Community perceptions of suspicious pigmented skin lesions: are they accurate when compared to general practitioners?

Abstract

Community responses (n = 925, response rate = 71%) of a series of eight photographs of pigmented skin lesions were compared against those of general practitioners (n = 114, response rate = 77%), considered to be the most relevant gold standard. The eight photographs included three melanomas, two potentially malignant lesions and three benign pigmented lesions. Over the pool of lesions examined, the average probability that community members thought a lesion was likely to be skin cancer (0.68 [99% CI = 0.66–0.69]) was higher (p < 0.0001) than that of the comparison general practitioners 0.58 [99% CI = 0.55–0.62]. This reflects a general (but not consistent) inflated propensity to over-diagnose among community members. The average probability that respondents indicated they would seek medical advice for a lesion was 0.71 [99% CI = 0.70–0.73]. As expected, this was strongly associated with their perceptions of the skin lesion. These results suggest that the community can play a valuable role in assessing the need for medical evaluation of pigmented skin lesions.

Introduction

Melanoma is an important cause of mortality and morbidity worldwide [1], [2] particularly in Queensland, which has the highest incidence rates of malignant melanoma and non-melanoma skin cancer in the world [1], [3]. Mortality trends probably give the best measure of progress in controlling melanoma, since incidence rates of melanoma can fluctuate according to public awareness and therefore trends are often difficult to interpret. In Queensland between 1982 and 2001, mortality from melanoma among men increased by about 1.2% per year, with an age-standardised (world population) mortality rate of 6.2 deaths per 100,000 population in 2001 [4], [5]. There was no statistically significant trend in the melanoma mortality rate among women between 1982 and 2001 (2.1 deaths per 100,000 population in 2001) [4], [5]. Increases in melanoma mortality rates have been reported elsewhere [6].

There are two strategies that have the potential to reduce the high incidence and mortality of melanoma: primary prevention and secondary prevention. Primary prevention aims to identify, eliminate or counteract the many factors that promote the development of melanoma, and is applicable only to those people for whom there is no current diagnosis of melanoma [7]. Although primary prevention is important [8], [9] and may prove useful for future generations, it may not lead to immediate benefits for the current population [10], especially for people who already have had some exposure to the factors that promote melanoma development. Primary prevention is also considered a more difficult and long-term solution [11]. The second method, secondary prevention or early detection, aims to detect melanoma at an earlier stage when prognosis after treatment is better [9], [12], [13]. Although no randomised trials have yet reported on the effectiveness of early detection [14], [15], clinical experience and knowledge of the stages of melanoma progression suggest that there is potential to reduce melanoma mortality through early detection [16].

Studies have shown that a substantial proportion of melanomas are first detected by individuals [17], [18]. The purpose of skin self-examination is not to diagnose skin cancer. Clearly, that is the role of physicians, dermatologists and ultimately, pathologists. Rather the purpose of skin self-examination is to be able to distinguish between those skin lesions displaying characteristics that warrant medical clarification and those skin lesions that require no action. By doing this we may increase the prevalence of skin cancer amongst the pool of lesions presented to physicians. Therefore, by improving skin self-examination techniques, an improvement in the sensitivity of physicians’ clinical diagnosis could also follow [19].

Previous studies have tended to investigate the effectiveness of skin self-examination by comparing an individual's response against the histological diagnosis of the lesion [20], [21], [22]. Discrepancies between individuals’ perceptions and the histological diagnosis of the lesion alone are attributed solely to the individuals’ lack of ability in skin self-examination. However, this ignores the anticipated role of skin self-examination in the early detection of skin cancer and its relationship to the role of medical services. There are three potential limitations with using histology as the comparison criteria. First, it could underestimate the current skill levels of individuals in the community. Second, it may underestimate the possible contribution of skin self-examination to an early detection program and ignores the differing roles of community and general practitioners in the treatment pathway. Finally, we might be setting ourselves an unattainable standard by which to measure skin self-examination skills. Combined, these limitations may lead us to discount skin self-examination when thinking about early detection programs if we use histology as the comparison criteria.

Therefore, more relevant criteria to assess the accuracy of skin self-examination would be the clinical impressions of general practitioners, and not histology. In Australia, the general practitioner acts as a gatekeeper to further specialist medical care. Using general practitioners as the comparison criteria enables us to better reflect the role of skin self-examination in the process of detecting skin cancer.

This study contributes to an understanding of the process of early detection by assessing the accuracy of community perceptions of skin lesions by comparing them against those of general practitioners, and their own reported actions that they would take if they noticed a particular lesion. By doing this we hope to provide more information about the role that the community can play in recognising the need for medical evaluation of pigmented skin lesions.