Retirement on grounds of ill health: cross sectional survey in six organisations in united kingdom
BMJ 1997; 314 doi: https://doi.org/10.1136/bmj.314.7085.929 (Published 29 March 1997) Cite this as: BMJ 1997;314:929- C J M Poole, consultant occupational physiciana
- Accepted 24 January 1997
Abstract
Objective: To assess the process and outcome of retirement due to ill health in six large organisations.
Design: Cross sectional study of the rate of retirement due to ill health by age, sex, and length of service. Principal diagnoses by age and length of service were also compared.
Setting: Four public and two private large employers in the United Kingdom.
Main outcome measures: Rates of retirement on the grounds of ill health by age, sex, and length of service of employees contributing to pension schemes.
Results: Rates of ill health retirement varied from 20 to 250 per 10 000 contributing members, and in two organisations the rate varied geographically within the same organisation. In the two organisations that provided data by sex, women retired at a greater rate than men under age 40 and over age 50. In four organisations the modal age or length of service coincided with enhancements in benefits. In the four that provided information on diagnoses, musculoskeletal and minor psychiatric illnesses were the most common reasons for retirement.
Conclusion: The granting of ill health retirement benefits may not be determined by illness. There is a need for some employers and pension schemes to improve their processes for granting benefits. Doctors should be wary of conflicts of interest and work to guidelines when they advise pension schemes about the merits of an application for benefits.
Key messages
The rate of retirement due to ill health varies greatly between organisations and may even vary within the same organisation
Applicants for ill health retirement may be motivated more by financial benefits than by ill health
Women may retire at a greater rate than men before age 40 and after age 50
Some pension funds need to improve their processes for granting ill health retirement benefits
Doctors should beware of conflicts of interest and work to guidelines when advising pension funds about the merits of an application for benefits
Introduction
Applications to retire from occupational pension schemes before the normal retirement age have increased over the past 15 years.1 Of the 45% of people who retire early, about a third will do so because of ill health.1 The criteria for awarding a benefit and the process and size of the benefit vary between pension schemes. The criterion may be as stringent as “permanent incapacity which is likely to prevent any gainful employment” or as loose as “incapacity which prevents the employee undertaking regular and effective duties.” The process may entail a report from the applicant's general practitioner or the company's medical officer only or from a doctor who is less likely to have a conflict of interest.
Benefits are usually related to the length of service of the applicant and take the form of an increase in the number of pensionable years of service once a minimum number of years has been spent in the organisation (less generous) or an increase in pensionable service as if the applicant had worked to normal retirement age (more generous). During periods of company rationalisation employees may be given financial incentives to leave by taking premature retirement or voluntary redundancy or even by being compulsorily made redundant, which will affect the number of applications for ill health retirement.
Doctors are often asked to advise on whether an applicant fulfils the criteria for ill health retirement benefits, but little is published to assist with these judgments.2 3 There is anecdotal evidence from employees and employers that decisions about ill health retirement may not always be fair, and there is evidence of poor correlation between doctors when they assess case scenarios for retirement benefits.4
I undertook a cross sectional study of six large organisations in the United Kingdom to determine the effects of age, sex, length of service, and diagnosis on retirement due to ill health.
Methods
Data were requested on numbers, age, sex, length of service, and principal diagnosis for employees retiring with ill health from Rover (a car manufacturer), the fire, police, and ambulance services, the Post Office, and the Teachers Pensions Agency for the period 1990-5. Details on age and length of service by sex of members contributing to the pension schemes were also requested. The criteria and number of doctors involved in the process were compared. It was agreed that because of the sensitivity of the data the results would be anonymised so that individual organisations could not be separately identified. Letters were used to represent each organisation.
Data analysis was by frequency distribution, 2 and Mann-Whitney tests, and logistic regression as appropriate.
Results
Rates of retirement due to ill health varied more than 10-fold between organisations (table 1). In two organisations, as a proportion of all retirements (1994-5) it varied geographically within the same organisation–from 37% (7/19) to 100% (11/11) in B and from 13% (8/60) to 69% (124/179) in C. Table 1) also shows modal and median ages and lengths of service at retirement. Rates of retirement generally increased with age. In four of the organisations–B, C, D, and E–the mode for age or length of service coincided with enhancements in benefits (table 1). In B, C, and D the modes coincided with maximum enhancement in benefits. By contrast, maximum benefit was paid to only three employees in organisation F (1994-5).
In organisation C the median (interquartile range) age for retirement was 47 (9.0) years for men and 31 (6.0) years for women (P<0.001). The respective median lengths of service were 25 (10.0) years and 12 (6.0) years (P<0.001; fig 1). Median ages of men and women in organisation C were 37 (12.0) and 30 (8.0) years, respectively. When the data were analysed in five year age bands by sex, the proportion of women retiring on grounds of health was greater than the proportion of men between ages 21 and 40 years (table 2). Table 3) shows the same data for organisation D, with women retiring at a greater rate than men at ages 26-30 and 51-60.
The criterion for retirement on the grounds of ill health varied between organisations. In A, B, and F ill health had to be permanent and prevent applicants from doing their job; in C ill health had to be permanent and prevent the applicants from doing any job within the organisation (not just the one they were doing at the time); in D ill health needed to last only a year or two; and in E ill health needed only to make applicants incapable of regularly and effectively undertaking the duties of their grade.
In B, C, and E retirement on the grounds of ill health could be granted with supporting evidence from just one doctor, either the applicant's general practitioner or the organisation's medical officer. This compared with two doctors in D and F (usually the general practitioner and a benefits agency doctor) and three doctors in A (the general practitioner, a company occupational physician or medical officer, and an independent occupational physician).
In the four organisations that provided the principal medical diagnoses (A, B, E, and F) musculoskeletal problems affecting the back or joints and minor mental ill health such as stress, anxiety, and depression were the most common reasons for granting ill health retirement benefits. When data for B and E (both predominantly manual workers) were analysed in five year age bands the ratio of diagnoses of musculoskeletal conditions and mental ill health to those of cardiovascular and respiratory conditions by age or length of service ranged from 4:1 to 10:1 (table 4 and table 5, respectively), whereas in A (also predominantly manual workers) the ratio was 1 or less (table 6).
Discussion
The greater rate of ill health retirement before age 40 and after age 50 in women compared with men has not, to my knowledge, previously been reported. The relative risk was almost fivefold in organisation C between ages 26 and 35 years. I cannot identify the reason for this sex difference from this study, but it needs to be examined if expensive training and pension costs are not to be wasted. Women have been recruited increasingly into this organisation over the past 20 years and now form 14% of the workforce. The lower median age of the women in C is unlikely to be the reason for their higher rate of retirement as even those with short service and under age 30 were retiring at four times the rate of men. D was the only other organisation that provided sex specific data. Women retired at a greater rate than men between ages 26 and 30, which coincides with a doubling of reckonable service–for example, five to 10 years–for pension benefits. Women also retired at a greater rate over age 50, but men retired at a greater rate between ages 46 and 50, when ischaemic heart disease is particularly prevalent in men.5
The concurrence of modes of ill health retirement by age or length of service with enhancements in benefits in four of the six organisations (B, C, D, and E) probably reflects an understandable desire by retiring employees to secure the optimum pension possible.
Peaks of ill health
There is no medical reason why ill health should peak at these times, though it is possible that employees may carry their ailments for some time before presenting them to a doctor for the purposes of securing ill health retirement. The small peaks at ages 45, 50, and 55 that were seen in A may have occurred for similar reasons at these psychological milestones. A gradual increase in the rate of ill health retirement similar to that of organisation A would be expected if medical reasons were the main factors that determined applications for benefits. The carrying of ill health is probably easier with musculoskeletal problems (such as joint or back pain) and psychiatric problems (such as stress, anxiety, or depression) than with cardiovascular or respiratory illnesses. The comparatively high proportion of musculoskeletal and psychiatric diagnoses around the modes in some of the organisations suggests that this is the case. By contrast, it is unfair that so few employees in organisation F achieved maximum pension benefits.
Influence of non-medical factors
Support for the notion of non-medical factors influencing applications for benefits comes from studies of patients who have undergone coronary artery bypass grafting. The proportion of abnormal electrocardiograms was found to be no different in those who received benefits from those who returned to work,6 and social, economic, or psychological rather than medical reasons have been reported as the principal reasons for not returning to work after coronary artery bypass grafting.7
Variations between organisations in the proportions of employees leaving by retiring on the grounds of ill health may simply be a reflection of the different ways in which employment contracts are terminated. A company with a low rate of retirement due to ill health may use other methods of dismissing staff, such as redundancy, premature retirement, or frustration of contract, as is the case with organisation A. By comparison, organisation C rarely makes use of these other methods, preferring the ill health retirement route. This has the effect of “medicalising” dissatisfaction, generating consultations with doctors and expensive investigations for what are comparatively minor illnesses.
Conflicts of interest
Wide variations in the proportions of employees leaving by retiring on the grounds of ill health within the same organisation (B and C) suggest that the process is out of control and in need of audit. Possible factors are pressure from management or unions for employees who leave to opt for ill health retirement and inconsistent judgments by doctors. The latter could be dealt with by doctors working to guidelines3 in which they had had some training. The decision to award ill health retirement is best not left to one doctor (whether the applicant's general practitioner or the organisation's occupational physician), who could be placed in a position of conflict of interest, but to a second doctor who is external to the organisation, trained in occupational health, and working to guidelines.
Acknowledgments
I thank the many people from the six organisations studied who supplied me with data; Ian Calvert at the Institute of Occupational Health, Birmingham, for statistical advice; and Paul Sims, medical illustration department, Dudley Group of Hospitals NHS Trust.
Funding: No external funding.
Conflict of interest; None.