In this prospective multicenter study, we investigated the course of depression and anxiety during hematopoietic stem cell transplantation (HSCT) until 5 years after transplantation adjusting for medical information. Patients were consulted before HSCT (n=239), at 3 months (n=150), 12 months (n=102) and 5 years (n=45) after HSCT. Depression and anxiety were assessed with the Hospital Anxiety and Depression Scale (HADS). Detailed medical and demographic information was collected. Prevalence rates were compared with an age- and gender-matched control group drawn from a large representative sample (n=4110). The risk of depression before HSCT was lower for patients than for the control group (risk ratio (RR), 0.56; 95% confidence interval (CI), 0.39/0.81). Prevalence rates of depression increased from 12 to 30% until 5 years post HSCT. Anxiety rates were most frequently increased before HSCT (29%, RR, 1.31; 95% CI, 1.02/1.68) and then reached a stable level comparable to the background population (RR 0.83, 95% CI, 0.56/1.22). This study confirms the low levels of depression in the short term after HSCT and identifies depression as a long-term effect. Furthermore, it confirms previous results of heightened anxiety before HSCT. Surveillance of symptoms of anxiety during the short-term phase of HSCT and of depression during the following years is crucial.
Survival after allogeneic hematopoietic stem cell transplantation (HSCT) has improved from 37% in the period 1993–1997 to more than 50% between 2003 and 2007.1 Despite the potential cure of an otherwise terminal illness, allogeneic HSCT is accompanied by high levels of morbidity, mortality and immense psychosocial distress.2 This is illustrated by a significantly increased risk of psychiatric conditions, for example, depression and anxiety with prevalence rates following HSCT reaching from 5% to more than 40%.3 Across prospective studies of depression in HSCT patients, it has been shown that the highest risk of depression is just before transplantation procedures are carried out4 or during hospitalization shortly after the transplantation.5, 6, 7 Moreover, it has been indicated that depression and anxiety can become chronic years after the transplantation.8, 9 In a cross-sectional assessment of 662 HSCT patients (mean follow-up 7 years), a fourfold increased risk of depression was observed compared with an age- and gender-matched peer-nominated control sample.9 However, findings with regard to persistence of depression are mixed, as some studies reported that depression declined to pre-HSCT levels,6, 10, 11 some that depression reached lower levels4 or remained high following HSCT.12, 13 A severe limitation of these studies of depression and anxiety is the fact that they are often based on small sample sizes of less than 30 patients4, 7, 14 and short follow-up between 14 and 100 days post HSCT.5, 7, 13
In prospective studies with long-term follow-up including a baseline measurement before transplantation, investigating between 6115 and 319 HSCT patients10 or a first measurement at 1 year after HSCT including 415 HSCT patients,16 the level of depression and anxiety did not change significantly at 1–3, 4 or 5 years of follow-up.12 Furthermore, prevalence rates of depression and anxiety in a cross-sectional assessment with the mean follow-up of 3–5 years after HSCT were similar to rates of an age- and gender-matched population sample (N=52 000).15 One study12 used data from patients being transplanted between 1987 and 1992, and transplantation procedures have changed considerably since then.1 Moreover, none of these four studies controlled for use of psychotropic medication,10, 12, 15, 16 which reduces the prevalence of psychiatric symptoms.17 It is also remarkable that two studies10, 16 only included recurrence-free patients at follow-up, which may explain the reduced risk of depression and anxiety observed following HSCT. One may speculate that these conceptual and methodological limitations explain the mixed findings with regard to the risk of depression. Here we present data from a population-based prospective study, which investigates the prevalence of depression and anxiety following HSCT with up to 5 years of follow-up compared with an age- and gender-matched control group randomly sampled in the German population, adjusting for detailed medical information.
Materials and methods
Our research was part of a prospective multicenter trial investigating cognitive functioning in cancer patients treated with allogeneic HSCT,18 recruiting patients at six German University Medical Centers. Eligible patients were ⩾18 years of age, diagnosed with a hematologic disease and scheduled for allogeneic HSCT. Exclusion criteria were second allogeneic HSCT, permanent or demand-oriented use of psychoactive drugs or substance abuse, preceding intrathecal chemotherapy, preceding irradiation of central nervous system (CNS, except TBI), preexisting neurological disease, insufficient knowledge of German language, uncorrectable sight or hearing impairment (for example, tinnitus), and new onset of cerebral neurologic disease after HSCT. Patients were contacted personally or by phone and were informed about the study. If they fulfilled inclusion criteria and agreed to participate in the study, questionnaires were mailed and returned at the next scheduled appointment at the outpatient clinic. Patients gave informed consent before start of the study. Study approval was obtained from the local ethics committees or the institutional review boards responsible for the respective medical centers (File number of the main assessment of the medical association in Hamburg: 2389). All data including follow-up assessments were collected over a 7-year period from 2 June 2005 to 11 June 2012.
Follow-up. For the follow-up measurements, the patients were once more contacted personally or by phone and—upon agreement to further participate in the study—questionnaires were mailed to study participants, who returned them to the outpatient clinic at the following appointment.18 Patients were assessed before conditioning for HSCT (T0), 100±20 days (T1) after HSCT, which is usually the end of the immediate and close after care phase, 12±1 months after HSCT (T2) and 5 years after HSCT (T3). For the 5-year follow-up only patients at four of the original six involved centers were contacted. Of the 92 patients contacted, 24 had died and 23 declined to participate leaving 45 patients (66%) for 5 years’ follow-up (Figure 1).
Control group. The age- and gender-matched control groups were drawn from a representative population-based sample that consisted of 4110 women and men randomly sampled from the German population by the random-route technique, leading to a random selection of street, house, apartment and target person contacted in the household. These participants were sampled during two phases in 1998 (n=1993) and in 2009 (n=2417) with participation rates of 73% and 62%, respectively. The average age of the final sample was 50 years and 56% were women. Of the complete sample, 24% of men and 24% of women had depression and 18% of men and 23% of women had increased levels of anxiety. Information on history of diseases was not assessed. Data and detailed description of the sample have been published elsewhere.19
Sociodemographic information was obtained via patients’ self-report (see Tables 1 and 2). On the basis of medical record review, medical information including diagnosis, pretreatment for the current hematological disease before conditioning, myelotoxicity of pretreatment and conditioning regimen, transplantation type (PBSC transplantation (PBSCT) versus bone marrow transplantation) and characteristics of the transplant regarding match and kinship to the donor was obtained. Furthermore, it was assessed during follow-up whether patients suffered from acute or chronic GvHD, had a relapse or progression of the disease or regularly used psychotropic drugs.
Hospital anxiety and depression scale—German version (HADS). The HADS is a validated screening instrument for depression and anxiety in somatically ill patients20 and excludes symptoms that may arise from somatic aspects of illness (for example, insomnia, weight loss and fatigue). The measure consists of 14 items on a 4-point Likert scale (range 0–3) comprising a depression and anxiety subscale. For both subscales a total score is calculated (ranging from 0 to 21). In addition, three cutoff scores can be calculated: 0–7 (normal), 8–10 (moderate) and ⩾11 (severe). The German translation has been used across many studies and shows satisfying reliability with Cronbach’s alpha =0.81 for the depression scale and 0.80 for the anxiety scale.21 Furthermore, the HADS is sensitive toward intra-individual change.22 In the current study, Cronbach’s alpha ranged between 0.73 and 0.89 for depression and between 0.77 and 0.81 for anxiety across the different time points.
We computed prevalence rates and risk ratios (RRs) among patients and the general population. Depression and anxiety symptom scores higher than 7 (moderate to severe levels of depression and anxiety) were considered as clinically meaningful and were coded with 1. We performed χ2-tests to investigate differences in depression and anxiety rates over the four time points. To obtain values from the general population, random samples were drawn from the comparison population with five age- and gender-matched norm participants per patient for each assessment point. We drew separate samples for each assessment point because patients got older during the study period and younger patients were more likely to dropout. For random selection of the age- and gender-matched control groups the R-package MatchIt23 was used. RRs and 95% confidence intervals were computed for patients to experience moderate or severe levels of depression and anxiety.
In a second analysis, individual growth curve models with random intercepts were computed to assess change in self-reported symptoms of depression and anxiety within individuals as opposed to changes in sample means across time points. The four assessment points (level 1) were nested within individuals (level 2). Depression and anxiety were measured at four different time points (pre-HSCT, 3, 12 months and 5 years after HSCT). Covariates were included in the models in backward elimination procedure and were tested for explanation of intercept variation. The following demographic and medical variables were included: age, gender, being in a relationship, education (>9 years of education), employment status (employed, unemployed and non-working), diagnosis (AML versus other), years since diagnosis, treatment before conditioning, acute and chronic GVHD, myelotoxicity of pretreatment and conditioning regimen, donor characteristics (matched versus unmatched, related versus unrelated) and occurrence of relapse or progression. Covariates with significance level ⩽10% were retained in the model. In a last step, each covariate was included in an individual model as a cross-level interactive term in order to find individual factors that explain slope variation, that is, the strength of increase in depression or decrease in anxiety. In avoidance of inflation of Type 1 error when 11 individual tests for cross-level interaction are carried out, the level was Bonferroni-adjusted to alpha=0.45% (5%/11). Finally, because of the high rate of dropout throughout the study period, pattern mixture modeling was applied for the outcome variables depression and anxiety.
Of the 239 patients at baseline, three did not fill in the questionnaires, which is why they were excluded from all analyses. Table 1 illustrates that patients with ⩽9 years of education, being in pretreatment with chemotherapy, being pretreated with autologous HSCT, being diagnosed with multiple myeloma and with myelotoxic pretreatment were significantly more likely to dropout of the study.
Prevalence of depression and anxiety compared with a normative sample
Table 3 shows that the risk of depression before admission for HSCT and 1 year after HSCT was significantly lower than for the age- and gender-matched control group. Among HSCT patients, prevalence rates of depression did not change until 1 year post HSCT, when rates significantly increased from 12 to 30% at 5 years post HSCT (Figure 2). Contrary to our findings with regard to depression, pre-HSCT patients were significantly more likely to experience anxiety than the general population with significantly lower and comparable prevalence rates at following time points. Prevalence of anxiety significantly decreased from 29 to 19% at 3 months after HSCT and stayed at this level.
Individual course and predictors of depression and anxiety following HSCT
The growth curve models revealed that individual depression levels significantly increased, whereas anxiety levels significantly decreased over a time period of 5 years. Table 4 shows that higher age, shorter education and having received pretreatment were significantly associated with elevated depression levels. Women, patients with shorter education and patients who were pretreated before conditioning were more likely to experience anxiety. All cross-level interactive terms were nonsignificant, meaning that none of the covariates significantly explained the strength of increase in depression or anxiety over time. Pattern mixture modeling revealed that patients who dropped out at T2 had a significantly stronger predicted increase in depression than completers or dropouts at the other time points (β=0.99, P=0.04).
Contrary to previous reports, we observe a pattern of initial low prevalence of depression in HSCT patients, followed by an increase at 5 years' follow-up.4, 5, 6, 10, 11, 12, 13, 14, 15, 24, 25 Our findings of a high prevalence of anxiety before transplantation and later decrease are in line with findings of previous studies.4, 6
One may wonder why our findings contradict earlier results, and we suggest the following reasons for this discrepancy. Previous studies10, 11, 12, 14 applied depression measures that incorporate ratings of physical aspects of depression like loss of energy or decreased sexual interest such as the Beck Depression Inventory-II26 and the Symptom Checklist-90 (SCL-90)27. We applied the HADS, which intentionally does not assess physical symptoms of depression to avoid overlap with the burdened physical condition that accompanies medical situations such as HSCT.20 This may explain why the studies using Beck Depression Inventory-I10, 11 and SCL-90-R12, 14 observe higher depression scores in the time around the HSCT procedures accompanied by a high burden of somatic symptoms.28 Studies that used the HADS to measure depression after HSCT report depression scores comparable to our finding of a low prevalence of depression illustrated by scores between 2.6 and 4.4 just before HSCT was carried out4, 5, 6, 15, 29 in line with our finding of 4 as a sum score. Further, the HADS scores were lower than4 or comparable to15 prevalence rates in the general population. Taken together, results from previous studies and our study point to a relatively low risk of depression in the first year following HSCT compared with the risk in the background population.
Our finding of an increase in depression after 5 years has not been observed in previous studies among HSCT survivors.6, 10, 12, 15, 16 An exception is a cross-sectional study reporting elevated depression at a mean follow-up of 7 years after HSCT compared with a healthy comparison sample.9 One may wonder whether our finding may be explained by the fact that depressed patients were more likely to respond after 5 years than non-depressed patients. A thorough analysis of this question revealed that patients who responded to questionnaires at all follow-up times did not differ in depression scores reported at any time point throughout the study compared with patients who dropped out at any time point. Further analyses showed that patients who dropped out 1 year after HSCT were predicted to have an even higher risk of developing depression at 5-year follow-up than completers or dropouts at the other time points. Furthermore, in a longitudinal study with 137 patients at 10-year follow-up, it was shown that 12% of HSCT survivors had elevated depression levels compared with 8% among the non-transplant controls.8 However, in this study 65% of HSCT patients without depression were on antidepressant medication. Provided that drug treatment for these patients was further indicated and not unnecessarily continued, it can be assumed that actual depression rates among the HSCT survivors were higher than the 12% observed. In line with our results, a large nationwide and population-based Danish register-based study showed that clinical depression in virus and immune-related cancer survivors (including leukemia and non-Hodgkin lymphoma) was almost 40% significantly increased until 4 years after diagnosis for women (RR, 1.41; 95% CI, 1.18–1.68) and 70% in men at 10 years' follow-up (RR, 1.73; 95% CI, 1.10–2.71).30 In sum, we believe that these findings combined with our results point to an increased risk for depression in HSCT patients even years after the procedure has been carried out.
Only short education increased the risk of depression and anxiety.31 In line with previous studies we observe that gender and medical factors did not predict severity of depressive symptoms.10, 16, 32
Strengths and limitations
The prospective design including a large sample of a relatively homogenous group of HSCT patients, access to detailed medical information, the use of validated questionnaires and our comparison to a random age- and gender-matched population-based sample contribute to the advantages of the current study. However, among limitations that may influence the interpretation of the results is the considerable dropout rate during the course of the study potentially underestimating the prevalence of depression and anxiety in the patients, especially since part of the dropouts denied participation because the psychological burden was too high. We cannot exclude that the decreasing prevalence of anxiety is explained by selection bias favoring the participation of more healthy patients. However, this explanation does not comply with our observation of an increased risk of depression over time. In addition, as the present research was part of a large study examining cognitive functioning following HSCT, patients receiving CNS prophylaxis (intrathecal chemotherapy and/or irradiation of the CNS) were excluded from the study in order to reduce potential bias. In the present article, however, excluding this patient group may reduce generalizability of our findings. It is also of importance to note that the control population was sampled in but not followed over time as was the case group. We somewhat addressed this limitation by sampling control groups at each follow-up time to ensure age and gender matching over time. Furthermore, use of the self-report instrument HADS does not allow for inferences concerning the diagnosis of depression and anxiety, which requires an interview, for example, the structural clinical interview for DSM-5.33 Results of the current study reflect the proportion of patients with moderate to severe symptom levels that are consistent with a diagnosis of depression or anxiety.
This study confirms the harmful short- and long-term effects of a troublesome cancer diagnosis and specifically when treated with HSCT. Shortly before HSCT, almost 30% of patients experience increased levels of anxiety. Until 5 years after HSCT, the psychological landscape changes as 30% of patients now experience depression. In the light of possibilities for preventing both anxiety and depression becoming a major problem in cancer patients, we find that health professionals treating these patients should pay attention to these conditions throughout the cancer trajectory. Our recommendation is further supported by the link between depression and increased mortality in HSCT patients.34
Gooley TA, Chien JW, Pergam SA, Hingorani S, Sorror ML, Boeckh M et al. Reduced mortality after allogeneic hematopoietic-cell transplantation. N Engl J Med 2010; 363: 2091–2101.
Fife BL, Fausel CA. Hematopoietic dyscrasias and stem cell/bone marrow transplantation. In: Holland JC (ed). Psycho-oncology, 2nd edn, Oxford University Press: New York, NY, USA, 2010, pp 191–195..
Mosher CE, Redd WH, Rini CM, Burkhalter JE, DuHamel KN . Physical, psychological, and social sequelae following hematopoietic stem cell transplantation: a review of the literature. Psychooncology 2009; 18: 113–127.
Goetzmann L, Boehler A, Halter J, Klaghofer R, Muellhaupt B, Wagner-Huber R et al. Quality of life and psyschosocial situation before and after a lung, liver or an allogeneic bone marrow transplant: results from a prospective study. Swiss Med Wkly 2006; 136: 281–290.
Prieto JM, Atala J, Blanch J, Carreras E, Rovira M, Cirera E et al. Patient-rated emotional and physical functioning among hematologic cancer patients during hospitalization for stem-cell transplantation. Bone Marrow Transplant 2005; 35: 307–314.
Hjermstad MJ, Loge JH, Evensen SA, Kvaloy SO, Fayers PM, Kaasa S . The course of anxiety and depression during the first year after allogeneic or autologous stem cell transplantation. Bone Marrow Transplant 1999; 24: 1219–1228.
Illescas-Rico R, Amaya-Ayala F, Jiménez-López JL, Caballero-Méndez ME, González-Llaven J . Increased incidence of anxiety and depression during bone marrow transplantation. Arch Med Res 2002; 33: 144–147.
Syrjala KL, Langer SL, Abrams JR, Storer BE, Martin PJ . Late effects of hematopoietic cell transplantation among 10-year adult survivors compared with case-matched controls. J Clin Oncol 2005; 23: 6596–6606.
Andrykowski MA, Bishop MM, Hahn EA, Cella D, Beaumont JL, Brady MJ et al. Long-term health-related quality of life, growth, and spiritual well-being after hematopoietic stem-cell transplantation. J Clin Oncol 2005; 23: 599–608.
Syrjala KL, Langer SL, Abrams JR, Storer B, Sanders JE, Flowers, Mary ED et al. Recovery and long-term function after hematopoietic cell transplantation for leukemia or lymphoma. JAMA 2004; 291: 2335–2343.
Chang G, Orav J, Mcnamara TK, Tong M, Antin JH . Psychosocial function after hematopoietic stem cell transplantation. Psychosomatics 2005; 46: 34–40.
Broers S, Kaptein AA, Le Cessie S, Fibbe W, Hengeveld MW . Psychological functioning and quality of life following bone marrow transplantation: a 3-year follow-up study. J Psychosom Res 2000; 48: 11–21.
Rischer J, Scherwath A, Zander AR, Koch U, Schulz-Kindermann F . Sleep disturbances and emotional distress in the acute course of hematopoietic stem cell transplantation. Bone Marrow Transplant 2009; 44: 121–128.
Beglinger LJ, Duff K, Van Der Heiden, Sara, Moser DJ, Bayless JD, Paulsen JS et al. Neuropsychological and psychiatric functioning pre- and posthematopoietic stem cell transplantation in adult cancer patients: a preliminary study. J Int Neuropsychol Soc 2007; 13: 172–177.
Hjermstad MJ, Knobel H, Brinch L, Fayers PM, Loge JH, Holte H et al. A prospective study of health-related quality of life, fatigue, anxiety and depression 3-5 years after stem cell transplantation. Bone Marrow Transplant 2004; 34: 257–266.
Bush NE, Donaldson GW, Haberman MH, Dacanay R, Sullivan KM . Conditional and unconditional estimation of multidimensional quality of life after hematopoietic stem cell transplantation: a longitudinal follow-up of 415 patients. Biol Blood Marrow Transplant 2000; 6: 576–591.
Fournier JC, DeRubeis RJ, Hollon SD, Dimidjian S, Amsterdam JD, Shelton RC et al. Antidepressant drug effects and depression severity: a patient-level meta-analysis. JAMA 2010; 303: 47–53.
Scherwath A, Schirmer L, Kruse M, Ernst G, Eder M, Dinkel A et al. Cognitive functioning in allogeneic hematopoietic stem cell transplantation recipients and its medical correlates: a prospective multicenter study. Psychooncology 2013; 22: 1509–1516.
Hinz A, Brähler E . Normative values for the hospital anxiety and depression scale (HADS) in the general German population. J Psychosom Res 2011; 71: 74–78.
Zigmond AS, Snaith RP . The hospital anxiety and depression scale. Acta Psychiatr Scand 1983; 67: 361–370.
Hermann C, Buss U, Snaith RP . Hospital Anxiety and Depression Scale – Deutsche Version (HADS-D). Hans Huber: Bern, Germany, 1995.
Hinz A, Zweynert U, Kittel J, Igl W, Schwarz R . Veränderungsmessung mit der Hospital Anxiety and Depression Scale (HADS): Änderungssensitivität und Änderungsreliabilität. Psychother Psychos Med Psychol 2009; 59: 394–400.
Randolph JJ, Falbe K, Manuel AK, Balloun JL . A step-by-step guide to propensity score matching in R. Pract Assess Res Eval 2014; 19: 380–92.
Lee SJ, Loberiza FR, Antin JH, Kirkpatrick T, Prokop L, Alyea EP et al. Routine screening for psychosocial distress following hematopoietic stem cell transplantation. Bone Marrow Transplant 2005; 35: 77–83.
Syrjala KL, Chapko M, Vitaliano P, Cummings C, Sullivan K . Recovery after allogeneic marrow transplantation: prospective study of predictors of long-term physical and psychosocial functioning. Bone Marrow Transplant 1993; 11: 319–327.
Beck AT, Steer RA, Brown GK . Manual for the Beck Depression Inventory-II. Psychological Corporation: San Antonio, TX, USA, 1996.
Derogatis LR, Savitz KL . The SCL–90–R and Brief Symptom Inventory (BSI) in Primary Care. Lawrence Erlbaum Associates Publishers, 2000.
Anderson KO, Giralt SA, Mendoza TR, Brown JO, Neumann JL, Mobley GM et al. Symptom burden in patients undergoing autologous stem-cell transplantation. Bone Marrow Transplant 2007; 39: 759–766.
Trask PC, Paterson A, Riba M, Brines B, Griffith K, Parker P et al. Assessment of psychological distress in prospective bone marrow transplant patients. Bone Marrow Transplant 2002; 29: 917–925.
Dalton SO, Laursen TM, Ross L, Mortensen PB, Johansen C . Risk for hospitalization with depression after a cancer diagnosis: a nationwide, population-based study of cancer patients in Denmark From 1973 to 2003. J Clin Oncol 2009; 27: 1440–1445.
Balfour J, Kaplan G . Social class/socioeconomic factors. In: JC Holland (ed). Psycho-Oncology, 1998, pp 78–90.
Mehnert A, Brähler E, Faller H, Härter M, Keller M, Schulz H et al. Four-week prevalence of mental disorders in patients with cancer across major tumor entities. J Clin Oncol 2014; 32: 3540–3546.
First MB. Structured clinical interview for the DSM (SCID). In: Cautin RL, Lilienfeld SO (eds). The Encyclopedia of Clinical Psychology. John Wiley & Sons Inc: Hoboken, NJ, USA, 2014, pp 1–6..
Andrykowski MA . Depression and survival after hematopoietic stem cell transplantation: where do we go from here? J Clin Oncol 2005; 23: 5878–5880.
This work was supported by the José Carreras Leukämie-Stiftung (grants DJCLS R 04/29pf, DJCLS R 07/37pf and DJCLS R 10/38p).
AS, LS, AS, LS and MK performed the research. AS, AM, UK, FS-K and AZ designed the research study. AZ and NK contributed essential reagents or tools. KK and PE analyzed the data. All authors wrote the paper.
The authors declare no conflict of interest.
About this article
Cite this article
Kuba, K., Esser, P., Mehnert, A. et al. Depression and anxiety following hematopoietic stem cell transplantation: a prospective population-based study in Germany. Bone Marrow Transplant 52, 1651–1657 (2017). https://doi.org/10.1038/bmt.2017.190
Important factors associated with sick leave after allogeneic haematopoietic stem cell transplantation—a 1-year prospective study
Journal of Cancer Survivorship (2021)
Journal of Cancer Education (2021)
REBOUND “Trained to live again”: The practice of great Olympic coaches improves and enhances the quality of life of cancer patients in remission after hematopoietic stem cell allogeneic transplantation
Bone Marrow Transplantation (2020)
Fertility Concerns and Access to Care for Stem Cell Transplantation Candidates with Sickle Cell Disease
Biology of Blood and Marrow Transplantation (2020)
Long‐term Follow‐up of Hematopoietic Stem Cell Transplant Survivors: A Focus on Screening, Monitoring, and Therapeutics
Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy (2020)