To our knowledge, this is the first study, which explored influencing factors of adherence to medication and aftercare in a large cohort of patients with acromegaly. In this exploratory approach, we investigated potential influencing factors based on a theoretical model of adherence with the aim to identify the most relevant factors for patients with acromegaly. We found different factors to be important for adherence to medication compared to adherence to aftercare. Overall, our data suggest, that those patients with a self-assessed higher disease severity struggle more with adherence to medication. They were younger, had a higher subjective symptomload and worse mental QoL. On the other hand, patients with lesser symptoms had more problems with adherence to aftercare. These patients had a lower subjective symptomload and a lower perceived need for treatment.

Adherence to acromegaly-specific medication in the study group was very good, with approximately one half of the investigated patients reporting perfect adherence to their prescribed medication, whereas the other half reported to skip or reduce a dose at least now and then. No patient reported not to take their prescribed mediation at all.

The main reason for taking acromegaly medication was the advice of their treating physician, followed by a self-perceived improvement of physical and mental capacity. Interestingly, almost 40% of the patients in our sample stated to take their acromegaly medication because they felt that it reduced pain. Unfortunately, our data do not allow to differentiate whether this is connected to a described effect of somatostatin analogs on reversibility of joint thickening in acromegalic arthropathy [16, 17] or other factors. However, it is an interesting finding, worth to be elucidated further in future studies.

Of all sociodemographic/economic factors potentially impacting on adherence and investigated in the present study, we found only age to be of relevance for medication adherence in the direction that patients in this study sample stating problems with adherence were younger than those without. The same finding also emerged in a previous study by our group on adherence to GH replacement therapy [18]. Younger age as an influencing factor for poorer adherence has also been uncovered in other population-based studies, for example from Germany and Sweden [19, 20] and has been discussed to be related to perceived side effects of medication [3, 19]. It has also been found to be a relevant factor for treatment persistence in patients with acromegaly [21].

Condition-related factors also played a crucial role in adherence to medication. Patients with any problems of adherence to acromegaly medication in this study reported a significantly poorer mental QoL and a higher acromegaly symptomload. Similarly, a recent Bulgarian study found inadherence to acromegaly medication to be related to impaired QoL in the SF-36 [22]. These findings do also not come as a surprise. Across studies, poor mental QoL is highly associated with depressive symptomatology [23]. Depression itself, which is known to be related to negative illness beliefs, helplessness and lack of perceived illness control [24] is a known barrier to adherence [25]. We were unable to determine from our data, whether the higher symptomload, that also accompanied adherence problems in our study, is an expression of depressive symptomatology or a consequence of adherence problems. Probably, these factors are mutually dependent. Patients with a higher symptomload and a poorer QoL find it more difficult to take their medication, potentially because of a higher prevalence of depression and depressive coping strategies in this group. Lack of adherence in turn reduces the likelihood of successful treatment. This relationship may contribute to the fact that the QoL of many patients with acromegaly remains permanently reduced [26, 27]. Supporting these patients in acquiring better coping strategies and achieving better adherence might be a valuable leverage point to improve their psychological well-being. Interestingly, IGF-1-normalization was not related to adherence, perhaps indicating that the perceived disease severity according to symptomload and subjective QoL may be more relevant to patient motivation than hormonal status.

With regard to health system related factors, we identified the length of the medical consultation to be impacting on adherence to medication in patients with acromegaly. The patients with adherence problems tended to have a shorter duration of consultations and were significantly more often dissatisfied with the duration of their medical consultations than patients stating no problems with adherence. They found that physician did not address potential adherence problems well, significantly more often, then patients without problems in adherence. The importance of a good patient-provider relationship for adherence has been consistently shown in previous studies in patients with chronic health conditions (for an overview see [28, 29]). This finding underlines the importance of the communication skills of the treating physician for the success of medical treatment of acromegaly. This notion is supported by patient interview studies in which some acromegaly patients felt unable to discuss adherence problems with their doctor [6].

Adherence to aftercare in our study was also excellent, with over 80% of all patients reporting to see their endocrinologist and/or neurosurgeon at least once a year. Only few of all constructs investigated impacted on adherence to aftercare: In contrast to patients who had problems with medication adherence, patients with problems in adherence to aftercare had a lower symptomload and perceived need for treatment. This is in line with findings from Kasuki, who reported absence of symptoms to be one of the main reasons for loss to follow-up in patients with acromegaly in a pilot study across three treatment centers [5].

Interestingly, none of the patients who did not visit aftercare appointments regularly lived closer than 50 km away from the treatment center. These results suggest that the regular attendance of aftercare seems to be dependent on the patients understanding of its necessity and their willingness to take on long travel times to the treatment center. The effect of increased distance to the treatment center has been shown to be associated with an increase in likelihood of non-adherence to treatment or follow-up plans and—in some illnesses, even survival—in other diseases [30,31,32] in pediatric and adult populations. Our findings serve as arguments for increasing necessity beliefs about aftercare in acromegaly patients as well as the expansion of telemedicine services, especially in those patients with a milder course of the disease. Experience with telemedicine during the Covid-19 pandemic suggests, that adherence rates in patients with acromegaly can be significantly improved if the treating physicians stays in contact via online visits [33].

The only factor influencing both adherence to medication and adherence to aftercare, was a higher degree of concerns about medication than the belief in their necessity. This result is in line with a meta-analytic review of the necessity-concerns framework published in 2013 [34] in which across studies, higher adherence was associated with stronger perceptions of necessity of treatment and fewer concerns about treatment, independently of study size, country in which the research was conducted and the type of adherence measure used. In the context of our study, it can be seen as an important reminder that the patients’ concerns should be addressed and resolved during medical consultations to improve adherence in patients with acromegaly.

The major strengths of our study are a novel approach to measure adherence and influencing factors, combining self-designed questionnaires with standardized ones, and taking into account all the five dimensions on adherence defined by the WHO. Despite the considerable size of the questionnaire package, we managed to achieve a sufficiently large sample to obtain statistically meaningful results. The scientific approach to construct a study design based on a theoretical model to then eliminate irrelevant factors in an explorative study, allows a systematic investigation of adherence in acromegaly.

To address questions for which standardized questionnaires are not available, this study had to rely largely on self-developed questionnaires that have not been used before and for which normal values are not available. It is therefore a limitation of this study that our results cannot be compared to a healthy population. Also, the greater time efficiency of the validated short scales allowed us to examine more different factors in the same sample, but does not allow for the same in-depth analysis as their respective longforms. The large number of items was necessary for the exploratory research approach, but may also have led to a higher number of missing values, which we have dealt with by reporting valid percentages.

Therefore, it has to be kept in mind that the factors identified in this study are in need of further confirmation. We purposefully decided to include factors that were only nearly significant and to refrain from corrections for alpha errors. This limits the certainty with which we can identify influencing factors in this study, but serves the main purpose of this study to narrow down future research to those variables that have the potential to predict adherence. It is important at this explorative stage of the scientific understanding of adherence, to not falsely exclude factors, which would then not be further investigated. We suggest, to conduct prospective studies including all factors we found to be associated with adherence in this study, to further prove their actual influence.

An unforeseen limitation of this study was, that due to a change of date protection legislation, which went into effect after the finalization of the study plan, it was no longer possible to trace back former patients of the participating institutions to enquire about their present aftercare status and medication adherence. This limited the information we were able to accrue on patients lost to follow-up. We also have to acknowledge a potential response bias in that we cannot rule out that predominantly highly motivated and adherent patients participated in this study, who might not constitute a representative sample of the general group of patients with acromegaly. On the other hand, self-reports are susceptible to error and filling in in terms of social desirability, thus potentially overestimating the degree of adherence in our patients. However, this is a problem shared by all studies relying on self-reporting measures and cannot be avoided in patient-reported outcome research.