Healthcare consumption among patients with stress-related exhaustion: a register-based study in Swedish primary care

Abstract

Objective

There is limited understanding of healthcare consumption among patients seeking care for stress-related exhaustion. This study examines the number of care contacts and treatments received by patients diagnosed with Exhaustion Disorder (ED) in primary care, as well as its association with psychiatric comorbidity, sex, age, location, and the governance of primary care centers (PCCs).

Methods

Data from the Region Västra Götaland healthcare database Vega was analysed including patients diagnosed with ED for the first time during 2018 and 2019, in total of 11,058 patients. Number of care contacts and treatments registered the 12 months before and after the individual date of first time ED diagnosis were compared.

Results

Number of care contacts and treatments and proportion of patients receiving care increased post-ED diagnosis. The variation is large, and patients diagnosed with another psychiatric condition, most often depression or anxiety, in addition to ED received more care. Female patients tended to receive more care. Minor variations were observed with respect to age and PCC population size, while larger differences were identified between municipalities. Public and private PCCs showed a similar pattern for both care contacts and treatments.

Conclusion

Healthcare consumption varies greatly among ED patients seeking primary care. This variation is mostly related to the burden of psychiatric comorbidity and, to a lesser extent, to gender and age. The wide variety of treatments used, along with the gender and age differences observed, warrant further analysis, as these patterns may not align with current evidence for treating patients with ED.

Introduction

Stress-related mental health problems have, since 2010, been the fastest growing cause of sick leave in Sweden according to the Swedish Social Insurance Agency [1], with incidences increasing by 25% between 2019 and 2024 alone [2]. The increase in sick leave due to stress at work is also a general trend seen across Europe, generating high societal costs [3]. Many authors have highlighted the challenges regarding which clinical diagnosis should be used for patients seeking care for stress-related exhaustion/burnout [4,5]. This is an important issue because the treatment provided to patients largely depends on the specific diagnosis of their condition. Burnout is included in the 11th Revision of the International Classification of Diseases (ICD) as an occupational phenomenon but is not classified as a medical condition. In Sweden, the criteria-based diagnosis of exhaustion disorder (ED) has been used since 2005 to define patients seeking care for stress-related exhaustion classified as F43.8A in ICD-10 but will receive different classification due to the revision of ICD-11.

ED has, in several studies, been characterized by a variety of symptoms both in the acute and chronic stage [6–8]. All patients with ED report extreme fatigue as this is included in the diagnostic criteria. Most patients also experience a wider range of symptoms beyond the minimum required for diagnosis, such as cognitive impairment, sleep disturbances, reduced stress tolerance, musculoskeletal pain, chest pain, and gastrointestinal issues. Comorbid depression and anxiety is also commonly reported [9,10]. A recent study exploring symptoms of more than 500 individuals with self-reported ED showed a complex picture encompassing several physical, cognitive, and behavioural symptoms additionally to the core symptoms that are required for fulfilling the criteria for [11]. ED is mostly associated with work-related factors, but private-related stress exposure is also evident for many of the patients seeking care [12–15].

Patients diagnosed with ED typically have longer sick leave durations compared to those with other stress-related disorders, often extending up to six months. This contrasts with sick leave for i.e. acute stress reactions, which usually lasts around one month. This difference is understandable, given that patients with ED report a higher burden of symptoms and greater functional impairment [1]. The long duration of sick-leave and the heavy burden of symptoms calls for effective treatment and prevention.

Although several treatments for ED have been studied, it is difficult to draw firm conclusions about their effectiveness in alleviating symptoms or facilitating a return to work. High-quality studies are few and show only marginal treatment effects, resulting in limited evidence for any specific treatment [16–21]. Although results suggest that sick leave decreases over time independent of intervention, the evidence is growing that an early and continuous contact with the workplace might decrease sick leave and improve return to work for patients on sick-leave for stress-related exhaustion [22–25].

The Swedish healthcare system is tax-funded and includes both public and private healthcare with first level primary care and specialist care (outpatient and inpatient care). Private care providers are mainly tax-funded and thus reimbursed by the county. Very few private clinics exists that are not connected to the tax-funded system. These typically offer health screenings or have insurance agreements to cover the costs. In Sweden, most individuals with stress-related mental disorders including ED are treated in the primary care [26]. The primary care is parallelly organized in primary care centres (PCC) and rehabilitation centres (RC). These two organizations cooperate but are not linked to each other with respect to i.e. financing and thus patients can seek care independently to a RC without having contact with a physician in a PCC first.

Patients in Sweden are free to choose their healthcare provider, and care must be given on equal terms and should encompass a broad range of medical competences (National Board of Health and Welfare, 2005). Recently published regional guidelines for treating patients with ED emphasize that those seeking care for stress-related exhaustion require structured care and treatment tailored to their individual needs (Västra-Götalandsregionen, 2024).

Currently, there is limited knowledge about the treatments being offered to patients with ED in primary care settings. Information on the consumption of care, including visits to different professionals, is also scarce. Therefore, this exploratory study aims to describe the care provided to patients with ED in a primary care context, focusing on the volume of care contacts with various professionals and the treatments offered. Additionally, the study seeks to determine whether the care provided is equitable by examining its dependence on factors such as the care provider, geographical location, patient comorbidity, age, and sex.

Aims

The primary aim of the study is to explore the number of care contacts to different care givers of patients diagnosed with ED in primary care, one year before and one year after receiving an ED diagnosis. Additionally, the study aims to explore whether there are any differences in the number of care contacts related to psychiatric comorbidity, sex, age, or the location and governance (public or private) of the PCC.

A secondary aim is to explore which treatments among those recommended in the regional medical guidelines for ED are offered to these patients during the same time periods, and whether there are differences in this respect related to psychiatric comorbidity, sex, age, or the location and governance (public or private) of the PCC.

Method

Study design and setting

The present study was based on register data from the healthcare database Vega, managed by Region Västra Götaland and is a part of a larger research project entitled ‘Treatment of stress-related exhaustion disorder in Swedish primary care (TED)’ with the overall aim of exploring the care-process of patients with stress-related exhaustion.

Vega contains healthcare information about the inhabitants in Region Västra Götaland which is located in the western part of Sweden and has a population of 1.8 million (Statistics Sweden, 2023) and which houses 49 municipalities. There are roughly 200 different PCCs and 130 different RCs in the area, of which approximately half are operated by the private sector, the other half being operated by the public sector.

Subjects

Working age patients (ages 18–67) with an ED diagnosis made by a physician in a PCC were included in the study. Patients who received their ED diagnosis between January 1, 2018, and December 31, 2019, without a record of ED diagnosis the year before the date of their diagnosis, were considered eligible. Altogether 11,058 patients met these inclusion criteria.

Data handling

Data from January 1, 2017, to December 31, 2020 was retrieved ensuring that care contacts and treatments one year before and one year after the patients received the ED diagnosis (in 2018 or 2019), were included in the data set. The individual date of first time ED diagnosis was used to compare the number of care contacts and treatments during the 12 months before (pre-ED) and the 12 months after (post-ED) the patient received the ED diagnosis.

Data from both PCC and RC was analysed using the following variables: 1) Care contacts: registered digital- or in-person care contacts (but no other contacts e.g. via email, letter, etc), 2) Registered treatments according to the Swedish National Board of Health and Welfare classification of Health Interventions (KVÅ) [27], 3) Other variables used to compare different patient groups: ICD- 10 F-diagnoses (mental, behavioural and neurodevelopmental disorders), sex, age, PCCs governance and location. The ICD-10 diagnosis was handled as a single binary variable - any F diagnosis vs. none, in addition to ED.

To increase the probability that the treatments analysed were connected to the ED diagnosis and not to other conditions that the patients might have sought care for during the same period, we only analysed treatments recommended in the specific regional medical guidelines for treatment of ED [28,29]. Three experienced clinicians—a physician (first author), a physiotherapist, and a psychologist—each with extensive experience in treating ED patients, reviewed all available KVÅ codes. All three clinicians have worked in primary care and have firsthand experience using the KVÅ codes. A complete list of the interventions included in each category is provided in the (supplement Table S1). This resulted in 160 different registered KVÅ interventions/treatments being included which were categorized into ten different treatment categories: (1) General information/education, (2) Individual information/education, (3) Collaborative Care, (4) Multimodal rehabilitation (MMR), (5) Diverse rehabilitation (pain, cognition, activities, adjustment) (6) Group treatment for stress, (7) Guided physical activity, (8) Guided physical activity with focus on mental health/relaxation, (9) Psychotherapy, 10) Other psychological treatment. Those KVÅ codes not included in the analysis were judged to be related to other symptoms and diagnosis such as allergy, infections etc. Information on pharmacological treatment and sick leave is not available in the VEGA database and thus not included.

The patients were divided into three age groups: young age, younger middle-age, and older middle-age, 18–29, 30–44 and 45–67 years in accordance with the age groups used in the National Public Health Survey, a collaborative effort between the Public Health Agency of Sweden and Swedish regions. The age groups represent different life transitions related to entering the labour market/higher education, starting a family/career building and changing careers/late working life.

The PCCs were divided into three groups based on the population size and geographic density according to the classification of Swedish municipalities by the Swedish Association of Local Authorities and Regions [30]: (1) Large cities and medium-sized towns and municipalities (≥40 000 inhabitants), (2) Small towns and municipalities (between 15 000, and 40 000 inhabitants), (3) Rural municipalities (˂15 000 inhabitants).

Data analysis

Statistical analyses were performed with SPSS statistics, version 28.0.1.1 (15). Chi-2 test for statistic difference in count/percentages and non-parametric Mann-Whitney test was used to analyse statistic difference in median for non-normally distributed data.

Results

Most of the patients were female, with the majority being over 30 years old. Slightly more than half were registered at a public PCC, while nearly half were registered at a PCC in large cities or medium-sized towns and municipalities (Table 1). A consistent female-to-male ratio of 3:1 was observed across different age groups, PCC organizations, and locations (data not shown).

Table 1.

Characteristics of patients with exhaustion disorder included in the study and distribution between different primary care centers during 2018–2019.

Characteristic	n (%)
Gender
Women	8452 (76)
Men	2606 (24)
Age groups
18–29	1328 (12)
30–44	4555 (41)
45–67	5175 (47)
PCC organisation
Public	5839 (53)
Private	5219 (47)
PCC location1
Large cities and medium-sized towns and municipalities2	5245 (47)
Small towns and municipalities3	3288 (30)
Rural municipalities 4	2525 (23)

¹ Based on SALAR classification.

² ≥40 000 inhabitants.

³ between 15 000, and 40 000 inhabitants.

⁴ ˂15 000 inhabitants.

Care contacts

The total number of care contacts per year post-ED compared to pre-ED increased by 75% (158,084 vs 91,003 care contacts). Further, as shown in Figure 1, the increase in visits follow a pattern of increased percentage of patients meeting with all health professions except for nurse and assistant nurse. The number of registered care contacts varied largely among the patients, with a small number of patients accounting for the large proportion of the contacts. Thus, the year before diagnosis the median was 5 visits (range 0–118) and the year after the ED diagnosis was set the median was 14 visits (range 0–156). Nine patients did not have any registered contacts in primary care during the two-year study period, indicating that they received their ED diagnosis without a registered care contact with a physician.

Figure 1.

The proportion of primary care patients with at least one registered care contact with various professions one year before and one year after the exhaustion disorder (ED) diagnosis in 2018–2019. Contacts with ‘other’ professions accounted for less than 2% of all registered care contacts.

Most of the patients consulted two or three professions and less than 10% of the patients consulted more than four professions, either pre- or post- ED.

Care contact patterns in relation to patients’ psychiatric comorbidity, sex, age, PCC location and management

Patients diagnosed with co-morbid psychiatric diagnoses (n = 8686), mainly anxiety (55%) and depression (44%), in addition to ED were registered with significantly more care contacts both pre- and post-ED, compared to those without reported psychiatric co-morbidities (n = 2372) (Figure 2).

Figure 2.

Distribution of number of care contacts of patients with or without psychiatric comorbidity, one year before and one year after the exhaustion disorder (ED) diagnosis in 2018–2019. The interior of the box indicates 50% of the distribution. The thin perpendicular lines outside the box indicate the minimum and maximum number of care contacts and a perpendicular line inside the box shows the median number of contacts. Outliers are represented individually by dot symbols.

Overall, significantly higher proportion of female than male patients had registered care contacts. Female patients also had significantly higher mean number of care contacts, both before and after receiving the ED diagnosis (Supplementary data, Figures S1, Tables S2 and S3).

Significant differences were observed in care contact patterns related to age (Supplementary data, Table S4). Among the older patient groups, it was more common to visit a physiotherapist, an occupational therapist, a nurse, or an assistant nurse whereas a visit to a psychologist was more common for younger patient, both pre-and post-ED diagnosis.

Care contact patterns with the various professions offered to patients with ED was similar when public and private PCCs were compared, with some significant differences regarding few professions (Supplementary data, Table S5).

The number of care contacts varied somewhat between different municipalities (mean number of care contacts ranging between 9 and 20 per patient per year post-ED) whereas smaller but significant differences were seen in care contact patterns when comparing PCCs located in cities, towns, and rural municipalities. Focusing on post-ED, the most notable difference was that a higher proportion of patients visited a psychologist (32%) in cities and medium-sized towns compared to small towns (26%) and rural municipal (24%) whereas counsellor was more common in small towns (15%) and rural municipals (15%) compared to large cities (8%). More frequent visits were also seen to assistant nurse in small towns (21%) compared to large cities (7%) and rural municipals (13%).

Treatments

Post-ED, a larger proportion of patients had a registered treatment initiative (Figure 3), and more than twice as many treatments were registered compared to pre-ED (165,105 vs 72,600 treatments).

Figure 3.

Proportion of patients registered to at least one session of various treatments in the primary care one year before and one year after receiving the exhaustion disorder (ED) diagnosis in 2018–2019.

The number of registered treatment sessions per patient varied largely both pre-ED (range 0–200) and post-ED (range 0–269). Also, 9% (n = 982) of the patients did not undergo any registered treatment during the two-year period.

Very few patients received MMR (Figure 3) along with other treatments, but these patients had relatively high number of care contacts, per patient post-ED (median = 34 care contacts) compared to the median for patients that did not participate in MMR. For example, patients registered to receive guided physical activity with focus on mental health/relaxation along with other treatments had a median of 25 care contacts and for those receiving collaborative care along with other treatments, the median was 16 care contacts.

Treatment pattern in relation to patients’ psychiatric comorbidity, sex, age, PCC location and management

Similarly to care contacts, patients diagnosed with co-morbid mental diseases in addition to ED where registered with significantly more treatment sessions both pre- and post- ED compared to those without co-morbid diagnoses (data not shown).

A higher proportion of female than male patients were registered for all treatment categories, both before and after receiving the ED diagnosis (Supplementary data, Figure S2).

The oldest group of patients was less likely than the two groups to receive psychotherapy and other psychological treatment both pre-and post-ED diagnosis and they were also more likely to receive guided physical activity post-ED diagnosis. Thus, post-ED, 45% of the 18–29 years received psychotherapy compared to 34% among those 45-67 years old (p < 0.001). Guided physical activity was offered to 31% of the oldest group of patients compared to 23% among the younger group (p < 0.001).

The proportion of patients receiving different treatments was similar for public and private PCCs with some marginal differences, the most notable being that psychotherapy and individual information/education were somewhat more commonly offered in private PCCs (40% vs 32% and 33% vs 28% of patients with at least one treatment post-ED) as opposed to collaborative care and group treatment being more common in public PCCs (32% vs 27% and 26% vs 22% of patients with at least one treatment post-ED) (Supplementary data, Table S6).

The treatments offered were generally consistent across urban and rural PCCs. However, notable differences emerged post-ED, with psychotherapy being more prevalent in rural municipalities (45% vs. 37%) and collaborative care more frequently implemented in small towns (34%) and rural areas (32%) compared to large cities (28%).

Discussion

One of the main findings of this study is the large variability seen regarding the frequency of care contacts and treatments offered to patients with ED seeking primary care both during the 12 months before and 12 months after the date of the ED diagnosis. Patients with psychiatric co-morbidity, mainly depression and anxiety, consume more care both prior to and after receiving the ED diagnosis, which is to be expected. However, the disparities related to gender, age, and to some extent municipalities suggest that some of the variations may be unwarranted, potentially indicating inconsistency and inequity in the care of patients seeking care for stress-related exhaustion. However, it appears that the type of management or the geographical location of the PCCs does not significantly impact the frequency of care contacts, or the decision of which treatment is offered to patients.

Before the ED diagnosis is established the patients presumably receive care in accordance with the symptoms being presented. After receiving the ED diagnosis there is a noticeable increase in both the number of care contacts and the treatment initiatives for patients both with and without co-morbid psychiatric diagnoses. One speculation is that many physicians probably await decisions on treatment before the appropriate diagnosis is made. Another explanation could be that the patients are initially too exhausted to take part in treatment interventions and thus the bulk of treatment initiatives are most probably being provided a few weeks after receipt of the diagnosis. One additional explanation could be related to accessibility to different treatments, in particular psychological treatments, as waiting lists could be long.

It is noteworthy that the number of visits varies considerably among patients, with the findings indicating that a small proportion exhibit high levels of healthcare utilization. Exploring potential explanations for this elevated use—beyond symptom burden—is essential. Thus, the finding that patients with co-morbid psychiatric diagnoses consume more healthcare services—aligns with previous studies. Thus, morbidity burden has been shown to differ between patients with high stress levels [31] and a higher morbidity burden is shown to increase healthcare consumption [32]. Furthermore, in a systematic review on health service utilisation, consistent evidence is reported that need factors such as comorbidity and severity of symptoms are associated with health service utilisation among patients with common mental disorders [33].

When searching for additional explanation to the observed large variation in care, it is important to consider the structural perspective within the PCC organization, including communication between different caregivers and the organization of care processes. Thus, a newly published study from our research group shows that the care processes for patients with ED was perceived by psychologists in primary care to be unstructured, ineffective and not congruent with the needs of the patients [34]. Furthermore, many of the treatments being offered to patients with ED take place in rehabilitation centres that are not always clearly connected to the PCC, thus hampering the collaborative care process needed for these patients [35]. As a result of the lack of a holistic overview, patients are compelled to manage and coordinate their own care. Consequently, some of the variation of treatments being provided may be linked to the fact that patients with the most resources and/or the strongest perceived need, regardless of symptoms or comorbidity, receive the most care. Different resources related to staffing and professions with variable interest and knowledge of the ED diagnosis and guidelines could also explain some difference between, for example, different municipalities regarding number of care contacts and treatments being offered to the patients.

The treatments included in the analysis are those listed in the regional medical guidelines developed for this patient group, which are primarily based on clinical experience and on evidence for treating different symptoms presented by patients with ED. However, there is lack of evidence for a specific treatment for ED as a syndrome [18,20], and this could contribute to explaining the variation of treatment initiatives being provided to these patients. It could be speculated that the treatments offered are indeed individualized, suggesting that the variation in treatments reflects the principle of equitable care. Therefore, the fact that different patients receive different treatments does not necessarily disadvantage them, but rather reflects their specific needs. However, this is somewhat contradicted by the observation that female patients are generally allocated more visits and more extensive treatments, despite previous research showing little difference in the burden of symptoms between male and female patients seeking care for stress-related exhaustion [7,9,10]. One possible explanation is that male patients with ED may refrain from participating in various treatment options. This aligns with previous studies showing that men with mental health issues tend to underutilize mental health services compared to women [29,32,36,37]. The aspect of underutilization warrants further studies, since it could also be the other way around, namely that female patients overutilize care that perhaps is not so effective for their symptoms. It has been shown that patient characteristics affect the patient-doctor relationship [38], including when different treatment options are provided to female and male patients [36,39]. In line with this, it has also been shown that women were more likely to receive more accessible, less expensive primary care, while men were more likely to receive specialist inpatient care [37].

The differences across age groups found in our study align with previous research, indicating that older age groups are less likely to receive psychological therapy [36,40] which is in line with previous studies indicating that younger patients are more likely than older patients to prefer psychological treatment [41]. Older patients were more likely to receive guided physical activity, possibly due to central barriers that limit their ability to engage independently—such as pain, discomfort, fear of falling, or difficulties with access associated with aging. As a result, more older adults may have been assessed as needing support to enhance their capability, opportunity, and motivation to participate in physical activity [42]. Another explanation could be that younger patients are generally more active and, therefore, leading healthcare professionals to perceive less need for recommending guided physical activity.

The governance of the PCC did not influence the pattern regarding the number of care contacts provided. This suggests that the needs of patients are more important than the governance of the PCC when providing care for stress-related mental health problems and that regulations and steering principles of different PCCs do not seem to affect which treatments are provided.

Notable from the results is that MMR treatment is not commonly offered to patients in Region Västra Götaland. MMR is a treatment that at the timepoint of this study was more commonly used in clinical settings in other regions in Sweden [18,19]. The effectiveness and cost-effectiveness of MMR for patients with a mental health diagnosis is unclear and thus MMR is a treatment that could plausibly be seen as unnecessary and hence contributes to wasting resources [43].

Strengths and weaknesses

The Vega register is a large-scale database that includes most primary care healthcare consumption of the inhabitants in Region Västra Götaland. However, care utilisation of occupational healthcare is not included, which is important for the treatment of ED [44]. Another limitation that should be mentioned is the accuracy and completeness of the data recorded in Vega. Thus, we are aware that errors do occur in these records, such as incorrect diagnoses and the fact that all professionals do not register all treatment interventions. Professionals other than physicians are those that primarily use the KVÅ codes that were included in this study. Physicians perform diagnostic procedures, issue sickness certificates and prescribe medication (the latter two is not recorded in the Vega database), but do not usually conduct the treatments that were analysed. Consequently, the data presented cannot be expected to include all treatment interventions offered to patients with ED.

Another weakness to raise is that we divided the patients into three age groups, and we might have missed some age-related differences that could have been captured with a more detailed division. However, we believe that such detailed analyses are better suited for future studies with specific aims related to age differences.

Conclusions

The observed differences in gender and age do not appear to be justified by symptom burden, suggesting that various patient groups may be receiving different treatments. The substantial variation in treatment approaches for patients with ED may partly stem from a lack of evidence on optimal interventions, as well as the influence of caregiver preferences on clinical decision-making. Further research is needed to determine whether this variation in healthcare utilization is truly individualized and based on patient needs, or if it reflects inequities in the care provided. Considering the long duration of sick leave and the heavy burden of disease associated with ED, it is crucial to address the implications of this variation to ensure optimal healthcare for all patients with ED.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Funding Statement

This study was financed by grants from the Swedish state under the agreement between the Swedish government and the county councils, the ALF agreement (ALFGBG-726011).

Ethical approval

This study was approved by the Swedish National Authority for ethical approval in January 2019 (Drn 1064 − 118/2019 − 00612).

Disclosure statement

The authors report that there are no competing interests to declare.