Skip to main content

Table 3 Multivariable associations between vulnerability cluster variables and receipt and time to care across four key care junctures in mBC (N = 418)

From: Assessing equity of care across metastatic breast cancer treatment junctures: a multi-site retrospective cohort study

Care Juncture

Outcome

Analysis used

N

Significant Factors at p < 0.05

Factor

β/OR/

RR/IRR

95% CI

lower

95% CI

upper

p-value

Diagnosis and Evaluation

E&M visita

N/Aa

418

-

-

-

-

-

Days to E&M visit

GLM-

Gamma

404

(missing n = 14)

Marital status Unknown

(vs. Married)

1.589

1.108

2.278

0.0119

Treatment Plan Design

Treatment Plan

Logistic

Y = 373, N = 31

Number of metastatic

sites

1.775

1.118

2.817

0.0150

Days to Treatment Plan date

GLM-

Gamma

373

Residential segregation index for

black/whiteb

1.024

1.006

1.044

0.0102

Distress Screening

Logistic

Y = 353, N = 51

Age at diagnosis

0.961

0.927

0.997

0.0338

Marital status Unknown

(vs. Married)

0.323

0.113

0.926

0.0354

Days to Distress Screening date

GLM-

normal

353

Age at diagnosis

0.0000056

0.000000025

0.001

< 0.001

Treatment Implementation

Anticancer Therapy Start

Logistic

Y = 372, N = 32

Number of metastatic

sites

1.686

1.088

2.613

0.0195

Days to Anticancer Therapy

date

GLM-

Gamma

368c

None

-

-

-

-

Radiation Start

Logistic

Y = 76,

N = 328

None

-

-

-

-

Days to Radiation

dated

N/Ad

76

-

-

-

-

-

Surgery Start

Logistic

Y = 43, N = 361

Age at diagnosis

0.957

0.922

0.994

0.0238

Days to Surgery

dated

N/Ad

43

-

-

-

-

-

ACP visit

Logistic

Y = 136, N = 268

None

-

-

-

-

Days to ACP date

GLM-

Gamma

136

Access to other primary

care providers

0.9993

0.9988

0.9998

0.0095

End of Life Care

Palliative Care

Logistic

Y = 14,

N = 390

Age at diagnosis

0.906

0.840

0.977

0.0101

Days to Palliative

Care referrald

N/Ad

14

-

-

-

-

-

Hospice Enrollment

Logistic

Y = 128,

N = 89

None

-

-

-

-

Days to Hospice Enrollment date

GLM-

Poisson

128

Race/ethnicity–Black

(vs. White)

0.868

0.845

0.891

< 0.0001

Race/ethnicity–Hispanic (vs.

White)

0.652

0.635

0.671

< 0.0001

Age at diagnosis

0.971

0.970

0.972

< 0.0001

Marital – Divorced

(vs. married)

0.723

0.704

0.743

< 0.0001

Marital – Single

(vs. married)

0.741

0.725

0.758

< 0.0001

Marital – Widowed

(vs. married)

0.935

0.919

0.952

< 0.0001

Marital – Unknown

(vs. married)

0.606

0.582

0.630

< 0.0001

No. of metastatic sites

0.991

0.986

0.996

0.0003

Access to primary care

1.0001

1.0001

1.0001

< 0.0001

High school completion

1.416

1.129

1.775

0.0026

Access to other primary care

providers

0.9999

0.9999

0.9999

< 0.0001

Residential segregation

index for black/white

1.0086

1.0077

1.0094

< 0.0001

Residential segregation index for white/non-white

0.982

0.981

0.983

< 0.0001

Chemotherapy within

14 days of deathd

N/Ad

18

-

-

-

-

-

Hospice Length of Stay (LOS)

GLM-

Poisson

82e

Race/ethnicity – Black

(vs. White)

0.754

0.645

0.882

0.0004

Race/ethnicity – Hispanic

(vs. White)

0.747

0.632

0.882

0.0006

Age at diagnosis

1.010

1.006

1.015

< 0.0001

Marital – Divorced

(vs. married)

2.094

1.855

2.364

< 0.0001

Marital – Widowed

(vs. married)

0.833

0.734

0.946

0.0048

No. of metastatic sites

0.871

0.842

0.901

< 0.0001

Access to primary care

1.0001

1.000

1.0001

0.015

High school completion

6.346

1.794

22.457

0.0042

Access to other primary care

providers

0.9999

0.9998

1.0000

0.0014

Residential segregation

index for black/white

1.027

1.021

1.033

< 0.0001

  1. ACP = Advance Care Planning; E&M = Evaluation and Management; GLM = Generalized Linear Model; IRR = incidence rate ratio;
  2. OR = odds ration; RR = rate ratio
  3. a 100% of the 418 patients had a record of the Diagnosis and Evaluation juncture E&M visit
  4. b The residential segregation index ranges from 0 (complete integration) to 100 (complete segregation)
  5. c Four patients whose days between the mBC diagnosis and the date of treatment start were greater than 365 days were removed from the analysis
  6. d Sample size is too small to conduct multivariable analysis
  7. e Among patients who have a record of both ‘hospice enrollment’ and ‘death’