This technical guidance note explains and illustrates the Information Commissioner's view of what is personal data for the purposes of the Data Protection Act 1998. It is designed to help data protection practitioners decide whether data falls within the definition of personal data in circumstances where this is not obvious.
We have been aware for some time of the need to replace our guidance on the implications of the Durant judgment. Inevitably that guidance reflected the fact that the Court of Appeal was widely understood to have adopted a rather narrower interpretation of "personal data" and "relevant filing system" than most practitioners and experts had followed previously. We recognised the need to produce guidance with a greater emphasis on what is covered than what is not. In June 2007 the Article 29 Working Party, an advisory committee composed of representatives of the national supervisory authorities, agreed an opinion on the "concept of personal data". Though our guidance is structured differently we are satisfied that it is consistent with the approach taken by the Working Party. Both the Opinion and our guidance make great use of practical examples to illustrate the key considerations when deciding what is personal data.
Our previous guidance covered the meaning of both "personal data" and "relevant filing system". This guidance covers only "personal data". We intend in the near future to publish guidance on the meaning of "relevant filing system". In the mean time we are retaining the appendix to our previous guidance, Frequently Asked Questions on "relevant filing systems". This includes the "temp test" to help organisations decide whether they hold information within a "relevant filing system".
The Data Protection Act 1998 (the DPA) applies only to information which falls within the definition of 'personal data'. The ICO, with other European data protection authorities, has been considering what is meant by 'personal data' in Directive 95/46/EC on the protection of individuals with regard to the processing of personal data and on the free movement of such data (the European Data Protection Directive or the Directive). This work has culminated in Opinion 4/2007 on the concept of personal data (01248/07/EN - WP136) adopted by the Article 29 Data Protection Working Party on 20 June 2007. This guidance draws on Opinion 4/2007 and applies the concepts discussed in that paper in a UK context.
The object of the European Data Protection Directive1 , implemented in the UK by the DPA, is to provide that "Member States shall protect the fundamental rights and freedoms of natural persons, and in particular their right to privacy with respect to the processing of personal data".
'Personal data' is defined in Article 2 of the Directive by reference to whether information relates to an identified or identifiable individual.
The Directive provides, in Article 3, that it applies only to the processing of personal data where the processing is wholly or partly by automatic means, or where it is non-automated processing of personal data which forms part of a 'filing system2' or is intended to form part of a 'filing system'.
The Directive therefore considers first whether the information relates to an identifiable individual and then describes the two different types of processing (processing by automatic means or non-automated processing within a 'filing system') which will bring information within the scope of the Directive.
The DPA repeats the substance of the Directive definition of 'personal data' but tackles the definition in reverse order to the Directive. The DPA first considers the nature of the processing to determine whether the information in question is 'data' (either processed by automatic means or non-automated processing within a filing system) and, secondly, considers whether the 'data' is 'personal data' in that it relates to an identifiable individual.
The Directive and the DPA cover two common categories of information:
In most circumstances it will be relatively straightforward to determine:
and consequently, to determine whether you are processing 'personal data'.
In most cases it will be obvious when you are processing personal data. In those relatively few cases where this is unclear, this guidance, and in particular the questions set out in the flowchart, aim to take you through the factors to consider when determining whether you are processing personal data. The guidance offers suggestions, for use in appropriate cases, of considerations which may help you reach a decision about the nature of the information in question.
The DPA introduces two more types of manual processing of information which, if the information relates to an identifiable individual, will involve processing of 'personal data'. These additional categories of processing are introduced in the DPA definition of 'data' and concern:
The DPA is therefore concerned with four types of data which can be broadly referred to as:
Whether information falls within any of the four categories of 'data' covered by the DPA is considered in our guidance 'What information is 'data' for the purposes of the DPA?'
This guidance aims to help you determine whether 'data' is 'personal data' for the purposes of the DPA and the Directive. The guidance is in the form of a flowchart of numbered questions which, when taken in order, aim to assist in identifying 'personal data'. The flowchart questions are supplemented by guidance and illustrative examples aimed at developing a practical understanding of the concept of personal data.
There are several steps to determining whether data (electronic or manual) is 'personal data'6 for the purposes of the DPA. Questions to help you are set out in boxes 1 to 8 below7.
Can a living individual be identified from the data, or, from the data and other information in the possession of, or likely to come into the possession of, the data controller?
Yes Go to next question.
No The data is not personal data for the purposes of the DPA.
See definition of 'personal data' section 1(1) DPA.
See also p. 6-8 Legal Guidance.
An individual is 'identified' if you have distinguished that individual from other members of a group. In most cases an individual's name together with some other information will be sufficient to identify them.
A name is the most common means of identifying someone. However, whether any potential identifier actually identifies an individual depends on the context. By itself the name John Smith may not always be personal data because there are many individuals with that name. However, where the name is combined with other information (such as an address, a place of work, or a telephone number) this will usually be sufficient to clearly identify one individual8 . (Obviously, if two John Smiths, father and son, work at the same place then the name, John Smith, and company name alone will not uniquely identify one individual, more information will be required).
Simply because you do not know the name of an individual does not mean you cannot identify that individual. Many of us do not know the names of all our neighbours, but we are still able to identify them.
Example: The tall, elderly man with a dachshund who lives at number 15 and drives a Porsche Cayenne.
Example: A description of an individual may be personal data where it is processed in connection with a neighbourhood watch scheme or by the police, when seeking to identify potential witnesses to an incident.
There will be circumstances where the data you hold enables you to identify an individual whose name you do not know and you may never intend to discover.
Example: Where an individual is not previously known to the operators of a sophisticated multi-camera town centre CCTV system, but the operators are able to distinguish that individual on the basis of physical characteristics, that individual is identified. Therefore, where the operators are tracking a particular individual that they have singled out in some way (perhaps using such physical characteristics) they will be processing 'personal data'.
Similarly, a combination of data about gender, age, and grade or salary may well enable you to identify a particular employee even without a name or job title.
Sometimes it is not immediately obvious whether an individual can be identified or not, for example, when someone holds information where the names and other identifiers have been removed. In these cases, Recital 26 of the Directive states that, whether or not the individual is nevertheless identifiable will depend on "all the means likely reasonably to be used either by the controller or by any other person to identify the said person".
Therefore, the fact that there is a very slight hypothetical possibility that someone might be able to reconstruct the data in such a way that the data subject is identified is not sufficient to make the individual identifiable for the purposes of the Directive. The person processing the data must consider all the factors at stake.
The starting point might be to look at what means are available to identify an individual and the extent to which such means are readily available. For example, if searching a public register or reverse directory would enable the individual to be identified from an address or telephone number, and this resource is likely to be used for this purpose, the address or telephone number data should be considered to be capable of identifying an individual.
When considering identifiability it should be assumed that you are not looking just at the means reasonably likely to be used by the ordinary man in the street, but also the means that are likely to be used by a determined person with a particular reason to want to identify individuals. Examples would include investigative journalists, estranged partners, stalkers, or industrial spies.
Means of identifying individuals that are feasible and cost-effective, and are therefore likely to be used, will change over time. If you decide that the data you hold does not allow the identification of individuals, you should review that decision regularly in light of new technology or security developments or changes to the public availability of certain records.
Taking this into account, a person who puts in place appropriate technical, organisational and legal measures to prevent individuals being identifiable from the data held may prevent such data falling within the scope of the Directive.
Does the data 'relate to' the identifiable living individual, whether in personal or family life, business or profession?
Yes The data is 'personal data' for the purposes of the DPA.
No The data is not 'personal data' for the purposes of the DPA.
Unsure See 3 to 8 below.
See definition of 'personal data'' in section 1(1) DPA.
It will often be clear where data 'relates to' a particular individual. However, sometimes this is not so clear and it may be helpful to consider in more detail what is meant by 'relates to'. Data which identifies an individual, even without a name associated with it, may be personal data where it is processed to learn or record something about that individual, or where the processing of that information has an impact upon that individual. Therefore, data may 'relate to' an individual in several different ways, the most common of which are considered below.
Is the data 'obviously about' a particular individual?
Yes The data is 'personal data' for the purposes of the DPA.
No Go to next question.
There are many examples of data that 'relate to' an individual because the records concerned are clearly about that individual.
Example: A medical history, a criminal record, or a record of a particular individual's performance at work or in a sporting activity.
With these types of information it is the content of the information that determines that it 'relates to' an individual.
There are many examples of records which will clearly be personal data where the information in question is not 'obviously about' an individual but is about their activities.
Example: Data such as personal bank statements or itemised telephone bills will be personal data about the individual operating the account or contracting for telephone services.
Where data is not 'obviously about' an identifiable individual the following question may help to determine whether the data is 'personal data'.
Is the data being processed, or could it easily be processed, to:
something about an identifiable individual,
as an incidental consequence of the processing, either:
Questions 4 to 8 may help when considering this issue.
Is the data 'linked to' an individual so that it provides particular information about that individual?
Yes The data is 'personal data' for the purposes of the DPA.
No Go to next question.
There will also be many cases where data is not in itself personal data but, in certain circumstances, it will become personal data where it can be linked to an individual to provide particular information about that individual.
Example: Data about the salary for a particular job may not, by itself, be personal data. This data may be included in the advertisement for the job and will not, in those circumstances, be personal data. However, where the same salary details are linked to a name (for example, when the vacancy has been filled and there is a single named individual in post), the salary information about the job will be personal data 'relating to' the employee in post.
Is the data used, or is it to be used, to inform or influence actions or decisions affecting an identifiable individual?
Yes The data is 'personal data' for the purposes of the DPA.
No Go to next question.
There are many other examples of data which 'relate to' a particular individual because it is linked to that individual and informs or influences actions or decisions which affect an individual.
Example: Data about an individual's phone or electricity account clearly determines what the individual will be charged.
Context is important here. Information about a house is often linked to an owner or resident and consequently the data about the house will be personal data about that individual. However, data about a house will not, by itself, be personal data.
Example: Information about the market value of a particular house may be used for statistical purposes to identify trends in the house values in a geographical area. The house is not selected because the data collector wishes to know anything about the occupants, but because it is a four bedroom detached house in a medium-sized town. As soon as data about a house is either:
- linked to a particular individual, for example, to provide particular information about that individual (for example, his address) (see 4 above); or
- used in deliberations and decisions concerning an individual (even without a link to the individual's name, for example, the amount of electricity used at the house is used to determine the bill the individual householder is required to pay);
then that data will be personal data.
In both these examples the data about the house relates to the individual because the purpose of processing that data is to learn something about the individual (his address) or to determine something about him (the extent of his liability).
Example: Data used in deliberations or decisions about an individual may include data about unauthorised alterations to a house in breach of planning law where that data is processed to determine whether to prosecute the individual house owner. The data about the unauthorised alterations may be processed by reference to the house address but the data clearly relates to the individual who carried out the alterations in that the data is being processed to determine whether to take action against that person.
Example: Where the value of a particular house is used to determine an individual liability for Council Tax, or is used to determine the assets of an individual or individuals in proceedings following divorce, then this will be personal data because the data about the house is clearly linked to the individual or individuals concerned.
Example: A utility company may not record the name of the occupier of the house to which it provides water, but may simply note the address of the property and address all bills to 'the occupier'. Data concerning the water consumption for a particular address will be personal data about the occupier in that this data determines what that individual will be charged.
In this last example, even without a name associated with the water consumption data, this data will be personal data in that it determines what the occupier will be charged and the occupier is identified, even without a name, as the person living at the property in question and is therefore distinguished from other individuals. Also, if necessary, the water company is likely to be able to easily obtain the name of, if not the occupier, then at least the registered owner of the property.
It is important to remember that the same piece of data may be personal data in one party's hands while it may not be personal data in another party's hands.
Example: At New Year celebrations in Trafalgar Square two almost identical photographs of the revellers are taken by two separate photographers and stored in electronic form on computer. The first photographer, a photo journalist, takes a picture of the crowd scene to add to his photo library. The second photographer is a police officer taking photos of the crowd scene to identify potential troublemakers. The data in the electronic image taken by the journalist is unlikely to contain personal data about individuals in the crowd as it is not being processed to learn anything about an identifiable individual. However, the photo taken by the police officer may well contain personal data about individuals as the photo is taken for the purpose of recording the actions of individuals who the police would seek to identify, if there is any trouble, so they can take action against them.
A single piece of data, which is not personal data for one data controller may become personal data when it is passed to another data controller.
Example: An estate agent takes a photograph of a high street shop to market the property. The photograph is held in digital form by reference to its address or by reference to the client name on the agent's computer. The photograph is used solely to produce photographic prints to display and distribute to potential purchasers.
The photograph of the shop includes images of pedestrians who were walking past the shop at the time the photo was taken. The estate agent is not processing the shop data to learn anything about any of the pedestrians whose images were captured by chance on the photo, nor is it likely that the estate agent would ever process the photo for that purpose. The estate agent is unlikely to possess the appropriate software to digitally enhance the photo to identify individuals. Therefore, in the hands of the estate agent, the photo does not contain personal data about the pedestrians as it is not processed to learn something about those individuals and nor is it likely to be processed by the estate agent for this purpose.
If we consider the example of the data contained in the images of the pedestrians captured on the shop photo by the estate agent in the above example, in certain circumstances, this data may be personal data about the pedestrians in the hands of another data controller.
Example: If, at about the same time as the photograph was taken by the estate agent, a bank raid took place on the same high street, the police might make a public appeal for information about movement on the high street at that time. The estate agent might supply the police with a copy of the photo in response to the appeal. The police would then process the digital photo, not to learn anything about the shop but, using photo enhancing technologies, to attempt to identify potential witnesses or suspects. The photo would then be being processed to learn something about the individual pedestrians and, in the hands of the police, may be personal data about such individuals
Therefore, data may not be personal data in the hands of one data controller (for example, the estate agent) but the same data may be personal data in the hands of another data controller (for example, the police) depending on the purpose of the processing and the potential impact of the processing on individuals.
Does the data have any biographical significance in relation to the individual?
Yes The data is likely to be personal data for the purposes of the DPA.
No Go to next question.
Unsure Go to next question.
It is important to remember that it is not always necessary to consider 'biographical significance' to determine whether data is personal data. In many cases data may be personal data simply because its content is such that it is 'obviously about' an individual. Alternatively, data may be personal data because it is clearly 'linked to' an individual because it is about his activities and is processed with the purpose of determining or influencing the way in which that person is treated. You need to consider 'biographical significance' only where information is not 'obviously about' an individual or clearly 'linked to' him.
When considering 'biographical significance', what is important is whether the data go beyond recording the individual's casual connection with a matter or event which has no personal connotations for him. Does the processing of this data affect, or is it likely to affect, the individual? Data may, for example, have personal connotations for an individual if it provides information about an individual's whereabouts or actions at a particular time.
Example: Where an individual is listed as an attendee in the minutes of a meeting then the minutes will have biographical significance for the individual in that they record the individual's whereabouts at a particular time.
The fact that an individual attended the meeting will be personal data about that person. However, this does not mean that everything in the minutes of that meeting is personal data about each of the attendees.
Whether the content of the minutes includes any additional personal data, beyond attendance data, about the attendees at the meeting may be determined by the focus of the minutes.
Does the data focus or concentrate on the individual as its central theme rather than on some other person, or some object, transaction or event?
Yes The data are likely to be personal data for the purposes of the DPA.
No Go to next question.
Unsure Go to next question.
Again, it is important to remember that it is not always necessary to consider 'focus' to determine whether data is personal data. In many cases data may be personal data because it is 'obviously about' an individual, or because it is clearly 'linked to' an individual because it is about the individual's activities.
You need to consider the 'focus' of the data only where information is not 'obviously about' an individual or clearly 'linked to' them.
It is often difficult to determine whether the contents of minutes of a meeting are personal data about either those attending the meeting or individuals whose conduct or condition is discussed at the meeting. Considering the 'focus' of the minutes may help determine whether any personal data is involved.
There will be circumstances where part of the record of a meeting will be personal data as the data is 'obviously about' or clearly 'linked to' an individual.
Example: Where an individual's suitability for a particular course or post is discussed (consideration being given to the individual's qualifications, personality and/or performance at work), the record of these discussions will be personal data about the individual in question.
In this last example, where a candidate's suitability for the job is only one of many topics discussed at the meeting, the whole of the record of the meeting will not necessarily be personal data about that candidate.
Example: Where a meeting is held to consider four candidates for a job, only the information which concentrates on the individual in question will be personal data about that individual. Information about other candidates, the need for a new person in the job or the creation of the new job, (that is, information which does not concentrate on the individual in question) will not be personal data about that individual. The minutes may therefore contain four separate sets of personal data about the four candidates respectively as well as information which is not personal data as it concerns the business requirement for the new employee.
If the whole of a meeting is about a particular individual then, assuming the minutes are held as data, they will be personal data about that individual. The meeting may concern the behaviour and actions or the condition of an individual. The personal data will include not only those facts about the condition or behaviour of the individual discussed at the meeting, but also any third parties' opinions about the individual in question and any indication of the intentions of any person in respect of that individual. These expressions of opinion or intention are personal data of the individual being discussed9 .
Example: A disciplinary hearing is held into the conduct of an individual employee. Everything discussed at the meeting is likely to be personal data about the individual in question. This will include the statements of fact about the employee's behaviour; opinions about the employee; and statements as to any proposed disciplinary measures provided by colleagues. The minutes of this meeting will be personal data about the individual as the information is clearly linked to the behaviour, condition or activities of the individual in question.
Where comments made at a meeting are, in the minutes, directly attributed to a particular individual, whether the comments are personal data about the speaker will depend on the capacity in which the speaker made the comments. That is to say, consider whether the individual is giving a personal opinion or is putting forward views on behalf of another individual, company or organisation (most commonly, the individual's employer).
Example: Where an individual attends a meeting in the capacity of an employee (for example, to discuss the provision of services by the company), if the employee expresses the views of the company, those views, when recorded in the minutes of the meeting, will not be personal data about the employee. The views will be information about the position of the company with regard to the service provision as expressed by its agent, the employee.
However, if allegations were made that the employee's representations failed to reflect the views of the organisation, information as to the representations made at the meeting could become personal data about the conduct of the employee.
The views of a company or organisation as expressed by its agent (either an employee or professional representative), are not personal data about the agent. The focus of the comments does not concern the employee's or agent's personal views but concerns the company's position.
When considering the 'focus' of information it may be helpful to consider whether the information is being processed to record something about an individual or to record information about an object.
Example: Information may be recorded about the operation of a piece of machinery (say, a biscuit-making machine). If the information is recorded to monitor the efficiency of the machine, it is unlikely to be personal data (however, see 8 below). However, if the information is recorded to monitor the productivity of the employee who operates the machine (and his annual bonus depends on achieving a certain level of productivity), the information about the operation of the machine will be personal data about the individual employee who operates it.
Whether information is linked to an individual, for example, to learn something about that individual, is the key factor in determining whether information about an object (for example, a biscuit-making machine) is personal data.
Also, if the information has potential to be used to learn something about an individual, it may be personal data as discussed below.
Does the data impact or have the potential to impact on an individual, whether in a personal, family, business or professional capacity?
Yes The data is 'personal data' for the purposes of the DPA.
No The data is unlikely to be 'personal data'.
Even though the data is not usually processed by the data controller to provide information about an individual, if there is a reasonable chance that the data will be processed for that purpose, the data will be personal data.
Example: A taxi firm may record the movements of the taxis in its fleet by using vehicle tracking devices. The data is used by the firm to help provide the taxi service in that the control centre will know where all the taxis are at any one time and will therefore, on receiving a request for a taxi, be able to direct the nearest taxi to pick up the new passenger. The data is not intended to be used to inform the taxi firm as to the whereabouts of each individual taxi driver, but to plot the location of the fleet of taxis.
Even though the data was not intended to be used to record individual drivers' movements, the taxi control staff will usually know which driver is driving which taxi at any particular time and the data could therefore be used, without any adjustment, to locate a driver.
If family members needed to contact a taxi driver, they could ask one of the taxi control staff to use the taxi location data to provide the location. Consequently the taxi location data may be personal data about the taxi drivers.
If, as a matter of fact, data is occasionally processed to learn something about an individual, even though it was not the data controller's intention to process the data for this purpose, this data will be personal data as the processing does, or is likely to, impact on the individual.
Example: If we consider the taxi location data referred to in the example given above, if the control centre occasionally uses the taxi data to locate individual drivers, even though this was not the data controller's primary purpose for processing, the taxi location data will be personal data about the individual drivers.
When considering data about objects, if the data is processed to provide particular information about an individual (for example, information about a biscuit-making machine is used to assess the productivity of the operator of the machine) the data will be personal data.
Where data about objects is not currently processed to provide particular information about an individual, but could be processed to provide information about an individual (for example, taxi location data) the data is likely to be personal data.
What is being considered here is whether the processing of the information has or could have a resulting impact upon the individual even though the content of the data is not directly about that individual, nor is there any intention to process the data for the purpose of determining or influencing the way that person is treated.
There will be circumstances where it remains uncertain whether particular data is personal data. Where this is the case we consider that, as a matter of good practice, you should still treat the information with care and, in particular, ensure it is held and disposed of securely.
Other issues concerning 'personal data' are addressed in the appendix attached to this guidance.
Inevitably, because man is a social animal, a record which is mainly about one individual, and therefore personal data about that individual, will often contain personal information about another person, a partner, child, relative or friend. The DPA contains provisions which address the implications of this when responding to a subject access request10 .
There are circumstances where the same information is personal data about two or more individuals. This may be due to one of three factors:
Example: Consider the record of the arrest of an individual by a policeman. Where the individual arrested, or the arresting policeman, records an account of the circumstances of the arrest, if the record is held as data subject to the DPA, it will be personal data of both individuals.
There is no sensible way of separating the account of the direct interaction between the two individuals involved into personal information about each one separately. Indeed, the precise nature of the interaction leading to the arrest and its immediate consequences is a crucial part of the record.
Example: For each child attending a particular school, the school records emergency contact details identifying the name, address and phone number of the adult to be contacted should the child have an accident. The emergency contact information will be personal data about the adult (in that the content of the information comprises their name and contact details). The information will also be personal data about the child as the purpose of holding the information is to contact the child's responsible adult in the event of an emergency.
Example: An investigation is carried out into allegations made by an employee of bullying by a manager. In the course of the investigation other employees are asked about their dealings with the employee and the manager concerned. In these circumstances, the views of the employees who have been interviewed are likely to be personal data about both the complainant and the subject of the complaint and will also be personal data about the interviewee.
There has been some confusion about whether the records associated with complaints can be personal data about individuals. Records relating to the consideration and investigation of complaints can be personal data about the person making the complaint, but this will depend on the circumstances.
Example: Where a newspaper complains that a government department has failed in its obligation to disclose information under the Freedom of Information Act 2000, the case file is unlikely to be personal data because the newspaper is not an individual. However, the complaint case file will almost certainly contain personal information identifying the journalist making the request on behalf of the newspaper and department officials dealing with the request.
It is possible that a case file could be personal data if, for example, it was about a particular official or Minister. If the request for information directly related to, or concerned, the conduct or activities of a particular Minister or other official, much if not all of the complaint case file is likely to be personal data about that Minister or official.
Example: Where an individual complains that a government department has not responded properly to a Freedom of Information request, and that therefore the individual's right to receive the requested information has been breached, the case file is likely to be personal data relating to the individual complainant.
Where a business requests a DPA assessment of the activities of another business, as they are entitled to do, the case file is unlikely to be personal data.
Example: Where an individual's complaint that a particular company has been fly-tipping prompts an investigation into the alleged incidents, the case file will contain information about the investigation and the case file will not be personal data about the individual complainant.
However, where an individual complains that a police force has not responded properly to a subject access request made under section 7 of the DPA, then the case file will be personal data relating to that individual.
Even in circumstances where the bulk of a case file is personal data about the individual complainant, it is likely to contain some personal information about other individuals. It may well also contain information which is not, in itself, personal data, for example, if it includes details of policies and procedures relevant to the case.
Article 1 of the Directive states that "personal data means any information relating to an identified or identifiable natural person." Recital 26 of the Directive states that "to determine whether a person is identifiable, account should be taken of all the means likely reasonably to be used either by the controller or any other person". This means that where, though it is conceivable that someone could identify a particular individual or individuals if they devoted sufficient effort and resources to the task, it is unlikely that anyone will do so then the individual(s) are not "identifiable" for the purposes of the Directive.
Data may be held as personal data by one organisation, because they can link the data to living identifiable individuals. The same data may be held by another organisation that is unable to link the data concerned to individuals. The question arises, therefore, whether the latter organisation holds personal data because the data, in the hands of another organisation, are linked to identifiable individuals.
The pragmatic line taken by the Article 29 Working Group is that where an organisation holds records which it cannot link, nor is ever likely to be able to link, to particular individuals, the records it holds will not be personal data. This will only be the case where it is unlikely that anyone else to whom the records may be released will be able to make such links. This will remain the case even if there is one organisation that would be able to make such a link as long as that organisation will not release information enabling such links to be made and adopts appropriate security. Where there is no likelihood of records being linked to individuals there is no real risk of any impact on those individuals. Therefore, where researchers hold samples of 'anonymised' individual census records released by the Office of National Statistics they will not be processing personal data even though the ONS may be able to link records to particular individuals.
There will be circumstances in which an organisation holds records relating to individuals where the obvious identifiers have been removed but where there is a need to be able to initiate contact with particular individuals if necessary.
Example: An EU based company carries out pharmaceutical research on identifiable individuals. They remove the obvious identifiers from the individual records (name, address etc) and key code them (that is they assign a unique code such as KLPR767805 to each individual record). They then release the 'anonymised' individual records to another pharmaceutical company which will use them for further research. In the event that the second company identifies that a particular individual might be at risk because of the combination of their illness and the drugs they are using, the second company can alert the first company, 'identifying' the individual in question by means of the code. The first company can then contact the individual. The key question is whether the second company holds the records in question as personal data.
The second company is able to isolate particular records where the medical histories and current medication give cause for concern and 'identify' them by means of the codes. Unless there are exceptional circumstances, for example where an individual has a very rare condition and there has been publicity in the press which named them, it is unlikely that the second company will ever find out the name and address or other information which would enable them to physically find the individual in question. However, by alerting the first company to their concerns they do cause the individual to be contacted and thus their processing has a clear effect on the individual. Nevertheless, because they do not contact the individual themselves and because they have no interest in the individuals themselves, merely in ensuring that where records give cause for concern the individual is contacted, we consider that for all practical purposes they do not hold the key coded records as personal data. A significant consideration here is that as long as the first company have appropriate security in place there is little or no chance that any other person who might have access to the coded records would be able to link an individual by name and or address to a particular record. In such circumstances the chances of an individual suffering detriment are negligible.
A question faced by many organisations, particularly those responding to Freedom of Information requests, is whether, in disclosing information that does not directly identify individuals, they are nevertheless disclosing personal data if there is a reasonable chance that those who may receive the data will be able to identify particular individuals.
Example: An organisation receives a Freedom on Information request for the full home addresses of its staff but without staff names attached.
Organisations need to consider whether, by releasing the addresses, they will have released personal data. The DPA refers to data which relate to a living individual who can be identified from that data or from that data and other information in the possession of, or likely to come into the possession of, the data controller. This emphasis on identification of individuals by "the data controller" might suggest that, even if there is a reasonable chance that someone other than the recipient of the address information might be able to link particular addresses to specific staff members, the data controller has not released personal data merely by releasing the addresses to the particular FOI applicant. However, the definition of personal data in the Directive suggests otherwise.
The Directive provides that "personal data shall mean any information relating to an identified or identifiable natural person... ; an identifiable person is one who can be identified, directly or indirectly...".
This definition would suggest that an organisation would be disclosing personal data where it releases information which can be linked to particular individuals. Taking into account the purpose of the Directive this seems a sensible view. It is a view which the Information Tribunal took when deciding whether a local authority should release the addresses of empty properties. The Tribunal held that releasing such addresses would involve releasing personal data where the properties were owned by individuals.
1 See Article 1 European Directive
2 As defined in European Directive Article 2 (c)
3 Data in electronic form is defined in section 1(1)(a) of the DPA.
4 'Relevant filing system' is defined in section 1(1)(a) DPA.
5 'Accessible record' is defined in section 1(1)(d) and section 68 DPA.
6 See definition of 'personal data' section 1(1) DPA.
7 See also p. 5-7 Legal Guidance
8 The term 'personal data' undoubtedly covers the name of a person in conjunction with his telephone number or information about his working conditions or hobbies - Paragraph 24 of the Opinion of Advocate General Tizzano in the Lindqvist case (Bodil Lindqvist v Aklagarkammaren i Jonkoping - Case Commissioner-101/01 - European Court of Justice) dlivered on 19 September 2002
9 DPA section 1(1) - 'personal data'
10 Section 7 DPA 1998