Abstract
Thalidomide was first synthesized in 1953 and was subsequently marketed as a mild hypnotic and sedative in more than 20 countries. By 2001 it was estimated that there were 5000 survivors from the 10,000 – 12,000 babies who were, as a result, born with severe abnormalities. For these survivors, recent concerns have emerged about their physical state, in particular their levels of pain and their ability to maintain independence. It was therefore wished to ascertain health status and current concerns amongst a sample of survivors living in the UK. A combination of qualitative and quantitative methodologies was employed. Qualitative interviews were undertaken with a sample drawn from the population affected by thalidomide known to the Thalidomide Trust. Selection of participants was based upon a theoretical sample frame. Postal questionnaires to examine health status and various psychosocial aspects concerned with living with the consequences of thalidomide were sent subsequently to these same participants and to a random sample of those not originally drawn for the qualitative interviews. For the qualitative interviews, 28 agreed to take part; two refused and the remainder did not respond. Those agreeing to interview were representative of the original compensation bands (Chi-Square = 3.929; p = 0.416). Several themes emerged from these interviews, including the effects on work and career; coping in terms of attitude to life, self-image, confidence, self-esteem, stress and emotion; relationships, independent living issues and emergent problems such as pain, quality of life, and anxiety about the future. A postal questionnaire was then sent to those who had agreed to interview (28), plus a random sample of the remaining group who were not initially chosen for the qualitative interviews. In total 82 people were sent the questionnaire, of whom 41 (50%) responded. Two-thirds of responders were female. Seven out of ten lived with a partner, and over half (56%) had children. Almost half (46%) were in work, but 32% reported they were permanently unable to work because of disability. Current levels of impairment were found to be similar across groups defined by the original compensation band. In contrast, the activity limitation measures showed a steep gradient across bands but only 37% considered themselves disabled. Nevertheless, despite the restriction in activities for some, levels of participation were similar across bands; likewise simple summary items on health status and quality of life were similar and 70% reported their quality of life (QoL) was good or better than good. Yet nine in ten believed that their body was less flexible than in the past. Almost as many reported they were less able to carry things. It turns out that when compensation bands were grouped (1 – 3, 4 – 5) to highlight those most severely affected according to the original assessment, then those in the higher band grouping reported significantly more musculoskeletal problems, high levels of fatigue and increasing dependency and feelings of vulnerability. It is clear that the original ranking of disability severity, as expressed through the compensation, bands (allocated in early childhood in most cases), is consistent with current ranking of limitations in activity and participation. Nevertheless, despite high levels of disability amongst some survivors, lifelong adjustments to the original impairments have resulted in more than two thirds reporting at least a good quality of life. However, survivors expressed increasing concern about emerging musculoskeletal and other problems which may compromise hard-won independence.
Introduction
Thalidomide was first synthesised in 1953 and was subsequently marketed as a mild hypnotic and sedative in more than 20 countries. It was claimed to be safer than alternatives. At the time barbiturates were the main alternative (Kelsey 1967). However, reports and letters began to appear (Lenz et al. 1962; McBride 1964) indicating that the drug used by pregnant woman was associated with severe foetal deformities, mostly of limbs. By 2001 it was estimated that there were 5000 survivors from the 10,000 – 12,000 babies who were born with these severe abnormalities (Warren 2001). For these survivors, recent concerns have emerged about their physical state, in particular their levels of pain and their ability to maintain independence (Kennelly et al. 2002).
As a consequence, the Thalidomide Trust commissioned the authors of this paper to review the current situation and consider whether assessments made during childhood to determine the level of benefits awarded, remain appropriate. Thus a number of questions were formulated with respect to ascertaining the current health status of those living with the consequences of Thalidomide; whether current levels of activity limitation were consistent with the original expectations as expressed through compensation bands (levels of expected disability), and whether commonly used standardized questionnaires could be used to review the situation given evidence of further deterioration.
In summary, the primary aims of this study were (a) to document current health status and any emerging problems in those with Thalidomide impairments in the UK and (b) to consider whether compensation bands based on childhood assessments of impairment and predictions of disability remain appropriate or whether new tests should be introduced to replace them.
Methods
A two-stage methodology was employed, using qualitative (Stage 1) and quantitative (Stage 2) methodologies. At both stages potential respondents were contacted through the Thalidomide Trust by letter and were free to decline participation without compromising status, and could seek further information through a free-phone number.
Stage 1: Qualitative interviews
From the names given by the Trust, a purposive sample was drawn. The framework for this sample included gender and the five levels of impairment related to the original compensation awards. In each of these 10 cells (gender by impairment) there were to be three respondents. Potential respondents were excluded from the primary sampling frame if they had difficulties of comprehension or communication which would require the help of a third party. Appointments for the interviews were made by telephone for a time and place to suit the interviewee and took place throughout the UK. Interviews were unstructured, and allowed the respondent to reflect on their lives, and their current circumstances. A sample of 70 individuals was drawn to provide the 30 respondents needed. Following the interview, the taped interviews were transcribed and analysed using QRS NUD∗ST software (Fielding and Lee 1991).
Stage 2: Quantitative methodologies
Stage 2 comprised a postal questionnaire measuring health status and quality of life. Stage 1 respondents were included as well as a further random sample of 1 in 6 beneficiaries, 54 possible additional subjects. The inclusion of Stage 1 participants was deliberate to ensure that the health status and quality of life of those who participated in the qualitative interviews was representative of the wider group, with no differences in health status in successive samples (i.e. to those additional respondents drawn for Stage 2).
Standardized measures were chosen to assess impairment, activity limitation; participation and quality of life. At the level of impairment, Self esteem was measured by the 10 item Rosenberg Self-Esteem Scale (Rosenberg 1995) giving a score range of 10 to 40, with higher scores representing higher self esteem. Stress was measured using the Perceived Stress Scale. This 14 item scale measures the degree to which respondents find their lives unpredictable, uncontrollable and over burdened (Cohen et al. 1983). Perceived self-efficacy was measured by the 10 item General Perceived Self-Efficacy Scale. A low score is associated with helplessness (Schwarzer 1993). Pain and fatigue were measured using the Bodily Pain and Vitality dimensions of the SF 36 where a high score represents no pain and maximum energy (Ware and Sherbourne 1992).
At the level of activity limitation (disability) the postal Barthel Index (Gompertz et al. 1994) [range 0 – 20 with 0 representing worst disability] and the Physical Function Scale of the SF36 [range 0 – 100 with 0 as no physical function] were used (Ware and Sherbourne 1992). For participation (handicap) the London Handicap Scale was chosen with a score range of 0 – 100, with a low score indicating no participation (Harwood et al. 1994). Overall, health status was measured by the EuroQoL (EuroQol Group, 1990) where a value of 1 represents perfect health. Although a zero score indicates death, it is possible to have a state worse than death (i.e. a negative score). Finally, subjective quality of life was assessed by the General Well-Being Index (McKenna and Hunt 1992). These questionnaires were put together into a postal booklet, which also contained a set of items for a draft ‘Late Effects of Thalidomide’ questionnaire, the contents of which were derived from the qualitative interviews using a standard protocol (Britten 1995).
The study was commissioned by the Thalidomide Trust but was nevertheless presented to and accepted by the Ethics Committee of the United Leeds Teaching Hospitals Trust.
Results
Stage 1: The qualitative study
Of the 70 people drawn in the first sample, with a target of 30 interviews, 28 (12 men and 16 women) agreed to take part, 2 refused and the remainder did not reply. The 28 who participated were found to be representative of the original compensation bands (levels of impairment) with no significant difference in agreement to interview across compensation bands (Chi-Square = 3.929; p = 0.416). The main themes which emerged from interviews were:
Life history (an account of childhood, schooling, education and training). The influence of the impairment on upbringing, attitudes and money issues. Effects on work and career. Coping in terms of attitude to life, self-image, confidence, self-esteem, stress and emotion. Relationships (with partners, family, friends). Independent living issues. Problems which have emerged later such as pain, quality of life, anxiety about the future.
Given the emphasis of the study was on current issues, this paper focuses mainly on themes 3 – 7 and, within the context of this paper, a limited set of illustrative quotations have been chosen for each theme.
Theme 3: The impact on work and career
This impact on work and career was found to be considerable. For example, one respondent said: “I did have one chap who interviewed me who was a fairly senior director for a fairly well known company and he was aware of Thalidomide but …. The company wasn't prepared to take the risk in terms of future liability on the impact of Thalidomide”. “On one occasion, even worse in some ways, was a job that I applied for and the bloke said to me one of the problems is that in this job you will have to go and see clients a lot. We have lunch with clients and they might not like being with you while you're eating”. “My one passion used to be I wanted to be a vet, but I was shot down in flames over that one because of my disability”. “it was a worrying time because I'd never had a job, not even a Saturday job because they all needed hands, so it was a bit worrying but it proved to be much easier to get a first job than I thought it would be. It was then more difficult to get out of that job when I realised I didn't like it, I was there for three and a half years”. “Yes, there have been times in certainly the last three or four years where I've had to cut the work down to four even three days a week just to let the joints rest”.
Theme 4: Coping
Many had strategies for dealing with the difficulties they encountered: “I've always had an attitude that when a situation arises I'll cope with it. I don't know how but I will. As you get older the more you realise, the more unusual situations you've experienced and you haven't fallen flat on your face yet”. “On the surface you can appear as confident as you like but inside your self image doesn't change because you can't change yourself, you are as you are and you either accept it which I've never done”. “ … that I do feel that the stress of me trying harder, I think, and combined with the physical deficiencies that I live with, I think it's left me where my health isn't as good as it could be”.
Theme 5: Relationships
The majority interviewed were married or had partners. Evidence exists to suggest that some had difficulties in starting or maintaining a relationship, but by no mean all did so: “I've never had a problem that concerned my disability. I don't think so anyway. I'm married again, this is my second time”. “… Always a problem I mean a lot of other male teenagers didn't want to be seen hanging round with a disabled kid and it was certainly true with girls. They didn't want to have a boyfriend who had no arms so they were difficult times”. “and I'm still very sensitive that disability is a factor but I'm still very sensitive in the sense that if things go wrong in relationship I tend to blame my disability”. “I wore a urine bag in my side and that had to be emptied every two or three hours so my brother had to know how to do that … and I think bitterly resented a lot of what had to be done and the way it fell on his shoulders, a lot of pressure and a lot of problems”.
Theme 6: Independence issues
Many issues emerged in relation to maintaining independence. For example, clothing had to be carefully chosen and help might be needed in dressing: “I think you learn certain clothes that are a ‘no no’ and you know you're not going to manage them so you learn what you need to buy that will be most useful and can help you to dress independently”. “I drive my own car, there's no way I could manage on public transport. My car is vitally important to me …. I've had the car since I passed my test when I was sixteen and that's how I get about”. “The only thing I refuse to use is the bus. I find it rather embarrassing when they have to get the lift down and you've got people looking thinking I'm making them late”.
Theme 7: Later and emerging problems
Several themes emerged which suggest that the group are experiencing increasing problems with pain, current needs and quality of life. These led to a real concern for the future: “Just the arthritis and the fact that I'm getting wrinkles, but I'd like to be a lot more mobile than I am. The problem is trying to keep moving all the time, I'm trying to get back to work at the moment”. “I have reach problems and my spine is affected, it's fused in three places and it gives me awful problems when I have to go to the dentist, I'm in absolute agony”.
Concerns expressed by more than 60% of sample through ‘late effects’ items (expressed as % of total sample).
Stage 2: The quantitative study
Twenty-two of those interviewed in Stage 1 also returned the postal questionnaire booklet for Stage 2 (i.e. 6 of those interviewed failed to do so). Of the 54 people additional people drawn for the Stage 2 sample, 19 subsequently returned the questionnaire booklet giving a total response of 41 out of 82 (50%). Responders were not significantly different to non-responders with respect to compensation bands (Chi-Square = 2.67; p = .0.614). There was also no significant difference in the distribution of compensation bands between responders who had taken part in stage 1 and those drawn specifically for, and subsequently participating in Stage 2 (Chi-Square = 6.506; p = 0.164). The samples were thus deemed to have come from the same population.
Two-thirds of responders were female; seven out of ten lived with a partner, and over half (56%) had children. Almost half (46%) were in work, but 32% reported they were permanently unable to work because of disability. For impairment, pain and fatigue (vitality – SF36) showed no discrimination across compensation bands, although all figures were well below normative scores for the age group (Brazier et al. 1992) (Table II). Cohen's Perceived Stress scale showed an average 38.7 (SD = 9.8) but no significant difference across bands (Kruskal Wallis: p = 0.184), as did the General Self Efficacy Scale (p = 0.589). Thus current levels of impairment, although substantial, do not vary by compensation band.
Impairment, activity limitation, participation, health status and quality of life across compensation bands. Expressed as median scores (and inter-quartile ranges).
In contrast, the activity limitation measures showed a steep gradient across bands. The SF36 Physical Function scale had a mean of 69.7 (SD = 37.9) which is well below population norms for this age, and significant for band (Kruskal Wallis: p = 0.003). The Barthel Index averaged 15.2 (SD = 4.9), and again was significant for the band (p = 0.029). Levels of participation from the SF36 were lower than normative scores, but they, along with the London Handicap scale, did not discriminate across bands; neither did simple summary items on health status and quality of life.
The median EuroQoL score of 0.59 (IQR = 0.21 – 0.69) suggests poor health status but despite this, and the low levels of activity participation observed, only 37% considered themselves disabled, 59% reported their health to be good and 70% reported their QoL was good or better than good. Indeed two respondents reported their quality of life to be ‘good’ yet their EuroQoL state was worse than death! This nicely illustrates the difference in the measurement of health status (EuroQoL) and subjective quality of life.
Those who did consider themselves to be disabled had significantly lower levels of physical function on the SF 36 than those not regarding themselves as such (Mann-Whitney U: p = 0.049). Levels of self efficacy were also significantly lower in this group (Mann-Whitney U: p = 0.02) but otherwise all measures of impairment, function, health and QoL did not vary significantly by perception of disability.
Yet 90% believed that their body was less flexible than in the past. Almost as many reported they were less able to carry things. It turns out that when bands were grouped (1 – 3, 4 – 5) to highlight those most severely affected according to the original assessment, then those in the higher band grouping reported significantly more musculoskeletal problems, high levels of fatigue and increasing dependency and feelings of vulnerability (Table III).
Late effects that differentiate across original impairment groups (Mann Whitney U < 0.05) (expressed as % within groups and in total).
Discussion
The study's purposive sample selected from across the range of compensation bands, confirmed previous anecdotal reports collated by the Thalidomide Trust that many were experiencing deterioration in their health. Pain and joint stiffness were reported by over 80% of those interviewed and both impinged on independence and ability to work. This may reflect early degenerative change in joints that have had abnormal contours since childhood. However, people who are limb deficient may be obliged to overuse their remaining joints or use them for unusual tasks. For example, an upper limb-deficient person may use the feet used for prehension and feeding.
Some problems may be almost unique to those with Thalidomide-related limb abnormalities where, typically, there are either symmetrical upper limb deficiencies or problems affecting all four limbs. Others may be shared by those with upper limb deficiencies arising from other aetiologies. A recent review of 92 such patients with limb loss (mainly due to trauma) at or above the elbow found that 40 – 45% experienced subsequent pain affecting the upper spine or contralateral shoulder, whilst a quarter experienced pain in the contralateral elbow, wrist or hand. The authors attributed these symptoms to overuse of the intact limb (Datta et al. 2004).
However, musculoskeletal problems have also been reported in those affected by childhood polio, and form part of the spectrum of problems grouped under the label, “Post-polio syndrome” (McGaughey et al. 2000). Adults who are ambulant, but whose gait is affected by cerebral palsy, also experience musculoskeletal symptoms and a deterioration in mobility as they grow older, again probably reflecting an energy demanding gait and altered use of their limbs leading to early onset musculoskeletal problems (Jahnsen et al. 2004). Some have recommended that those with limb deficiencies are counselled against overuse of remaining limbs (Jones and Davidson 1999). However, this strategy alone may place unacceptable constraints on independence and lifestyle. Instead, the population affected by thalidomide has been very active in pursuing the use of new technologies to assist and promote independence.
The authors also explored the relationship between the compensation bands based on childhood assessments and current levels of health status and quality of life, using a range of standard tests. It is clear that the original judgement about expected disability is consistent with current levels as expressed by both the Barthel index and the SF36 Physical Function scale. This would imply that either could be used if it were felt necessary to re-assess, given further deterioration (as opposed to a full clinical examination). Nevertheless, despite the clear gradient across bands, outliers are evident such that some individuals have levels of disability worse than expected for their compensation band, and thus reappraisal would seem appropriate even now.
This study has a number of limitations. The number of subjects was inevitably small. There are only around 400 people known with Thalidomide-related problems known to the Thalidomide Trust in the UK. The authors deliberately over-sampled at each stage of the study, knowing that participation rates were likely to be low. Only 30 out of 70 (43%) drawn for the initial qualitative interviews agreed to take part, and then a further two became not available, making the overall Stage 1 response to be just 40%. Likewise Stage 2 only achieved a 50% response rate. However, the samples drawn were representative of the original compensation bands. Likewise, the level of compensation band for those who participated in stage 1 did not differ from those in Stage 2 so there is reasonably confidence that the findings can be extrapolated to the whole group. It is unknown whether or not those refusing to take part are, in any way, qualitatively different than those who did, despite representation of compensation bands. The authors would also have liked to model the impact of impairment and activity limitation upon participation and quality of life, and to identify mediating factors such as levels of compensation, but our numbers were too small for this.
The emergence of Thalidomide related limb deficiencies acted as a spur to the development of new prosthetic limb technologies in the 1960s. Initially, gas-powered prostheses were developed. These were fraught with problems and eventually superseded by myoelectric prostheses. Although few of those with Thalidomide-related upper limb problems continued to use prostheses, others with limb loss or deficiencies arising from different aetiologies continue to take advantage of the prosthetic designs developed then (Millstein et al. 1986).
It now seems that a new range of strategies will need to be developed to address the emerging musculoskeletal and other problems in the population affected by Thalidomide. The lessons learnt from this small group of people may well help to respond to similar needs in the much larger populations affected by cerebral palsy and childhood polio.
Surprisingly, there was no obvious relationship between compensation bands and measures of self-esteem, self-efficacy, impairment and well-being. A majority were married or had partners and many were in paid employment. Furthermore, most considered their quality of life to be good and most did not consider themselves as disabled. This determined group of people who have survived thus far seem to have defied many of the gloomy predictions made in the 1960s about the lives they were likely to lead.
Footnotes
Acknowledgements
The authors wish to thank the Thalidomide Trust for sponsoring this research, and all those living with the consequences of thalidomide who gave their time to this study.