Effect of advertising on awareness of symptoms of diabetes among the general public: the British Diabetic Association Study

BMJ 1994; 308 doi: (Published 05 March 1994) Cite this as: BMJ 1994;308:632
  1. B M Singh,
  2. J J W Prescott,
  3. R Guy,
  4. S Walford,
  5. M Murphy,
  6. P H Wise
  1. Department of Endocrinology, Charing Cross Hospital, London W6 8RF Kilmartin-Baker Ltd, London WC2E 9OS Basingstoke District Hospital, Basingstoke RG24 9NA New Cross Hospital, Wolverhampton WV10 0QP British Diabetic Association, London W1M 0BD
  1. Correspondence to: Dr Singh.
  • Accepted 30 December 1993


Objective: To determine the impact of posters advertising symptoms of diabetes on public knowledge of these symptoms.

Design: Structured street interviews of members of the general public before, at the end of, and 10 weeks after a campaign advertising the main symptoms of diabetes.

Setting: Basingstoke and Wolverhampton. Subjects - Three samples of 1000 members of the general public were interviewed. Samples were selected randomly but stratified to match the local population's age (20 -75), sex, social class, and racial characteristics.

Main outcome measures: Knowledge of symptoms of diabetes; perceived seriousness of diabetes; and induction of anxiety about symptoms in the target population.

Results: Advertising significantly raised knowledge (without prompting) of symptoms: thirst, 245 before v 411 at end of campaign (P<0.0001) v 341 after (P=0.0012 v before); polyuria, 72 v 101 (P=0.0211) v 92 (P=0.5169); lethargy, 180 v 373 (P<0.0001) v 298 (P<0.0001); knowledge of weight loss and visual disturbance was unaffected. The number of subjects lacking knowledge of any symptoms was reduced from 550 to 388 (P<0.0001). The perceived seriousness of diabetes was unaffected 20(mean 7.6 in each phase on a scale of 1 (not) to 10 (very). Before advertising, 449 (45%) claimed to have one or more symptoms of diabetes, but this number fell at the end of the campaign (403; P=0.0419) and 10 weeks afterwards (278; P<0.0001).

Conclusions: An advertising campaign raised public knowledge of diabetes symptoms without inducing fear of diabetes or anxiety about symptoms. Its potential for achieving earlier detection of non-insulin dependent diabetes should be evaluated.

Clinical implications

  • Clinical implications

  • Patients with non-insulin dependent diabetes present late, despite prolonged symptoms before diagnosis, with a high prevalence of established micromacular and macrovascular disease

  • This may be because the knowledge base of symptoms of diabetes in the general population is poor

  • Poster based advertising of diabetes symptoms can be effective in educating the public without inducing fear of diabetes or anxiety about symptoms

  • Refining this approach may be a cost effective way of encouraging non-insulin dependent diabetic patients to present earlier


In patients with non-insulin dependent diabetes mellitus, microvascular disease at diagnosis is attributed to prolonged asymptomatic hyperglycaemia.*RF 1-3* We have recently shown, however, that non- insulin dependent diabetes is infrequently asymptomatic. When systematically questioned, 93% of newly diagnosed patients reported classic symptoms, often previously ignored; 40% had had these symptoms for 12 months or more.4 This failure to recognise symptoms may reflect the general public's lack of knowledge of the symptoms of diabetes - half are unable to name any symptoms and only 4% could volunteer the important combination of thirst and polyuria.5 In view of the increased morbidity, and possibly mortality, associated with delayed diagnosis,6 reducing the time from onset of non- insulin dependent diabetes to formal diagnosis is likely to be important and might be achieved by increasing general awareness of symptoms of diabetes, aided by public health education programmes. Therefore the aim of the present study was to determine the ability of a campaign advertising diabetes symptoms to raise knowledge of the symptoms of diabetes.


The study was conducted simultaneously in two English towns, Basingstoke and Wolverhampton, between May and November 1992. It consisted of three consecutive 10 week phases: before advertising, the advertising, and after advertising had been removed.

During the advertising phase, posters (91 cm x 152 cm) advertising symptoms were displayed in a quarter of four-sheet sites in all shopping precincts and other pedestrian areas in each town. Advertisements also appeared in the local press twice weekly. The six classic symptoms selected (thirst, poluria (“passing excessive urine”), weight loss, lethargy (“tiredness”), genital irritation, visual disturbance) are those most commonly given by newly diagnosed patients with noninsulin dependent diabetes.4,6 Two posters (fig 1) were specifically designed (by Kilmartin-Baker Ltd, London) to be educational and optimistic in tone without being alarmist or threatening. They were text based, in contrast with the highly visual posters of the British Diabetic Association depicting a mother injecting her child, for example (fig 2).

Fig 1
Fig 1

Posters advertising diabetes symptoms displayed in Basingstoke and Wolverhampton for 10 weeks

Fig 2
Fig 2

Example of British Diabetic Association fund raising poster advertising

Structured street interviews were conducted in the final week of each phase of the study by a professional market research organisation (Business Planning and Research International Ltd, London). Altogether 3000 subjects were questioned - 500 in each phase in each town. Subjects were randomly selected in the age range 20-75 years to meet quotas that proportionately reflected the age, sex, social class, and racial mix of the populations in the two local authority areas. The samples were statistically matched for these characteristics between the phases. No subject was questioned twice, either within or between phases. People with diabetes and health professionals (doctors, nurses, and pharmacists) were excluded. Of the 3000 subjects, 1493 (50%) were male; 700 (23%), 1146 (38%), and 1154 (39%) were in social classes A- B, C1-C2, D-E respectively; ethnic origin was determined as white, Asian, Afro-Caribbean, or other in 2689 (89.5%), 181 (6%), 114 (4%), and 16 (0.5%) respectively; age distribution was 747 (25%) 20-29 years, 609 (20%) 30-39 years, 594 (20%) 40-49 years, 467 (16%) 50-59 years, 434 (14%) 60-69 years, and 149 (5%) 70-75 years. Overall, 765 (26%) claimed to have a family member with diabetes and 783 (26%) had a friend with diabetes; there was no significant difference in these characteristics between phases.

The question regarding knowledge of symptoms was first asked unprompted: “Can you tell me any of the symptoms of diabetes? ... Anything else?” and all volunteered responses were recorded. Knowledge of symptoms was subsequently determined, prompted by the use of a letter coded, randomly ordered list of symptoms. Whether subjects themselves had any of the advertised symptoms was assessed in a similar way. When awareness of advertising was assessed an unprompted question, “Have you seen any advertising on the subject of diabetes?” was followed by a question prompted by pictures. Awareness of certain concurrent commercial advertisements was also determined by using picture prompts. Questions regarding the perceived seriousness of diabetes were asked with the aid of cards bearing combined numerical and text scales.

To determine the impact of the advertising either on new presentations of patients with non-insulin dependent diabetes or on the induction of anxiety about symptoms among the target populations, general practices were recruited to monitor the number of patients who attended specifically requesting tests for diabetes and the number of these who were confirmed as having diabetes. Other newly diagnosed patients who presented routinely with symptoms or who were found at routine screening were documented separately. Twenty one practices in Basingstoke and 22 in Wolverhampton volunteered to collect these data; they collected data during the first and second phases of the study only. New presentations of non-insulin dependent diabetes were also monitored in the local hospital diabetic clinics (Basingstoke District Hospital and New Cross Hospital, Wolverhampton).

Statistical analysis was by the X2 test for differences between proportions. Results were considered significant at P<0.05.


Awareness of advertising

At the end of the advertising phase, the number of subjects who, without prompting, described the diabetes symptoms posters (fig 1) when asked to recall diabetes related advertising was 324, with 235 recalling the poster with text only and 142 the poster with pictorial material (P<0.0001); the combined total fell to 215 (P<0.0001) at the follow up 10 weeks after advertising was withdrawn. With prompting, 445 people recognised the poster with text only and 372 the other poster (P=0.0009) at the end of the advertising phase. Twenty five people remembered the standard British Diabetic. Association posters without prompting in the pre-advertising phase; the number remained low in the next two phases (35 (P=0.2381) and 14 (P=0.1059)), while with prompting 158, 240 (P=<0.0001), and 214 (P=0.0016) remembered these posters in each of the three phases. The prompted recall of concurrent poster advertising of a well known washing powder, chocolate bar, and proprietary analgesic - 121, 84, and 72 respectively - indicated the greater success of both the diabetes related campaigns.

Knowledge of symptoms

In the pre-advertising phase, unprompted knowledge of individual symptoms (table I) was low. The combination of thirst and polyuria was given by only 51 of the 1000 people interviewed. Advertising led to a considerable increase in the awareness of thirst and lethargy, which was sustained after the campaign. There were also significant but small improvements in the awareness of polyuria and genital irritation, these were not maintained, and there was no change in awareness of weight loss or visual disturbance. The rise in awareness of thirst and polyuria combined was small (P=0.0506).

Table I

Number of non-diabetic members of the general public knowing symptoms of diabetes (unprompted) before, during and after a campaign advertising symptoms of diabetes

View this table:

In the pre-advertising phase, lack of knowledge (unprompted) of any symptoms of diabetes (table 1) was significantly associated with male sex (P<0.0001), younger age (P=0.0002), lower social class (P=0.0382), not having a friend or family member with diabetes (44% (227/515) with v 67% (323/485) without; P<0.0001), but not with race. Table I shows unprompted knowledge of 0, 1, 2, or 3 or more symptoms and table II shows knowledge of symptoms prompted from a symptoms list.

Table II

Prompted knowledge of symptoms of diabetes among non-diabetic members of the general population before, during and after a campaign advertising symptoms of diabetes

View this table:

Perceived seriousness of diabetes

Respondents were asked to rate the seriousness of diabetes on a scale of 1 to 10 (“not at all serious” to “very serious”). The mean score in each of the three phases was 7.6 (P=0.55). The perceived seriousness of diabetes relative to other diseases was determined on a scale of 1 to 5 (1=much less serious, 3=about the same, 5=much more serious). Small but significant movements occurred between categories; table III shows the mean scores in the pre-advertising, advertising, and post-advertising phases. The order of seriousness of these diseases relative to diabetes, ranked by mean scores, was constant between phases.

Table III

Perceived serious of diabetes relative to other diseases

View this table:

The worried well

Among the sampled population that was presumably non-diabetic, the number of claiming to have symptoms of diabetes (499) was high in the pre-advertising phase but fell significantly both during and after advertising (table IV). The prevalence of each symptom also fell, as did the numbers of subjects volunteering 1, 2, or 3 or more of these symptoms.

Table IV

Prevalence of symptoms of diabetes in presumed non-diabetic members of the general population before, during and after a campaign advertising symptoms of diabetes

View this table:

Impact on new presentation of patients with diabetes

During the 10 weeks before and during advertising, 43 general practices monitored the numbers of their patients who specifically requested tests to exclude diabetes. Compared with the pre-advertising phase, there was a 71% increase in such presentations during the advertising phase (99 v 58). Since the proportion confirmed as having diabetes remained constant between the two phases (17%), the numbers of new patients with non-insulin dependent diabetes who presented themselves requesting tests similarly increased by 70% (17 v 10). This occurred while the number diagnosed as having diabetes after a routine presentation either in general practices (64 v 70) or in the local hospital diabetic clinics (75 v 90) fell by 9% and 17% respectively.


Our hypothesis that health education about diabetes symptoms might lead to earlier diagnosis of noninsulin dependent diabetes is founded on our observation that patients often have symptoms for prolonged periods before they present,4,6 together with the general population's poor knowledge of the symptoms of diabetes.5 In our initial exploration of this hypothesis, we have determined the ability of advertising to influence the public's knowledge of diabetes symptoms.

In advertising terms this short duration, moderate intensity campaign achieved an exceptionally high level of penetration,7 compared not only with advertisements for commercial products but also with diabetes advertisements more oriented towards fund raising. The reasons for this are not clear, but the study advertisements were deliberately designed to educate by factually and clearly presenting the symptoms of diabetes, using a low key approach with a positive tone in order to avoid inducing anxiety and consequently denial.

The advertising campaign proved capable of raising the public's awareness of diabetes symptoms, although the effect varied between symptoms. Significant and sustained improvements occurred in the unprompted knowledge of thirst and lethargy, but the rise in knowledge of polyuria was of little practical significance and resulted in a disappointing lack of effect on the important combination of osmotic symptoms, thirst and polyuria. The prompted recall of the combined osmotic symptoms did, however, show significant improvement. Neither unprompted nor prompted knowledge of weight loss or visual disturbance was affected by advertising. Future diabetes symptoms advertising research will focus on lethargy and the osmotic symptoms and will be designed to have greater impact on polyuria.

Advertising to promote symptoms raised the spectre of inducing anxiety about symptoms in the general population, but we found no evidence of this. There was no alteration in the perceived seriousness of diabetes, and a significant decrease occurred in those claiming to suffer from these symptoms, perhaps indicating the converse - a positive health benefit in reduced anxiety about symptoms as a result of health education.

Prompting non-specific symptoms opened the potential for a significantly increased workload in primary care but, in the event, this was small. Nevertheless, the number of patients specifically requesting tests for diabetes was increased by 70% during the 10 week advertising period and among these there remained a high diagnosis rate for diabetes. The proposal that health education about diabetes symptoms may lead to earlier diagnosis of non-insulin dependent diabetes is supported. Although the impact of advertising on numbers of cases of new diabetes was not large, the short advertising phase and the possibility of improving the design of the poster make this a likely underestimate of the potential value of such a campaign.

Cost effectiveness is difficult to determine. The total advertising costs in this study were pounds sterling 21 000. An improved, high profile, six month campaign throughout the United Kingdom using posters and advertising in national newspapers would cost about pounds sterling 350 000. If our results are extrapolated to all general practices in the two areas surveyed, 22 additional patients would have presented during the advertising phase (a cost of pounds sterling 1000 per case), but considering we reached 1% of the target United Kingdom population aged 20-75, the potential nationwide cost per case over six months (2.6x10 weeks; pounds sterling 350 000/(22x2.6x100)) would be pounds sterling61, easily competitive with more conventional screening methods.8

In conclusion, a poster based public health education advertising campaign proved capable of improving knowledge of symptoms of diabetes among the general population without creating anxiety about symptoms or fear of diabetes. This approach will now be refined and its impact on achieving earlier diagnosis of non-insulin dependent diabetes further evaluated.

We gratefully acknowledge the help of general practitioners and practice nurses in Basingstoke and Wolverhampton who collected data for this study.


View Abstract