an underestimation of other actors.[15] Social workers could underestimate the importance of other private sector and public sector bodies, groups, and organisations that offer services to older adults. Other private sector and also public sector bodies, groups, and organisations offer services to older adults with mental health problems either because the government tasks them with doing so, or because they can make money from doing so.[16] In reality this weakness of using the evidence based approach when drawing up social work practices is not really a problem. This weakness is not a profound or serious problem because social work departments are used to working with private sector as well as public sector bodies, companies, groups, and organisations when they set out to deliver social services to the general public taken as a whole.[17] Providing the appropriate services for older adults with mental health conditions really is an example of public services being delivered by a whole host of private sector as well as public services groups, organisations, and agencies.[18] Alongside social work departments, the DWP, the NHS, private sector care homes, local authorities, as well as charities for older adults and people with mental conditions provide services.[19] The weaknesses of using the evidence based approach for developing social work practices for assisting older adults is that the evidence might not be as complete as would be helpful to social workers. Research into the affects and consequences of mental health issues or problems for older adults has arguably proved insufficient until recently given the ageing populations within the majority of Western societies.[20] Older adults with dementia or similar mental health conditions frequently need a great deal of care and support as do their carers, and also their families in order to cope with the consequences of severe mental illnesses. Sometimes it is the spouses, the children, or indeed other relatives of the older adults with the more pronounced mental health conditions who can be most adversely affected by the worst consequences of their loved ones’ illness.[21] There are weaknesses that the evidence based approach towards setting social work practices for older adults with mental health conditions relating to the limited scope of that approach. Research into older adults with mental health problems and conditions has not always examined the ill affects that such conditions have upon the carers and the relatives of the people with the illnesses or health complaints. [22] On a practical level social work practices can be as much about supporting tired or distressed close relatives and carers as it is about assisting the older adults afflicted with or by poor and worsening mental health conditions.[23] When mental health conditions, issues, or problems adversely affect older adults then it can strain their relationships with their close relatives, especially their spouses or children, which in turn causes stress to all of them. Strained relationships between older adults with mental health conditions and the close relatives caring for them can lead to those older adults moving into care homes and hospitals rather than been cared for at home. Helping older adults with mental health conditions get better can have the benefit of allowing some of them to continue working or indeed to start working again.[24] To conclude it has been argued that the use of the evidence based approach to draw up and shape social work practices in relation to older adults with mental health conditions can offer strengths as well as weaknesses to social workers. The evidence based approach to drawing up or shaping social work practices for older adults with mental health conditions offers the strengths of providing social workers with relevant information and data to draw effective policies. The information and data from the evidence based approach research allows social work departments to increase the effectiveness of their social work practices towards older adults with mental health problems. The main weaknesses of using the evidence based approach towards forming and amending social work practices is that it can cause a delay in relevant information being used by social workers, and that it can underestimate the part that other organisations play in providing services to older adults. Bibliography Audit Commission / Better Government for Older People – Older People, independence and well-being: The challenge for public services, Public Sector Briefing Bowers H, Eastman M, Harris J, Macadam A (2005) Moving out of the Shadows – A report on mental health and wellbeing in later life, Health Care Development Ltd, London Brooke L and Taylor P, Older workers and employment: managing age relations, Ageing society 25, 2005, 415-429, Cambridge University Press Department of Health, A Sure Start to later life, Ending inequalities for older people, January 2006 Estes, C.L. Biggs, S. and Phillipson, C. (2003), Social Theory, Social Policy and Ageing A critical introduction, Open University Press, Maidenhead Maria Evandrou and Karen Glaser, Combining work and family life: the pension penalty of care, Ageing and Society 23, 2003, 583-601, Cambridge University Press House of Commons Committee of Public Accounts, Improving Public Service for Older People, Twenty- Ninth Report of Session 2003-04 (May 2004) Riseborough M Jenkins C (April 2004), Now you see me†¦now you don’t How are older citizens being included in regeneration? Age Concern, London Vincent, J., Phillipson, C. Downs M., (eds) (2006) The Futures of Old Age, Sage 1 Footnotes [1] Vincent et al, 2006 [2] Brooke and Taylor, 2005 [3] Vincent et al, 2006 [4] Brooke and Taylor, 2005 [5] Vincent et al, 2006 [6] Brooke and Taylor, 2005 [7] Evandrou Glaser, 2003 [8] Brooke and Taylor, 2005 [9] House of Commons Committee of Public Accounts, May 2004 p. 10 [10] Vincent et al, 2006 [11] Department of Health, 2006 [12] Audit Commission / Better Government for Older People, p.2 [13] Riseborough Jenkins, April 2004 p. 6 [14] Evandrou Glaser, 2003 [15] Department of Health, 2006 [16] Brooke and Taylor, 2005 [17] Riseborough Jenkins, April 2004 p. 6 [18] Audit Commission / Better Government for Older People, p.2 [19] Vincent et al, 2006 [20] Riseborough Jenkins, April 2004 p. 6 [21] Brooke and Taylor, 2005 [22] Riseborough Jenkins, April 2004 p. 6 [23] House of Commons Committee of Public Accounts, May 2004 p. 10 [24] Department of Health, 2006

Pretentious Mothers in Williams’ The Glass Menagerie and Amy Tans’ Two

In the play The Glass Menagerie by Tennessee Williams and the short story â€Å"Two Kinds† by Amy Tan, a theme of embattled control is established through the association with their children. William’s long-winded Amanda is an overpowering, delusional Southern belle mother. Dead set on finding her slightly handicapped daughter Laura a suitable husband, and dictating how her creative, yet bored, son Tom should conduct himself in life. Amanda, through her nagging and domineering instructions over everything each of her children do, from how they eat, to how they should live out their lives, pushes them into mental seclusion. The subsequent overbearing behavior by Amanda in due course drives Tom away leaving Laura in complete solitude with her mother. In a likewise manner, Tan depicts her character Suyuan as a very ambitious overconfident Chinese mother with visions of grandeur for her daughter Jing-Mei. Suyuan after being influenced by a television show decides th at Jing-Mei is to become a child prodigy. Through strict educational instructions, she drives Jing-Mei to a point of contentious revolt. The consequences of Suyuan’s authoritarian treatment to make Jing-Mei a star result in an outburst, after a talent show, causing a deep rift of silence between the two that lasts for twenty years. The parental domineering nature of Amanda in The Glass Menagerie and Suyuan in â€Å"Two Kinds† ultimately fail to force any lasting influential direction on their children, compelling them to follow entirely contradictory paths than the ones preferred by their mothers, forcing their children into rebellion. Nonetheless both mothers wield strong authority over their children, and the motives behind their dictatorial control are directly rela... ...ious physical and mental control over her daughter to become a child prodigy results in a belligerent resistance by Jing-mei. The confrontation between Suyuan and her daughter ultimately drives a rift between the two that lasts for many years. In summary Amanda and Laura would appear to have yielded similar disappointing results by their endeavors to control the destinies of the children. Works Cited Tan, Amy. "Two Kinds." Kirszner, Laurie G. and Stephen R. Mandell. Literature: Reading, Reacting, Writing, Compact Seventh Edition. Ed. Karen Mauk. 7th. Upper Saddle River, New Jersey: Michael Rosenberg, 2009. 694-702. Williams, Tennessee. "The Glass Menagerie." Kirszner, Laurie G. and Stephen R. Mandell. Literature: Reading, Reacting, Writing, Compact Seventh Edition. Ed. Karen Mauk. 7th. Upper Saddle River, New Jersey: Michael Rosenberg, 2009. 1628-1677. Pretentious Mothers in Williams’ The Glass Menagerie and Amy Tans’ Two In the play The Glass Menagerie by Tennessee Williams and the short story â€Å"Two Kinds† by Amy Tan, a theme of embattled control is established through the association with their children. William’s long-winded Amanda is an overpowering, delusional Southern belle mother. Dead set on finding her slightly handicapped daughter Laura a suitable husband, and dictating how her creative, yet bored, son Tom should conduct himself in life. Amanda, through her nagging and domineering instructions over everything each of her children do, from how they eat, to how they should live out their lives, pushes them into mental seclusion. The subsequent overbearing behavior by Amanda in due course drives Tom away leaving Laura in complete solitude with her mother. In a likewise manner, Tan depicts her character Suyuan as a very ambitious overconfident Chinese mother with visions of grandeur for her daughter Jing-Mei. Suyuan after being influenced by a television show decides th at Jing-Mei is to become a child prodigy. Through strict educational instructions, she drives Jing-Mei to a point of contentious revolt. The consequences of Suyuan’s authoritarian treatment to make Jing-Mei a star result in an outburst, after a talent show, causing a deep rift of silence between the two that lasts for twenty years. The parental domineering nature of Amanda in The Glass Menagerie and Suyuan in â€Å"Two Kinds† ultimately fail to force any lasting influential direction on their children, compelling them to follow entirely contradictory paths than the ones preferred by their mothers, forcing their children into rebellion. Nonetheless both mothers wield strong authority over their children, and the motives behind their dictatorial control are directly rela... ...ious physical and mental control over her daughter to become a child prodigy results in a belligerent resistance by Jing-mei. The confrontation between Suyuan and her daughter ultimately drives a rift between the two that lasts for many years. In summary Amanda and Laura would appear to have yielded similar disappointing results by their endeavors to control the destinies of the children. Works Cited Tan, Amy. "Two Kinds." Kirszner, Laurie G. and Stephen R. Mandell. Literature: Reading, Reacting, Writing, Compact Seventh Edition. Ed. Karen Mauk. 7th. Upper Saddle River, New Jersey: Michael Rosenberg, 2009. 694-702. Williams, Tennessee. "The Glass Menagerie." Kirszner, Laurie G. and Stephen R. Mandell. Literature: Reading, Reacting, Writing, Compact Seventh Edition. Ed. Karen Mauk. 7th. Upper Saddle River, New Jersey: Michael Rosenberg, 2009. 1628-1677.

Brave New World Vs Reality Essay -- essays research papers fc

Brave New World vs. Reality In many cases when you read a novel you may find comparisons between the "fictional" society and your realistic one. The author may consciously or unconsciously create similarities between these two worlds. The novelist can foresee the future and write according to this vision. In Brave New World, Adlous Huxley envisions the future of our society and the dangerous direction it is headed in. Brave New World is greatly dependant upon soma, as in our world where prescribed drugs and drug abuse are prominent. This is evident when Bernard and Lenina return from the Savage Reservation. Lenina is devastated from her experiences, so decides to take soma. It illustrates how like our world when something upsets us instead of trying to solve the problem we use drugs to mask them. Linda's addiction to soma is also an illustration of the similarities of drug abuse between our two worlds. Linda's return to Brave New World after many years brings her to the abuse of soma. She uses it as an escape from reality. Some of us use drugs to escape from the harshness and the tough brutality of reality. We always dream of the perfect utopia and expect our world to transform into it. Some of us always look for the easy way out and drugs allow us that. A further similarity of Brave New World to us, is when John is in the hospital after his mother's death due to soma abuse, and witnesses the workers receiving their soma rations. John begins to throw the soma out ...

Hopes And Dreams :: essays research papers

Hopes and Dreams Science has been the backbone of my life ever since my high school days in Malaysia. Although I studied many science subjects like Physics and Biology, I was especially fond of chemistry. My fondness of chemistry was attributed to my chemistry teacher Mr. Ang, in the eleventh grade who motivated me to explore every topic in great depth. He always gave me great advice, and I was greatly influenced by his spirit and values. Before I applied to universities in Malaysia, I attended various seminars which gave advice to students who were applying to universities and colleges. It is here that I got the most important advice of all, "Chose a major in a field that interests you." It did not take long before I was accepted to the xxxx University, and I chose Chemical Engineering as my field of study. Unfortunately, I had to deny the offer since my family had decided to move to the United States of America. Moving to the States has not been easy. I have not only encountered cultural differences, but a lot of difference in the educational system as well. But my goal to major in Chemistry still remains unchanged, and my options have widened. Last year, I learned about the Center for Science Excellence in one of my Chemistry classes, but I wasn’t aware of the many advantages and rewards of being part of such a remarkable program. As a newcomer, I have made numerous mistakes in choosing my courses and I need a lot of good advice and guidance. The Center for Science Excellence is exactly the type of program that can fulfill these needs, and help to put me on the right track instead me personally learning from costly mistakes. CSE is center which guides many other science students. It would be very helpful to make friends with other science students with whom I can exchange knowledge, ideas, and friendship. The internships that CSE provides to its students would help me gain experience and exposure to various areas in science which interest me, and hopefully guide me to the job of my dream.

Freudian approaches to Tennessee Williams’ Cat on a Hot Tin Roof Essay

I will discuss how Freudian theory ties in with the issues featuring in Cat on a Hot Tin Roof. Sigmund Freud the founder of psychoanalysis came up with a theory of the unconscious mind, he identified that sexual desires are the primary motivational energy of human life. William shows this in play by Margaret’s sexual desires for Brick keeps her energy up and gets her to keep trying to get Brick’s affection. Big Daddy and Big Mama have fears and anxieties about Margaret and Bricks lack of children, and so does Margaret herself because she wants a share of the plantation. However this does not worry Mae and Gooper who are more than happy that they are prime candidates to become the owner of the plantation, Mae especially enjoy sneer at the fact that Margaret is childless â€Å"Mae: she’s childless because that big beautiful athlete of hers won’t go to bed with her† (97). The family especially Big Daddy and Maggie have doubts about Brick’s sexuality and try to confront him about it. Brick: â€Å"You think that me and Skipper were a pair of dirty old men?† (pg 77). Symbolism definitely shown in ‘Cat on a Hot Tin Roof’, Bricks crutch can symbolise that he needs support, â€Å"Brick: no just give me my crutch† (27). The crutch can be seen as symbol that Brick is weak without skipper hence needs support. But it can also represent the fact that Brick has lost his masculinity ‘Brick: I’m getting softer Maggie’ (pg 31) and the crutch is his masculinity hence why he cannot go anywhere without it. The alcohol symbolises that Brick is trying to escape reality by making himself unconscious or in his words the ‘click’ which he says to be when he is peaceful with himself. However the excessive drinking by Brick can also mean that Brick cannot cope without Skipper. Repression of homosexuality in Cat on a Hot Tin Roof ties in with Sigmund Freud ideology, as Brick represses his feeling towards Skipper Brick: â€Å"I had friendship with Skipper – you are naming it dirty!† (pg 42).Repression is a defence mechanism and there are two stages, firstly the desires are pushed out from conscious into unconscious but this does not work for Brick as his feeling still surface. So the second line of defence is started this is where Brick is at in the book. Williams shows Brick as being reluctant to come out about his sexuality because he is brought up in an upper class family. One type of second line defence is overcompensation which is when a weakness is covered up by overcompensating on something else, in Bricks case is excessive sport. Then comes projection which is when you blame someone else; Brick blames â€Å"Maggie(79)† because she got â€Å"the idea (79)† that Brick and Skipper relationship was abnormal and that one reason why Skipper committed suicide. Then comes denial when protecting you from unpleasant reality which is seen in Brick as he tries to deny the truth when Big Daddy talks about it. Brick â€Å"YOU THINK SO TOO? You think me an skipper did sodomy†(77). Brick tries to justify it was Maggie who put dirty things into Skipper minding doubting himself about his sexuality, Brick trying to repress any arguments down to Margaret being jealous. Brick comes from an upper-class family who are highly respected from other people. The time when the story was set (1955) was a time when homosexuals were looked down upon and brought shame to their family. Brick is from a rich family, so he hides his sexuality so not to embarrass himself or his family, and marries Maggie to show that he is a man and tries to show his masculinity by being an athlete. Death features massively in the play, especially for Big Daddy who has cancer yet has been lied to by the family saying that he has a spastic colon. Big Daddy is also in denial because he himself knows everything is not alright because of the pain yet is putting a front on. Big Daddy and Big Mama have been told that he does not have cancer when he does and the only people that know are Mae and Gooper and Brick and Maggie. Big Mama is told her husband still has cancer later, but she says â€Å"it’s all a mistake† (93). Skipper is dead which makes Brick angry and blames his wife for the death. Margaret say is begging for Brick attention when she says â€Å"Skipper is dead! I am alive!†, but Brick does not want to believe that Skipper is dead â€Å"I tried to kill your Aunt Maggie, but I failed.† Brick definitely does not love Maggie and is shown in different way for example[She kisses Brick on the mouth which he immediately wipes with the back of his hand, (55pg).Bricks sexual desire keep surfacing no matter how much he is trying to suppress, so he tries to get his manhood back by jumping hurdles at the school athletic field but ends up fracturing his ankle. It shows that Brick is weak and helpless without Skipper. But also can be also interpreted to show Freud theory of sexual desires becomes energy for human life. Freud theory can relate to Cat on a Hot Tin Roof because unconscious and sexual desire relate to the character Brick who is trying to suppress his abnormal feeling from conscious to unconscious. In addition there is fear of death by Big Daddy and there are sexual desires.

Local national and International Economic Conditions Affecting the Industry

The Saudi Arabia royal embassy in the Washington dc is highly likely to be affected by a number of different national and international impacts and changing conditions. Most importantly, it should be remembered that the Saudi Arabia tends to be one of the most important oil exporter to the US. As a result relations and trade between the two countries due to the increasing amount of oil trade often tends to be peaceful and cordial if not always warm. It should be noted that when locally speaking, the territory of Saudi Arabia is owned and controlled by the Arabs in general and by the royal kingdom comprising of the kings in particular. In other words, in contrast to a democratic leadership which is most often followed, Saudi Arabia follows its monarchy. This may mean that the two countries have, politically speaking very little in common. This may mean that the background setting in which the relations have to built up between the two nations may be somewhat problematic. Apart form this, it should also not be forgotten that in a number of cases, Saudi Arabia is mostly comprised of Arab population. Moreover, most of the Arabic population of Saudi Arabia tends to be Muslims. In such a case the strain and coldness in the relations may often be even more than the usual. Apart form this, the fact that Saudi Arabia uses the sharia form of government also implies that unlike the US and the other western nations, the similarities and the relations that the two countries may have are more distanced then assumed to be. Therefore, in these circumstances, Saudi Arabia and US relations may be difficult to manage out properly. The number of embassies formed, including the Saudi Arabian embassy formed at Washington plays a phenomenal role in improving the relations the two countries have. It should be remembered that relations between countries are most often based upon the common concerns of the two countries. It should be remembered that the US and Saudi Arabia share a common concern regarding the oil exports and imports, regional security and sustainable development. These facts lead to a need for there to being an embassy and were one of the reasons which may have initially leaded to the development of the royal embassy at Washington dc. Moreover, the Middle East peace process and the gulf issue also imply that the international relations improved and consequently had a positive impact upon the embassy of Saudi Arabia. The embassy as also affected by a number of international events. One of such events happens to be the increasing amount of post 9/11 and including the 9/11 attacks on the US. all these events and following the fact that most of the individuals involved in the terrorists attacks were Muslims and Saudi Arabians meant that the Saudi Arabian embassy has to face a great deal of criticism from the public and also to some extent from the US government. Not only this, but the Saudi Arabian legislations and the system of law also imply that the Saudi embassy in the US has to have an increasing amount of troubles. One of the examples of such an event tends to be the one which occurred on 26th June, 2010. This was when a Canadian woman was not allowed to leave Saudi Arabia due to the strict Saudi law which implied that no women can leave or travel anywhere without the authorization of a male guardian. Since, the husband of the Canadian woman, named Nathalie was not willing to give the required authorization, her condition after being help as a hostage continued deteriorating. Following this event there was a great deal of protest and chaos by the local citizens of the US outside the US embassy. Therefore, it can be seen from these instances as to how the embassy of the Saudi Arabian government has to deal with the variety of changing circumstances and conditions due to events at both local, national as well as on international levels. Factors affecting the increase in demand and supply analysis along with the cost structure It should not be forgotten that just like many other industries, even the Saudi Arabian embassy set up at the US tends to be an industry set up as a result of there being an increase in the demand for it. More recently, the increase in the demand for the embassy in particular has been so due to the desires for an increasing amount of investments which the members of the Saudi Arabia wish to do so in the US. Consequently, the embassy actively seeks to allow for conditions which could allow for the demand of the Saudi Arabians to wish to invest into the US to be satisfied easily. Thus, the Saudi embassy sets to provide for a means by which investors wishing to invest in the US should do some with a greater ease. However, there needs to be a clearer analysis to identify the reason why there has been such an increase in the demand for investing in the US. Perhaps one of the most important reasons why this has occurred has been due to the increasing population of Saudi Arabia. This has allowed for a greater deal of prospects which were previously not looked upon to, to being identified. Another reason why there has been such an increase is also due to the government policies and regulations. Now and then, the Saudi government has been encouraging a number of individuals from all over Saudi Arabia to come and invest in the agencies and thereby maximize as much of the profits as it is possible for them to do so. Another important reason for such an increasing demand for the embassy also tends to be due to the growth in the member of individuals from Saudi Arabia wishing to visit the US. These visits vary in their types, ranging from being pure business to being for tourism perhaps for the reason of emigration. Thus, these embassies serve to help the individuals who wish to visit or are visiting the US with their travel. Apart from this, there has also been an increase in the demand for the Saudi embassy due to the need of the locals of the US to visit Saudi Arabia. These visits may be again for business and may also b for tourism. Perhaps one of the most important reasons why people from the US may visit Saudi Arabia also tends to be for the religious purposes of pilgrimage which all the Muslims may wish o perform for once in their entire life. These factors may therefore lead to a greater demand for the Saudi embassy in Washington DC. When looking at the supply analysis of the industry it is important to note that the supply for the service is primarily dependant upon the state policies of both the US as well as that of Saudi Arabia. Therefore, provided there being good trading conditions and a belief of having strong and cordial relations probably implies an increase in the industry service. On the other hand if there are better conditions then the embassy may experience it being un able to provide for the required services. These factors along with the official budget to be allocated on foreign relations and the economic stability thereby, affects and impacts the cost structure of the state as well. Key embassies in the industry There are numerous embassies set all around the world in order to promote diplomatic relations between the various countries where they have been set up. it is important to note that their various locations and set up have an important impact on the type of relations that have been or are intended for the country. One of the key embassies of the world is the US embassy set up in Iraq. This embassy is the largest and perhaps also one of the most expensive embassies of the world. The location of the embassy has been set up at the green zone. Note that the green zone in Iraq is the most important and commercial area which had and still continues to be the most important of all the zones in Iraq. Note that the way the embassy has been set up reflect the increasing amount of power and domination which the US wants and has portrayed on Iraq. t should also be seen that the fact that there are so many facilities and the fact that the embassy is so large has been indented to do so to show how important and powerful the US continues to be in Iraq. Another such important agency is the Israeli agency set up is the Israeli agency in the US at the Washington DC. The agency has been built up to strengthen the cordial relations between the US and Israel. This perhaps tends to be one of the most important reas ons why the agency has been taken so much care of and has been provided with the utmost level of security. Looking at the se two embassies we can make an analysis in relation to their strengths and weaknesses. It is important to note hat when talking of the US embassy that has been set up in Iraq, the embassy tends to be fully guarded and protected. Note that the Israeli embassy in the US has the same strength. This implies that the importance that the US embassy gets is one of its important strength. Apart from this, however, the type of relations that the government tends to have with the embassy is also an important factor that can be both the strength and the weakness of the embassy. Cultural and ethical dimensions There are an increasing amount of cultural and ethical issues that these embassies have and run according to. In particular when speaking of the ethical and the cultural dimensions of the industry of embassies should be noted that the basic policies concerning g the human rights for most of the industry remain the same. Thus, the industry as a whole may be favoring the basic human right of there being the freedom f speech and the need for an individual to live freely without being harmed by anyone. Apart from this, the charges against harming others for instance theft, robbery, kidnapping or the murder of others is also the same in most of the embassies held all over the world. Apart from this, a number of other issues for instance the use of hard drugs or alcohol may also tend to be more or less the same. There may be a slight deal of variation however depending upon the type of country and its local population and policies. For instance, many Muslim countries may put a ban or may discourage alcohol use. Part from these policies one can also look at other ethical issues like those of abortion, prostitution, euthanasia as well as that of same sex marriages. In all these cases, the type of country and the population may determine weather or not such practices are allowed or not. For instance, in Canada same sex marriages have been allowed in recently perhaps due to the local policies and needs. Another important issue to look upon is the cultural dimension. In this case again it should be noted that each of the embassies take and make a great deal of effort to promote the local culture as much as possible. However, there may be cases when there is in fact no culture at all like it does in the case of the US whose culture is nothing but a melting pot of a variety of cultures. Constraints/threats The Saudi embassy which has been set up in the US has been facing an increasing amount of constraints and threats. Most of these arise out as a result of the changing relations and policies of the US as well as that of the Saudi Arabia. In this case, note that as it had been highlighted before on, these changing relations are a major indicator of where the embassy will head to. For instance, post 9/11 events implied a major threat to the embassy. However, as soon as the government of the Saudi Arabia condemned these acts of terrorism the constraints on the industry loosened. Therefore, the changing relations of the two countries have a great deal of impact on the embassy. Another important constraint that the embassy has is also that of the changing oil prices and the relative impact that the US-Saudi relations tend to have. Therefore, it should be noted that the changing economic and political setup have a great deal of impact on the embassy and acts as a major treat or constraint to it Opportunities Looking at the opportunities, it should be noted that the major trade and the increasing demand for oil provides for a new dimension to the embassy and acts as a growing opportunity. apart from this, the increasing level o investments that have been carried out and are likely to increase in the future by the Saudis to the US and vice versa acts as another important means of growing opportunity and prospects for the industry. Effectiveness it should be noted that even after an increasing deal of problems and threats, the fact that the relationship between the US and the Saudi government have still managed to hold on and manage cordial relationships with a great deal of diplomacy is an important factor which shows how effectively the Saudi government and the agency has managed to deal with most of it's problems. Embassies culture and ethics It is important to note that the Saudi embassy traces its culture and ethics form the cultural and the ethical heritage of Islam. In particular looking at the many qualities which individuals at the time of the advent of Islam had, it can be seen how the Saudi embassy seeks to promote those cultural and ethical dimensions. These include the muslin qualities of bravery, hospitality, respect for the elders and humility. In addition to this, Saudi Arabia seeks to promote peace and serenity just like the way it had been taught to them by Islam. Apart form this, when looking at the cultural aspects, trade form an important part of their culture and the embassy of the Saudi Arabia seek to promote trade on an extensive way all around the world.