Statistical aspects of the analysis of data from retrospective studies of disease.

@article{Mantel1959StatisticalAO,
  title={Statistical aspects of the analysis of data from retrospective studies of disease.},
  author={Nathan Mantel and William M. Haenszel},
  journal={Journal of the National Cancer Institute},
  year={1959},
  volume={22 4},
  pages={
          719-48
        }
}
The role and limitations of retrospective investigations of factors possibly associated with the occurrence of a disease are discussed and their relationship to forward-type studies emphasized. Examples of situations in which misleading associations could arise through the use of inappropriate control groups are presented. The possibility of misleading associations may be minimized by controlling or matching on factors which could produce such associations; the statistical analysis will then be… 
View on PubMed
archive.neicon.ru
14,355 Citations
Highly Influential Citations
342
Background Citations
445
Methods Citations
4,665
Results Citations
35

Tables from this paper

14,355 Citations

Citation Type

Citation Type

Statistical analysis of stratified 2x2 tables
TLDR
Methods appropriate for cross-sectional, prospective, and retrospective studies using matched or unmatched samples were described and methods for determining the extent of association between a dichotomous disease state and a dichotomyous risk factor were described.
Some aspects of retrospective studies.
Synthetic retrospective studies and related topics.
TLDR
It is brought out that in a large prospective study in which comparatively few cases of disease have occurred, computational problems can be so burdensome as to preclude a comprehensive and imaginative analysis of the data.
Methodologic issues in hospital epidemiology. IV. Risk ratios, confounding, effect modification, and the analysis of multiple variables.
TLDR
The data are reanalyzed through the use of stratification by severity of underlying illness in order to demonstrate the impact of confounding and effect modification by a third, extraneous variable.
Combination of log relative risk in retrospective studies of disease.
  • P. Sheehe
  • Mathematics
    American journal of public health and the nation's health
  • 1966
TLDR
The combination of estimates of relative prevalence or incidence, commonly referred to as relative risk, obtained from retrospective 2 X 2 tables, is discussed, involving the iterative solution of sets of quadratic equations.
Epidemiological Statistics—II
TLDR
Epidemiological studies use statistical methods to understand the factors that cause, reduce, and prevent diseases in human populations, and can be qualitative or quantitative, descriptive or comparative, and observational or experimental.
Statistical analyses of the relative risk.
  • J. Gart
  • Mathematics
    Environmental health perspectives
  • 1979
TLDR
The relation between the relative risk and the odds ratio (or cross-product ratio) is developed and can be considered a parameter of an exponential model possessing sufficient statistics, which permits the development of exact significance tests and confidence intervals in the conditional space.
Use and misuse of statistics for epidemiological studies of multiple sclerosis
  • P. Hibberd
  • Medicine, Psychology
    Annals of neurology
  • 1994
TLDR
The difficulties inherent in the use of any type of descriptive study to investigate etiology of disease are reviewed and an update of appropriate statistical methodology to analyze study results is provided.
Problems in the assessment of relative risk of chronic disease among biological relatives of affected individuals.
Statistical analysis of K 2 x 2 tables: a comparative study of estimators/test statistics for association and homogeneity.
TLDR
The results of a Monte Carlo study was presented to determine the size and power of a number of tests of association and homogeneity when the data are sparse and the performance of three estimators of the common odds ratio.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 55 REFERENCES
A Statistical Problem Arising from Retrospective Studies
TLDR
In prospective studies of lung cancer an estimate of relative mortality is usually preferred to one of relative incidence or prevalence because (a) the death registration system provides a complete enumeration which is lacking for newly developed or for existing cases and (b) diagnosis of cause of death is usually more accurate.
A study of environmental factors in carcinoma of the cervix.
Study of the Aetiology of Carcinoma of the Lung
TLDR
It is concluded that smoking is a factor in the production of carcinoma of the lung, and this conclusion was in conformity with the results of some other investigations.
Cancer and tobacco smoking; a preliminary report.
TLDR
Reviews of the literature for the past twenty years reveals that it is often conflicting and that it consists for the most part of studies which are inconclusive because of lack of adequate samples, lack of random selection, no proper controls or failure to age-standardize the data.
Lung cancer in women; a study of environmental factors.
TLDR
The present study is based on environmental data of proved cases of lung cancer in women, all of which were diagnosed in the period 1953 to 1955, and the purpose has been to study etiologic factors in the development of lung cancers in women.
The association of smoking with cancer of the urinary bladder in humans.
TLDR
A comparison of the histories of tobacco use of a group of patients with urinary bladder cancer with those of Patients with other cancers and noncancerous conditions to determine if in humans a similar relationship exists.
Occurrence of pregnancy, abortion, and artificial menopause among women with coronary artery disease: a preliminary study.
Relationship Between Cancer of Stomach and the ABO Blood Groups
TLDR
It occurred to one of us that there might be a correlation between the high northern incidence of cancer of the stomach and the genetic differences which are reflected in the ABO blood groups, the frequency of blood group 0 being greater, that of A less, in northern than in southern populations, B and AB being similar in all geographical areas in England and Scotland.
Cancer of the Uterine Cervix and Social Conditions
  • P. Stocks
  • Medicine
    British Journal of Cancer
  • 1955
TLDR
Examination of social factors which might be responsible for the class gradient shown by cervical cancer death rates at ages before 65 found environmental influences seemed important particularly in textile towns were many women work in the mills.
Lung cancer and smoking.
...
1
2
3
4
5
...