Secondary Logo

Journal Logo

Research Papers

Determinant factors for diagnostic delay in operable breast cancer patients

Montella, M1; Crispo, A1; D'Aiuto, G2; De Marco, M1; de Bellis, G1; Fabbrocini, G3; Pizzorusso, M2; Tamburini, M1; Silvesrtra, P2

Author Information
European Journal of Cancer Prevention: February 2001 - Volume 10 - Issue 1 - p 53-59
  • Free



Breast cancer is the most frequent malignancy among women in developed countries (Parkin et al., 1997). Randomized trials of mammographic screening have provided strong evidence that early diagnosis and treatment of breast cancer can reduce the specific mortality (Nystrom et al., 1993). Moreover, in a recent systematic review of published studies, delays of 3–6 months between symptom onset and treatment have been found to be clearly associated with lower survival rates for breast cancer patients (Richards et al., 1999).

The potential beneficial effects of reducing diagnostic delay for breast cancer patients have also been emphasized in Italy, where persistent inequalities in the health service between the north and south of the country have led to lower cancer survival rates for the population of the southern regions (GIVIO, 1986;Tomatis, 1997). A preliminary examination of the factors affecting diagnostic delay was carried out in southern Italy. An advanced stage of malignancy was found to be associated with socio-demographic factors, such as low educational levels and residence in rural areas (Montella et al., 1995).

However, a separate examination of the individual elements that make up overall delay, from the clinical examination of the breast to the definitive treatment of breast cancer is desirable, since each element requires quite different corrective actions (Coates, 1999).

Several Anglo-American groups have recently performed such an analysis (Caplan et al., 1996;Ramirez et al., 1999). However, their conclusions should be compared with studies of different populations and national health services.

The aim of this study was to examine separately the different delays registered among breast cancer patients in southern Italy so that their determining factors can be recognized, thus giving women a better opportunity for survival.

Material and methods

The study design was a cohort-study of 644 women referred to the National Cancer Institute of Naples (Campania Region), between January 1998 and November 1999. All patients underwent surgery for histologically confirmed breast cancer (ICD 174.0–174.9) (Avtrandilov et al., 1990) and were interviewed by a single trained physician while awaiting their first post-surgery examination in the breast unit.

Since inoperable women with late stage breast cancer are becoming very rare in southern Italy (Montella et al., 2000), the study was carried out exclusively on operable patients because all studies have clearly shown that women with locally advanced/metastatic breast cancers tend to have had significantly longer delays than women suffering from an operable disease (Richards et al., 1999;Ramirez et al., 1999).

Interviews were conducted on the basis of a semistructured questionnaire developed for the study by an expert team composed of a surgeon, an epidemiologist and a psychologist.

The following socio-demographic data were obtained from the personal interview: age at diagnosis: <50 (reference category), 50–64, ≥65; education (≤5, >5 school years); marital status (married, single); occupation (housewife, employee); geographic area (chief town, its province, other provinces); residence, categorized on the basis of a simplified ISTAT classification (Istituto Centrale di Statistica, 1993) into urban (urban and semi-urban) and rural (rural and semi-rural).

Other data obtained from the interview were: symptom status at first presentation – symptomatic and asymptomatic (this latter category represents the patients with health problems discovered by a preventative mammography); nature of symptom as seen by patients at first presentation (lump, others); date of first symptom presentation; date of first consultation with a health provider; consulted provider – senologist (oncologist exclusively dedicated to breast cancer operation), general practitioner, gynaecologist, general surgeon.

In order to better assess the relevance of socio-demographic factors on diagnostic delay we considered the following time intervals: time from first symptom to initial medical consultation (patient delay), this statistic was not used for asymptomatic women who had received a preventive mammography (before the appearance of any real symptom); the time from initial medical consultation to hospital admission (medical delay); the time from first symptom to surgical treatment (overall delay). Time intervals were categorized as follows: <1 month, 1–3 months and >3 months for patient delay and medical delay (<1 month reference category), 1–3 months, 3–6 months, >6 months for overall delay (1–3 months: reference category).

Date of hospital admission, date of surgery, intervention type (mastectomy versus conservative surgery, i.e. quadrantectomy, wide excision, lumpectomy), tumour size and nodal status, according to the pTNM system (IUAC, 1992) were collected from the hospital clinical records.

Tumour size was categorized into two groups according to the TNM classification and was independent from any nodal involvement: ≤2 cm (pTis–pT1), >2 cm (pT2–pT3). These groupings better reflect the biologic history of the disease and the possibility of early clinical detection and therefore yield results relevant to diagnostic delay (Carter et al., 1989;MacMahon, 1994). Cases with unassessable (pTx) or multifocal primary tumours were excluded from the analysis.

Nodal status was categorized into two groups: pN– (pN0), pN+ (pN1–pN2) and pNx. A statistical analysis was performed using chi-square test with corresponding P -value (in the tables) and odds ratios (OR) for the examined variables, with a corresponding 95% confidence interval (95% CI) (in the text). These were derived using unconditional multiple logistic regression models fitted by the maximum likelihood method (Mantel and Haenszel, 1959;Breslow and Day, 1980).

The regression equations included terms for age, years of education, residence and occupation.


The characteristics of the study population are shown in Table 1

Table 1
Table 1:
Patient characteristics

. The average age was 54 years old (range 29–83). Patients aged <50 were 242 (40%), 50–64 were 234 (35%) and ≥65 were 168 (25%). Three hundred and fifty-six patients (55%) had a low level of education (≤5 school years), 288 (45%) had >5 years school attendance. Five hundred and fifty patients (85%) were symptomatic at diagnosis, 94/644 (15%) were detected by a preventive mammography. In 456 patients (70%) a lump was the first symptom (in 362 cases a lump remained the only symptom, in 94 it was associated with other symptoms), 94/550 patients (15%) displayed other symptoms at presentation. In 300 patients (47%) tumour size was ≤2 cm, in 272 (42%) >2 cm, in 60 (9%) pTx or multifocal. There were no data for 12 (2%) patients. In 334 patients (52%) the nodal status was pN–, in 236 (37%) pN+, in 68 (10%) pNx. Data were missing for 6 (1%) patients.

Patient delay

Distribution of 550 evaluable patients according to patient delay is shown in Table 2

Table 2
Table 2:
Patient delay

. In 94 cases patient delay was invalid, as all patients had undertaken a preventive mammography.

Three hundred and sixty patients (65%) received their initial medical consultation in less then 1 month, 82 (15%) within 1–3 months, 108 (20%) after 3 months. Significant differences in patient delay were found: there is a significant risk for women of ≥65 years in the >3 months category (OR 2.1, 95% CI 1.2–3.5, P  = 0.005), another significantly higher risk of >3 months delay was found among women with ≤5 years of schooling (OR 3.3, 95% CI 2.0–5.6, P  = 0.001).

No significant differences were found among the other variables: marital status, occupation, residence and the nature of the first symptom (the assessment was not made at the time of diagnosis, but the first symptom was noted by the patient).

Medical delay

The distribution of 644 patients according to medical delay is shown in Table 3

Table 3
Table 3:
Medical delay

. Four hundred and forty patients (68%) were admitted to hospital less than 1 month after their initial medical consultation, 136 (21%) within 1–3 months and 68 (11%) after 3 months.

Three hundred and seventy-four patients (58%) were examined by a senologist, 126 (20%) by a general practitioner and 144 (22%) by a different specialist (gynaecologist, general surgeon).

Medical delay was associated with age: a significant protective risk of >3 months delay was found in the ≥65 age group (OR 0.1, 95% CI 0.0–0.3, P  = 0.05) while the odds ratio for women in 50–64 age category was not significant.

A significant association between medical delay and the professional figure involved in the first medical consultation was found. In fact, if we consider the senologist as the reference category, a higher risk of >3 months delay was found in patients examined by a GP (OR 3.0, 95% CI 1.2–7.5) and for those examined by different specialists (OR 3.5, 95% CI 1.5–8.4). A higher risk of 1–3 months delay (OR 2.1, 95% CI 1.1–4.0) was also found for the latter category. If we exclude the 94 asymptomatic women, 72 of whom sought help from a senologist, 2 from a family doctor (GP) and 20 from other specialized doctors, then the analysis is exclusively significant for the senologist comparison with other specialized doctors at 1–3 months (OR 2.2, 95% CI 1.4–3.6) and >3 months (OR 1.9, 95% CI 1.0–3.8).

Another significant association was found between symptomatic presentation and medical delay of 1–3 months. In fact, symptomatic women displayed a significantly higher risk of delay when compared with asymptomatic patients, with the latter discovering their health problem by means of a preventive mammography (OR 7.7, 95% CI 3.0–21.5, P  < 0.001).

No significant differences were found among education, palpable lump, tumour size and nodal status.

Overall delay

Distribution of 644 valid patients according to overall delay is shown in Table 4

Table 4
Table 4:
Overall delay

. Four hundred and twenty-two patients (65%) received surgery less than 3 months from first symptom, 102 cases (16%) within 3–6 months, and 120 (19%) after 6 months.

The following factors were found to be significantly associated: age, symptom status and tumour size.

Compared with the <50 age category, women in the 50–64 age category have a significant odds ratio at >6 months (OR 0.6, 95% CI 0.4–1.0, P  = 0.001).

Symptomatic women were more likely to experience a 3–6 months or >6 months overall delay than asymptomatic ones. Their OR was 12.8 (95% CI 3.1–52.9, P  = 0.005) and 4.9 (95% CI 2.1–11.4, P  = 0.001), respectively.

If we consider tumour size we see that in cases where the tumour is >2 cm, for 3–6 months delay the OR value was 2.3 (95% CI 1.4–3.7, P  = 0.0001) and for >6 months delay OR was 2.4 (95% CI 1.5–3.7, P  = 0.005). No significant association was observed between overall delay and education levels and nodal status.


The factors involved in the delayed presentation of symptomatic breast cancer have been investigated in a recent review of published studies mostly concerned with Anglo-American populations (Richards et al., 1999).

Our study analysed a Latin female population that was referred to the National Cancer Institute of Naples (320 beds, 9000 admissions per year and the main oncological institution of southern Italy) together with different time intervals of breast cancer delay (patient delay, medical delay, overall delay), as commonly used in the majority of similar studies (Richardson et al., 1992;Richards et al., 1999;Ramirez et al., 1999).

In southern Italy the survival rate for breast cancer is lower than that in other regions of Italy (Gatta et al., 1997). One of the major reasons for this is that in southern Italy there is a lack of medical structures such as primary care facilities, hospitals (hospital beds), radiotherapy facilities, teaching universities, etc. (Regione Campania, 1995;Istituto Nazionale di Statistica, 1999). Because of these factors the women living in remote areas of southern Italy are particularly at risk for diagnostic delay of breast cancer. Furthermore, in southern Italy women with lower educational levels were found to be subject to increased patient delays when compared with other female patients as previously demonstrated with a larger series (Montella et al., 1995). This is particularly true for women over 65 years of age, in whom breast cancer is more frequent and the prevention campaigns are accepted with some difficulties. This suggests that their help-seeking behaviour is sensitive to social influences and that there is a need for public educational programmes. This is supported by the work of other authors (Montella et al., 1995;Coates, 1999) even if numerous studies report that the association with these factors is not very evident (Ramirez et al., 1999).

In another study patient delay was a phenomenon mostly attributed to difficulties in receiving appointments, but this result may be biased by the fear of being reprimanded by the physician and by the natural tendency of patients to find a rational explanation for past events.

Unlike cases reported by other authors (Ramirez et al., 1999), the lumps in our study were not associated with patient delay as they were discovered by the patients themselves. Instead, our results strongly emphasize the role of the professional figure involved in the diagnostic process: a significant risk of >3 months delay was found to be associated with the intervention of physicians other than senologists. This risk was at its highest when other specialists were involved (especially gynaecologists and general surgeons). Even after eliminating the 94 asymptomatic patients (preventive mammography) this result shows no significant change. This is because the specialist (senologist) is able to evaluate the patient even in the absence of a mammography. The relationship appears more favourable when it involves the family doctor (GP), as the latter probably knows the patient's case history. To the best of our knowledge no other studies have focused on this particular aspect of diagnostic delay in breast cancer patients.

Furthermore, our results are consistent with the frequent involvement of non-senologists in breast cancer malpractice lawsuits (Mant et al., 1987;Rossi et al., 1990;Richards et al., 1999;Ramirez et al., 1999). These findings support the need for more coordination between primary and specialist care teams, as well as greater diffusion of current medical knowledge among non-senologists.

Two main factors could explain the association we found between younger age and medical delay. The first factor is the greatly reduced attention that is paid to the examination of young women. This is due to the fact that they represent a low risk group. The second, and perhaps most significant factor, is the inadequate diagnostic capacity of mammographies of young women's breasts (<40 years old) (Duffy et al., 1997;Baines and Miller, 1997).

The influence of delay on tumour size is clearly seen in the results concerning overall delay because the admission of women with large tumours greatly reduces survival rates (Richards et al., 1999;Ramirez et al., 1999;Sainsbury et al., 1999). Over the last 10 years there has been an improvement in the early diagnosis for breast cancer, including in south Italy (Montella et al., 2000), but from our study it emerges that subjective variables such as patient's age and level of education, doctor's specialization and the practice of preventive mammography have more influence on the length of the diagnostic process.

Breast cancer survival rates have increased, although they vary from country to country (Tomatis, 1997;Gatta et al., 1997;Sant et al., 1998), and these rates can be further improved by information campaigns, especially for women with a low level of education. For this reason the present study contains information that could be of great value to the Italian Health Authorities. This is not only because of the large numbers of patients facing delay (214/550, 39%, of woman with operable breast cancer had an overall delay in excess of 3 months) but also because of the recognition of the causes (old age concerning patient delay, young age for medical and overall delay, low education level, type of physician at first consultation). Many of these factors can be resolved through training programmes for members of the medical profession and education for women, as has been suggested in other studies (Bedell et al., 1995). To improve the efficiency of information delivery, educational training programmes for women should be developed specifically for each age category (i.e. in older women more attention to education in prevention; in younger women clear information about mammography and specialized structures).

Uncited references

GIVIO (Interdisciplinary Group for Cancer Care Evaluation) 1986, IUAC (International Union Against Cancer) 1992, Regione Campania, Assessorato Ricerca Scientifica ed Informatica, 1995


This study was conducted with the contribution of the Italian League Against Cancer.


1. Avtrandilov G , Berg JW , Carleschi R et al. ( 1990 ). International Classification of Diseases for Oncology, 2nd edn World Health Organization , Geneva .
2. Baines CJ , Miller AB ( 1997 ). Mammography versus clinical examination of the breasts . J Natl Cancer Inst Monogr 22 : 125 – 129 .
3. Bedell MB , Wood ME , Lezotte DC et al. ( 1995 ). Delay in diagnosis and treatment of breast cancer: implications for education . J Cancer Educ 10 : 223 – 228 .
4. Breslow NE , Day NE ( 1980 ). Statistical Methods in Cancer Research, Vol I International Agency for Research on Cancer , Lyon .
5. Caplan LS , Helzlsouer KJ , Shapiro S et al. ( 1996 ). Reasons for delay in breast cancer diagnosis . Prev Med 25 : 18 – 24 .
6. Carter CL , Allen C , Henson DE ( 1989 ). Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases . Cancer 63 : 181 – 187 .
7. Coates AS ( 1999 ). Breast cancer: delays, dilemmas, and delusions . Lancet 353 : 1112 – 1113 .
8. Duffy SW , Day NE , Tabar L et al. ( 1997 ). Markov models of breast tumor progression: some age-specific results . J Natl Cancer Inst Monogr 22 : 93 – 97 .
9. Gatta G , Buiatti E , Conti E et al. ( 1997 ). Variations in the survival of adult cancer patients in Italy . Tumori 83 : 497 – 504 .
10. GIVIO (Interdisciplinary Group for Cancer Care Evaluation). ( 1986 ). Reducing diagnostic delay in breast cancer. Possible therapeutic implications . Cancer 58 : 1756 – 1761 .
11. IUAC (International Union Against Cancer). ( 1992 ). TNM Classification of Malignant Tumours, 4th edn Springer-Verlag , Berlin .
12. Istituto Nazionale di Statistica (1999). Annuario Statistico Italiano 1999, Rome.
13. Istituto Centrale di Statistica (1993). Classificazione dei Comuni Secondo le Caratteristiche Urbane e Rurali [in Italian]. ISTAT, Rome.
14. MacMahon B ( 1994 ). A biologic framework for the risk factors for breast cancer . Adv Oncol 10 : 3 – 9 .
15. Mant D , Vessey MP , Nell A et al. ( 1987 ). Breast self examination and breast cancer stage at diagnosis . Br J Cancer 55 : 207 – 211 .
16. Mantel N , Haenszel W ( 1959 ). Statistical aspects of analysis of data from retrospective studies of disease . J Natl Cancer Inst 22 : 719 – 748 .
17. Montella M , Biondi E , De Marco MIR et al. ( 1995 ). Sociodemographic factors associated with the diagnostic staging of breast cancer in Southern Italy . Cancer 76 : 1585 – 1590 .
18. Montella M , Buonanno M , Biondi E et al. ( 2000 ). Changing patterns of breast cancer stage at diagnosis in Southern Italy: hospital data as indicators of progressive changes . Prev Med 30 : 174 – 177 .
19. Nystrom L , Rutqvist LE , Wall L et al. ( 1993 ). Breast cancer screening with mammography: overview of the Swedish randomized trials . Lancet 341 : 973 – 978 .
20. Parkin DM, Muir CS, Whelan S, et al. (1997). Cancer Incidence in Five Continents, Vol VII. International Agency for Research on Cancer, World Health Organization, Lyon.
21. Ramirez AJ , Westcombe AM , Burgess CC et al. ( 1999 ). Factors predicting delayed presentation of symptomatic breast cancer: a systematic review . Lancet 353 : 1127 – 1131 .
22. Regione Campania, Assessorato Ricerca Scientifica ed Informatica (1995). Annuario Statistico Campano, Napoli.
23. Richards MA , Westcombe AM , Love SB et al. ( 1999 ). Influence on delay on survival in patients with breast cancer: a systematic review . Lancet 353 : 1119 – 1126 .
24. Richardson IL , Langholz B , Bernstem L et al. ( 1992 ). Stage and delay in breast cancer diagnosis by race, socioeconomic status, age and year . Br J Cancer 65 : 922 – 926 .
25. Rossi S , Cinini C , Di Pietro C et al. ( 1990 ). Diagnostic delay in breast cancer: correlation with disease stage and prognosis . Tumori 76 : 559 – 562 .
26. Sainsbury R , Johnston C , Haward B ( 1999 ). Effect on survival of delays in referral of patients with breast-cancer symptoms: a retrospective analysis . Lancet 353 : 1132 – 1135 .
27. Sant M , Capocaccia R , Verdecchia A et al. ( 1998 ). Survival of women with breast cancer in Europe: variation with age, year of diagnosis and country. The EUROCARE Working Group . Int J Cancer 77 : 679 – 683 .
28. Tomatis L ( 1997 ). Inequalities in survival from cancer . Tumori 83 : 505 – 507 .

Age; breast cancer; education; medical delay; medical profession; overall delay; patient delay; tumour size

© 2001 Lippincott Williams & Wilkins, Inc.