Abstract
BACKGROUND:
General practitioners (GPs) have expressed difficulties in issuing sickness certificates and problems may arise if this work is not performed in an adequate manner. There is scant knowledge about how collaboration with other professions could be organized to enhance this work.
OBJECTIVE:
Evaluate the feasibility of occupational therapists (OTs) performing supplementary assessments for persons on sick leave.
METHODS:
Four healthcare centres (HCs) tested a working approach intervention where sick-listed patients were offered a complementary occupational therapy assessment. The OT assessments were intended to provide useful information for GP issued extended sickness certificates. Data on sick leave, sickness certificates and patient questionnaires were collected at different HCs. Interviews were conducted with GPs and OTs and the Consolidated Framework for Implementation Research was used to analyse the intervention’s implementation.
RESULTS:
No major differences in the sickness certificate quality was found. Available data on sick leave increased for all HCs during the project. Not all GPs used the OT assessments, which indicates that the implementation of the intervention was insufficient.
CONCLUSION:
Testing a new working approach in primary healthcare requires an implementation strategy. To improve sickness certification quality, this work needs to be prioritized as an important healthcare task.
Keywords
Introduction
In Sweden and many other countries, sickness certification is issued by physicians. In primary healthcare, sickness certification is a common and frequent task for general practitioners (GPs) [1, 2]. Previous research has shown that many physicians find that dealing with sick leave issues is difficult, uncomfortable, and can create conflict between doctor and patient [1–3]. A high proportion of GPs report that work related to sickness certification constitutes a work environment problem [4]. Work involving sick leave issues also covers rehabilitation, and the sickness certificate must communicate plans for medical treatment and rehabilitation, and also recommend whether vocational rehabilitation is needed. Nilsing et al. [5] found that rehabilitation is infrequently prescribed in sickness certificates, and this may indicate that rehabilitation needs were not identified [5].
Early intervention may be beneficial for the individual’s likelihood to return to work [6, 7]. A long sick leave period can affect the individual’s skills and motivation, which can then lead to difficulties in the return to work process [8, 9]. Early assessment to determine individual functioning and rehabilitation needs is crucial for beginning any intervention. Another important factor in sickness certification and rehabilitation is healthcare professional interactions with sick-listed patients. Perception of the healthcare encounter can affect the individual’s ability to return to work [10–12].
Studies indicate that a rehabilitation staff or team may support the sick leave assessment [13–16]. Mannelqvist et al. [17] found that sickness certificates tend to be good quality when they are based on team assessments or when occupational therapists (OTs) or physiotherapists contribute in the assessment. Despite this knowledge, GPs rarely consult with other healthcare professionals regarding work-related problems of patients [18]. In occupational therapy, the focus is on how human activity affects the individual in various situations in life, and activities that are highly important for health and well-being [20]. Occupational therapists (OTs) assess the client’s ability to perform daily activities in interaction with environmental conditions [21]. Such activity assessments could be a useful adjunct to the physician assessment because the sickness certificate requests information on how the disease affects the individual’s activity. Clarification of the individual’s daily activities and activity limitations in everyday life and at work are important for rehabilitation planning for return to work [13, 23]. OTs have knowledge and skills that can provide information about an individual’s ability to work, but they are seldom asked to do assessments during the sick leave process [19].
A working approach where OTs assess a patient’s work-related activity may provide complementary information for the GP who is issuing a sickness certification and identifying the need of rehabilitation. Such a working approach may support GPs and increase collaboration between GPs and OTs. Owing to the GP-reported problems with issuing sickness certification, and the knowledge gap surrounding how collaboration with other professions can be of assistance, this study set out to examine how OT assessments can contribute in such a collaboration. This study aimed to evaluate the feasibility of a working approach in which supplementary assessments by occupational therapists were carried out for persons on sick leave. The research questions were as follows: Does a collaborative working approach between GPs and OTs at HCs result in more sickness certificates with more complete information? Does this approach make any difference in patient experience of healthcare encounters compared to HCs without this intervention? How is this intervention received in a primary care context? How can the implementation process affect the outcome?
Methods
An intervention project was designed, aimed at developing work ability assessments in primary healthcare. Sick leave patients at four healthcare centres were offered a supplementary work ability assessment by an occupational therapist. Patients with an acute infection or pregnancy were excluded. Patients who were scheduled to return to work within 14 days were excluded. A rehabilitation coordinator provided the patients with oral and written information about the work approach. Patients who agreed to participate provided signed informed consent and met with an OT for a complementary assessment and interview to investigate activity- and work-related problems. The OT recorded gender, age, diagnosis/symptoms, whether the sick leave period was new or an extension, which methods or instruments should be used for evaluation, and whether further intervention was needed. The rehabilitation coordinator also recorded the same data for individuals who declined to participate.
The OT explored patient resources and barriers for work using observations, interviews (inspired by Assessment of Work Performance [24–26], the Work Environment Impact Scale [27] and the Worker Role Interview [22, 28]) and patient self-assessments. They investigated the patient’s current work and its requirements, the patient’s occupational performance, psychosocial and environmental factors that could affect work ability, and the patient’s perception of his or her ability to work. The OT assessments were documented in the patient’s electronic health record. The GP had access to and could use the OT assessment when deciding whether to extend a sickness certification or if rehabilitation was necessary.
Settings and participants
The intervention project took place in a county in northern Sweden, starting in September 2010 and lasting until June 2011. We tested the feasibility of this work approach at four healthcare centres. Other, non-intervention healthcare centres in the county were used as comparison group.
Recruitment process
Population-based data and sick leave data for the county healthcare centres were reviewed. The recruited HCs were similar to the comparison HCs in sickness rates. The intervention project concept was presented at a meeting attended by OTs from different county healthcare centres. The OTs were asked to consider whether their HC could be used to test the intervention. The project manager visited the HCs that met the inclusion criteria for sickness rates and were interested in participating. Four HCs of varying size, representing both urban and rural areas were recruited. The project manager gave a detailed presentation and review of the project to the GPs, OTs and clinical department managers at interested HCs. An agreement on compensation for the additional OT time was made between the clinical department manager and the project management. Additional funds were distributed to the participating HCs, depending on their size (i.e., the number of patients registered at each HC). OTs at participating HCs also had assignments as rehabilitation coordinators.
The project started with three half-day seminars with the four participating OTs. The group agreed on which instruments would be used for the assessments. Since patients could have varying types and degrees of disease or illness, activity and work limitations, the choice of instruments and assessment methods would need to be adapted for each individual. During the project period, the OTs participated in a course, ‘Assessment of Work Ability’ that included education on the instruments Assessment of Work Performance [24–26], the Work Environment Impact Scale [27], and the Worker Role Interview [22, 28].
The project manager assured that the OTs had access to materials, instruments, and had regular meetings with the OTs. During the first four months, the group met twice a month. During the remaining intervention time they met once a month. Permanent meeting agenda items were the process logistics, how sick-listed patients came to the rehabilitation coordinator’s attention, content of the assessments, instruments and methods, and the vocabulary to use in documentation (and information transfer). Vocabulary standardization included use of the International Classification of Functioning (ICF) [29].
Anonymised cases were discussed at the meetings, with emphasis on assessment of activity, how to clarify and express activity limitations, and how activity relates to impairment of body function and to the diagnosis. Two OTs from a specialty clinic were invited to present their experiences with assessment and documentation. A follow-up meeting with the GPs was conducted at each of the participating HCs (January–February 2011). The project manager was in contact with the clinical department managers to keep them apprised of the plan.
Data collection
Sickness certificates and patient questionnaires were collected from the intervention and non-intervention HCs. Participating OTs and GPs from intervention HCs were interviewed after the project was completed (Table 1).
Number of participants from each group and the type of data collected
Number of participants from each group and the type of data collected
HCs = healthcare centres; GPs = general practitioners; OTs = occupational therapists.
The primary outcome measure was the quality of the sickness certificate, and this was based on provision of sufficient information for the social insurance officer’s assessment and prescription of rehabilitation.
During the project, 122 patients were identified at the intervention HCs by the rehabilitation coordinator (72% women, 28% men). Of these, 57 (47%) agreed to an OT assessment, and 40 (33%) agreed to answer questionnaires. Of those who declined the OT assessment, 65 (75%) were women. The reasons they did not participate included weakness, lack of ability to read and write, referral to another clinic, patient had returned or was returning to work, had chosen another HC, or did not come to their appointment.
Rehabilitation coordinators at the intervention HCs sent coded, anonymised sickness certificates to the project manager for the 57 participating patients who had a new sickness certificate issued after the OT assessment. Excerpts of the OT medical record documentation on current patients were coded and sent to the project manager. To obtain comparison material, ten officers employed at the Swedish Social Insurance Agency (SSIA) were asked to copy incoming sickness certificates received from county HCs two days a week for the duration of the project. These certificates were sent to the project manager after removal of identifying information. The project received 338 certificates as comparison material. Five of these were issued by specialty clinics, four were issued by one of the intervention HCs, and one certificate lacked the second page. If more than one certificate for the same patient was received, only the first certificate was used (n = 62 [18% ]).
Questionnaire
Three months after initial contact with the healthcare centre, the project manager sent a follow-up postal questionnaire to the participants. The questionnaire was designed by the research team and consisted of 25 questions. The introduction gathered demographic information such as gender, age, educational level, and HC, and included questions on whether the patients were on sick leave, whether or to what extent they had returned to work, and whether they were seeking work. This was followed by 18 questions (with fixed response alternatives) on the quality and content of the HC encounter and were designed with positive statements presented in a Likert scale. Reminders were sent to those who had not responded after three and five weeks. Twenty-five patients from the intervention HCs (63%) answered the questionnaire (Table 2). To have comparable data for the same time period, questionnaires were sent to consecutive patients who were issued sick leave certificates from non-intervention health centres within the same geographical area. Rehabilitation coordinators from six other HCs provided names and addresses of persons who had been on sick leave. In total, 345 questionnaires were sent to patients who had contact with non-intervention HCs. Reminders were sent after three weeks. A total of 142 (41%) questionnaires were received from patients as comparison material (Table 2).
Demographic data for sick-listed persons who responded to the follow-up questionnaire
Demographic data for sick-listed persons who responded to the follow-up questionnaire
*Return to work; HCs = healthcare centres.
Data on sick leave were retrieved from the County Council ‘Diver’ system, a production management system (software designed to retrieve and analyse) that reports data from the patient record.
Occupational therapist and general practitioner interviews
After completion of the project, the project manager conducted in-depth interviews with the four participating occupational therapists and 12 of the 22 GPs at the intervention HCs.
Analysis
Since data collection included questionnaires and interviews, convergent parallel design, a mixed method design, was used for data analysis [30].
Sickness certificates
Evaluation of the quality of the sickness certificates was performed by an insurance specialist at the SSIA. The methods used were in accordance with the criteria used as in the Swedish national ‘Sick Leave Billion’ quality evaluations of sickness certificates. The certified insurance specialist had participated in those standardized evaluations. The quality in the sickness certificates and gender differences in quality, have been published elsewhere [31]. We also analysed whether the OT assessment was used in issuance of a new sickness certificate for extension of the sick leave period.
Quality and content of the encounters
Comparisons between the answers from intervention HC patients and patients from other HCs were made. The proportion in each group that answered Agree, Strongly agree and Totally agree on the positive statements were compared. Analyses of gender differences between answers were also made.
To compare different variables between groups, Pearson’s chi-square or Fisher’s Exact Test was used. Differences were considered significant at a P value≤0.05. All analyses were calculated using two-sided tests. Statistical analyses were performed using Statistical Package for Social Sciences 21 (SPSS Inc., Chicago, IL, USA).
Sick leave
Mean number of persons on sick leave during the first halves of 2010 and 2011 were compared between the intervention (n = 4) and non-intervention (n = 30) HCs. Number of persons on sick leave per 1000 inhabitants, number of persons on long-term sick leave (>90 days) per 1000 inhabitants, and number of adjusted sick days per 1000 inhabitants were examined. Adjusted sick days were based on level of sick leave. For example, 10 days at a 50% level of sick leave was calculated as 5 adjusted days.
Implementation
We used the Consolidated Framework for Implementation Research (CFIR) [32] to analyse the project implementation process. CFIR is a compilation of theories relevant to a number of determinants of implementation outcomes and is composed of five major domains: intervention characteristics, outer setting, inner setting, characteristics of the individuals involved, and the process of implementation. CFIR can be used to evaluate and build implementation knowledge and is further described in the Findings section. Data from the OT and GP interviews at the intervention HCs was used in analysis of the implementation.
Ethical approval for all parts of the study was granted by the Regional Ethical Review Board in Umeå, Sweden [Dnr 2010-177-31M].
Findings and discussion
Sickness certificates
One third of the sickness certificates did not contain all the required information. No differences could be identified between the intervention HCs and the other HCs in quality aspects of the mandatory certificate information. However, work tasks were more commonly described in certificates from intervention HCs, and those certificates more often contained information about ongoing or planned treatment and rehabilitation.
It proved very difficult to evaluate whether GPs used the OT assessments. In a few cases, the GP’s information on activity limitation was almost verbatim that of the OT, which indicates that the GP had noted the OT assessment. Some certificates lacked information on activity limitations, even though the information was in the OT documentation, suggesting that the GP had not accessed the OT evaluation. In some cases, the SSIA had requested supplementation but information on activity limitations was still missing in the new certificates. In most cases, it was not possible to determine whether the GP had utilized the OT assessment; certain words and descriptions might be the same, but it was hard to know whether the GP was influenced by the OT assessment.
Quality and content of the encounters
A higher proportion of the patients from the intervention HCs reported that they had talked about their work tasks (p = 0.03) and home situation (p = 0.01) during their HC encounters. Patients from intervention HCs more commonly reported that HC personnel performed what was agreed upon (p = <0.01), and that they were given enough time in the encounter (not significant).
A higher proportion of men than women agreed on the positive statements. We found significant differences in the statements I felt involved in my care and treatment (p = 0.01), I got information on where to go if I need help or have more questions (p = 0.04), During the encounter, we talked about my work ability (p = 0.01), and During the encounter, I received information about the sickness certificate (p = 0.02). Fewer women than men felt that the staff had listened (p = 0.05) and that they were given the opportunity to express their opinions (not significant).
Data on sick leave
The mean number of persons on sick leave, the number on long-term sick leave per 1000 inhabitants and the number of adjusted sick days at the intervention HCs were similar to the other HCs in the county at the start of the project. The mean sick leave data increased during the first half of 2011 compared to the same period in 2010 in the intervention and other HC groups. The increase was not significantly lower at the interventions HCs (Table 3).
Data on sick leave for the healthcare centres, January–June 2010 and January–June 2011
Data on sick leave for the healthcare centres, January–June 2010 and January–June 2011
HCs = healthcare centres.
This section presents an analysis of the intervention implementation according to the five domains of CFIR [32].
Intervention characteristics
These comprise stakeholder perceptions of the intervention, whether it is externally or internally developed, whether there is supportive research, whether there are advantages to its use, and its adaptability [32].
The intervention was externally developed and had not been not previously tested. Although previously published studies indicate that assessment of work ability may be improved if several professionals are involved [14–17, 19], this working approach had not been tested and experienced evidence could not be shown. The intervention may therefore have lacked legitimacy and the fact that the users had not developed the method themselves may have been a barrier. How well an intervention fits into existing, ongoing processes and opportunities for adaptation affects how easily an intervention can be implemented [33]. Our intervention process could be adapted to the different HC conditions. The ability of devices to adjust also means that there are a number of choices that can be difficult to relate to. The complexity of the intervention required a change in procedures to handle patients on sick leave.
Outer setting
This refers to the probability of the intervention meeting patient needs, and whether there are external strategies to spread the intervention, such as policies, recommendations or guidelines [32].
Although patients were informed, benefits of the intervention might not have been obvious to them. Therefore, some patients may have dismissed the opportunity to participate. Since this was a research study, patient consent was required, and attempts to convince patients to participate were not allowed.
“I think the patient may really benefit from the cooperation with OT, if they want it. Everyone doesn’t.” [Interview GP7]
“Sometimes I want (too) so much, but it’s impossible to force anyone.” [Interview OT1]
“If the patient feels that the physician trusts me as an OT, that’s a big advantage.” [Interview OT3]
This intervention did not originate from national guidelines or policy documents. National directives were mandated to county councils to improve their work in the sick leave process and increase the quality of sickness certificates. The current intervention was part of the local county council’s internal development of the sick leave process.
Inner setting
This refers to the organisational social structure and size, including network and communication within the unit, as well as its norms, values, and implementation climate [32].
The project management had not examined and had no detailed knowledge about informal communication in the units where the intervention was tested. The internal culture of norms and values was unknown. The strategy for communication involved the clinical department manager, workplace meetings, and physician group meetings. The intervention did not include performance incentives, rewards, or other incentives that could attract a change in approach.
Interest in the intervention may have been different among the individuals involved. The clinical department manager may have seen this as an opportunity to obtain additional resources. The occupational therapists may have seen an opportunity to advance their profession. GPs may have viewed this as an opportunity to receive support. There may have been too little focus on the fundamental motive of the ‘patient’s best interest’ for those who were involved.
The intervention was designed to fit into the regular work routine. The amount of added work that was due to the project was replaced with additional resources. The National Board of Health and Welfare in Sweden has pointed out the importance of quality sick leave process work for patient safety [34], and that this can be supported by local routines and management systems. Quality work in sick leave processes has historically had a low priority in healthcare from the top management down through the organization [3, 36].
In order to analyse the intervention effect, all GPs at the intervention HCs had to follow the same routines. To discern whether ‘no result’ can be attributed to the intervention itself or lack of fidelity, adherence must be confirmed [37]. Primary healthcare is not organized in a way that supports collaboration or time for reflection and feedback. There may be a conflict between time for patient care and time for reflection and education [38]. This intervention did not have strong support for following up on the goals and providing feedback to the GPs. On the contrary, the implementation had an advantage for occupational therapy resources, and the OTs received education and supervision during the project. The OTs were well-placed for combining their regular work duties with the intervention. Innovations developed within their profession are easier to embrace and incorporate than methods originating from other professions [39]. To promote an implementation, the target group must be involved and have the opportunity to adapt it to their situation [40].
The clinical department manager had a role in getting employees to participate in the working model. Leadership is crucial for healthcare development. In Sweden, work involving sick leave issues has historically been handled by physicians and there has been a lack of leadership guidance and control [3, 36]. Although a system for managing sick leave work issues has been introduced, tendencies may be retained of allowing the healthcare profession to determine where and how the work will be performed [41].
Characteristics of individuals
These encompass individual attitudes, the value of the intervention, and familiarity with facts and principles related to the intervention. They include individuals’ confidence in their abilities and knowledge, willingness to change, and identification with the organisation [32].
In Sweden, only physicians (and in some cases, dentists), have the right to issue sickness certificates. The attitude towards work with sick leave issues may have differed between individuals.
“There is no one who says that GPs do a great job with sickness certifications ... so I think they [the legislator] should start from scratch [change the system]” [Interview GP10].
Problems that the intervention was meant to solve had to be understood to be an area given priority and in need of quality improvement. Although assessment of work ability and sickness certification were considered problematic, GP opinions about who has responsibility to solve the problem varies. The intervention might not have been considered the preferred solution. Some expressed the opinion that if the SSIA needs more information in the sickness certificate, it is a problem for SSIA, not healthcare. Healthcare and the SSIA have a shared responsibility, but sometimes physicians hand their responsibility to the SSIA [42].
GPs do not have the practise of using other professions’ assessments when issuing sickness certificates. The GPs may have been hesitant to use this working approach.
“I do not have much experience in working with an occupational therapist” [Interview GP2]
“I think that many GPs think, to a great extent, that they should do everything themselves.” [Interview GP10]
“I work mostly on my own when I handle sickness certifications” [Interview GP11]
A study from the UK [43] found that expanding sickness certification beyond GPs, and including other professions is complex and more than organisational barriers. The difficulty is related to deeply ingrained values that are supported by professional identities. The OTs may have had differences in confidence that their assessment skills and knowledge could be useful for the GPs.
It’s sad, if GPs do not think OTs can supply with something ... but different occupations have a different status. I’m probably at the bottom. [Interview OT1]
Although occupational therapists have always assessed activity and activity limitations, they generally have limited experience in assessing these in relation to patient work. Occupational therapy instruments for assessing work ability have been recently developed, but it takes some time to find routines and become fully familiar with new methods. The identity of the individuals involved in the organization is also very important.
The intervention required increased cooperation between GPs and OTs. Cooperation requires a willingness from both sides, and perceptions of advantages to cooperation is a positive factor.
“It goes slowly when several people are involved. It will be an ungainly organization ... It can create waiting time” [Interview GP2]
An employee who has worked for a long time in a profession has a variety of experiences. These past experiences may have affected the process [39]. Some individuals believe and experience that collaboration is enriching. Others prefer to work by themselves [16, 44]. Even if collaboration in work with sickness certification is seen as important [42], collaboration may be difficult to achieve, and patient needs are often handled from each professional’s perspective [35].
Process of implementation
This entails the implementation strategy and consists of four parts: planning, attracting and involving appropriate individuals, carrying out the plan, and consecutive feedback [32].
No advance analysis of the participating HCs systems and processes were made, and this may be a weakness in the implementation process. OTs were attracted to the new working approach and became the opinion leaders. During the project, they had support and the opportunity to reflect and get feedback through continuous meetings with the project manager. However, opinion leaders were missing within the group of GPs. If better preparation had been made by inspiring the HC GPs with an insurance medicine mandate, they might have become ‘champions’ and supported and promoted the working approach. A follow-up meeting was held at each HC during the project. Disseminating information to everyone at a HC unit is a well-known problem. Not all of the GPs were able to attend the initial meeting, or the follow-up meeting. This means that some of the GPs received information from a rehabilitation coordinator.
If several professionals are involved in the same case, they do not automatically cooperate. We found that the GPs lacked a routine of reading and participating in other professions’ records and therefore the system for certification was used to alert the GP.
“I like the system for certification but there are many colleagues who are completely upset about it.” [Interview GP11]
“Certification of records can be a good thing, but if there is too often and too much, the function can lose its worth” [Interview GP1]
Multi-professional assessment can benefit the patient if the professional perspectives are made available to each other through conversation [45]. The results show the importance of context, and the implementation researchers concur with this [46]. The HCs previously established routines for sick leave process play a significant role in implementation of the working approach. Furthermore, the staff is not stable, which means that the team changes often, and there are variations in behaviour among the personnel involved in a tested intervention [47]. Improvements in collaboration between individuals must take explicit account of the group processes, and this demonstrates the need for more research on the impact of group process and teamwork across and between disciplines [48].
General discussion
Although there is a general goal for better internal collaboration in healthcare work with persons on sick leave, there were deficiencies in fidelity among users in this study. Limited use of other healthcare professions for work with sickness certification in primary care has been shown in other studies [35]. Physiotherapists and occupational therapists report that their competence was not solicited [19, 49], and physicians have reported problems with counselling by other professionals when it was needed [3].
These results had a significantly lower number of participating patients than anticipated. Based on historical sickness absence data, we estimated that approximately 200 patients would be identified by the rehabilitation coordinators at the participating healthcare centres during the project. At the end of the study, 57 patients had agreed to meet the OT for a complementary assessment, 40 of those agreed to participate in a follow-up survey, and 25 answered the questionnaire. The patients who declined (n = 65) may have thought they did not need more assessments and encounters; they had received their sickness certificates and were satisfied. Studying the reasons why some patients chose not to participate in the study clarifies that OT assessments are not necessary for all patients with limited work ability. In some cases, and special health conditions, such assessments may be considered inappropriate. Some of those who declined were referred to a specialist, and some already had contact with other professionals at the HC (e.g., a psychologist). Another explanation for declining, was the research project itself. Consent was necessary, and we wanted patients to answer questionnaires, which some perceived as an obstacle.
A similar project at Specialist Psychiatry at Sahlgrenska University Hospital, Gothenburg, Sweden [50] returned positive results. The aim of that study was to examine whether OT knowledge regarding activity and work ability, as well as the specific skills to assess and describe them, could contribute to an increase in quality of sickness certificate contents. There are two major differences between that project was in specialty care and ours in primary care. In the Gothenburg project, the OTs chose and booked, through the physicians’ booking system, the patients who were currently being seen for sick leave. These patients were informed that this was a trial working approach. The Gothenburg project had a higher patient participation rate, and this may relate to the fact that patients did not have to give written informed consent answer follow-up questionnaires. Fewer patients were missed, and this may be partly explained by the OT choosing and notifying current patients, rather than the physician.
Not all patients who were sick-listed by the GPs were identified to the rehabilitation coordinator. This could have occurred because of several factors. Availability of GPs varied during the project period. A number of GPs terminated their employment, and the HCs hired temporary GPs. This lack of continuity made it hard to maintain the routine of notifying the rehabilitation coordinator of sick-listed patients. Some GPs may have considered themselves competent in issuing sickness certificates that had sufficient information, and therefore not in need of another professional’s assessment. In some cases, the GPs may have lacked awareness of what information is required for high quality certificates and identification of the patient rehabilitation needs.
Health professionals at the intervention HCs raised the issue of patient work tasks in the questionnaires and the sickness certificates. We cannot conclude that the intervention influenced this based on this study, but it is important to consider that insufficient knowledge of patients’ work demands [35] and GP restriction of the assessment and lack of inquiry about a patient’s work situation were found in another study [51].
Information on what the OT could provide, whether a multidisciplinary assessment is suitable, and identification of which patients would benefit, and who would benefit from extended assessments is necessary. The healthcare system structure is often linked to diagnoses. The diagnosis can be important, but accurate assessment of work ability is the critical element in the sick leave process. The individual’s work (work requirements, working conditions) constitutes a major influencing factor that affects the assessment and support an individual may need. Finding variables other than the diagnosis, which can identify the patients who will benefit from multidisciplinary assessment and deploying different healthcare professionals effectively is important [42].
The CFIR analysis showed that this intervention could be adapted to meet local needs. This project intervention’s flexibility may have created an ambiguity which could be part of the lack of fidelity. This indicates the need for a clear intervention structure. To change working practices, there has to be awareness of the problem and readiness for implementation of changes [52, 53].
Changing a workplace process involves changing employee behaviour. According to Kielhofner [54], human activity and occupational behaviour comprise three subsystems: volition, habituation and performance. The volitional system is based on personal values, interests and self-awareness that affect motivation. The habituation system consists of habits and roles that support the organization of activity patterns or routines. Performance capacity refers to the physical and mental abilities that underlie skills. The way people act is a process that is controlled by motivation and is an intrinsic part of humans [55]. Creating new habits and roles requires a lot of willingness and motivation. If personal values and interest in improving work with sick leave issues are perceived as unattractive and are not prioritised [35], there may be difficulties in changing the approach to work.
The aim of this intervention was to develop and improve assessment of work ability prior to a decision on certifying sick leave. Studies of quality improvement are mainly applied science and a social process that is dependent on context [56]. Prior to implementation of this intervention, a more comprehensive analysis of participating HC conditions should have been made. More involvement by the clinical department manager might have facilitated implementation, although organizational support and environmental context need to be considered [57, 58]. Special coaching and education is needed for all the users [37].
Methodological considerations
The study aimed to test an intervention in a natural context. In many research fields, studies focus on efficacy and are therefore conducted to investigate the effects of actions under ideal circumstances [59]. Results from such research are often difficult to transfer into practice. Therefore, conducting studies in a clinical context that focuses on feasibility is important so that results can be applied in reality [59]. Primary healthcare has no tradition of being a research environment and there are barriers to intervention studies in HCs. In this study, we did not get all users to embrace our intervention, and there is a need to continue to strive for practice-oriented study designs. Since implementation of change in healthcare takes a long time [37], we have to persevere when we try new approaches. The use of CFIR [32] to analyse the project implementation process was useful, since that framework of determinants helped us to understand strengths and weaknesses of the implementation. The analysis with CFIR also clarified that further development of the current intervention will require more effort and time, and an applied implementation strategy. For example, strategies for improving patient care by Grol et al. [40] could provide support to plan and/or execute an implementation. The interviews with participating OTs and GPs will provide additional knowledge about how this intervention can be developed and used within primary healthcare, and this may facilitate the feasibility of this approach (to be published). The use of a mixed method design for data analysis was useful, as it allowed us to report on sick leave and the experiences of the participating OTs and GPs. An alternative study design such as ‘action research’ [60] needs to be considered for further research on developing multi-professional assessment and cooperation for persons on sick leave in primary healthcare.
Conclusions
To facilitate implementation of a collaborative working approach, there must be careful preparation of all the users with respect to knowledge of, and trust in, the intervention. Regular meetings for users are suggested. A collaborative working approach to improve sickness certification can only be implemented if the users a) perceive work with sick leave issues to be problematic, b) consider that healthcare has a responsibility to improve the quality of sickness certification, and c) are convinced that collaboration between different healthcare professionals is crucial for improvement.
Conflict of interest
None to report.
Footnotes
Acknowledgments
This work was supported by AFA Insurance Sweden (grant number 090261).