Main findings of this study
This study reviewed evidence for factors that are associated with self-testing and similar activities among UK adults. Forty nine eligible papers were identified. Most (n = 28) were identified by Medline and most (n = 30) were related to CAM. There were 46 unique studies and most (n = 36) had high or medium quality scores. Most (n = 37) studies were, or used data from established questionnaire surveys, but only 16 of them were population-based. Survey analysis methods varied from simple descriptive to adjusted analyses, but studies generally looked at five areas – demographic factors (sex and age), affluence/deprivation, education, health status, and health behaviours and beliefs.
Being female was associated with the activity in 13 of 27 studies about CAM, eight of 13 studies about OTC medicine, and the only study about home BP monitors. Ten studies about CAM, nine studies about OTC medicine, and three studies about the private sector reported that people in the 30 to 64 year age range were most likely to undertake the activity or that use declined after 60 or 70 years. One study, however, found that private insurance was more common in men than women and as age increased [47
], and the study about home BP monitors reported that use became more likely as age increased [52
A link with affluence was presented in eight studies about CAM, five studies about OTC medicine, all six studies about the private sector, and the study about home BP monitors. Higher educational attainment was associated with use in three studies about CAM, one study about OTC medicine, two studies about the private sector and the study about home BP monitors. Education and affluence are likely to be associated though and many studies used unadjusted analyses or looked at education or affluence. Even so, no study found that the activity was less common in more affluent or educated people.
Six studies about CAM, three studies about OTC medicine, and the study about home BP monitors suggested a link with various measures of poor health, but the analyses were often unadjusted. Conflicting evidence was also presented by two studies: one found that CAM patients had longer symptoms than GP patients but also lower pain scores [31
], and the other reported that private care was less likely among people limited in their daily activities than among those not limited [49
Four studies about CAM and one about OTC medicine suggested a link with healthy lifestyles or being health conscious and knowledgeable about health, and no studies contradicted this. Two studies also found that CAM users were more likely to believe that they control their health or less likely to believe that doctors control their health than users of orthodox medicine.
What is already known on this topic
The aim of this study was to review evidence about who uses self-tests among UK adults. No relevant studies about self-testing were initially identified though. As a result, the scope was widened to look at similar activities, and this study is the first to draw together evidence from different areas about common factors that are associated with self-care activities.
What this study adds
People who use CAM and OTC medicine appear similar. There is general consensus that they are usually female, middle-aged and affluent and/or educated. There is also fairly general agreement that people who undertake these activities have some measure of poor health. People who use home BP monitors appear similar to people who use CAM and OTC medicine except that use was associated with increasing age. The study about home BP monitors did not ask about hypertension though, and rising use could be related to the increased prevalence of this condition with age, perhaps unlike other conditions for which CAM and OTC medicines are used.
Similar to CAM, OTC medicines and home BP monitors, people who use the private sector appear to be affluent and/or educated. In contrast to users of other activities though, there is some evidence that males tend to have private insurance and people with good health tend to have private care. This suggests that it may not be appropriate to group studies about the private sector with studies about other self-care activities.
Limitations of this study
Only studies conducted in the UK were included because it was felt that the health care environment would have affected choices about the activities under consideration. The review may not, therefore, be applicable to other settings.
As there was no clear equivalent activity, studies about activities with similarities to self-testing were reviewed. Each activity, however, also has differences to self-testing, for example CAM is generally considered to be outside conventional medicine, whereas self-tests could be considered an extension of orthodox medicine. There were also differences between the activities, and grouping together results from such studies may not be appropriate, for example people who use the private sector may be different to people who use the other activities. As only one study about home BP monitors and self-tests was identified, it is probably not appropriate to draw firm conclusions about users of these activities.
Self-tests were defined for this study as tests that are bought and used without involving a doctor, nurse or other conventional health professional. As a result, studies were excluded where there was evidence that the activity was always initiated by a conventional health professional. Activities could also be simply recommended by health professionals though, and this was often not asked about or, if asked about, used to group people. It was not possible, therefore, to determine whether people who used an activity after a conventional professional's recommendation were different to people who used them without any such recommendation. Similar to this, use of private care and some CAM facilities may require referral from a conventional health professional, but the idea may have come from the patient. Again though, this was often not reported or used to group people. This may be important as someone who is affluent and/or educated may be more able to influence a GP to arrange a referral, which could, at least, partly explain the link between affluence and the use of these activities.
Eligible studies often defined use in different ways. Identified studies also used different data collection and analysis methods. Questionnaire surveys were often not population-based or did not have a relevant comparison group so descriptive analyses were presented about the group using the activity. Even where there was a comparison, analyses were often not adjusted for confounding variables so it was difficult to see if associations, such as education and affluence, were independent. Factors examined also varied widely, even though more basic characteristics, such as ethnic group, were infrequently studied. It was also sometimes difficult to tell whether papers related to the same study, for example three papers used similar methods and the same authors, but the number of participants varied [27
]. These issues meant that it was not possible to formally pool analyses from studies within each area.
The government is encouraging self-care because of evidence that this will improve health outcomes and appropriate use of conventional services. The promotion of self-care may mean that self-tests are seen as desirable. It will be important, therefore, given the potential disadvantages of self-tests, to assess the impact of this policy on their use. There is a lack of evidence about who currently uses self-tests and why they use them though and a need for research in this area. It seemed sensible to collate evidence from studies about similar activities as a starting point, but it remains important, particularly given the potential disadvantages of self-tests, to use this knowledge to directly study who is using self-tests and why they are being used. The evidence from this review will, therefore, be used to design a questionnaire to look at factors that are associated with using self-tests. This review will also be useful to policy makers wishing to consider how best to promote general self-care activities by highlighting those groups who are not engaged in such activities. Further qualitative research among these groups about why self-care is not used and how this could be facilitated would be useful.