Analysis of Medico-Social Factors for Job Retention Among Patients Presenting With Haematological Malignancy (Adamentine): Results of a ‘Pilot Study’

Objective The aim of this study was to describe return to work determinants in patients with haematological malignancy. Results This medico-social pilot study included patients with haematological malignancy in the département of Calvados, aged 18 to 55 years, diagnosed between 1st January and 31st December 2010 and alive at 1st January 2015. Patients were identified via consultation of the Lower Normandy haematological malignancy Registry. They completed a specially developed self-questionnaire, in addition to validated questionnaires for anxiety-depression, quality of life and fatigue. Of the patients contacted, 50% accepted to participate. The mean age at diagnosis was 49.8 years, and the majority of patients (79.2%) was professionally active at the time of diagnosis. Only 64.9% of subjects had stopped work due to illness. The psychological impact (demonstrated anxiety) was significantly greater in men (p = 0.01). The majority of subjects returned to work after treatment (80.7%) and among them, the mean duration of absence from work was 16.1 months. Only 52.6% of subjects had informed their occupational physician and 56.7% had benefited from a pre-return visit.


Introduction
Cancer is a public health issue affecting more than 380,000 subjects every year in France [1][2]. In 2018, 45,000 new cases of haematological malignancy were diagnosed in mainland France, of which 25,000 (55%) were males [1][2]. Heavy treatment and prolonged work stoppages are often required.
Even if the global incidence of haematological malignancy is increasing, average patient survival is improving thanks to scientific progress, both in terms of screening and overall cancer care [2].
Factors that may impact the return to work, in particular the duration of sick leave and its social and occupational determinants, as well as the possible specific modalities of such return to work and the associated support offered to patients, are poorly documented [3][4]. The aim of the present study was to describe the modalities involved in returning to work and the determinants of maintaining or reintegrating work early after treatment in patients with haematological malignancy.

Methods
The study population included professionally active patients with haematological malignancy in Calvados, aged 18 to 55 years inclusive at the date of diagnosis of haematological malignancy, diagnosed between 1st January and 31st December 2010 and alive at 1st January 2015. The identification of patients with haematological malignancy was conducted via consultation of the Lower Normandy haematological malignancy Registry. Consultation of the RNIPP (National Directory for the Identification of Natural Persons) was conducted by investigators from the cancer registry to ascertain the vital status of included patients.
Patients completed a self-questionnaire, accompanied with an information notice. If the patient did not respond within 8 weeks, a reminder was mailed. An identification list was established, associating each patient with an identification number.
The study was approved by the hospital ethics committee. All participants received information on the study and provided written informed consent.
Medical data included: date of diagnosis of haematological malignancy; tumour stage at diagnosis (Ann Arbor classification); initial treatment: type and duration of treatment; possible previous history of cancer(s) and/or concomitant cancer; possible recurrence of a previously treated haematological malignancy; comorbidity factors at time of diagnosis and 3 years after diagnosis.
We developed a specific self-questionnaire, which was sent together with other validated questionnaires on anxiety-depression (validated French version of the HADS, Hospital Anxiety and The specifically developed questionnaire included data on: patient sociodemographic characteristics (sex, age); socioprofessional data at the time of diagnosis and at the time of study, together with any potential change in this situation (occupied position, type of work contract, socioprofessional category, working conditions -in particular strenuousness and psychosocial risks, specific arrangements); potential socioprofessional difficulties associated with cancer (sick leave, financial support, disability recognition, etc.).
In order to study the factors associated with return to work as well as those related to employment preservation or occupational reintegration of patients, we compared 2 groups of subjects from our population. Group 1 consisted of subjects who had not ceased their professional activity and those who had resumed their professional activity after diagnosis; group 2 consisted of subjects who had not resumed their professional activity.
We calculated a 'Job Evaluation Score Prior to Diagnosis', which is the sum of the scores for each Job Evaluation Item prior to diagnosis. For each item: 0 points if 'not at all' or 'a little' in items 1, 2, 6, 8; or if 'moderately' or 'completely' in items 3,4,5,7. 1 mark if 'not at all' or 'a little' in items 3, 4, 5, 7; or if 'moderately' or 'completely' in items 1,2,6,8.
Descriptive statistics were produced using Statistical Analysis System (SAS) software, version 9.3.
Exploratory data analysis provided, for qualitative variables, their frequency and their 95% confidence interval and, for quantitative variables, the mean, the standard deviation for the mean, the median and the quartiles. All missing data were systematically reported.
Statistical tests and confidence intervals have been calculated with a bilateral α= 5%.

Results
The general characteristics of the 72 patients who accepted to participate are provided in Table 1.
The study population consisted of 45.8% women, mean age was 49.8 (9.5) years and the majority of subjects were active at the time of diagnosis, mostly under permanent contracts.
Among the active subjects at diagnosis (N, number=57 (79.2%)), the majority of subjects (75.4%) were working full-time and on fixed day schedules. The average length of service was 15.7 years.
The majority of subjects (71.9%) rated their professional activity prior to diagnosis as 'completely satisfactory', with no significant difference between men and women, or between ages. Forty-two percent of the subjects declared that their professional activity involved a high psychological burden and only 26.3% of subjects described a high physical burden. In terms of psychological support from co-workers, 43.9% of subjects declared themselves as 'well supported'. 57.9% of patients considered they had sufficient freedom of movement in their work (responsibilities, satisfactory decision-making margins).
Only 64.9% of subjects reported a sick leave at diagnosis, no significant difference being observed in terms of age or sex. The average duration of sick leave for the entire study population was 19.9 months.
Almost half of patients (47.4%) had not informed their occupational physician and 56.7% had benefited from a pre-return visit, no difference being observed for sex. There was an absence of adapted workstation layout for 90% of patients. The majority of subjects had resumed their professional activity in the same company as the one prior to diagnosis (65.2%). The majority of subjects had taken up the same type of employment contract (86.7%), and in the same job (83.3%).
Ninety percent of patients benefited from no adjustment to their working conditions. Arrangements for working conditions included: 'ergonomic adaptation of your workstation', 'arrangement of your workstation schedules' and 'other' (table 2).
Of those who reported having a current occupation (table S1) 67.7% worked full-time. Return to work conditions were considered sufficiently anticipated for 59.2% of subjects, with no significant difference for age and sex.
The majority of subjects rated the current occupational activity as 'completely' satisfactory (62.1%), and 10.3% rated it as 'not at all' satisfactory. The analysis shows no significant difference for age and sex. The majority of subjects reported feeling more tired at work than before ('completely' for 39.3%, 'moderately' for 25%, 'a little' for 17.8%, and 'not at all' for 17.8%), with no significant difference in age and sex.
The majority of subjects reported feeling that they were working 'the same as before' when they discovered their disease (71.4%), and 25.0% reported feeling that they were working 'less than before 'when they discovered their disease, with no significant difference for age or sex. Of those returning to work, 17.8% reported feeling penalised in their work because of their illness. No significant age and sex differences were observed (table S1).
Psychological impact was significantly greater in men (demonstrated anxiety disorder 73.0 versus 41.4, p=0.001), but no differences were observed for age.
Regarding the depression score, 65.6% of subjects had 'suspected depressive disorders' and 28.1% had 'proven depressive disorders'.
The average score for the overall health and quality of life scale was 65.9/100. No significant differences in these items were found for age and sex. For the assessment of fatigue, the average score was 55.5/100. There was no significant difference for sex, age, overall or sub-scale (table S1).
There were no significant differences between Group 1 (subjects who had not ceased their professional activity) and group 2 (subjects who had not resumed their professional activity) for sex, age, length of service, type of employment contract, social protection scheme, working time and working hours (table 3). The mean duration of absence from work among those who returned to work was 16.1 months (3.1) versus 42.2 months (8.2) for the group 2 (p=0.002).
The'Job Evaluation Score Prior to Diagnosis' (which is the sum of the scores for each Job Evaluation Item prior to diagnosis), was not significantly different for the 2 groups studied (p=0.08). In contrast, for the assessment of fatigue, the average score was significantly higher among subjects who did not return to work (p=0.004) ( In our study, we included 6 cases of MGUS: this disease often had a small impact on general status and professional life. This may explain the higher rate of return to work observed in our study. Other studies have observed that the longer the sick leave, the more difficult the return to work among patients suffering from cancer, hence highlighting the importance of early medico-social intervention [3-4]. In our study, the mean duration of the absence from work among subjects who returned to work was 16.1 months vs 42.2 months for the group which had not returned to work (p=0.002). This encourages patients to quickly seek employment adaptations, and to consider well in advance the possible layout of their workstation, in order to anticipate recovery and avoid occupational exclusion.
The role of the occupational physician in the return to work for cancer patients has been previously demonstrated and a significantly higher level of return to work was observed in patients who benefited from arrangements in their working conditions [20-21]. Unfortunately, according to another study, only one in 4 patients benefited from a consultation with the occupational physician immediately prior to return to work and only 50% benefited from special work arrangements [21].
Similarly, we found that almost half of patients had not informed their occupational physician about their disease and that 56.7% had benefited from a pre-return consultation. As a result, 90.0% of subjects did not benefit from any adjustment to their working conditions.

LIMITATIONS
One of the limitations of our pilot study is its low statistical power, which has not allowed us to analyse other determinants of job retention such as occupational category, education level and income level. Some factors were close to significant (psychological burden and physical burden between subjects who had returned to work or not, in particular).
The originality of our study was to exhaustively recruit all subjects with haematologic malignancies from a well-known regional cancer registry. This ensures the completeness and reliability of diagnosis.
This study yielded promising results which led to its extension to include other medical centres. The    'not at all' or 'a little' in items 3, 4, 5, 7; or if 'moderately' or 'completely' in items 1,2,6,8.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. TableS1.docx