Drinking Water Inspectorate Logo

Information Letters

 

DWI Information Letter 8/2001

15 June 2001

To: Board Level Contacts of Water and Sewerage Companies and Water Companies in England

Dear Sir or Madam

CRYPTOSPORIDIUM IN WATER SUPPLIES: MONITORING DATA AND ANOMALY REPORTING FOR THE WATER SUPPLY (WATER QUALITY) REGULATIONS 2000 AND WATER SUPPLY (WATER QUALITY) REGULATIONS 1989 AS AMENDED

Purpose

  1. The purpose of this letter is to:

    • advise companies of the checks carried out on the regulatory Cryptosporidium data submitted monthly;

    • advise companies of changes to the reporting by the Inspectorate of the outcome of investigations of Cryptosporidium anomalies;

    • seek companies’ views on proposed changes to the ways monitoring data and anomalies are reported to the Inspectorate; and

    • clarify the requirements for the retention of information.

Introduction

  1. Information Letters 10/1999, 4/2000, 10/2000, 13/2000 and 28/2000 set out the arrangements for Water Companies to report to the Inspectorate the results of analysis of regulatory Cryptosporidium sampling. Operational data needs to be reported as a separate file as specified in Information Letter 10/2000 and will not be subjected to data checks.

  2. The arrangements for reporting by the Inspectorate the outcomes of investigations of anomalies required a full report for all anomalies. These arrangements have been changed.

  3. The present arrangements for reporting monitoring data require the data to be returned in Excel97 spreadsheet format, with anomalies being reported using the pro-formas given in annexes B and C of Information Letter 28/2000. The Inspectorate proposes making changes to these procedures and the classification of anomalies to improve the efficiency of data handling and to allow cross referencing of anomalies and data.

Checks carried out on regulatory data returns

Formatting errors

  1. Data received by the Inspectorate are transferred into an Access97 database. Any incorrectly formatted data, for example data entered in number format instead of a date or text format, is rejected. The structure of the database ensures that any sample with a duplicated evidence bag number is identified and the new set of data is rejected. When formatting errors are detected, the Company concerned will receive a letter notifying it of the errors and asking them to correct and resubmit the data. The Company will normally be given fourteen days to respond.

  2. As mentioned in paragraph 2 above, operational data must be kept separate from regulatory data, but should be in exactly the same form to permit the data to be pooled if required. Currently this should be done by returning separate Excel97 workbooks for regulatory and operational data. Operational data is not subject to data checks. Fields for which their is no data from operational sites should be left blank.

  3. At sites which have been designated as regulatory and are subject to a Notice issued under either regulation 23A(7) of the Water Supply (Water Quality) Regulations 1989 as amended or regulation 28(7) of the Water Supply (Water Quality) Regulations 2000, regulatory monitoring must continue until a revised Notice is issued removing that site from the list. This applies to all sites irrespective of the reason for their change of status.

Data errors

  1. Once the data has been successfully imported into our database checks are carried out to ensure the integrity of the data return. Many of these checks are carried out in Access97 but some are carried out in Excel97. Data errors are issued to the relevant Company and the Company is normally given fourteen days to reply. The details of the checks are given in Appendix A. It is expected that companies will wish to carry out their own checks on the integrity of the data before it is submitted.

  2. Electronic versions of the formulas and queries used are available on request from Paul Irving (telephone number 0207 944 5988). E-mail: Paul_Irving@detr.gov.uk.

  3. When complete and accurate data are received further checks will be carried out on the data to identify any unreported breaches of the regulations and other important anomalies. Companies will be notified of these checks in due course.

Results to be reported

  1. Only the confirmed result of analysis, as defined in part 2 of the Protocol, must be reported. If, following investigation by the Inspectorate or as a result of discussions with the Inspectorate, a result is changed this changed result must be the result reported. Any such changes to the results of analysis and related data that take place after the data has been submitted must be notified to the Inspectorate in writing. The information required is the analysis ID (evidence bag number), field(s) requiring amendment, the old value(s), the new value(s) and the reason(s) for the change.

Proposed changes to the reporting of monitoring data and anomalies by companies

  1. Currently, the Inspectorate receives all monitoring data in Excel97 format. This data is then imported into an Access97 database and then exported into Excel97 for further checking. As mentioned in paragraph 5 above there have been problems in the past with the formatting of data in Excel97 by some companies, which resulted in data not being loaded into Access97. This problem was brought to the attention of companies, and such problems are now very rare. The Inspectorate is grateful to the companies for their assistance in resolving this problem.

  2. A number of the failed data checks are as a result of notifiable matters, which cannot currently be cross referenced with the anomalies recording system. There are also a number of events notified under the anomalies reporting system that are not anomalies, but which relate to the data checks.

  3. The Inspectorate proposes to request that all monitoring data be reported in Microsoft Access97 format. Extra fields will be used to identify the reference number of any relevant anomaly in addition to the current comments field, which will be retained for providing fuller details where required, and information relating to failed checks.

  4. A number of failed data checks have arisen as a result of the sampler failing to take readings at the correct time. Details of the correct time to take readings are given in appendix B. The Protocol will be amended to reflect the new advice on the taking of readings when Part 1 is next revised. In the meantime appendix B should be kept with Part 1 of the Protocol.

Changes to the notification of anomalies

  1. Under existing arrangements, companies do not know the reference number of an anomaly. It is proposed that this be addressed by the company assigning the reference number. Additionally, to make other cross checks easier, it is proposed that companies include in the notification and report the evidence bag number(s) of the affected sample(s). The notification and report pro-formas will be amended to allow the reference number and evidence bag numbers to be included. The proposed format for the reference is XXX/yy/mm/nnn

    2. WhereXXX is the company identification (provided by the Inspectorate)
     yy is the year in which the anomaly occurred eg 01 for 2001
     mm is the month in which the anomaly occurred eg 01 for January
     nnn is a sequential number assigned by the company, restarting at one for the first anomaly each month.

  2. The Inspectorate proposes to revise the list of anomaly notification codes to remove sampling equipment repairs and maintenance and tagging from the list of anomalies. This will require all sampling units to have their own tag log. Companies will be responsible for maintaining this log to demonstrate continued chain of evidence from the DWI Inspection Report. More detail of the proposed change is given in appendix C. An example page from such a book is also included in appendix C. Other changes include the splitting of some of the remaining anomaly types to allow automated processing of data. A list of the proposed new anomaly codes and notification requirements is given in appendix D. A comparison with the current arrangements is given in appendix E.

Changes to the reporting of the outcome of investigations of Cryptosporidium anomalies by the Inspectorate

  1. The Inspectorate previously provided a report of the outcome of its investigation of each reported anomaly. Most of these are routine matters (eg maintenance) or are regarded as having been fully and satisfactorily addressed by the company. These reports had to be produced by the Inspectorate and read by the company. The Inspectorate has adopted a system of exception reporting. Reports will no longer be produced for routine matters, for anomalies that have been addressed fully and satisfactorily by the company, or for those where any remaining deficiencies are regarded as trivial.

  2. A significant number of anomalies have been caused by maintenance work being carried out on sampling units at inappropriate times. All maintenance work on sampling units should be undertaken during the one hour sample changeover period. This extended period was specifically written into the regulations to allow maintenance work to be carried out. Apart from maintenance work overrunning due to unforeseen problems, the only exceptions are emergency work required to:

    (a) maintain or restore satisfactory water treatment;

    (b) restore Cryptosporidium sampling following loss of sampling; or

    (c) address an immediate health and safety issue.

    As an alternative to scheduling maintenance to coincide with sampling, there is also leeway in the Regulations for a company to retime sampling to coincide with maintenance work.

  3. The Inspectorate will continue to request further information relating to an anomaly whenever this is required, but will only produce a report if it has a recommendation or other relevant comment to make. One company has suggested that a periodic statement allowing them to close their own files would be useful. The Inspectorate accepts this suggestion and will provide companies with such statements. The precise details are not yet available, but will probably take the form of a statement that all anomalies identified before a specific date are closed with the exception of specified anomalies.

Retention of records

  1. Paragraph 3.5.1 of Part 1 of the Protocol specifies that all documentation and receipts must be kept in a secure location for a minimum period of twelve calendar months following sampling. This applies to all documents associated with the sample. On rare occassions it may be necessary to retain the records relating to a specific sample for longer. The Inspectorate will notify a company in writing when this is the case. Such records must be stored under secure conditions until written authorisation is given by the Inspectorate for their disposal. The requirement to retain records for one year is in addition to the requirements to retain information entered on the public record.

  2. Companies’ views are sought on the proposals in paragraphs 12 to 15 by 12 July 2001. Comment is also welcomed on the changes detailed in paragraphs 16 and 17. Companies should address their views and comments to David Drury.

Enquiries

  1. Any enquiries about this letter should be addressed to David Drury (020 7944 5976) or Paul Irving (020 7944 5988).

  2. Copies of this letter and the attachments are being sent to Pamela Taylor, Chief Executive, Water UK; Rodney Anderson, Water Supply and Regulation Division, Department of the Environment, Transport and the Regions; Bob Macey, Environment Division, The National Assembly for Wales; Tim Hooton, Water Services Unit, Scottish Executive; Randal Scott, Drinking Water Inspectorate for Northern Ireland; Rowena Tye, Office of Water Services and David Harper, Department of Health.

    Yours faithfully



    Michael Rouse
    Chief Inspector

Appendix A

Checks carried out on regulatory cryptosporidium data returns

NUMBER OF CHECK
EXPLANATION
PROGRAMME USED
1A “Company code” is valid Access97
1B “Laboratory code” is valid Access97
1C “Site id” is valid Access97
2 Blank fields1 Access97
3 “Sample date & time end” not before first of month & not after the last of the month Access97
4 “Sample date & time start” is not more than one hour after previous sample “date & time end”2 Excel97
5 “Sample date & time start” is not be before sample date & time end of the previous entry (including the first entry of each month) Excel97
6 “Report date” is not before “sample date & time end” Access97
7 “Report date” is not greater than “sample date & time end” + 3 days3 Access97
8 “Timer reading” must not be less than the previous “timer reading” for that site Excel97
9 “Elapsed time” = “timer reading” – previous “timer reading” for that site (+/- 0.2 hours) Excel97
10 “Water meter reading” is not less than previous “water meter reading” for that site Excel97
11 “Sample volume” = “water meter reading” – previous “water meter reading” to the nearest 1 litre Excel97
12 “Oocyst concentration” not less than 0.01 Access97
13 If “oocyst <” then oocyst concentration must be 0.014 Access97
14 “Sample volume” is not be less than 200 or is not less than 40 times “elapsed time” Access97
15 “Elapsed time” is > 0 and <48 hours Access97
16 “Elapsed time” is not greater than “sample date & time end” – “sample date & time start” (+/- 0.2 hours) Access97
17 Number of samples reported for site in the month not Excel97
18 Data have been included in the return for each “Site id” with the date for commencement of monitoring before the last day of the month Access97
19 “Sample date & time end” is not before “sample date & time start” Access97

1 This query checks whether the following fields are blank:

  • Sample Date & Time Start
  • Sample Date & Time End
  • Report Date
  • Timer Reading
  • Elapsed Sample Time
  • Water Meter Reading
  • Sample Volume
  • Headloss at end of run
  • Deposit Volume
  • Oocyst Concentration
    •

2 Regulation 29(7)(a) of the 2000 Regulations and regulation 23B(6)(a) of the 1989 Regulations allows an hour change over period between samples. Companies are expected to carry out maintenance of the sampling equipment during this time.

3 Regulations 29(12) & 29(15) of the 2000 Regulations and regulations 23B(11) & 14 of the 1989 Regulations require that a collection device is analysed and reported within three days of the date it was removed from the sampling equipment.

4 Result should be <0.01 oocysts per 10 litres if no oocysts detected, except for low sample volumes. If it is greater than 0.01 the number must equal 10 divided by Sample Volume (rounded to two decimal places).

Appendix B

Correct times for the taking of readings when sampling for Cryptosporidium

Reading
When to take reading
Inlet Pressure Finish5 Just before switching sampling unit off to change filter unit
Flow Finish5 Just before switching sampling unit off to change filter unit
Headloss Finish Just before switching sampling unit off to change filter unit
Timer Reading Finish Immediately before sampling unit is switched off to change filter unit
Finish Time6 When sampling unit is switched off to change filter unit
Water Meter Reading Finish Immediately after disconnecting the used filter unit
Water Meter Reading Start Immediately before sampling unit is switched on after changing filter unit
Start Time6 When sampling unit is switched on after changing filter unit
Timer Reading Start Immediately after sampling unit is switched on after changing filter unit
Headloss Start Just after taking Timer Reading Start, once reading has stabilised
Flow Start5 Just after taking Timer Reading Start, once reading has stabilised
Inlet Pressure Start5 Just after taking Timer Reading Start, once reading has stabilised

5 The recording of this information is optional, but is recommended.

6 A clock, watch or the time display on the Trendview unit may be used provided it is calibrated twice a year and is known to be accurate to within one or two minutes between calibrations. Calibration can be carried out using the speaking clock in March and October. Clocks, watches and timers that are not calibrated or are of unsuitable or unknown accuracy should not be used.

Appendix C

Proposed Changes to Anomalies Reporting

Introduction

The idea behind the requirements laid down in the Protocol was to ensure that a chain of evidence was in place that could not be challenged effectively in the event of a contravention of the Cryptosporidium treatment standard that might lead to a criminal prosecution.

The present system of having to report all anomalies and the number of re-inspections required because of the anomalies is very cumbersome and inefficient both from the companies’ and the Inspectorate’s points of view.

The Inspectorate believes there is another way of achieving the objective of ensuring that there is a chain of evidence in place in the event that a contravention of the standard does occur. The following proposals would reduce the paperwork and the need for so many re-inspections. In addition to taking expert advice, the Inspectorate would wish to receive comments from companies on these proposals before proceeding to change the current sampling protocol and anomaly reporting requirements.

Sampling Line

Once the initial inspection has been carried out and the site approved, the sample line and all enclosures and cabinets along it will have been initially tagged with blue tags. If in the course of time one or more of these tags are broken the present system requires the Company to fit a temporary white tag and to report the details to the DWI of why the tag was broken, the number of the tag broken, and the number of the temporary white tag fitted in an anomaly report. The Inspectorate believes there is no need for such a report provided all these facts have been properly recorded in a secure logbook provided on site for this purpose. Providing companies comply strictly with the requirement to fill in the logbook (timing and dating entries) the chain of evidence will still be intact. The log book must be bound such that removal, insertion or substitution of pages would be evident. Each page must be sequentially numbered when the book is bound, the pages used sequentially and no gaps and no lines must be left blank on any page. Any unused pages, lines or boxes must be crossed through and signed and dated. The status of tags and the logbook would be audited when the Inspectorate re-inspected the site. If the logbook was not in order or was not a true record of all changes the Inspectorate would consider the matter to be a serious breach of the requirements of the Protocol in so far as the chain of evidence would have been broken. Such a breach could be considered a matter for possible prosecution.

Sampling Cabinet

On all regulatory sites it is a requirement that the sampling cabinet is fitted with an approved, unique padlock and a white security tag. The interior of this cabinet can therefore be regarded as a secure area. The Inspectorate believes in view of this it should be able to dispense with the fitting of security seals over the Pressure Reducing Valve, Sample Pump, and the Terminal Box. Dispensing with these seals would obviate the need for anomaly reports in the event that the Pressure Reducing Valve needed to be adjusted or the Sample Pump needed to be serviced or changed.

In addition, the only blue tag fitted within the sampling cabinet is that fitted on the flushing valve. Provided details of the breaking of this tag were correctly logged together with the details of the replacement white tag fitted and the event dated and timed, an anomaly report would not be required.

Any other fitting within the sampling cabinet could be allowed to be changed without sending in an anomaly report providing that the details (reason for change and date and time the change was made) are correctly entered in the appropriate logbook. The exception to this would be if the water meter is replaced, when both the reading and identity number of the old meter and those of the replacement meter together with the reason for, and date and time of, changing would be required. This information could be recorded in the logbook, and included in the comments field of the appropriate data return.

At the time of re-inspecting any site, the security paint within the cabinet can be re-applied.

In addition the Inspectorate intends allowing a pressure relief valve to be fitted within the sampling cabinet where water pressure at the tapping-off is in excess of 8 bars and the company is concerned about safety. A requirement will be that a pressure relief valve must only be fitted when there is no sampling pump fitted within the sampling cabinet. The pipe work leading to the pressure relief valve must be teed off after the inlet solenoid valve and the pressure relief valve must be a Swagelok "RL4" Series Low Pressure Proportional Relief Valve set for a cracking pressure of 8 bars with an approved check valve fitted downstream of the pressure relief valve to doubly ensure no chance of back siphonage. The pipe work from this check valve must then be teed back into the drainage line within the sampling cabinet downstream of the flow meter. The DWI must be informed where such a device is fitted to a regulatory sampling unit.

Control/Display Cabinet

On the Hydraulic Modelling Services unit a seal is affixed to the terminal door on the programmable logic controller. The breaking of this seal would be considered a serious matter that requires a full anomaly report. Likewise on the Watercare unit the breaking of any of the security tags or seals fitted to the Trendview unit at the time of inspection would be considered a serious matter that requires a full anomaly report.

Maintenance and Repairs

With the above arrangements there would be no need to report almost all maintenance or repairs completed during the one hour sample changeover period. In addition to the matters mentioned above the only exceptions are those that result in modification to either the sampling line, the pipe work inside the sampling cabinet or a change of the K075 key.

Restructuring of Remaining Anomalies

The remaining anomaly codes, approximately 20 of an original 35, have been reviewed. A few have been consolidated, but others have been sub-divided to allow automated and more precise analysis of reported anomalies. With the sole exception of a new anomaly of failing to make correct records in the new site log, the restructuring will not result in more anomalies. The proposed new list, with revised priorities, is given in appendix D.

Example of a typical page from a sampling unit/sample line logbook

Date
Time
Type of Security Device Changed
No. of Security Device Removed
No. of Replace-
ment Security Device
Location of changed Security Device
Reason for Change
Details of work carried out (including serial numbers of parts changed)
Name (Print)
Position
Signature
 

          
 

          
 

          
 

          
 

          

Comments

 

Appendix D

Proposed new anomaly notification

Notification timescale and reporting requirements

Priority
Initial Notification Timescale
Reports required
1 Immediate by telephone7, using cascade system if necessary 72 hour event report.
30 day incident report may be required.
2 Immediate by telephone7, using cascade system if necessary Written notification within one working day of telephone notification8.
Written report of investigation within seven working days of telephone notification9.
3 By Email or telephone10 before end of next working day Written notification within one working day of telephone notification8.
Written report of investigation within seven working days of initial notification
4 Not required Written report of investigation with monthly data return

7 The anomaly must be notified to an Inspector in person. A voice mail message, Email message or Fax will not suffice.

8 Separate written notification is not required if the full report is received within one working day of telephone notification.

9 72 hour and/or 30 day incident reports may be required in some circumstances

10 Telephone notification must be to an Inspector in person. A voice mail message or Fax will not suffice. Telephone notification is not required if the initial notification is sent by Email. However, telephone notification should be made if receipt of the Email is not acknowledged within one working day.

Anomalies to be reported and priority

Notification Code
Anomaly
Comment
Priority
1 Preliminary or confirmed result above regulatory limit Will be handled as a water quality event. 1
2 No sample taken or sample lost before receipt at laboratory. May be handled as a water quality event. 2
3 No result obtained by laboratory or


inability to obtain a valid result		 May be handled as a water quality event.

Eg partial loss of sample during analysis. May be handled as a water quality event		 2
4 Failure to sample for part of the sampling period Including maintenance work carried out outside changeover period or overrunning work scheduled to be completed during changeover period. May be handled as a water quality event. 2
5 Sample container transit bag not sealed or evidence of tampering   3
6 Open evidence bag, incorrectly sealed evidence bag or evidence of tampering with Evidence Bag after sealing Up to receipt at laboratory 3
7 Evidence of tampering with sample after receipt at laboratory   3
8 Unauthorised breaking of security devices on sampling lines and units   3
9 Failure to record correctly authorised security device changes   3
10 Sampling cabinet found unlocked or otherwise in an insecure condition   3
11 Incorrect white tag fitted to sampling unit door Report incorrect recording of number on evidence bag by issuing laboratory as code 17 3
12 Key to sampling cabinet or key safe lost or stored insecurely   3
13 Any gap in the chain of evidence or failure of security not covered above Including missing signature, or correction not initialled, timed or dated 3
14 Sample not analysed within specified timescale or result not reported within specified timescale 3 days for routine samples
1 day for express analysis		 4
15 Volume filtered less than 40 litres per hour over sampling period Only to be used when actual average rate of flow is less than 40 l/hr 4
16 Deviations from approved sampling procedure   4
17 Incorrect recording of filter unit numbers or other information on the evidence bag by the laboratory or failure of laboratory to record information Include tag numbers and insecure green tag 4
18 Deviations from approved analysis procedure Not covered by codes 3, 7, 13, 14 or 17 above. 4
19 Sampling information missing from evidence bag other than chain of evidence records. Use code 17 for laboratory omissions.
Use code 13 for missing signature or initials or continuity block not completed.		 4
20 Sampling information incorrectly recorded on evidence bag other than chain of evidence records If no correct written contemporaneous record exists. Use code 17 for laboratory errors.
Use code 13 for chain of evidence information.		 4
21 Minor errors or inconsistencies in recording sampling information on the evidence bag other than chain of evidence records For use only if a contemporaneous written record of the correct information exists (eg sampler’s log). If information inferred or deduced from other records eg preceding or subsequent evidence bags use code 20.
Use code 17 for laboratory errors.
Use code 13 for chain of evidence information.		 4
22 Any contravention of part 1 of the protocol (sampling) not included in this list   4
23 Any contravention of part 2 of the protocol (analysis) not included in this list Also parts 2a and 2b if appropriate 4

Each event must be recorded as a separate anomaly.
Eg if an analysts drops a tray of slides, this is one anomaly affecting a number of samples.
If a sampler fails to sign all the evidence bags on a particular day, each failure is a separate event and an anomaly reported for each sample affected.
If less than 40 l/hr is filtered at a site for 2 or 3 days while a replacement part is obtained and fitted, this is one anomaly affecting a number of samples.

Appendix E

Comparison of current anomalies list with the proposed new list

IL 28/2000 Anomaly code
Proposed new code
01
01
02
02 , 03 and 04
03
05
04
06 and 07
05
08
06
09 and 17
07
10
08
11 and 13
09
13
10
15
11
02 or 04 (only if sampling interrupted)
12
02 or 04 (only if sampling interrupted)
13
None
14
None
15
None
16
None
17
02 or 04
18
04
19
14, 16, 17 and 18
20
12
21
17
22
17 (broken in lab)
16 (broken by sampler)
13 (broken while fitted to sampling device and not done by sampler)
23
17
24
17 (by laboratory)
19 and 20 (by sampler)
13 (if chain of evidence broken)
25
22 and 23
26
None
27
21
28
17
29
22
30
None
31
None
32
None
33
None
34
None
35
None

A - Z Index | Business and Technical | Consumer Information | Contacting us |
| DWI Home Page | News | Product Approval | Search our site | Wales |

Drinking Water Inspectorate,
Floor 2/A1, Ashdown House, 123 Victoria Street, London, SW1E 6DE
Telephone : 020 7944 5956 - Facsimile : 020 7944 5969
E-mail: dwi_enquiries@detr.gov.uk

Published 19 June 2001 Updated 11 July 2001
Return to Current Information Letters Index
Return to Information Letters Index
Return to Drinking Water Inspectorate Home Page
Department for Environment, Food & Rural Affairs (DEFRA)Home Page
The National Assembly for Wales / Cynulliad Cenedlaethol Cymru Home Page
Web site terms
© Crown copyright 2001