Key words
treatment adherence - growth hormone deficiency - growth hormone therapy - recombinant
growth hormone - health knowledge
Introduction
Growth hormone deficiency (GHD) is a chronic, etiologically diverse and multi-faceted
disease. It leads to abnormal body composition (decreased lean body mass and increased
fat mass), osteoporosis, altered lipid and glucose metabolism, as well as reduced
quality of life (QoL) and is associated with an increased incidence of cardio-vascular
events and increased mortality [1]
[2]
[3]
[4]
[5]. Treatment consists in replacement with recombinant human growth hormone. Currently,
available formulations stipulate daily administration via subcutaneous (s.c.) injections.
Growth hormone replacement therapy (GHRT) has proven to be safe and effective [6].
However, symptoms develop gradually and the treatment effect is not immediately noticed
by patients due to the lack of acute symptoms, whereas the discomfort of daily injections
is a tangible inconvenience. Thus, adherence of GH replacement therapy is an issue
that has been explored, particularly in the setting of childhood-onset GHD where low
adherence was consistently observed [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]. Adherence in adult GHD patients is especially affected by dissatisfaction with
the perceived therapy results and a bad clinician-patient relationship [15]. A significant reduction of adherence and persistence within the first four years
of treatment has been described [16]. There have been a number of attempts to improve treatment adherence by utilizing
different non-daily dosing regimens including every other-day [17] and 3 times weekly dosing [18]
[19]
[20]. Clinical outcomes were similar, but the effect on adherence remained unclear as
some studies showed a better adherence while others did not [21]. Several long-acting rhGH preparations are currently under clinical investigation
and thought to potentially improve adherence [22].
Investigations of adherence in patients with adult growth hormone deficiency (AGHD)
are scarce [23]. We sought to determine patients’ knowledge of the disease and treatment, as well
as attitudes towards current treatment options versus newly developed LAGH in patients
through a questionnaire-based approach, especially by comparing perception of patients
receiving replacement therapy with patients who refuse it. Further objectives were
to assess potential factors that influence adherence and persistence.
Methods
Design of the questionnaire
A systematic literature research did not reveal any validated questionnaire eligible
for our specific approach. Based on pre-existing adherence questionnaires and published
data on specific advantages and disadvantages of their items [24]
[25] we created a tailored questionnaire covering the domains of medication taking behaviour,
barriers to adherence and beliefs associated with medication adherence. Notably, we
adopted most adherence-related items from the Medication Adherence Questionnaire
[26], the Hill-Bone Adherence Scale
[27] and the Medication Adherence Rating Scale
[28] and modified them to fit the context of GH treatment. In particular, we asked patients
not currently undergoing GHRT treatment about their reasons for treatment discontinuation.
The questionnaire also included a section capturing the knowledge domain concerning
GHD by asking the patients to write down symptoms and effects on health of GHD in
a short-answer format. The questionnaire also assessed attitudes regarding long-acting
GH formulations and requested demographic data including sex, age and level of education.
The latter was operationalised by the highest degree of general education following
the recommendations of the German statistical federal office. The questionnaire was
sent by mail with a prepaid return envelope enclosed, which allowed the patients to
remain completely anonymous and not assume any costs for participation. All questions
and their possible answers are presented as supplementary material (Table 1S).
Questionnaire evaluation
Non-adherence score
The questionnaire included six questions pertaining to adherence. A score ranging
from 0 to 18 was computed for every respondent. Four questions were coded with a 5-point
Likert scale ranging from “never” (=0 points) to “more than 3 times per week” (=4
points): “How often do you forget your growth hormone injections?”, “Do you skip injections
when feeling healthy and having the feeling that your symptoms are well controlled?”,
“Do you sometimes skip injections when you’re sick?”, and “Have you ever injected
less than recommended or nothing at all because of feeling worse after injection?”
A fifth 5-point Likert scale question (“How often do you forget your medication when
traveling or spending the night elsewhere?”) was accounted for by dividing the points
by 4 to factor in the non-routine role of travel in most people’s lives. The last
question contained binary information and was awarded one point if disagreed (“Have
you injected growth hormone every single day in the past 7 days?”). The score result
was then correlated with non-adherence and treated as ordinal data.
GHD knowledge score
The free-text statements from patients about the effects of GHD on health were grouped
into different symptom categories. These categories are listed below with exemplary
answers as given by patients:
-
Physical performance: fatigue, rapid exhaustion (“one is not fit”, “one performs worse
in sport”).
-
Metabolism: altered lipid metabolism, altered glucose metabolism (“disturbed metabolism“,
“diabetes”).
-
Body composition: reduced muscle mass, increased fat mass, increased abdominal fat
mass (“one becomes fat”, “one loses muscles”).
-
Cardiovascular risk factor (“risk for heart diseases”).
-
Bone mineral density (“osteoporosis“).
-
QoL and psychological disturbances (“depression”, “bad mood”, “less well-being”).
-
Increased mortality (“reduced life expectancy”).
A score for knowledge of GHD was calculated by counting the number of different symptom
categories mentioned by each patient, regardless of the number of symptoms stated
in each category. Hence, a score result of 0 indicates the patient was not able to
make any correct statement about the effects of GHD, while a result of three indicates
the patient mentioned symptoms belonging to three different categories. The result
was treated as ordinal data.
Study population
The questionnaire was distributed amongst 106 adult patients with GHD who attended
our Endocrinology Outpatient Clinic at least once since 2000, and who participate
in the observational study of the Network of Excellence for Neuroendocrine Tumours
Munich (NeoExNET). The local ethics committee granted ethical approval. All study
procedures were performed in accordance with the Declaration of Helsinki and all patients
gave informed consent. 70 patients completed the questionnaire correctly and returned
it (response rate: 66%). The study population was well balanced in regard to sex distribution
(49% female, 51% male), onset type (61% adult-onset, 33% childhood-onset, 6% not disclosed)
and level of education (data not shown), cf. [Table 1] for an overview of the study population. Mean age of patients with childhood-onset
GHD was 48.7±12.9 years and 60.5±12.1 years in AoGHD patients.
Table 1Demographic data of surveyed patients.
|
patients with GHRT
|
patients without GHRT
|
all patients
|
|
n (% of total)
|
46 (66%)
|
24 (34%)
|
70
|
|
male/female patients (%)
|
24: 22 (52%: 48%)
|
10: 14 (42%: 58%)
|
34: 36 (49%: 51%)
|
|
age (mean±SD)
|
57.1±12.7
|
55.4±15.1
|
55.9±13.5
|
|
childhood: adult onset (%)
|
15: 31 (32%: 68%)
|
9: 15 (37%: 63%)
|
24: 46 (35%: 65%)
|
Adherence, attitudes and beliefs of growth hormone deficient patients – A questionnaire-based
cohort study.
Statistics
Statistical analysis was performed using IBM SPSS Statistics (IBM Corporation, Armonk,
NY, USA). Correlations were calculated using Kendall’s tau b (ordinal – ordinal),
Mann-Whitney U test (nominal – ordinal) and chi-square test/Fisher’s exact test (nominal
– nominal). For all calculations concerning adherence only patients under GHRT were
considered. P values ≤ 0.05 were considered significant.
Results
Adherence and burden of therapy
76% of the patients receiving GHRT reported forgetting a GH injection “never” (41%)
or at the most “less than once per month” (35%), implying good adherence in our study
group. On the other hand, 18% skipped injections occasionally (“less than once per
week”) and 6% reported skipping injections several times weekly. None stated forgetting
injections more than 3 times per week. 96% reported they “never” (80%) or “seldomly”
(16%) skipped an injection because of feeling healthy or the lack of disease symptoms.
98% of the patients reported to “never” (76%) or “seldomly” (22%) skip a dose when
ill. Furthermore, patients did not skip doses because of feeling worse after injection
(“never” 94%, “seldomly” 6%). Accordingly, the mean non-adherence score was very low
(2.0±2.5 points; range: 0–11) with a median of 1 out of 18 possible points.
All patients except for one injected GH in the evening, as recommended. Only 10% reported
they felt burdened and 27% admitted to “sometimes” feeling burdened by GH therapy.
In addition, several patients added a handwritten comment reporting they felt burdened
“while travelling” or “on vacation.” 18% of the patients reported forgetting their
GH injections “often” or “most of the times” when travelling or leaving home, with
12% of the patients reporting to forget at least “occasionally.”
Reasons for treatment refusal
Of the 70 respondents in our study 24 patients received no GHRT (34%). Among six of
these 24 patients, GHRT was stopped by their treating physicians. The reasons listed
for stopping treatment were tumour recurrence/growth in 4 cases, hyperglycaemic dysregulation
in one case and one patient stopped GHRT after suspected apoplexy. The other patients
refused GHRT for various non-medical/personal reasons. The most common reason why
these patients refused GHRT was fear of side effects (n=9). It was also mentioned
that the benefit of GHRT was considered insignificant (n=6), and some patients did
not feel well informed about GHD and its treatment (n=5). Four of the five patients
who reported feeling uninformed also indicated they were fearful of adverse effects.
Two patients stated that the daily injections were too uncomfortable.
Perceived GH effects
A majority (76%) of the patients receiving GHRT stated they didn’t notice any differences
during the day whether they had or had not injected GH on the evening before; 22%
stated feeling better. Although in the long term, 42% noticed a significant difference,
with only 9% reporting they did not (49% “did not know”). As expected, currently untreated
patients differed significantly and negated a feeling of efficacy (p=0.002). About
half (52%) of all participants estimated that GH therapy has a strong impact on their
health and QoL, whereas only 8% considered this as negligible or non-existent (40%
couldn’t say). Again, patients receiving active treatment were more optimistic about
the effects of therapy than untreated patients (strong impact 61 vs. 32%; p=0.012).
Correspondingly, almost half (47%) of the participants judged the effect of GH therapy
on their body shape and appearance as positive, 11% as negative, and 42% did not perceive
an effect.
Knowledge of GHD
34 participants (49%) could not provide any correct symptom of GHD. Of the remaining
patients (n=36) a majority stated impairment of performance (n=26). Additional mentioned
symptom categories were abnormal body composition (n=19), reduced QoL and other psychological
disturbances (n=11), altered lipid and glucose metabolism (n=10), reduced bone mineral
density (n=10), increased cardiovascular risk (n=3), and increased mortality (n=1).
The mean GHD knowledge score was 1.1±1.4 (range: 0–4).
Attitudes towards LAGH therapy
When asked whether or not patients would want to start or switch to a LAGH that was
administered once weekly 36% of the patients responded affirmatively and 44% considered
the possibility to do so. 20% indicated no interest. Of note, 12 of respondents had
previously participated in a clinical trial with LAGH. When comparing answers, LAGH
experienced patients significantly favoured switching to LAGH as compared to LAGH
inexperienced patients (p=0.048). Childhood-onset GHD (CoGHD) patients were more willing
to make the switch than adult-onset (AoGHD) patients (57% vs. 23%; p=0.031). Patients
with and without GHRT were regarded separately. The willingness to switch or start
such a treatment did not differ between patients receiving active daily GH therapy
and untreated patients. Opinions about the risk of forgetting weekly administration
were divided: 37% considered the probability higher for weekly than for daily administration,
34% considered it lower, and 28% considered it equal. Patients without prior LAGH
experience seemed to judge this risk higher than the study participants, although
this failed to reach significance (p=0.162). Similarly, treated patients seemed more
wary than untreated patients and estimated weekly injections to bear a higher risk
of forgetting injections without reaching significance (p=0.066). Of the LAGH study
participants eight patients stated that weekly administration has had an advantage
over daily administration, especially “less expenditure of time” and “fewer injections”
were mostly named as beneficial effects. Two patients reported feeling better under
LAGH therapy compared to conventional GHRT. Overall, 10 of the 12 LAGH patients preferred
weekly injections, two preferred daily injections.
Parameters influencing adherence and treatment readiness
Adult- versus childhood-onset
There was no difference in treatment refusal or GHD knowledge score between AoGHD
and CoGHD patients. Although treated CoGHD patients had worse non-adherence scores
(mean 3.9±3.3) than treated AoGHD patients (2.5±2.2) the difference failed to meet
the level of significance.
Level of education and knowledge of GHD
Patients receiving current GH treatment had significantly higher GHD knowledge scores
as compared to untreated patients (1.4±1.4 vs. 0.6±1.2; p=0.023), see [Fig. 1] for a histogram of the scores for both groups. Equally, the level of education (having
obtained a level 3 degree according to the International Standard Classification of
Education) was higher in treated patients (e. g., 29.5% vs. 12.5%; p=0.017).
Fig. 1 Histogram depicting GHD knowledge scores of those patients with (dark grey) and those
without (light grey) current growth hormone replacement therapy; higher scores indicate
a better knowledge of the disease.
Discussion
In this study we surveyed growth hormone deficient patients regarding adherence, attitudes
towards and beliefs about GHRT and knowledge of the disease. 76% of our GHD patients
currently under treatment were well adherent to it, with less than one missed injection
per month. This result is well in agreement with the literature. Abdi et al. reported
good adherence in 69% (less than two missed injection per month) [21] , Rosenfeld and Bakker reported no missed injection during three months in 70% of
their patients [15].
However, 24 (34%) of our survey participants were currently not receiving treatment
or any GH substitution. Of those, 75% had no clear medical reason for discontinuation
of their treatment. Amongst the reasons mentioned were fear of side effects, low expectation
of therapeutic benefits as well as lack of information about GHD and GHRT. A majority
of patients refusing treatment did not believe GHRT influenced their health and/or
QoL. Only two patients reported pain from the injections as a reason for discontinuation.
Furthermore, we could show that knowledge of GHD is limited in these patients. Quintessentially,
in most cases a lack of belief in treatment efficacy due to misinformation seems to
precipitate unwillingness to pursue GHRT. This is in agreement with the results of
Abdi et al., who described the perceived lack of therapeutic effect as the most common
reason for discontinuation of GHRT [21]. It has previously been shown that generally adherence and patients’ perception
of disease severity strongly correlate [29]. As patients not receiving treatment had a lower level of education, it is to be
suspected that in these patients the aptitude to comprehend their disease and the
consequences of non-treatment is limited. To improve treatment of these patients,
more focused educational and behavioural strategies may be needed [30]
[31]
[32]
[33].
We found no significant difference between patients with CoGHD as compared to AoGHD,
even though the data indicates lower adherence in CoGHD patients. Adherence of GHRT
has thoroughly been investigated in children, but only limited data are available
in adult patients with CoGHD. In a recent study Auer et al. report higher adherence
in AoGHD than CoGHD patients, but this finding was not independent of the age of the
patient, indicating that younger patients independent of their age of onset are less
adherent to therapy [23]. As our CoGHD cohort is older than in the study by Auer et al. this might explain
the discrepancy of these results.
Willingness to initiate therapy with LAGH was lower than we predicted. Only 36% of
the respondents affirmed they would initiate treatment with a LAGH if it were available.
This attitude towards LAGH did not differ between patients currently treated or not
treated. That is possibly due to a sceptical or observing attitude regarding new pharmaceuticals
and maybe more patients would opt for them when they have proven to be safe. Interestingly,
CoGHD patients were more willing to make the switch than AoGHD patients. About a third
each rated the probability of forgetting a weekly injection compared to a daily injection
as higher, equal, or lower. This is particularly noteworthy because missing an injection
of LAGH has a much bigger impact than missing an injection of daily GH. As a historic
comparator one might think of bisphosphonates where switching from daily to weekly
administration improved adherence [34]
[35] and reduced the rate of discontinuation of therapy [35]
[36].
12 patients were included in our survey who had previously gained first-hand experience
with LAGH as participants of clinical trials. Overall, 10 of the 12 LAGH patients
preferred weekly injections, two preferred daily injections. This might imply a selection
bias of patients who are highly motivated to adopt novel therapies, but it might also
indicate that the adoption of new therapeutic regimens necessitates its application.
As this sample is small this can be only regarded as indicative data that need to
be expanded on in larger studies.
Our data were collected using a self-reporting survey. That raises the problem of
overestimating adherence, because patients might report better than actual adherence
due to a feeling of guilt. Allowing them to fill and post the questionnaire completely
anonymously was supposed to prevent that effect. This in turn disallowed us from comparing
survey results with actual clinical data from the medical records so that we cannot.
A drawback of every retrospective survey is the risk of recall bias. An alternative
and allegedly more objective method of assessing adherence is evaluating prescription
data [37]. However, this method is prone to bias as well. E.g., it cannot be excluded that
patients who indeed get their prescriptions regularly throw away leftover medication,
do not correctly apply the medication or just don’t have their prescriptions filled.
It would have aided to identify false-positive reports of adherence in patients not
receiving prescriptions, which we cannot exclude. Indeed, we emphasized opinions and
beliefs about the benefits and discomforts of therapy which evaluation of prescription
data cannot assess as is equally true for the assessment of adherence through measurement
of metabolite levels. Moreover, IGF-I has been shown to be a poor marker for GHRT
adherence because patients tend to inject GH more reliably prior to an appointment
[23]. Hence, we determined a survey was the most eligible method of data collection for
our approach.
A strength of our study is the high return rate and relatively large single center
study population. It is also, to our knowledge, the first study to present data on
attitudes towards LAGH. While in theory novel long-acting growth hormone formulations
are expected to increase adherence [22] , patients seem to be wary to adopt it. and a small fraction seems to prefer staying
on daily rhGH treatment. Long-term efficacy and safety of those new drugs, as well
as their implication for costs and adherence, remain to be elucidated [38]. As Caicedo & Rosenfeld indicate, there are still some challenges in the delivery
of growth hormone therapy to be overcome [39].
In conclusion, we could show that patients’ educational level and beliefs regarding
therapeutic effect is a strong driver of treatment adherence. Treating endocrinologists
should identify patients refusing treatment and consider intensifying disease and
treatment education. Furthermore, we demonstrated a reluctance of GHD patients to
initiate treatment with LAGH. Patients with prior LAGH experience seem to be much
in favour of these new drugs.
