Treatment burden on patients receiving intravitreal anti‐VEGF for wet age‐related macular degeneration

The aim of this study was to map the treatment burden for patients with wet age‐related macular degeneration (wAMD).


| I N T RODUC T ION
Age-related macular degeneration (AMD) is the most common cause of blindness among the elderly population in the developed part of the world (WHO, 2012).In earlier studies, its neovascular form (wet AMD or wAMD) accounted for an estimated 20% of the patients with AMD but nearly 90% of legal blindness caused by AMD (Ferris 3rd et al., 1984).Untreated wAMD leads to severe visual impairment in three fourths of patients within 3 years (Wong et al., 2008).In the last decade, treatment for wAMD has been revolutionized by injections of medications that hinder the development of new blood vessels (Comparison of Age-related Macular Degeneration Treatments Trials Research et al., 2016).Therefore, patients with wAMD may today maintain their vision and hence their autonomy and quality of life (Roque et al., 2021).
The treatment for wAMD demands a lot of resources, as it usually needs to be repeated often, typically every 4-12 weeks, sometimes for the rest of the patient's life.In 2020, almost 100 000 injections were administered in Sweden to over 20 000 patients, of whom 96% had wAMD.Compared to 2011, this represents a 160% increase in the number of patients and a 380% increase in the number of injections (Styrgruppen för Svenska Makularegistret, 2021).The registered drugs for wAMD are also considerably expensive, at around 6000 SEK (560€) per injection (Fass 2022), and on top this, there are considerable costs for hospital staff to manage the treatment (Hodgson et al., 2016).Besides direct costs of treatment, there are also indirect expenditures such as transportation and commitments of caregivers.The treatment could also be associated with anxiety and discomfort.It is therefore important to study if the potential increase in quality of life is worth the cost to get the treatment.
Treatment burden is an emerging concept in health care literature.Despite a shortage of a precise definition, it often means the actions and resources patients devote to their health care (Alsadah et al., 2020).In the context of wAMD, we interpret this as the psychological and physical efforts of the patients and caregivers to receive the treatment with special focus on the time.Earlier studies of injection treatment have analysed factors such as anxiety and discomfort before and during treatment, with the purpose to develop ways to deliver the treatment as gently as possible (Boyle et al., 2015).An average time per intravitreal injection appointment (including transportation but excluding recovery after the treatment) of 2.7 h has been reported in data from the USA.In 72% of cases, a caregiver drove the patient to the appointment, which had to take time off work in 22% (Prenner et al., 2015).Corresponding numbers of 4.4 h per intravitreal injection appointment have been reported from Australia, where indirect costs such as travelling and involvement of caregivers constituted onethird of the treatment costs (Spooner et al., 2018).A recent national study from Norway revealed that 38% of patients with wAMD needed caregiver assistance at the appointments and 58% felt anxious prior to the treatment (Reitan et al., 2023).No previous studies of the time per wAMD treatment within a European cohort have been published.However, a study including Germany, Italy and the UK has shown an average time of 4.5 h per intravitreal treatment for diabetic macular oedema or retinal vein occlusion (Sivaprasad & Oyetunde, 2016).
The aim of this study was to map the treatment burden for patients with wAMD in a Swedish setting.

| M ET HOD
This was a cross-sectional, non-interventional study conducted in 2021 at the Department of Ophthalmology, Höglandssjukhuset Eksjö, Sweden.Eksjö is a small municipality located in the southern part of Sweden with 12 000 residents in 2021 and the clinic is responsible for the wAMD treatment in an area of roughly 118 000 people (Statistics Sweden, 2021).Hence, the majority of the patients need to travel to the clinic.The patients completed the survey in direct connection to their treatment.It was conducted face-to-face with the patients, physician or assistant nurse and sometimes together with their caregiver.The questions were read out and explained to the patient so that the question was well defined for the patient.The survey did not contain any open-ended questions, all questions had pre-specified alternatives.
Recruitment of study participants was done at random among patients with active treatment.Patients with other diseases than wAMD, such as diabetic macular oedema (DME) or retinal vein occlusion (RVO), were excluded as these groups are considerably younger than wAMD, average age of 64 years for DME and 72 years for RVO compared to median 80 years for wAMD (Styrgruppen för Svenska Makularegistret, 2021).Patients with decreased cognitive capabilities (e.g.due to dementia) that were deemed unable to give a reliable informed consent and unable to give adequate answers to questions and were also excluded.All patient that were given treatment on the days that the study was conducted were given the opportunity to participate.The participants were not selected in any way to represent any demographic or other specific group of patients.With this strategy, the study population was considered to be reasonable representative of the Swedish population of patients with wAMD.
The survey included questions about time spent to receive treatment, caregiver assistance, way of transportation and self-rated vision (bilateral with correction).The patients were also asked about eventual negative experiences (discomfort, anxiety or transportation problems) associated with the actual or previous wAMD treatments.The questions concerning discomfort, anxiety and both in combination were formulated so that the answer was to represent only problems related to the injection procedure (before, during or after) and not the underlying disease (e.g.fear of losing vision or problems reading etc.) (Table S1).Information about current visual acuity, date of treatment start, number of treatments and current treatment interval was obtained from medical records.
Confidence intervals were calculated with z-tests for proportions and t-tests for mean values.The association between bilateral best-corrected vision and the self-reported vision on a five-grade scale was analysed with linear regression.A multivariate logistic regression analysis was performed to identify possible risk factors for negative experiences associated with the treatment.Findings were reported as odds ratios with a 95% confidence interval, and p-values calculated with the Wald test.The alpha level of p < 0.05 was always used.
The study participants received verbal and written information regarding the study and the type of information collected from the patient records.Ethical approval from the Swedish Ethical Review Authority had been obtained before the study commenced (reference number 2020-02753).The study adhered to the Declaration of Helsinki.

| R E SU LT S
The average age of the 93 patients was 79.9 years, and the proportion of women was 68%.The median number of previous treatments was 23 and the mean previous treatment duration was 3.3 years.The median best-corrected bilateral visual acuity (BCVA) measured in ETDRS and converted to Snellen was 0.77.Active bilateral treatment in both eyes was reported in 26%.Compared to patients with unilateral treatment only, these patients had a higher number of total injections and worse bilateral visual acuity (Table 1).
On average the spent 2.7 h (2.4-2.9:95% CI) to receive one treatment.A caregiver assisted the patient in 58%.Spouse (wife/husband) was the most common caregiver (46%) followed by son/daughter (20%) and friend (15%).The average time from home or work for the accompanying caregiver was 2.6 h (2.4-2.9:95% CI).One fifth of caregivers needed to take time from work.The most common way of transportation was with a caregiver (48%) followed by own car (26%) and hospital transportation (14%).The patients rated their vision as excellent or good in 41% and bad or very bad in 19%.The proportion of participants who experienced discomfort or anxiety was 22% and 14% respectively.The majority (91%) did not experience any problems travelling to the eye clinic (Table 2).
No strong association was found between the bilateral best-corrected vision and the self-reported vision on a five-grade scale in a linear regression model (β = 0.1, r 2 = 0.08, p < 0.05) (Figure 1).A multivariate logistic regression analysis showed a significant lower odds ratio for discomfort with higher self-rated vision and a significant higher odds ratio for discomfort with a longer treatment interval (Table 3).

| DI SC US SION
This study of patients with wAMD in Sweden showed that the treatment places a considerable burden on patients and their relatives regarding time.On the other hand, the majority of patients did not experience any problems travelling to the eye clinic.No strong association could be found between self-reported visual function and bilateral BCVA.
Only two factors were found to be associated with negative experiences.Firstly, the higher the self-rated vision, the lower the risk for discomfort.This could be a consequence of patients with lower vision being more afraid to lose autonomy caused by vision loss.Somewhat unexpectedly, a relation was not found between self-rated vision and anxiety.The other factor associated with discomfort was the treatment interval, with the surprising association that longer treatment intervals were associated with a higher risk of discomfort.A possible explanation is that patients with shorter intervals may be more used to the clinic and the treatment.It is also possible that short treatment intervals have a psychologically calming effect on patients and they have the feeling that they are more closely observed if anything bad were to happen.In future studies, it could therefore be interesting to evaluate if close examinations of patients with wAMD, even when treatment is not needed, would have the same effect.
Compared to previous research, the average time commitment per treatment of 2.7 h is identical to previous data from the USA (Prenner et al., 2015) but shorter than 4.4 h reported in Australia (Spooner et al., 2018) and is, of course, dependent on local demographics.The mean treatment interval of 7.3 weeks is also very similar to 7.6 weeks reported by Spooner.The assistance of a caregiver in 58% is lower than the 72% reported by Prenner but higher than 38% reported by Reitan, which had a comparable younger population, probably due to the voluntary survey participation online platform.As our study was conducted during 2021, we can further not exclude an influence of the SARS-CoV-2 pandemic.For the same reason, the number of patients who went with public transportation may be lower than before the pandemic.
The median bilateral BCVA of Snellen 0.77 is relatively high and makes it possible for as many as 26% of the patients to drive to the clinic themselves, as the minimum BCVA to drive in Sweden is Snellen 0.5 (TSFS, 2010:125).In our study, one patient drove themselves even though the bilateral BCVA was only 0.4.From a traffic safety perspective, it is therefore also important to discuss this issue with patients with treatment for wAMD.
Individuals with both high and low best-corrected bilateral vision had both good and bad self-reported vision and could not be fitted into a linear regression model, a finding that reflects the multidimensional aspect of self-reported vision.A previous study of the relationship between self-rated visual function and visual acuity showed a strong relationship but also used the validated NEI VFQ-25 questionnaire instead of a five-grade scale (Revicki et al., 2010).
To the best of our knowledge, this is the first study of time burden and experience in patients with treatment for wAMD in Sweden.The study size is comparable to similar studies conducted in the USA and Australia.However, the study also had important weaknesses.As the interviews were performed by clinicians also responsible for the treatment, some social desirability bias cannot be ruled out.Besides, due to an update in the questionnaire, only the last 65 of the 93 participants have a registered way of transportation to the treatment.Since this is still a majority of participants, the ratio of way of transportation is not expected to change dramatically.The exclusion of patients with cognitive impairment also means that we possibly excluded the patients with the biggest need for caregiver or healthcare personnel participation.However, this group of patients is small and may therefore not have given any major difference to the outcome.
Anti-VEGF treatment has revolutionized the prognosis for wAMD but also implies a significant treatment burden.On the other hand, the majority of patients do not experience problems with the treatment.Future medications or developed administration of existing medication could potentially decrease the number of treatments.This would decrease the burden for both patients and their relatives.

AC K NO W L E DGE M E N T S
This study was funded by Futurum-Akademin för vård och hälsa Region Jönköpings län and Edwin Jordans stiftelse för oftalmologisk forskning (2021-00024).

R E F E R E NC E S
Alsadah, A., van Merode, T., Alshammari, R. & Kleijnen, J. (2020) A systematic literature review looking for the definition of treatment burden.Heliyon, 6, e03641.F I G U R E 1 Bilateral BCVA in relation to self-rated vision with linear regression trend line.T A B L E 3 Multivariate logistic regression models of factors of meaning in regard to transportation to injection treatment.
Background data for patients receiving treatment in this study.[Correction added on 25 October 2023, after first online publication: The unit of Treatment interval was corrected in this version.]Results of questionnaire regarding current vision conditions and regarding transportation to the treatment.
T A B L E 1