Introduction
Childhood maltreatment (CM) significantly increases the risk of developing post-traumatic stress disorder (PTSD) and other mental health issues. CM is defined as any act of commission or omission by a parent or caregiver that results in intended or unintended harm, potential for harm, or threat of harm to a child [Reference Ehring, Welboren, Morina, Wicherts, Freitag and Emmelkamp1–Reference Leeb, Paulozzzi, Melanson, Simon and Arias4]. CM encompasses various forms, including physical abuse, sexual abuse, psychological or emotional abuse, neglect, and witnessing intimate partner violence [Reference Gilbert, Widom, Browne, Fergusson, Webb and Janson2]. In Europe, the estimated prevalence of CM is approximately 23% for physical abuse, 10% for sexual abuse, and 30% for psychological or emotional abuse [Reference Sethi, Bellis, Hughes, Gilbert, Mitis and Galea5]. A meta-analysis found the global prevalence of neglect to be around 18% [Reference Stoltenborgh, van Ijzendoorn, Euser and Bakermans-Kranenburg6]. The prevalence of witnessing intimate-partner violence during childhood was reported to range between 8 and 24% in surveys from the USA and Sweden [Reference Gilbert, Widom, Browne, Fergusson, Webb and Janson2, Reference Dong, Anda, Felitti, Dube, Williamson, Thompson, Loo and Giles7, Reference Janson, Långberg and Svensson8].
CM is associated with PTSD in adolescence and adulthood, particularly in cases of physical or sexual abuse and neglect [Reference Gilbert, Widom, Browne, Fergusson, Webb and Janson2, Reference Werner, McCutcheon, Challa, Agrawal, Lynskey, Conroy, Statham, Madden, Henders, Todorov, Heath, Degenhardt, Martin, Bucholz and Nelson3]. Symptoms of PTSD related to CM (PTSD-CM) often include the repeated occurrence of intrusive thoughts and memories, sleep disturbances, and feelings of detachment or numbness [Reference Gilbert, Widom, Browne, Fergusson, Webb and Janson2]. Additionally, individuals with PTSD-CM often exhibit high levels of complex symptomatology beyond PTSD, such as difficulties with emotion regulation difficulties, interpersonal issues, impulsive and self-destructive behaviors, and high levels of dissociation [Reference Briere, Kaltman and Green9–Reference Cloitre, Stolbach, Herman, van der Kolk, Pynoos, Wang and Petkova11].
Individuals with PTSD are typically treated with trauma-focused psychotherapies, including trauma-focused cognitive-behavioral therapy, eye-movement desensitization and reprocessing, and prolonged exposure therapy, as well as non-trauma-focused cognitive-behavioral therapies that address trauma-related thoughts, emotions, and anger management [Reference Bisson, Roberts, Andrew, Cooper and Lewis12–16]. For PTSD-CM particularly individual trauma-focused psychotherapies are generally recommended as first-line treatment with adaptions to meet the specific needs of individuals with PTSD-CM, such as phase-based approaches like Skills Training in Affect and Interpersonal Regulation/Narrative Therapy (STAIR/NT) [Reference Ehring, Welboren, Morina, Wicherts, Freitag and Emmelkamp1, Reference Bisson, Roberts, Andrew, Cooper and Lewis12, Reference Cloitre, Koenen, Cohen and Han17, Reference Forbes, Creamer, Bisson, Cohen, Crow, Foa, Friedman, Keane, Kudler and Ursano18].
Despite those recommended and presumably cost-effective therapies, it is known that individuals with PTSD cause a high economic burden, particularly those with more severe symptoms [Reference von der Warth, Dams, Grochtdreis and König19]. In 2010, the total annual costs of PTSD in Europe was approximately €8.4 billion, affecting around 7.7 million people [Reference Olesen, Gustavsson, Svensson, Wittchen and Jonsson20]. A recent systematic review of economic evaluations and cost analyses found that annual excess health care costs of PTSD, that is health care cost differences between individuals with PTSD and those without, ranged from about €460 to €17,400, and annual excess costs of absenteeism of €4500 per person [Reference von der Warth, Dams, Grochtdreis and König19]. High excess health care costs of PTSD were associated with greater utilization of both outpatient somatic, psychiatric, psychosomatic, psychological as well as nonmedical services among individuals with PTSD [Reference Chan, Medicine, Air and McFarlane21–Reference O’Donnell, Creamer, Elliott and Atkin24]. However, analyses of excess costs related to somatic, psychiatric, and psychosomatic hospital utilization were inconclusive.
To our knowledge, excess costs of PTSD have rarely been analyzed so far with studies conducted only in Australia, the USA, Canada, and the Netherlands [Reference Chan, Medicine, Air and McFarlane21–Reference Eekhout, Geuze and Vermetten25]. There are no existing studies on the excess costs of PTSD in Germany or specifically on the excess costs of PTSD-CM. Previous excess cost analyses primarily focused on victims of motorcycle accidents [Reference Chan, Medicine, Air and McFarlane21, Reference O’Donnell, Creamer, Elliott and Atkin24] and veterans with PTSD [Reference Chan, Cheadle, Reiber, Unutzer and Chaney22, Reference Lamoureux-Lamarche, Vasiliadis, Preville and Berbiche23, Reference Eekhout, Geuze and Vermetten25]. Therefore, this study aimed to compare health care service utilization, associated costs of health care, and absenteeism in individuals with PTSD-CM to those in the general population in Germany, ultimately determining the excess health care and absenteeism costs per person associated with PTSD-CM from a societal perspective.
Methods
Sample of individuals with PTSD-CM
Data on individuals with PTSD-CM were obtained from the baseline sample of a multicenter randomized controlled trial (Enhancing treatment and understanding of PTSD-CM [ENHANCE]; trial registration number: DRKS 00021142) [Reference Leichsenring, Steinert, Beutel, Feix, Gündel, Hermann, Karabatsiakis, Knaevelsrud, König, Kolassa, Kruse, Niemeyer, Nöske, Palmer, Peters, Reese, Reuss, Salzer, Schade-Brittinger, Schuster, Stark, Weidner, von Wietersheim, Witthöft, Wöller and Hoyer26]. This study aimed to compare methods of STAIR/NT and of trauma-focused psychodynamic therapy against a minimal attention waiting list for PTSD-CM. In Germany, a significant proportion of care for mentally ill people is provided by psychosomatic-psychotherapeutic clinics, clinics with a focus on specialized multimodal psychotherapeutic treatment, as well as medical and psychological psychotherapists in the outpatient sector. Thus, the study was conducted in university psychosomatic-psychotherapeutic outpatient clinics and university psychological institutes in Giessen, Dresden, Berlin, Mainz, and Ulm, Germany.
Participants were included if they had a primary diagnosis of PTSD-CM, experienced sexual or physical abuse by a caregiver or authority figure before the age of 18, and were aged 18–65 years. Exclusion criteria included current psychotic disorders, ongoing maltreatment, acute suicidality requiring emergency care or hospitalization within the past three months, substance dependence not in remission for at least three months, borderline personality disorder, dissociative identity disorder, organic mental disorder, severe medical conditions incompatible with psychotherapy, newly applied pharmacotherapy and concurrent psychotherapy.
The ethics committee of the Faculty of Medicine at Justus Liebig University Giessen granted ethical approval for the ENHANCE trial (AZ 168/19). A total of n = 361 persons diagnosed with PTSD-CM were included in the randomized controlled trial from August 2020 to May 2023. All participants were required to provide written informed consent prior to study participation. A detailed description of the ENHANCE trial can be found elsewhere [Reference Leichsenring, Steinert, Beutel, Feix, Gündel, Hermann, Karabatsiakis, Knaevelsrud, König, Kolassa, Kruse, Niemeyer, Nöske, Palmer, Peters, Reese, Reuss, Salzer, Schade-Brittinger, Schuster, Stark, Weidner, von Wietersheim, Witthöft, Wöller and Hoyer26].
Sample of individuals without PTSD
Data on individuals without PTSD were obtained from a representative telephone survey of the German adult general population conducted in March and April 2014 [Reference Grupp, König and Konnopka27]. Self-reported diagnoses were used to identify potential PTSD cases, with the question “Have you ever been diagnosed by a doctor with PTSD?.” Of the total sample from the general population (n = 5005), n = 245 persons were indicated with a PTSD diagnosis and were excluded, resulting in a final sample of n = 4760 persons without PTSD. A detailed description of the representative telephone survey of the German adult population can be found elsewhere [Reference Grupp, König and Konnopka27].
Health care service utilization and other measures
Health care service utilization and absenteeism from work of individuals with PTSD-CM and those without PTSD were assessed retrospectively over six months using an adapted self-report version of the German Client Socio-Demographic and Service Receipt Inventory (CSSRI) [Reference Herdman, Gudex, Lloyd, Janssen, Kind, Parkin, Bonsel and Badia28]. Participants provided information on their utilization of psychiatric and psychosomatic hospital or daycare, somatic hospital, daycare or rehabilitation, outpatient psychiatric, psychosomatic and psychological services, outpatient somatic medical services (e.g. general practitioner, orthopedist, dentist), and outpatient nonmedical services (e.g. occupational therapist, physiotherapist).
In both samples, participants provided information on their sex, age, marital status, educational attainment, professional training, employment status, health insurance, and the number of (underage) persons living in their household. In the sample of individuals without PTSD, participants were asked about the lifetime prevalence of various diseases, including lung diseases, metabolic diseases, diabetes, and cardiovascular conditions. Since data on comorbid chronic diseases were unavailable for the sample of individuals with PTSD-CM, prevalence estimates were derived from medication use data based on the World Health Organization’s Anatomical Therapeutic Chemical (WHO-ATC) classification [Reference Huber, Szucs, Rapold and Reich29].
For individuals with PTSD-CM, PTSD severity was assessed using the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) [Reference Weathers, Blake, Schnurr, Kaloupek, Marx and Keane30], a structured 30-item interview, evaluating past-month symptom severity on a five-point scale ranging from absent to extreme/incapacitating [Reference Marx, Lee, Norman, Bovin, Sloan, Weathers, Keane and Schnurr31, Reference Weathers, Bovin, Lee, Sloan, Schnurr, Kaloupek, Keane and Marx32].
Calculation of health care costs
Costs associated with health care service utilization were calculated by evaluating their quantities with standardized unit costs for the German health care system [Reference Bock, Brettschneider, Seidl, Bowles, Holle, Greiner and König33–Reference Muntendorf, Brettschneider, Konnopka and König35]. Informal care hours were valuated with the gross hourly labor costs of persons in the social care sector, sourced from the Federal Statistical Office of Germany’s gross labor cost database [36]. Days absent from work were evaluated with the gross hourly labor costs (including non-wage benefits) of persons in the manufacturing and services sectors, assuming an average eight-hour working day.
Total costs were assessed from a societal perspective, encompassing health care and absenteeism costs. All unit costs and hourly labor rates were inflated to 2022 price levels using the German consumer price index [37]. A detailed list of unit costs and hourly labor costs can be found in Supplementary Table S1.
Statistical analysis
Missing data in the samples of individuals with PTSD-CM and individuals without PTSD ranged from 0.02 to 0.80% across the 49 included variables, with 369 (0.14%) of a total 262,934 records being incomplete among n = 27 (7.48%) and n = 136 (2.72%) individuals, respectively. To enhance the accuracy and statistical power of the analyses, missing data were imputed under the assumption of missing at random using multiple imputations by chained equations, with predictive mean matching and m = 20 imputations [38].
The data sets of individuals with PTSD-CM and individuals without PTSD were balanced with regard to sociodemographic characteristics using entropy balancing [Reference Hainmueller39]. The entropy balancing model included the covariates sex, age, marital status, educational attainment, professional training, employment status, health insurance, and the number of (underage) persons in the household. Furthermore, (comorbid) chronic diseases were added as dummy-coded covariates. The means, variances, and skewnesses of the covariates were balanced between the two data sets. The sociodemographic characteristics of the samples of individuals with PTSD-CM and individuals without PTSD before balancing are presented in Supplementary Table S2.
Health care costs of individuals with PTSD-CM and those without PTSD were analyzed using two-part models. The first part of the models was a logit specification to account for potential substantial zero costs, while the second part was a generalized linear model with a gamma family and log-link function to account for the skewed cost distributions. The models incorporated the entropy balancing weights to adjust for differences in sociodemographic characteristics. Marginal effects between individuals with PTSD-CM and individuals without PTSD were estimated, representing the excess health care costs of PTSD-CM.
All data analyses were conducted using Stata/MP 18.0 (StataCorp, TX, USA). Multiple imputation was applied using Stata’s ‘mi’ package, entropy balancing was applied using the ‘ebalance’ package [Reference Hainmueller and Xu40] and two-part models were computed with Stata’s ‘tpm’ package [Reference Belotti, Deb, Manning and Norton41]. All statistical tests were two-sided, with a significance level set at p < 0.05.
Additional analyses
A subgroup analysis was conducted for individuals with mild to moderate PTSD symptoms and those with severe to extreme PTSD symptoms. The median CAPS-5 total score for the sample of individuals with PTSD-CM was used to differentiate between mild to moderate symptoms (CAPS-5 total score < 34) and severe to extreme symptoms (CAPS-5 total score ≥ 34). The data sets of individuals with mild to moderate PTSD symptoms and individuals with severe to extreme PTSD symptoms and individuals without PTSD were each balanced using entropy balancing for sociodemographic characteristics. Health care costs of individuals with mild to moderate PTSD symptoms, individuals with severe to extreme PTSD symptoms, and individuals without PTSD were analyzed using two-part models incorporating the respective entropy balancing weights.
Additionally, a further analysis explored potential determinants of total costs (including absenteeism costs) and total health care costs among individuals with PTSD-CM. Generalized linear models with gamma family and log-link function were used to examine these costs, with the covariates CAPS-5 total score, sex, age, marital status, educational attainment, professional training, employment status, health insurance, comorbid chronic diseases, and number of comorbid mental and behavioral disorders included in the models.
Results
Sample characteristics
The sociodemographic characteristics of the samples of individuals with PTSD-CM and individuals without PTSD after balancing are presented in Table 1. The average age of the samples was 39 years. Most participants were female (80%), single (62%), and had an academic secondary school qualification (62%). In terms of professional training, 37% had completed vocational training and 33% had a university degree. Approximately 30% were employed full-time, 22% part-time, and 26% were not in employment. The prevalence of (comorbid) chronic diseases was 7% for lung diseases, 22% for metabolic diseases, 3% for diabetes mellitus, and 11% for cardiovascular diseases.
Sociodemographic characteristics of the samples of individuals with post-traumatic stress disorder related to child maltreatment and individuals from the general population without PTSD after balancinga
SE: standard error, PTSD: post-traumatic stress disorder, PTSD-CM: post-traumatic stress disorder related to child maltreatment.
a The entropy balancing-model included the covariates age, sex, marital status, educational attainment, professional training, employment status and number of (underage) persons in household.
b ‘Separated’, ‘divorced’ and ‘widowed’ are not shown.
c ‘No school-leaving qualification’, ‘special-needs school’, and ‘still a pupil’ are not shown
d ‘Not applicable/not specified’ is not shown.
e ‘Other health insurance’ and ‘no health insurance’ are not shown.
Excess health care costs and costs of absenteeism
The average six-month total health care costs in individuals with PTSD-CM were €6131, compared to €1569 for those without PTSD (Table 2). This results in total excess health care costs associated with PTSD-CM of €4562 per person (95% CI: €3182 to €5942; p < 0.001). The average six-month costs of absenteeism in individuals with PTSD-CM were €4846, compared to €646 for those without PTSD, leading to excess absenteeism costs associated with PTSD-CM of €4200 per person. Overall, the six-month total excess costs associated with PTSD-CM amounted to €8762 per person (95% CI: €6736 to €10,788; p < 0.001). The average six-month total costs including absenteeism costs, for individuals with PTSD-CM were €10,977, compared to €2215 person in those without PTSD.
Average day/contacts, health care costs, and excess health care costs of post-traumatic stress disorder related to child maltreatment (six months, in Euro 2022)
SE: standard error, CI: confidence interval, PTSD: post-traumatic stress disorder, PTSD-CM: post-traumatic stress disorder related to child maltreatment.
* p < 0.05, ** p < 0.01, *** p < 0.001.
a Average days are shown for hospital/daycare/rehabilitation, average contacts are shown for outpatient medical, psychological and nonmedical services, average hours are shown for nursing care and absenteeism.
b Excess health care costs were calculated by a two-part model with logit specification for the first part and a generalized linear model with gamma family and log link function for the second part.
Individuals with PTSD-CM incurred significantly higher costs in several categories: hospital/daycare/rehabilitation (+€3267; 95% CI: €2167 to €4367; p < 0.001), outpatient medical and psychological services (+€395; 95% CI: €293 to €498; p < 0.001), and outpatient nonmedical services (+€79; 95% CI: €32 to €127; p = 0.001). Notably, individuals with PTSD-CM spent approximately 28 times more days in psychiatric and psychosomatic hospitals than those without PTSD (5.37 days versus 0.19 days). Additionally, they utilized outpatient psychiatric, psychosomatic, and psychological services about six times more frequently (4.21 contacts versus 0.68 contacts). In terms of nursing care, those with PTSD-CM had significantly higher costs for informal care (+€784; 95% CI: €219 to €1349; p = 0.007), spending roughly three times more hours on informal care compared to individuals without PTSD (37.08 hours versus 13.56 hours).
Additional analyses
The total excess health care costs associated with PTSD-CM for individuals with mild to moderate PTSD symptoms amounted to €2663 per person (95% CI: €680 to €3996; p < 0.001), while for individuals with severe to extreme PTSD symptoms, the total excess health care costs amounted to €6369 per person (95% CI: €4057 to €8482; p < 0.001; Table 3). The excess costs of absenteeism associated with PTSD-CM for individuals with mild to moderate PTSD symptoms were €3308 per person (95% CI: €1911 to €4705; p < 0.001), compared to €5042 per person (95% CI: €3400 to €6685; p < 0.001) for individuals with severe to extreme symptoms. Consequently, the six-month total excess costs associated with PTSD-CM were €5971 per person (95% CI €3813 to €8128; p < 0.001) for those with mild to moderate PTSD symptoms and €11,312 per person (95% CI €8081 to €14,542; p < 0.001) for those with severe to extreme symptoms. The sociodemographic characteristics of the samples of individuals with mild to moderate PTSD symptoms and individuals with severe to extreme PTSD symptoms are presented in Supplementary Table S3. The samples differed statistically significantly with regard to marital status, educational attainment, and employment status.
Excess health care costs of post-traumatic stress disorder related to child maltreatment (six months, in Euro 2022): subgroup analysis by post-traumatic stress disorder symptom severity
SE: standard error, CI: confidence interval, PTSD: post-traumatic stress disorder, PTSD-CM: post-traumatic stress disorder related to child maltreatment.
* p < 0.05, ** p < 0.01, *** p < 0.001.
a CAPS-5 total score ≥ 34.
b CAPS-5 total score < 34.
c Excess health care costs were calculated by a two-part model with logit specification for the first part and a generalized linear model with gamma family and log link function for the second part.
Among individuals with PTSD-CM, total health care costs (+€336; 95% CI €78 to €594; p = 0.010) and the total costs including absenteeism costs (+€419; 95% CI: €101 to €737; p = 0.011) were significantly associated with the CAPS-5 total score. The total health care costs and the total costs including costs of absenteeism from work were not associated with age. The generalized linear models of total health care costs and total costs, PTSD severity, and selected sociodemographic characteristics in patients with PTSD-CM are shown in Supplementary Table S4.
Discussion
This study aimed to determine the excess costs associated with PTSD-CM in Germany. The six-month total excess costs associated with PTSD-CM amounted to €8762 per person, with the primary contributors being absenteeism (€4200) and hospitalization (€3267). Among all individuals with PTSD-CM, those with severe to extreme PTSD symptoms incurred nearly twice the excess costs compared to those with mild to moderate symptoms (€11,312 vs. €5971).
Compared to a similar analysis of annual excess costs of PTSD conducted in the Netherlands, this difference in six-month total excess costs associated with PTSD-CM between individuals with severe to extreme PTSD symptoms and those with mild to moderate symptoms was notably higher. In the Dutch study, the difference in excess costs associated with PTSD between individuals with more severe PTSD symptoms (above the 95th percentile) and those with less severe PTSD symptoms (below the 9th percentile) was approximately €460 [Reference Eekhout, Geuze and Vermetten25]. However, as the Dutch sample consisted of veterans and the PTSD severity assessed using the Self Report Inventory for PTSD, direct comparability between the two studies is limited.
The six-month excess absenteeism costs associated with PTSD-CM in this study were higher than those reported in another analysis of annual excess costs of PTSD, which also accounted for absenteeism costs (€4540) [Reference von der Warth, Dams, Grochtdreis and König19, Reference Chan, Medicine, Air and McFarlane21]. However, comparability is limited since the referenced study was conducted in Australia in 2003 and focused on victims of traffic accidents [Reference Chan, Medicine, Air and McFarlane21].
Costs for hospitalization in psychiatric, psychosomatic, and somatic facilities, outpatient psychiatric, psychosomatic and psychological, somatic medical, and nonmedical outpatient services were significantly higher among individuals with PTSD-CM compared to those without PTSD. In contrast, a systematic review indicated only non-significantly higher costs for outpatient medical, psychological, and nonmedical services between individuals with and without PTSD [Reference von der Warth, Dams, Grochtdreis and König19, Reference Chan, Medicine, Air and McFarlane21–Reference Lamoureux-Lamarche, Vasiliadis, Preville and Berbiche23]. Regarding hospitalization costs, the review yielded inconclusive results, with two identified studies reporting positive excess costs of hospitalization [Reference Chan, Medicine, Air and McFarlane21, Reference O’Donnell, Creamer, Elliott and Atkin24] and two others reporting negative excess costs [Reference von der Warth, Dams, Grochtdreis and König19, Reference Chan, Cheadle, Reiber, Unutzer and Chaney22, Reference Lamoureux-Lamarche, Vasiliadis, Preville and Berbiche23, Reference Eekhout, Geuze and Vermetten25]. Notably, only two studies [Reference Chan, Medicine, Air and McFarlane21, Reference Chan, Cheadle, Reiber, Unutzer and Chaney22] found significant differences in costs between individuals with and without PTSD.
The current study identified significantly higher costs for informal care among individuals with PTSD-CM compared to those without PTSD. A cost-of-illness study reported annual informal care costs of approximately €4710 for war-affected adults with PTSD in Germany, which exceeds the six-month informal care costs of €1236 identified in this study [Reference von der Warth, Dams, Grochtdreis and König19, Reference Priebe, Matanov, Jankovic Gavrilovic, McCrone, Ljubotina, Knezevic, Kucukalic, Franciskovic and Schutzwohl42]. These elevated informal care costs suggest a greater need for assistance from family members, friends, and acquaintances due to health issues faced by individuals with PTSD-CM, particularly for tasks typically managed independently. However, the specific activities involved in informal care, such as emotional support or assistance with everyday tasks, remain unclear as do the underlying health issues prompting this need for help, such as social isolation or impaired functioning.
Generalizability and policy implications
The excess costs associated with PTSD-CM identified in this study may be merely applicable to individuals who sought treatment in a university psychiatric, psychosomatic, and psychological outpatient clinic or a university psychological institute in Germany. However, it is important to note that routine care for individuals with PTSD occurs in outpatient settings outside of hospitals [Reference Schäfer, Gast, Hofmann, Knaevelsrud, Lampe, Liebermann, Lotzin, Maercker, Rosner and Wöller43].
To potentially reduce these excess costs associated with PTSD-CM in the German health care system, cost-effectiveness should be especially explored for hospital care which has been the primary driver of total excess health care costs. As hospitalized individuals with PTSD-CM are predominantly severely and not often chronically ill, adequate inpatient and outpatient treatment is difficult. Multimodal specialized inpatient and outpatient treatment for patients with PTSD-CM should be strived for. Also stepped care depending on the patient’s symptom severity with the option of preceding trauma-focused outpatient medical and psychological psychotherapy should be targeted as an alternative treatment option. However, in order to be able to refer patients with PTSD-CM to outpatient medical and psychological psychotherapy, it is necessary to have a sufficient number of qualified psychotherapists available who are also willing to treat individuals with severe PSTD. This could subsequently contribute to reduced admissions to hospital care.
Additionally, understanding and addressing the underlying factors contributing to work absenteeism among individuals with PTSD-CM is crucial. It should also be acknowledged that much of the caregiving of individuals with PTSD-CM is provided by family members, friends, and acquaintances at no additional cost to the health care system. Finally, health care services and policies should specifically target those individuals with severe to extreme PTSD symptoms, as their hospitalization, informal care, and absenteeism costs are notably high.
Strengths and limitations
One significant strength of this analysis is extensive data on health care service utilization and work absenteeism for a large cohort of individuals with PTSD-CM in Germany. Additionally, the adjustment for sociodemographic differences between individuals with PTSD-CM and those without PTSD from the general population enabled the isolation of health care and absenteeism costs specifically attributable to PTSD-CM. It is worth mentioning that individuals with PTSD-CM in Germany differed with regard to sociodemographic characteristics compared to those in the general population in Germany. For example, there were differences in health insurance status, with about only 2% of all individuals with PTSD-CM being privately insured, whereby about 10% of all individuals from the German general population were privately insured in 2022 [44]. This difference could be explained, at least in part, by the younger age of those individuals with PTSD-CM and by an association between posttraumatic stress and socioeconomic disadvantage [Reference Lowe, Galea, Uddin and Koenen45].
However, this study has further limitations. First, data on medication use, medical aids, and presentism were not available for the general population sample without PTSD, which may have led to an underestimation of the total excess costs associated with PTSD-CM. Second, health care service utilization was assessed using an adapted self-report version of the German CSSRI, which does not cover specific medical and nonmedical outpatient services for people with mental illnesses, such as psychiatric counseling, psychosocial care, assisted living, and occupational integration. Third, the recruitment of individuals with PTSD-CM was supported by the application of additional measures, such as information about the study in mass media, in psychiatric, psychosomatic, and psychological outpatient clinics and practices, which may have introduced a potential selection bias. Fourth, the data for the general population was collected through a representative telephone survey conducted in the year 2014, which may limit comparability regarding health care service utilization and absenteeism due to significant differences in time periods and data collection methods (telephone survey versus patient interviews). Nevertheless, health care service utilization and absenteeism were evaluated using standardized unit costs for the German health care system [Reference Bock, Brettschneider, Seidl, Bowles, Holle, Greiner and König33–Reference Muntendorf, Brettschneider, Konnopka and König35]. and gross hourly wages from the Federal Statistical Office’s gross labor cost database [36], which were inflated to 2022 price levels using the German consumer price index [37], ensuring an increased comparability. Lastly, the data on individuals with PTSD-CM was collected during the COVID-19 pandemic, which may affect health care service utilization and absenteeism patterns compared to periods outside the pandemic.
Conclusion
The six-month excess health care and absenteeism costs associated with PTSD-CM were substantial, with absenteeism accounting for approximately half of the total excess costs. Notably, individuals with severe to extreme PTSD symptoms faced more than twice the total excess costs compared to those with mild to moderate PTSD. Further research is essential in order to explore the cost-effectiveness of hospital care of individuals with PTSD-CM, as well as to identify and address the underlying factors contributing to work absenteeism in this population.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1192/j.eurpsy.2025.6.
Data availability statement
The data sets generated and/or analyzed during the current study are not publicly available due to ethical and confidentiality concerns but are available from the corresponding author upon reasonable request.
Acknowledgments
We extend our gratitude to our study therapists, participants, and especially to Michael Wittenberg for their valuable contributions to the ENHANCE trial and Alexander Karabatsiakis for his helpful comments on an earlier version of the manuscript.
Financial support
This study was funded by the German Federal Ministry of Education and Research (BMBF) under grant number 01KR1801A. The funding agency had no influence on study design, data collection, management, analysis, interpretation, writing, or publication process.
Competing interest
T.G., H.-H.K., F.L., M.E.B., L.F., H.G., A.H., M.H., C.K., I.-T.K., J.K., H.N., F.N., S.S., K.S.S., P.S., C.S., K.W., J.v.W., J.H. and J.D. declare none.
Open access
Comments
No Comments have been published for this article.