Contributing and limiting factors to guideline-adherent therapy in senior and elderly breast cancer patients: a questionnaire-based cross-sectional study using clinical and cancer registry data in Germany

Breast cancer remains the most common cancer in Germany (Robert Koch Institute 2022). As age demographics shift and patients live longer, increased attention must be paid to their unique characteristics and needs (Gomes et al. 2020). Guidelines have partly taken up the topic of ageing, as in the case of Germany’s “S3” guideline (“step 3” indicating the highest level of empirical evidence and expert consensus) for breast cancer (Leitlinienprogramm Onkologie 2017). Special needs of elderly patients are touched upon therein but rarely are age-specific treatment recommendations given.

Preliminary work has shown that older patients with cancer (70 years and older) are more often treated with deviation from guideline recommendations than younger patients. Older breast cancer patients are less likely to receive breast conservation therapy, adjuvant therapy, or radiation (Waldmann et al. 2008; Hancke et al. 2010; Peters et al. 2015; Wallwiener et al. 2016). It has also been shown that deviations from specific guideline recommendations are associated with higher mortality or lower recurrence-free survival in older patients (Wöckel et al. 2010). The interpretation of this finding is not straightforward, as various factors influence the treatment decision and its success. For example, medical factors, such as diminished organ functions, may stand in the way of guideline-congruent therapy. In addition to medical factors, many other conceivable aspects can cause therapy to deviate from the guideline recommendations. If, for example, a guideline-congruent treatment option is not pursued due to the patient’s values or individual preferences, then a deviation from the corresponding recommendation is not deficient but desirable in the sense of patient-oriented care, provided that this decision was made in an informed manner and with consideration of the evidence. Deviation from guideline recommendations is particularly problematic when due to poor communication about personal opinions and preferences. These considerations touch on fundamental questions about the “sense” or “meaningfulness” of specific therapy options concerning the patient’s life situation. Doctors, patients, or relatives can influence the therapy decision with such assessments (Gieseler et al. 2021). What is considered “still reasonable” for the older patient may deviate from the guideline recommendations. There is insufficient knowledge about the underlying factors contributing to this discrepancy and the general reasons for non-guideline compliant treatment. A recent study examined facilitators of and barriers to guideline-concordant care of breast cancer patients in the United States (McDougall et al. 2021). The authors identified social support, patient support services, and communication issues as intervention targets for improving breast cancer care. However, the authors enrolled patients aged 26 to 69, and transferring the findings to older populations is not readily admissible. Furthermore, their study did not address patients’ attitudes and preferences, and aspects of medical communication and cooperation were not dealt with in depth, although those were repeatedly identified as issues that complicate the treatment process and cause a lack of comprehension of medical knowledge (Kruse 2013; Füeßl 2015; van Vliet et al. 2015; Hoffmann et al. 2018; Groen-van de Ven et al. 2018; Jack et al. 2019). Therefore, we conducted a questionnaire-based cross-sectional study using clinical and cancer registry data in Northern Germany, analyzing clinical data and querying seniors (50–69 years at diagnosis) and elderly (70 years at diagnosis) breast cancer survivors, covering attitudes and preferences in addition to physical and psychosocial factors.

Out of 3075 patients contacted via the cancer registries of Hamburg and Schleswig–Holstein, 1315 (42.8%) returned a questionnaire—590 (32.7%) and 725 (57.1%), respectively. Responders were slightly younger than non-responders (66.1 years vs 68.6 years; P < 0.001). On average, questionnaires from patients recruited via cancer registries were answered 79 weeks after diagnosis; this was 34 weeks for patients recruited via cooperating clinics. Overall, 1,112 responders from cancer registries and 38 patients recruited in cooperating clinics submitted a usable questionnaire, and sufficient clinical data were available to assess at least one relevant guideline recommendation. Participants were predominantly diagnosed in earlier stages of cancer, with UICC stage I being most commonly observed in both senior and elderly patients (66% and 51%, respectively). Elderly cancer patients were more frequently in need of care (OR: 4.56; 95% CI: 2.56–8.44) and more likely to take more than four medications (OR: 1.81; 95% CI: 1.39–2.35) but were less frequently diagnosed with depression compared to younger patients (OR: 0.46; 95% CI: 0.33–0.64). The proportion of frail patients does not differ significantly by age group (OR: 1.68; 95% CI: 0.91–1.50). Patient characteristics are displayed in Table 2.

Deviations from guideline recommendations were observed in 206 (17.9%) patients, with 95 (13.6%) in the younger age group vs 111 (24.6%) in the elderly (OR: 2.07; 95% CI: 1.52–2.80). In both younger and elderly patients, deviation from guideline recommendations was associated with the subjective notion of not having received treatment suitable to their individual situation (OR: 0.48; 95% CI: 0.27–0.87 and OR: 0.52; 95% CI: 0.29–0.95, respectively). Of patients for whom more than one guideline recommendation was relevant, 3.2% in the age group 70 years and older were treated in deviation from the guideline recommendations in more than one manner. In the age group of 50–69 years, this was 1.5% (OR: 2.24; 95% CI: 0.82–6.08) (see Table 3). The guideline recommendation for adjuvant chemotherapy for HER2-positive or triple-negative tumors was the most frequently not followed in both age groups, with a non-adherence rate of 18.6% in younger patients and 40.8% in older patients (OR: 3.02; 95% CI: 1.60–5.67). Proportions of guideline adherence per recommendation differentiated by age groups are displayed in Fig. 1.

The univariate analysis of potential explanatory variables showed that different factors were associated with adherence to guideline recommendations depending on the age group. Depression before cancer diagnosis was linked with deviations from guideline recommendations in older patients (OR: 1.80; 95% CI: 1.00–3.22). The other variables assessing comorbidities (frailty, intake of medications, and need for care) did not significantly affect adherence to guideline recommendations in any age group. Conversely, for the younger age group, independent mobility was associated with adherence to guideline recommendations (OR: 2.98; 95% CI: 1.01–8.78). While the patient’s age at diagnosis did not matter in the age group 50 to 69 years, it was a significant factor in the group of elderly patients. Therein, older age reduces the chance of treatment according to the guidelines (OR: 0.94; 95% CI: 0.90–0.99). The satisfaction with the therapy decision-making process was significantly associated with guideline-adherent therapy in the elderly (OR: 0.52; 95% CI: 0.33–0.83). The associations between adherence to guideline recommendations and potential explanatory variables, differentiated by age groups, are shown in Table 4.

In the logistic regression model for younger patients, the preference for “quality of life” over “prolongation of life” remained a relevant factor after adjusting for covariates. Patients who stated that quality of life had been the decisive factor in weighing the treatment decision have a significantly higher chance of treatment deviating from guidelines compared to patients for whom quality of life and the prolongation of life were equally important (AOR: 2.08; 95% CI: 1.11–3.92). The strongest factor in the univariate analysis of younger patients—independent mobility—is no longer significant after adjustment for the covariates. In the elderly, the patient’s age at diagnosis remained positively associated with deviations from guideline recommendations (AOR: 1.06; 95% CI: 1.01–1.11). Likewise, the chance of receiving guideline-adherent treatment was lower in patients who were diagnosed with depression prior to their cancer diagnosis (AOR: 1.84; 95% CI: 1.00–3.40). In contrast, patients satisfied with the therapy decision-making process were more likely to be treated in accordance with guideline recommendations (AOR: 0.56; 95% CI: 0.34–0.91). The logistic regression models are displayed in Table 5.

The generalized variance inflation factor of the covariates was between 1.008 and 1.044 in the model for the younger patients and between 1.005 and 1.016 in the elderly patients’ model.

Using medical data and questionnaire responses from a sample of 1,150 women with breast cancer aged 50 to 91, we investigated potential age-related factors associated with adherence to guideline recommendations. Older women were considerably less likely to be treated per guideline recommendations than younger patients (75.4% overall adherence rate vs 86.4%). The association between age and guideline adherence was also found within the older age group, and is consistent with previous and comparable research (Wöckel et al. 2010; McDougall et al. 2021).

Deviations from guideline recommendations are required when this fits the patient’s individual situation and her needs and preferences. Accordingly, in the younger age group, we found that patients who reported that quality of life was guiding their therapy decision were twice as likely to be treated in deviation from guidelines compared to younger patients with no such preference. This suggests that in these cases, omissions of guideline-recommended therapy components were conscious, likely due to patients fearing their quality of life would be unduly impaired. Such seeming violations of the guidelines are not problematic when it is ensured that patients have made an informed decision with a balanced consideration of the expected side effects, including the availability of supportive care options (Jordan et al. 2017), but have not acted out of irrational fears.

On the other hand, in our study, both younger and elderly patients were more likely to be retrospectively convinced that their therapy was not appropriate when treated in the deviation of guideline recommendations. In these cases, the patients’ needs might not have been met sufficiently. This assumption is supported by two findings in the older age group: first, elderly patients with depression were significantly more likely to have received treatment deviating from guideline recommendations after adjustment for covariates. Depression in and of itself should not influence a treatment decision. However, it is conceivable that a depressive disorder impairs motivation for demanding guideline-based therapy (Hindmarch et al. 2013). All the more important is psycho-oncological counseling, which, according to the guidelines, should be offered to all patients. In our data, however, this was less often the case in older patients (68.1% compared to 46.2%). Second, elderly patients’ satisfaction with the decision-making process was significantly associated with adherence to guideline recommendations in univariate analysis and after adjustment for covariates. In cases where the patient’s personal situation suggests to deviate from guideline recommendations, pros and cons of the alternatives should be communicated and discussed particularly well. This in turn should be reflected in higher satisfaction with the decision-making process. The fact that this is not the case here is again an indication that patients’ needs have not been met sufficiently.

In the previous studies, a lack of social support and inclusion has been identified to inhibit successful cancer treatment (Kroenke et al. 2006; Lambert et al. 2018). This is not reflected by higher rates of guideline-discordant therapy in our sample; the level of social support was not related to the treatment decision in either younger or older people.

Research from the broader context of this study suggests that in elderly patients, informed and autonomous therapy decision-making should not be taken for granted due to those patients’ declining independence and subsequent dependence on third parties, like confidants and caregivers (Gieseler et al. 2021; Heidenreich et al. 2023). In Germany, for example, it has been criticized that cancer treatment is too seldom organized routinely by a managing or guiding physician (Pfaff and Schulte 2012). Then, the challenging task of coordinating the complex treatment process falls to the patient. We have not assessed the skills required for this in detail. However, it seems likely that those are less pronounced in older people. In our study, we found that the chance of receiving guideline-discordant care increased by 6% for each year of age. To some extent, the influence of age may represent a decline in communication and organizational skills. Future investigations that take a more differentiated look at the therapy decision-making process and allow for minute reconstruction of the reasons underlying therapy decisions are necessary here: which decisions are informed by the patients’ explicit wishes, and which treatment options were presented and recommended as allegedly suitable for their situation? Only in this way can a meaningful distinction be made between informed and autonomous decisions and those that either presume the patient’s will or do not even consider it.