predicting the uptake of health checks in general practice: invitation methods and patients'...

Sot. Sri. Med. Vol. 37, No. I, pp. 53-59, 1993 0277s9536/93 $6.00 + 0.00 Printed in Great Britain. All rights reserved Copyright 0 1993 Pergamon Press Ltd

PREDICTING THE UPTAKE OF HEALTH CHECKS IN GENERAL PRACTICE: INVITATION METHODS AND

PATIENTS’ HEALTH BELIEFS

PAUL NORMAN

Department of Psychology, University College of Swansea, Swansea SA2 8PP, Wales, U.K.

Abstract--General practice is in an ideal position to encourage people to adopt healthier lifestyles. One way in which this might be achieved is through the offering of health checks. This paper sought to identify some of the factors which may be important in predicting the uptake of health checks. These were seen to fall into two broad areas-the way in which patients are invited and patients’ health beliefs. For the purposes of the study, 299 middle-aged male patients registered at a single practice were randomly allocated to be invited to a health check via two invitation methods. In the first, patients were sent an invitation letter with an appointment. In the second patients’ notes were tagged so that they could be invited opportunistically by their GP. After one year, 63.5% of patients invited opportunistically had attended a health check compared with 61.2% of patients sent an invitation letter. However, only 51.6% of the opportunistic group had been invited after one year, meaning that the letter invitation method produced twice as many patients at the health check over the same period of time. Examination of health belief questionnaires sent to patients prior to the commencement of the health checks (50.2% response rate) indicated that attenders had a stronger initial intention to attend and placed a greater value on their health. More detailed analysis revealed intention to show strong relationships with beliefs about the benefits of, and barriers to, attendance at a health check. These results suggest that in order to increase uptake of health checks letter invitations need to be employed. Such letters should emphasize the benefits of health checks and address the potential barriers to attendance.

Key words-health belief model, general practice, screening

INTRODUCTION

Coronary heart disease is the commonest single cause of death in the U.K. accounting for over a quarter of all deaths. Its consequences are far reaching. It has been estimated to have cost society over f900 million in lost production and over E350 million in treatment costs in 1981/82 alone [l]. The role of lifestyle factors in the etiology of coronary heart disease has attracted considerable attention with the modification of these factors offering one of the best prospects for reduced morbidity and mortality. It is within this context that the Coronary Prevention Group, which includes both the British Medical Association and the Royal Col- lege of General Practitioners, has advocated the encouragement of the adoption of healthy lifestyles.

This approach has been reinforced by current developments in general practice, brought about by the recent general practitioner contract, which have put a greater emphasis on the provision of preventive services. Under the tenns of the contract [2], general practitioners are committed to “the initial surveil- lance for disease, disability and other health prob- lems” and to giving “general advice and counselling on the maintenance of good health and well-being by the adoption of a healthy lifestyle”. So, in line with aims of the new contract, services targeted at the prevention of coronary heart disease have typically focused on the recording of known risk factors and the giving of health promotion advice.

In relation to the first focus, it is generally acknowl- edged that base-line measures of these risk factors in general practice is poor, with studies showing only 50% of patients’ notes to have a record of blood pressure, 25% a record of smoking status and 10% a record of weight [3]. One way of improving this level is through the offering of health checks, and recent years have seen an increase in the provision of health checks in general practice [47]. Through the provision of health checks it is also possible to tackle the second focus, that is, the giving of appropriate health promotion advice. However, early studies have shown the uptake of health checks to be disappoint- ingly low with 50% attendance rates not Cncommon [6]. Such a performance seriously undermines the potential effectiveness of health checks, on a population level, and points to the need to identify those factors which may hinder or encourage their uptake. This paper focuses on two sets of factors, the first being the way in which patients are invited to health checks and the second being patients’ health beliefs.

To date, two main ways of inviting patients to health checks have been employed. The first, an opportunistic method involves general practitioners, or other members of the primary health care team, inviting patients when they attend a routine consultation. This procedure can be. enhanced by tagging patients* notes so that the health professional is reminded to invite the patient. The appeal of this

SSM 3711-E 53

54 PAUL NORMAN

approach is obvious given that over 90% of patients will consult their GP at least once every three years and that there are over one million primary care contacts each day [8]. In fact, results obtained by this method are, on first inspection, impressive with attendance rates in excess of 90% being reported [9, IO]. Unfortunately, the actual coverage achieved in these studies is less than impressive. For example. an Oxford practice reported that 94% of patients attended a health check in response to an opportunistic invitation. However, after a period of two and a half years only 25% of their targeted population had actually attended a health check [IO], thus suggesting that many patients hadn’t been invited.

In contrast. other practices have invited patients through the use of invitation letters. This approach involves the practice first identifying those patients they wish to invite, and then sending them a letter. It has the advantage that all patients in a target group can be invited over a relatively short period of time. The uptake of health checks in response to letter invitations has been mixed, with reported attendance rates ranging from 36% to 72% [9, I I], although it is likely that such an approach will have produced greater numbers at health checks.

Though not mutually exclusive (in fact, many practices employ both methods), there is a clear need for the potential and limitations of each method to be established so that when they are used in con- junction, they are used to maximum effect. Unfortu- nately, it is difficult to make firm conclusions about the relative merits of each method on the basis of previous research as much of it has been conducted across different practices. As a result, other factors, such as the size and location of the practice and the make-up of the target groups may have contributed to the results. In short, there is a necessity for studies which compare the effectiveness of different invitation methods as they run in parallel in single practices with random allocation of the target group to the different invitation methods. To date, there have been relatively few studies which have taken this approach [I 1. 121. The present study therefore followed this approach to compare the effectiveness of opportunistic and letter invitations, both in terms of attendance rates and in terms of numbers (i.e. coverage) attending a health check.

In addition to examining the way in which patients are invited to health checks, it is also possible to look to other factors which may be important in determining the uptake of the service. Health psychologists have long been interested in examining the factors which underly health-relevant decisions, and have paid particular attention to the role attitudes and beliefs play in these decisions. In the present context the beliefs patients hold about their health and about health checks may influence the likely uptake of the service. Knowledge of those

beliefs which are associated with attendance and non-attendance at health checks can inform the way in which the service is presented to patients and, ultimately, increase uptake.

Most of the work in this area has been guided by the Health Belief Model (HBM) [I 31. Briefly, accord- ing to the HBM, the likelihood of an individual following a health-related action is determined by four psychological variables. First and second, is the individual’s percieved susceptibility to, and severity of, a particular condition or illness in general. Indi- viduals who perceive their health to be under threat are seen to be more motivated to pursue a health-related action. Which action is followed is determined by a cost-benefit analysis of the benefits of, and barriers to, the various alternatives. Actions with many benefits and few barriers are more likely to be followed. These considerations need to be triggered by a ‘cue to action’, be it internal (e.g. physical symptom) or external (e.g. invitation to attend a health check). Furthermore. it has been argued that certain individuals may be pre-disposed to respond to such cues because of the value they place on their health [I31 and the extent to which they see their health to be under their own control, the control of powerful others (i.e. health professionals) or chance [14]. Finally, some researchers have suggested that an individual’s intention to act is a mediating variable between the variables in the HBM and behaviour [15-171. Hence, given that an intention to act may form the basis of behaviour, an examination of the determinants of intention may also identify variables important in increasing the uptake of health checks.

Although there have been few studies which have applied the HBM to the prediction of attendance at health checks [6, I I], studies which have looked at the uptake of a range of screening services lend some support to HBM. For instance, the value placed on health has been related to attendance at a health check [6], health locus of control beliefs to the uptake of cervical cytology screening [l8], perceptions of susceptibility and severity to colon cancer screening [19], the perception of benefits to attendance at a breast cancer screening clinic for women [20], the perception of (a lack of) barriers to attendance at tuberculosis screening [21], and intention to attendance at a high blood pressure screening clinic

u51. The present study therefore considered the role

of health beliefs, as presented in the HBM, in predicting both intention to attend a health check and actual attendance behaviour. In addition the study sought to compare the effectiveness of inviting patients opportunistically and by letter, in terms of attendance rates and coverage. On the basis of previous research it was hypoth- esised that the opportunistic invitation method would produce a higher attendance rate but a lower coverage.

Predicting the uptake of health checks in general practice 55

METHOD

Patients

The study was conducted by a University department in a single general practice in Sheffield, England, consisting of 4 general practitioners with a combined list size of approximately 6000 patients. For the purposes of the study, 299 men aged between 40 and 50 were identified to be invited to attend a health check at the practice.

Invitation methods

The 299 patients were randomly allocated to one of two invitation groups. In the first, letter (n = 148), group patients were sent an invitation letter, with an appointment time, asking them to attend a health check (see Table 1). In the second, opportunistic (n = 151), group patients’ notes were tagged so that members of the primary health care team could invite them personally to a health check when they attended a routine consultation at the practice. This procedure continued for a year, after which those patients who had not been invited were contacted by letter.

The health check

At the health check patients were seen individually by a practice nurse for between 20-30 min. Details about a range of preventive health behaviours were taken including smoking behaviour, alcohol con- sumption and exercise levels, as well as some simple clinical measures such as weight and blood pressure. Furthermore, the health check gave patients an opportunity to talk generally about their health and any concerns or questions they wanted to raise.

Health belief questionnaire

Before patients were invited to attend a health check, the University department sent health belief

questionnaires to the 299 patients included in the study. The questionnaires were returned in prepaid envelopes, and after two mailings 150 completed questionnaires had been returned (50.2% response rate). The questionnaire was based on the Health Belief Model, and used items employed in earlier studies [l 1, 17,221. All items were scored on 4 point response scales, with the exception of the health locus of control items which were scored on 6 point scales.

A number of general motivating factors were firstly considered; these being health value (4 items; internal reliability Cronbach’s [23] tl = 0.77: e.g. “How important do you think it is that people take special care of their health?“) and the internal(4 items, a = 0.62: e.g. “Whatever goes wrong with my health is my own fault”), powerful others(4 items, c( = 0.67: e.g. “I can only maintain my health by consulting health professionals”) and chance(4 items, CL = 0.59: e.g. “When I become ill, it’s a matter of fate”) subscales of the Multidimensional Health Locus of Control Scale [14,22]. In addition, a short scale measuring patients’ beliefs about the eficacy ofdoctorswas constructed (4 items, CL = 0.78, e.g. “For most kinds of illness, it is the doctor who can help me most”).

In line with the overall aims of the health checks, patients were asked to rate their perceived susceptibility to (4 items, c1 = 0.91) and perceived severity of (4 items, tl = 0.97), a number of conditions including heart disease and cancer. The perceived benefits, or efficacy, of health checks were measured using a single item (“How effective do you think health checks are in reducing your chances of getting a serious illness?“). Three scales were constructed to assess the potential barriers to attendance. These focused on motivational barriers(4 items, u = 0.75, e.g. “It would be too much effort”), time barriers (2 items, a = 0.58, e.g. “It would take up a lot of my spare time”) and worries(3 items, tl = 0.66, e.g. “It

Dear Mr Smith,

Table I. The invitation letter

As part of the Medical Centre’s policy of continually improving the services available to its patients we are offering you the chance to have a general health check. This is being offered to all males aged between 40 and 50 who are registered with the Centre and is available as part of the National Health Service.

The health check will give you the opportunity to ask any questions you may have about looking after your health and will also involve the checking of your blood pressure, etc.

We have made you an appointment with a practice nurse

at

The appointment will take approximately 20 minutes. If you would like to bring your wife/partner with you they will be most welcome.

If for any reason you cannot make your appointment please let the reception staff know as soon as possible so that the appointment slot can be used for another patient and you can be offered a more convenient time.

Yours sincerely,

Practice Nurse

56 PAUL NORMAN

would be embarrassing”). Finally, intention to attend was measured using a single item (“If you had the opportunity, how likely is it that you would attend a health check at your doctor’s surgery?).

Attendance behariour

Attendance behaviour was dummy coded as a dichotomous variable of attendance or non-attendance.

RESULTS

Imitation methods

Of the 299 patients identified to be invited to attend a health check, 19 of those who were to be invited by letter and 26 of those who were to be invited opportunistically, were removed during the course of the study either because they had moved or because their general practitioner felt that it would be inappropri- ate to invite them. This left 129 patients to be invited by letter and I25 opportunistically.

The two invitation methods ran in parallel for a year. At the end of this period, no difference was found between the attendance rates of those patients invited by letter (61.2%) and those invited opportunistically (63.2%) (x’ = 0.01, df= I, n.s.). How- ever, whereas all 129 patients in the letter invitation group had been invited, only 52 (41.6%) of those in the opportunistic group had been invited at the end of the year period. As a result, when considering the coverage obtained by the two methods, it was found that the letter invitation method produced more than twice as many patients at the health checks than the opportunistic method (79 vs 33; x’= 31.23, 4f = I, P <O.OOl).

Those patients in the opportunistic group who hadn’t been invited during the first year of the study were identified (n = 73). At this point a further 15 patients were removed from the study, leaving 58 patients to be sent an invitation letter. Of these 58 patients, 29 (50.0%) attended a health check. The difference between this attendance rate and those obtained for patients invited by letter and opportunistically during the first year of the study was not found to be significant (x’ = 2.64, &= 2, ns.). Simi- larly, no difference was found between the combined attendance rate for all those patients originally in the opportunistic invitation group (56.4%) and that obtained for patients in the letter invitation group (61.2%) (x’=O.58. 4f = I, ns.). As a result, the coverage obtained overall for the opportunistic invitation group was broadly similar to that obtained for the letter invitation group (62 vs 79; x’ = 0.58, df = I, n.s.).

Patients’ health beliefs

The first step in the analysis of the health belief data was to examine the correlates of patients’ intentions to attend a health check. A number of signifi-

Table 2. Correlations between the Health Belief Model variables and intention to attend a health check and

attendance behaviour

Intention Attendance Health behefs lo attend behaviour

Health value 0.35*** 0.21’ Internal HLOC 0.14 0.12 Powerful others HLOC 0.1 I -0.05 Chance HLOC -0.17 - 0.05 Efficacy of doctor5 0 27** 0.02 Susceptibility - 0.04 --0.05 Severity 0.17 0.09 Benefits 0.63”’ 0.01 Motivational barrwr 0 59*** fJ.14 Tune barriers 0 39”’ 0 I7 Worries 0.26” 0. I7 Intention 0.1n*

‘P < 0.05. **p < 0.01. ***p < 0.001. n varies between 121.128.

cant correlates emerged, these being: health value. efficacy of doctors, benefits of health checks, motivational barriers, time barriers and worries (see Table 2). Thus. patients who intended to attend a health check were seen to place a greater value on their health, to be more likely to believe in the efficacy of doctors and the benefits of health checks, and to be less likely to perceive a range of barriers as preventing them from attending. The significant correlates of intention were then entered into a stepwise regression analysis in order to assess their predictive value. Three significant independent predictors emerged, these being benefits of health checks (fi = 0.49) motivational barriers (p = 0.32) and time barriers (p = 0.16). Together these three variables accounted for 59% of the variance in intention (R’= 0.59, F(3, I 15) = 54.20, P < 0.0001).

The next step in the analysis of the health belief data was to correlate each of the measures with attendance behaviour. As can be seen from Table 2. only health value and intention to attend were found to correlate significantly with attendance behaviour. These correlations showed attenders to place a greater value on their health and to be more likely to have stated that they intended to attend a health check if given the opportunity. In addition, the perception of time barriers and worries were mar- ginally short of reaching significance, suggesting that attenders were less likely to see these factors as barriers to attendance. Again, the significant correlates of attendance behaviour were entered into a stepwise regression analysis. Only health value (/I = 0.21) emerged as a significant independent predictor, accounting for only 4% of the variance in attendance behaviour (R? = 0.04, F( 1,125) = 5.73, P <0.05).

DISCUSSION

The present study sought to identify some of the factors which are important in determining the uptake of health checks in general practice. These were seen to fall into two broad areas; the first being the


way in which patients are invited to the health check, and the second being the health beliefs patients bring into play when reaching their decision whether or not to attend.

Considering firstly the role of invitation methods, this study sought to compare the effectiveness of inviting patients by letter or opportunistically. The two methods ran in parallel in a single practice for a period of a year. The results showed that there was no difference between the attendance rates of those invited by letter (61.2%) and those invited opportunistically (63.2%). This is in contrast with earlier work which has compared the methods in operation across different practices and concluded opportunistic methods to be superior [9], but in line with other studies which have shown no differences in attendance rates when the two methods run in parallel within a single practice [l 11. It is clear therefore that letter invitation methods can be just as effective, in terms of attendance rates, as opportunistic methods. Part of the reason for this may lie in the nature of the letter employed in the present study which can be seen to present the health check in a relatively positive ‘frame of reference’. The content of invitation letters may therefore play a pivotal role in encouraging patients to attend health checks. This is clearly an issue which is worthy of further research. Alterna- tively, it may be the case that having already received a health belief questionnaire, patients are in some way ‘primed’ to respond in a positive fashion to the invitation letter. Against this view, earlier work has failed to find a difference between the attendance rates of patients who have, or have not, received a health belief questionnaire prior to receiving an invitation to attend a health check [ll].

While there was no difference between the attendance rates of the two methods, there was a substantial difference in the coverage obtained. Inviting patients by letter was found to produce twice as many patients at the health check during the first year of the study. This was because more than half of the patients in the opportunistic group had not been invited during this period. Similar findings have been reported in other studies which have compared the two methods oper- ating within a single practice [1 11. Clearly the potential of opportunistic invitation methods is limited by the fact that patients cannot be invited if they don’t consult. In addition, if patients, for whatever reason, are not invited when they do consult, this will further limit the coverage obtained. As a result, purely opportunistic approaches are likely to take considerable time to obtain adequate rates of coverage [24]. One response to this problem is to send invitation letters to those patients who haven’t been invited opportunistically. Doing so in the present study led to the overall coverage in the opportunistic invitation group being of a similar level to that in the letter invitation group.

Considering letter invitation methods in more de- tail though, earlier work has highlighted a number of

drawbacks with invitation letters which include appointments. Firstly, they have been found to produce a higher proportion of wasted slots through patients failing to attend appointments without informing the practice [25]. In addition, patients have rated such appointments harder to keep than those made after an opportunistic invitation [26]. One response to the problem of wasted slots may be to send patients open invitations asking them to make their own appointments. However, while this has been found to reduce the number of wasted slots, it has also been found to reduce attendance rates [12]. A more appropriate response to the problem of wasted slots may be to slightly overbook clinics.

Overall then, the results of the present and earlier studies suggest that, by themselves, neither invitation method may be able to provide a high coverage of a target population. As shown in the present study, opportunistic invitation methods are likely to require the additional support of letter invitations. Conversely, there may be considerable scope for patients to be invited opportunistically when they have failed to respond to a letter invitation.

The second focus of the present study was on patients’ health beliefs, as based on the HBM, and their relationship with intention to attend and actual attendance behaviour. Although the 50% response rate to the health belief questionnaire was disappointing, it is broadly in line with those reported in other studies in the area [l 1, 171. What is a crucial import- ance though, is that the responses to the questionnaire were obtained prior to patients receiving their invitations to attend a health check. This is important inasmuch as it allows a consideration of the predictive

value of variables taken from the HBM. Other studies have assessed patients’ beliefs after attendance or non-attendance [6, 271. While such studies may provide valuable information, they are clearly limited to the extent to which patients may be merely offering post-hoc justifications for attendance or non-attendance [17]. Clearly, there is a need for prospective studies to supplement earlier work.

Given that an intention to act may form the basis of behaviour [15-171, the first step in the analysis of the health belief data was to examine the correlates of intention to attend a health check. A range of health beliefs were found to correlate with intention; these being health value, perceived benefits, perceived motivational and time barriers, and worries. Thus those patients who intended to attend a health check were more likely to value their health and believe in the benefits of health checks, and less likely to perceive time, motivation or worries about the health checks as barriers to attendance. In a regression analysis three of these variables, namely benefits, motivational barriers and time pressures, were able to account for 59% of the variance in intention. This is in line with an earlier study in which variables taken from the Health Belief Model were found to be highly

58 PAUL NORMAN

predictive of patients’ intentions to attend a health screening appointment [ 171.

In contrast though, when the predictors of attendance behaviour were considered, only health value and intention were found to correlate with attendance behaviour, with those patients who placed a high value on their health and those who said that they would be likely attend a health check if given the opportunity being more likely to attend. This is in contrast with earlier research which had found a range of health beliefs to predict attendance behaviour at health checks, including health value and health locus of control beliefs [6] as well as perceptions of severity and barriers [l I]. Moreover, when the significant correlates were entered into a regression analysis, only health value emerged as a significant independent predictor accounting for a meagre 4% of the variance in attendance behaviour. This disappointing finding is consistent with other studies which have noted the poor predictive value of health beliefs in relation to the uptake of a range of screening services [22,28] and have questioned the utility of cognitive models such as the HBM.

There are a number of plausible reasons for this poor performance [29]. Firstly, cognitive models of health-related behaviour may have been inadequately tested and their constructs inadequately operational- ized. This may be particularly relevant in relation to the HBM which has seen a multitude of operational definitions across different studies [30]. Part of the reason for this is that the HBM can be viewed as a collection of variables rather than a formal model as such [31]. A second plausible reason for the poor performance of cognitive models centres on the models themselves. For example, the HBM may not adequately capture individuals’ representations of illness threats. These may be more concrete and categorical than probabilistic, as implied by the HBM [32]. Thirdly, it is likely that variables other than patients’ health beliefs may be important in the determination of health-related behaviour. As such there may be a need to consider broader social, economic and environmental influences on behaviour

]331. However, the poor performance of variables taken

from the HBM in predicting attendance behaviour can be contrasted with the strong predictive power of the same variables in predicting patients’ intentions to attend a health check. This is evident both in the present and earlier studies, and points to the need to question, not necessarily the utility of models such as the HBM, but rather the gap between prediction of intention and the prediction of behaviour. Why is it that such models can be highly predictive of intentions but consistently poor predictors of behaviour?

On a methodological note, it is perhaps not too surprising that strong relationships were found between variables taken from the HBM and patients’ intentions to attend a health check given that, in the present study, they were measured at the same point

in time. In contrast, for many of the patients in this study, there may have been a considerable time gap between completing the questionnaire and receiving an invitation. This time gap is likely to have attenu- ated the relationship between health beliefs and behaviour [34].

Of perhaps more interest in the present context is an examination of the nature of the behaviour which is being predicted. Even if patients have a strong intention to attend a health check with supporting beliefs there may be difficulties in translating this intention into action. Thus, attendance behaviour may be influenced by a whole range of idiosyncratic events which are difficult to incorporate into a model, such as missing the bus, unexpectedly having to be somewhere else or simply forgetting. The occurence of such events will weaken the relationship between health beliefs and attendance behaviour. Another way of viewing this problem is as a measurement issue. In the present study, attendance behaviour was measured on a single occasion in a single context; that is, patients either attended or failed to attend in response to an invitation. A more appropriate behaviour measure would assess the target behaviour over several occasions or situations. In the present context, this would require an examination of patients’ attendance behaviour in response to a number of invitations. Such a measure would be expected to correlate more strongly with patients’ health beliefs, and so enhance our understanding of the psychological variables underlying patients’ decisions to attend a health check.

To conclude, it is possible to draw a number of practical implications from the findings of the present study. Firstly, from a comparison of two invitation methods it appears that the use of letter invitations which include an appointment time is likely to be needed in order to obtain an acceptable coverage of a target population within a realistic time span. However, such an approach may produce a high proportion of wasted appointment slots, which points to the need for practices to pay special attention to the suitability of the appointment times they offer. This coincides with an analysis of patients’ health beliefs which suggests that those patients who perceive time pressures to be a barrier to attending are less likely to intend to attend. The health belief data also highlight the fact that the letters need to empha- sise the benefits of attending in order to increase interest in health checks.

REFERENCES

Holland W. W. and Stewart S. Screening in Health Care-Beneft or Bane.7 Nuffield Provincial Hospitals Trust, London, 1990. Department of Health and the Welsh Office. General Practice in the National Health Seroice. A New Contract. HMSO, London, 1989. ’ Fleming D. M. and Lawrence M. S. T. A. Evaluation


of recorded information about preventive measures in 38 practices. J. R. College G. P. 31, 615620, 1981.

4. Fleming D. M., Lawrence M. S. T. A. and Cross K. W. List size, screening methods, and other characteristics of practices in relation to preventive care. Br. Med. J. 291, 869-872, 1985.

5. Fullard E., Fowler G. and Gray M. Promoting prevention in primary care: controlled trial of low technology, low cost approach. Br. Med. J. 294, 108&1082. 1987.

6. Pill R., French J., Harding K. and Stott N. C. H. Invitation to attend a health check in a general practice setting: comparison of attenders and non-attenders. J. R.Coll. 6. P. 38, 53-56, 1988.

7. Robson J.. Boomla K.. Fitzpatrick S.. Jewel1 J. A.. Taylor J., Self J. and Coyler I.-A randomised controlled trial of recording preventive activities using health promotion nurses and computer assisted follow up. Br. Med. J. 298, 433436, 1989.

8. Stott N. C. H. and Davis R. H. The exceptional potential in each primary care consultation. J. R. CON. G. P. 29, 201-205, 1979.

9. Mann J. I., Lewis B., Shepherd J., Winder A. F., Fenster S., Rose L. and Morgan B. Blood lipid concen- trations and other cardiovascular risk factors: distri- butions, prevalence and detection in Britain. Br. Med. J. 297, 1702-1706, 1988.

10. Sacks G. and Marsden R. Evaluation of a practice- based programme of health checks: financial costs and success at risk detection. J. R. Co/l. G. P. 39, 3699372, 1989.

11. Norman P. and Fitter M. Predicting attendance at health screening: organizational factors and patients’ health beliefs. Counsell. Psychol. Q. 4, 143-155, 1991.

12. Norman P., Conner M. T., Willits D. G., Bailey D. R., Hood D. H. J. and Coysh H. L. Health checks in general practice: a comparison of two invitation letters. Br. J. Gen. Pratt. 41, 432433, 1991.

13. Becker M. H. (Ed.) The health belief model and per- sonal health behaviour. Hlth Educ. Monographs 2, 324508.

14. Wallston K. A. and Wallston B. S. Health locus of control scales. In Research with the Locus of Control Construct, 2nd edn (Edited by Lefrourt H.). Academic Press, New York, 1981.

15. King J. The impact of patients’ perceptions of high blood pressure on attendance at screening. Sot. Sci. Med. 16, 1079-1091, 1982.

16. Calnan M. The health belief model and participation in programmes for the early detection breast cancer. Sot. Sci. Med. 19, 823-830, 1984.

17. Norman P. and Fitter M. Intention to attend a health screening appointment: some implications for general practice. Counsell. Psychol. Q. 2, 261-272, 1989.

18. Murray M. Locus of control, self-concept, attitudes to cancer and attendance for screening. International Con - ference on Health Psychology, Card%, 1989.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Macrae F. A., Hill D. J., St John D. J., Ambikapathy A. and Garner J. F. Predicting colon cancer screening behaviour from health beliefs. Preventive Med. 13, 115-126, 1984. Hobbs P., Smith A., George W. D. and Sellwood R. A. Acceptors and rejectors of an invitation to undergo breast cancer screening compared to those who referred themselves. J. Epidemiol. Community Hlth 34, 19-22, 1980. Wurtele S. K., Roberts M. C. and Leeper J. D. Health beliefs and intentions: predictors of return compliance in a tuberculosis detection drive. J. appl. Sot. Psychol. 12, 128-136, 1982. Norman P. Social learning theory and the prediction of attendance at screening. Psycho/. Hlth 5, 231-239, 1991. Cronbach L. J. Coefficient alpha and the internal struc- ture of tests. Psychometrika 16, 297-334, 1951. Norman P. and Fitter M. The potential and limitations of opportunistic screening: evidence from a computer simulation of a general practice screening programme. Br. J. Gen. Prac. 41, 188-191, 1991. Fitter M. and Norman P. Improving the effectiveness of . - screening in general practice. In Current Perspectives in Health Care Comwtina (Edited bv de Glanville H. and Roberts J.). The ‘British‘Journal of Health Care Com- puting, London, 1990. Norman P. and Fitter M. Patients’ views on health screening in general practice. Fam. Practice 8, 129-132, 1991. Pill R. and Stott N. C. H. Invitation to attend a health check in a general practice setting: the views of a cohort of non-attenders. J. R. Coil. G. P. 38, 5760, 1988. Marteau T. M., Johnston M., Kidd J., Michie S., Slack J., Shaw R. W. and Cook R. Psychological models in predicting uptake of prenatal screening. Psychol. Hlth 6, 13-22, 1992. Marteau T. M. Health beliefs and attributions. In Health Psychology: Processes and Applications (Edited by Broome A. K.). Chapman and Hall, London, 1989. Champion V. L. Instrument development of the health belief model constructs. Adu. Nursing Sci. 6, 73-85, 1984. Oliver R. L. and Berger P. K. A path analysis of preventive health care decision models. J. Consumer Res. 6, 113-122, 1979. Leventhal H. and Nerenz D. The assessment of illness cognition. In Measurement Strategies in Health Psvchol- ogy (Edited by Karoly P.). Wiley, New York, 1985. Winnett R. A. Ecobehavioural assessment in life-styles: concepts and methods. In Measurement Strategies in Health Psychology (Edited by Karoly P.). Wiley, New York, 1985. Ajzen I. and Fishbein M. Factors influencing the intention behaviour relation. Hum. Relations 27, I-15, 1974.

predicting the uptake of health checks in general practice: invitation methods and patients'...

Documents