Retrospektive Analysen zu unipolarer Depression unter besonderer Berücksichtigung des Alters und somatischer Komorbiditäten

The aim of this work was to investigate the relationship between depression and comorbid somatic health risks, whose impacts are known to be exacerbated by depression. To illustrate this relationship, three studies were conducted with the following aims i) Investigate whether age at depression diagnosis has an impact on anti-depressant medication prescribed by physicians ii) Assess whether the risk and incidence of common cancers is increased in patients with depression iii) Differentiate the risk conferred by depression from the risk associated with other comorbid psychiatric disorders on the risk of cancer, illustrated with the case of gastrointestinal (GI) cancers This dissertation is based on these three publications published in the Journal of Psychiatric Research, Brain Sciences, and Journal of Cancer Research and Clinical Oncology. All three analyses used secondary data from the Disease AnalyzerTM (DA) database (IQVIA). In the first study published in 2021 as “Age effects on treatment patterns in 138,097 patients with unipolar depression followed in general practices in Germany“ in the Journal of Psychiatric Research we investigated within a retrospective cohort analysis whether treatment prescribed varied with age at depression diagnosis. A cohort of 13,553 patients with depression diagnosed first-time between January 2015 and December 2018 within the DA database was selected, to allow for at least 12 months follow-up after depression diagnosis at the time of data extraction. Patients aged over 18 years were grouped by age (18-30, 31-65 and >65 years) and Odds Ratio (OR) for specific treatment types assessed between the groups matched by gender and insurance type, with OR adjusted for by gender, insurance type, treatment site and Charlson-Comorbidity-Index. Less than half of patients in each group received treatment, with patients aged 65 or older most likely to receive medication (48.3%) compared to 31–65-year-olds (42.4%) or 18–30-year-olds (34.8%). Older patients showed an increased probability (OR: 1.3 [1.26–1.34 95% CI], p < 0.0001) for tri- and tetra-cyclic medication, while younger patients showed an increased probability for SSRIs and SNRIs (OR: 1.23 [1.16–1.30 95% CI], p <0.0001). The second study published in 2023 as “Depression is associated with an increased risk of subsequent cancer diagnosis: a retrospective cohort study with 235,404 patients“ in Brain Sciences aimed at assessing whether cancer risk was affected by depression diagnosis. We compared 117,702 patients with a depression diagnosis documented in the DA database between January 2015 and December 2018 with a comparator group of 117,702 patients 1:1 matched based on index year, age, gender, visit frequency, and treating site, allowing for a minimum follow-up of three years at time of Seite 83 data extraction. Depression patients with previous bipolar disorder (F31), mania (F30) or schizophrenia (F20–29) and cancer before index date were excluded, while from the comparator group any patients with any previous psychiatric disorder or cancer were excluded. 4.9% of patients with depression compared to 4.1% without depression received at least one cancer diagnosis over 3.9 years median follow-up. The depression group showed an 18% increase in risk for a cancer diagnosis overall, with largest increased risk in lung cancer (HR: 1.39 [1.21–1.60], p < 0.0001), cancers of the gastro-intestinaltract (HR: 1.30 [1.15–1.46], p < 0.0001), breast (HR: 1.23 [1.12–1.35], p < 0.0001) and urinary (HR: 1.23 [1.06–1.43],p < 0.01). Similarly, the incidence of cancer diagnosis overall increased by 22% for depressed patients. IRs showed no difference across cancer types. The third study published in 2023 as “Psychiatric disorder and its association with gastrointestinal cancer: a retrospective cohort study with 45,842 patients in Germany” in the Journal of Cancer Research and Clinical Oncology investigated whether there was an association between psychiatric illness and cancer, and whether the association differed by type of psychiatric illness. Cases were selected from the DA database if a first GI-cancer diagnosis occurred between 2015 and 2020. Comparators with no diagnosis of cancer at any time were matched 1:1 using nearest neighbour matching based on propensity scores calculated from age, sex, visit frequency and total observation time. The final study cohort of 44,582 patients aged ≥18 years with sex documented was assessed for presence or absence of psychiatric disease before the index date (first cancer diagnosis or random visit between 2015 and 2020). 46.8% of patients with GI-cancer had a psychiatric diagnosis before the index date, as did 45.6% of patients without cancer. Depression was the most common psychiatric diagnosis (21%), followed by psychosomatic disorder (13-16%) and PTSD (9%). Patients in both groups received previous psychiatric treatment (22-23%). There was no association between previous psychiatric diagnosis overall, as well as for previous diagnoses of depression, PTSD, anxiety, schizophrenia, or ADHD with cancer risk, as the ORs approached or straddled 1.0. Only the association between psychosomatic disorder and risk for GI-cancer was significant at the adjusted α-level (OR: 0.85, 0.81- 0.90 95%CL, p<0.0001). The results were consistent regardless of whether the model was adjusted only for relevant somatic comorbidities, or also by previous psychiatric treatment. In conclusion, this body of work was able to show age of patients at depression diagnosis impacted the type of anti-depressant treatment prescribed. We further showed depression increased the risk for cancer over a 4-year observation period, with different strengths of association depending on the type of cancer examined. We were not able to replicate this finding in a cohort of cancer patients with previous depression, however we were able to show psychosomatic disorder conferred a reduced risk of GI cancer in an 8-year pre-index observation period. This association was not altered when taking previous psychiatric treatments into account, which are known to impact the risk of cancer. We demonstrate that depression treatment is impacted by somatic health risks such as age, and Seite 84 depression in turn impacts the risk of certain types of cancer. Further research is required to understand the relationship between depression and cancers, considering also psychiatric and somatic comorbidities and different cancer types.