The acceptability and impact of a randomised controlled trial of welfare rights advice accessed via primary health care: qualitative study

The protocol was approved by the Newcastle upon Tyne joint universities and NHS research ethics committee.

Twenty-five semi-structured, in-depth interviews were undertaken (14 intervention, 11 control). Ten participants were interviewed with partners who made active contributions. Fourteen participants received some financial award. The median weekly income gain was £57 (€84, $101) (range £10 (€15, $18) to £100 (€148, $178)) representing a 4–55% increase in household income. Eighteen participants were in receipt of benefits, either as a result of the current intervention or because of claims made prior to the study. Although 96 of the total sample (126) agreed to participate, data saturation was reached after interviewing 25 respondents.

Participation in the RCT required considerable commitment. All participants were interviewed four times over 24 months with the same structured interview schedule and the qualitative sub-sample participated in a further in-depth interview. Participation rates for the RCT remained high (87%) after two years (five deaths; one moved, 11 declined follow-up).

The research process was a positive experience overall (table 2). Factors which contributed to this included: enjoying the company of others; learning through participation; the domiciliary nature of the intervention and research; trust in the institutions involved (NHS and University); and building a relationship with the researchers. No one was concerned about the time commitment, divulging personal information or breaches of confidentiality.

During the qualitative interview, participants re-read and commented on the participant information leaflet. There was unanimous agreement that it was clear, straightforward and also reflected their experiences of being in the study. However, it became apparent that a number of people had agreed to take part in the study without being fully aware of what it entailed. Some participants disclosed that they had not realised there was the potential for gaining additional resources, despite this being clearly stated during invitation and consent. The main reasons given for taking part were: altruism – helping others and highly regarded institutions like the NHS; importance of research for generating new knowledge; and, finding out about welfare benefit entitlements (table 3). Their GP's signature on the invitation letter was another reason given for participation, as it conferred a degree of legitimacy on the research. Participants also felt they were reciprocating for the care they had received from their GPs, whose involvement in the research, they felt, signified a caring attitude.

Although the randomisation procedure was fully explained, some participants were not aware of its implications and a number were unaware to which group they were randomised (table 4). However, when the procedure was explained, most participants, whether in the intervention or control group, thought that randomisation was fair. Two participants did voice reservations about the control condition on the basis that this resulted in a denial of income.

The intervention was regarded positively by participants, irrespective of whether they gained personally (table 5). The reasons for this were linked to the fact that this particular welfare rights service did not operate in a conventional way. The assessment was offered rather than sought and any claims arising were as a result of a knowledgeable professional actively assisting respondents to make claims. This effectively sanctioned claims as well as lessening the chances of rejection. Providing the service in the respondent's home meant that it was convenient, safe and particularly accessible for those with sensory or mobility problems. The most common description of the service was 'helpful', but many participants spoke at length about having the chance to talk, and having someone take an interest, suggesting that the welfare consultation itself had a therapeutic effect. Among those who did not qualify for benefits, there was no evidence that their hopes had been raised and dashed by no subsequent claim being made.

The process of the intervention required the disclosure of a considerable amount of information about personal finances. No-one objected to this in the context of the welfare rights assessment, although one participant did not wish to disclose actual amounts. Most participants saw this as a necessary part of the process, and many were already required to disclose finances for other matters, such as qualifying for council tax rebates. Participants highlighted the fact that the service made them feel relaxed, at ease and confident about applying for benefits.

Fourteen participants received additional financial resources. The perceived impact of extra money was considerable and it was used on a wide range of items. From participants' accounts, four linked categories were identified (table 6). Firstly, increased affordability of necessities, without which maintaining independence and participating in daily life was difficult. This included accessing transport, maintaining social networks and social activities, buying better quality food, paying bills, preventing debt and affording paid help for household activities. Secondly, occasional expenses such as clothes, household equipment, furniture and holidays were more affordable. Thirdly, extra income was used as a means of dealing with potential emergencies and to increase savings. Fourthly, all participants described the easing of financial worries as bringing 'peace of mind'. Figure 2 summarises the ways in which these categories may be linked.

Participation in this RCT was regarded as a positive experience by most interviewees; the intervention was highly regarded and had wide-ranging impacts, particularly for those who gained new benefits.

Participants' perspectives in RCTs have generally been measured by using attitude questionnaires with a view to improving participation rates[3]. Although limited by the questionnaire format, such studies show that the main reasons given for anticipated participation in trials were altruism[23]and personal benefit[24], and major disadvantages identified were time involved and problems with travel[25, 26]. Altruism, generating new knowledge through research and personal benefit were the main reasons given for participation in this study. Because of its domiciliary design, the study did not require participants to travel, something that many, particularly those with limited mobility or sensory impairments, appreciated and which may have increased recruitment and reduced attrition. Our study also suggests that the involvement of a trusted health professional in the recruitment can increase participation rates.

Although every effort was made to ensure that consent was fully informed, it became clear that a number of people did not understand the nature of the intervention or were not fully aware that they were being randomised. This occurred despite ethical committee approved information leaflets that were regarded as clear and a good reflection of participants' experiences in the trial. This has also been found in clinical research[4]. With the exception of two participants, no-one was unhappy about randomisation to the control condition which involved waiting six months for welfare rights advice. Participants rationalised this in three main ways: firstly that it was a fair way to 'ration' a service that was otherwise not usually available; secondly, that people must face the consequences of their informed consent; and, thirdly, a degree of fatalism and/or ignorance about group allocation. Participants in this study appeared to have a more relaxed attitude to randomisation than has been reported in clinical trials[3]. This could be explained by differences in the nature and context of clinical and social interventions, the former concerning treatment for an existing condition with anticipated health benefits; the latter, as in this case, concerning access to a potentially valuable service, but with unknown health effects. This study demonstrates that it was acceptable to randomise the intervention with a delay of six months to the control group. However, the pilot RCT findings suggest that this is unlikely to be adequate time for any measurable health benefits to become evident [18]. A longer delay between the study and control group's receipt of the intervention raises ethical issues that require further consideration [27].

The qualitative data were derived from a sub-sample of the RCT participants and it is possible that only those willing to be interviewed further held such positive views. Comparison between the quantitative and qualitative samples on a number of social and economic variables indicated similarities according to age, proportion of those living alone/as a couple, council tax band and long term limiting illness, suggesting that both samples were not qualitatively different [18, 28].