Neuromuscular disorders in general and the ability to work

Whilst it is possible for some persons with neuromuscular disorders to work, the proportion who have never been employed plus those who are not currently working is more than half. For example, Bostrom, Netterlund, and Ahlstrom (Reference Bostrom, Natterlund and Ahlstrom2005) undertook a longitudinal study and followed 44 people with muscular dystrophy over a 10-year period (1991 and 2001). The number of people who were working dropped from 21 in 1991 to 11 in 2001. In 2001, six were working full-time and five were working part-time. Fowler et al. (Reference Fowler, Abresch, Koch, Brewer, Bowden and Wanlass1997) compared employment outcomes with the perceptions of 175 rehabilitation counselors and 76 physicians. Counselors (95%) thought that clients with neuromuscular conditions could work in the open labor market and just over half of the counselors estimated success rate at 30% or less. Almost all physicians (98%) thought that persons with neuromuscular conditions had vocational potential, but in that study, 92 out of 154 individuals (approximately 60%) were not working.

The broad outline of a picture of quasi-employment potential is confirmed by other studies. Andries et al. (Reference Andries, Wevers, Wintzen, Busch, Howeler, de Jager and Wokke1997) analyzed the work situation of 1072 Dutch patients with a neuromuscular disorder. They reported that “a labour career is in reach of most NMD [neuromuscular disorder] patients” (p. 255), but labor force participation decreases after age 34. The conditions selected were dystrophia myotonic, hereditary motor and sensory neuropathy, spinal muscular atrophy, and myasthenia gravis. In Phase 1 of the study (n = 1072), 41% were currently working, 50% had worked previously, and 9% had no previous work experience (p. 260). In Phase 2 of the study, adaptation to the workplace was studied. Amongst those who retired (n = 499), 22% had some work adaptation, 30% needed some work adaptation, and 32% met with no acceptance of the disorder in their workplace. For those in Phase 2 who were presently working (n = 410), work adaption or accommodation was provided for 42%, but another 27% still expressed the need for workplace adaptation. Adverse physical labor conditions affected 18% of those currently working and sadly some 30% reported no acceptance of their disorder in work (p. 265). The authors concluded that “…for NMD patients presently working, a labour career of 31 years duration (male patients) and 26 years duration (female patients), is quite probable” (Reference Andries, Wevers, Wintzen, Busch, Howeler, de Jager and Wokke1997, p. 270).

The nature of employment conditions was examined from the reports of 86 patients visiting a neuromuscular center (Minis et al., Reference Minis, Cup, Heerkens, Engels, van Engelen and Oostendorp2012). Less than half were employed (43%) and most of them (80%) were working in clerical-administrative work. It was reported that the most frequent adjustments involved “working conditions, terms of employment and task content” (Minis, Reference Minis, Cup, Heerkens, Engels, van Engelen and Oostendorp2012, p. 2279).

Other studies have reported employment rates of 44.5% amongst 119 persons with mixed neuromuscular conditions (Natterlund & Ahlstrom, Reference Natterlund and Ahlstrom1999); or 39% out of 46 patients from a 10-year follow-up from an initial group of 58 in which 10 had died (Bostrom & Ahlstrom, Reference Bostrom and Ahlstrom2005). In a small-scale study of physical activity, 8 out of 13 ambulatory people with neuromuscular disease worked compared with all 18 in the control group (Phillips, Fleming, & Tsintzas, Reference Phillips, Flemming and Tsintzas2009). A particularly low rate of working was that of Baumann, Newman, and Diserens (Reference Baumann, Newman and Diserens2013). They evaluated the medical records of 267 patients aged 16–25 years, 43 of whom had neuromuscular diseases. In the final analysis, only 1 out of the 43 persons with neuromuscular diseases was working. These results must be qualified as it excluded those studying.

Whilst the reports on working across neuromuscular conditions in general indicate that there is limited vocational potential, it is reasonable to ask whether the grouping together of neuromuscular conditions may mask substantial differences between the specific disorders (this issue is taken up in the following section).

Specific neuromuscular disorders and their relationship to work

Specific neuromuscular disorders and their relationship to employment have been investigated. In one of the earliest studies of social identity, Ville, Ravaud, Marchal, Paicheler, and Fardeau (Reference Ville, Ravaud, Marchal, Paicheler and Fardeau1992) surveyed 100 patients with facioscapulohumeral muscular dystrophy, of whom 68 agreed to participate in a study and 83.8% of the 68 were currently working. In a study on job perspectives of 126 adult patients with facioscapulohumeral muscular dystrophy, Wevers, Brouwer, Padberg, and Nijboer (Reference Wevers, Brouwer, Padberg and Nijboer1993) reported that 44% were working, 47% were no longer employed, and 9% had no work experience at all. The average female and male career was 15–30 years, respectively. It was noted that those who entered the workforce after the onset of their illness tended to choose office work. Just over half (51.5%) with an employment history had retired and the mean age was 50 years. The study emphasized that only 9 out of 61 jobs were adjusted to the disabilities of the patient. Bakker, Schipper, Geurts, and Abma (Reference Bakker, Schipper, Geurts and Abma2017) followed up the qualitative experiences of 25 people with facioscapulohumeral muscular dystrophy and reported only 2 out of 9 were working full-time and 10 had taken early retirement, but equally importantly, they reported that 32% were wheelchair dependent.

Fowler et al. (Reference Fowler, Abresch, Koch, Brewer, Bowden and Wanlass1997) noted significant differences when they examined the employment rates across four conditions of 154 individuals with neuromuscular conditions. Those with limb-girdle syndrome and myotonic muscular dystrophy had the lowest rates of employment. For limb-girdle syndrome, 7 out of 27 were employed, 9 out of 16 for facioscapulohumeral dystrophy, 5 out of 9 for Becker’s muscular dystrophy, 10 out of 17 for spinal muscular atrophy, 18 out of 43 for hereditary moor sensory neuropathy, and 13 out of 42 for myotonic muscular dystrophy. They reported a link between age and unemployment but no relationship between sex and employment status.

A relevant distinction that takes into account those with ambulatory disabilities was made by van der Linden et al. (Reference van der Linden, Kalkman, Hendricks, Schillings, Zwarts, Bleijenberg and van Engelen2007) in a study of patients with hereditary motor and sensory neuropathy type 1. The return to work rate for those with an ambulatory disability (n = 54) was 51.9% compared with 71% for those with no ambulatory disability (n = 21). At the same time, the age of the two cohorts was also significantly different (44.6 and 36.5 years, respectively).

Jeppesen, Madsen, Marquardt, and Rahbek (Reference Jeppesen, Madsen, Marquardt and Rahbek2010) conducted a mixed methods design of the conditions of living for 29 adults with spinal muscular atrophy type 2, and in the process reported that 14 (48%) were employed, 4 had been employed (14%), and 11 (38%) had never been employed. They stated that “The majority were financially supported by the public early retirement pension, while half of the participants supplemented their income with a part time job” (p. 12).

In a secondary analysis of data, Minis et al. (Reference Minis, Kalkman, Akkermans, Engels, Huijbregts, Bleijenberg and van Engelen2010) contacted 591 patients with neuromuscular conditions (facioscapulohumeral muscular dystrophy, n = 138; hereditary motor and sensory neuropathy, n = 135; and myotonic dystrophy n = 318). The largest proportion of those employed were in the facioscapulohumeral muscular dystrophy group (70.3%), followed by hereditary motor and sensory neuropathy group (63.7%) and 47.8% for myotonic dystrophy. Voluntary work also followed the same monotonic pattern. Patients’ reports of the impact of symptoms routinely mentioned fatigue (Hearwole et al., Reference Hearwole, Johnson, Bode, Dekdebrun, Dilek, Hilbert and Moxley2015), and in the case of myotonic dystrophy type 2 meant that only 33 (44.6%) out of a sample of 74 from a national registry were employed. This outcome is affected first by the participation rate in that only 74 out of 120 responded and by the advanced age vis-à-vis employment (mean = 57.0, SD = 12 years).

Amongst those with spinal muscular atrophy (Kruitwagen-Van Reenen et al., Reference Kruitwagen-Van Reenen, Van der Pol, Schroder, Wadman, Van Den Berg, Visser-Meily and Post2018), only 24 out of 62 had paid employment and only 10 had full-time work (36 h per week). From 48 patients with hereditary motor and sensory neuropathy 1a, Videler, Beelen, van Schaik, de Visser, and Nollet (Reference Videler, Beelen, van Schaik, de Visser and Nollet2009) reported 32 were working.

In the case of amyotrophic lateral sclerosis, Sandstedt et al. (Reference Sandstedt, Littorin, Johansson, Gottberg, Ytterberg and Kierkegaard2018) undertook 6-month follow-ups of a group of 60 patients over 3 years. They indicated that 8 out of 60 were working at Time 1, but by Time 5 none were employed. Equally poor outcomes were reported by Rahbek et al. (Reference Rahbek, Werge, Madsen, Marquardt, Steffensen and Jeppesen2005). They interviewed 65 adult males aged 18–42 years with Duchenne Muscular Dystrophy. They described graphically the functional ability of a typical adult patient:

There were problems in the area of education with only 23.1% completing high school. Only eight persons had work experience and none were employed at the time of interview.

Quantitative summary of work rate from previous studies

There were 19 relevant studies of neuromuscular conditions comprising a total of 2926 participants of whom 1266 or 43.26% were working. These outcomes came from seven nations (Canada 1, Denmark 2, France 1, Netherlands 8, Sweden 4, UK 1, and USA 2). As most nations had reported only one or two studies, the most meaningful comparison might be between the overall 47.0% work rate for Netherlands (n = 8 studies) and the overall 37.5% for Sweden (n = 4 studies). The difference in proportions is statistically significant (z = 2.64, P < 0.005), but there are no a priori theoretical justifications for this difference in proportions.

Methodological factors played no role in the outcomes. The correlation between the year of publication and the work rate was −0.034, meaning that neither the earlier nor the later studies had a better employment rate. The size of the study and the work rate were also unrelated (r = −0.019) as was the response rate for a study (r = −0.028).

Demographic factors did not play a significant role in the outcomes. The number of males is important as males have a higher rate of labor force participation (70.6% participation rate) compared with females (60.2%; Australian Bureau of Statistics, 2018). Overall, there was a negative correlation (−0.303) between the proportion of males and the work rate (i.e., studies with lower proportions of males had higher work rates). The mean age of the sample correlated −0.262 with work rate (i.e., studies with younger participants had a higher work rate).

Seven studies reported unambiguously that 36.1% were working, 18.1% had never worked, and 45.6% had worked previously at some time. This breakdown into three categories gives a more specific description of labor market participation than the overall work rate. The work rate of 36.1% for these seven studies differs markedly from that reported for the other 12 outcomes (63.9%), which did not evaluate those who had never worked and those who had worked previously. It suggests that the proportions who are working might well be overestimated when one does not take into account those who had never worked or who had worked at some time and dropped out of the labor market.

The breakdown according to the specific conditions is relevant and provided in Table 2. The conditions with more than reasonable likelihood of working were facioscapulohumeral muscular dystrophy and hereditary motor and sensory neuropathy, whereas amyotrophic lateral sclerosis and limb-girdle syndrome recorded very low rates; but these were all based on single studies. The mixed group of neuromuscular disorders based on six results had an overall work rate of 39.3%, which is around the median of the specific disorders.

Table 2.

Work rates for specific neuromuscular conditions

The relationship between particular validity criteria and work rate was also examined, and the results are summarized in Table 3. There was no a priori hypothesis as to how the application of a validity criterion would impact on the resulting work rate, but it seemed reasonable to consider that studies with fewer methodological controls might report higher work rates.

Table 3.

Specific validity criteria and the work rate

na, Kruskal-Wallis not calculated for sample sizes less than 5; ns, not significant two-tail test.

Eight criteria were examined. In every instance, the application of a methodological control (e.g., follow-up, the participation rate greater than 80%) or the reporting of a feature (e.g., the proportion of males in a study, the number who were ambulatory) resulted in a lower work rate. In simple terms, the better controlled the study the lower the work rate that was reported. No study satisfied all eight criteria, but studies that met 4–6 criteria had a median work rate of 52.8% compared with a median of 43.0% for those that satisfied 0–3 criteria (Mann–Whitney test U = 38, P = 0.652, z = −0.454, ns).

Specifically, seven findings from the studies did not distinguish between neuromuscular conditions and reported a median work rate (40.2%) that was below the 56.1% for those 12 studies that examined a particular condition (e.g., ALS, FHSN, DM); but this difference was not statistically significant (Mann–Whitney test U = 25, P = 0.164, z = 1.39). Studies with participation or response rate of less than 80% had a much lower work rate 41.2% (n = 15) compared with 64.4% for those four studies that had a participation rate of more than 80%. The six studies that did not report the mean age of the sample had a lower median work rate (40.7%) compared with a median of 55.6% for those 13 studies that indicated the mean age. Only a small number of studies took into account those that had died, those that were ambulatory or undertook a follow-up, and when they did lower work rates were reported.