Internal Audit Report Waiting Times November 2012
Internal Audit Report Waiting Times November 2012 1 Introduction... 1 2 Background... 1 3 Audit Approach... 2 4 Summary of Findings... 3 5 Executive Summary... 4 6 Management Action Plan... 6
1 Introduction Between September and November 2012 we carried out a review of waiting times at. The review was directed by the Scottish Government Health and Social Care Directorates (SGHSCD) in response to a request by the Cabinet Secretary for Health and Well-being for all relevant health boards in Scotland to audit local waiting times management arrangements. 2 Background Patient waiting times is a key indicator of the accessibility of services and the quality of patient care. Reduced waiting times can result in earlier diagnosis and better outcomes for patients and can also improve the patient journey by reducing uncertainty and unnecessary worry. The Scottish Government introduced a HEAT target to reduce waiting times across NHS Scotland. The overall national target is that, from December 2011, at least 90% of patients should receive treatment within 18 weeks of being referred (known at the Referral To Treatment time RTT). There are also related targets within this area, for example 12 week outpatient and 9 week admission "Stage of Treatment" targets. As at December 2011, 92% of patients in Scotland whose journey could be fully measured were treated within 18 weeks, compared to 82% in January 2011, an increase of around 12,000 patients. has exceeded the HEAT target and consistently performed around or above the national average in the nine months to June 2012. Month Patient Journeys within RTT target (%) NHS Highland Patient Journeys within RTT target (%) Scotland Oct 2011 92.7 89.5 Nov 2011 92.2 89.6 Dec 2011 93.6 92.0 Jan 2012 92.8 91.4 Feb 2012 91.9 91.4 Mar 2012 94.9 91.5 Apr 2012 95.2 92.0 May 2012 94.6 92.4 June 2012 95.5 92.4 To support consistency and fairness in managing and reporting on patient waiting times across NHS Scotland, SGHSCD introduced the New Ways approach in January 2008. The waiting time clock was introduced which records the time between referral and treatment or outpatient admission, excluding periods when the patient is unavailable. To ensure that all patients are being managed in line with the guidance, health boards are required to record data clearly and accurately. All stages of the patient s journey must be recorded to calculate the waiting time, with using the isoft (North Highland area) and Helix (Argyll & Bute area) patient management systems for this purpose. The reason for having two different systems relates to the Board taking on responsibility in 2006 for part of the area previous covered by Argyll and Clyde health board. Waiting times - national context The management of waiting times has come under national scrutiny following the publication of a report on waiting times in NHS Lothian. The main findings of the report include observations of: Excessive and inappropriate use of periods of patient unavailability; Manual adjustments to those patients who were breaching waiting times before reporting to more senior management levels; A practice of don t minute or record, which prevented full details of waiting times issues progressing up the operational framework where a more strategic and collective approach could have been November 2012 1
taken; and An encouragement to local operational staff to resolve issues through adjustments of waiting times figures rather than actually resolving delays. As a result of these observations the Cabinet Secretary for Health and Well-being requested all relevant NHS boards to commission an internal audit review of local waiting times arrangements. These findings are to be reported to SGHSCD. 3 Audit Approach The scope of this review was set by SGHSCD in a letter to NHS boards on 3 May 2012 seeking audit assurance in relation to the specific objectives listed in Section 4 of this report. We have completed our work in accordance with those terms of reference. This review was supported by an analysis of waiting list information from the electronic patient management system and covered the two quarters ending March 2012 and June 2012. The extraction and analysis of this data was to be performed for all 14 territorial health boards and the National Waiting Times Centre Board by a single third party contractor appointed by SGHSCD, PwC. In line with the terms of that appointment, we used the output we received from the analysis to inform the nature and extent of our audit work and associated sample testing. However, the vast majority of the data analysis was unable to be performed in the originally planned way as relevant data was not extracted successfully from the North Highland patient management system, which covers the North Highland area. PwC spent time from the date of their appointment looking to understand and analyse the systems and data. They concluded that all data queries for the isoft system could not be run. Accordingly, in mid October 2012, approximately 20% of the planned data queries were provided to us by PwC. Mindful of reporting deadlines, we developed and implemented our own contingency arrangements to deal with these data extraction issues in North Highland. This enabled us to analyse and select data for testing in an alternative way. This led to an extensive sample population being tested, including focussing on unavailability/suspensions, journey times around breach dates and review of entire patient journeys. This end-to-end approach gives greater insight into the entire patient journey compared to more aspect-specific queries which would have resulted if the data analysis had been successful in North Highland. However, this alternative approach could not initially be as directive and exception-based without the full PwC data. A similar data analysis and extraction problem was encountered at NHS Greater Glasgow and Clyde, which also uses the isoft patient management system. We liaised closely with the internal audit team responsible for that waiting times review to ensure consistency of approach and confirm a robust audit methodology was employed across both audits. We have also played an active part in national discussions involving NHS internal auditors across Scotland, including representation from SGHSCD and coordination with Audit Scotland, to promote a rigorous and consistent approach to this audit across all boards. The following is a summary of the waiting times key milestones since the publication of the initial internal report on NHS Lothian s waiting times arrangements, through to our final reporting to : Action Dates: Argyll & Bute Dates: North Highland 1. Publication of NHS Lothian internal report on waiting times 9 Jan 2012 2. Publication of PwC NHS Lothian Report 19 Mar 2012 3. Cabinet Secretary for Health and Well-being response to PwC report 4. SGHSCD internal audit of Waiting Times Terms of Reference to NHS Boards 31 Mar 2012 3 May 2012 November 2012 2
5. Data analysis - tendering process completed by SGHSCD (PwC appointed) 16 July 2012 6. Start of fieldwork on governance, reporting and local guidance 5 Sept 2012 7. Data pack received from PwC 13 Sept 2012 12 Oct 2012* 8. Selected sample information from data pack received 19 Sept 2012 16 Oct 2012* 9. Data testing commenced 28 Sept 2012 23 Oct 2012* 10. Data testing completed (fieldwork completed) 14 November 2012 11. Draft report issued for formal management responses 19 November 2012 12. Final report presented to Audit Committee 4 December 2012 13. Final report issued to Scottish Government 17 December 2012 *The substantial delay to the North Highland data extraction and analysis explains the respective dates when we were able to undertake our work. Our detailed findings are discussed in the Management Action Plan (section 6) of this report. 4 Summary of Findings The table below summarises our assessment in relation to each of the audit objectives specified by SGHSCD. All control weaknesses identified are included in the Management Action Plan in section 6. No. Audit Objective Assessment Grading & no. of agreed actions 5 4 3 2 1 1 2 3 Individual patient records are accurate and systems are in place to ensure that the patient administration system cannot be inappropriately changed. Reporting on the target waiting times is accurate and consistent at every level of the organisation up to and including the Board. The local guidance is consistent with national guidance and its implementation is both valid and reliable (i.e. not open to different interpretation in use). RED 1 3 YELLOW 1 1 YELLOW 1 TOTAL NUMBER OF AGREED ACTIONS - 1 5 1 - Assessment BLACK RED YELLOW GREEN Definition Fundamental absence or failure of key control procedures - immediate action required. The control procedures in place are not effective - inadequate management of key risks. No major weaknesses in control but scope for improvement. Adequate and effective controls which are operating satisfactorily. November 2012 3
5 Executive Summary Conclusion Our findings do not indicate the existence of systematic and deliberate mis-recording or mis-reporting of waiting times which would materially impact on achievement of waiting times targets. However, in a small number of cases tested we identified inconsistencies in the implementation of the waiting times guidance which resulted in the avoidance of a breach against the 18 week RTT target, when correct implementation of the guidance would have resulted in a breach. From wider analysis we have performed and extended sample testing undertaken (including focussing on suspensions applied to journeys near to breaching) this would not appear to have a high risk of a material impact on reporting against waiting times targets. Main findings The Board takes the management of waiting times seriously, with regular reporting and scrutiny at Board and Committee levels. This is coupled with detailed analysis by senior management and operational groups to scrutinise the detail and trends within this complex area. We performed extensive testing over patient records and waiting times recording processes, in addition to reviewing the reporting and governance of those processes. This involved testing some 200 individual patient records, across both North Highland and Argyll & Bute. In North Highland we tested 133 patient journeys from referral to treatment (RTT), to ensure that the patient journey was recorded in line with the New Ways guidance and the Waiting Times Recording Manual. This testing identified 6 potential issues. These issues may relate to the extent and method of recording of these specific cases, within a system which management have advised is having to be used in a way not originally anticipated. We identified 4 cases where periods of social unavailability had been applied. However, in these 4 instances the narrative to support the period of unavailability did not support the dates that had been recorded on the patient management system, including one case where an additional 26 days of unavailability had been applied, apparently without sufficient justification. In all these cases, the unavailability avoided a breach against RTT target, although these were reported appropriately against stage of treatment targets. We also identified 2 instances where the patient had a period of unavailability applied because they had not responded to a patient-focussed booking (PFB) contact. In these cases a period of unavailability was applied for 14 days, providing 11 days more unavailability than we expected. This was because the unavailability was applied to the initial tranche of the PFB (ie between initial letter and the reminder letter) rather than from the date of the reminder letter until a response was received from the patient. In both these cases the unavailability avoided a breach against RTT target, although we identified no issues in relation to reporting against stage of treatment targets. Given our wider analysis and extended testing, these cases do not appear likely to have contributed to a material impact on reporting against waiting times targets. This is because we tested a large number of unavailabilities, including those which were applied around the RTT target time. This enabled us to assess the materiality of these issues in context. Our work included cross referencing unavailability against RTT times using computer analysis techniques, to help specifically target this extended testing. Our targeted patient record testing in Argyll and Bute did not lead to any reportable exceptions. In the Management Action Plan we have identified a number of further areas which, if addressed, would strengthen the organisation s waiting times processes. These include: When staff apply a period of patient unavailability (i.e. a patient is suspended from the waiting list) they often do not record any or sufficient detail of why the unavailability is being applied. The Waiting Times Recording Manual states that an explanation to support the period of unavailability should be recorded. This occurred in around one third of the sample tested, including cases where the RTT November 2012 4
target was nowhere near breached. Audit trail capabilities in place within the North Highland patient management system are not sufficiently robust. During testing we sought to analyse a sample of unavailability periods to establish when they were applied. However, we found that the audit trail within the system was often incomplete with the system regularly auto-erasing historic adjustments. We appreciate the need for pragmatism and costefficiency in dealing with systems, but sufficient audit trail is vital in such a high profile area. System access privileges should be reviewed to ensure that access is commensurate with user roles and responsibilities. The local access policy needs further review and update against revised national guidance and regulations. Patient and wider stakeholder input could support this process. These issues are discussed further in the Management Action Plan at Section 6 of this report. Basis of opinion In giving our opinion it should be noted that assurance can never be absolute. The most that the internal audit service can provide is reasonable assurance that there are no significant weaknesses in the areas within the scope of this review. In particular, the absence of clear and evidenced audit trails and issues with the specific cases highlighted above prevents us being able to provide fuller assurance. Acknowledgements We would like to thank all staff consulted during this review for their assistance and co-operation. November 2012 5
6 Management Action Plan Grading of recommendations Our grading structure helps management assess the significance of the issues raised and prioritise the action required to address them. The grading structure is as follows: Grade Definition 5 Very high risk exposure - Major concerns requiring immediate Board attention. 4 High risk exposure - Absence / failure of significant key controls. 3 Moderate risk exposure - Not all key control procedures are working effectively. 2 Limited risk exposure - Minor control procedures are not in place / not working effectively. 1 Efficiency / housekeeping point. November 2012 6
Management Action Plan 1 Key control objective: Individual patient records are accurate and systems are in place to ensure that the patient administration system cannot be inappropriately changed. Observation and Risk Recommendation Management Response 1.1 Waiting times In North Highland we tested 133 patient journeys to ensure they had been recorded in line with the New Ways guidance and Waiting Times Recording Manual. This identified 4 cases where a period of social unavailability was applied but the narrative to support the unavailability did not agree with the data recorded on the patient management system. The impact of this excess unavailability was that the patient journey did not breach the RTT deadline. We also identified 2 instances where the patient had a period of unavailability applied because they had not responded to a patient-focussed booking (PFB) contact. In these cases a period of unavailability was applied for 14 days, allowing 11 days more unavailability than we would otherwise expect. Given our wider analysis and extended testing (as explained further in the Executive Summary, above), there does not appear to be a high risk of a material impact on reporting against waiting times targets. However, these are still notable findings in themselves. We cannot give absolute assurance whether these specific cases were deliberate manipulations, due to the lack of a sufficient audit trail within the North Highland patient management system and related records. All staff should be clearly instructed to record waiting times accurately and with full reference to relevant back up. Further controls should also be put in place; for example, we suggest management should design an exception report to identify all cases which would have breached had a period of unavailability not been applied. These cases should be double checked and signed off by an independent officer in full in the short term, and potentially on a sample basis thereafter (dependent on the outcome of the checking in the more immediate term). The results of this work should be reported to relevant management for assurance. Whilst awaiting further national guidance we are taking a range of management actions: 1. Standard Operating Procedure has been developed across Northern to ensure that User IDs are recorded, start and end date to be agreed at PAS Users Group in December 2012. 2. has a report on the Intranet detailing all patients who have a period of unavailability over 10 weeks which is reviewed at Operational Level. 3. has developed an Experienced Wait Report that identifies patients who would have breached had a period of unavailability not been applied. The report will be tested from 1 st December at Raigmore Hospital with the intention of implementing this report across the whole of Highland 4. A report will be developed to identify PFB non responders that highlights any patient with unavailability greater than 7 days. We will make any further changes required once we receive the awaited national guidance for management of nonresponders, along with refresher training for all PFB staff to confirm compliance with the policy. To be actioned by: Chief Operating Officer No later than: January 2013 (impact of national guidance will depend on published date.) Grade 4 November 2012 7
Management Action Plan Observation and Risk Recommendation Management Response 1.2 Periods of unavailability (suspensions) Where a patient is unable to undergo treatment due to either medical or social reasons the patient should be recorded as unavailable (effectively suspended from the waiting list). This should only be applied if legitimate medical or social reasons are given, such as the patient going on holiday. The periods of unavailability are deducted from the patient s total waiting time to give a true reflection of the actual waiting time. We performed detailed testing over the creation of periods of unavailability to ensure adequate evidence was in place to support the suspensions and they were in line with the New Ways guidance. This identified some one-third of cases where there was not sufficient evidence to explain why the period of unavailability had been created or amended. There is a risk that periods of unavailability are being created or amended without appropriate patient contact/audit trail. This is non-compliant with New Ways guidance. Staff should be reminded that they should record sufficient and appropriate details of why the patient is unavailable, to support any recorded period of unavailability. Standard Operating Procedure has been developed across Northern to ensure that User IDs are recorded, start and end date to be agreed at PAS Users Group in December 2012. In the meantime the operational staff will continue to be reminded of the need to record unavailability details through the regular waiting times management meetings. To be actioned by: Chief Operating Officer No later than: January 2013 Grade 3 November 2012 8
Management Action Plan Observation and Risk Recommendation Management Response 1.3 North Highland PMS audit trail/timestamps To gain further assurance on a number of patient journeys we sought to analyse the audit trail and timestamps within the North Highland patient management system to determine when each adjustment was made and by whom. However, in all but one case we found that the audit trails and timestamps were incomplete as the system autoerases historic adjustments, given the limitations of the system. There is a risk that inappropriate changes are made to the patient journey and these are not recorded appropriately. Management will not be able to identify, investigate and resolve any potential issues or misuse of the system. The Board should liaise with the software provider to ascertain if the audit trails can be improved. Ideally the auto-erase of historic actions should be removed (or at least curtailed). This will allow management to conduct their own internal investigations, as and when appropriate, to satisfy themselves that patient journeys are being treated correctly. The existence of this trail would also act as a deterrent, to discourage anyone from amending records inappropriately. (This issue may be superseded, dependent on action taken and timing to address Action Plan Point 2.3, below) Board are considering a business case in December 2012 for the introduction of the national PMS system. Discussions with Isoft confirmed the amendments could not be made within the timescales. Subjected to Board approval the implementation of the new PMS system will commence in April 2013 To be actioned by: Head of ehealth No later than: 1 st April 2013 Grade 3 November 2012 9
Management Action Plan 1.4 System access Observation and Risk Recommendation Management Response Within the patient management systems there are multiple user access privileges that have been created over a number of years. These privileges are used to grant access to certain parts of the system and are generally granted based on the individual s job role. For example, if an existing Helix user has certain privileges, a new Helix user with the same job role will also be granted those privileges. The detail of access allowed by each privilege is not always clear and there is a risk that staff may be granted access to parts of the system that may not be aligned to their specific role or responsibility. We recommend that management reviews the list of privileges to ensure that the list is current, relevant and access is commensurate with user roles and responsibilities. 1. A review of the Privileges in existence across has identified a significant piece of work to take forward the recommendation. Consideration is being given to whether this is beneficial given the likely timescale for the introduction of a new PMS System. 2. Work is underway to establish the SOPs that are required to be undertaken by each Job Profile, and then identify the PAS user profile that is required for each job family. To be actioned by: Head of ehealth No later than: October 2013 Grade 3 November 2012 10
Management Action Plan 2 Key control objective: Reporting on waiting times is accurate and consistent at every level in the organisation up to and including the Board Observation and Risk Recommendation Management Response 2.1 Waiting times system integration operates two separate patient management systems: Helix in Argyll & Bute and isoft in North Highland. The reason for having two different systems relates to the Board taking on responsibility in 2006 for part of the area previous covered by Argyll and Clyde health board. As well as presenting the Board with dual training/ recording/systems management costs, waiting times data for both areas needs to be merged in order to submit the Monthly Management Information (MMI) spreadsheet to SGHSCD and to inform Board-wide reporting. Whilst there are quality control checks in place over the the merging and publication process which help safeguard against errors, there remains an ongoing risk that the data may not be consolidated correctly and reports an inaccurate waiting times position. Further, there are wider cost/overhead and operational implications of continuing to run two separate systems. We are aware that initial work has been done to look at this area within. We believe the Board should take this initial work further and prepare a formal, detailed business case which looks at all qualitative and quantitative aspects of future patient management system provision. From our initial assessment, it would seem sensible to hold an initial preference for one, unified system for the whole Board area to resolve the current situation (which was the product of external circumstance). A unified approach would remove current duplication and reduce consolidation/data risks. System update/integration may also help resolve some of the issues the Board is experiencing with the North Highland patient management system. Board meeting in December 2012 is considering a Business Case to replace the PAS systems across the Board. To be actioned by: Head of ehealth No later than: December 2012 Grade 3 November 2012 11
Management Action Plan Observation and Risk Recommendation Management Response 2.2 Waiting times data available to consultants We interviewed consultants, including to discuss their interaction with the management on waiting times within their respective specialities. A common issue was that they felt patients could benefit from consultants getting earlier warning of potential breaches, including from access to more detailed reports on waiting times data. Consultants felt this could enable them to proactively plan additional capacity (if needed) and to review the reasons for referral and outcomes. Making information for consultants as tailored and user-friendly as possible could minimise reactive action and lead to a more responsive and integrated approach. We recognise that that is progressing with significant additional and improved reporting from their PMS system through the Highland Information Portal. The discussions within ehealth to better engage consultants and establish a forum to discuss, determine and agree the data needs of the clinicians should be taken further. It may also be useful to analyse the needs and wants of other key stakeholder groups to consider the cost/benefit (and appetite) for other, tailored reports. Consultants currently have ready access to patient level information via their individual service managers whose role it is to ensure service planning and delivery of waiting time targets through detailed discussions with clinicians. However to compliment this information and to respond to our discussions with consultants on how the information for each patient could be presented, work has commenced with the Urology specialty to design patient pathway information similar to the reports that have been designed for use by NHS Western Isles for patients treated in. To be actioned by: Head of ehealth No later than: 1 st April 2013 Grade 2 November 2012 12
Management Action Plan 3 Key control objective: The local guidance is consistent with national guidance and its implementation is both valid and reliable (i.e. not open to different interpretation in use). Observation and Risk Recommendation Management Response 3.1 Managing Access for Patients policy At time of writing, the Managing Access for Patients policy was published in November 2009 and has not been reviewed or updated since that time. This is the main, overarching policy has in place for managing patient waiting times. A paper was presented to s Board in April 2011 that recommended updates to the policy. These do not appear to have been completed and approved. Without an up-to-date policy, staff and patients may not be fully aware of the processes in place and respective roles/responsibilities. In addition, an out-ofdate policy may result in not complying with the NHS Scotland National Access Policy and The Patient Rights (Treatment Time Guarantee) (Scotland) Regulations 2012. should review and update the Managing Access for Patients policy against updated NHS Scotland Treatment Time Guidance and the NHS Scotland National Access Policy. In line with the recommendations in the National Access Policy, the Board s local policy should be developed with appropriate patient participation and published on the Board s website. Once published, a training and awareness programme for all staff involved in waiting times should be developed. We agree with the Recommendation and NHS Scotland s Central Legal Office approval of the business processes will shape the Managing Access for Patients Policy. The existing policy is available on NHSH website To be auctioned by: Chief Operating Officer No later than: 31 st March 2013 Grade 3 November 2012 13
(www.scott-moncrieff.com), one of Scotland s leading independent professional services firms, provides industry-focused audit, tax, business advisory and corporate consulting services for commercial, public, not-for-profit and private clients. Chartered Accountants 2012. All rights reserved. refers to Chartered Accountants, a member of Moore Stephens International Limited, a worldwide network of independent firms. Chartered Accountants is registered to carry on audit work and regulated for a range of investment business activities by the Institute of Chartered Accountants of Scotland.