Introduction

Top Abstract Introduction Methods Results Discussion References

Type 2 diabetes is a serious condition, with implications for the mortality, morbidity, and social functioning of patients. The prevalence of type 2 diabetes is increasing, and the number of patients in the United Kingdom is expected to rise from just over one million in 1997 to just under three million by 2010.¹ This will inevitably be reflected in rising costs. An estimated 7-8% of the total NHS budget is spent on patients with type 2 diabetes,² and the burden of caring for these patients is falling increasingly on primary care. ^{3 4}

Impaired glucose tolerance, typically characterised by hyperglycaemia and insulin resistance, is considered to be a stage in the development of type 2 diabetes. Up to half of all people with impaired glucose tolerance will progress to type 2 diabetes within 10 years of diagnosis.⁵ In addition, people with impaired glucose tolerance are known to be at significantly increased risk of cardiovascular disease, which may present before the onset of diabetes.⁶ Studies in the United Kingdom have reported the prevalence of impaired glucose tolerance in the 35-65 year age group to be around 17%.⁷

Increasing evidence indicates that intervention can favourably influence the clinical course of impaired glucose tolerance,^8-10 with some studies showing a 58% reduction in progression to diabetes. ^{11 12} However, information is lacking on British general practitioners' awareness of the clinical significance of impaired glucose tolerance and their current management of these patients. This study aimed to ascertain levels of awareness among general practitioners of the prevalence and clinical significance of impaired glucose tolerance and to explore their attitudes to its detection and management.

	Methods

Top Abstract Introduction Methods Results Discussion References

Participants and setting

Focus groups
We used stratified random sampling (to give a representative male:female ratio from each primary care group) to recruit participants for the focus groups. We chose general practitioners from lists supplied by Derwentside, Sunderland West, South Tyneside, and Gateshead West and Central Primary Care Groups. We initially contacted potential participants by telephone and then invited them by letter to take part. We contacted 56 general practitioners, of whom 28 agreed to participate; two of these failed to attend. The remaining 26 general practitioners (18 men, eight women) participated in four focus groups (three groups of seven participants and one group of five participants). The mean age of participants was 44 (range 30-58) years; all were principals, with a mean of 11 (1-27) years' experience in general practice. The 30 general practitioners who either declined or failed to attend were similar in terms of sex and practice characteristics. The table shows the characteristics of individual participants.

Semistructured interviews
We chose the eight participants (six men, two women) in the semistructured interviews purposively¹³ from a list of all general practitioners in one health authority. All were principals, with a mean age of 41 (31-46) years and an average of 12 (4-24) years' experience in general practice. Participants were chosen to reflect diversity in terms of age, sex, practice characteristics (list size, geographical location), and involvement in diabetes care (little involvement, lead general practitioner, clinical assistant in diabetes, primary care group adviser on diabetes) (table). All general practitioners invited agreed to take part. We conducted the interviews at participants' surgeries, and interviews had an average duration of 35 minutes. Participants received an honorarium of £25.00.

Collection and analysis of data
The lead investigator (GW) carried out all the focus groups and interviews, which were audiotaped for later transcription. Before each focus group, participants completed a questionnaire designed to evaluate their knowledge of the clinical significance and prevalence of impaired glucose tolerance. Their responses were then explored in the focus group discussion. The lead investigator then gave a short presentation, based on a review of the literature, on the anticipated rise in prevalence of type 2 diabetes, together with the clinical significance, prevalence, and management of impaired glucose tolerance. ^{1 2 5-11} Further focus group discussion centred around participants' attitudes to impaired glucose tolerance in the light of what, for most of them, was new knowledge.

We took a similar approach with the semistructured interviews, administering the questionnaire verbally and following this with open ended questions concerning knowledge of the clinical significance and prevalence of impaired glucose tolerance. We then gave a short presentation, as above, and used subsequent open ended questions to explore attitudes to and perceptions of the detection and management of impaired glucose tolerance.

We used a "pragmatic variant" grounded theory approach to analyse the data by generating categories and themes. ^{14 15} GW and APSH coded the data independently to increase the reliability of the study. We adopted an iterative approach to data analysis, with analysis beginning after the first focus groups and interviews, to allow emerging themes to be explored in subsequent interviews. The coders agreed that no new themes were emerging after four focus groups and eight semistructured interviews, and saturation was achieved.¹⁵

Validation
To increase confidence in the validity of the findings, we sent all 34 participants a report summarising the outcomes of the study. Twenty eight (82%) replied stating that they "strongly agreed" (10 respondents) or "agreed" (18) that the report was a true representation of their opinions.

	Results

Top Abstract Introduction Methods Results Discussion References

QuestionnairesAll participants were aware of impaired glucose tolerance as a clinical entity. However, 16 (47%) participants were unaware of the risk of impaired glucose tolerance progressing to type 2 diabetes, and 21 (62%) were unaware of the increased risk of cardiovascular disease. In addition, 17 (50%) participants had no idea how many patients with impaired glucose tolerance might be known to their practice, and 13 (38%) estimated prevalence at less than 1% (figure). Focus groups and interviewsThree main themes emerged from data collected before participants received a presentation detailing the anticipated rise in prevalence of type 2 diabetes, together with the clinical significance, prevalence, and management of impaired glucose tolerance (box 1). Eight main themes emerged after the presentation (box 2).

View larger version (13K): [in this window] [in a new window]   General practitioners' estimation of the number of patients known to have impaired glucose tolerance in their practice

	Discussion

Top Abstract Introduction Methods Results Discussion References

The aim of this study was to investigate general practitioners' knowledge of the prevalence and clinical significance of impaired glucose tolerance and their attitudes to its detection and management. The results indicate that awareness of the existence of impaired glucose tolerance was good but that awareness of the prevalence and clinical significance of impaired glucose tolerance was poor. In addition, general practitioners seem to be uncertain about how best to manage and follow up these patients. This has implications for the training and education of general practitioners, and not least for patient care in a field that is likely to expand exponentially in the next few years.

The study also indicates that only a small proportion of patients with impaired glucose tolerance are currently known to practices and that general practitioners are reluctant to pursue more aggressive case finding and management, even after being presented with a critical appraisal of the literature. Several factors seem to contribute to this unwillingness to screen and intervene in this condition. General practitioners were afraid of being overwhelmed by the workload generated, admitting to reluctance to divert finite resources away from other clinical areas, and were pessimistic about the effectiveness of lifestyle intervention. Furthermore, some felt strongly that screening patients for impaired glucose tolerance and subsequent lifestyle intervention would medicalise an essentially social problem and that a health educational approach, involving schools and the media, should be adopted instead. Other studies have shown that general practitioners lack confidence in their ability to change lifestyle behaviour and suggested that the solution would be to increase training and support. ^{16 17} Pill et al have shown how difficult it is for primary healthcare professionals to change patients' behaviour in the context of type 2 diabetes.¹⁸ Our findings are in broad agreement with those of Lawlor et al, who found that general practitioners were in favour of measures to tackle social and environmental determinants of ill health.¹⁹

Conversely, some of the general practitioners in this study expressed a positive attitude towards pharmacological intervention in patients with impaired glucose tolerance, even though the effectiveness of this has yet to be shown in large scale clinical trials. This may be because of the perception that pharmacological intervention is ultimately likely to be more effective than lifestyle intervention. However, it could also be argued that the act of prescribing drugs for these patients essentially defines impaired glucose tolerance as a medical problem, and general practitioners have been shown to be more comfortable managing illness than preventing disease.²⁰

The general practitioners who participated in this study broadly supported the provision of a guideline for the management of patients with impaired glucose tolerance in primary care, but we were unable to find any suitable published guideline. This has implications for guideline development. General practitioners have positive attitudes towards the use of guidelines,²¹ although problems exist with regard to implementation.²²

Methodological considerations
The rigour of this study was increased by triangulation,²³ both of methods and analysis, and by validation by respondents. However, few data were available on the 30 out of 64 general practitioners who either declined or failed to participate. Although sex and size of partnership were broadly similar, we had no way of knowing if the knowledge and attitudes of these general practitioners differed from those participating in the study. In addition, we cannot exclude the possibility that the personal and intellectual bias of the lead investigator (GW) may have shaped the data. This has been minimised by making the account as reflexive as possible and by reporting a wide range of different perspectives, a method described by Mays and Pope as "fair dealing."²³ Factors relating to the ethnic origin of patients did not emerge as a theme from our data. Two out of eight general practitioners interviewed were of Asian ethnic origin, and each focus group contained one or two such participants. The population prevalence of patients of Asian ethnic origin in the study area is low (0.3% for County Durham, 0.5% for Gateshead and South Tyneside, and 0.6% for Sunderland, compared with 3.3% for England and Wales²⁴). This may limit the generalisability of our study to areas with large ethnic minority populations.

Conclusions
The recently published Diabetes National Service Framework: Standards document recommends that the NHS and partner organisations adopt both a "population" approach (tackling obesity and sedentary lifestyles) and a "targeted" approach (identifying and intervening in high risk groups, such as patients with impaired glucose tolerance) to reducing the incidence of type 2 diabetes.²⁵ The national service framework recognises that such interventions are also likely to have an impact on reducing cardiovascular disease. Similarly, early treatment of macrovascular risk factors may be more important than screening for and treating asymptomatic type 2 diabetes itself.²⁶ Although studies from other countries have shown encouraging results, ^{11 12} questions about the feasibility of primary prevention of type 2 diabetes in the United Kingdom remain unanswered. Our findings clearly show that general practitioners have major reservations about the appropriateness and effectiveness of giving lifestyle advice to patients in this context. Similarly, we have shown that general practitioners perceive the need for considerable extra resources if they are to be given the task of screening for impaired glucose tolerance and intervening in patients at high risk of progression to type 2 diabetes. This has important implications, both for the implementation of the diabetes national service framework and for primary care research.