PubMed Health⌕ Search

Biomedical subjects

Tim Churches

Publications and source records attributed to Tim Churches.

8 recordsLinked to original sources
PubMed2005

Effect of a chemical manufacturing plant on community cancer rates.

BACKGROUND: We conducted a retrospective study to determine if potential past exposure to dioxin had resulted in increased incidence of cancer in people living near a former manufacturing plant in New South Wales, Australia. During operation, from 1928 to 1970, by-products of the manufacturing process, including dioxin and other chemical waste, were dumped into wetlands and mangroves, discharged into a nearby bay and used to reclaim land along the foreshore, leaving a legacy of significant dioxin contamination. METHODS: We selected 20 Census Collector Districts within 1.5 kilometres of the former manufacturing plant as the study area. We obtained data on all cases of cancer and deaths from cancer in New South Wales from 1972 to 2001. We also compared rates for some cancer types that have been associated with dioxin exposure. Based on a person's residential address at time of cancer diagnosis, or at time of death due to cancer, various geo-coding software and processes were used to determine which collector district the case or death should be attributed to. Age and sex specific population data were used to calculate standardised incidence ratios and standardised mortality ratios, to compare the study area to two comparison areas, using indirect standardisation. RESULTS: During the 30-year study period 1,106 cases of cancer and 524 deaths due to cancer were identified in the study area. This corresponds to an age-sex standardised rate of 3.2 cases per 1,000 person-years exposed and 1.6 deaths per 1,000 person-years exposed. The study area had a lower rate of cancer and deaths from cancer than the comparison areas. The case incidence and mortality due to lung and bronchus carcinomas and haematopoietic cancers did not differ significantly from the comparison areas for the study period. There was no obvious geographical trend in ratios when comparing individual collector districts to New South Wales according to distance from the potential source of dioxin exposure. CONCLUSION: This investigation found no evidence that dioxin contamination from this site resulted in increased cancer rates in the potentially exposed population living around the former manufacturing plant.

Adolescent↗
PubMed2004

Some methods for blindfolded record linkage.

BACKGROUND: The linkage of records which refer to the same entity in separate data collections is a common requirement in public health and biomedical research. Traditionally, record linkage techniques have required that all the identifying data in which links are sought be revealed to at least one party, often a third party. This necessarily invades personal privacy and requires complete trust in the intentions of that party and their ability to maintain security and confidentiality. Dusserre, Quantin, Bouzelat and colleagues have demonstrated that it is possible to use secure one-way hash transformations to carry out follow-up epidemiological studies without any party having to reveal identifying information about any of the subjects - a technique which we refer to as "blindfolded record linkage". A limitation of their method is that only exact comparisons of values are possible, although phonetic encoding of names and other strings can be used to allow for some types of typographical variation and data errors. METHODS: A method is described which permits the calculation of a general similarity measure, the n-gram score, without having to reveal the data being compared, albeit at some cost in computation and data communication. This method can be combined with public key cryptography and automatic estimation of linkage model parameters to create an overall system for blindfolded record linkage. RESULTS: The system described offers good protection against misdeeds or security failures by any one party, but remains vulnerable to collusion between or simultaneous compromise of two or more parties involved in the linkage operation. In order to reduce the likelihood of this, the use of last-minute allocation of tasks to substitutable servers is proposed. Proof-of-concept computer programmes written in the Python programming language are provided to illustrate the similarity comparison protocol. CONCLUSION: Although the protocols described in this paper are not unconditionally secure, they do suggest the feasibility, with the aid of modern cryptographic techniques and high speed communication networks, of a general purpose probabilistic record linkage system which permits record linkage studies to be carried out with negligible risk of invasion of personal privacy.

Computer Communication Networks↗
PubMed2003

A proposed architecture and method of operation for improving the protection of privacy and confidentiality in disease registers.

BACKGROUND: Disease registers aim to collect information about all instances of a disease or condition in a defined population of individuals. Traditionally methods of operating disease registers have required that notifications of cases be identified by unique identifiers such as social security number or national identification number, or by ensembles of non-unique identifying data items, such as name, sex and date of birth. However, growing concern over the privacy and confidentiality aspects of disease registers may hinder their future operation. Technical solutions to these legitimate concerns are needed. DISCUSSION: An alternative method of operation is proposed which involves splitting the personal identifiers from the medical details at the source of notification, and separately encrypting each part using asymmetrical (public key) cryptographic methods. The identifying information is sent to a single Population Register, and the medical details to the relevant disease register. The Population Register uses probabilistic record linkage to assign a unique personal identification (UPI) number to each person notified to it, although not necessarily everyone in the entire population. This UPI is shared only with a single trusted third party whose sole function is to translate between this UPI and separate series of personal identification numbers which are specific to each disease register. SUMMARY: The system proposed would significantly improve the protection of privacy and confidentiality, while still allowing the efficient linkage of records between disease registers, under the control and supervision of the trusted third party and independent ethics committees. The proposed architecture could accommodate genetic databases and tissue banks as well as a wide range of other health and social data collections. It is important that proposals such as this are subject to widespread scrutiny by information security experts, researchers and interested members of the general public, alike.

Computer Security↗
PubMed2003

Exploratory data analysis using set operations and ordinal mapping.

Exploratory data analysis requires the ability to issue ad hoc queries to filter and summarise data sets. As the sizes of health data sets grow, traditional methods of processing data have difficulty in providing acceptable response times for such queries. An alternative method is described which combines complete vertical partitioning of data with set operations on ordinal mappings (SOOM). An initial implementation of the technique provides significantly better performance than a conventional SQL database on typical exploratory data analysis queries. The use of parallel, distributed computation to further increase the performance of the technique appears to be feasible.

Algorithms↗
PubMed2002

Preparation of name and address data for record linkage using hidden Markov models.

BACKGROUND: Record linkage refers to the process of joining records that relate to the same entity or event in one or more data collections. In the absence of a shared, unique key, record linkage involves the comparison of ensembles of partially-identifying, non-unique data items between pairs of records. Data items with variable formats, such as names and addresses, need to be transformed and normalised in order to validly carry out these comparisons. Traditionally, deterministic rule-based data processing systems have been used to carry out this pre-processing, which is commonly referred to as "standardisation". This paper describes an alternative approach to standardisation, using a combination of lexicon-based tokenisation and probabilistic hidden Markov models (HMMs). METHODS: HMMs were trained to standardise typical Australian name and address data drawn from a range of health data collections. The accuracy of the results was compared to that produced by rule-based systems. RESULTS: Training of HMMs was found to be quick and did not require any specialised skills. For addresses, HMMs produced equal or better standardisation accuracy than a widely-used rule-based system. However, accuracy was worse when used with simpler name data. Possible reasons for this poorer performance are discussed. CONCLUSION: Lexicon-based tokenisation and HMMs provide a viable and effort-effective alternative to rule-based systems for pre-processing more complex variably formatted data such as addresses. Further work is required to improve the performance of this approach with simpler data such as names. Software which implements the methods described in this paper is freely available under an open source license for other researchers to use and improve.

Data Collection↗