1. |
Methods and Preliminary Findings in Assessing the Economic and Health Services Consequences of Smoking, with Particular Reference to Lung Cancer |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 297-312
A. B. Atkinson,
T. W. Meade,
Preview
|
PDF (1031KB)
|
|
摘要:
SummaryA framework for the economic and health services consequences of smoking is proposed, which includes an itemization of specific elements concerned, and a discussion of different views (based on levels of awareness of health risks and on the addictive nature of smoking) of the consequences of the habit. Detailed consideration is given to the medical care costs of lung cancer within a model which is being constructed to relate government intervention(viataxation on tobacco, etc.) to changes in tobacco consumption, and thus to changes in mortality, morbidity and the need for medical care in lung cancer. Sources are considered from which data may be obtained on hospital and community resources devoted to lung cancer; data from cancer registration are especially valuable.
ISSN:0959-5341
DOI:10.2307/2344952
出版商:Wiley
年代:2018
数据来源: WILEY
|
2. |
The Classification of Smoking by Factorial Structure of Motives |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 313-333
M. A. H. Russell,
J. Peto,
U. A. Patel,
Preview
|
PDF (1203KB)
|
|
摘要:
SummaryUsing a 34‐item self‐completion questionnaire relating to a wide variety of smoking motive themes gleaned from previously published work, responses of 175 normal smokers were subjected to factor analysis. Six oblique factors were obtained representing the following six types of smoking: psychosocial, indulgent, sensorimotor, stimulation, addictive and automatic. Previous work suggesting a sedative type of smoking was not confirmed. The most striking finding was a major “pharmacological addiction” dimension which completely separated the stimulation, automatic and addictive factors and their items from the rest. It was these three factors which were correlated (·50, ·56, ·63 respectively) with cigarette consumption and which differentiated the sample of normal smokers from a criterion sample of 103 addicted heavy smokers attending smoking withdrawal clinics. It is suggested that it may prove more useful to classify smokers according to their position on the single dimension of pharmacological addiction to nicotine rather than in terms of their profiles on the six types of smoking. The interpretation and limitations of factor analysis of questionnaire‐type data are discussed.
ISSN:0959-5341
DOI:10.2307/2344953
出版商:Wiley
年代:2018
数据来源: WILEY
|
3. |
General Discussion |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 333-346
Preview
|
PDF (1039KB)
|
|
ISSN:0959-5341
DOI:10.1111/j.2397-2327.1974.tb03385.x
出版商:Wiley
年代:2018
数据来源: WILEY
|
4. |
The Distribution of Doctor‐Patient Contacts in the National Health Service |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 347-371
J. R. Ashford,
R. G. Hunt,
Preview
|
PDF (1531KB)
|
|
摘要:
SummaryThe confrontation between doctor and patient may be regarded as the basic unit of medical practice in any comprehensive health care system. This paper is concerned with the marginal and joint distributions of the numbers of doctor–patient contacts in the National Health Service. The general structure of this system of care and the range of services available to the patient are described and it is emphasized that the processes which determine the pattern of contacts by a particular patient are complex and for the most part poorly understood.A simple but general model to represent the joint distribution of patient contacts is derived. This model is based upon the concept that the individual patient is subject to episodes of ill health which start at defined points in time and generate varying numbers of contacts of different types. It is assumed that patients vary in their general health and that susceptibility to illness follows a characteristic distribution within a population of patients. A more specific class of models is then considered. These models embody the assumption that, for a particular patient, episodes are events in a Poisson process, the rate parameter varying within the population according to some specified distribution. It is shown that the joint probability generating function (p.g.f.) of the total number of contacts in a given period of time may be expressed in a simple form, in terms of the Laplace transform of the probability density function of the distribution of the Poisson rate parameter and the joint p.g.f. of the numbers of contacts generated by a particular episode. Further and more specific assumptions are then considered and a model for the joint p.g.f. of the numbers of contacts of a given type, in which the rate parameter of the Poisson process of episodes follows a Gamma distribution and the numbers of contacts generated by particular episodes are independent random variables following modified logarithmic distributions is described. The application of this model to the practical situation, in which it is not possible to differentiate reliably between contacts generated by specific episodes, is then considered and a set of recurrence formulae for the calculation of the probabilities involved in the maximum likelihood estimation process is derived.The Poisson/Gamma/modified log distribution model is then applied to empirical data concerning the joint distribution of general practice surgery attendances, general practice home visits and hospital out‐patient consultations arising from a general population of some 70,000 persons living in the City of Exeter over a period of one calendar year. The basic data for males and females are subdivided separately in terms of 5‐year age‐groups. With the exception of the very young and the very old, the model provides a good fit to the data for males and, when contacts involving pregnancy are change in the accessibility or in the delivery of care can result in very substantial changes in the overall usage. In this way, it is possible to assess the “tip‐of‐the‐iceberg” effect which is believed to apply to many aspects of the delivery of medical care.
ISSN:0959-5341
DOI:10.2307/2344954
出版商:Wiley
年代:2018
数据来源: WILEY
|
5. |
Discussion on the Paper Professor Ashford and Dr Hunt |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 371-383
Preview
|
PDF (1060KB)
|
|
ISSN:0959-5341
DOI:10.1111/j.2397-2327.1974.tb03387.x
出版商:Wiley
年代:2018
数据来源: WILEY
|
6. |
Voting Research and the Population Census 1918–71: Surrogate Data for Constituency Analyses |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 384-411
W. L. Miller,
Gillian Raab,
K. Britto,
Preview
|
PDF (1468KB)
|
|
摘要:
SummaryThe formation of political partisanship is a social and historical process, not only, or even mainly, an individual response to personal social or economic circumstance. Consequently we need to examine environmental influences explicitly and test for them empirically, before inferring a socio‐political relationship at one level of aggregation of data from a relationship observed at another.The incompatibility of census areas with parliamentary areas prevents constituency level analysis before 1966, and the revision of parliamentary constituency boundaries three times between 1918 and 1970 also causes difficulties. By working with borough level units, census and voting data can be found for 173 units and these units remain unchanged throughout 1918‐70.It is important to distinguish two types of environmental characteristics that can influence individual partisanship: those local characteristics which are highly correlated with the spatial clustering of significant individual characteristics, and those local characteristics that are not. The first type biases regression slopes, the second affects only the variance of estimates of regression slopes.We have shown that in predicting the individual's partisanship from occupational class, the class of the individual and of his immediate neighbourhood—an area equal to or smaller than the constituency—are both important. There are also strong local influences operating at the wider levels of the borough, county or region but these influences are not correlated with the area's occupational class. Consequently it is possible to infer the results of constituency level regressions from regressions with borough data. We have given two different methods of doing this, each with advantages in different situations.We have thus developed and tested methods of estimating the results of constituency level socio‐political analyses throughout the entire 1918‐70 period using borough data. For completeness we have also considered the use of county and regional data for such a purpose. County and regional level data are much less useful than borough level data.
ISSN:0959-5341
DOI:10.2307/2344955
出版商:Wiley
年代:2018
数据来源: WILEY
|
7. |
Joint Meeting of Royal Statistical Society and Institute of Statisticians |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 412-427
Preview
|
PDF (1171KB)
|
|
ISSN:0959-5341
DOI:10.1111/j.2397-2327.1974.tb03389.x
出版商:Wiley
年代:2018
数据来源: WILEY
|
8. |
Comparative Statistical Inference |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 428-429
D. M. Grove,
Preview
|
PDF (160KB)
|
|
摘要:
1. Comparative Statistical Inference.By Vic Barnett. London and New York, Wiley, 1973. xv, 287 p. 9″. £5·25.
ISSN:0959-5341
DOI:10.2307/2344957
出版商:Wiley
年代:2018
数据来源: WILEY
|
9. |
Time Series |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 430-430
P. R. Fisk,
Preview
|
PDF (88KB)
|
|
摘要:
2. Time Series.By M. G. Kendall. London, Griffin, 1973. ix, 197 p.912″. £3·60.
ISSN:0959-5341
DOI:10.2307/2344958
出版商:Wiley
年代:2018
数据来源: WILEY
|
10. |
Stochastic Models for Social Processes |
|
Journal of the Statistical Society of London,
Volume 137,
Issue 3,
2018,
Page 431-431
S. Vajda,
Preview
|
PDF (82KB)
|
|
摘要:
3. Stochastic Models for Social Processes.By D. J. Bartholomew. 2nd edition. London, Wiley, 1973. xi, 411 p. £6·95.
ISSN:0959-5341
DOI:10.2307/2344959
出版商:Wiley
年代:2018
数据来源: WILEY
|