Table 2.

Breast cancer-specific and overall mortality among patients undergoing nipple-sparing and non-nipple-sparing unilateral mastectomy in California, 1988–2013

	Number of deaths	Total person-years	Age- and stage-adjusteda			Fully adjustedb,c
			HR	95 % CI	p value	HR	95 % CI	p value
Breast cancer-specific mortality
1988–2013 diagnoses
Non-nipple-sparing	25,948	1451,617	1.0a			1.0b
Nipple-sparing	37	4553	0.67	0.49–0.93	0.02	0.71	0.51–0.98	0.04
1996–2013 diagnoses
Non-nipple-sparing	13,469	767,098	1.0a			1.0c
Nipple-sparing	17	2518	0.61	0.38–0.98	0.04	0.86	0.52–1.42	0.55
Overall mortality
1988–2013 diagnoses
Non-nipple-sparing	62,784	1451,617	1.0a			1.0b
Nipple-sparing	118	4553	0.91	0.76–1.09	0.31	0.92	0.76–1.12	0.41
1996–2013 diagnoses
Non-nipple-sparing	29,707	767,098	1.0a			1.0c
Nipple-sparing	32	2518	0.59	0.42–0.83	0.003	0.74	0.50–1.08	0.12

n = 157,592 for 1988–2013; n = 106,181 for 1996–2013

^a Cox regression with time from diagnosis (days) as the time-scale; stratified by American Joint Committee on Cancer (AJCC) stage (0, I, II, III); and adjusted for age at diagnosis

^b Cox regression with time from diagnosis (days) as the time-scale; stratified by AJCC stage (0, I, II, III) and histology (ductal, lobular or with lobular component, other); adjusted for age, race, tumor size, lymph node involvement, adjuvant treatment, neighborhood socioeconomic status (SES), marital status, patient SES distribution of reporting hospital, National Cancer Institute-designated cancer center, and year of diagnosis; and adjusted for clustering by hospital

^c Same as the model in footnote b, but additionally adjusted for grade, estrogen and progesterone receptor status, and insurance status, which were not available before 1996