1. Presenter: Elizabeth Lumley RGN/Research Associate
Examining the relevance of PROMs to patients: A systematic review of qualitative research capturing which QoL domains are important to patients.
Presenter: Elizabeth Lumley
RGN/Research Associate
This presentation presents independent research funded by the National Institute for Health Research (NIHR) under the Programme Grants for Applied Research programme (RPPG ). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

2. Background Part of NIHR Programme Grant (RP-PG-1210-12009).
Relates to other work including the identification and evaluation of existing PROMs for vascular conditions.
A primary qualitative research study examining QoL issues.
Integral part of the development of a new PROM, an electronic patient questionnaire for vascular patients (ePAQ-VAS).
The review and synthesis of qualitative research discussed in this presentation forms part of an NIHR Programme Grant, and is an integral part of the development of a new vascular PROM.
It relates to other work-streams including the identification and evaluation of existing PROMs for vascular conditions, and a qualitative study examining the impact of vascular conditions on quality of life.

3. Patient Reported Outcome Measures (PROMs)
PROMs should include domains that are relevant to patients.
FDA (2009) advises that measures show evidence that their items have been generated through taking into account the experiences and perspective of the patient group.
The use of qualitative research methods is the best way of examining patients viewpoints.
As PROMs allow measurement of outcomes elicited from patients; they should therefore include the items or domains that are the most relevant ones to patients to ensure their content validity.
After examining the development of PROMs the US Food and Drug Administration advised that developers should ensure that items have been generated taking into account the experiences and perspectives of the patient group it will be used with.

4. Qualitative Research
Qualitative research can be used to understand and describe patients’ experiences of their condition, and any impact it may have on their quality of life.
Qualitative research methods, such as interviews and observation, often produce rich data and intricate details, that would be missed using quantitative methods such as questionnaires.
Stand alone primary qualitative studies are common in healthcare research, however sample sizes are often small, with studies conducted in one location, which makes the results hard to generalise to a wider population.
One method of establishing if the findings of one qualitative study, are similar to the findings of any other, is to conduct a review of existing qualitative research and synthesise the results.
This enables researchers to establish if themes are typical solely of one study, or if they are common across a specific condition.

5. Aim of Evidence Synthesis
Primary aim of this qualitative evidence synthesis was to examine the symptoms and QoL domains that are important from the perspective of a patient with varicose veins (VV).
Secondary aim was to compare the findings to existing PROMs domains that are currently used with VV patients.
This presentation will demonstrate how a qualitative review can be used to examine the experiences that are common within a condition group, specifically varicose veins in this example, and will show how the findings can be used to evaluate whether or not existing PROMs capture the impact varicose veins have on patients.
The primary aim….

6. Varicose Veins (VV)
Relatively common affecting at least a third of the UK population.
Wide range of reported symptoms, and severity can differ.
Treatment can vary from compression hosiery through to invasive surgical procedures.
Varicose veins are common in the UK, and although relatively few people develop serious complications they are considered a chronic condition with implications for quality of life.

7. PROMs data collection in the NHS
Since 2009 NHS providers have collected PROM data for four surgical procedures, including VV treatments.
PROM completion rates for VV patients are much lower than the other procedures.
2013/14 VV completion rate was 40.5% compared to 86% and 94% for hip and knee surgeries respectively.
Since 2009 NHS providers have been required to ask patients undergoing hip replacements, knee replacements, hernia repair and VV surgery to complete a PROM.
Participation is voluntary and patients are under no obligation to take part.
PROM completion rates for VV patients are considerably lower than for the other procedures.
There does not appear to be any established reason why VV response rates are so poor comparatively although one explanation could be that the PROMs being used do not contain items that are relevant to VV patients.

8. VV PROMs used in the NHS
Patients complete the condition specific Aberdeen Varicose Vein Questionnaire (AVVQ).
Also complete two generic measures – the EQ-5D Index and the EQ- Visual Analogue Scale (EQ VAS).
A separate systematic review concluded the most appropriate PROMs were the AVVQ and the SF-36 based on psychometric analysis.
As part of the Vascular Programme Grant another systematic review was conducted to identify PROMS that have undergone some kind of published validation with VV patients, and assess the psychometric properties .
This review identified one generic, the SF-36, and three condition specific PROMs validated in VV patients. The condition specific ones were the Aberdeen VV Questionnaire, the VV Symptom Questionnaire and the Specific Quality of Life and Outcome Response-Venous SQOR-V.
The review concluded based on their psychometric properties the that most appropriate generic and condition specific PROMs were the SF-36 and the AVVQ.
A major limitation noted in the review was the lack of content validity of the PROMs, with items developed using expert opinion rather than patient experience.
The PROMs currently used with VV patients in the NHS are the EQ5D, and the condition specific Aberdeen VV Questionnaire. There is currently no published evidence that the EQ5D has been validated in the VV population.

9. VV Qualitative Review Eight electronic databases were searched.
1804 citations were identified; after screening only three studies met the inclusion criteria.
Three included studies – Palfreyman et al 2004, Hudson et al 2015, Franz & Wann-Hansson 2015.
So to move on to the qualitative review we conducted – eight electronic databases were searched to identify qualitative research, published in English, of the experiences of adults with VV.
A total of 1804 citations were identified; after screening only three studies met the inclusion criteria.
Thematic analysis was conducted and resulting themes were compared to items in validated VV PROMs.
This approach was taken in order to evaluate if these PROMs capture the impact that VV have on QoL as identified by the findings of the qualitative review.
(lack of qual studies, lack of reporting in mixed methods studies and lack of quality.)

10. Findings of the Review
Physical Impact
Psych Impact
Social Impact
Adapting
Reasons for Treatment
Five overarching themes were identified; physical impact, psychological impact, social impact, adapting to VV and reasons for seeking treatment.
Themes
In our synthesis we identified five key themes of physical impact on Qol, psychological impact on Qol, social impact on quality of life, ‘reasons for seeking treatment’ and ‘adaptation’.
There is not scope to discuss the full findings of the review in great detail, however the main themes will briefly be discussed.

11. Physical Impact No definitive list of symptoms.
“The last few days I literally crawled through the door and had to sit down with my feet up high to get the blood and swelling down.” (Patient 14 – Hudson et al 2015)
No definitive list of symptoms.
Not all symptoms affect people in the same way.
Generic PROMs won’t capture specific symptoms, such as swelling.
The range of physical symptoms reported across the three papers was varied.
The subjective nature of symptom experience has implications for the use of PROMS, as specific VV symptoms may not be included.
Or there may symptoms, such as itching, that are included in some condition specific PROMS, but that are not typical to all patients.
NICE guidance for the management of VV actually states that most QoL measures are unlikely to reflect severity of disease because of variations in perception of symptoms.

12. Psychological Impact Identified a strong psychological impact.
“I was just very aware that my legs shouldn’t be like this and that if they were more normal and less ugly and swollen, then I would feel a bit happier.”
Patient 2 – Hudson et al
Identified a strong psychological impact.
Cosmetic appearance was a significant finding.
Generic measures don’t capture ‘body- image’ concerns.
The psychological impact of living with VV appeared strongly in this review, with all three papers identifying a significant impact.
This demonstrates that any PROM used should include items that would enable patients to report psychological, emotional or mental impacts.
The significance of the cosmetic appearance of VV has perhaps historically been trivialised, however this review has found that appearance was a significant issue for patients.
In the NICE vascular referral guidelines ‘body –image concerns’ is included as an important measurable outcome of patient-assessed symptoms.
‘Appearance’ is an included domain in two condition specific measures however it would not be captured at all by a generic measure.

13. (Participant 11 – Palfreyman et al 2004)
Social Impact
All papers identified an impact on social function – often led to social isolation.
Link to psychological impact as social isolation led to low mood.
Social activities are included in most PROMs.
In all three papers social function was clearly found to be affected, often leading to varying degrees of social isolation.
The impact of some VV symptoms resulted in a number of participants being unable to undertake usual leisure activities and hobbies.
For others it was down to avoidance due to a reluctance to expose their legs.
Impact of VV on social activities does appear generally to be measured by both generic and VV condition specific PROMs.
“I don’t feel comfortable I feel like a lot of people stare at them or feel repulsed by them.”
(Participant 11 – Palfreyman et al 2004)

14. Reasons for Seeking Treatment
Symptom relief rather than cosmetic appearance was the primary reason.
Improvement in symptoms may be an unmet expectation.
Patients may be unaware of risk of failure.
No PROM would capture why patients seek treatment.
“More than anything is that it won’t be as it is now, so that the pain factor, the heaviness, everything that goes with it hopefully will have gone.”
(Participant 14 – Palfreyman et al 2004)
Symptom relief rather than cosmetic appearance was the primary reason for seeking treatment.
Current VV PROMs do not include any way of patients highlighting their reasons for seeking treatment, therefore it is difficult for clinicians to judge whether or not patients have realistic treatment expectations.

15. Adaptation Adaptation was a key theme to emerge.
Participants adopted a wide range of adaptations in order to continue to live their lives as fully as possible.
This was seen in all aspects regarding the impact of VV on QoL.
No available PROM, validated for VV populations, currently exists that would fully capture how VV patients adapt their lives to their condition.
Two condition specific PROMs do ask if clothing choice is affected by VV, and of these, one also asks if choice of activities is influenced by VV; however this does not comprehensively cover how patients ‘adapt’ or employ ‘coping strategies’ to live their lives fully.
No generic PROM would capture this aspect of having VV.

16. Construction of PROMs
The construction of PROMs is often based on clinical or researcher expertise, rather than patient experience.
Only one condition specific PROM, the VVSymptomQuestionnaire, incorporated qualitative patient based findings into development.
The Aberdeen VV Questionnaire items were based on common clinical assessment questions of VV patients, and the Specific Quality of Life and Outcome Response-Venous SQOR-V used a committee of ‘experts’ to review existing questionnaires and analyse relevant literature.
The developers of one of the VV specific measures that didn’t consult patients actually stated that their aim was, to develop a questionnaire that considered the patients main concerns – how they accomplished this without consulting a single patient about the construction of items is interesting.
The developers also concluded that ‘content validity could be inferred as questions were created for clarity and completeness by a group of experts in the field’ – no doubt far more expert than any patient experience.

17. Conclusion
“HRQoL, patient-assessed symptoms (including pain, discomfort, body-image concerns, swelling, aching and heaviness), and progression……were considered to be the most important outcomes to identify who would benefit from a referral to a vascular service” NICE 2013
Health-related quality of life, patient-assessed symptoms (including pain, discomfort, body image concerns, swelling, aching, and heaviness), and progression have been cited as the most important outcomes to identify which people would benefit from a referral to a vascular service (NICE 2013).
The use of PROMS to gather such information is well established in the NHS but as this review has shown many PROMs currently used with VV patients, may not capture the full extent of the impact on patient’s QoL.
Familiarity with the findings of existing qualitative literature, through qualitative reviews, or employment of qualitative research methods such as interviews, enables an in-depth understanding of the range and severity of symptoms experienced by patients.
Qualitative research methods should be used in the development of future PROMs to ensure that they reflect the most relevant issues patients face, therefore enhancing the content validity of such measures.

18. Co-authors and Collaborators
Patrick Phillips
Rosie Duncan
Helen Buckley-Woods
Ahmed Aber
Stephen Radley
Georgina Jones
Jonathan Michaels