2.1 Research Strategy

We sought articles about epilepsy and employment on PubMed, Google Scholar, and Embase.

Our search string was “Epilepsy AND Employment OR Job OR Work”. We used the Rayyan review system, an online tool allowing for flawless collaborative systematic review [12].

Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines, we performed a systematic review on epilepsy and employment [13].

We selected 55 among 7272 articles from 1958 to 2023 (Figure 1).

We then grouped those articles between papers reporting employment status and papers only describing factors contributing to employment/unemployment. We also considered specific cases: surgical patients and people with specific syndromes (e.g., juvenile myoclonic epilepsy, childhood-onset epilepsy, absence epilepsy).

2.2 Inclusion and Exclusion Criteria

We have only included articles written in English and published in journals with good impact factors (IF) regarding employment conditions in different European countries, as a specific deliverable of the European Project EpilepsyPOWER (2021-1-IT02-KA220-ADU-000028349). We considered articles regarding people living in the UK, as it was part of the geographic area of the European Union (EU) and politically included in the EU until 2023. We excluded articles with wrong types of publication (e.g., letters to editors) or wrong populations (e.g., people with disabilities and not suffering from epilepsy).

2.3 Study Selection

We screened papers using the artificial intelligence automatic Rayyan review system [12]. Abstracts were selected by two investigators (GL and the Principal investigator, FN), based on titles and abstracts. Two researchers independently reviewed selected articles and results. GA supervised the quality of selected papers and adherence to the proposed aim, trying to minimize the risk of selection biases.

We selected all papers containing the world “employment” and/or “employed” in the title or in the abstract, dividing them into manuscripts reporting rates of employment, articles reporting specific situations (including surgical outcomes), and disclosure issues.

2.4 Employment Status

Selected manuscripts considered different sample dimensions, and they often did not report the specific rate of employment/unemployment.

Only a few articles (eight) specified rates of retired PwE; the others included them as unemployed people, likewise for housewives and people receiving disability pensions. No article specifically mentioned self-employment, and only three papers mentioned part-time jobs. Therefore, despite the limitations, this category was considered among employed PwE.

Therefore, due to the above limitations, the evaluation of the real unemployment rate was not completely clear. Thus, we decided to list articles reporting the same/different rates of employment/unemployment compared with the general population, also to find out a trend in recent years. We specified the type of study, rates of employment/unemployment reported, and comparison with the general population.

3.1 Employment Status and Unemployment Rate

Unemployment rates reported among PwE were different across Europe.

Most of the articles came from the UK (12), whereas other articles examined the situation of the Netherlands (6), Germany (5), Sweden (4), Estonia (3), Spain (3) Ireland (2), Denmark (2), Poland (2), and Finland (2).

Here, we list articles comparing rates of employment/unemployment with the general population, in chronological order (Tables 1 and 2).

Scambler in 1980, interviewing general practitioners, described the following employment situation in the UK: 31/42 (74%) of men of working age, 6/19 (32%) of married women, and 5/12 (42%) of unmarried women were employed, with respective rates in the general population of 81% (men), 48% (married women), and 42% (unmarried women) [14].

Several studies conducted by Ann Jacoby analyzed the employment situation in the UK in the 90s, both from the employers and employees' perspective [5, 15, 16]. Among those studies, in 1992 and 1995, she demonstrated a similar rate of employment in people with well-controlled epilepsy compared with the general population. In 1992, rates reported among PwE were 91% of men and 74% of women under 60 employed; of those unemployed, only 3% reported epilepsy as the main unemployment reason [15]. Similar results were described in 1995: among 494 PwE of working age, 71% were employed, 26% unemployed, and 3% of those unemployed reported seizures as the reason for unemployment [16]. Collings reported a rate of unemployment of 12% compared with the 9% national unemployment rate in the UK in 1994 [4].

In 1992, Callaghan described the employment situation in Ireland in a population of 343 PwE: 34% of males and 28% of females were unemployed; 53% of males and 24% of females were gainfully employed; 1% of males and 41% of females did house duties; 8% of males and 7% of females were involved in sheltered employment; and 4% of males and less than 1% of females were retired. They also highlighted higher underemployment among PwE (34%) compared with the general population (13%) in the same years [17].

On the contrary, Kourchonov evidenced a lower rate of employment in PwE in Germany, despite the promulgation of the Law on Support of Employment in 1996 [18].

In more recent years, Herodes and Rätsepp both demonstrated similar rates of unemployment compared with the general population in Estonia [19, 20].

Geerts reported similar rates of employment among people with childhood-onset epilepsy compared with the general Dutch population in a longitudinal study, regardless of the specific kind of epilepsy: idiopathic epilepsy 91.5%, symptomatic epilepsy 91.3%, cryptogenic epilepsy 97.1%, epilepsy in remission 93.2%, active epilepsy 91.2%, all compared with Dutch age peers 92% [21].

A cross-sectional multicentric epidemiological study in Spain also showed similar rates of employment and unemployment in PwE compared with the general population [22].

Recently in Poland, Majkowska demonstrated lower rates of employment in PwE (49%) compared with the Polish and European populations (59.3% and 64.6%, respectively) [23].

As for surgical patients, Eldevik compared the employment rate of seizure-free patients after surgery in a prospective longitudinal study. In the Swedish general population, 65%–71% of those between 25 and 54 worked full time, compared with 36% of seizure-free patients 5 years after surgery and 65% 10 years after surgery [24].

Considering surgical treatment outcomes, Reinholdson showed that seizure-free PwE with intelligence quotient IQ > 70 gain the same level of employment as the general population [25].

In the last decades (2000–2023), a slight prevalence of articles reporting similar rates of employment (five vs. three) has emerged.

3.2 Factors Contributing to Unemployment and Underemployment

Unemployment and underemployment in PwE derive from a combination of multiple and different factors. Internal factors include individual characteristics, clinical features, and self-concepts. Whereas, work atmosphere, knowledge, and employers' attitudes represent the main external determinants of employment status in epilepsy (Table 3) [10].

Among individual factors, younger age and early onset of disease seem to negatively affect the chance of obtaining a good job position [19, 26, 27]. Indeed, in a Finnish study, Sillanpaa evidenced that onset at age less than 6 is one factor promoting employment in PwE [28]. According to Geerts, childhood-onset epilepsy is strictly connected to lower educational success, and, so, to a lower job attainment [21]. On the contrary, Haag found that older people had a lower probability of being employed. Jacoby demonstrated comparable results in a large cohort study in the UK in 1996 [29].

When considering clinical factors, the most important determinants of employment are seizure freedom and remission. Multiple studies indicated seizure severity as the main explanation for unemployment in PwE: in 1992, Jacoby, focusing on a population of well-controlled PwE, stated that active epilepsy (seizures occurring in the last 2 years) impaired employment conditions [15]. Three years later, the same group established in a similar population that seizure severity, frequency, and controllability negatively affect employment in PwE [16]. Refractory epilepsy may hamper long-term social outcomes, including employment [22]. Conversely, seizure remission and having no history of status epilepticus are positive factors in people with childhood-onset epilepsy [28]. Later, other authors asserted that a higher seizure frequency was linked to unemployment [14, 16, 20, 25, 30-32]. When considering long-term outcomes, treatment with antiseizure medications prompts seizure control and, in most cases, seizure freedom: thus, it can be considered as another clinical factor connected to employment [33, 34]. Seizure type can also influence employment: tonic–clonic seizures are linked to a worse employment status [16, 34].

Another relevant element is cognition: cognitive deficit and intellectual constraints consistently impair job attainment [32, 35, 36]. Academic difficulties and lower educational levels also contribute to unemployment [37, 38]. Social difficulties [36] and psychiatric comorbidities [33, 38, 39], anxiety and depression [16, 37, 38, 40] might also impair employment. In general, comorbid disabilities lead to lower educational levels and chances of employment, as demonstrated in a population of surgical patients in Germany [41]. Polytherapy also affects employment status and quality of life [22, 42]. Clearly, a higher number of drugs can affect cognitive capacities and concentration, and so the ability to work normally.

Stigma, misconceptions, and misbeliefs surrounding epilepsy lead to discrimination from employers and self-denial of employees in the workplace [14, 30, 36]. Employers' attitudes and an unsupportive familial environment can hamper gaining a good job position [5, 32, 36, 43]. Limits posed by specific work issues, such as working with machinery or driving license restrictions, can affect employment status in PwE [43, 44].

To summarize, we can refer to Gloag's paper in which he highlighted three aspects impairing job position in PwE: the hazards posed by specific jobs in case of seizures; anxiety and prejudice on the side of employers and fellow workers; behavioral and mental abnormalities [3]. Geerlings has also drawn a specific predicting risk profile score made up of unsupportive family environment, lower intelligence, and higher seizure frequency and ongoing seizures [32].

3.3 Underemployment and Lower Salaries

Researchers have shown evidence of disadvantages for PwE in the workplace, even when a higher rate of unemployment is not reported. PwE earned lower salaries [9, 33] and were underemployed compared with the general population [45-47]. These issues may derive from lower educational levels [21], employers' attitudes, misconceptions, and self-denial.

3.4 Factors Favoring Employment

Factors opposed to those described previously facilitate employment. The most reported factor favoring employment is seizure remission or good seizure control [4, 23, 28, 30, 48].

People with normal intelligence are prone to get a better education and, consequently, better employment [25, 28, 49]. Focusing on specific functions, Partanen demonstrated that better working memory and executive functions were associated with improvement in employment status in surgical patients [50].

A higher educational level allows PwE to reach a more suitable job [20, 24, 31, 51, 52].

Experiences from Ireland, UK, and Belgium demonstrated that training and job counseling favored the employment of PwE more than other elements [30, 49, 53].

Having a favorable atmosphere at work, strong relationships, and a positive individual perception increase the chances of being employed [4, 51].

Although social aspects are less explored, we can assume that emotional stability (including having no psychiatric comorbidities) and proper social skills positively contribute to PwE's employment status [36].

As declared before, employment status is strictly connected to seizure control and seizure freedom [24, 26, 30, 37, 41, 49, 54]. Other clinical factors, such as being treated with a lower number of antiseizure medications, can contribute to a good employment situation [55]. When considering long-term outcomes, surgery has a modest but significant influence on employment status in surgical candidates [26, 56, 57]. Finally, having a driving license also contributes to independence and, hence, improves the employment situation in PwE [44].

Considering external factors, apart from employers' attitudes, the promulgation of laws favoring inclusion of PwE in the workplace represents one of the most relevant determinants supporting employment, as proven in Germany [18].

3.5 Long-Term Outcomes and Specific Conditions

When analyzing specific conditions:

3.6 Disclosure Issues

Though it was not our primary aim, we could not leave unmentioned papers that shed light on such a controversial issue as disclosure. Most PwE fear discrimination, stigmatization, and negative consequences (such as being fired, demoted or unemployed) when revealing their condition. They frequently conceal their disorder, especially if well controlled [4, 15, 39]. In different studies, about one-half or one-third of the population examined did not disclose their condition [14, 23, 51]. Lassouw evidenced that in a Dutch working population, 77% of respondents disclosed their condition, although they refused to answer questions about the impact of epilepsy on their job [9].

3.7 Employment Programs and Interventions for PwE in Europe

In Europe we found a report of only a few experiences with inclusion programs. In 1992, Carroll demonstrated the usefulness of a rehabilitation program in the Irish population: 58% of the considered population had found employment at the end of the program. Furthermore, about 66% of the surveyed population thought that programs are useful to develop self-confidence and social skills [53]. Similarly, in Sweden, Wedlund demonstrated improvement in work and education participation in 38 among 124 PwE involved in a rehabilitation program [58].

4.1 Strengths and Limitations

Our review seems to be comprehensive and extensive, describing the unemployment situation in Europe over the last 60 years, including longitudinal studies about specific conditions. The wide extension of years and conditions considered makes this review noteworthy, laying the background for future studies in the social field.

We have focused on European literature as this review is a deliverable of the EpilepsyPOWER European Project Erasmus + (project 2021-1-IT02-KA220-ADU-000028349).

Main limitations are variability among reported rates of employment/unemployment, different sample dimensions, different times of observation (longitudinal vs. transversal studies), and methodologies of reported studies. Another limitation is the heterogeneity of considered populations as employed (e.g., part-time) and unemployed (e.g., housewives).

Most of the reviewed papers used surveys, so they could be affected by compilation biases, and finally, not all the papers considered the comparison with the general population.