Patient-reported outcome and experience measures (POEM) of a community-based glaucoma clinic in Cambridge, UK: an observational study

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • Limited previous research exists on patient’s experiences of community settings and how this impacts patient’s perceptions of their glaucoma.

  • Data was gathered using a patient-reported outcome and experience measure designed specifically for glaucoma.

  • Data were collected in one community hub and may not be representative for the whole UK.

  • Only patients with lower risk scores are referred to community-based clinics and the experience of those with higher risk scores may be different.

Introduction

Primary open-angle glaucoma (POAG) affects approximately 2% of people over 40 years within the UK1 with the prevalence increasing as people get older.2 Due to an increase in ageing population, the demand for glaucoma care is increasing globally. The Royal College of Ophthalmologists, The Way Forward project predicts that from 2015 to 2035, the number of people in the UK with glaucoma will increase by 44%, ocular hypertension by 16% and those identified as glaucoma suspects by 18%.3 The NHS Ophthalmology Transforming Elective Care Services have made several recommendations of how to meet demand, including developing and implementing community-based clinics.4

Community-based glaucoma clinics are a well-established pathway for the delivery of glaucoma care. They can involve virtual clinics or a shared care pathway. There are different types of community-based clinics for glaucoma. In virtual clinics, technicians work under the supervision of a consultant ophthalmologist (who is present or not at the site), collecting patients’ quantifiable diagnostic information following a standard protocol.5 The diagnostic information is then reviewed remotely by a secondary care specialist optometrist or consultant ophthalmologist to decide on treatment, ongoing management options or referral to a consultant-led clinic.6 7 Within the shared care pathway, patients triaged as a low-risk are referred to the community-based clinic. Clinical assessment and management are performed by a primary care optometrist with a special interest in glaucoma under remote consultant supervision.8 In addition, there are types of remote community-based clinics in which patients are seen in the community by hospital doctors or optometrists with no virtual element to it. Community-based clinics enable secondary care clinicians to review more cases and maximise appointment capacity9 to improve sustainability of the system.4 Patients benefit from a reduced journey time through the service,10 and access to glaucoma care and advice closer to home in a community setting.4

It is unknown how many community clinics exist in the UK. In 2018, 50% (n=42) of all the clinical lead ophthalmologists surveyed in the UK ran a virtual glaucoma clinic, operating from within their ophthalmology units with a further 21% reported to have plans to establish one in the near future.8 The Ophthalmology Getting It Right First Time (GIRFT) programme national specialty report in 2019 visited 120 Trusts and recommended providers to introduce virtual clinics for glaucoma within 1 year;11 however, there is no updated data of how many trusts followed this recommendation. Current virtual glaucoma clinics are designed for low-risk patients with early glaucoma, ocular hypertension and glaucoma suspect patients;12 however, a recent study has investigated whether virtual glaucoma clinics eligibility criteria could be expanded to include other disease stages and glaucoma phenotypes and found the care to still be safe, effective and achieving high levels of patient’s satisfaction.12 Increasing the capacity of virtual clinics by including a wider patient eligibility criterion could help to provide a long-term strategy to the increasing demand for glaucoma care.

Several studies have investigated the quality of care received in community-based glaucoma clinics in comparison to hospital-based clinics. It has been shown community optometrists perform as reliably as Hospital Eye Service glaucoma review when using specified measurement techniques,13 and specialist optometrists in glaucoma demonstrate acceptable levels of accuracy with a glaucoma consultant.7 As well as clinical accuracy, patients’ acceptability of community-based clinics is important. One previous study found satisfaction rates between patients attending a specialist optometrist community-based glaucoma clinic and a doctor-led hospital clinic were comparable for the information given to patients (77% community-based, 69% HES, rated good/excellent), answering questions (95% community-based, 83% HES, rated good/excellent) and the clinician overall (75% community-based, 73% HES, rated good/excellent).7 Few studies have investigated patient’s experiences of community-based glaucoma clinics and how this affects patient’s outcomes and perceptions of their disease. The aim of this observational study was to investigate this using a modified glaucoma patient-reported outcome and experience measures (POEM) instrument.14 15 The modified glaucoma POEM is a novel concise eight-item questionnaire designed to capture patient’s perceptions of their glaucoma experience, and outcomes as a measure of quality in healthcare.15 Within Australian private practice, the POEM questionnaire found younger patients were more worried about losing their vision and felt glaucoma interfered more with their daily lives. Female patients were found to more strongly agree that they understood their glaucoma diagnosis and management than male patients.15

Methods

Design and setting

This is an observational study. Ninety-six consecutive patients were recruited attending a community-based glaucoma clinic at the Vision and Eye Research Institute in Cambridge from July to September 2022.

Population and sample size

Three months was chosen as the fixed study period as this was determined as sufficient to obtain a meaningful sample size based on previous research.16 17 Patients had the option of completing the survey independently or to complete the survey with the researcher. This was to eliminate the potential of glaucoma severity impacting the ability to respond to questions and record responses. Seven patients did not take part as they were outside the inclusion criteria (Alzheimer’s/dementia (n=4), arthritis (n=1), stroke (n=1) and anxiety (n=1)). Five did not take part due to time limitations. Four participants did not wish to take part with no further explanation. Informed consent was obtained from all eligible patients. The study followed the Declaration of Helsinki and had ethical approval from Anglia Ruskin University’s School Research Ethics Panel (MED-SREP-21-011). Patients either had a confirmed diagnosis of glaucoma, were under investigation for suspect glaucoma, or had ocular hypertension.

Data collection

The eight-item Australian glaucoma POEM15 was used which has been slightly adapted from the six-item glaucoma POEM that was developed from the ‘National Glaucoma Think Tank’ event in the UK in 2012, a concise questionnaire that was developed to address areas on glaucoma experience and outcome of greatest significance. The questionnaire underwent minor modifications to include two extra items, a question regarding the understanding of how their condition is managed, and whether patients think their glaucoma care is well organised. The recording of responses was modified to include a Visual Analogue Scale (VAS) rather than an ordinal categorical answer selection. This enabled a quantitative analysis potential.15 The modified eight items in the Australian POEM consist of the understanding of diagnosis and management, acceptability of glaucoma treatment, whether they feel their glaucoma is getting worse, interference with daily life, worry of losing vision, feeling safe under the care of their clinical team and organisation of care. As not all the patients in our study had an official diagnosis of glaucoma, the word glaucoma was removed from items 3, 4, 6 and 7. Item four was adapted from ‘I do not think my glaucoma is getting worse’ to ‘I do not think my vision is getting worse’. Part A of figure 1 features the Australian POEM and part B features the modified Australian POEM we used during our study.

Figure 1
Figure 1

Part A shows the Australian glaucoma patient-reported outcome and experience measure (POEM) developed by Fraenkal et al,15 and part B shows the Australian glaucoma POEM adapted for this study.

Additional information regarding the patient’s gender, age, best corrected visual acuity (BCVA) in the worse eye and best eye on the day of survey completion, duration of diagnosis and number of years under hospital care were collected. Socioeconomic data were gathered for each patient including their postcode, type of secondary school attended and their highest qualification. Participants were also asked about their experience of using the community-based clinic including whether the clinic was easy to find, was the experience comfortable and if they felt the experience was the same as what would be expected from a hospital. The continuous patient variables were age, duration of disease and BCVA (Snellen converted to LogMAR) in the worst and best eye. The categorical variables were gender and diagnosis. Diagnosis was split into POAG, suspect glaucoma, ocular hypertension and other. The other category consisted of pigment glaucoma, uveitic glaucoma, angle closure glaucoma and patients at an increased risk of glaucoma (table 1). Patients responded on the VAS for each question. The VAS scale was 10 mm long with each 1 mm being one VAS unit. A score between 0 and 100 was reported. The scale was designed from strongly disagree to strongly agree; therefore, a score closer to 100 reflected a more positive response. Statistical analysis was performed using IBM SPSS Statistic Software (Version 28; SPSS Inc., Chicago, IL, USA).

Table 1

Patients’ demographics, socioeconomic information and disease characteristics (n=96)

Patient postcodes were used to find the corresponding Lower-Layer Super Output Areas (LSOAs) which are small areas of the UK controlled for population size. LSOAs have an average population size of 1500 and there are 375 LSOAs in Cambridgeshire. Socioeconomic status is determined using the Index of Multiple Deprivation (IMD) which ranks LSOAs from 1 to 10 (most deprived to least deprived). IMD is a measure of relative deprivation in England encompassing a wide range of individual living conditions including income, employment, health deprivation and disability, education, crime, barriers to housing and services and living environment. LSOAs within decile one are in the top 10% of socioeconomically deprived LSOAs in the UK. The patients’ socioeconomic and IMD decile data are shown in table 1.

Statistical analysis

Descriptive statistics were carried out followed by one-way analysis of variance tests to determine if there were any statistically significant differences between the demographic, socioeconomic and disease characteristics data and the VAS scores. A p value of <0.05 was considered as being statistically significant. For better understanding of the practical significance, the effect size of the ANOVA results were also reported using eta squared η2. The effect sizes were interpreted as small effect size 0.01–0.059, medium effect size 0.06–0.139 and large effect size >0.14. Further subgroups were analysed to investigate any effect of age, duration of disease, years under the Hospital Eye Service and diagnosis. Any missing data were excluded from the analysis.

Patient and public involvement

None.

Results

Ninety-six patients participated in the study and are included in the analysis. Fifty one per cent (n=49) were female and 49% (n=47) male. The mean and SD of participant age was 70 years±12 years (range 36–94). Majority of participants had POAG (55%, n=53), followed by suspect glaucoma (14%, n=13), ocular hypertension (13%, n=12) and other (19%, n=18). The mean and SD of duration of disease was 9.4 years±10.8 years. The descriptive statistics are shown in table 1.

Six out of eight questions showed strong positive responses indicating patients were positive regarding their perceived glaucoma experience and outcome. These were question 1 (‘I understand my diagnosis’), question 2 (‘I understand how my condition is managed’), question 3 (‘My treatment (and any side effects) are acceptable to me’), question 5 (‘my glaucoma does not interfere with daily life’), question 7 (‘I feel safe under the care of my clinical team’) and question 8 (‘my care is well organised’).

Question 4 (‘I do not think my vision is getting worse’) and question 6 (‘I am not worried about losing my vision’) showed lower VAS scores. This indicated patients had concerns regarding their vision getting worse and the loss of vision. The mean and SD of the VAS scores for each POEM question are shown in table 2.

Table 2

The mean and SD of the Visual Analogue Scale (VAS) score for each patient-reported outcome and experience measure (POEM) question

Age

The age of participants ranged between 36 and 94 years. The POEM VAS scores were compared across three age groups, those <60 years, 60 to 80 years and >80 years. A one-way ANOVA found a statistically significant difference between the age groups on their understanding of their diagnosis (F(2, 91)=[3.755], p=0.027). A medium effect size was found η2=0.076. Those <60 years scored the lowest with a mean VAS score of 79, in comparison to those 60–80 years with a mean VAS score of 90 and those aged >80 with a mean VAS score of 80. Post hoc comparisons found a significant difference between <60 years and 60–80 years (mean difference −10.97, SE 5.13, p=0.035) and between 60–80 years and >80 years (mean difference 9.91, SE 4.49, p=0.030). The mean VAS score and 95% CI for question 1 in each age group are shown in figure 2.

Figure 2
Figure 2

(A) Graph showing the mean Visual Analogue Scale (VAS) score and 95% CI for question 1: I understand my diagnosis for those aged <60 (n=17), 60–80 (n=56) and >80 years (n=23). (B) Graph showing the mean VAS score and 95% CI for question 2: I understand how my condition is managed for those with a disease duration of ≤5 years (n=44) and those >5 years (n=48). (C) Graph showing the mean VAS score and 95% CI for question 5: My glaucoma does not interfere with my daily life for those how have been under the care of the hospital ≤10 years (n=74) and those >10 years (n=14). (D) Graph showing the mean VAS score and 95% CI for question 1: I understand my diagnosis for those diagnosed with primary open-angle glaucoma (POAG; n=51), suspect glaucoma (n=14), ocular hypertension (OH; n=11) and other (n=19). (E) Graph showing the mean VAS score and 95% CI for question 6: I am not worried about losing my vision and those diagnosed with POAG (n=51), suspect glaucoma (n=14), ocular hypertension (OH; n=11) and other (n=19). *Indicates a significant difference p<0.05.

Duration of disease

The duration of disease ranged from 1 month to 72 years. The POEM VAS scores were compared across two different durations, those with a duration of disease 5 years and under and those greater than 5 years. A statistically significant difference was found between patients understanding of how their condition is managed and the duration of disease. Those with a duration of 5 years or less were found to have a poorer understanding of how their condition was managed (F(1,90) = [5.076], p=0.027). A small effect size was found η2=0.053. The mean VAS score and 95% CI for question 2 and those with a disease duration ≤5 years and those >5 years are shown in figure 2.

Years under the Hospital Eye service

Years under the Hospital Eye Service ranged from 1 month to 72 years. The years were split into two different durations, 0 to 10 years and greater than 10 years. The interference on daily life was statistically greater in those patients who had been under the care of the hospital for longer than 10 years (mean VAS score, 65) in comparison to those who had been under the care of the hospital for less than 10 years (mean VAS score, 83) (F(1,86) = [6.453], p=0.013). A medium effect size was found η2=0.070. The mean VAS score and 95% CI are shown in figure 2.

Diagnosis

Diagnosis was split into four categories, POAG, suspect glacuoma, ocular hypertension and other. The other category consisted of patients with pigmentary glaucoma, uveitic glaucoma, angle closure glaucoma and those at risk for reasons such as congenital anomaly and Posser-Schlossman syndrome. The understanding of diagnosis was statistically different depending on the diagnosis (F(3,91) = [2.785], p=0.045). A medium effect size was found η2=0.085. Post hoc comparison found a significant difference between a diagnosis of suspect glaucoma (mean VAS, 72) and POAG (mean VAS, 88; mean difference −15.43, SE 5.62, p=0.044). The mean VAS score and 95% CI are shown in figure 2. A statistically significant difference was found between the fear of losing vision and diagnosis (F(3,91) = [3.394], p=0.021). Those diagnosed with other forms of glaucoma were found to be less worried about losing their vision (mean VAS, 65) in comparison to those diagnosed with POAG (mean VAS, 39; mean difference 26.84, SE 9.31, p=0.030). The mean VAS score and 95% CI are shown in figure 2.

BCVA in worse eye

BCVA in the worse eye was measured as Snellen acuity converted into LogMAR. Patients were split into two groups: those who had BCVA in the worse eye of ≤1.00 (≤6/60) and those who had BCVA in the worse eye of >1.00 (>6/60). No statistically significant differences were found between the level of vision and any of the POEM responses. Although no statistically significant differences were found, patients with more severe sight loss reported more positive perceptions of understanding diagnosis, understanding how the condition is managed, and whether their treatment and side effects were acceptable. Patients with more severe sight loss also reported more negative perceptions of feeling vision is getting worse, interference on daily life and worry about losing their vision. The level of vision did not impact the patients’ experience of feeling safe under the care of the clinical team or the organisation of care p>0.05.

BCVA in the better eye

All patients had better than 1.00 (6/60) vision in their better eye. Patients were split into two groups: those with BCVA ≤0.3 (≤6/12) and those with VA >0.3 (>6/12) as this is the minimum visual acuity required for driving.18 A statistically significant difference was found between the two groups and the understanding of how their condition is managed, and the impact on daily life. Those with BCVA in the better eye ≤6/12 were found to have a better understanding of how their condition is managed (mean VAS, 88) compared with those with VA >6/12 in the better eye (mean VAS, 73; F(1,73) = [4.147], p=0.045). A small effect size was found η2=0.054. The impact on daily life was found to be significantly less in those patients with VA ≤0.3 (≤6/12) in the better eye (mean VAS, 82) in comparison to those with VA >0.3 (>6/12) in the better eye (mean VAS, 55; F(1,73 = [4.805], p=0.032). A small effect size was found η2=0.016.

Gender

Although women’s average score was lower for being worried about losing vision (mean VAS, 41) compared with men (mean VAS, 53), there were no statistically significant differences in any of the POEM questions.

Socioeconomic status

Those who had attended a secondary school outside of the UK were found to feel less safe under the care of their clinical team (mean VAS, 73) compared with those educated at a UK state school (mean VAS, 93; F(3,83)=[4.494], p=0.006). A large effect size was found η2=0.140. Similarly, those who attended school outside of the UK felt that their care was less organised (mean VAS, 71) in comparison to those educated at a UK state school (mean VAS, 90; F(3,83)=[3.605], p=0.017). A medium effect size was found η2=0.117.

Patient’s experiences around the community-based clinic

Eighty-four per cent of participants said the community-based clinic based in the Vision and Eye Research Institute was easy to find. Ninety-six per cent of patients said their experience of attending the remote community-based clinic was comfortable. When patients were asked if they felt the experience was the same as what they would expect from the hospital, 93% said yes, 5% said not applicable (as they had not attended a hospital-based clinic) and only 2% said no. No demographic, socioeconomic or disease characteristics were found to impact the patient’s experience of attending the community-based clinic.

Discussion

The modified POEM was developed with a ‘patient-generated’ approach designed to gather glaucoma patient feedback in clinical practice, perceived outcomes and experiences.14 15 The tool is simple and short making it easy to use in clinical practice. The VAS scores for six out of the eight questions had a mean score of 80 or higher. Question 4 ‘I do not think my vision is getting worse’ and question 6 ‘I am not worried about losing vision’ had lower mean VAS scores, with the range of results suggesting that there was widespread variation around their concern about vision getting worse, and worried about losing their vision. The large range of responses for this question was influenced by the severity of glaucoma which varied greatly between patients with some having no perceived vision loss up to counting fingers. A cross-sectional survey of UK adults found sight to be the sense people feared losing the most which is reflected by the lower scores.19 Patients with BCVA >0.3 (>6/12) in the better eye were found to have a significantly poorer understanding of how their condition is managed and had a significant impact on their daily life. The type of glaucoma and the worry of losing vision were found to be significant (p=0.021) with differences between POAG patients and those with other types of glaucoma (pigmentary glaucoma, uveitic glaucoma, angle closure glaucoma and those at risk for reasons such as congenital anomaly and Posser-Schlossman syndrome). Those diagnosed in the other category were found to be less concerned about losing their vision. These results may also highlight that those patients with less common forms of glaucoma such as pigmentary glaucoma, uveitic glaucoma and angle closure glaucoma may lack understanding regarding their condition and potential impact on vision, in comparison to those diagnosed with POAG, requiring more active monitoring and consultation time which may contribute to easing their concern.

This study demonstrates that longer duration of registration of glaucoma diagnosis was significantly associated with more interference on daily life. Interestingly, relatively younger age (<60 years) was significantly associated with the poorest understanding of their diagnosis. In addition, those with lower disease duration (5 years or less) had poorer understanding of how their condition was managed in comparison to those with longer duration.

These results are new for the UK and are consistent with those found by a study in Norway exploring the feelings and experiences of patients living with glaucoma.20 They found patients under 60 years of age reported a lack of information given to them by their practitioner regarding their disease, the cause of glaucoma and the risk of going blind twice as often as those above 60 years of age, and reported more frequent feelings of anxiety, depression or fear of going blind when being diagnosed with the disease.20

Our results suggest that patients newly diagnosed with glaucoma may need more initial consultation time and educational materials to improve their understanding of their diagnosis, how their condition will be managed and the possible impact on their vision if untreated. Those who presented at the community clinics for suspect glaucoma reported a statistically significant poorer understanding of their diagnosis and how their condition is managed. It is therefore vital to ensure that patients understand their diagnosis is important in order to maximise appointment attendance and adherence to treatment plans, as it has been shown that patients who do not understand the likely impact of their long-term condition are more likely to abandon their treatment. It has previously been shown that patients with glaucoma and ocular hypertension have similar illness perceptions despite having a different prognosis.16 Separate information needs to be given to both glaucoma and ocular hypertension patients to highlight the differences in diagnosis and prognosis and reduce anxiety in those diagnosed with ocular hypertension.

The socioeconomic information gathered was decided from the Cabinet Office evaluation of 12 socioeconomic background measures and their accuracy.21 Socioeconomic status was not found to impact POEM responses or perceptions of the community-based glaucoma clinic even though 48% of the patients were in the top 20% of least deprived LSOAs in the UK.

Although the patient’s highest qualification and IMD decile did not influence any of the POEM question responses, patient’s perception of feeling safe under the care of their clinician team was lower in those educated outside of the UK as were their perceptions of their care being well organised (p=0.017). This may be due to their understanding of healthcare systems, cultural differences or language barriers.22

The responses of patients visiting the community-based clinic at the Vision and Eye Research Institute with regards to questions about feeling safe under the care of their clinical team, and whether their care was well organised were not impacted by the demographics, socioeconomic status or disease characteristics. Ninety-six per cent of patients said their experience was comfortable. Only 2% reported that their experience was not the same as the hospital. Fifteen per cent reported that the experience was better than at the hospital due to fewer people in the waiting area, quieter, shorter waiting times, better parking and easier access. A similar study at Manchester Royal Eye Hospital and Bristol Eye Hospital investigating patients’ experiences and perceptions of attending glaucoma virtual clinics were positive, with 93.5% reported that they would recommend the clinic to family and friends.23

Community-based clinics help to meet the significant demand in glaucoma care11 by maximising appointment capacity and reducing the time spent in clinic.24 These findings show these services were found to be acceptable to patients and are an important step forward to help deal with the increasing demand on Hospital Eye Services.3

The findings of this study are potentially limited as data were collected in Cambridge and are not representative for the whole UK. It is important to note that only patients with lower risk scores are referred to community-based clinics and the experience of those with higher risk scores may be quite different. Patients within community-based clinics are deemed to have stable glaucoma. Patients with severe eye diseases, comorbidities, unstable disease or developing any complications are seen within the Hospital Eye Service. The study gives an insight into community-based clinics and patient’s experiences, which may help in the development of commissioning community-based clinics in the future. Future studies should be developed to gather clinical data and severity of disease on patients to further investigate the impact on anxiety and fear of blindness.

Conclusion

POEM responses found younger, newly diagnosed and suspected glaucoma patients would benefit from improved information regarding their diagnosis and treatment. Those with secondary glaucoma would benefit from a greater understanding of their prognosis and potential impact on vision. Patients educated outside of the UK may require more information about community-based clinics to improve their perception of feeling safe under the care of their clinical team. Overall, this study shows patients had a positive response to community-based glaucoma clinic. Rather than attending outpatient Hospital Eye Services, patients received their glaucoma care within the community and remarked improved access, easier parking, shorter waiting times and a better experience than in the hospital.

Data availability statement

Data are available upon reasonable request. The data is available from the corresponding author upon reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by Ethical approval was granted from Anglia Ruskin University’s School Research Ethics Panel (MED-SREP-21-011). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors thank Majidah Khaliq for helping with data collection.

This post was originally published on https://bmjopen.bmj.com