Gathering Patient Views

IMPROVING THE PATIENT’S EXPERIENCE 3

a review of patient feedback processes in THE nhs IN wales

Marcus Longley and Nick Gould

Welsh Institute for Health and Social Care | University of Glamorgan

December 2009

purpose

This paper reports on a recent exercise to describe the range and nature of processes routinely undertaken by NHS Wales to understand how patients’ experience healthcare, and their priorities for improvement. It is part of a series of three papers on the patient’s experience, the others being:

• Paper 1: Patients’ Priorities for the NHS in Wales reports the results of a consultation exercise with patients carried out across Wales in May 2009, which focused on top priorities for improvement from a patient’s perspective
• Paper 2: Common Themes from the Literature summarises some of the themes which have emerged from various other recent attempts to find out patients’ priorities for improvement.

GATHERING PATIENT VIEWS – part 1

OVERVIEW OF PART 1

This part of the research report is based on the analysis of data submitted by health and healthcare related organisations in Wales during February/March 2009. NHS Trusts, Local Health Boards, Community Health Councils and other interested ‘entities’, e.g. networks, were requested to supply data on their recent activity concerning patient’s views. Details of data collection are recorded below in the section on methodology.

The following account stems from the processing and analysis of data submitted in respect to the first prompt of the data collection tool:

“How have you gathered patients’ views of services during 2007 and 2008?”

KEY FINDINGS

Purpose: findings

‘Gaining views’ is the most frequent purpose for requesting data from

the public.

‘Gaining views’ seems to be the default or all-purpose term favoured

by all health(care) agencies in Wales in relation to the reason for

gathering data from the public.

Typically, audit / monitoring / review represents roughly a third of all

activity in Wales connected to gathering data from the public on

health(care).

Consultation is a comparatively small part of the ‘citizen engagement

portfolio’ in NHS Wales.

For NHS Trusts, consultation represent under 5% of citizen engagement activity measured by frequency.

Respondent / Participants: findings

The most frequently used ‘audience’ to which requests for data and information are directed is users / patients.

Views are also sought from carers and age-based groups relatively frequently.

There appears to be little evidence that the views of individuals are sought on the basis of ethnicity. As a corollary, Welsh language issues perhaps need consideration.

The NHS in Wales has contact with at least 50 permanent groups from which it gathers views.

Data and information was gathered from groups on 145 occasions during 2007 and 2008.

Comparison between Table 2 and Table 3 indicates a bias towards gathering data from individuals rather than groups – with possible implications for hearing, and listening to, voices in person.

Methods: findings

The NHS in Wales uses a wide variety of methods to canvas the views of citizens.

In respect to surveys / questionnaires, LHBs, CHCs and Others have roughly the same levels of activity relative to other methods.

NHS Trusts use impersonal methods for roughly 60% of related activity.

LHBs and Others make proportionately high use of group interactions.

Substantive issues: findings

In relative terms, there is measurable activity in relation to the three core issues of: citizen’s influencing decisions; service delivery; and specific clinical issues.

In relative terms, the distribution of frequencies within and between the NHS bodies is in line with expectations due to their core functions.

Trusts are most interested in the patients’ views on clinical matters.

LHBs are most interested in the people’s views on service delivery issues.

There is some evidence to support the view that the NHS in Wales is interested in developing infrastructure to support citizen engagement – notably LHBs and Others.

RESULTS & DISCUSSION

The results are presented in the following way.

-For each of the five main categories generated, there is a page of graphics summarising and representing the results.

-In each case there is an accompanying explanation of the main category and its sub-categories.

-For each category, there is a discussion in the form of interpretation, observations and summary findings.

Purpose - explanation

Why are health(care) agencies asking people for information?

One way of looking at this question is to investigate core intentions. For example: ‘Was feedback being sought?’ Or: “Was this a consultation?”

(In the section on ‘substantive issues’, we respond to this question in a different way.)

‘Purpose’ is one of the original headings of the data response tool. Data processing and analysis suggested that ‘patients’ views’ were being gathered essentially in respect to a time orientation: past, present and future. That is, a commonsense approach to the purpose of gathering patients’ views. Either patients were being asked their views:

-about what had happened, e.g. audit;

-or, what was currently happening, e.g. mystery patients;

-or, what should happen in the future, e.g. service re-design.

In order to achieve time continuity and restrict the number of sub-categories, ‘purpose’ is facetted by the sub-categories shown:

-current / forward looking;

-looking back /current;

-and, past / present / future.

On reading ‘Graphics 1’, it is to be hoped that the reader agrees with the logic of this schema and its correspondence with reality as indicated by the coded data responses. For example, respondents used the term ‘consultation’ and the reality corresponding to the term has a current / future orientation.

However, the sub-category past / present / future - and the codings it contains – remains somewhat problematic. This sub-category is in many ways a ‘catch-all’. Whereas ‘audit’ has a reasonably sound definition, the same cannot be said of ‘views’. What exactly are ‘views’? The research answer is to group ‘views’ in the catch-all sub-category of ‘past / present / future’. ‘Awareness’ was classified likewise.

To reiterate aspects of the methodology: Table 1 provides an optimum structure for categorising data recorded under ‘Purpose’.

Purpose - discussion

When discussing the results for purpose, it needs to be kept in mind that frequency of types of activity is being considered. Behind each data response item, there is likely to be a significant story about resources. For example, the resource implications for an extended consultation compared to a ward survey of patients’ views are likely to be significantly different. Hence, when interpreting results, it is vital to remember that no weighting – for example, using resource inputs - has been assigned to processed data.

Views

Visually, blue is the most significant colour of the bar chart, reflecting the distribution of numbers in the table.

Gaining patients’ views is the most frequent purpose for contacting members of the public.

As noted above, this type of purpose may well reflect a general or catch-all approach to the reason for contacting patients for their opinions.

Looking back / current

For Trusts, LHBs and CHCs this type of purpose represents relatively significant activity. Audit / monitoring / review features prominently.

Consultations

Looking across the bars, the relative sizes of the segments representing consultation activity is notable.

For LHBs, Trusts and CHCs, consultations represent a comparatively minor part of their ‘purpose portfolio’.

For NHS Trusts, consultation is the least favoured purpose.

Summary Findings 1

SF1.1 – ‘Gaining views’ is the most frequent purpose for requesting data from

the public.

SF1.2 – ‘Gaining views’ seems to be the default or all-purpose term favoured

by all health(care) agencies in Wales in relation to the reason for

gathering data from the public.

SF1.3 – Typically, audit / monitoring / review represents roughly a third of all

activity in Wales connected to gathering data from the public on

health(care).

SF1.4 – Consultation is a comparatively small part of the ‘citizen engagement

portfolio’ in NHS Wales.

SF1.5 – For NHS Trusts, consultation represent under 5% of citizen

engagement activity measured by frequency.

Respondents / participants (1) - explanation

From whom do health(care) agencies get information?

In this section, we look people contacted by health(care) agencies on the basis of their individual characteristics, for example, age or residency. This approach to contacting people may well be based on, for example, the specific requirements of a survey related to the healthcare needs of younger people in rural areas where a wide cross-section of views is sought.

Constructing the sub-categories was straightforward; standard demographic data for the most part. ‘Users / patients’ presented the only categorisation dilemma, with the final choice of this term echoing the selections made by this survey’s respondents.

Respondents / participants (1) - discussion

In this case, the graphics require minimal overall interpretation and we proceed directly to the discussion.

Users / patients

Obviously, users / patients is the dominant category.

Age and carers

Both these categories have relatively high frequencies. The patterning within the age category is worth noting due to the high frequency counts for children/ young people/ students. The low count for adult may be accounted for due to un-processed data pre-assigned to other categories, for example, ‘public general’.

Ethnicity

Only 1 patient / public data gathering exercise related to Black and Minority Ethnic (BME) groups appears to have taken place in the two year period being researched. There may well have been other exercises in which ethnicity is recorded and BME data is processed and analysed a sub-set – but the fact remains of only 1 formally recorded exercise. [But also see ‘Target groups’ below.]

Summary Findings 2

SF 2.1 – The most frequently used ‘audience’ to which requests for data and

information are directed is users / patients.

SF 2.2 – Views are also sought from carers and age-based groups relatively

frequently.

SF 2.3 – There appears to be little evidence that the views of individuals are

sought on the basis of ethnicity. As a corollary, Welsh language

issues perhaps need consideration.

Respondents / participants (2) - explanation

From whom do health(care) agencies get information?

This question repeats that at the beginning of the previous section. Clearly, it needs a different kind of response.

This section is concerned with the gathering of data from people on the basis that they belong to groups with a shared interest or a cohort in a study over a period of time. Immediately, the issue arises: aren’t the categories in the preceding section also groups? Except for the case of ‘expert patients’, probably not – expert patients were included in Table 2 on the basis of giving a easily noticed comparison to the related category of users / patients. If the expert patients of Table 2 meet regularly as a group then certainly they would feature in this section. But, another reason for categorising expert patients in Table 2 is that from the data returns for this particular research project, expert patients appeared to be approached as individuals.

Why does any of this matter?

Without getting bogged down in the technicalities of research design, it matters because, for instance, the responses given by people in a focus group will be different to their responses to a telephone questionnaire. Essentially, respondents / participants have been considered in two distinct sections because whoever was conducting the research to gather the views had two research strategies in mind. Either, they were seeking views from a random sample of the population or they wanted to views of a known group.

So, the categories shown in Table 3 tend to reflect that the research participants meet with each other face to face – and they probably do that regularly. However, meeting face-to-face probably does not apply to ‘panels’ – the members of which are probably selected randomly in the first instance. Panels were placed in this section on the basis of the on-going data collection involved. Contrast this fact with the one-off data collection which also typifies Table 2.

So as to give greater insight into what is involved in the classificatory process, it should be useful to record all the instances that were eventual categorised in Table 3.

This long list deserves a comment (below).

Finally, ‘target groups’ and ‘focus groups’ require a few words of explanation.

The membership of target groups may well be individuals selected randomly on the basis of some demographic or other characteristic. Some BME target groups might feature here – but there was no means of knowing.

As for focus groups – also recorded in the Table 4. Again, there was no means of knowing to what extent these groups were convened just for the purpose of data collection or whether they were drawn from individuals already known to each other.

Respondents / participants (2) - discussion

The number of types of group recorded in List 1 is a useful indicator for the variety of different audiences to which the NHS in Wales listens.

Permanent groups

In this category, we have included groups such as forums and patient groups – characterised by people known to each other and who meet regularly.

The concern here is very much with hearing very directly the voice of patients, users and carers on health and healthcare issues that directly concern them. For these groups, the intention is that the NHS should hear very directly and very personally about service planning, organisational and delivery issues.

There are 67 instances of such groups being contacted for their views over a two year period. Focus groups accounted for 39 instances in this period – giving a total of 106 face-to-face events. However, in the same period the total of instances of contacting individuals is 195 (Table 2) – where, generally, the methods used in Table 2 will be impersonal, e.g. postal survey. Crudely, this could be interpreted as the NHS preferring to gain the views of citizens by at arms-length. However, as will be shown in the next section, this observation tends to hold for NHS Trusts but much less so for LHBs.

Summary Findings 3

SF3.1 – The NHS in Wales has contact with at least 50 permanent groups

from which it gathers views.

SF3.2 – Data and information was gathered from groups on 145 occasions

during 2007 and 2008.

SF3.3 – Comparison between Table 2 and Table 3 indicates a bias towards

gathering data from individuals rather than groups – with possible

implications for hearing, and listening to, voices in person.

Methods - explanation

By what means does the NHS obtain the views of citizens?

The extent of Table 4 indicates the difficulties in finding a small number of categories into which coded data could be slotted.

Chart 4 represents data arising from merging ‘surveys’ and ‘questionnaires’ so as to enable a representation open to sensible representation.

Overall, the decision was taken toadhere to the large variety of descriptors used in the data response sheets. So, for example, in Table 4 ‘survey’ and ‘questionnaires’ have been distinguished form each other – though, in reality, they might well be the same.

Methods - discussion

The graphics for ‘method’ clearly display the variety of ways in which NHS bodies gain the views of citizens.

Comparing and contrasting methods both within and between LHBs, Trusts, CHCs and Others is instructive.

Although all four types of organisation show a variety of methods,surveys / questionnaires account for over half of Trust activity in this respect. This fact confirms the hypothesis that NHS Trusts tend to use impersonal methods of contacting patients, users and publics.

LHBs and Others have proportionately high levels of ‘group interactions’ referenced both internally and externally.

Summary findings 4

SF4.1 – The NHS in Wales uses a wide variety of methods to canvas the

views of citizens.

SF4.2 – In respect to surveys / questionnaires, LHBs, CHCs and Others have

roughly the same levels of activity relative to other methods.

SF4.3 – NHS Trusts use impersonal methods for roughly 60% of related

activity.

SF4.4 – LHBs and Others make proportionately high use of group

interactions.

Substantive issues - explanation

On what issues does the NHS ask for opinions?

The generic sub-categories require little in the way of explanation. They were also relatively straightforward to construct. Suffice to say:

-information provision refers to activities related to the quality and usability of information provided by the NHS. For example, the ‘readers panels’ recorded previously will scrutinise NHS documents for ‘simple language’;

-communications issues are more general than their information counterparts and partly overlap with citizen engagement issues.

However, as considered in a note in the methodology section, specific substantive issues with the sub-categories of service delivery and clinical proved problematic. These two categories were invoked to accommodate the recording by informants of what were at least two types of activity – though precisely where the dividing line should be drawn remains uncertain. A glance at Chart 5 – and noting the relative proportions of ‘specific’ in LHBs, Trusts and CHCs – suggests there may be a degree of validity and reliability in the sub-categories of clinical and service delivery. The expectation is for Trusts to have a greater focus on purely clinical matters than service delivery – and vice-versa for LHBs.