NHS Digital Data Release Register - reformatted

Health & Safety Laboratory (hsl)

Project 1 — DARS-NIC-351522-Y6W3L

Opt outs honoured: Y

Sensitive: Non Sensitive, and Sensitive

When: 2016/12 — 2018/02.

Repeats: One-Off, Ongoing

Legal basis: Section 251 approval is in place for the flow of identifiable data, Approved researcher accreditation under section 39(4)(i) and 39(5) of the Statistical Registration Service Act 2007

Categories: Anonymised - ICO code compliant, Identifiable

Datasets:

  • Hospital Episode Statistics Admitted Patient Care
  • MRIS - Cause of Death Report
  • MRIS - Cohort Event Notification Report

Benefits:

The Pesticide Users’ Health Study is a large study of individuals potentially exposed to low levels of pesticides over a long period. As such, it is a valuable resource that can make a worthwhile contribution to the wider literature on pesticides and health, and help to elucidate some of the current inconsistencies in the literature. The study and its results has already paved the way for a new separate cohort study, which will collect more detailed information but on a smaller number of pesticide users. The Pesticide Users’ Health Study forms an integral part of the Health & Safety Executive’s commitment to protecting the health of people at work. Hence, the results of this study, and a greater understanding of the potential risks involved in pesticide use, could help to inform future policy. In addition, all of the results will be publically available, which will raise the awareness of the potential effects of working with pesticides among the public, workers, and bystanders, and encourage care and vigilance in the way that pesticides are used. All of this will help to protect the health of users of pesticides.

Outputs:

Results of all analyses will be published in Health & Safety Executive Research Reports (http://www.hse.gov.uk/research/rrhtm/), peer-reviewed journal publications, and presented at conferences where appropriate. The most appropriate journals will be decided upon at the time of publication. Examples of journals this study and other similar studies have published in previously include Occupational Medicine, British Journal of Cancer, and Annals of Occupational Hygiene. As with journals, the most appropriate conference will be decided upon at the time. Examples of conferences this study and other similar studies have presented at previously include The International Epidemiology in Occupational Health (EPICOH) Conference, The UK & Ireland Occupational & Environmental Epidemiology Conference, and The British Occupational Hygiene Society Annual Conference. The results will take the form of aggregated data, and will typically include summary statistics, and standardized mortality/incidence/admission ratios or relative risks. Individuals will not be identifiable in these results and if there are small numbers involved in any aggregated data, these will be suppressed in accordance with current guidelines. All outputs will be aggregated with small numbers suppressed and will follow the HES analysis guide. The data will not be used for commercial purposes. Objective 1: Ongoing cancers, deaths, and emigrations Previous analysis of mortality and cancer incidence included follow-up to 2005, and so it is expected that an updated analysis will be completed shortly using the additional 10 years’ follow-up data. This is an on-going study, and updated analyses will be undertaken periodically after this. In addition to these analyses, the Health and Safety Executive may request additional analyses of the data to address specific questions. Objective 2: One-off linkage and tabulation of hospital admissions This analysis is expected to be completed and the results published by April 2018.

Processing:

The data will be processed, only by substantive employees, at the Health & Safety Laboratory. All data transfers between NHS Digital and HSL will be undertaken using NHS Digitals secure file transfer system. Any information received from NHS Digital will be downloaded directly onto a restricted access network drive, dedicated to the study. The data will not be accessible by a third party organization. The data will only be used for the purpose as stated. Objective 1: Ongoing cancers, deaths, and emigrations Linkage has already been undertaken by NHS Digital, with updates on cancer and death registrations, and embarkations and returns previously received on a quarterly basis. Hence, there will be no information transferred from the Health & Safety Laboratory to NHS Digital. It is hoped that NHS Digital will continue to send the event notification file, the death notification file, on a quarterly basis using their secure file transfer system. This will be downloaded directly onto a restricted access network drive, dedicated to this study. Identifiable data is requested for data quality purposes, analysis however, is conducted on data with identifiers removed. To process the data ready for analysis and reporting in the specified outputs, the data are extracted from the database in three csv files: one file contains details on the participants and includes NHS numbers and dates/causes of death (if relevant); one contains details on the certifications the individuals have undertaken; and one on cancer registrations that have been entered onto the database. These files are saved on the restricted access network drive, which is dedicated to this study. Linkage of the different files is through the unique study ID number. Stata statistical software is then used to link the files and de-identify the data as much as possible before analysis. Information such as name, address and NHS number is removed. Information such as date of death, cause of death, date of cancer registration, sex, postcode and study ID number remain for the analysis. Only the de-identified dataset is used during analysis. Analysis is conducted using Stata statistical software, and all files relating to the analysis are saved on the restricted access network drive for the study. Objective 2: One-off linkage and tabulation of hospital admissions For the linked episode-level information: Linkage will be undertaken by NHS Digital. They will then provide the HSL study team with the linked hospital episode data requested, but pseudonymised with the study ID. The analysis that HSL plans to undertake will require linking the episode level data to the following information on study participants (held by HSL and linked, by HSL, through the study ID): 1) Baseline information: date of birth, sex, address, pesticide certificates held. Collected at the time of recruitment. 2) Date of death or emigration (if applicable). 3) Information on pesticide use, collected from a subset of participants. Stata statistical software will be used to link the files and pseudonymise the data as much as possible before analysis. Information such as name, address and NHS number will be removed. Information such as date of death, date of emigration, sex, postcode and study ID number will remain for the analysis. All information requested on hospital admissions will be retained for the anlaysis. Only the pseudonymised dataset will be used during the analysis. The analysis will be conducted using Stata statistical software, and all files relating to the analysis will be saved in the restricted access network drive for the study.,

Objectives:

The Pesticide Users’ Health Study was established in the late 1990s, with the aim to monitor the long-term health of people who use pesticides as part of their work. Monitoring mortality and cancer incidence among study members is an important part of this. Mortality up until 2005 and cancer incidence up until 2004 have already been analysed and published in Health & Safety Executive Research Reports and a peer-reviewed publication (http://www.hsl.gov.uk/resources/major-projects/puhs). Additional follow-up will help to clarify some of the results observed, and will add to the literature on the potential link between pesticides and ill health. Cancer and death registrations do not necessarily capture conditions that do not tend to lead to death. Conditions such as neurological, eye, respiratory, and skin diseases have all been linked to potential pesticide exposure, but the evidence is inconclusive. These conditions will not necessarily be captured on death certificates, but individuals with these diseases could have been admitted to hospital. The specific objectives of the processing will be to; 1) analyse cancer incidence and mortality among members of the Pesticide Users’ Health Study, comparing these rates to the general population and investigating trends between groups where possible (for example, between regions and types of pesticide use). 2) Analyse hospital admissions due to neurological, eye, respiratory, or skin disease among members of the Pesticide Users’ Health Study, comparing these rates to the general population and investigating trends between groups where possible (for example, between regions and types of pesticide use). HSL are therefore requesting the following data to answer the two objectives above: 1) Objective 1: Individual-level data relating to cancers, deaths, and emigrations, linked to study members. In the past, this has been received from NHS Digital (and its predecessors) on a quarterly basis, and HSL wish this to continue into the foreseeable future. These data will be analysed on a regular basis. 2) Objective 2: Episode-level hospital admission data, linked to study members. This is a one-off request.