Delayed treatment and continued growth ofnonmelanoma skin cancer

Murad Alam, MD, MSCI,a,b,c Leonard H. Goldberg, MD, FRCP,h,l Sirunya Silapunt, MD,f Erin S. Gardner, MD,m

Sara S. Strom, PhD,g Alfred W. Rademaker, PhD,d,e and David J. Margolis, MD, MSCE, PhDi,j,k

Chicago, Illinois; Houston, Texas; Philadelphia, Pennsylvania; and St Louis, Missouri

From

of

ge

Fa

w

ve

D

Te

lo

an

A

Background: Patients may delay treatment for skin cancer for various reasons. Prior research on treatmentdelay has focusedonmelanoma rather than nonmelanoma skin cancer (NMSC),which ismuchmore common.

Objective: We sought to clarify the reasons for delay in the presentation for diagnosis and treatment of NMSC.

Methods: This was a prospective cohort study in a Mohs micrographic surgery private practice in an urbansetting. Eligible subjects were 982 consecutive patients presenting for Mohs micrographic surgery for NMSCbetween March and December 2005. No enrolled subjects were withdrawn for adverse effects. The surveywas a 4-page written self-administered questionnaire, eliciting patient medical history, skin cancer history,demographic information, initial and subsequent lesion size, and reasons for delay in presentation forevaluation and management. Outcome analyses addressed the: (1) frequency of specific reasons fordelayed presentation, as provided by self-report; (2) association between reasons for delay withdemographic or other patient-specific factors; and (3) change in lesion diameter from the time of detectionby the patient to the time of presentation to the doctor.

Results: Among the reasons for waiting, denial (including: thought it would go away, thought it wasn’timportant, too busy, thought they could self-treat, afraid it might be something dangerous) was the mostfrequent, accounting for 71% of cases; difficulty scheduling was associated with 10% of the instances ofdelay. Older patients (age[64 years) were more likely to wait to seek care than younger patients (odd ratio[OR] = 0.5; 95% confidence interval [CI] 0.4-0.7). Patients with a prior skin cancer were more likely to wait(OR = 1.4; 95% CI 1.1-2.0), as were patients with major life problems (OR = 2.6; 95% CI 1.6-4.3) and patientswith a history of any cancer (OR = 1.8; 95% CI 1.3-2.4). Weighted kappa analysis comparing tumor size atthe two time points yielded a kappa of 0.72 (SE = .02; 95% CI 0.68-0.77). When the data were separated intotwo groups, one including those tumors that had decreased in size or remained the same (698 patients),and those that had increased in size (120 patients), the median delay-to-presentation intervals associatedwith these two groups (2.5 vs 6.0 months, respectively) were found to be significantly different (P \.0001).

Limitations: This study may have limited generalizability to the extent that it reflects the characteristicsonly of the subpopulation of patients with skin cancer who eventually received treatment at a referral-based, urban, dermatology private practice. Overall, these patients may have been better insured and bemore affluent than the general population.

Conclusions: Denial is themost commonpatient-specific factor accounting for delayedpresentation forNMSCdiagnosis and treatment. Patients younger than65years,witha skincancerhistory,withmajor lifeproblems, andwitha historyof any cancerweremost likely towait to see adoctor. Therewas a significant increase in tumor sizefrom the time when tumors were noticed by patients to the time when patients presented to a physician.Increased delay was associated with increased tumor growth. ( J Am Acad Dermatol 2011;64:839-48.)

Key words: basal cell carcinoma; delay; denial; nonmelanoma skin cancer.

the Section of Cutaneous and Aesthetic Surgery, Department

Dermatology,a the Departments of Otolaryngology,b Sur-

ry,c and Preventive Medicine,d and the Biostatistics Core

cility, Robert H. Lurie Comprehensive Cancer Center,e North-

estern University, Chicago; Department of Dermatology, Uni-

rsity of TexaseHoustonf; Department of Epidemiologyg and

ivision of Dermatology, Department of Medicine,h University of

xas, MD Anderson Cancer Center; Departments of Dermato-

gyi and Epidemiology,j and Center for Clinical Epidemiology

d Biostatistics,k University of Pennsylvania; DermSurgery

ssociates, Houstonl; and Aesthetic Derm Surgery, St Louis.m

Funding sources: None.

Conflicts of interest: None declared.

Accepted for publication June 6, 2010.

Reprint requests: Murad Alam, MD, MSCI, Department of

Dermatology, Northwestern University, 676 N St Clair, Suite

1600, Chicago, IL 60611. E-mail: m-alam@northwestern.edu.

Published online November 9, 2010.

0190-9622/$36.00

ª 2010 by the American Academy of Dermatology, Inc.

doi:10.1016/j.jaad.2010.06.028

839

J AM ACAD DERMATOL

MAY 2011

840 Alam et al

Approximately 3 to 4 million incident cases of

basal cell carcinoma and 100,000 of cutaneoussquamous cell carcinoma occur each year in theUnited States. Although most nonmelanoma skincancers (NMSCs) are not fatal, they are capable ofdestroying facial sensory organs, including the nose,ear, lips, and eyelids.

CAPSULE SUMMARY

d When patients with nonmelanoma skincancer delay presenting for diagnosisand treatment, the most frequentpsychologic association appears to bedenial regarding the severity of theircondition.

d Older patients and patients without afamily history of skin cancer wererelatively more likely to delay presentingfor diagnosis and treatment fornonmelanoma skin cancer.

d Delay to diagnosis intervals weresignificantly greater for patients whoreported visible growth of theirnonmelanoma skin cancers, as comparedwith patients who did not report tumorgrowth.

There is often a delay be-tween the clinical emergenceof a NMSC and the point intime at which the patientpresents for definitive diag-nosis and treatment. It hasbeen hypothesized that thisdelay can result in anincrease in the size of thecancers. Treatment delaymay potentially lead to largerand more disfiguring exci-sions. Increase in the size ofinvasive cutaneous squa-mous cell carcinomas isassociated not only withmore local organ destruc-tion, but also a heighteneddanger of metastatic spread.

Diverse patient-specificfactors appear to be respon-sible for the delay in thetreatment of cancers in gen-

eral, and skin cancer in particular.1-19 Patients maymisunderstand the signs and symptoms of theirlesions, which they consider unimportant, or likelyto go away. They may also experience fear or denial.Alternatively, other concurrent major life events mayresult in competing demands for their time thatprevent prompt attention to their medical needs.Patients with physical limitations are reliant onothers for care and may be unwilling or unable toimpose on these caregivers to help them reach aphysician. Indigent patients or those patients with-out health insurance may believe they are unable toafford coverage. In some cases, there may bephysician-related scheduling delays.

The evidence for the reasons for delayed treat-ment remains sparse. Virtually all past studies ondelays in skin cancer treatment have examinedmelanoma1,4-9,11-13,16 rather than NMSC. Much ofthe prior work elucidated only the psychologicproblems (emotions and fears) that prevent patientsfrom timely management of their tumors; physical,social, and financial obstacles have received littlescrutiny.

The purpose of this questionnaire study was toattempt to clarify the reasons for delayed

presentation for diagnosis and treatment of NMSC.We administered a questionnaire to 982 consecutivepatients undergoing Mohs micrographic surgery todetermine the patient- and physician-specific rea-sons, including physical, financial, social, intellec-tual, and psychologic factors, to which patientsattribute delays in diagnosis or treatment. To the

extent that factors contribut-ing to treatment delay weresuccessfully identified, futureinvestigators may be able todevise targeted interventionsto reduce such delays.

METHODSSetting and patientpopulation

The site of the study was amultiphysician dermatologyprivate practice in Houston,TX. Patients eligible for thestudy were those previouslygiven a biopsy diagnosis ofNMSC who were presentingfor treatment of their skin can-cer by Mohs micrographicsurgery. Patients who pre-sented with more than onecancer requiring treatmentwere excluded from the study.

The median age of subjects was 64 years andthe mean age was 63.3 years; 57.8% were male;and 37.4% had a family history of skin cancer.Written informed consent was elicited from allpatients before questionnaire administration, withthe consent process in accordance with institu-tional review board recommendations. The studywas approved by the Western Institutional ReviewBoard (2002-1078).

Questionnaire measureThe questionnaire elicited patient demographic

information, general medical history, and dermato-logic and skin cancer history (Table I). The question-naire had 3 specific purposes: (1) to determinewhether the propensity to delay seeking diagnosisand treatment for NMSC was associated with demo-graphic and medical history factors (eg, age, gender,family history of skin cancer, history of other majorlife problems, marital status, family history of cancer,living arrangement, nature of initial presentation,person first noticing skin lesion, or specialty of doctorto whom the lesion was initially presented); (2) todetermine if the propensity to delay was associatedwith subjective patient-specific factors (eg, whether

Table I. Questionnaire assessing how long and why patients with skin tumors postpone diagnosis andtreatment

Question Answer choices

1. What is your gender? , Male , Female2. What is your marital status? , Single

, Married, Divorced, Widowed, Separated

3. What is your living arrangement? , Live alone, Live with spouse, Live with roommate, Live with parents, Live with children, Live at a care facility

4. Your early symptoms were (check all that apply): , Unattractive appearance, Pain, Ulcer, Bump, Bleeding, Itching, Other__________________

5. Who first noticed that you may have a problem skinlesion?

, You, A close friend, A friend, Your doctor, Someone else, who? ______

6. When did you first become aware of your problemlesion?

Month ______Year ______

7. How long since first noticing your problem lesion didyou wait before seeing a doctor?

Month(s)_____Year(s)_____

8. Why did you wait to go to the doctor? (provide only themost important reasons)

, Thought it would go away, Thought it was not important, Thought you could treat it yourself, Too busy, Physician scheduling delay, Had other personal/professional problems at the time, Had other medical problems at the time, Was afraid it might be something dangerous, Had a bad experience with doctors or with surgery in the

past, Financial reasons/insurance problems Comments:______

9. Is this your first skin cancer? , Yes , No10. Do you have a family history of SKIN cancer in particular? , Yes , No11. Were you concerned about the financial burden to

yourself or your family to treat your lesion?, Yes , No

12. Did you have major life problems around the time younoticed your lesion?

, Yes , No

13. Before your surgery, how concerned were you aboutdisfigurement or scarring after surgery?

, Most concerned, Very concerned, Moderately concerned, Barely concerned, Not concerned at all

14. Before your surgery, how concerned were you about theside effects of surgery?

, Most concerned, Very concerned, Moderately concerned, Barely concerned, Not concerned at all

Continued

J AM ACAD DERMATOL

VOLUME 64, NUMBER 5

Alam et al 841

Table I. Cont’d

Question Answer choices

15. How big was your lesion when you first noticed it was aproblem?

, Small as a pinhead, About the size of a pimple, Dime size, Nickel size, Bigger than a nickel

16. How big was your lesion when you first showed it to adoctor?

, Small as a pinhead, About the size of a pimple, Dime size, Nickel size, Bigger than a nickel

J AM ACAD DERMATOL

MAY 2011

842 Alam et al

the patient: had financial concerns/insurance prob-lems; was concerned about disfigurement/scarringafter surgery; or was concerned about the side effectsof surgery); and (3) to compare the self-reported sizeof lesion at initial detection with the size at presen-tation to a physician (to make size determinationmore accessible to patients, patients were asked toselect from a list of common objects: pinhead, pim-ple, dime, nickel, bigger than nickel). The question-naire was not subjected to a formal assessment ofvalidity and reliability.

Statistical analysis‘‘Time waited’’ was defined as the patient-

reported time in months from when the lesion wasfirst noted to when a doctor first examined the lesion.This time was analyzed both as a continuous variableand as a categorical variable ( # 1 month,[1 month).The continuous time measure was compared be-tween categories of 6 demographic and patienthistory characteristics (each dichotomous) usingthe Wilcoxon rank sum test. The categorical timemeasure was related to those characteristics usingthe x2 test. For each characteristic, odds ratio (OR)was defined as the odds of waiting in the non-reference category divided by the odds of waiting inthe reference category. OR and confidence intervalsfor OR were calculated using logistic regressionanalysis. OR and confidence intervals adjusted forage and history of skin cancer were obtained usinglogistic regression. No adjustments were made formultiple testing. Weighted kappa was used to assessthe change in apparent tumor size over time.Specifically, this statistic was used to evaluate theextent to which the initial sorting of tumors into the 5ordinal categories (small as a pinhead, about the sizeof a pimple, dime size, nickel size, bigger than anickel) associated with size (ie, at the time of prob-lem lesion detection by the patient) concurred withthe final sorting of these tumors by size (ie, at the

time of presentation to a physician). Lesion size wascompared between the two time points usingMcNemar test. Waiting time was compared betweenpatients whose lesions increased and those whosedid not increase using the Wilcoxon rank sum test.

RESULTSData collection and response rate

The questionnaire was administered to 982 con-secutive patients presenting for Mohs micrographicsurgery for NMSC between March and December2005. On average, patients spent 30 to 60 minutescompleting this questionnaire. Patients presentingfor surgery accompanied by friends, family mem-bers, or both were permitted to confer with theseothers during the process of responding to question-naire items. Elderly patients and those with limitedsensory and writing capacities were assisted bynursing staff members, who slowly read the ques-tionnaire items and transcribed the patient re-sponses. When necessary (in approximately 5% ofcases), nursing staff or investigators explained ques-tionnaire items in greater detail to patients who wereunclear about their meaning; patient responses werethen recorded.

Because of incomplete responses, there weremissing data across questionnaire items. Waitingtime was available for 860 patients, but analysesrelating another characteristic to waiting time werebased on smaller sample sizes, depending on theamount of missing data for the other characteristic.Lesion size data at both times (when initially noticed,at doctor examination) were available for 918patients.

Association of waiting time with demographicand patient history characteristics

Median waiting time was related to demographicand patient history characteristics (Table II). Therewas longer delay in younger patients (P \ .0001),

Table II. Time waited (months) to see a doctor by demographic and patient history characteristics of skincancer cases

Min Q1 Median Q3 Max P value*

Age, y# 64 (n = 458) 0 1 4 10 240[64 (n = 402) 0 1 2 6 180 \.0001

Family history of skin cancerNo (n = 522) 0 1 3 6.5 240 .35Yes (n = 318) 0 1 3 7 186

GenderMale (n = 509) 0 1 3 6 240Female (n = 351) 0 1 3 9 192 .10

Major life problems when lesion first noticedNo (n = 712) 0 1 3 6 240 \.0001Yes (n = 130) 0 2 6 12 200

Marital statusMarried (n = 653) 0 1 3 6 240Not currently married (n = 207) 0 1 2 8 200 .89

Family history of cancerNegative (n = 386) 0 1 2 6 240 .006Positive (n = 436) 0 1 3 8 192

Max, Maximum; median, 50th percentile; min, minimum; Q1, 25th percentile; Q3, 75th percentile.

n = 860 (n smaller for some characteristics because of missing data on those characteristics).

*Wilcoxon rank sum test.

J AM ACAD DERMATOL

VOLUME 64, NUMBER 5

Alam et al 843

patients with a major life problem when lesion firstbecame noticed (P \ .0001), and patients with afamily history of cancer (P \ .006). There was noassociation of delay time with family history of skincancer, gender, or marital status.

Association of delay in presentation to aphysician with demographic and patienthistory characteristics

After noticing a suggestive lesion, most patientsdid wait to see to a physician. In total, 581 (67.6%)patients reported waiting (defined as [1 monthbetween first noticing lesion and doctor examina-tion), and 279 (32.4%) did not wait. The medianwaiting time among patients who did wait was 6months, with an interquartile range of 3 to 12months; the mean waiting time was 12.5 months.Of patients, 88% presented to a physician within1 year of noticing their lesions, but 8.3% said theywaited 3 years or more, with 1.2% having waitedmore than 10 years.

Waiting time as a dichotomous variable wasrelated to demographic and patient history charac-teristics (Table III). There was more waiting inyounger patients (75% vs 59%, P \ .0001), patientswith a family history of skin cancer (72% vs 64%, P =.002), patients with a major life problem when lesionfirst was noticed (83% vs 65%, P \ .0001), andpatients with a family history of cancer (73% vs 61%,

P = .0001). There was no association of waiting withgender or marital status. Using a multivariate logisticregression model simultaneously adjusting for ageand skin cancer history, the significance of the resultsfor gender, major life problems, marital status, andfamily history of cancer remained unchanged.

Association of delay in presentation to aphysician with other patient characteristics

A comparison of waiting groups on other patientcharacteristics (Table IV) indicated that approxi-mately 75% of patients in both the waiting and not-waiting groups lived with a spouse, but fewer whowaited lived alone whereas more who waited hadother living arrangements (P = .014). In both waitingand not-waiting groups, the 3 most common initialsymptoms reported by patients were bump, unat-tractive appearance, and bleeding, in order of de-scending frequency. In the waiting group, morepatients found their skin cancer lesion themselves(77% vs 64%, P = .0001). The 3 most commonly listedreasons why patients waited to see their doctor werethought it would go away (36%), thought it wasn’timportant (24%), and couldn’t get an earlier appoint-ment (10%). Of the patients, 86% in the waitinggroup and 89% in the non-waiting group noted‘‘dermatologist’’ as the specialty of the doctorto whom they first presented. Monetary concerns(financial reasons/insurance problems) did not differ

Table III. Waiting status by demographic and clinical characteristics of skin cancer cases

Wait* (N = 581) Not waity (N = 279) P valuez cOR (95% CI) aOR (95% CI)

Age (mean 6 SE), y 60.86 6 0.56 67.04 6 0.77# 64 (n = 458) 342 (75%) 116 (25%) \.0001 1.0[64 (n = 402) 239 (59%) 163 (41%) 0.5 (0.4-0.7) e

Family history of skin cancerNo (n = 522) 336 (64%) 186 (36%) .002 1.0Yes (n = 318) 230 (72%) 88 (28%) 1.4 (1.1-2.0) e

GenderMale (n = 509) 335 (66%) 174 (34%) 1.0Female (n = 351) 246 (70%) 105 (30%) .21 1.2 (0.9-1.6) 1.2 (0.9-1.6)

Major life problems when lesion first noticedNo (n = 712) 463 (65%) 249 (35%) 1.0Yes (n = 130) 108 (83%) 22 (17%) \.0001 2.6 (1.6-4.3) 2.6 (1.5-4.2)

Marital statusMarried (n = 653) 439 (67%) 214 (33%) 1.0Not currently married (n = 207) 142 (69%) 65 (31%) .73 1.1 (0.8-1.5) 1.2 (0.8-1.7)

Family history of cancerNegative (n = 386) 235 (61%) 151 (39%) 1.0Positive (n = 436) 320 (73%) 116 (27%) .0001 1.8 (1.3-2.4) 1.6 (1.1-2.9)

aOR, Odds ratio adjusted for age and family history of skin cancer; CI, confidence interval; cOR, crude odds ratio.

Odds = odds of waiting.

n = 860 (n smaller for some characteristics because of missing data on those characteristics).

*Time to see doctor [ 1 mo.yTime to see doctor # 1 mo.zx2.

J AM ACAD DERMATOL

MAY 2011

844 Alam et al

between waiting groups. Those who waited weremore concerned about disfigurement/scarring aftersurgery (P = .037) whereas concerns about sideeffects of surgery did not differ between the groups.

Association of delay in presentation fortreatment with an increase in tumor size

A total of 918 patients provided information aboutthe size of their tumor at each of two time points.Overall, 12 tumors decreased in size, 778 stayed inthe same category, 113 increased by 1 category, 9increased by 2 categories, 5 increased by 3 cate-gories, and 1 increased by 4 categories.

Weighted kappa analysis comparing the two timepoints yielded a kappa of 0.72 (SE = .02), with aconfidence interval of 0.68 to 0.77. It was concluded,based on McNemar test (P\.0001) and the observeddirection of the change, that there was a significantincrease in tumor size from the time when tumorswere noticed by patients to the time when patientspresented for treatment.

Data were then separated into two groups, oneincluding tumors that had decreased in size orremained the same (790 records), and the other fortumors that had increased in size (128). For 698 of the790 tumors in the decreased/same group, the mediantime between each pair of tumor size measurementswas 2.5 months. For 120 of the 128 records in the

increased group, the median time between each pairof tumor size measurements was 6.0 months (P \.0001). It was concluded that those who presentedwith increased tumor sizes had waited longer thanthose who presented with tumors unchanged in size.On average, skin cancers grew while patients werewaiting, with the average lesion enlarging frompimple sized (2-3 mm) to between pimple anddime sized (10 mm).

DISCUSSIONThe results of this study indicate that most patients

with NMSCs tended to delay presenting to physiciansfor definitive diagnosis and treatment. Those whowere younger ( # 64 years), those with a skin cancerhistory, those with major life problems, and thosewith a family history of any cancer waited relativelylonger to seek medical care after noticing skinlesions. Overall, waiting to present to a physicianwas associated with a significant increase in lesiondiameter.

The self-reported measures studied to track tumorsize were subject to patient bias. Patients may haveerroneously remembered their tumors to have beensmaller than they were initially (recall bias), andpatients may have been less precise in their initialtumor measurements than their later measurements(measurement bias or information bias). These

Table IV. Waiting status by other patient factors

Variables Wait (N = 581) Not wait (N = 279) P value

Living arrangement (n = 858) .014Live with spouse 437 (75%) 209 (75%)Live alone 95 (16%) 60 (22%)Other living arrangements 47 (8%) 10 (4%)

Initial presentation (n = 847) \.0001Bump 189 (33%) 82 (30%)Unattractive appearance 150 (26%) 35 (13%)Bleeding 74 (13%) 25 (9%)Ulcer 27 (5%) 16 (6%)Itching 30 (5%) 17 (6%)Pain 18 (3%) 9 (3%)Other 86 (15%) 89 (33%)

First person who noticed lesion (n = 859) .0001Self 446 (77%) 179 (64%)

Other 135 (23%) 100 (36%)Reason wait to see doctor (n = 550 in wait group)

Thought it would go away 198 (36%)Thought it wasn’t important 130 (24%)Couldn’t get an earlier appointment 57 (10%)Too busy 30 (5%)Thought they could treat themselves 26 (5%)Had other personal/professional problems at the

time20 (4%)

Had other medical problems at the time 16 (3%)Physician scheduling delay 13 (2%)Afraid it might be something dangerous 6 (1%)Financial reasons/insurance problems 6 (1%)Previous bad experience with doctors 1 (\1%)Other miscellaneous reasons 47 (9%)

Specialty of initial doctor (n = 851) .27Dermatologist 497 (86%) 246 (89%)Other 78 (14%) 30 (11%)

Concern of financial burden (n = 851) .07Yes 100 (17%) 35 (13%)No 474 (83%) 242 (87%)

Concern about disfigurement/scarring aftersurgery (n = 853)

.037

Most concerned 55 (10%) 17 (6%)Very concerned 58 (10%) 18 (6%)Moderately concerned 171 (30%) 87 (31%)Barely concerned 133 (23%) 57 (20%)Not concerned at all 158 (27%) 99 (36%)

Concern about side effects of surgery (n = 814) .67Most concerned 30 (5%) 11 (4%)Very concerned 48 (9%) 17 (7%)Moderately concerned 149 (27%) 69 (27%)Barely concerned 143 (26%) 66 (25%)Not concerned at all 184 (33%) 97 (37%)

Size of lesion when first noticed (n = 828) .07Small as a pinhead 53 (9%) 33 (13%)About the size of a pimple 424 (75%) 178 (68%)Dime size 72 (13%) 36 (14%)Nickel size 13 (2%) 6 (2%)Bigger than a nickel 5 (1%) 8 (3%)

Continued

J AM ACAD DERMATOL

VOLUME 64, NUMBER 5

Alam et al 845

Table IV. Cont’d

Variables Wait (N = 581) Not wait (N = 279) P value

Size of lesion when first showed todoctor (n = 823)

.0008

Small as a pinhead 26 (5%) 32 (12%)About the size of a pimple 384 (68%) 171 (66%)Dime size 117 (21%) 40 (15%)Nickel size 24 (4%) 8 (3%)Bigger than a nickel 13 (2%) 8 (3%)

n = 860 (n smaller for some characteristics because of missing data on those characteristics).

J AM ACAD DERMATOL

MAY 2011

846 Alam et al

caveats notwithstanding, we are not aware of afeasible method of gauging initial tumor size thatdoes not rely on patient self-reporting. One benefitof using patient self-reporting for both initial andsubsequent measurements was the minimization ofinterrater measurement differences. Further, facevalidity studies confirmed that the self-reportingsize categories were understandable by patientswith grade-school education and were sufficientlydifferent from each other to permit discriminationbetween adjacent categories. Although patients’ tu-mor size estimates were only of visible cutaneousmanifestations, there does not appear to be a plau-sible rationale for a systematic overestimate of thelatter versus the initial size estimates. Although thetumor size self-reports were obtained after theemergence of the tumors, the reports were obtainedbefore the surgeries; thus patients were not subject torecall biases deriving from the size of their finalwound defect or duration of surgery (eg, a patientwith a protracted, complex surgery could not beinfluenced by this fact to recall a larger initial tumorsize or a greater duration of delay before presenta-tion for treatment).

For clarification, the waiting interval began whenthe patient first noticed a problem lesion. That is, theinitial time was not when a spot was first noted at theultimate site, but rather when, based on symptoms,persistence, or unusual appearance, the patientcame to the conclusion that something troublesomewas occurring. The waiting interval ended when thepatient actually saw a physician, not when thepatient began the attempt to make a doctor appoint-ment. Because the wait for a dermatologist or otherdiagnosing physician can be considerable, this def-inition of end time may have artifactually lengthenedthe waiting period by 1 month or longer.

Subjects were asked to estimate the size of thelesion when they presented to the initial diagnosingphysician, not when they presented for treatment. Inaddition, most of the cancers were biopsied beforepresentation for treatment to the dermatology prac-tice where the survey was administered. This creates

the possibility that in some cases the type or size ofbiopsy specimen (eg, broad shave) could haveinfluenced the patient’s perception of the tumorsize at that time. However, most patients had rela-tively diminutive shave biopsy specimens that weresignificantly smaller than the histologically verifiedtumor extent, as delineated by the final wounddefect after Mohs micrographic surgery. So it couldbe argued that if patients used biopsy defect size as aproxy for late tumor size, they were more likely tounderestimate size as a consequence of small biopsyspecimens, than overestimate it based on largebiopsy specimens.

The reasons for the delay in diagnosis and treat-ment of skin cancer have previously not been wellstudied. The most relevant investigations that havebeen done are those pertaining to melanoma, and toa lesser extent, other types of cancer.

The results of this study indicate that most of thereported reasons for patient delay in presentation formanagement of NMSC are forms of denial of theirillness.18-25 Denial may be a normal response tohealth concerns, with patients minimizing the psy-chologic burden of their illness, or it may be amaladaptive coping mechanism. Denial can alsointerfere with obtaining treatment, and result indelay in presenting to a physician and noncompli-ance with treatment plans. In the patients understudy, denial resulted in delayed treatment of inher-ently treatable NMSC, with correspondingly in-creased morbidity. In some cases, denial may notbe a psychologic defense mechanism but indicativeof a lack of recognition or understanding of theproblem. However, because the delay period in thisstudy was defined as beginning at the point at whichpatients realized that they had a ‘‘lesion’’ that was a‘‘problem,’’ it seems reasonable that during theinterval examined, most patients were at some levelaware that they had a serious concern but were notalways able to cope effectively with this information.

Although most of the patients in this study initiallypresented to dermatologists, patients with NMSCsmay first present to internists, family physicians,

J AM ACAD DERMATOL

VOLUME 64, NUMBER 5

Alam et al 847

gynecologists, or other primary care physicians. Thisstudy may have limited generalizability to the extentthat it best reflects the characteristics only of thesubpopulation of patients with skin cancer who areinitially seen by a dermatologist and then referred toa dermatologic surgeon for treatment. Moreover,although the dermatologic surgery practice in thisstudy did also serve a nearby rural, less affluentpopulation, the majority of the patients seen in thispractice were likely more affluent, better educated,and in possession of better insurance than thegeneral population.

By educating patients about the need for earlydetection and management of NMSC and biopsyingsuggestive lesions early in their clinical course,physicians can reduce the personal and societalburden of NMSC. Early diagnosis and treatment willusually result in cure and in markedly reducedfunctional loss and disfigurement. Further researchis likely to improve understanding of delayedpresentation for NMSC26-28 and melanoma skincancer.29-37

Summary and conclusionsDelayed treatment of NMSC may result in tumor

enlargement, functional loss, and the need for largerexcisions that may impact cosmetic and motor func-tion. Early detection and treatment of NMSC mark-edly reduces morbidity and is associated with a highlikelihood of cure.

The current study explores the reasons patientsdelay seeking care for NMSC.26-28 In this cohort,younger patients, patients with a skin cancer history,patients with major life problems, and those with afamily history of any cancer waited longer for diag-nosis and treatment of their skin cancers. Mostprominent among the underlying reasons for delayappeared to be the patients’ psychologic unwilling-ness to accept the need for care (ie, denial) ratherthan physical impediments such as difficulty inscheduling or major concurrent life or health prob-lems. During the period of delay, tumors grew insize, and the longest delays were associated withrelatively greater increases in tumor size.

This study suggests that the timely management ofNMSC may dramatically improve associated morbid-ity and mortality. One potential intervention wouldbe patient education programs that specifically ad-dress the problem of denial and communicate thepractical benefits of early NMSC detection. If patientsunderstand that early diagnosis and treatment canlead to briefer surgeries, smaller scars, and decreasedmorbidity, they may be less inclined to delay seeing aphysician.

REFERENCES

1. Henrikus D, Girgis A, Redman S, Sanson-Fisher RW. A commu-

nity study of delay in presenting with signs of melanoma to

medical practitioners. Arch Dermatol 1991;127:356-61.

2. Hollows P, McAndrew PG, Perini MG. Delays in the referral and

treatment of oral squamous cell carcinoma. Br Dent J 2000;

188:262-5.

3. Burgess CC, Ramirez AJ, Richards MA, Love SB. Who and what

influences delayed presentation in breast cancer? Br J Cancer

1998;77:1343-8.

4. Blum A, Brand CU, Ellwanger U, Schlagenhauff B, Stroebel W,

Rassner G, et al. Awareness and early detection of cutaneous

melanoma: an analysis of factors related to delay in treatment.

Br J Dermatol 1999;141:783-7.

5. Doherty VR, MacKie RM. Reasons for poor prognosis in British

patients with cutaneous malignant melanoma. Br Med J (Clin

Res Ed) 1986;292:987-9.

6. Krige JE, Isaacs S, Hudson DA, King HS, Strover RM,

Johnson CA. Delay in the diagnosis of cutaneous malignant

melanoma: a prospective study in 250 patients. Cancer

1991;68:2064-8.

7. Temoshok L, DiClemente RJ, Sweet DM, Blois MS, Sagebiel RW.

Factors related to patient delay in seeking medical attention

for cutaneous malignant melanoma. Cancer 1984;54:3048-53.

8. Rampen FH, Rumke P, Hart AA. Patients’ and doctors’ delay in

the diagnosis and treatment of cutaneous melanoma. Eur J

Surg Oncol 1989;15:143-8.

9. Holden R, Damato BE. Preventable delays in the treatment of

intraocular melanoma. Eye 1996;10:127-9.

10. Chakravorty SG, Chakravorty SS, Patel RR, DeSouza CJ,

Doongaji DR. Delay in specialist consultation in cancer

patients. Indian J Cancer 1993;30:61-6.

11. Richard MA, Grob JJ, Avril MF, Delaunay M, Gouvernet J,

Wolkenstein P, et al. Delays in diagnosis and melanoma

prognosis (I): the role of patients. Int J Cancer 2000;89:271-9.

12. Cassileth BR, Temoshok L, Frederick BE, Walsh WP, Hurwitz S,

Guerry D, et al. Patient and physician delay in melanoma

diagnosis. J Am Acad Dermatol 1988;18:591-8.

13. Oliviera SA, Christos PJ, Halpern AC, Fine JA, Barnhill RL,

Berwick M. Patient knowledge, awareness, and delay in

seeking medical attention for malignant melanoma. J Clin

Epidemiol 1999;52:1111-6.

14. Love N. Why patients delay seeking care for cancer symptoms.

What you can do about it. Postgrad Med 1991;89:151-2.

15. Michielutte R, Dignan MB, Sharp PC, Boxley J, Wells WB. Skin

cancer prevention and early detection practices in a sample of

rural women. Prev Med 1996;25:673-83.

16. Temoshok L, DiClemente RJ, Sweet DM, Blois MS, Sagebiel RW.

Prognostic and psychological factors related to delay behavior

in patients with cutaneous malignant melanoma. Prog Clin

Biol 1984;156:169-79.

17. Blackford S, Roberts D, Salek MS, Finlay A. Basal cell carcino-

mas cause little handicap. Qual Life Res 1996;5:191-4.

18. Wool MS. Understanding denial in cancer patients. Adv

Psychosom Med 1988;18:37-53.

19. Moyer A, Levine EG. Clarification of the conceptualization and

measurement of denial in psychosocial oncology research.

Ann Behav Med 1998;20:149-60.

20. Henderson JG. Denial and repression as factors in the delay of

patients with cancer presenting themselves to the physician.

Ann N Y Acad Sci 1966;125:856-64.

21. Salander P, Windahl G. Does ‘‘denial’’ really cover our everyday

experiences in clinical oncology? A critical view from a

psychoanalytic perspective on the use of ‘‘denial.’’ Br J Med

Psychol 1999;72:267-79.

J AM ACAD DERMATOL

MAY 2011

848 Alam et al

22. Brock G, Gurekas V, Deom P. Denial among cancer patients:

tips and traps. Can Fam Physician 1993;39:2581-4.

23. Kreitler S. Denial in cancer patients. Cancer Invest 1999;17:

514-34.

24. Phelan M, Dobbs J, David AS. ‘I thought it would go away’:

patient denial in breast cancer. J R Soc Med 1992;85:206-7.

25. Greer S. The management of denial in cancer patients.

Oncology (Huntingt) 1992;6:33-6.

26. Maser E, Berg D, Solish N. Changes in patient perception and

behavior following Mohs micrographic surgery. J Cutan Med

Surg 2001;5:14-7.

27. Freidenbergs I. Psychosocial management of patients with

cutaneous cancers: a preliminary report. J Dermatol Surg

Oncol 1981;7:828-30.

28. Robinson JK, Rademaker AW. Skin cancer risk and sun

protection learning by helpers of patient with nonmelanoma

skin cancer. Prev Med 1995;24:333-41.

29. Temoshok L, Heller BW, Sagebiel RW, Blois MS, Sweet DM,

DiClemente RJ. The relationship of psychosocial factors to

prognostic indicators in cutaneous malignant melanoma.

J Psychosom Res 1985;29:139-53.

30. Creagen ET. Psychosocial issues in oncologic practice. Mayo

Clin Proc 1993;68:161-7.

31. Kneier AW. The psychological challenges facing melanoma

patients. Surg Clin North Am 1996;76:1413-21.

32. Gibertini M, Reintgen DS, Baile WF. Psychosocial aspects of

melanoma. Ann Plast Surg 1992;28:17-21.

33. Richard MA, Grob JJ, Avril MF, Delaunay M, Gouvernet J,

Wolkenstein P, et al. Delays in diagnosis and melanoma

prognosis (II): the role of doctors. Int J Cancer 2000;89:280-5.

34. Metzger S, Ellwanger U, Stroebel W, Schiebel U, Rassner G,

Fierlbeck G. Extent and consequences of physician delay in

the diagnosis of acral melanoma. Melanoma Res 1998;8:

181-6.

35. Sollner W, Zingg-Schir M, Rumpold G, Mairinger G, Fritsch P.

Need for supportive counselingethe professionals’ versus

the patients’ perspective: a survey in a representative sample

of 236 melanoma patients. Psychother Psychosom 1998;67:

94-104.

36. Richard MA, Martin S, Gouvernet J, Folchetti G, Bonerandi JJ,

Grob JJ. Humor and alarmism in melanoma prevention: a

randomized controlled study of three types of information

leaflet. Br J Dermatol 1999;140:909-14.

37. Rhodes AR. Public education and cancer of the skin: what do

people need to know about melanoma and nonmelanoma

skin cancer? Cancer 1995;75(Suppl):613-36.