52003SA0003

Special Report No 3/2003

on the invalidity pensions scheme of the European institutions, together with the institutions' replies

(pursuant to Article 248(4), second subparagraph, of the EC treaty)

(2003/C 109/01)

CONTENTS

>TABLE>

SUMMARY

I. The Court carried out an audit of the Invalidity Pensions Scheme in order to assess the cost of invalidity pensions, identify potential savings, determine whether invalidity pensions are only granted where a real invalidity has been duly recognised, and evaluate whether the institutions have set up the management systems required for adequate monitoring of, and effective control over, the operation of the scheme.

II. There are approximately 200 invalidity pensions awarded each year representing a total annual net actuarial cost of the order of 74 million euro, with recent increases in annual expenditure being attributable, in the main, to increases in the numbers and age of the workforce at the institutions. However, the modifications to the Staff Regulations as part of the current reform package are likely to reduce the cost of invalidity pensions.

III. The Court's audit revealed a complex picture. On the one hand the rate of invalidity retirement has remained stable over the last 15 years; and, in the opinion of the Court's medical adviser, invalidity pensions are awarded in a justified way. On the other hand, retirement on invalidity grounds is more common in some grades than normal retirement, and there is evidence that frustration in the working environment affects some staff who are eventually retired on ill-health grounds.

IV. Shortcomings were found concerning the overall policy and management systems for absences due to illness, characterised by a medical rather than administrative approach, a lack of clarity and coordination between the various departments involved, a lack of resources, unsuitable IT management tools, and obstacles resulting from certain provisions of the Staff Regulations.

V. The result is frequent and costly delays in the opening and progress of the invalidity procedure with the length of the process itself associated with deteriorating health and consequently with extremely low rates of reinstatement, especially in the 50 % of cases involving psychological disorders.

VI. The audit found scope for financial savings through the adoption of adequate administrative measures for prevention and early treatment, particularly in cases where the grounds for invalidity are psychological. Such measures should include the development by institutions of an overall policy on absences due to illness, clearer allocation of roles and responsibilities between the medical services and personnel departments, earlier and more intensive medical examinations, implementing management information systems capable of identifying patterns of absence and risk factors associated with eventual invalidity, and modifications to the Staff Regulations.

VII. In particular, modifications to the provisions of the Staff Regulations should be considered to deal with situations where a member of staff impedes or deliberately evades a medical examination, to provide for an easier and swifter appeals and arbitration mechanism, to lower the threshold period of sick leave required before the institution can start the invalidity procedure, to lay down conditions under which medical examinations after retirement should be performed, and to modify the composition of the Invalidity Committee to create a more permanent and inter-institutional structure.

AUDIT OBJECTIVES AND APPROACH

1. The general objective of the audit was to evaluate the European institutions' policies and practices in respect of the Invalidity Pensions Scheme. The audit aimed to assess the cost of invalidity pensions, to identify potential savings and to ascertain whether the institutions have set up the management systems required for adequate monitoring of, and effective control over, the operation of the scheme. One of the audit specific objectives was to ensure that invalidity pensions are only granted where a real invalidity has been duly recognised.

2. The audit consisted of a series of analytical procedures designed to reveal any risks of abuse of the scheme. It included an evaluation of the internal control system relating to absences due to illness, focusing on the issues connected with the medical and administrative monitoring of long-term absences due to illness, which usually culminate in retirement on an invalidity pension. Substantive tests were carried out on a sample of administrative files on invalidity pensions. The files were selected randomly, but partly also within the framework of an approach which focused on a section of the population that presented possible higher risks of being granted invalidity pensions for which there is no justification(1). This resulted in the selection, in the great majority of cases, of files connected with psychological illnesses, which are assumed to be more difficult to evaluate. A questionnaire was also sent by the Court to a sample of beneficiaries concerned(2). In addition, the Court commissioned a medical expert (hereinafter referred to as the Court's medical expert) to carry out an examination of half the medical files concerned in order to establish whether the opinions of the Invalidity Committees were based on sufficient and adequate work and whether retirement due to invalidity appeared justified on medical grounds(3). The audit work focused on the invalidity pensions awarded in the period from 1996 to 2000. For comparison purposes, the Court has visited five Member states(4), reflecting a wide range of administrative traditions present within the Community, as well as a number of international organisations(5), which face challenges similar to those of the institutions. Information was gathered on the basis of interviews with representatives of these Members States and organisations.

THE INVALIDITY PENSIONS SCHEME

Rules

3. The legal basis for invalidity pensions for officials and temporary members of staff is to be found in Article 78 of the Staff Regulations of officials of the European Communities and in Article 33 of the Conditions of employment of other servants of the European Communities(6). Officials or other members of staff are entitled to an invalidity pension in the case of total permanent invalidity preventing them from performing the duties corresponding to a post in their career bracket.

4. The invalidity retirement procedure is initiated either by the institution, when a member of staff has been absent due to illness for more than 12 months in any three-year period (Article 59(1) of the Staff Regulations), or at the request of the member of staff concerned.

5. An Invalidity Committee is set up (Article 9(1)(b) of the Staff Regulations). It is called upon to decide whether the member of staff is suffering from invalidity within the meaning of the Staff Regulations (Article 13 of Annex VIII to the Staff Regulations). The Committee consists of three doctors (Article 7 of Annex II to the Staff Regulations):

(a) one appointed by the institution;

(b) one appointed by the member of staff;

(c) one appointed by agreement between the first two doctors.

6. If no agreement can be reached on the choice of the third doctor, he/she is automatically appointed by the President of the Court of Justice. The Invalidity Committee delivers a unanimous opinion or, failing this, a majority opinion.

7. If the Invalidity Committee confirms the member of staff's invalidity, he is automatically retired on the last day of the month in which the Appointing Authority recognises his permanent incapacity to perform his duties (Article 53 of the Staff Regulations). The right to receive payment of invalidity pension has effect from the first day of the calendar month following the decision on his retirement (Article 14 of Annex VIII to the Staff Regulations).

8. While a former member of staff drawing invalidity pension is less than 60 years of age, the institution may have him medically examined periodically to ascertain that he still satisfies the requirements for payment of the pension (Article 15 of Annex VIII to the Staff Regulations). Where a former member of staff who has been drawing invalidity pension is reinstated, the time during which he received invalidity pension is included for the purpose of calculating his retirement pension, without payment by him of arrears of contributions (Article 16 of Annex VIII to the Staff Regulations).

9. The invalidity pension is equal to the retirement pension to which the member of staff would have been entitled at 65 years of age if he had remained in the service until that age. Where the invalidity arises from an accident in the course of or in connection with the performance of his duties, from an occupational disease, from a public-spirited act or from risking his life to save another human being, the invalidity pension is 70 % of his basic salary. The invalidity pension is calculated by reference to the basic salary that the member of staff would have received in his grade if he had still been in the service at the time of payment of the pension (which means that a member of staff who has retired due to invalidity has the benefit of increases in step within his grade). The invalidity pension may not be less than 120 % of the minimum subsistence figure, an amount equivalent to the basic salary of an official in grade D 4, step one (Article 78 of the Staff Regulations). This implies that from grade D 4 up to grade C 2 fifth step, grade C 1 second step, grade B 5 and grade B 4 second step, the maximum amount of a retirement pension (70 % of the basic salary) is less than the minimum amount of an invalidity pension. According to the proposal for amended Staff Regulations submitted by the Commission to the Council in April 2002, the minimum amount of the invalidity pension will be reduced to 100 % of the minimum subsistence figure.

Statistics and analyses

10. Appropriations intended to cover the invalidity pensions of officials and other staff of all the institutions are entered under the Commission's budget item A-1 9 0 1. Table 1 retraces budgetary expenditure during the period from 1985 to 2000. In 2000 the payments made amounted to 123,7 million euro. At constant prices, annual expenditure has almost tripled during this period. As this table and Diagram 1 show, this expenditure has increased considerably less than expenditure on retirement pensions, which has increased almost sixfold.

Table 1

Budgetary expenditure on invalidity pensions and retirement pensions for all the institutions

>TABLE>

Source:

Revenue and expenditure account, Volume II, Section III, Commission.

Diagram 1

Comparison of expenditure on retirement pensions and invalidity pensions between 1985 and 2000 (payments in constant prices)

>PIC FILE= "C_2003109EN.000601.TIF">

11. Over the 1985 to 2001 period, the number of retirements due to invalidity averages 166 per year (see Table 2). In 1986 and 1987 the figure is markedly higher by comparison with the preceding period. The number fell sharply in 1988, followed by moderate fluctuations until 1997. The threshold of 200 retirements per year is passed in 1998, with a strong upward trend thereafter. In 2001, 226 persons retired due to invalidity, that is 0,78 % of the staff employed at 31 December 2000. Diagram 2 shows (at

Table 2

Number of retirements due to invalidity during the 1985-2001 period

>TABLE>

Note:

The figures are based on the date on which the invalidity retirement decision took effect.

Source:

The Commission's Sysper database.

Diagram 2

Evolution of the invalidity rate(7) at the Commission

>PIC FILE= "C_2003109EN.000701.TIF">

>TABLE>

Source:

The Commission's Sysper database.

intervals of five years) that the invalidity rate at the Commission(8) fell between 1985 and 1995, but had risen again by 2000, with a slight decrease in 2001.

12. The variation in retirements due to invalidity over the whole period does not, in itself, reflect an abnormal trend, since, as the analysis carried out in respect of the Commission shows, it seems to be mainly attributable to the increase in the number of staff and age of staff. As Table 3 shows, during the period from 1970 to 2000 the number of staff tripled and the average age of staff rose from approximately 40 to 45. While the percentage of staff aged over 50 was only 13,7 % in 1970, it was 28,6 % in 2000. As Table 4 reveals, the rate of exposure to the risk of invalidity increases significantly for staff aged over 50, who account for more than three quarters of retirements due to invalidity.

Table 3

Changes in the age structure of the Commission's staff

>TABLE>

Source:

The Commission's Sysper database.

Table 4

Retirements due to invalidity by age group (%) 1968 to 2000

>TABLE>

Source:

The Commission's Sysper database.

13. As the Court has already highlighted in its Annual Report concerning the financial year 1987, the majority of those retired due to invalidity are in categories C and D, who have the characteristics of, firstly, retiring more frequently because of invalidity than staff in other categories and, secondly, of leaving more frequently on an invalidity pension than on a retirement pension. In fact, as Table 5 shows, at the end of 2000 categories C and D accounted for 62 % of beneficiaries of invalidity pensions but only 21,7 % of beneficiaries of retirement pensions. Likewise, pensions paid in conformity with the provisions concerning the

Table 5

Distribution by grade of the number of beneficiaries of invalidity pensions and retirement pensions between the end of 1987 and the end of 2000

>TABLE>

Note:

The number of beneficiaries = the number of staff retired - deaths - staff returning to work.

Source:

The Commission's Sysper database.

minimum subsistence amount still account for approximately one third of all invalidity pensions and are still paid practically in full, i.e. at the rate of 97 %, to category C and D staff(9). Grades C 3 to D 4 receive this systematically.

14. Approximately two thirds of the Invalidity Committees are convened on the initiative of the institution when the member of staff has accumulated more than 12 months' absence due to illness over a three-year period, one-third at the member of staff's request. In more than 90 % of cases the procedure ends in retirement due to invalidity. Psychological disorders account for approximately half the medical causes of invalidity. The other main causes are osteo-muscular (approximately 20 %) and cardio-vascular (approximately 15 %) ailments.

Psychological illnesses are also the major cause of invalidity (30 % to 60 % of the cases of invalidity) in national administrations and international organisations(10).

THE COURT'S PREVIOUS WORK AND THE FOLLOW-UP TO THIS

The Court's previous work

15. In its Annual Report concerning the financial year 1987(11), the Court highlighted a number of facts that might give rise to concern that the system, which is far from restrictive and predominantly social in nature, may have certain adverse effects, for example:

(a) a significant increase in retirements due to invalidity over the period 1983 to 1987;

(b) a high incidence of invalidity pensions for category C and D staff, for the vast majority of whom the invalidity pension is higher than a retirement pension at the maximum rate of 70 % (application of the minimum subsistence figure rule laid down in the Staff Regulations).

Follow-up

Recommendations of the "Caston Report"

16. In response to the Court's work and at the request of the heads of administration, an ad hoc interinstitutional committee was set up and submitted its report in 1989 (dubbed the Caston Report after the chairman of the committee) to the Committee on Budgetary Control. The report recommended that a combination of amendments to the Staff Regulations and administrative measures should be adopted to prevent abuse of the system and to reduce the financial attractiveness of invalidity pensions. The report also recommended early detection and treatment by appropriate measures of problems relating to career, mobility, relations with hierarchical superiors or conduct that are liable to aggravate a member of staff's health problems.

17. The interinstitutional committee noted a lack of valid statistics and checks with respect to absences due to illness. It suggested improving the coordination of medical and administrative measures in order to identify, at an early stage, persons who might retire due to invalidity. It also emphasised that, in some cases, there was a need for swift administrative measures in order to avoid situations where, at the end of the day, an invalidity pension is the easy solution.

Action taken

18. The report was warmly welcomed by the heads of administration, who forwarded the conclusions of the report to the Commission so that it could take the necessary action with regard to measures which involve a reform of the Staff Regulations, with the administration of each institution having to take the appropriate administrative measures.

19. The Commission drew up proposals for amendments to the Staff Regulations in respect of invalidity pensions. These met with the opposition of the trade union and staff organisations. In 1991, the Commission withdrew the proposals on the grounds that it had come to the conclusion that the arrangements for invalidity pensions as provided for in the Staff Regulations were sound and that it would be more appropriate to intensify checks and penalise any abuses that were detected. As part of the current proposal for a reform of the Staff Regulations, i.e. ten years later, the Commission is proposing a series of amendments to the Staff Regulations that are very similar to those envisaged at that time and then withdrawn. Concerning the administrative measures, as explained in this report, these have not been taken or remain broadly insufficient, and at present there are no plans to take any significant action in this area.

OBSERVATIONS ON THE ABSENCE MANAGEMENT SYSTEM IN RESPECT OF ABSENCES DUE TO ILLNESS

Absence rates

20. Absences have a considerable financial impact. According to the information submitted to Parliament by the Commission in July 2001(12), the rate of absence calculated for 2000 varies, depending on the institution, between 3,3 % and 5,9 % of theoretical working time(13), which equates to about seven to 12 working days per person a year(14).

Average absence rates in some national administrations or international organisations showed a similar picture, although different calculation methods may have been applied.

General policy in the institutions

21. Although certain internal rules and procedures have been put in place, no institution has adopted an overall policy, with clear objectives and the range of measures required to deal with all aspects of absences in an appropriate manner, i.e. fairly, constructively, but also firmly. This does not allow the institutions to ensure:

(a) that support is provided, when needed, to staff who are unable to attend work for long periods;

(b) that due efforts are made to maximise attendance for the good of the staff and their employing department; and

(c) that attendance is not adversely affected by inappropriate assignments of duties or inappropriate working conditions.

As set out below, particular shortcomings and weaknesses were observed concerning the role of management, the coordination of the various administrative and medical departments and the organisation of checks.

Roles and coordination of the various parties concerned

General observations

22. Absence management calls on the services of three parties: the employing department, the personnel department and the medical service. The effectiveness of the system is undermined by an unclear allocation of roles and responsibilities and the lack of or delays in the exchange of the necessary information. While roles and responsibilities are generally clear as regards reporting absence and maintaining absence records, it is not clear:

(a) who is supposed to be the first point of contact with the member of staff during his absence, and what frequency and method of contact are appropriate;

(b) who is responsible for monitoring absence rates both on an individual basis and across the whole institution, and according to which benchmarks;

(c) who is responsible for identifying patterns of absence which cause concern and for taking the necessary action;

(d) who is responsible for deciding whether to carry out medical checks, and in what circumstances;

(e) who is responsible for providing any additional support needed by the member of staff, for example in diagnosing his health problems in good time and ascertaining their connection with, and repercussions on, his working environment, and for taking measures to help his rehabilitation;

(f) whether return-to-work interviews should be conducted after the end of the absence, in what circumstances, by whom, how and for what purpose.

23. The lack of clarity concerning the allocation of responsibilities in respect of the decision to conduct medical checks on absences and their primary purpose may explain why, at the Commission, few requests for checks come from the personnel department or the employing departments, as they consider that this is the responsibility of the medical service. The medical service interprets this low number of requests for checks from the other departments as evidence of their limited interest in what is mainly a management responsibility. On this particular issue, institutions should consider following the good practice of the Court of Justice and the Council, which have set up a joint body comprising members of the personnel departments and the medical services, with regular meetings to examine absences and take decisions on checks.

Shortcomings specific to certain institutions

24. In 1999 the Commission began the process of decentralising leave and absence management in its directorates-general. This reform was carried out under conditions which were a source of major weaknesses in the control systems and which would appear to have led to a certain loss of central control on the part of the institution, as revealed by the difficulties encountered by the Commission, in the context of the Court's audit, in producing data and information which are nevertheless essential to everyday management.

25. The directorates-general(15) did not call for this decentralisation, which they perceived as an additional administrative burden rather than as a tool for the management of their human resources. The sharing of responsibilities with the personnel and administration directorate-general (DG ADMIN) is also confused, as it is not clear if the latter merely has an advisory and support role or if it must play a guiding, coordinating and supervisory role. Likewise, every directorate-general has appointed a human resources manager and a leave manager but it is not clear whether their duties with regard to absences should go beyond the purely administrative tasks, such as recording absences, to which they appear to limit themselves.

26. Although the Commission previously had a centralised database which provided it with general information on all its staff, decentralisation has been accompanied by a fragmentation of this system with the installation of a separate database, called SIC Congés, for each directorate-general. This SIC Congés application offers relatively few functions for performing data analysis and producing data and reports, and those which are available are not always suitable for monitoring the application of the provisions of the Staff Regulations. The obligation for each directorate-general to develop its own data analysis programmes, which, of necessity, are similar to those of the other directorates-general seems inappropriate.

27. In addition, the Commission's medical service does not have access to the SIC Congés application and has developed its own database, called Sermed, which, unlike SIC Congés, does not include absences due to illness which are not supported by a medical certificate. As a rule, the medical service is not informed of these absences. This duplication is a source of delays in forwarding information and difficulties in reconciling information.

28. At Parliament there is an unnecessary duplication of the databases for leave and absences as both the employing directorates-general and the personnel directorate-general (personnel department and medical service) maintain their own databases which contain broadly the same information. There is no link between the two databases, and since the employing directorates-general and the personnel directorate-general do not have reciprocal access to their respective databases, the exchange of information is carried out in hard copy format.

29. In the Council the employing directorates-general do not have access, even for consultation, to the database of the personnel and administration directorate-general, which includes the personnel department and the medical service. The directorates-general notify absences on paper, causing delays in the recording of absences in the database and, as a consequence, in the carrying out of any checks required. The employing directorates-general receive extracts of the absence records from the personnel and administration directorate-general but are not sent the reports, analyses and warning lists which the latter produces on a regular or occasional basis, and are not in a position to produce this type of data themselves.

Medical checks on absences

30. Apart from difficulties already mentioned above, medical checks on absences are complicated by a number of shortcomings and constraints, which have a considerable impact on their effectiveness.

Multiple duties of the institutions' medical officers

31. The medical officers of the institutions assume a wide range of responsibilities. Their principal tasks cover:

(a) occupational health, pre-recruitment examinations, annual check-ups, questions concerning health and safety;

(b) medical control, including checks on absences;

(c) medical opinions, serving on invalidity committees;

(d) medical issues provided for in the Staff Regulations, opinions on the medical aspects of applications for entitlements (e.g. doubling of family allowance for families with a handicapped child, etc.);

(e) emergencies, and treatments and opinions requested by members of staff.

32. According to the rules concerning professional conduct set out in the International Labour Organisation's Convention No 161(16) and in the national laws(17) adopted to implement Council Directive 89/391/EEC of 12 June 1989 on the introduction of measures to encourage improvements in the safety and health of workers at work(18), the duties of a medical officer are incompatible with those of an examining doctor. Contrary to these provisions, the medical officers of the Court of Justice and the Court of Auditors combined, during the audited period, the duties of occupational medical officer and examining doctor. At Parliament the medical checks in relation to absences are carried out by examining doctors employed for this purpose under the supervision of in-house doctors. These situations raise problems of professional ethics for the doctors(19) and are a source of confusion with regard to perceptions of their roles in any specific situation, both for the institution and its staff.

Intensity of checks

33. The resources available to the institutions in terms of examining doctors are limited. For instance, for its staff in Brussels (approximately 16000) the Commission had, until recently, only two part-time auxiliary examining doctors, whose workload was so heavy that it precluded virtually any possibility of home calls(20) and early or unannounced checks. The Council has one part-time examining doctor; at the Court of Justice and the Court of Auditors, the medical officer, who is employed on a part-time basis must carry out these checks himself, as far as possible, in addition to all his other tasks.

34. Although the Court has only been able to obtain incomplete information on this subject, it seems that, for these reasons, checks are relatively infrequent. For instance, at the Commission the checking rate is approximately 6 % of absences; figures allowing to establish the intensity of checks could not be obtained from Parliament; at the Court of Justice and the Court of Auditors checks are rare; at the Council they are somewhat more frequent. Generally speaking, checks focus on repeated prolonged absences and absences which represent a particularly high risk of abuse. The fact that the system does not provide for earlier and more intensive checks means that the system provides neither the necessary disincentives(21) nor timely support to those members of staff who are genuinely in need (see paragraphs 53 and 54).

Deadlines for the submission of medical certificates and for carrying out checks

35. The Staff Regulations (Article 59(1)) allow a total of 12 days per year for absences due to illnesses lasting no longer than three days and provide that a certificate must be produced for the fourth and subsequent days of absence. However, the Staff Regulations do not set a deadline for submission of the medical certificate. Each institution has adopted implementing provisions or practices which vary slightly from those of the others(22). These rules, which in some cases lack clarity, are applied, in practice, with considerable flexibility and this is likely to have a bearing on the effectiveness of controls on absence, in particular in respect of absences of shorter duration.

36. The time allowed for submission of the medical certificate, when added to the time required by the internal procedures for transmitting the information between the various departments, means that, as a rule, no checks can possibly be scheduled in the two weeks following the first day of absence. The checks are therefore carried out late sometimes, which is normal practice at the Council, even after the member of staff has returned to work. This has repercussions on their effectiveness in cases of absences suspected not to be genuine, as it is then no longer possible to call into question the legitimacy of the absence, even in the most doubtful of cases. In its established case-law, the Court of First Instance(23) has prohibited the calling into question of the validity of a medical certificate covering the period which predates the check.

The rights of the institution to demand a medical examination

37. The Staff Regulations stipulate (Article 59(1)) that a member of staff absent due to illness may be required to undergo a medical examination arranged by the institution and (Article 60) that, without prejudice to any disciplinary measures that may apply, any unauthorised absence which is duly established shall be deducted from his annual leave and, if this has been used up, shall give rise to a salary deduction. The Staff Regulations say nothing about the measures that should be applied when a member of staff impedes or deliberately evades a medical examination.

38. It would seem reasonable to assume that an institution's inability to exercise control as a result of a deliberate refusal on the part of the member of staff might in itself be grounds for declaring the absence unauthorised, as from the date fixed for the medical examination. The General Secretariat of the Council laid down clear internal provisions to this effect. Parliament's internal provisions only provide that, in such cases, the member of staff is failing to fulfil his obligations and is liable to disciplinary measures, but in practice it has followed an identical approach to the one laid down by the Council. However, both institutions had to abandon this practice following a decision by the Court of First Instance(24) which established that there is no legal basis for it. According to the Court of First Instance, the presentation of a medical certificate creates a presumption of regularity in respect of the absence. Deducting an absence from annual leave presupposes that the unauthorised nature of the absence has been duly established. The member of staff is required to undergo an examination, but failure to fulfil this duty does not in itself authorise the institution to reverse the presumption of regularity in respect of the absence due to illness which attaches to a medical certificate that has been duly presented. On the other hand, this failure can, where appropriate, form the subject of disciplinary proceedings. In reality, the institutions are therefore practically powerless in the face of those situations, happily rare, where there is reason to doubt the legitimacy of the absence, as this cannot be resolved, whatever the case, by any disciplinary procedure.

In France, pursuant to provisions applicable to officials of the State(25), if the person concerned is unable to justify not undergoing the medical examination, the administration may withhold the payment of his salary, declare the absence unauthorised and launch a procedure "for desertion of post" which may lead to his removal from his post.

39. The examining doctor's conclusions with regard to the member of staff's fitness to resume work may be challenged, which is normal. The appeal and arbitration mechanism is, however, extremely unwieldy and inappropriate. If the member of staff submits a new medical certificate and if, following a second examination, the difference of opinion between the attending physician and the examining doctor persists, the case should then, according to the Court of First Instance's case-law(26), be referred to the Invalidity Committee. This procedure, with the amount of time it entails, is completely unsuited to handling disputes of this type as it is difficult to imagine that the Invalidity Committee could legitimately rule on whether, several months before, the member of staff had been fit to resume work or not(27).

In France, pursuant to provisions applicable to officials of the State(28), the person must resume work as soon as he receives the administrative decision to do so. He may challenge the examining doctor's conclusions in front of a medical college, composed of doctors appointed for three years by the administration (and meeting at least twice a month). In cases where the dispute is about the renewal of an absence of more than six consecutive months, he may make an appeal against the opinion of the medical college to a high medical college (appointed for three years by the Ministry of Health).

40. These factors have a bearing on the effectiveness of the medical checks, especially in cases where strong doubt exists as to the legitimacy of the absence.

OBSERVATIONS CONCERNING INVALIDITY PENSIONS

Quantifying and monitoring the cost of the scheme

41. The institutions have not equipped themselves to establish and monitor the cost of the invalidity pensions scheme. This shortcoming also means that the staff, who also contribute to the funding of the scheme, are not aware of their responsibilities and not in a position to exercise any control over the functioning of the scheme. This is undoubtedly due to the fact that since the cost of invalidity pensions is financed from the budget year by year there was not felt to be any need to establish supervisory and control mechanisms similar to those used to safeguard the equilibrium of a pension fund. For the institutions other than the Commission, the budgetary cost of the pensions awarded to members of their staffs is not shown in their financial statements, so that there is even less incentive to perceive that such mechanisms are desirable.

42. The budgetary expenditure charged to Commission item A-1 9 0 1 does not provide any information about the cost of this scheme, for several reasons. Firstly, it only reflects the cost of the pensions paid during the financial year. The specific future cost of invalidity pensions is unknown and is included in the total future cost of pensions shown in the balance sheet. Secondly, unlike other schemes where an invalidity pension ceases to be paid once the recipient reaches retirement age and starts to receive a retirement pension, under the Community scheme an invalidity pension continues to be paid until the beneficiary dies. The Court estimates that, for the financial year 2000, out of the 123,7 million euro of budgetary expenditure against A-1 9 0 1, around 40 %, or 50 million euro, was paid to beneficiaries over 65 and was thus more in the nature of retirement pensions(29). The changes to the Staff Regulations that are currently being considered aim to make this distinction in future.

43. The Court estimates the average net cost of one case of invalidity at around 370000 euro(30). At the current rate of retirement, i.e. around 200 members of staff, the invalidity pensions awarded each year thus represent a total net cost of the order of 74 million euro. As explained in the note to this paragraph, this net cost is not the amount of the invalidity pension paid to the former member of staff.

The assessment of invalidity

44. Article 78 of the Staff Regulations provides that an official or a member of the temporary staff "shall be entitled to an invalidity pension in the case of total permanent invalidity preventing him from performing the duties corresponding to a post in his career bracket".

45. If "total permanent invalidity" were intended as a state involving total loss of physical and/or psychological integrity, almost none of the persons retired on an invalidity pension and included in the sample examined by the Court's medical expert would be classified as total invalids. In practice, according to the spirit of the Staff Regulations, it is a question of determining invalidity as incapacity for work, that is to say, the impossibility of the member of staff performing the duties corresponding to a post in his career bracket. In consequence invalidity covers many factors and is assessed, on the one hand, on the basis of strictly medical factors which can be identified and measured by means of clinical examination and tests, and on the other hand, by means of non-medical facts which rely on subjective rather than quantifiable criteria. The Staff Regulations do not require consideration of other possible responses to ill health, such as capacity to work part time or to work in a less demanding level in the organisation.

Most of the national administrations and international organisations visited by the Court use a similar concept. However, in France, pursuant to provisions applicable to officials of the State(31), the degree of disability is determined by reference to an official disability scale. In Sweden, the invalidity can be total or partial; in the case of partial invalidity (e.g. 75 %) the member of staff must work part-time (e.g. 25 %)(32).

46. To assess incapacity of performing the duties corresponding to a post in a career bracket, a sound knowledge of the working background and good medical and administrative synergy are necessary, especially in the case of psychological illness(33). It is particularly important to be familiar with the specific working background of the member of staff and to explore properly and in good time the possible alternatives to retirement on an invalidity pension. In this respect the systems applied in the European institutions show significant weaknesses.

47. In the institutions, invalidity is currently approached only from a medical angle, something that the medical officers of the institutions are the first to find regrettable. The departments responsible for personnel management and the employing departments mainly confine themselves to the purely administrative aspects of the procedure. The Invalidity Committee is therefore left with the responsibility of assessing not only the purely medical part of the problem, but the non-medical aspects as well(34). Very often the Invalidity Committee does not have sufficiently precise information about the nature of the duties performed by the patient and those which he could reasonably perform. Incapacity is then assessed in relation to a theoretical reference framework, which assumes that posts within the category in question are uniform, whereas the reality is diverse, with the working environment, and the relational aspects in particular, constituting one of the major factors of difference. The system thus lends itself to a broad interpretation of the concept of invalidity.

Monitoring, analysis of causes and corrective action

48. The institutions have not established a monitoring system that enables them, on a continuous or regular basis, to quantify, in particular, the frequency of retirement on an invalidity pension and to obtain information about the pathology of such retirements, to analyse trends using a set of relevant parameters (age, sex, category, length of service, etc.), to pinpoint the population groups that are most at risk, analyse the causes of the highest frequencies, or frame and implement possible measures concerning prevention, early detection, research into alternative solutions. This shows a certain lack of interest on the part of the institutions, both as regards the financial management of the scheme and the aspects of the working environment and personnel management that are more likely to have an incidence on the problem.

49. Some statistics were drawn up on or immediately after publication of the Court's observations in the Annual report concerning the year 1987. They essentially comprise information supplied by the Commission in replying to two parliamentary questions in October 1991 and June 1993, and a report which it presented in December 1993, at the Council's request, in connection with the examination of the preliminary draft budget for 1994. However, the analysis is very cursory. The Court did not find any evidence of attempts to quantify or produce a general analysis of the scheme's operation after those dates.

50. Fragmented or very specific information is sometimes available, for example on the pathology of invalidity retirement, but it is the result of individual initiatives on the part of individual medical officers rather than a response to requests by the institutions. Moreover, the information is either held by the medical services for their own information, or, when it is forwarded to the administration, it is not followed up(35). An interinstitutional medical board's suggestion for an epidemiological study did not receive the necessary funding, and the more recent idea of the Joint Sickness Insurance Scheme to carry out a study on the general state of health of the staff ran into opposition from representatives of the staff and of some administrations, on the grounds that there were insufficient guarantees as to how the data might be used.

51. In the context of analysis of the causes of invalidity, a noteworthy initiative was that taken by Parliament's medical officer in Luxembourg. He produced five-year reports for the period 1983 to1998 that are a model of what could be done systematically for the institutions as a whole. These reports are full of information, including information on the frequency of retirement on invalidity pension and the pathology of it, and they put forward a number of hypotheses as to the causes of invalidity, with recommendations about the nature of possible action, especially as regards the organisation of work and the working environment and certain aspects of personnel management.

52. According to these reports, the organisation of work and the psychological and social environment are significant factors in stress and loss of motivation. They impact on the state of health of members of staff and may contribute to the high rate (around 50 %) of psychological illness among the medical conditions that underlie retirement on an invalidity pension. Perceptions of unfair distribution of work, exclusion from the decision-making process and shortcomings at managerial level are included among the causes of lack of job satisfaction. If a member of staff is affected by a physical and/or psychological condition these may be decisive factors in the process leading up to retirement on an invalidity pension. These problems particularly affect staff in the lower career brackets, who perceive their work as being under-valued and under-recognised. This analysis is widely shared by the medical officers in the other institutions and by the personnel departments. It also seems to be borne out by the replies which the Court received in response to the questionnaire sent to a sample of former members of staff on invalidity pensions (see paragraph 2). In fact, in around a third of cases, the former members of staff said that they had been dissatisfied with the nature of their duties and/or the conditions in which they had to perform them and, in two thirds of these cases, said that that had had an influence on the fact that they had applied for or agreed to retirement on an invalidity pension. They generally blamed over-work, pressure of deadlines, shortage of human and material resources and, quite frequently, the role of management in relation to these problems.

Early detection, medical and administrative synergy and alternative measures

53. Due to the abovementioned weaknesses in the absence control systems the institutions are not in a position to detect systematically early signs of persons at risk whose health and/or performance is declining. They therefore cannot take appropriate measures in good time (e.g. mobility, organisation of work, etc.), that is to say, while it is still possible for the medical service and the administration to make a concerted effort to avoid a chain of events that ineluctably leads to removal from the working environment and, therefore, to invalidity.

54. The replies to a questionnaire sent by the Court to a sample of former members of staff indicated that in approximately half the cases they had not received any enquiry about their state of health or the possibility of their resuming work from the administrative departments or their senior officers during the periods when they were on sick leave in the year prior to the opening of the invalidity procedure. They interpreted this as a general indifference towards them, precisely at a time when they were experiencing difficulties and were looking for support and concern on the part of the institution in general and their hierarchical superiors in particular.

55. The study by the Court's medical expert showed that, with some rare exceptions, the files examined did not contain any indication that serious action had been taken beforehand with the aim of reaching an alternative solution and suggestions from the medical officer, mainly regarding a change of post, were practically never followed up. An alternative to an invalidity pension was seldom offered, or was offered too late. One quarter of the sample of former members of staff receiving invalidity pensions would probably have been able to continue working if the institution had detected and treated their cases at an early stage. The replies received in response to the questionnaire confirm this feeling: 20 % of the former members of staff thought that they would have been able to continue working had an alternative solution been offered. On that basis, the Court estimates that, in principle, savings of about 10 million euro could be made every year on the net cost of the invalidity pensions granted in the year, by implementing a policy of early detection and treatment of repeated or prolonged periods of absence due to illness.

Opening the invalidity procedure

56. Article 59(1) of the Staff Regulations authorises the institution to refer the matter to the Invalidity Committee only if the member of staff has had a total of more than 12 months' sick leave in any period of three years. This raises the difficult question of the relevance of the length of temporary incapacity before the procedure is started. For any doctor with experience of social services and any human resources specialist, it is an acknowledged fact that a person who has been away from the working environment for an extended period can seldom return to it fully, because of a combination of medical, psychological, technological or sociological factors. This is confirmed by the fact that in these cases the procedures almost always culminate in retirement on an invalidity pension. A review of the sample of cases submitted to the Court's medical expert for study suggests that a threshold of six months' absence would be more appropriate, on the one hand, for dealing with the cases where coordinated action is useful and still possible and, on the other hand, so as not to maintain in a situation of undue financial advantage(36) people who, actually, will never return to work on a permanent basis. From the information available to the Court it also appears that where the invalidity procedure is opened on a member of staff's initiative it is, on average, after a period of absence of five months in a three-year period.

Invalidity schemes of national administrations and international organisations vary considerably regarding the length of the period of incapacity before the procedure can be opened on their initiative. Compared to the Community scheme, the period is shorter in Germany (three months over a six-month period) and at the EBRD (six months). At the ILO (18 months over a four-year period) the conditions are more similar to those of the Community scheme(37).

57. Moreover, although the smallest institutions (Court of Justice and Court of Auditors) refer matters to the Invalidity Committee as soon as the threshold of 12 months' absence has been reached, this does not apply in the largest ones, which often delay. In the sample examined, in 20 % of cases at the Commission and 38 % of cases at Parliament, the invalidity procedure was initiated after more than 14 months absence, with the total period of absence averaging around 17 months(38).

The Invalidity Committee

Constitution

58. One feature of the Community system that makes it rather cumbersome, compared with other invalidity schemes (see below), is that two of the three doctors that make up the Invalidity Committee are not appointed by the organisation that funds the pension. The two doctors in question are the doctor representing the member of staff and the third doctor who is appointed by agreement with the doctor representing the institution.

The civil services of the Member States visited act as follows.

In Austria, the "Bundespensionsamt" (Federal Pensions Office) commissions two medical examinations, each of which is carried out by a single doctor. In Germany the medical opinion is provided by the medical officer of the administration. In the United Kingdom it is provided by the Medical Services Advice Board, in Sweden by the doctor of the pension scheme, and at the EBRD by the doctor of the insurance company. At the UNO, a recommendation is made by the Staff Pension Committee, made up of representatives of staff, the administration and the Member States, on the basis of a medical report by the Joint Medical Service. In France, advice is provided by the "Commission de réforme" of the ministry concerned, which is made up of the head of the department, the financial controller, two staff representatives and the three doctors of the medical committee. The medical officer of the department concerned may get access to the file, express his views and attend the meeting of this commission in a consultative role. The commission may also choose to hear the doctors chosen by the member of staff and by the administration(39).

59. Cooperation by the member of staff, who must appoint one of the members of the Invalidity Committee, appeared to be deficient in cases where the procedure was instigated on the institution's initiative. In fact, it was found that, for the sample examined, whilst it took an average of 50 days before the Invalidity Committee could meet in cases where the request had come from the member of staff, it took twice as long in cases where it was the institution that took the initiative(40). In 20 % of cases the average time was more than six months.

60. This phenomenon can be explained to some extent by the fact that the records under consideration are essentially linked to psychological illness, that, in some cases, there may be a conflictual relationship with the institution and that the former member of staff receives a full salary while on sick leave.

61. The length of time taken can also be attributed to difficulties in reaching an agreement between the medical officer of the institution and the doctor appointed by the member of staff on the choice of the third doctor. The medical officers mentioned that they were quite often obliged to oppose proposals putting forward the names of doctors who had a known propensity to decide in favour of invalidity. Lastly, delays also result from the frequent difficulty of convening the three doctors.

62. The study by the Court's medical expert nevertheless leads to the conclusion that in all the cases in the sample, the choice of the third member of the Invalidity Committee was very appropriate as to his professional ability and reputation, independence vis-à-vis the parties, familiarity with advisory work and knowledge of the specialist medical fields that the work concerned.

Basis of opinions

63. As set out in paragraph 1, one of the audit objectives was to verify that invalidity pensions are granted only when invalidity has been duly recognised. In all 63 cases in the sample studied by the Court's medical expert he concluded that the opinions were based on sufficient and adequate work, demonstrating the existence of a clear and intelligible (to a physician) link between the findings and the conclusions of the Invalidity Committee. They were systematically set out and supported by details of the diagnosis and development of the condition, the prognosis of improvement or recovery or, on the contrary, the probability or certainty that the causes of the incapacity would persist. They took account of the medical aspects as well as those concerning the member of staff's integration in his working environment, though subject to the constraints on the Invalidity Committee's work set out in paragraph 47. The study did not reveal any cases where retirement on an invalidity pension was clearly not justified; in two cases, however, the severity of the condition and the disabling nature of it seemed to be open to question, which is not abnormal, given the inevitable leeway inherent in any assessment of this kind.

64. Non-psychiatric conditions were established by means of appropriate tests, which provided a sound basis for an accurate and precise diagnosis. The opinions were based on a standard advisory approach, correctly performed, which provided a sound basis for a reliable, reproducible opinion, which was impeccably drafted. In the case of the psychiatric conditions the items on the file also support the diagnosis and permit an assessment of the relative severity of the condition.

Decision on invalidity

65. When the Invalidity Committee establishes a member of staff's invalidity, he is automatically retired on the last day of the month in which the appointing authority recognises his permanent incapacity to perform his duties, and the entitlement to invalidity pension commences on the first day of the calendar month following the calendar month of the decision. In almost 25 % of the cases in the sample the Court observed delays in adopting the appointing authority's decision, which had the effect of unnecessarily delaying the member of staff's retirement on an invalidity pension by one month.

Medical examinations after retirement on an invalidity pension

66. The Staff Regulations (Article 15 of Annex VIII) provide that the institution may have a former official drawing invalidity pension examined periodically, up to age 60 years, in order to ascertain that he still satisfies the requirements for payment of the pension.

67. The frequency of such checks varied markedly from one institution to another. The Court of Justice and the Court of Auditors systematically have former members of staff examined every two years, except where the Invalidity Committee considers such examination unnecessary, because the illness is irreversible. At the Commission, examinations are carried out with the frequency prescribed by the Invalidity Committee, usually annual or every two years. At the Council there is only one examination, between one-and-a-half and three years after retirement on an invalidity pension, and in a small proportion of cases (around 17 % according to the medical officer). Medical checks of this type are rare at Parliament.

68. In the sample studied by the Court's medical expert, the Invalidity Committee had not made a medical examination obligatory in 60 % of the cases(41). The decision of the Invalidity Committee was in all cases justified by appropriate considerations regarding the condition of the person in question and the prognosis. Where provision for an examination had been made, the examination had been correctly carried out and the conclusions reached as a result of it were well-founded.

69. In practice the possibility of medical examination provided for by the Staff Regulations does not really allow any reversal of the situation of invalidity. In fact, an interruption of invalidity very rarely results in successful reinstatement. For the reasons mentioned at paragraph 56, in the rare cases where former members of staff return to work, they are almost always retired on an invalidity pension after a relatively short time.

Draft reform of the scheme

70. In the context of the proposed reform of the Staff Regulations the Commission put forward a set of proposals seeking to reduce the financial attraction of invalidity pensions. These proposals are essentially identical to those put forward in 1989 in the Caston report (see paragraphs 16 and 17):

(a) replacing the invalidity pension with an invalidity allowance which is then replaced by a retirement pension once retirement age is reached (between ages 60 and 65, when the maximum rate is reached);

(b) calculating the invalidity benefit as 70 % of basic salary, instead of basing it on the rate of retirement pension to which the former member of staff would be entitled at age 65 if he continued in active employment up to that age;

(c) abolishing advancement in step for the duration of the period of invalidity;

(d) reducing the minimum subsistence figure from 120 % to 100 % of the basic salary of an official in the first step of the lowest grade;

(e) introducing an anti-aggregation rule, with deduction from the invalidity allowance of the part of the income from gainful activity that exceeds the final total remuneration received while in active service;

(f) making the invalidity allowance subject to pension contributions;

(g) calculating the amount of retirement pension on the basis of the last basic salary received during employment.

71. In the Court's opinion the measures would result in a decrease of around 25 % in the total net cost of the invalidity pensions awarded each year. At current retirement rates this would decrease from around 74 million euro (see paragraph 43) to around 55 million euro. Furthermore, the Court estimates that the measures would result in an average reduction of around 10 % in the amount of the invalidity pension paid to former members of staff; the incidence would be higher for category C and D staff receiving the minimum subsistence level, which would be cut by some 14 %.

72. The change to the minimum subsistence rule might be expected to result in savings of around 2 million euro in the net cost of invalidity pensions awarded each year. For the beneficiaries concerned the change would produce a cut of around 6 % of the total pension received. The aim of this measure is to provide greater equity for members of staff who continue to work until normal retirement age and, also, to offer an incentive to keep on working by making work financially more rewarding than invalidity. In certain cases, the minimum invalidity pension will nevertheless still be higher than a retirement pension.

73. The Commission did not discuss the matter with the institutions' medical officers, who are, however, the principal players in the operation of the scheme, and it omitted to take into account parameters other than financial factors, which, as the Court has shown in this report, have a predominant effect. The invalidity problem is both a human and a financial one. The proposed reform omits to take into account the action that should have been taken beforehand, as part of a policy of prevention (working conditions, information campaigns, monitoring absences, mobility, etc.), and likewise the action that should be taken afterwards, with a policy of rehabilitation or redeployment (part-time working, adjusting working hours, etc.), in order to encourage staff who are in poor health to continue working under reasonable conditions, without necessarily having to ask for their retirement on an invalidity pension.

CONCLUSIONS

74. The main conclusions of the audit are the following:

(a) the invalidity pensions awarded each year represent a total net cost of the order of 74 million euro, with an average net cost for each case of invalidity at around 370000 euro (paragraph 43);

(b) in all cases of invalidity pension examined the invalidity had been duly recognised (paragraph 63);

(c) the institutions have not developed an overall policy regarding absence and invalidity management due to a lack of senior management commitment (paragraphs 21 and 41);

(d) the management systems suffered certain shortcomings in different areas and the management of absences and invalidity was approached solely from the medical point of view (paragraphs 20 to 40 and 44 to 54);

(e) financial savings could be obtained through adequate administrative measures for prevention, early diagnosis and treatment (paragraph 55).

75. The Court's audit reveals a complex picture. On the one hand, over the last 15 years the rate of invalidity retirement has remained stable and the increase in retirement on invalidity pensions may be explicable by the increase in the numbers and age of the workforce in the institutions. In the opinion of the Court's medical adviser invalidity pensions are awarded in a justified way, the work of the Invalidity Committees is sufficient and adequate to assess the reality and severity of the pathology and the resultant incapacity. On the other hand, retirement on invalidity grounds is more common in certain grades than normal retirement. Reports by the institutions' medical officers and surveys of invalidity pensioners reveal frustration in the working environment frequently to be a significant element in demotivating staff who are eventually retired on ill-health grounds. A significant part of invalidity retirements could, moreover, be avoided if adequate administrative measures for prevention and early treatment of medical problems and the associated employment problems were taken in good time.

76. Shortcomings were found concerning the overall policy and management systems for absences because of illness, which generally suffer from lack of clarity and coordination between the various departments involved, lack of resources, unsuitable IT management tools, and the obstacles resulting from the provisions of the Staff Regulations. The absence of medical and administrative synergy in dealing with absences due to long-term and/or repeated illness and the lack of priority given to finding alternative solutions in good time results in the relative ease of setting in train a process which inevitably leads to removal from the working environment and to retirement on an invalidity pension. To a large extent the management of invalidity is approached from the medical angle alone, as the departments responsible for human resources management and the employing departments do not involve themselves in the process and by and large confine themselves to dealing with it by purely administrative means.

77. Frequent delays in opening the invalidity procedure and in the progress of a significant part of the invalidity retirement procedure entail a substantial cost to the budget and misuse of human resources. The difficulties of setting-up and convening an Invalidity Committee, two of the three members of which are appointed by or with the agreement of the member of staff are the main cause of this. Excessive delays in adopting the administrative decision to retire a member of staff on an invalidity pension were also observed as being frequent.

78. The provisions of the Staff Regulations regarding medical examinations after retirement on an invalidity pension are theoretically satisfactory, but, in practice, have no perceptible effect, because it is extremely rare for them to lead to successful reinstatement, i.e. to an effective, long-term return to work. A person whose state of health is deteriorating, especially where psychological problems are involved, and who is absent from the working environment for an extended period can rarely return to it successfully and, with a few very rare exceptions, invalidity is always final in nature.

79. As part of the draft reform of the Staff Regulations the Commission put forward proposals which are limited to the financial aspects of the invalidity pensions scheme and which are essentially a reworking of the proposals that were put forward in 1989 and subsequently abandoned. It did not carry out a preliminary evaluation of the overall scheme in order to identify all the factors that impinge on the invalidity phenomenon, and it did not involve the medical officers of the institutions in its preliminary studies, even though the medical officers are currently the principal players in the process and could have made a useful contribution by widening the field under review.

RECOMMENDATIONS

80. The institutions need to develop an overall policy on absences due to illness and on invalidity, with performance indicators, strong support from senior management, clearly allocated roles and responsibilities, strong medical and administrative synergy, and with careful and resource-intensive attention given to the needs of those members of staff who need support. This policy should focus both on actions to be taken in the early stages through preventative measures that consider the organisation of work and working conditions, and on those actions required at a subsequent stage to help rehabilitation and encourage members of staff who are in poorer health to continue to work under reasonable conditions.

81. Although medical matters are obviously involved, this policy is above all the responsibility of management. Managers have therefore to be made aware of their roles and responsibilities and they should receive the necessary guidance and training. An adequate management information and monitoring system should be set up with a view to ensuring the timely identification of patterns of absence and of cases of retirement on an invalidity pension which cause concern and the adoption of appropriate action. Annual reports should be established by the institutions in order to monitor the performance of the system.

82. Modifications to the Staff Regulations may need to be considered to better cope with situations where a member of staff impedes or deliberately evades a medical examination, and to provide for an easier and swifter appeal and arbitration mechanism when the examining doctor's conclusions are challenged. As far as provisions on invalidity are concerned, it may be appropriate to reconsider the threshold period of sick leave before the institution can start the procedure and the need for medical examinations after retirement or the conditions under which they are performed. A modification of the composition of the invalidity committee should also be examined, and thought should be given to creating a more permanent and interinstitutional structure.

This report was adopted by the Court of Auditors in Luxembourg at its meeting of 27 February 2003.

For the Court of Auditors

Juan Manuel Fabra Vallés

President

(1) The risk factors taken into account were the following:

- no recognition of serious illness by the Joint Sickness Insurance Scheme (JSIS),

- no major surgery,

- low level of reimbursements for medical expenses by the JSIS during the two years preceding invalidity retirement,

- staff who had been the subject of an administrative enquiry, disciplinary proceedings, or proceedings for incompetence, or who had been marked "inadequate" in their staff reports,

- the awarding of the minimum subsistence amount,

- cases where the retirement pension which would have been obtained at the age of 60 is less than the maximum rate.

(2) The rate of response to this questionnaire was 57 %.

(3) The Court's findings have been presented in such a way that the rules governing medical confidentiality have not been infringed, the medical expert's report contains no information which would enable a connection to be made between an identifiable person and a pathological condition.

(4) Austria, France, Germany, United Kingdom and Sweden.

(5) European Investment Bank (EIB), European Bank for Reconstruction and Development (EBRD), International Labour Office (ILO) and World Health Organisation (WHO), United Nations Joint Staff Pension Fund (UNO).

(6) Council Regulation (EEC, Euratom, ECSC) No 259/68 (OJ L 56, 4.3.1968, p. 1) as last amended by Council Regulation (EC, ECSC, Euratom) No 1986/2001 (OJ L 271, 12.10.2001, p. 1).

(7) The invalidity rate can be defined as the number of agents retiring on medical grounds each year divided by the total number of Commission officials and temporary staff.

(8) Commission staff account for approximately 75 % of the beneficiaries of invalidity pensions.

(9) In the case of the A/LA and B categories, the minimum subsistence figure rule is only applied in rare cases where invalidity occurs after a relatively short period of employment.

(10) 30 % for Germany in 2000 (Daten und Schlussfolgerungen zum Zweiten Versorgungsbericht der Bundesregierung, Bundesministerium des Innern, 12 September 2001), 60 % in 1998 to 1999 for UNO according to data obtained during the visit at the United Nations Joint Staff Pension Fund.

(11) OJ C 316, 12.12.1988, p. 151.

(12) SEC (2001) 1128 final (report attached to proposal No 17/2001 for a transfer of appropriations to cover expenditure for auxiliary staff).

(13) Absence due to illness is entered in calendar days. As a general rule, it is measured over a reference period of one calendar year and is expressed as an average percentage rate of absence for the whole population for the year (number of days of absence divided by 365).

(14) On the basis of 205 working days a year.

(15) Interviews were carried out in three DGs.

(16) Article 15 of ILO Convention No 161 provides, "Occupational health services shall be informed of occurrences of ill health amongst workers and absence from work for health reasons, in order to be able to identify whether there is any relation between the reasons for ill health or absence and any health hazards which may be present at the workplace. Personnel providing occupational health services shall not be required by the employer to verify the reasons for absence from work."

(17) In the case of Luxembourg, Article 12 of the law of 17 June 1994 concerning occupational health services provides that "le médecin du travail exerce sa fonction en toute indépendance professionnelle par rapport à son employeur et aux travailleurs. En aucun cas le médecin ne peut vérifier le bien-fondé des congés de maladie. La fonction de médecin du travail est incompatible avec l'exercice libéral de la profession." (The medical officer shall exercise his duties in complete independence of his employer and the staff. Under no circumstances may the doctor verify the legitimacy of absences from work due to illness. The duties of the medical officer are incompatible with external professional activities. (ECA translation)).

(18) OJ L 183, 29.6.1989.

(19) For these reasons, moreover, Parliament's medical officer in Luxembourg recently decided to give up all verifications of the legitimacy of absences from work due to illness.

(20) Since the end of 2000 the Commission has had recourse to an external company for conducting checks at home.

(21) In addition, although this has no impact in the context of invalidity, it must be noted that, to a large extent, short periods of absence escape controls, including those absences of less than three days which are not supported by a medical certifcate. This presents a risk of abuse of a deadline which might be interpreted by some as a right to 12 days' additional annual leave.

(22) The periods laid down in these internal rules are as follows:

- Commission: no precise period (as soon as possible),

- Parliament: on the fourth calendar day at the latest,

- Council: on the sixth working day at the latest,

- Court of Justice: no written rules; generally, one week at the latest,

- Court of Auditors: no precise period; generally, at the latest when resuming work.

(23) In particular, Case T-36/96 Gaspari v Parliament, Judgment of 10 July 1997.

(24) Joined cases T-110/99 and T-260/99 F v Parliament, Judgment of 13 December 2000.

(25) Fonctionnaires et stagiaires de l'Etat, Protection sociale contre les risques maladie et accidents de service (Circulaire du 30 janvier 1989).

(26) Case T-171/98 Biasutto v Council, Judgment of 22 February 2000.

(27) As an example, this arbitration procedure took eleven months in one case at the Court of Justice.

(28) Fonctionnaires et stagiaires de l'Etat, Protection sociale contre les risques maladie et accidents de service (Circulaire du 30 janvier 1989).

(29) The estimation is based on information extracted from the Commission's database Sysper, which contains personal data of active and retired staff.

(30) The net cost of invalidity is defined for the purposes of this report as the actuarial net impact (expenditure - income) on the Community budget of awarding an invalidity pension. The impact equals the actuarial difference between the net budgetary cost of retirement on grounds of invalidity (i.e. the cost of an invalidity pension plus the salary of the person recruited to replace the recipient of the invalidity pension) offset against the net budgetary cost if the member of staff continued to follow a normal career (i.e. the salary received by the member of staff up to normal retirement age plus his retirement pension and the salary of his replacement after that date).

(31) Fonctionnaires et stagiaires de l'Etat, Protection sociale contre les risques maladie et accidents de service (Circulaire du 30 janvier 1989).

(32) Data obtained during the visit to the Swedish Agency For Government Employers.

(33) These account for almost half the cases of invalidity in the European institutions.

(34) Examples of such non-medical aspects are: the level of job satisfaction, quality of relationships with colleagues, mastery of new technology, etc.

(35) A Commission doctor in Brussels has, since 1997, kept up-to-date statistics on the medical causes of invalidity, but they are not circulated to other services. Since 1999, the Council's medical officer has submitted similar annual statistics to the administration: measures are being envisaged.

(36) Continuation of full salary.

(37) Data obtained during the visits.

(38) It was not possible to make similar estimates for the Council as the latter did not provide the necessary data.

(39) Data obtained during the visits.

(40) This difference is partly inevitable, however, in that when the procedure is opened on the member of staff's initiative he usually gives the name of the doctor representing him immediately when making the application.

(41) 38 cases out of 63.

The Institutions' replies

REPLY OF THE COUNCIL OF THE EUROPEAN UNION

The Court of Auditors' Report on the invalidity pensions scheme of the European Union does not give rise to any observations on the part of the Council.

THE EUROPEAN PARLIAMENT'S REPLIES

1 and 2. Objectives and methodology

Parliament welcomes the fact that, on the basis of a detailed examination of 127 of the more difficult cases (invalidity pensions awarded on psychological grounds), including a review by a Court-appointed medical expert of 63 such cases, the report finds no evidence of invalidity pensions having been awarded improperly or as a substitute for other measures. It also welcomes the report's emphasis on improved monitoring and preventive measures as the best approach to this difficult problem.

10 to 14. Statistics and analysis

The report points to the higher percentage, both relatively and absolutely, of category C and D staff who leave with an invalidity pension and the significant proportion of such staff who benefit from the provisions relating to the minimum subsistence amount. It is not entirely clear whether, in invoking these figures, the report is implying the kind of incentive effect referred to earlier (see points 1 and 2). If that is indeed the case, Parliament would reiterate its view that, to the extent that the audit itself appears to show that invalidity pensions for all categories of staff have been properly awarded, the question of incentives should be discounted. The main medical causes of invalidity (see point 14) show that the European institutions are far from having at their disposal all possible means which might have a real impact on these causes (for instance, non-professional aspects).

21 to 23. General policy and coordination

Parliament accepts without reserve the report's comments (point 22) about the need to define proper procedures and the respective roles of the actors and services involved in operating them. Indeed, it is precisely these principles which, together with the close involvement of employing services, are the basis of the changes that it proposes to its own system of managing absences (see comments in relation to points 30 to 34). However, it is not entirely accurate to suggest (point 23) that there is a lack of clarity as regards the allocation of responsibilities in respect of the decision to conduct medical checks on absences. In Parliament, the decision in question is vested in the Head of the Personnel Division, acting on his or her own initiative, on a recommendation from the Medical Officer or at the request of the director-general responsible for the employing service involved. Again, there is nothing to suggest that these procedures are not fully understood by all the parties involved.

28. Computer systems

The Directorate-General for Personnel is currently heavily engaged in preparatory work with a view to acquiring a comprehensive human resources software package designed according to modern best practice with integrated information across all departments. When installed, this should obviate the problems alluded to in the report, including the current inability to keep account of sickness leave other than in working days.

30 to 34. Medical checks

As the report indicates, medical checks are carried out by doctors employed for this purpose, on a contract basis, both in Luxembourg and Brussels, where, respectively, 60 and 21 checks were carried out in 2001 (while controls were also carried out in the previous years, no statistics are available for those years).

These examining doctors do, however, currently operate under the aegis of the in-house medical officers, and the report is quite right to draw attention (point 32) to the inherent problems of incompatibility of roles for the institution's medical officers and the fact that current arrangements are in breach of international standards and norms in this regard. With this in mind, however, Parliament's Directorate-General for Personnel is currently working on a series of reforms which should go a long way towards meeting this and other criticisms made in the report. These reforms will involve:

- the transfer of all responsibilities in relation to medical checks, including responsibility for the examining doctors, from the Medical Service to the Personnel Department,

- improved handling of medical certificates, possibly based on the Personnel Department rather than (as is currently the case) the Medical Service,

- close monitoring, by the Leave Office, of sick leave with a view to early detection of anomalies and the initiation of appropriate follow-up action,

- close and early involvement of the Social Affairs Division where problem cases are detected, with the latter being responsible, in conjunction with the Medical Officer and, where appropriate, other support services, for ensuring adequate follow-up measures, including, where appropriate, making recommendations to the appointing authority as regards the administrative situation of the individual concerned.

It is envisaged that the system outlined above will operate in a context of close liaison between the services of the Directorate-General for Personnel and the employing departments. The intention is not merely to resolve the problem of incompatibility of roles referred to earlier but also to provide a framework within which the roles of all participants are adequately defined and which will encourage the close monitoring and early detection of problems to which the auditors rightly attach so much importance. It is hoped that a system based on the above principles will be in operation by the start of 2003 and that it will also lead to the speedier convening, where necessary, of invalidity committees and earlier implementation of the latter's recommendations.

35 and 36. Deadlines for the submission of medical certificates

Parliament's rules stipulate that medical certificates must be submitted on the fourth day of absence. That is a relatively early deadline and, in practice, a rather longer period would normally elapse before action is taken to determine the status of the person concerned and, where appropriate, to endeavour to ensure that the situation is regularised. In such cases, action normally means the Leave Office first liaising with the Medical Officer to check whether there is a certificate in the pipeline and, subsequently, with the employing department to check whether the latter has any information which might explain or account for the apparent irregularity. The ultimate 'sanction', where no reasonable explanation or medical certificate is forthcoming, is, after referral to the Head of the Personnel Division, to deduct the relevant number of days from the individual's annual leave (or salary) and to send him a letter apprising him of that decision, although, in practice, the subsequent production of a medical certificate and a reasonable explanation for its late submission would normally lead to such a decision being rescinded.

Such a system, as the auditors point out, inevitably calls into question the institution's ability to set medical checks in train early enough (in those cases - almost certainly a minority - where such a procedure would otherwise seem appropriate). On the other hand, any attempts at draconian enforcement of the fourth-day rule would be of dubious value and would, at all events, appear also to be fatally undermined by the implications of the Court of First Instance in Cases T-10/99 and T-260/99 (the presumption of regularity established by the subsequent submission of a medical certificate). Possible solutions to this difficulty will therefore be considered in conjunction with the other aspects of the judgment in question (see below).

37 to 40. Implementing the right to control

As the report points out, the presumption of regularity established in the judgment of the Court of First Instance referred to above has driven a coach and horses through the institutions' undoubted (if now possibly largely theoretical) right to insist on medical checks just as, if the reasoning in the previous paragraph is correct, the right to enforce the rules on prompt submission of medical certificates is similarly undermined.

When trying to find a way out of this seeming impasse, which as the auditors point out, might well make it extremely difficult to deal with the small minority who are determined to exploit the weaknesses of the system, we must recognise that what has effectively been ruled out is a system based on declaring absences irregular in cases where the individual evades the requirement to submit to a medical check. That requirement is not in itself undermined, nor is the possibility of resorting to other forms of disciplinary sanction to enforce it. The problem, of course, is that the alternative, recourse to formal disciplinary procedures, is something of a heavy and blunt instrument and can hardly be applied indiscriminately. Even where the requirement to submit to medical checks is successfully imposed, it is, as the auditors point out (point 39), further undermined by the cumbersome mechanisms for dealing with disputes between the examining doctor and the attending physician.

In such a context, it is clear that the institutions will have to rely less on standardised mechanisms, such as the declarations of irregular absence referred to earlier, and more on management systems, such as those proposed above for Parliament, which would enable the more flagrantly problematical cases to be identified, targeted and pursued, however lengthy and cumbersome the procedures, with the kind of determination which realistically may be deployed only in a small number of cases. Such an approach would certainly not be easy and is unlikely to produce swift results. If pursued energetically, however, its pedagogical effect on the staff as a whole might well be substantial.

47. The method of dealing with invalidity

As the auditors point out, the current approach to invalidity is essentially a medical one. The measures set in the reply to points 30 to 34 of the report are designed, inter alia, to counteract this tendency. The role envisaged for the Social Affairs Division, in particular, should enable the decisions taken in individual cases to be set in a wider framework of assessment, thereby solving the problem of the theoretical reference framework to which the report alludes.

48 to 55. Monitoring, analysis of causes and corrective action

Useful work in monitoring invalidity and its causes has been done by the institution's own medical officers. The proposals set out in response to points 30 to 34 above, which should progressively favour improved monitoring, early detection and remedial action of the kind called for in points 53 to 55 of the report, are themselves a response to the medical officers' concerns. A number of other measures have also been taken, namely:

- the appointment of two psychologists to work within the Medical Service in Brussels, with a similar approach planned for Luxembourg,

- the creation, within the Social Affairs Division, of a unit tasked with analysing the causes of repetitive or prolonged absences and contacting the officials concerned with a view to seeking solutions to their problems,

- the creation, within the same division, of a unit to deal with alcohol and drug-related problems in close collaboration with the Medical Service and the institution's social workers.

These specific measures also need to be seen in the context of wider aspects of the institution's staff policy, with particular regard to its emphasis on improved human resources management by way of management training, improved career planning and guidance, better matching of skills and job profiles via the ROME-PE project, the action taken to deal with harassment and victimisation in the workplace and measures to facilitate part-time working.

Suggestions from the medical officers regarding a change of employment are followed up. Given the rigidity of the institutions' employment structures, such solutions are not always possible. The cases where they are successfully implemented are obviously not apparent in the sample available to the auditors.

56 and 57. Invoking the invalidity procedure

The report suggests that the threshold for invoking the invalidity procedure should perhaps be reduced from 12 months to six, arguing that any person who has been away from the working environment for an extended period (presumably, in this case, meaning six months or more) can seldom return to it fully. Anecdotal evidence, based on cases of long-term illness (six months or more) which have not led to invalidity procedures, would suggest that this conclusion is unduly pessimistic. Parliament therefore sees no real reason to alter, other perhaps than on a very limited and discretionary basis, the 12-month threshold.

The delays in invoking the procedure even after the 12-month period to which the auditors refer are not invariably the result of bureaucratic inertia. It is not the case that invalidity is universally sought after and, where certain individuals are battling against long-term and, possibly, even terminal illness, a decision by their institution to invoke the invalidity procedure may be profoundly damaging psychologically. In such cases, the financial interests of the institutions have to be balanced against their welfare obligations to their staff. Moreover, given that, as the report itself points out, decisions to grant invalidity pensions are almost invariably definitive and the scheme itself represents a substantial charge on the Community budget, it is not self-evident that speedier invoking of the procedure (for example, by reducing the qualifying threshold as suggested in the report's conclusions (point 82)) would necessarily always be in the best financial interests of the institutions.

65. Decision on invalidity

As was pointed out earlier (see the comment on 30 to 34), Parliament hopes that the administrative reforms that it has in mind will lead to speedier implementation of the decisions taken by invalidity committees.

66 to 69. Medical examinations after retirement on an invalidity pension

According to point 56 of the audit, 'For any doctor with experience of social services and any human resources specialist, it is an acknowledged fact that a person who has been away from the working environment for an extended period can seldom return to it fully, because of a combination of medical, psychological, technological or sociological factors'. In these circumstances, Parliament intends, as the above comments suggest, to focus its efforts on early detection and prevention of invalidity.

THE COMMISSION'S REPLIES

SUMMARY

I to VII. The Commission welcomes the Court's observations concerning the invalidity scheme. They substantially correspond to its own analysis.

It notes that the audit did not detect serious operational anomalies, such as cases of unjustified retirement on health grounds.

It shares the view that appropriate support for officials suffering from ill health, improved preventive measures, greater administrative monitoring of files, or even the creation within the institution of a multi-disciplinary standing group for monitoring each problem case, could lead to an improvement in the situation. The Commission intends to examine the feasibility of implementing the aforementioned measures in the light of the human resources at its disposal.

In any case, there would be a gradual, rather than an immediate, effect from such measures.

The Commission believes that the measures currently being developed to combat harassment at work, the pilot experiments under way in the area of family counselling, but also the reforms which it hopes will soon be adopted by the Council concerning working conditions such as the new career system, or even the new arrangements for staff reports, may help to enhance staff motivation and, as the Court noted, thereby reduce factors contributing to invalidity.

THE COURT'S PREVIOUS WORK AND THE FOLLOW-UP TO THIS

19. It is true that previous administrative measures (in particular as regards organisation) concerning absence through ill health and invalidity were insufficient. For this very reason, the Commission has proposed amendments to the Staff Regulations in its "reform" package which the Court summarised in point 71.

OBSERVATIONS ON THE ABSENCE MANAGEMENT SYSTEM IN RESPECT OF ABSENCES DUE TO ILLNESS

21. The Commission shares the view that an overall policy comprising improved preventive measures, appropriate support for officials suffering from ill health and greater administrative monitoring of files could lead to an improvement in the situation. Some aspects of the administrative reforms such as measures to combat harassment at work, psychological support, improved working conditions and the new system for career development and internal transfers should also help to enhance staff motivation.

In the specific case of individual medical support to be provided in the event of ill health, arrangements are currently being made within the Medical Service. Responsibility will lie with a psychiatrist who will be assisted by social workers; the unit will work in close cooperation with the human resources services.

The decentralisation does not mean that there is a lack of clarity Regarding the Respective roles of the Medical Service and, in the Directorates-General, of human resoursces managers and their staff who manage absences and leave.

From the central point of view, the Medical Service plays a key role since, with the aid of an "inspection" unit which is administratively highly structured, it singles out problem cases which require individual medical supervision by a doctor who is not an inspector. It should also be noted that the inspection unit shares with human resources managers the right to initiate inspections using IT applications to target the most relevant cases.

23. The method employed by the Court of Justice and by the Council would be difficult to apply at the Commission, which probably carries out 10 times as many checks. In addition, an increasing number of checks are made at the request of the DGs' heads of unit. This situation is the result of a staff-awareness campaign which will continue.

24. The decentralisation measures implemented in 1998/1999 as part of the "MAP 2000" reform had a number of negative effects which are stressed by the Court. In particular, the objective of devolving responsibility to the directorates-general was not always fully achieved in all directorates-general. The situation has improved and is continuing to improve, in particular due to the measures referred to in points 21 and 22.

25. See comments in point 24.

26 and 27. The SIC Congés application is now working satisfactorily following a major effort to improve performance.

The Sermed database is a working tool used exclusively by Medical Service inspectors. The data are, of course, subject to medical confidentiality. The tool used to manage staff absences is the SIC Congés application.

Duplication is necessary as the Medical Service is not responsible for staff absences. However, it is clear that an improvement must be sought as regards the speed at which information necessary for management is transferred from the Medical Service's inspection unit to the SIC application.

33. The Commission must point out that the examining doctors have no other internal responsibilities. Furthermore, it believes that the quality or relevance of the examinations are currently satisfactory and that both of these criteria are more important than the actual number of examinations (which are sometimes opposed on the grounds of harassment).

In addition, the Medical Service also employs the services of a medical inspection company (an outside firm working under contract) which, upon request, carries out examinations at patients' homes almost immediately.

34. The Commission believes that the relevance of the examinations is of more value than their frequency.

In this context, it has at the end of 2001 developed a system for improving the "targeting" of the examinations carried out at the initiative of the inspection unit. It also believes that the departments to which staff are assigned are often best placed to draw attention to the most interesting cases. The recent increase in inspection requests by the various departments is a favourable aspect in terms of the relevance of the examinations.

35, 36 and 38. The Commission does not believe that the rules for submitting medical certificates are applied too flexibly. On the contrary, it takes the view that the rules are applied so strictly that an increasing number of complaints have been made under Article 90 of the Staff Regulations.

As regards the deadlines for submitting certificates, the Commission undertakes to quantify the expression "as soon as possible".

It fears that greater severity would lead to complaints or even appeals to the Court of First Instance, which unfortunately does not usually find in its favour in this area.

37. Despite the lack of specific rules in the Staff Regulations, the Medical Service recommends in such cases that the Appointing Authority should declare the individual to be absent without justification. Consequently, Article 60(1) of the Staff Regulations applies: days of annual leave may be deducted or a proportion of the salary docked, without prejudice to other disciplinary measures (certain cases are referred to the Investigatory and Disciplinary Office (IDOC)).

39. The Commission shares the view that the current procedure for disputes concerning absence through ill health is inappropriate. An overall review should no doubt be envisaged.

OBSERVATIONS CONCERNING INVALIDITY PENSIONS

41. The Commission has not published figures for the long-term cost of invalidity, although this cost does appear in the budget under a specific heading. The information is therefore in the public domain and, as such, is available to staff.

In this context, it should be borne in mind that the staff have voiced opposition, through certain trade unions, to the amendments proposed by the Commission as part of the reform.

The Commission doubts that an awareness of the cost of invalidity and of the resulting expenditure would reduce the number of instances of retirement through ill health. The respective decisions are based on medical assessments and are taken on a case-by-case basis.

42. An overview of the cost of the "invalidity" system in terms of long-term debt could be produced, as the Commission provides an estimate each year of the total liabilities of the pension scheme. By evaluating the effect of each component of the system, in particular for invalidity, an overview could be produced when the next actuarial study of the system is carried out.

It is true that once a beneficiary reaches a certain age, he/she should receive a retirement pension. Consequently, some of the expenditure currently listed under invalidity pensions should be listed under retirement pensions. The matter will be resolved by the Commission's proposed reform, which establishes a distinction between the actual invalidity payment and the retirement pension.

43. The Commission would like to point out that, as is the case in each Member State, an invalidity scheme is absolutely necessary. In the absence of such a scheme the productivity of an invalid is considerably lower than that of a healthy official. The model applied by the Court to calculate the net cost of invalidity is based on the assumption that the official concerned, had continued his career until the normal age of retirement.

45. It is true that the current system only recognises total invalidity. On a human level, the Commission acknowledges the value of "partial invalidity" which would enable the individual to remain in the workplace. However, under the current system where the establishment plan contains "full" posts, such posts would be filled by employees working part-time, which in principle would not be in the department's interests as it would not have a full complement of staff at its disposal.

Nevertheless, the Commission has proposed a specific provision as part of the reform of the Staff Regulations which would authorise an official over 55 to prepare for retirement by working part time.

47. As far as the Commission is concerned, the observation that invalidity is not currently regarded as total inability from the medical point of view to discharge the duties required of an official is pertinent.

A solution to this problem must no doubt be sought before the matter is referred to the Invalidity Committee. Action has already been taken to this effect: the duties of the Internal Ombudsman have been redefined, a specialised unit has been set up within DG ADMIN with responsibility, inter alia, for career guidance (the Central Department for Professional Guidance and Career Development (SCOP)); an attempt is often made to find a more suitable position for the official concerned. SCOP will, of course, play a part in the structure mentioned in the replies to the Summary (third indent).

48. The Commission is endeavouring to introduce a simple system of statistical analysis.

52. The Commission believes that the measures currently being developed to combat harassment at work, the pilot experiments under way in the area of family counselling, but also the reforms which it hopes will soon be adopted by the Council concerning working conditions such as the new career system, or even the new arrangements for staff reports, may help to enhance staff motivation and, as the Court noted, thereby reduce factors contributing to invalidity.

53. The Commission wholly agrees with this analysis. It believes that over and above the measures referred to in point 21 administrative tools should also be developed (e.g. authorisation by the budgetary authority of "transitional" posts which would make it easier for officials to return to their posts).

In addition, it is currently setting up a "medical monitoring" facility (see comments on the Summary).

55. The rules in force at present, and in particular Article 78(1) of the Staff Regulations, which states that an official is entitled to an invalidity pension "in the case of total permanent invalidity preventing him from performing the duties corresponding to a post in his career bracket", limit the range of alternatives. Without questioning the validity of the sample, the Commission believes it is difficult to accept conclusions reached on the basis of impressions given by individuals who decide, after the event, that their invalidity could have been avoided if they had been offered alternative solutions. Consequently, the Commission does not consider that savings in the amount estimated by the Court could be made. All the same, the Commission believes that, following the measures that will be adopted in the context of the Reform, it will be possible to make savings in future, compared with the present situation.

56. The Commission agrees with this interpretation, but has doubts about the short-term implementation of suggestions which it believes would require the Staff Regulations to be amended.

57. The Commission believes that since the Court of Auditors' audit, significant progress has been made as regards the periods involved.

As things stand, the medical status of officials suffering from ill health is systematically examined as soon as the total period of absence over the previous 36 months exceeds 350 days.

65. These delays have recently been cut by approximately 50 %.

70. The Commission stresses that its proposal to modify the invalidity scheme essentially aims to establish a more coherent system and to rectify certain discrepancies, such as the fact that some invalidity pensions may be larger than retirement pensions.

73. The Medical Service took part in the preparatory phase of the Reform. In addition, the comments made concern behaviour or administrative structures rather than the reform of the Staff Regulations. The Commission is endeavouring to act accordingly (see comments on the Summary).

CONCLUSIONS

76. The Commission accepts the comments made but believes that it is important that the concept of invalidity should remain first and foremost a medical one. As was suggested elsewhere, the involvement of other parties must be considered before the procedure is initiated.

77. Progress has been made and will continue to be made as regards delays (see comment on point 65).

RECOMMENDATIONS

80. As stated above (point 21), the Commission, in the framework of an overall policy, is going to step up its efforts to inform and raise awareness in the directorates-general. It is introducing a counselling and support facility. Once the multi-disciplinary group responsible for monitoring individual cases (as mentioned in the replies to the summary) has been operational for a significant period, it will decide whether it is appropriate to introduce a broader or more ambitious plan of action while bearing in mind that large-scale measures call for resources which are currently unavailable.

81. The training programme for heads of unit already includes a course on staff management (and therefore covers absences).

A system for monitoring the causes of invalidity will be swiftly introduced.

82. See comment on point 73. The Commission is not convinced that altering the composition of the Invalidity Commission would have significant effects.

REPLIES OF THE COURT OF JUSTICE

The Court of Justice notes that the report does not contain criticisms specifically levelled at the way in which its administration manages the invalidity scheme or the related administrative and medical spheres.

It would, however, point out in connection with the observations relating to the monitoring of absences on account of illness that its administration pays particular attention to the administrative and medical supervision of officials and other members of staff on long-term sick-leave. Such absences may conceal professional or personal problems, and in certain cases even lead to future invalidity.

With that in mind, an informal group (the medical officer, the nurse, the welfare assistant, the Director of Personnel and Finance, the Head of the Personnel Division and the Principal Administrator responsible for social and medical matters) meets, in principle once a week, to discuss those absences and in particular whether or not it is necessary in each case to have a medical examination carried out.

Last, it is to be noted that the Court's administration is at present in the final stages of appointing an external doctor to carry out those medical examinations.

REPLY BY THE COURT OF AUDITORS

The Court of Auditors has carefully noted the observations made in this report. It has already taken steps, in particular to improve the monitoring of these absences. Because of its size and the limited resources at its disposal, the Court cooperates wherever possible with the competent departments at the Commission.

THE ECONOMIC AND SOCIAL COMMITTEE'S REPLIES

The Economic and Social Committee agrees with the Court of Auditors about several points raised in the preliminary observations, and in particular are the following:

- checks on the circumstances leading to invalidity, with review where necessary,

- simplification and easing of the measures proposed by the reform in the event that a check on absence due to illness is challenged.

The Economic and Social Committee presented amendment proposals to the Interinstitutional Staff Regulations Committee in respect of these matters in early 2002. The proposals were then sent to the rapporteurs of the various European Parliament committees that are required to comment on the reform.

In the same vein, the Economic and Social Committee would unreservedly support the introduction of additional measures such as those proposed in the Court of Auditors' observations, especially:

- an overall policy to intensify checks on absences due to illness (during the past two years these have been carried out far more frequently at this institution),

- clear and automatic measures in the event of refusals to submit to a check on absence due to illness,

- the analysis, preferably at interinstitutional level, of the main causes of invalidity, and increased monitoring.

THE COMMITTEE OF THE REGIONS' REPLY

The Committee of the Regions shares the Court's opinion on the subject and has no further observations to make.