University of Southern California Dissertations and Theses

The role of physicians in the late-stage diagnosis of melanoma among: a population-based study in Los Angeles County

(USC Thesis Other)

The role of physicians in the late-stage diagnosis of melanoma among: a population-based study in Los Angeles County

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content i
The Role of Physicians in the Late-Stage Diagnosis of Melanoma Among Hispanics
A Population-Based Study in Los Angeles County
by
Rosario Agüero Ureta, M.D.
A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(CLINICAL AND BIOMEDICAL INVESTIGATION S)
May 2023
Copyright 2023 Rosario Agüero Ureta
ii

Table of Contents

List of Tables ................................................................................................................................. iii
List of Figures ................................................................................................................................ iv
Abbreviations .................................................................................................................................. v
Abstract .......................................................................................................................................... vi
Chapter 1: Introduction ................................................................................................................... 1
Chapter 2: Materials & Methods..................................................................................................... 3
Population and data sources ........................................................................................................ 3
Survival time ............................................................................................................................... 3
Statistical analysis ....................................................................................................................... 3
Variables...................................................................................................................................... 4
Chapter 3: Results ........................................................................................................................... 6
Population characteristics. ........................................................................................................... 6
Patients..................................................................................................................................... 6
Physicians ................................................................................................................................ 9
Survival Trends ........................................................................................................................... 9
Multivariate analysis ................................................................................................................. 11
Chapter 4: Discussion ................................................................................................................... 13
Physician experience and specialty ........................................................................................... 14
Socioeconomic and insurance status ......................................................................................... 15
Stage .......................................................................................................................................... 16
Chapter 5: Strengths and Limitations ........................................................................................... 17
Chapter 6: Conclusion and next steps ........................................................................................... 20
References ..................................................................................................................................... 22
Appendices .................................................................................................................................... 25

iii

List of Tables
Table 1: Sociodemographic and clinical characteristics of patients with melanoma (ages 18+)
by ethnicity, LACSP, 2005 - 2019. ................................................................................................25
Table 2: Sociodemographic and clinical characteristics of patients with melanoma (ages 18+)
by physician volume, LACSP, 2005 - 2019. .................................................................................26
Table 3: Number of melanoma cases diagnosed (all races, ages 18+) for each attending
physician specialty by ethnicity and total number of physicians for each specialty with column
percentages, LACSP, 2005 - 2019. ...............................................................................................27
Table 4: Number of melanoma cases diagnosed (all races, ages 18+) for each attending
physician specialty by ethnicity and total number of physicians for each specialty with row
percentages, LACSP, 2005 - 2019. ...............................................................................................28
Table 5: Hazard ratio for risk of all-cause death by ethnicity group, in patients with melanoma,
LACSP, 2005-2019. .......................................................................................................................29
Table 6: Hazard ratio for risk of all-cause death, Hispanics and NHWs patients with melanoma,
LACSP, 2005-2019. .......................................................................................................................30

iv

List of Figures
Figure 1: Melanoma survival by race/ethnicity for all stages combined (ages 18+), Los Angeles
County, 2005-2019 ........................................................................................................................31
Figure 2: Melanoma survival by attending physician volume for Hispanics and Non-Hispanic
Whites (ages 18+), Los Angeles County, 2005-2019 ....................................................................32
Figure 3: Melanoma survival by stage at diagnosis for Hispanics and Non-Hispanic Whites
(ages 18+), Los Angeles County, 2005-2019 ................................................................................33
Figure 4: Melanoma survival by stage and attending physician volume for Hispanics and Non-
Hispanic Whites (ages 18+), Los Angeles County, 2005-2019. ....................................................34
Figure 5: Melanoma survival by primary physician´s specialty for Hispanics and Non-Hispanic
Whites (ages 18+), Los Angeles County, 2005-2019 ....................................................................35

v

Abbreviations

NHW Non-Hispanic white
LACSP Los Angeles County Cancer Surveillance Program
LAC Los Angeles County
SES Socioeconomic status
US United States
SEER National Cancer Institute’s Surveillance, Epidemiology,
and End Results

KM Kaplan-Meier
CI Confidence interval
HR Hazard ratios

vi

Abstract
Background: Melanoma is the most lethal form of skin cancer. Hispanics are less likely to be
diagnosed with melanoma, but they are diagnosed at later stages and have lower survival rates than
their non-Hispanic white (NHW) counterparts. Reasons for the later stage at diagnosis among
Hispanics are unknown and may be related to patient, clinician, healthcare systems issues, or a
combination of several factors. This study had two objectives:
Objective 1: Describe the sociodemographic and clinical characteristics, such as socioeconomic
status, marital status, insurance status, age, and stage at diagnosis, among others of Hispanic and
NHW patients with melanoma as well as the characteristics of patients seen by physicians with
high or low experience and different specialties.
Objective 2: Identify if the physician's specialty or level of experience (in terms of the number of
melanomas diagnosed) is related to melanoma survival among Hispanics and NHW.
Methods: We used data from the Los Angeles County Cancer Surveillance Program (LACSP) and
compared melanoma survival among Hispanics and NHW, considering the specialty and
experience level (defined as the number of melanomas diagnosed per year in a 15-year period,
with high volume defined as more than 15 melanomas, and low volume defined as 15 or less) of
the treating.
Results: The study identified a total of 20,744 melanomas diagnosed in adults in Los Angeles
County (LAC) between 2005 and 2019, with 2,077 cases of melanoma in Hispanic and 18,667 in
NHW patients. NHW melanomas were more likely to be treated by an experienced physician than
Hispanics (67.5% vs. 52.2). The 5-year survival for melanomas that had an inexperienced
physician as the attending doctor was significantly lower than for those melanomas treated by
experienced physicians (HR: 1.31, 95%CI 1.24-1.38). The worse outcome among Hispanics is
vii

mostly related to their lowest SES and not to physician experience or specialty. Hispanic
melanomas are concentrated in the lower socioeconomic status (SES) quintiles, while NHW
melanomas are in the highest SES quintiles. Patients from lower SES quintiles were less likely to
be diagnosed by an experienced physician. Low-experienced physicians diagnose melanomas in a
more advanced stage than highly experienced physicians.
Conclusion: Physicians’ experience and specialty are not associated with a worse outcome for
Hispanics with melanoma. However, it appears that being treated by a physician with less
experience or a specialty not commonly related to melanoma influences survival in the general
population. When studying survival trends in Hispanic populations with melanoma in LAC, the
specialty and experience of the physician does not account for survival variability; instead, lower
SES is found to correlate with higher mortality.

1
Chapter 1: Introduction

Melanoma is the most lethal form of skin cancer. Over the last twenty years, the global incidence
of melanoma has been progressively increasing over and above the impact of screening. (1)
Melanoma survival is closely correlated to the stage at diagnosis, with early-stage diagnosis having
improved outcomes. Screening and early intervention are widely accepted as a method to improve
survival.
Hispanics are the largest ethnic group in the United States and have rising rates of melanoma. (2,3)
While Hispanics are less likely to be diagnosed with melanoma than their NHW counterparts, they
also have higher mortality rates than NHWs, corresponding to a later stage of diagnosis. (4)
Delayed diagnosis in Hispanics could be attributed to a variety of modifiable factors, such as lack
of knowledge of self-screening and warning signs, poor access to appropriate medical care, or lack
of understanding of appropriate screening approaches or concerning findings by physicians, or a
combination of these factors. (5,6)
California is one of the states with the highest incidence rate of melanoma in the United States,
with 10,269 new cases diagnosed in 2022. (4) California is also home to a large Hispanic
community, particularly in LAC, where according to the latest United States (US) census, 49.1%
of the population identifies as Hispanic or Latino. Population-based registry data is available to
support the collection of a large sample.
The objective of this study was to assess the possible influences of physician specialty and
physician experience with melanomas (defined as the number of melanomas diagnosed per year
in a 15-year period, with high volume defined as more than 15 melanomas, and low volume defined
as 15 or less) and physician specialty on melanoma survival among Hispanics and non-Hispanic
2

whites (NHWs) using population-based data. The findings will be used to design, test, and
implement effective interventions to improve earlier detection and reduce melanoma-related
disparities for Hispanics.

3
Chapter 2: Materials & Methods
Population and data sources
Cases of cutaneous melanoma in male and female adults 18 years or older of all races and
ethnicities, residents of LAC, diagnosed between 2005 and 2019, were identified from the LACSP.
We only included patients 18 or older because childhood and adolescent melanoma are uncommon
and have different sociodemographic characteristics. Cutaneous melanomas were identified using
the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program
classification for melanoma of the skin (code 25010). For each melanoma case, the primary
physician was identified. The primary physician is defined as the physician who was responsible
for the overall management of the patient during diagnosis and/or treatment for this cancer. All
study variables were obtained from routinely collected registry data.
Survival time
The primary outcome of this study was observed survival, reflecting survival from all causes of
death after a melanoma diagnosis. For the survival analysis, we included only the first tumor of
each Hispanic and non-Hispanic white patient.
Statistical analysis
Differences in patient characteristics were evaluated using chi-square tests. Survival was estimated
using the Kaplan-Meier (KM) method. Survival time was determined by calculating the number
of months elapsed between the patient's date of diagnosis and the date of last follow-up or death,
with a study cut-off date of December 2019. KM curves and one- and five-year survival estimates
were calculated in R version 4.1.0 (7) using the survival package version 3.2-11. (8) Statistical
significance was determined using 95% confidence intervals (CIs). Multivariate Cox proportional
4

hazard regression was used to control for confounding and estimate adjusted hazard ratios (HR)
with 95% CIs for the overall risk of death by ethnicity group (Hispanic and non-Hispanic white).
HRs were estimated using SAS software (version 9.4, SAS Institute Inc, Cary, NC).
Variables
Stage. The stage of disease was determined using the SEER Summary Stage classification. We
used Summary Stage to classify tumors as in situ/localized, regional, distant, and unstaged. We
included unstaged tumors with distant tumors for analysis based on similar survival outcomes.
Race/ethnicity. We included patients in our data listed under the Hispanic and non-Hispanic white
categories. Hispanic ethnicity was coded using a standardized identification algorithm. (7)
Other Patient Characteristics. Patients were characterized according to their age, sex, marital
status, insurance status, comorbidities, and socioeconomic status (SES). Marital status was
categorized as married (or in a domestic partnership), single (including widowed, separated, or
divorced), and unknown. Insurance was classified as insured (including private insurance and
Medicare), Medicaid, and not insured or unknown based on primary payer at the time of diagnosis.
Medicare coverage was combined with private insurance due to similar acceptance rates among
physicians overall (8) and dermatologists specifically. (9,10) Comorbidity was classified as low,
medium, high, and unknown based on the Charlson Comorbidity Index. SES was based on an
existing index and classified into quintiles. (11)
Physician volume. Attending physician volume was based on the number of melanoma tumors
seen by a physician under their role as a primary physician in the California Cancer Registry. The
number of tumors includes all melanomas diagnosed among adult (males and females, ages 18+)
residents of Los Angeles County between 2005-2019. All races and ethnicities were included in
this total. Low-volume captures attending physicians who diagnosed up to 15 melanomas during
5
this period, and high-volume refers to physicians who diagnosed over 15 melanomas (more than
one melanoma per year). This number was chosen based on what we believe is clinically relevant
and to ensure that each category represented a sufficient number of physicians. A total of 7,056
(34%) physicians were classified as low volume, and 13,688 (66%) were classified as high volume.
Physician specialty. To determine physician specialties, we conducted a manual search on the
California Department of Consumer Affairs website using each physician's license number. After
obtaining their names, we searched for their specialties on Google. When we could not find
information for a physician either due to a wrong license number that could not be found on the
website or a lack of information on Google, which was mostly due to the physician being deceased
or retired, that physician was classified as "no information." For data analysis, we combined certain
specialties into groups based on their likely expertise with melanoma. Specifically, general surgery
and surgical subspecialties, including plastic surgery, vascular surgery, orthopedics, urology,
neurological surgery, thoracic surgery, oral and maxillofacial surgery, and colorectal surgery, were
grouped together as ´´other surgical specialties.´´ Additionally, oncology, hematology-oncology,
and radiation oncology were combined into a category labeled as ´´oncology´´. Family medicine,
internal medicine, and all internal medicine subspecialties were also grouped together as ´´primary
care subspecialties.´´ All other specialties were labeled as "other." Dermatology and Surgical
oncology were listed alone due to the high volume of melanomas associated with each specialty.

6

Chapter 3: Results
Population characteristics.
Table 1 outlines the key demographic and clinical features of melanoma cases of Hispanic and
non-Hispanic white (NHW) patients included in the study. Additionally, we sought to investigate
whether there were any disparities in the outcome of melanoma cases based on the volume of
patients treated by the physicians involved. To this end, Table 2 presents a breakdown of the
characteristics of melanoma cases seen by low and high-volume physicians.
Patients
The study identified a total of 20,744 melanoma tumors diagnosed in adults in LAC between 2005
and 2019, with 2,077 cases of melanoma in Hispanic and 18,667 in NHW patients (Table 1).
Age, gender, comorbidities, and marital status
The mean age of melanoma diagnosis was 59.3 (SD 16.5) for Hispanics and 64.4 (SD 15.8) for
NHW (p < 0.0001). There were no differences in mean age for patients seen by high and low-
volume physicians. Hispanic melanomas were mostly found in females (55.4%), while NHW
melanomas were mostly present in males (60.6%) (p < 0.0001). Gender did not have an impact on
the chance of being seen by an experienced physician. The distribution of comorbidities and
marital status was similar between Hispanics and NHW. Being married or in a domestic
partnership was associated with a slightly higher probability of being seen by a high-volume
physician (67% versus 64.3% for single, separated, divorced, or widowed patients and 65.9% for
unknown marital status) (p = 0.0011). Patients with high comorbidities were less likely to be seen
by a high-volume physician (58.3%) compared to those with low (68.8%) and moderate (61.8)
comorbidities (p < 0.001) (Tables 1 and 2).
7
Socioeconomic status
Regarding SES, the majority of melanomas in the Hispanic population were concentrated in the
lowest income quintiles, with the lowest and second-lowest quintiles accounting for 48.5% of the
cases. In contrast, melanomas in NHW patients were predominantly concentrated in the highest
quintiles, with the two top quintiles accounting for 72.3% of the melanomas and the lowest and
second-lowest quintiles accounting for only 12.4% of melanomas (p < 0.0001). Among cases
categorized in the highest SES quintile, 71.5% of melanomas were seen by a high-volume
physician. The probability of seeing a high-volume physician decreased as the SES quintile
decreased. Melanoma cases categorized in the lowest SES quintile had a 53.2% probability of
being seen by a high-volume physician (p < 0.0001). Among Hispanics, those in the lowest SES
quintile are less likely to see a dermatologist (25.7%) compared to those in the highest SES quintile
(31.8%). However, this is not the case for NHW, as those in the highest SES quintile have a similar
chance of seeing a dermatologist compared to the lowest SES quintile (32.1% vs. 30.3%) (Tables
1 and 2). Finally, our results show that Hispanics within the lowest SES quintile were less likely
to be seen by a dermatologist (25.7%) than Hispanics in the highest SES (31.85). This difference
is not seen in the NHW population, where those in the lowest and highest SES quintile had a
similar likelihood of being seen by a dermatologist (30.3% and 32.1%, respectively).
Insurance status
In terms of insurance status, Hispanics were more likely to be uninsured or have unknown
insurance (12.8%) compared to NHW patients (9.5%). Medicaid coverage was also more common
among Hispanic patients (12.9%) than NHW patients (1.9%). Furthermore, 88.6% of NHW were
categorized as insured compared to 74.3% of Hispanics. (p < 0.0001). Patients with Medicaid had
8

a lower probability of being seen by a high-volume physician (39.4%) compared to those insured
(67.8%) and not insured/unknown (58%) (p < 0.0001) (Tables 1 and 2).
Stage of disease
Melanoma was detected at an early stage (in situ/localized) more frequently in both groups. Still,
the percentage of regional and distant/unstaged melanomas was higher for Hispanics (14.9%
regional, 10.5% distant/unstaged) compared to NHW (7.2% regional, 6.1% distant/unstaged)
(p<0.0001). Notably, more advanced melanomas were found to be more likely to have a low-
volume attending physician. Specifically, 69.5% of in situ/localized melanomas were diagnosed
by physicians in the high-volume group, whereas only 58.9% of regional melanomas and 28.4%
of distant/unstaged melanomas were in this group (< 0.0001) (Tables 1 and 2).
Physician volume and ethnicity
Hispanics were seen by a high-volume physician in 52.2% of the cases, compared to 67.5% among
NHW (p<0.0001) (Table 1).
Physician specialty and ethnicity
Regarding physician specialties, 29.5% of Hispanic melanomas and 32.1% of NHW melanomas
were treated by dermatologists. NHW patients were twice as likely to be seen by a surgical
oncologist compared to Hispanic patients (24.7% of NHW and 11.8% of Hispanics). 25% of
Hispanics were seen by general surgeons or surgeons with other subspecialties. Finally, Hispanics
were more likely to be diagnosed by primary care and internal medicine subspecialties than NHW
(11.3% of Hispanics, 8.4% of NHW) (Table 1). When comparing physicians within specialties,
we observe that 78% of the melanoma cases diagnosed by a dermatologist were diagnosed by high
volume, compared to 98.3% for surgical oncology, 47.1% for general surgery and other surgical
subspecialties and only 4.6% for primary care and internal medicine subspecialties (Table 2).
9
Physicians
A total of 3,696 unique physicians were listed as the primary physician for one or more of the
34,541 melanoma tumors diagnosed among 30,524 patients (all races and ethnicities) across the
15-year study period.
The most common specialty among primary physicians was dermatology (18.3% of physicians),
followed by internal medicine (15.6%) and family medicine (12.7%). The specialty with the
highest number of melanomas diagnosed was dermatology (31.7% of melanoma cases), followed
by surgical oncology (23.4%) and general surgery (12.1%) (Table 3).
Although surgical oncologists represented only 1.5% of the physicians, they accounted for 23.4%
of melanoma cases (Table 3). They were more likely to be in the high-volume group, with 52.7%
of Surgical Oncologists classified as high-volume. However, when we look at the number of
melanoma cases seen by surgical oncologists, we can see that 98.3% of the cases were seen by a
high-volume surgical oncologist (Table 4). If we look at dermatologists, they accounted for 31.7%
of melanoma cases, with 77.9% of these cases seen by high-volume dermatologists (Tables 3 and
4). 99.8% of family medicine physicians, 99.5% of internists, and 90.6% of general surgeons were
classified as low volume (Table 4).
Survival Trends
Hispanics had a lower 5-year survival probability than NHW (Figure 1). Overall, melanomas
treated by low-volume physicians had worse survival outcomes for both groups than those treated
by high-volume physicians. When treated by low-volume physicians, Hispanic survival was worse
than NHW (Figure 2).
10

Hispanics had a slightly higher survival probability when diagnosed at an early stage of disease
(in situ/localized) but significantly worse survival when diagnosed at a late stage (regional and
distant/unstaged), compared to NHW (p<0.0001) (Figure 3).
When comparing stage and physician volume, the disparity due to physician volume was greater
among non-Hispanic whites (NHW) than Hispanics, particularly for melanomas diagnosed at a
late stage (p<0.0001). NHW patients with distant melanoma seen by a high-volume physician had
better survival outcomes than those with regional melanoma seen by a low-volume physician.
However, this was not the case in the Hispanic population, particularly for regional melanomas,
where there was almost no difference in survival between those diagnosed by high and low-volume
physicians (Figure 4).
If we examine the disparities in survival rates among different medical specialties for Hispanic
and non-Hispanic white patients, we find that the outcomes vary significantly. Specifically,
Hispanic patients who received care from dermatologists experienced the highest survival rates,
with those treated by surgical oncologists and other surgical specialties also showing positive
outcomes. Conversely, those seen by oncologists and primary care/internal medicine
subspecialties exhibited the lowest survival rates. For NHW patients, surgical oncology emerged
as the specialty associated with the best survival rates, followed by dermatology. Outcomes for
those seen by oncologists and primary care physicians were the worst, echoing the results observed
for Hispanic patients (Figure 5).

11
Multivariate analysis
Hispanics had worse melanoma survival than NHW (Hispanic crude HR 1.14, 95% CI 1.05-1.25).
This disparity became even more pronounced when we adjusted for age only (Hispanic HR 1.55,
95% CI 1.42-1.69). After adjusting for race/ethnicity, attending physician experience, stage, age,
sex, year of diagnosis, SES quintile, insurance status, and comorbidities, we can see that Hispanics
and NHW have similar survival. This is primarily due to the relationship between SES and survival
probability. In the fully adjusted model, patients treated by less experienced physicians had worse
survival outcomes than those treated by experienced physicians (HR: 1.31, 95% CI 1.24-1.38).
Stage, SES, and comorbidities were statistically significant prognostic factors of cutaneous
melanoma survival. Regarding insurance status, patients with Medicaid had the worst survival
(HR: 2.61, 95% CI 2.27-3), followed by those uninsured or with unknown insurance status (Table
5).
When comparing Hispanics to NHW, we see that being seen by a low-volume physician had a
greater effect on survival for NHW (HR: 1.33, 95% CI: 1.26-1.41) than for Hispanics (HR: 1.13,
95% CI: 0.94-1.35). This is mostly due to Hispanics being diagnosed at later stages (Table 5).
Males had worse survival than females in both groups (Hispanic HR: 1.41, 95% CI 1.18-1.67;
NHW HR: 1,3 95% CI 1.22-1.38). Having Medicaid insurance, being uninsured, or having an
unknown insurance status was associated with a higher risk of death in both groups. Hazard ratios
were larger among NHW with Medicaid versus Hispanics (NHW HR: 2.78, 95% CI 2.34-3.3;
Hispanic HR: 1.99, 95% CI 1.56-2.54) (Table 5).
In the fully adjusted model, the Hispanic HR for regional melanoma was 3.49 (95% CI 3.5-4.85)
and 6.87 (95% CI 5.5-9.57) for distant or unstaged melanomas, larger than the NHW HR for
regional melanoma which was 2.76 (95% CI 2.54-2.99) and distant or unstaged melanoma which
was 5.54 (95% CI 5.11-6.02) (Table 5).
12

When comparing the survival outcome of Hispanics with a dermatologist as their primary
physician to those with other specialty physicians, we can see that the survival probability is
significantly lower for those who did not see a dermatologist (HR 1.97, 95% CI: 1.57 - 2.46). This
is not the case for NHW, where there is no statistically significant survival difference between
those seen by dermatologists or other specialists (Table 5).
When considering volume and specialty, we observe no statistically significant differences in
survival among Hispanics seen by high-volume dermatologists, high-volume physicians with other
specialties, and low-volume dermatologists. However, there is a significant difference in survival
for Hispanics who were seen by low-volume physicians with other specialties (HR 2.28, 95% CI:
1.7-2.92). Among NHW, patients seen by low-volume physicians with other specialties are also at
a higher risk (HR 1.57, 95% CI: 1.45-1.69), although not to the same degree as Hispanics (Table
5).
In our crude data, patients seen by oncologists had the worst survival (HR 2.18, 95% CI: 1.94 -
2.44), followed by primary care (and associated subspecialties) physicians (HR 1.75, 95% CI: 1.54
- 1.94). Being seen by a surgical oncologist was a good predictor of survival (HR 0.72, 95% CI:
0.67 - 0.79) (Table 6).

13
Chapter 4: Discussion

Hispanics represent an underserved and understudied population when it comes to melanoma
diagnosis and outcomes. They are diagnosed with melanoma at later stages than their NHW
counterparts, leading to an increased likelihood of metastasis and worse survival. (2,6,12) The
latter implies a lack of access to appropriate screening, adherence to screening recommendations,
understanding of appropriate screening approaches in the primary care setting (among physicians),
or a combination of these factors. (13)
Our knowledge about physician-related factors impacting melanoma survival among Hispanics is
limited. Studies have shown that the prognosis of melanoma patients is generally minimally
influenced by factors related to the treating physician. (14) However, our study shows that
melanoma survival is significantly lower for patients diagnosed by a non-dermatology physician
with low experience. Further studies are needed to understand the role that the physician plays in
the timely detection of disease among Hispanics, including physician knowledge and awareness
of melanoma risk factors among Hispanics, physician screening and referral practices for
melanoma, physician experience with melanoma, and physician perceived barriers to timely
detection of melanoma.
Melanomas are detected and diagnosed by multiple types of physicians, including primary care,
dermatologists, medical oncologists, surgical oncologists, and others. Physician type has been
linked to melanoma treatment delay (15,16), but virtually no data exist regarding the influence of
physician type and melanoma survival in Hispanic populations. Based on our findings, the survival
rate of Hispanics diagnosed with melanoma is almost half when they are diagnosed by a non-
dermatologist physician, as compared to those who are diagnosed by a dermatologist.

14

In this study, we utilize population-based records of the California Cancer Registry to examine
survival among 2,077 Hispanic patients and 18,677 NHW diagnosed with melanoma. Our primary
goal was to assess whether melanoma survival among Hispanics and NHW was related to the
primary physicians' level of experience and specialty.
Physician experience and specialty
Our findings indicate that Hispanics are more likely than NHW to be diagnosed by a physician
who has seen fewer than one melanoma per year. However, the impact of being diagnosed by a
low-volume physician is greater on survival for NHW than for Hispanics. While the difference in
survival between high and low-volume physicians for Hispanics disappears when adjusting for
stage, this is not the case for NHW patients. This finding suggests that when Hispanics are
diagnosed by a low-volume physician, they may already have a more advanced stage of melanoma,
and their poor outcomes are primarily a result of the stage at diagnosis. On the other hand, poor
outcomes seen among NHW patients may result from not receiving appropriate care from low-
volume physicians.
Furthermore, our analysis shows that patients with melanoma diagnosed by a general practitioner
or physician with a specialty lacking rigorous melanoma-specific training had significantly worse
survival outcomes than those seen by dermatologists, general surgeons, or surgical oncologists.
These results suggest the importance of educating non-dermatology physicians in the early
detection and management of melanoma. Several studies have evaluated the role of general
practitioners in melanoma diagnosis and have concluded that additional training in skin cancer,
melanoma, and biopsy could make a significant difference in melanoma outcomes, particularly for
patients who do not have access to specialist care. (17,18)

15
When examining the survival patterns of patients treated by different medical specialties, the
results are largely consistent with our expectations. Patients seen by oncologists generally
exhibited the worst survival outcomes, likely due to their advanced stage of disease requiring
radiation therapy or chemotherapy. Similarly, patients diagnosed by primary care physicians
experienced poor outcomes, potentially due to a lack of specialized expertise in melanoma
diagnosis and treatment, leading to delayed diagnosis and treatment initiation. Conversely, patients
seen by general surgeons had relatively positive outcomes, likely because early-stage melanomas
are easily removed through surgery, leading to better chances of survival.
Socioeconomic and insurance status
Our analysis reveals that controlling for SES eliminated the statistically significant difference in
survival between Hispanics and NHW. In addition, SES has a stronger impact on dermatology
access among Hispanics, but this is not the case for NHW. Hispanics are also more likely to be
diagnosed with melanoma when they belong to low SES quintiles. On the contrary, most
melanomas among NHW are concentrated in the highest SES quintiles. Finally, a direct
relationship exists between the SES quintile and the likelihood of being seen by a high-volume
physician. All these findings indicate that the main reason for the poorer survival outcomes among
Hispanics is their lower SES status, which leads to later diagnosis. This highlights the need for
better access to high-quality healthcare services for individuals from lower SES backgrounds.
Our data shows that a significantly higher proportion of Hispanics were uninsured or had Medicaid
than NHW. Moreover, being uninsured or having Medicaid negatively impacted the survival rates
of Hispanics and NHW. This is consistent with previous research highlighting uninsured and
Medicaid patients' challenges in accessing timely and appropriate care, including dermatology
appointments, and experiencing longer wait times (9,13,14,19). Efforts should be made to address
16

the healthcare disparities faced by underserved populations, particularly those with lower income
or those who are uninsured, to ensure that all patients have access to the same level of care and
improve survival outcomes.
Stage
Our findings showed that Hispanics are more likely to be diagnosed with advanced stages of
melanoma than NHW and that advanced cases are more likely to be diagnosed by physicians with
lower experience levels. This is concerning, as the optimal management of advanced melanomas
typically requires specialized expertise. It is noteworthy that this is the opposite of what one would
expect, as in situ melanomas can often be managed by physicians who are not specialists in
dermatology, whereas advanced cases require a more experienced approach. One possible
explanation for this phenomenon, as described in the literature, is that physicians without
experience in skin cancer may be more adept at recognizing advanced melanomas, while early-
stage melanomas require a higher level of clinical suspicion. Studies have shown that melanomas
diagnosed by dermatologists tend to be thinner than those identified by patients or referred by
primary care physicians. (20–22)

17
Chapter 5: Strengths and Limitations
Strengths
This study has several strengths that make it valuable for advancing our understanding of
melanoma diagnosis and treatment. First, the study is population-based, which means that it
includes a diverse group of individuals with melanoma. This reduces the potential for reporting
bias, as all individuals with melanoma have an equal chance of being included in the study.
In addition, because we are using data from LAC, the study includes a large number of Hispanic
patients. This allows for a more comprehensive understanding of melanoma diagnosis and
treatment patterns among this population.
Furthermore, the study includes a large variety of treating physicians, allowing for a more
complete understanding of the factors that influence melanoma diagnosis and treatment, including
physician experience and training.
Another strength of your study is the long follow-up period, which allows for identifying long-
term survival trends among individuals with melanoma.
Finally, the study design also reduces the potential for recall bias, as it collects data directly from
medical records rather than relying on patient self-reporting. This enhances the accuracy and
reliability of the study findings and allows for more confident conclusions to be drawn regarding
melanoma diagnosis.
Limitations
Physicians' specialties were manually searched, which could have led to miscategorization. First,
there might have been physicians with the same name as other physicians. To avoid this, we used
the school where they got their medical degree since this information was available on the license
lookup website and in most of the ‘‘physicians' description’’ information in Google. Also,
18

physicians may be double-boarded, with only one specialty clearly indicated, which may
contribute to the large variety of physician specialties related to melanoma diagnosis. Finally, some
physicians with a specialty could be working as general practitioners due to the lack of personnel
in the area. Even though there are some specialties we would not have expected to be related to
melanoma diagnosis, most physicians are dermatologists and surgical oncologists, suggesting that
our data correlate with reality.
Secondly, we could not find information for 257 (7%) physicians who accounted for 1,297 (6.3%)
melanoma cases. Some were associated with invalid license numbers in the dataset, and others had
valid ones, but we could not find information about their specialty. The latter was mostly because
those physicians were retired or deceased. This is a major limitation because even though it is only
6.3% of the cases, it represents the fourth category regarding the number of melanoma cases.
69.8% of the physicians in this group were categorized as highly experienced (the third group with
the highest experience percent, after surgical oncology and dermatology), which leads us to think
that most of the physicians with no information are dermatologists, surgical oncologists, and
general surgeons. If data were available for these physicians, it would probably increase the
number of physicians as well as the overall level of expertise in the aforementioned specialties.
Additionally, there might be differences in how reporting facilities determine who the primary
physician is for each patient. In order for this to incorporate bias into our analysis, it would need
to be consistent throughout the data; for example, having one big hospital or facility that is
reporting the wrong physician. If this happened, we could end up listing a physician who is not the
primary physician and therefore has nothing to do with the timely diagnosis of melanoma.
Furthermore, the 15 melanoma cut point to classify physicians into the high and low experience
was set by us based on what we thought was clinically significant and also making sure that the
low volume group had a fair number of melanomas that would allow us to compare it to the high
19
volume group. However, no sensitivity analyses were performed. Since almost no data have been
published regarding this subject, we could not base our categorization on previous study designs
or results. We also assume that all physicians within the same specialty group have similar
capacities. Further studies are needed to compare our results and define a validated cut point.
Additionally, there is a chance that some physicians have seen multiple other melanomas outside
of California that we are not capturing. Finally, not every physician in the low-volume group had
the same years of active practice. For example, there might be physicians who have been active
for a few years and have diagnosed a considerable number of melanomas but are still grouped into
low experience. In our future work, we will stratify low-experience physicians into groups
depending on the number of melanomas diagnosed yearly. We may notice that differences are even
bigger than what we see now.
Finally, the SES variable employed is based on the SES of the patient’s residential census block
group rather than the individual patient's SES. Therefore, in certain instances, there could be
discrepancies between an individual's SES and the SES of their block group.

20

Chapter 6: Conclusion and next steps
This study examined significant disparities in the diagnosis and treatment of melanoma between
Hispanic and non-Hispanic white patients, showing that SES and insurance status were significant
factors in melanoma outcomes. Physicians’ experience doesn’t seem to be the cause of the poorer
survival among Hispanics specifically; however, this variable appears to impact Hispanics and
NHW.
This study highlights the need for targeted interventions to improve access to high-quality care for
underserved populations, particularly strategies to increase awareness of the importance of early
detection and prevention and interventions to improve access to high-volume physicians,
particularly dermatologists, for the Hispanic population. It additionally highlights the importance
of educating primary care providers on melanoma detection. Studies are needed to categorize
further the discrepancy in melanoma survival differences between Hispanic and NHW populations
to allow effective and efficient interventions.
We have planned several next steps to improve our understanding of melanoma diagnosis. The
first step involves a study where we will send a comprehensive survey to a diverse group of
physicians with varying levels of experience in melanoma diagnosis. The survey will inquire about
their personal experience with melanoma (and particularly Hispanic melanoma), including their
diagnostic practices, referral patterns, and any challenges they may have encountered. By
collecting this data, we will better understand the factors that may influence melanoma diagnosis,
such as physician experience, training, and access to resources.
In addition, we will analyze the survey data to identify any common patterns or trends among
physicians who diagnose melanoma. This information will then be used to identify gaps in current
21
diagnostic practices and guide the development of targeted interventions to improve melanoma
diagnosis rates.
Furthermore, we will conduct surveys with patients diagnosed with melanoma to better understand
their experiences with diagnosis, including any factors that may have delayed their diagnosis or
impacted their access to care. By linking patient-reported data with physician information, we will
develop a more comprehensive picture of the factors influencing melanoma diagnosis and
treatment. This will enable us to develop more effective strategies to improve melanoma diagnosis
and, ultimately, improve patient outcomes.

22

References
1. Cortez JL, Vasquez J, Wei ML. The impact of demographics, socioeconomics, and health
care access on melanoma outcomes. J Am Acad Dermatol [Internet]. 2021;84:1677–1683. doi:
10.1016/j.jaad.2020.07.125. Cited: in: : PMID: 32783908.

2. Pollitt RA, Clarke CA, Swetter SM, Peng DH, Zadnick J, Cockburn M. The expanding
melanoma burden in California hispanics [Internet]. Cancer. 2011. p. 152–161. Available from:
http://dx.doi.org/10.1002/cncr.25355.

3. Cockburn MG, Zadnick J, Deapen D. Developing epidemic of melanoma in the Hispanic
population of California. Cancer [Internet]. 2006;106:1162–1168. doi: 10.1002/cncr.21654. Cited:
in: : PMID: 16429450.

4. Fejerman L, Ramirez AG, Nápoles AM, Gomez SL, Stern MC. Cancer Epidemiology in
Hispanic Populations: What Have We Learned and Where Do We Need to Make Progress?
[Internet]. Cancer Epidemiology, Biomarkers & Prevention. 2022. p. 932–941. Available from:
http://dx.doi.org/10.1158/1055-9965.epi-21-1303.

5. Srivastava R, Wassef C, Rao BK. The Dayanara effect: increasing skin cancer awareness
in the Hispanic community. Cutis [Internet]. 2019;103:257–258. Cited: in: : PMID: 31233576.

6. Sanchez DP, Maymone MBC, McLean EO, Kennedy KF, Sahni D, Secemsky EA, Vashi
NA. Racial and ethnic disparities in melanoma awareness: A cross-sectional survey. J Am Acad
Dermatol [Internet]. 2020;83:1098–1103. doi: 10.1016/j.jaad.2020.04.137. Cited: in: : PMID:
32380221.

7. Howe HL. NAACCR Guideline for Enhancing Hispanic-Latino Identification: Revised
NAACCR Hispanic/Latino Identification Algorithm [NHIA v2]. North American Association of
Central Cancer Registries [Internet]. 2011;

8. Hing E, Decker S, Jamoom E. Acceptance of new patients with public and private
insurance by office-based physicians: United States, 2013. NCHS Data Brief [Internet]. 2015;1–
8. Cited: in: : PMID: 25932895.

9. Creadore A, Desai S, Li SJ, Lee KJ, Bui A-TN, Villa-Ruiz C, Lo K, Zhou G, Joyce C,
Resneck JS Jr, et al. Insurance Acceptance, Appointment Wait Time, and Dermatologist Access
Across Practice Types in the US. JAMA Dermatol [Internet]. 2021;157:181–188. doi:
10.1001/jamadermatol.2020.5173. Cited: in: : PMID: 33439219.

10. Resneck J Jr, Pletcher MJ, Lozano N. Medicare, Medicaid, and access to dermatologists:
the effect of patient insurance on appointment access and wait times. J Am Acad Dermatol
[Internet]. 2004;50:85–92. doi: 10.1016/s0190-9622(03)02463-0. Cited: in: : PMID: 14699371.

11. Yang J, Schupp C, Harrati A, Clarke C, Keegan THM, Gomez SL. Developing an Area-
Based Socioeconomic Measure From American Community Survey Data. Fremont, CA: Cancer
23
Prevention Institute of California; 2014. 2018.

12. Zell JA, Cinar P, Mobasher M, Ziogas A, Meyskens FL Jr, Anton-Culver H. Survival for
patients with invasive cutaneous melanoma among ethnic groups: the effects of socioeconomic
status and treatment. J Clin Oncol [Internet]. 2008;26:66–75. doi: 10.1200/JCO.2007.12.3604.
Cited: in: : PMID: 18165642.

13. Escobedo LA, Crew A, Eginli A, Peng D, Cousineau MR, Cockburn M. The role of
spatially-derived access-to-care characteristics in melanoma prevention and control in Los
Angeles county. Health Place [Internet]. 2017;45:160–172. doi:
10.1016/j.healthplace.2017.01.004. Cited: in: : PMID: 28391127.

14. Richard MA, Grob JJ, Avril MF, Delaunay M, Gouvernet J, Wolkenstein P, Souteyrand P,
Dreno B, Bonerandi JJ, Dalac S, et al. Delays in diagnosis and melanoma prognosis (II): the role
of doctors. Int J Cancer [Internet]. 2000;89:280–285. doi: 10.1002/1097-
0215(20000520)89:3<280::aid-ijc11>3.0.co;2-2. Cited: in: : PMID: 10861505.

15. Adamson AS, Zhou L, Baggett CD, Thomas NE, Meyer A-M. Association of Delays in
Surgery for Melanoma With Insurance Type. JAMA Dermatol [Internet]. 2017;153:1106–1113.
doi: 10.1001/jamadermatol.2017.3338. Cited: in: : PMID: 28979974.

16. Lott JP. Delay and Disparity in Time to Surgical Treatment for Melanoma [Internet].
JAMA Dermatol. 2017. p. 1094–1095. Available from:
http://dx.doi.org/10.1001/jamadermatol.2017.3339.

17. Buckley D, McMonagle C. Melanoma in primary care. The role of the general practitioner.
Ir J Med Sci [Internet]. 2014;183:363–368. doi: 10.1007/s11845-013-1021-z. Cited: in: : PMID:
24091614.

18. Harkemanne E, Baeck M, Tromme I. Training general practitioners in melanoma
diagnosis: a scoping review of the literature. BMJ Open [Internet]. 2021;11:e043926. doi:
10.1136/bmjopen-2020-043926. Cited: in: : PMID: 33757946.

19. Resneck JS Jr, Isenstein A, Kimball AB. Few Medicaid and uninsured patients are
accessing dermatologists. J Am Acad Dermatol [Internet]. 2006;55:1084–1088. doi:
10.1016/j.jaad.2006.07.012. Cited: in: : PMID: 17097404.

20. Melanoma screening by means of complete skin examinations for all patients in a
dermatology practice reduces the thickness of primary melanomas at diagnosis [Internet]. Journal
of the American Academy of Dermatology. 2014. p. AB129. Available from:
http://dx.doi.org/10.1016/j.jaad.2014.01.537.

21. McGuire ST, Secrest AM, Andrulonis R, Ferris LK. Surveillance of patients for early
detection of melanoma: patterns in dermatologist vs patient discovery. Arch Dermatol [Internet].
2011;147:673–678. doi: 10.1001/archdermatol.2011.135. Cited: in: : PMID: 21690529.

24

22. Kantor J, Kantor DE. Routine dermatologist-performed full-body skin examination and
early melanoma detection. Arch Dermatol [Internet]. 2009;145:873–876. doi:
10.1001/archdermatol.2009.137. Cited: in: : PMID: 19687416.

25
Appendices
Table 1: Sociodemographic and clinical characteristics of patients with melanoma (ages 18+) by
ethnicity, LACSP, 2005 - 2019.

26

Table 2: Sociodemographic and clinical characteristics of patients with melanoma (ages 18+) by
physician volume, LACSP, 2005 - 2019.

27

Table 3: Number of melanoma cases diagnosed (all races, ages 18+) for each attending physician specialty by ethnicity and total number
of physicians for each specialty with column percentages, LACSP, 2005 - 2019.
28

Table 4: Number of melanoma cases diagnosed (all races, ages 18+) for each attending physician specialty by ethnicity and total
number of physicians for each specialty with row percentages, LACSP, 2005 - 2019.
29
Table 5: Hazard ratio for risk of all-cause death by ethnicity group, in patients with melanoma,
LACSP, 2005-2019.

30

Table 6: Hazard ratio for risk of all-cause death, Hispanics and NHWs patients with melanoma, LACSP, 2005-2019.
31
Figure 1: Melanoma survival by race/ethnicity for all stages combined (ages 18+), Los Angeles
County, 2005-2019

32

Figure 2: Melanoma survival by attending physician volume for Hispanics and Non-Hispanic
Whites (ages 18+), Los Angeles County, 2005-2019

33
Figure 3: Melanoma survival by stage at diagnosis for Hispanics and Non-Hispanic Whites (ages
18+), Los Angeles County, 2005-2019

34

Figure 4: Melanoma survival by stage and attending physician volume for Hispanics and Non-
Hispanic Whites (ages 18+), Los Angeles County, 2005-2019.

35
Figure 5: Melanoma survival by primary physician´s specialty for Hispanics and Non-Hispanic
Whites (ages 18+), Los Angeles County, 2005-2019

Abstract (if available)

Abstract Background: Melanoma is the most lethal form of skin cancer. Hispanics are less likely to be diagnosed with melanoma, but they are diagnosed at later stages and have lower survival rates than their non-Hispanic white (NHW) counterparts. Reasons for the later stage at diagnosis among Hispanics are unknown and may be related to patient, clinician, healthcare systems issues, or a combination of several factors. This study had two objectives:
Objective 1: Describe the sociodemographic and clinical characteristics, such as socioeconomic status, marital status, insurance status, age, and stage at diagnosis, among others of Hispanic and NHW patients with melanoma as well as the characteristics of patients seen by physicians with high or low experience and different specialties.
Objective 2: Identify if the physician's specialty or level of experience (in terms of the number of melanomas diagnosed) is related to melanoma survival among Hispanics and NHW.
Methods: We used data from the Los Angeles County Cancer Surveillance Program (LACSP) and compared melanoma survival among Hispanics and NHW, considering the specialty and experience level (defined as the number of melanomas diagnosed per year in a 15-year period, with high volume defined as more than 15 melanomas, and low volume defined as 15 or less) of the treating.
Results: The study identified a total of 20,744 melanomas diagnosed in adults in Los Angeles County (LAC) between 2005 and 2019, with 2,077 cases of melanoma in Hispanic and 18,667 in NHW patients. NHW melanomas were more likely to be treated by an experienced physician than Hispanics (67.5% vs. 52.2). The 5-year survival for melanomas that had an inexperienced physician as the attending doctor was significantly lower than for those melanomas treated by experienced physicians (HR: 1.31, 95%CI 1.24-1.38). The worse outcome among Hispanics is
mostly related to their lowest SES and not to physician experience or specialty. Hispanic melanomas are concentrated in the lower socioeconomic status (SES) quintiles, while NHW melanomas are in the highest SES quintiles. Patients from lower SES quintiles were less likely to be diagnosed by an experienced physician. Low-experienced physicians diagnose melanomas in a more advanced stage than highly experienced physicians.
Conclusion: Physicians’ experience and specialty are not associated with a worse outcome for Hispanics with melanoma. However, it appears that being treated by a physician with less experience or a specialty not commonly related to melanoma influences survival in the general population. When studying survival trends in Hispanic populations with melanoma in LAC, the specialty and experience of the physician does not account for survival variability; instead, lower SES is found to correlate with higher mortality.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Disparities in gallbladder, intra-hepatic bile duct, and other biliary cancers among multi-ethnic populations: a California Cancer Registry study

PDF

Psychosocial and cultural factors in the primary prevention of melanoma targeted to multiethnic children

PDF

Magnetic resonance imaging (MRI) staging for breast cancer in a diverse population

PDF

Predictive factors of breast cancer survival: a population-based study

PDF

Medical comorbidities and multiple sclerosis in the Hispanic population: linking healthcare disparities

PDF

Racial/ethnic differences in patient experiences with health care in association with the Gleason’s score level at prostate cancer diagnosis: findings from the SEER-CAHPS data

PDF

Clinical outcomes of allogeneic hematopoietic stem cell transplant in acute lymphoblastic leukemia patients: a quality improvement project and systematic review meta-analysis

PDF

Integrated management of atrial fibrillation in women in an underserved, safety-net health care system: a multicenter, single health system randomized control efficacy trial protocol

PDF

Model development of breast cancer detection and staging via rare event enumeration from a liquid biopsy: a retrospective descriptive clinical research study

PDF

The pharmacokinetics and pharmacodynamics of vincristine in the adolescent and young adult population compared to younger patients

PDF

Melanoma in children, adolescents, and young adults

PDF

Multilevel sociodemographic correlates of the health and healthcare utilization of childhood cancer survivors

PDF

Construction of a surgical survival prediction model of stage IV NSCLC patients-based on seer database

PDF

Hispanic ethnicity in multiple sclerosis

PDF

Survival trends and related outcomes of survivors of childhood and young adult cancer

PDF

Induction therapy in relapse adult acute lymphoblastic leukemia

PDF

A cross-sectional study of the association of PTH on bone quality across levels of propionic acid among adult patients with uremia

PDF

Disparities in exposure to traffic-related pollution sources by self-identified and ancestral Hispanic descent in participants of the USC Children’s Health Study

PDF

Comparison of secondary and primary thyroid cancer in adolescents and young adults (AYA)

PDF

The Filipino family initiative: preliminary effects of an evidence-based parenting intervention offered in churches on parent and child outcomes

Asset Metadata

Creator Agüero Ureta, Rosario (author)

Core Title The role of physicians in the late-stage diagnosis of melanoma among: a population-based study in Los Angeles County

School Keck School of Medicine

Degree Master of Science

Degree Program Clinical and Biomedical Investigations

Degree Conferral Date 2023-05

Publication Date 04/17/2024

Defense Date 04/17/2023

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Hispanics,Melanoma,OAI-PMH Harvest,physician,population based

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Cockburn, Myles (committee chair)

Creator Email aguero@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113017571

Unique identifier UC113017571

Identifier etd-AgeroUreta-11649.pdf (filename)

Legacy Identifier etd-AgeroUreta-11649

Document Type Thesis

Format theses (aat)

Rights Agüero Ureta, Rosario

Internet Media Type application/pdf

Type texts

Source 20230418-usctheses-batch-1024 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu