[Time-dependent confounding in the estimation of treatment effects in randomised trials with multimodal therapies--an illustration of the problem of time-dependent confounding by causal graphs]. / Verzerrte Schätzung von Behandlungseffekten in randomisierten Studien mit multimodalem Behandlungsprozess--Darstellung des Problems zeitabhängiger Störvariablen anhand kausaler Graphen.

Biased effect estimates induced by unconsidered confounding variables are a known problem in observational studies. Selection bias, resulting from non-random sampling of study participants, is widely recognised as a problem in case-control and cross-sectional studies. In contrast, possible bias in randomised controlled trials (RCTs) is mostly ignored. This paper illustrates, by applying directed acyclic graphs (DAGs), possible bias in the effect estimates of first-line therapy, caused by subsequent changes in therapy (time-dependent confounding). Possible selection bias, induced by not only random loss to follow-up, will be explained as well using DAGs. Underlying assumptions of standard methods usually used to analyse RCTs (like intention-to-treat, per-protocol) are shown and it is explained why effect estimates may be biased in RCTs, if only these conventional methods are used. Adequate statistical methods (causal inference models as marginal structural models and structural nested models) exist. Higher documentary efforts, however, are necessary, because any changes in medication, loss to follow-up as well as reasons for such changes need to be documented in detail as required by these advanced statistical methods. Nevertheless, causal inference models should become standard along side the currently applied standard methods, especially in studies with high non-compliance due to changes in therapy and substantial loss to follow-up. Possible bias cannot be excluded if similar results are obtained from both methods. However, study results should be interpreted with caution if they differ between both approaches.

Prevalence and effectiveness of first-, second-, and third-line systemic therapy in a cohort of unselected patients with advanced non-small cell lung cancer.

Systemic therapy is the most relevant option for patients with advanced non-small-cell lung cancer (NSCLC) and many receive therapies beyond first-line. Little is known on response, progression free survival (PFS) and overall survival (OS) and their prognostic factors after second- and third-line therapy in daily clinical practice. Between January 2003 and July 2007, 406 consecutive patients were included in this prospective observational study and followed up until August 2010. At each treatment line the timing and kind of therapy, best response achieved, sites and time of progression were documented. Multiple logistic and Cox-regression models were used to analyse prognostic factors for achieving disease control (DC: response or disease stabilization), PFS and OS after different therapy lines. DC rate and median PFS decreased from 64% and 146 days, to 41% and 49 days, and to 39% and 51 days in response to first-, second- and third-line, respectively. A strong predictor for a worse outcome after second-line was development of new metastases after first-line therapy (DC: OR=2.50; 95% CI: 1.30-4.83; p-value=0.006; PFS: HR=1.53; 95% CI: 1.13-2.06; p-value=0.005) or achieving no DC after first-line (OS: HR=1.41; 95% CI: 1.01-1.97; p-value=0.041). Achieving no DC after second-line was a strong negative predictor for all outcome measures after third-line therapy (DC: OR=5.10; 95% CI: 1.56-16.6; p-value=0.007; PFS: HR=2.00; 95% CI: 1.23-3.27; p-value=0.005; OS: HR=1.69; 95% CI: 1.02-2.79; p-value=0.042). In conclusion, response in previous line and no involvement of new metastases after progression were relevant positive prognostic factors. However, further research is necessary to identify optimal therapy sequences.

Every-day clinical practice in patients with advanced non-small-cell lung cancer.

Little data is available on the treatment of patients with advanced non-small-cell lung cancer (NSCLC) in every-day clinical practice. Clinical trials hardly reflect reality due to strict selection criteria and a focus on the treatment line of interest. Thus, we aimed to describe the prevalence of treatment measures in a cohort of unselected patients with advanced NSCLC. From January 2003 to July 2007, we included 416 consecutive NSCLC patients treated at a single institution in this observational study, with a follow-up until August 2008. At each treatment line the time, duration and kind of therapy (drugs used, radiotherapy, surgery), response and time of progression were documented. Of the 405 patients receiving first-line systemic therapy, 52%, 26% and 10% received a second-, third- and fourth-line systemic therapy, while 31%, 40% and 50% of the patients died during or after their first-, second- and third-line therapy, respectively. About 35% of all patients received radiotherapy at any time. About half of the patients with advanced NSCLC undergoing systemic therapy in first-, second- or third-line also receive subsequent therapy. Although this is expected to influence overall survival, previous or further lines are seldom presented in trials in detail and are usually not included in statistical analyses.

[Social contact and depression in geriatric patients: is there an influence of gender?]. / Soziale Kontakte und Depression bei geriatrischen Patienten: Gibt es einen Einfluss von Geschlecht?

Depression is one of the most common affective illnesses. The investigation of changeable factors that are associated with depression is an important condition for the establishment of preventive measures. In a cross-sectional study, data on social factors were recorded from 580 geriatric patients of the municipal hospital in Munich. Ordinal logistic regression was used to analyse their association with depression (clinical diagnoses by ICD-10) and depressive symptoms (GDS). The results indicate that the occurrence of depressive symptoms and of depression was associated with less social contacts and less support, even after adjustment for other risk factors (for example, physical and cognitive impairment). This association was more distinct for women (for example, depressive symptoms: moderate versus much contact: OR=2.7; 95% CI: 1.8-4.1) than for men (OR=1.3; 95% CI: 0.7-2.4). Further research is needed to investigate whether women could benefit more than men from programmes promoting social support.

[Relation between certain diseases and frequency of depression in geriatric patients]. / Depressionshäufigkeit in Abhängigkeit von verschiedenen Erkrankungen bei geriatrischen Patienten.

The higher prevalence of depression in specific diseases and older persons is discussed. This prevalence varies greatly according to the method used to collect data. A risk group can only be defined if information on diseases and other influencing factors are collected uniformly. The target diagnoses Parkinson's disease, stroke, myocardial infarction, cancer, diabetes mellitus, chronic pain, multiple infarct syndrome, Alzheimer's and other dementia were recorded from 1208 geriatric patients of the ZAGF municipal hospital in Munich, Germany. Logistic regression was used to identify chronic pain as the main cofactor for an association with depression (clinical diagnoses by ICD-10) and depressive symptoms (via GDS [Geriatric Depression Scale]). This association was also found for multimorbid patients with chronic pain. Impairment of the activities of daily living and the clinical setting were important additional cofactors. Pain patients are therefore at higher risk for depression.

Sex-specific usage patterns of sleeping sites in grey mouse lemurs (Microcebus murinus) in northwestern Madagascar.

Sleeping sites are a potentially important resource for grey mouse lemurs since they are confronted with high daily temperature fluctuations and a high predation pressure. In order to determine the existence and degree of resource competition, sleeping site characteristics, locations, and usage patterns as well as sleeping group compositions were investigated in a 3 month field study in a dry deciduous forest of northwestern Madagascar. The daily sleeping sites of females were on average better insulated and safer than those of males. Males used more sleeping sites and changed the site more often than females. During the whole study, males slept alone, whereas the females formed stable sleeping groups in on average 83.7% of the days. Sex-specific differences in usage patterns might be explained by intersexual resource competition and female dominance and could possibly be related to differential parental investment of the sexes. The underlying study indicates that sleeping sites may be a restricted and defendable resource for grey mouse lemurs. The investigation gives new insights into the distribution patterns and social organization of this species.