Home health aides’ experiences of their occupational health: a qualitative meta-synthesis

ABSTRACT Home health aides (HHA) have high sickness absence while the need for home care services is rapidly growing. The aim of this study was to derive new conceptual understandings by identifying, describing and interpreting key concepts across qualitative studies on how HHA experience their occupational health related to their working conditions. A qualitative ethnographic meta-synthesis was used as a method to analyze 27 articles included from systematic searches in CINAHL, MEDLINE and PsycINFO. HHA experience physical strenuous work task demands in combination with unfortunate organizational conditions in an uncontrolled and ever-changing psychosocial and physical working environment as the main obstacle to their occupational health, although many positive presence factors with opposite effects were reported. More research is needed to investigate whether physical demanding work tasks can have positive effects on HHA’s occupational health by reorganizing their work while preserving patients’ empowerment at their home.

Today, home care services are regarded as a potentially cost-effective way to maintain people's independence, keep people physically, mentally and socially active as long as possible and enhance quality of life (Genet et al., 2011;Spasova et al., 2018). Addressing the influence of working conditions on occupational health among HHA is imperative in order to ensure well-functioning, high-quality home care services able to cope with future health challenges in society (Bloom et al., 2015;Spasova et al., 2018). Whilst some qualitative studies exist on the topic, they risk remaining "islands of knowledge" (Glaser & Straus, 1971) unless some form of synthesis is undertaken. Therefore, the objective of this study was to derive new conceptual understandings by identifying, describing and interpreting key concepts across qualitative studies to gain a deeper understanding of HHA's occupational health reflections and experiences related to their working conditions.

Methods
We used the meta-ethnographic approach, developed by Noblit and Hare (Noblit & Hare, 1988), to synthesize data from the included studies as it transcends differences in qualitative methodologies and has been used in other meta synthesis of qualitative health literature (Britten et al., 2002;Frostad Liaset & Lorås, 2016;MacEachen, Clarke, Franche, & Irvin, 2006). The objective was to develop new interpretations and conceptual insights that go beyond the findings of the individual studies synthesized, by focusing on the "translation of qualitative studies into one another" (Noblit & Hare, 1988). The studies deemed to be relevant were treated in a similar way to primary data. As Noblit and Hare (Noblit & Hare, 1988) argued, "Each account to be synthesized is already an interpretation of interpretations. The translation of accounts raises this to another level: interpretations of interpretations of interpretations" (p.35). The meta-ethnographic approach was conducted in seven phases where all authors contributed (Noblit & Hare, 1988). In line with our reporting so far, in the first phase we decided the focus of the synthesis (Noblit & Hare, 1988).

Search strategy
In the second phase, we conducted a systematic literature search to identify and select relevant study accounts to synthesize (Noblit & Hare, 1988). The first author searched CINAHL, MEDLINE (OVID platform) and PsycINFO (OVID platform) to identify relevant research articles following these criteria: Inclusion criteria: 1. Qualitative methods; mixed-methods studies could be included if the qualitative part could be seen separated from the quantitative part of the study; 2. Written and published in English; 3. Peer-reviewed; 4. Containing HHA's perspectives and personal experiences of occupational health related to working conditions within a western context; 5. Published between 2009 and 2019.

Review articles
To increase the validity of the study by ensuring the use of relevant databases and search words, the first author conducted literature searches with two librarians at separate times. We used search techniques inspired by Booth (Booth, Sutton, & Papaioannou, 2016) to identify relevant articles based on the theme. To identify relevant articles, we conducted searches via subjectrelated wordlists in the databases, including synonyms. We supplemented the search with keywords we manually selected from article keywords or abstracts. After wide-ranging pilot searches, we used the keywords and combinations listed in Table 1 to cover the theme and search the databases.

Search procedure
Three hundred and forty-five relevant articles were immediately identified. The first author read the abstracts of all 345 articles and excluded 315 articles based on the inclusion and exclusion criteria. The remaining 30 articles were read in full by the first author. Additional five articles were excluded this way and a total of 25 articles was included in this step. Most of the articles were included from CINAHL (N = 17) because we searched this database first. Furthermore, seven articles are included from MEDLINE and one from PsycINFO. Further searchers were conducted by reviewing the reference lists of the included articles. An equal procedure as described above was conducted and two additional articles were included during this process. Thus, we included 27 articles in this meta-synthesis (Table 2). A flow chart of the search procedure and reasons for exclusion is shown in Figure 1.

Evaluation of methodological quality
Extensive attention was given to the details in selected study accounts in the third phase of the meta-ethnographic approach. Repeated reading, identification and noting of interpretative metaphors were conducted to gain an overview on relevant topics in each study (Frostad Liaset & Lorås, 2016;MacEachen et al., 2006). Critical Appraisal Skills Program (CASP) (Critical Appraisal Skills Programme) was utilized for appraising the qualities of the included study accounts.

Meta-synthesis
In the phase 4, the relationship between study accounts was addressed. This was done by examining the focus studies had, their theoretical approach and the meaning of their concepts, themes or metaphors. We concluded that studies were "dissimilar but related" (Noblit & Hare, 1988) (p.64) and "a line of arguments" (p.63) was built up in order to discover a new "whole" -how and why the concepts are relevant (Frostad Liaset & Lorås, 2016;MacEachen et al., 2006).
In the next step (phase 5), we translated the study accounts into one another (Noblit & Hare, 1988). Reciprocal translation was deemed relevant as we found that study accounts can be "added" together (Noblit & Hare, 1988). Connections and differences between studies were explored by reciprocal translation, seeking to understand and transmit topics or concepts, ideas and metaphors across studies. Table 4 shows the outcome of the translation process. Potentially relevant key second-order concepts were identified and checked against the individual studies to maintain a close connection to the original studies. For a second-order key concept to be included, at least four studies had to be relevant to the concept (Frostad Liaset & Lorås, 2016;MacEachen et al., 2006). Second-order key concepts were identified and a description of each was provided (Table 4, third column). In the phase 6, those concepts were compared to identify overreaching themes and develop new interpretations from these. Four themes (third-order concepts) were defined. Thus, the findings from this study are a re-interpretation of "key concepts" according to how they relate to each other, on the main theme of the analysis and the purpose of the individual studies (Frostad Liaset & Lorås, 2016;MacEachen et al., 2006). In the last phase, we designed a suitable format for the dissemination of our findings.

Results
The main findings of the 27 included studies are summarized in Table 2.

Methodological quality
The assessment of individual studies based on the Critical Appraisal Programme (CASP) showed good methodological quality in studies included in the present synthesis (Table 3). Nonetheless, the relationship between researchers and participants has not been explicitly considered in several studies. However, studies are approved by ethical authorities and peer-reviewed, ergo we assume ethical considerations to be substantially evaluated.

Participants
The total number of participants from the included studies is 1,380, including 602 females and 37 males. The remaining 639 participants were not specified by gender (Andersen & Westgaard, 2013;Butler, 2018;Butler et al., 2010Butler et al., , 2012Franzosa et al., 2018;Hoppe et al., 2015;Mabry et al., 2018;Muramatsu et al., 2018Muramatsu et al., , 2019Samia et al., 2012;Schoenfisch et al., 2017), but the majority of participants were reported as women, with a percentage of 80-100 in all the included studies. The age of participants ranged from 18 to 73 years, and the average age was between 42 and 50.

Key concepts
HHA elaborated on their occupational health by addressing the following key concepts: physical strenuous work task demands, physical working environment, organizational conditions and psychosocial working environment (Table 4).
Informants described lack of required equipment at patients' homes (Franzosa et al., 2018). Equipment such as wheelchairs, walkers, canes and mechanical lifts could have been broken, missing parts or inappropriately sized for the clients, not fit the space or difficult to operate (Love et al., 2017;Polivka et al., 2015;Sims-Gould et al., 2013;Smith et al., 2019;Tourangeau et al., 2014). When assistive devices worked as required, several studies reported benefits such as more ergonomic and injury-reducing postures at work (Andersen & Westgaard, 2013;Mabry et al., 2018;Markkanen et al., 2014).
Stress created mental strain and lead to multitasking, lack of focus, rushing and moving too fast and increased fear of injuries or falling (Love et al., 2017;Muramatsu et al., 2018). Skipping lunch and cutting All included studies

7.Have ethical issues been taken into consideration?
All included studies 8. Was the data analysis sufficiently rigorous?
All included studies 9. Is there a clear statement of findings?

Verbal violence
Psychological distress due to unpleasant behavior's and rude comments from some clients. (Markkanen et al., 2017, Muramatsu et al., 2018, Schoenfisch et al., 2017, Tourangeau et al., 2014, Hittle et al., 2016, Andersen & Westgaard, 2013,Butler et al., 2010,Butler, 2018) ORGANIZATIONAL CONDITIONS Low pay Experiences of lack of/ inconsistent compensation and/or not being reimbursed for mileage. Exhaustion and strain because they had to work more. (Markkanen et al., 2017, Sims-Gould et al., 2013, Butler, 2018, Samia et al., 2012 Working alone Emotional strain due to little/no contact with colleagues. Individual responsibility for appraisal and initiative for facilitation of working environment.

Lack of training and unclear expectations
Lack of control over own practice and environment, as well as participation in decisionmaking (e.g. misunderstandings, miscommunication, lack of prior information, decisions made by administration). (Hoppe et al., 2015, Muramatsu et al., 2019, Andersen & Westgaard, 2015, Butler et al., 2012 Time pressure Work overload due to poor economy, increased demands of efficiency and paperwork created strain and burnout (e.g. restricted hiring, heavy patient caseloads, performing extra tasks) (Critical Appraisal Skills Programme, Franzosa et al., 2018, Hoppe et al., 2015Muramatsu et al., 2018, Schoenfisch et al., 2017, Muramatsu et al., 2019, Muramatsu et al., 2018, Smith et al., 2019, Mabry et al., 2018, Karlsson et al., 2009,Andersen & Westgaard, 2015, Wills et al., 2016, Samia et al., 2012 (Continued) down on the time spent on assignments in order to get through the work list on time were some of the negative trends toward increasingly hectic work situations (Andersen & Westgaard, 2015;Karlsson et al., 2009). Informants described increased indirect work tasks -for instance, paperwork -as endeavors that did not directly involve patient care but increased the time pressure and workload, were difficult to complete during paid work hours and were a source of overtime (Agbonifo et al., 2017;Andersen & Westgaard, 2015;Samia et al., 2012;Tourangeau et al., 2014).

Psychosocial working environment
In several studies, HHA reported a lack of respect for their professional expertise, lack of appreciation, lack of positive feedback and insufficient support from agency supervisors (Butler, 2018;Butler et al., 2010Butler et al., , 2012Franzosa et al., 2018;Muramatsu et al., 2019;Tourangeau et al., 2014). The informants described the lack of support as stressors (Muramatsu et al., 2019), negative experiences, such as no staff meetings and supervisors who were never available (Tourangeau et al., 2014), and feelings of being invisible (Karlsson et al., 2009).
Some HHA had also experienced verbal abuse from clients with mental or cognitive disorders (Butler, 2018;Butler et al., 2010Butler et al., , 2012Hittle et al., 2016;Mabry et al., 2018;Muramatsu et al., 2019;Schoenfisch et al., 2017;Sims-Gould et al., 2013;Smith et al., 2019). In contrast, many informants described that they developed meaningful, trusting relationships and close bonds with clients and their families. These bonds were a major reason to continue the job Franzosa et al., 2018;Karlsson et al., 2009;Mabry et al., 2018;Markkanen et al., 2014;Tourangeau et al., 2014). The informants emphasized the importance of seeing that they made a difference in people's lives (Butler et al., 2012;Nielsen & Jørgensen, 2016) and client's and/or the client's family's appreciation had positive mental health effects (Butler et al., 2012;Karlsson et al., 2009;Markkanen et al., 2014;Muramatsu et al., 2019; Franzosa et al., 2018, Muramatsu et al., 2018, Andersen & Westgaard, 2015 Nielsen & Jørgensen, 2016;Smith et al., 2019;Zoeckler, 2018). In several studies, the HHA indicated that their close relationships with clients contributed to emotional stressors, emotional drain and grief because they became very attached to clients who were ill, sad or died (Butler, 2018;Butler et al., 2010Butler et al., , 2012Mabry et al., 2018;Muramatsu et al., 2019;Smith et al., 2019;Zoeckler, 2018). Several studies described that HHA felt isolated and lonely at work because they had little to no contact with other members of the home care services (Franzosa et al., 2018;Karlsson et al., 2009;Mabry et al., 2018;Samia et al., 2012;Schoenfisch et al., 2017;Tourangeau et al., 2014;Zoeckler, 2018). This factor contributed to isolation from conventional personal support and the development of emotional stress (Mabry et al., 2018;Schoenfisch et al., 2017). Despite this, the informants emphasized the importance of home care as a profession. They viewed the value of their work as related to other care options and recognized themselves as professionals (Schoenfisch et al., 2017). Indeed, several studies described that HHA were often happy, found their job satisfying, enjoyed most aspects of the work and looked forward to going to work (Andersen & Westgaard, 2013;Butler et al., 2012;Nielsen & Jørgensen, 2016).
The variety of patients necessitated a diversity of skills, a factor that the HHA viewed as motivating to remain in home care (Tourangeau et al., 2014). Some felt a greater sense of autonomy and control at work and increased satisfaction when they gained responsibility for transferred tasks . Despite tight schedules, HHA expressed that they took the needed time to listen to the patient (Nielsen & Jørgensen, 2016). Informants reported job flexibility as a positive and rewarding factor to choose a home care job because it allowed them to meet the client needs as well as their own personal requirements or family responsibilities (Markkanen et al., 2014;Nielsen & Jørgensen, 2016;Tourangeau et al., 2014).

Discussion
Despite reports about challenges with occupational health and high sick leave, several studies in this synthesis reported that HHA often find themselves happy and find their job satisfying. Indeed, they seem to enjoy many aspects of the work and look forward to going to work (Andersen & Westgaard, 2013;Butler et al., 2012;Nielsen & Jørgensen, 2016). Conditions at work that make the employee want to go to work are positive presence factors which can contribute to their occupational health. As such, they should be mapped and emphasized. The findings from this study suggest following positive presence factors for HAA; the variety of patients necessitates a diversity of skills (Tourangeau et al., 2014), self-recognition as a professional (Schoenfisch et al., 2017), sense of autonomy and control at work , job flexibility that allowed them to meet client needs as well as their own personal requirements or family responsibilities (Markkanen et al., 2014;Nielsen & Jørgensen, 2016;Tourangeau et al., 2014), meaningful, trusting relationships and close bonds with clients and their families Franzosa et al., 2018;Karlsson et al., 2009;Mabry et al., 2018;Markkanen et al., 2014;Tourangeau et al., 2014). However, trusting relationships could also be a source of psychosocial stress (Muramatsu et al., 2018(Muramatsu et al., , 2019, and the informants described it as challenging to maintain boundaries related to assigned hours and job duties as their relationship with the client or the client's family became closer Mabry et al., 2018). Occasionally, informants experienced even verbal abuse (Butler, 2018;Butler et al., 2010Butler et al., , 2012Hittle et al., 2016;Mabry et al., 2018;Muramatsu et al., 2019;Schoenfisch et al., 2017;Sims-Gould et al., 2013;Smith et al., 2019). Research has shown that a poor psychosocial working environment by itself does not permanently reduce workability, but it may lead to poorer mental and physical health (Hanson, Perrin, Moss, Laharnar, & Glass, 2015) and result in a disability pension (Stapelfeldt et al., 2014).
Several studies in this synthesis found that HHA described their occupational health as impaired due to physical strenuous work tasks demands (Agbonifo et al., 2017;Andersen & Westgaard, 2013;Butler, 2018;Love et al., 2017;Mabry et al., 2018;Markkanen et al., 2017Markkanen et al., , 2014Muramatsu et al., 2018Muramatsu et al., , 2019Polivka et al., 2015;Schoenfisch et al., 2017;Wills et al., 2016;Zoeckler, 2018). Previous studies have also shown that assisting clients with physical tasks ranging from lifting to bathing to household chores are physically demanding (Delp, Wallace, Geiger-Brown, & Muntaner, 2010). These findings are consistent with the results from other studies that showed musculoskeletal pain and injuries in the neck/shoulder and lower back are commonly expressed and a major health problem among HHA (Denton et al., 2002;Genet et al., 2011;Spasova et al., 2018). It has been common to try to reduce the physical strain at work in recent decades, but positive results have so far been modest (Driessen, Proper, Anema, Bongers, & Van Der Beek, 2010). However, it is a paradox that physical activity at work can make you sick, while physical activity in your free time provides health benefits (Holtermann et al., 2018;Holtermann, Hansen, Burr, Søgaard, & Sjøgaard, 2012). HHAs' strenuous high mechanical strain with awkward postures in accordance with findings in the present study can contribute as a part of the explanation. Research also suggests that repetitive movements with lower intensity and little rest might add to the burden (Holtermann et al., 2018). A solution can be to reorganize strenuous physical work tasks, so they instead get more of the effect we are left with when we train in our free time (Holtermann, Mathiassen, & Straker, 2019). However, such a solution remain to be designed, tried and proven in practice.
Reorganizing physical work tasks to be more health promoting can be especially challenging due to HHA provide care in an uncontrolled and everchanging environment; patients` home. It can create an unsafe and unhealthy working space (Agbonifo et al., 2017;Butler, 2018;Hittle et al., 2016;Markkanen et al., 2017Markkanen et al., , 2014Muramatsu et al., 2018Muramatsu et al., , 2019Polivka et al., 2015;Schoenfisch et al., 2017;Sims-Gould et al., 2013;Tourangeau et al., 2014) and expose them to a myriad of safety hazards (Gershon et al., 2012;Howard & Adams, 2010;Lavender et al., 2019). Changes required to make physical work tasks more health promoting can deteriorate patients` feeling of being home and empowerment. Moreover, HHA work much alone, often with a lack of sufficient support (Karlsson et al., 2009;Muramatsu et al., 2019;Tourangeau et al., 2014). They reported time pressure during hectic workdays (Andersen & Westgaard, 2015Franzosa et al., 2018;Hoppe et al., 2015;Karlsson et al., 2009;Nielsen & Jørgensen, 2016;Smith et al., 2019), lack of training, unclear expectations, inconsistent scheduling (Butler et al., 2010;Denton et al., 2015) and increased indirect work task are sources of negative emotions, such as frustration, stress, workload and overtime hours (Agbonifo et al., 2017;Andersen & Westgaard, 2015;Samia et al., 2012;Tourangeau et al., 2014). Stress at work created mental strain and an increased risk or fear of injuries (Love et al., 2017;Muramatsu et al., 2018). Excessive sick leave was reported as a problem in home care (Andersen & Westgaard, 2013) and created strain and pressure also for those who were at work (Andersen & Westgaard, 2013) by requiring them to take on extra responsibilities, perform extra tasks beyond the work schedule and work unpaid hours to fill care gaps and to meet the client's needs (Franzosa et al., 2018;Hoppe et al., 2015;Schoenfisch et al., 2017). Consequently, HHA's physically strenuous work task demands are being even more strenuous due to physical, psychosocial and organizational working environment. Therefore, all these factors should be addressed in order to promote their occupational health.

Methodological considerations
The gender distribution in this synthesis is skewed as most of the participants in the included studies were women. However, current research shows a similar pattern across Europe, namely that home care is mainly provided by women (Spasova et al., 2018). Thus, we argue that the informants are representative for HHA. Furthermore, one should be aware that the number of participants reported in this study can be overestimated as some of the included studies are mixed methods studies where the number of participants is reported as one joined number. Furthermore, the nature of the primary study data (first-order concepts) available to synthesis can have implications for the results. It is a well-established understanding that rich descriptive data are best suited to develop conceptual insights in a meta-ethnography approach (France et al., 2019). Data from telephone interviews compared to those from focus group interviews are rather "thin" descriptive data which is difficult to interpret and have limited explanatory ability. Consequently, their contribution to in-depth insights would be limited. However, we decided to include studies with such data due to their contribution to a comprehensive grasp of the topic we wanted to shed light on.
Finally, although only studies from a western context are included in this meta-synthesis, health-care systems within this context can differ a great deal. Therefore, it is important to consider that not all results are valid across health-care systems.

Conclusions
The findings in this synthesis suggest that HHA experience many positive presence factors at their work which contribute to their job satisfaction and occupational health. Those factors should be recognized and strengthened. Physical strenuous work tasks have a negative influence on their occupational health. The impact is worsened by challenges in an uncontrolled and everchanging physical and psychosocial working environment. Organizational conditions add to this burden, especially colleagues' sick leave, time pressure and inconsistent scheduling. There is a need for reorganizing the way HHA work so that physical demanding work tasks could have a larger positive effect on HHA's occupational health while preserving patients' empowerment at their home. In order to achieve this aim and ensure high-quality home care services that are able to cope with future health challenges, more investigation is needed.

Consent for publication
All authors have consented to publication