COVID-19 Vaccination Initiation and Completion in a Multicenter Cohort Study of Insured US Transgender and Cisgender Adults

Abstract

Purpose:

Transgender and gender-diverse (TGD) people may have been disproportionately impacted by the COVID-19 pandemic, yet little is known about vaccination status in this population. This multicenter cohort study of insured adults examined the rates of COVID-19 vaccine initiation and completion in TGD persons compared to matched cisgender persons.

Methods:

A cohort of TGD persons and matched cisgender persons enrolled in Kaiser Permanente health plans in Northern and Southern California between 12/1/2020 and 7/31/2021 were analyzed. COVID-19 vaccination initiation and completion rates were compared across groups using Cox regression models.

Results:

Among transmasculine persons, the HR (95% CI) estimates for COVID-19 vaccination initiation and completion were, respectively, 1.35 (1.30–1.40) and 1.78 (1.71–1.85) compared with cisgender women and 1.34 (1.29–1.40) and 1.81 (1.73–1.88) compared with cisgender men. Among transfeminine persons, the corresponding HRs (95% CIs) for vaccination initiation and completion were 1.35 (1.30–1.40) and 1.78 (1.71–1.85) compared with cisgender women and 1.34 (1.29–1.40) and 1.81 (1.73–1.88) compared with cisgender men.

Conclusion:

Findings from this cohort of insured adults demonstrated that TGD persons initiated and completed COVID-19 vaccination at higher rates compared to matched cisgender persons. Further work is needed to understand vaccination rates and determinants in the broader TGD populations.

Introduction:

The COVID-19 pandemic may have disproportionately impacted sexual and gender minority groups, including transgender and gender diverse populations (TGD)^1–3, who historically experience challenges accessing and receiving healthcare.^4,5 The advent of COVID-19 vaccines provides an opportunity to reduce health disparities; however, the marginalization of TGD communities due to lack of trust in medical providers, fear of stigma, and discrimination may contribute to reduced vaccination uptake.^6–8 Though vaccination is an important tool for reducing the impact of COVID-19, there is limited information about COVID-19 vaccine receipt among TGD persons. This multicenter cohort study of insured adults examined the rates of COVID-19 vaccine initiation and completion in transgender persons compared to matched cisgender persons.

Methods:

Study Design & Setting:

Data for analysis were retrieved from the Study of Transition Outcomes and Gender (STRONG), a multicenter cohort study of transgender persons and matched cisgender persons enrolled in Kaiser Permanente (KP) health plans in Northern (KPNC), Southern California (KPSC), and Georgia (KPGA). STRONG was initiated in 2013 with the primary goal of assessing the long-term health status of transgender individuals. In response to the COVID-19 epidemic, additional information for an expanded cohort was collected to assess incidence and severity of COVID-19 infection and vaccination rates at KPNC and KPSC. This study was carried out in collaboration with Emory University which acted as the coordinating center. All procedures detailed in this manuscript underwent review and approval by the Institutional Review Boards (IRBs) at the participating sites with the exemption for informed consent. This study was reported in accordance with the STrengthening the Reporting of OBservational Studies in Epidemiology (STROBE) guidelines.⁹

Participants:

Approximately 3.99 and 3.5 million KNPC and KPSC members, respectively, receive comprehensive care services through these prepaid integrated care systems. Enrollment is open to individuals and their families through employers, state or federal programs like Medicaid (which offers health coverage to low-income individuals and families), or direct enrollment. The methods for cohort attainment have been detailed previously.¹⁰ In summary, two steps were employed to validate eligibility for inclusion in the cohort: 1) identification of TGD people, and 2) determination of transmasculine (TM) or transfeminine (TF) status based on sex assigned at birth. First, to validate TGD status, patients with relevant International Classification of Diseases, 9^th Edition and 10^th edition (ICD-9 and ICD-10) codes and keywords in free-text notes were identified. Second, each validated TGD person was categorized as TM or TF based on a separate program that searched for keywords that reflected sex anatomy (e.g., testes or ovaries), surgical procedures (e.g., orchiectomy or hysterectomy) or hormone therapy (e.g., estrogen or testosterone). An evaluation of text strings containing TM-/TF-specific keywords was performed by 2 trained reviewers independently with discordant results adjudicated a senior member of the research team.

Each validated TGD cohort member was matched with up to 10 male and 10 female cisgender KP members based on race/ethnicity (non-Hispanic white, non-Hispanic black, Asian/Pacific Islander, Hispanic, and other/unknown), year of birth (within a 5-year interval), study site (KPNC or KPSC), and calendar year of membership at the index date to ensure comparable follow-ups.

Follow-up extended from 12/1/2020 (the approximate starting date of mass vaccination) until the first receipt of COVID-19 vaccine, disenrollment from the health plan for more than 90 days, death, or the end of the study period (07/31/2021).

Exposure and Outcomes:

COVID-19 vaccinations were described as administered in a Kaiser-affiliated setting, externally administered, or self-reported. Vaccine initiation was defined as any COVID-19 vaccine date (internally administered, externally administered, or self-reported) between 12/1/2020 and the end of the study. Vaccine completion was defined as one dose of Janssen/Johnson & Johnson (J&J), or two doses of Moderna, Pfizer-BioNTech, or other vaccines.

Analyses:

Rates of COVID-19 vaccination initiation and completion in TGD persons and matched cisgender persons were compared using Cox regression models with results expressed as hazards ratios (HR) and 95% confidence intervals (CI). Multivariable models adjusted for socioeconomic variables including Medicaid coverage and residence in high poverty census tract, and Charlson comorbidity index in addition to matched variables (age, race/ethnicity, date of enrollment, site of enrollment). Matching was accounted for by including the cluster identification number in the stratum statement of each model. Proportional hazards assumptions were examined by log-log plots and by performing a goodness-of-fit test using Schoenfeld residuals. Cohort members with HIV were excluded as part of a sensitivity analysis. Data analyses were performed with SAS, version 9.4 (SAS Institute, Cary, NC).

Results:

The analysis dataset included 4,973 TM and 3,680 TF adults, matched with 73,662 cisgender men and 74,958 cisgender women. TM cohort members were Asian (9.4%), Black (7.6%), Hispanic (24%), Other/Unknown (5.8%), and White (53%). Most TM cohort members were <35 years old (73%), and had no major medical comorbidity defined by the Charlson Comorbidity Index (69%). HIV prevalence among TM cohort members was 0.06% (Table 1). Among TM cohort members, just under one-half (49%) of vaccinations were administered at a Kaiser facility or were externally administered (49%), and 2% were self-reported. (Table 2). TF cohort members were Asian (13%), Black (5.3%), Hispanic (20%), Other/Unknown (6.2%), and White (56%). Most TF cohort members <35 years old (57%) and had no major medical comorbidity defined by the Charlson Comorbidity Index (69%). HIV prevalence among TF cohort members was 2.3% (Table 1). Among TF cohort members just over one-half (52%) of vaccinations were administered at a Kaiser facility, 46% were received externally and 2% were self-reported (Table 2). A greater proportion of TM persons initiated and completed COVID-19 vaccination compared with matched cisgender men (73% vs. 55% and 65% vs. 47%, respectively) and cisgender women (74% vs. 62% and 64% vs. 53%, respectively; all P<0.0001). TF persons were also more likely to initiate and complete COVID-19 vaccination relative to matched cisgender men (65% vs. 47% and 64% vs. 53%, respectively) and cisgender women (65% vs. 53% and 64% vs. 59%, respectively; all P<0.0001). Among TGD persons who completed COVID-19 vaccines, 59% received Pfizer-BioNTech, 41% Moderna, and <1% J&J. After adjusting for potential confounders, the HR (95% CI) estimates for COVID-19 vaccination initiation and completion among TF persons, were, 1.35 (1.30–1.40) and 1.78 (1.71–1.85), respectively, compared with cisgender women, and 1.34 (1.29–1.40) and 1.81 (1.73–1.88), respectively compared with cisgender men (Table 3). Among transfeminine persons, the corresponding HRs (95% CIs) for vaccination initiation and completion were 1.35 (1.30–1.40) and 1.78 (1.71–1.85) compared with cisgender women and 1.34 (1.29–1.40) and 1.81 (1.73–1.88) compared with cisgender men (Table 3). Medicaid insurance and residence in poverty areas were independently associated with lower vaccination rates, whereas Charlson index of at least 1 was independently associated with higher vaccination rates (Table 3). Sensitivity analyses that excluded cohort members with HIV yielded similar results (data not shown).

TABLE 1.

Characteristics of the TGD cohort members (N=8,653)

Participant Characteristic	TF Cohort	TM Cohort
	n (%)	n (%)
Membership Site
KPNC	2721 (74)	3283 (66)
KPSC	959 (26)	1690 (34)
Age at baseline, years
18–24	978 (27)	2102 (42)
25–34	1095 (30)	1551 (31)
35–44	608 (17)	738 (15)
45–54	382 (10)	324 (6.5)
55–64	332 (9.0)	193 (3.9)
65 +	285 (7.7)	65 (1.3)
Race/Ethnicity
Asian/Pacific Islander	466 (13)	467 (9.4)
Non-Hispanic Black	195 (5.3)	378 (7.6)
Hispanic	726 (20)	1181 (24)
Other/Unknown	228 (6.2)	287 (5.8)
Non-Hispanic White	2065 (56)	2660 (53)
Medicaid Insurance Indicator
Yes	558 (15)	759 (15)
No	3122 (85)	4214 (85)
Percent of census tract households living in poverty
<20%	3381 (92)	4530 (91)
20% or more	299 (8.1)	443 (8.9)
Charlson Comorbidity Index
0	2553 (69)	3407 (69)
1 +	1127 (31)	1566 (31)
HIV Status
Positive	84 (2.3)	3 (0.06)
Negative	3596 (98)	4970 (99)
Follow-Up (Days), Mean (SD)	161 (57)	160 (60)
Total	3,680	4,973

Abbreviations: TF = transfeminine; TM = transmasculine; TGD = transgender and gender-diverse; KPNC = Kaiser Permanente Northern California; KPNC = Kaiser Permanente Southern California; SD = standard deviation

TABLE 2.

COVID-19 vaccination initiation and completion among TM and TF participants as compared to matched cisgender men and cisgender women

N (%)	TM cohort	CF referents	P	CM referents	P	TF cohort	P	CF referents	P	CM referents
Initiated vaccine series
No	1328 (27)	16699 (39)	<. 01	18871 (45)	<.0 1	953 (26)	<. 01	10943 (34)	<. 01	12149 (38)
Yes	3645 (73)	26143 (61)	<. 01	23230 (55)	<.0 1	2727 (74)	<. 01	21173 (66)	<. 01	19412 (62)
JJ - External	128 (3.5)	936 (4)		1118 (5)		133 (5)		667 (3)		902 (5)
JJ - Internal	101 (3)	532 (2)		675 (3)		92 (3)		468 (2)		562 (3)
JJ – Self-Reported	9 (<1)	26 (<1)		17 (<1)		6 (<1)		23 (<1)		20 (<1)
Moderna - External	369 (18)	5043 (19)		4400 (19)		440 (16)		4174 (20)		3658 (19)
Moderna - Internal	789 (22)	4845 (19)		4172 (18)		589 (22)		4170 (20)		3579 (18)
Moderna - Self-Reported	27 (<1)	133 (<1)		86 (<1)		22 (<1)		115 (<1)		74 (<1)
Pfizer - External	1075 (<1)	9042 (35)		8034 (35)		752 (28)		6902 (33)		6281 (32)
Pfizer - Internal	827 (23)	5370 (21)		4590 (20)		662 (24)		4471 (21)		4232 (22)
Pfizer – Self-Reported	49 (1)	215 (<1)		136 (<1)		31 (1)		182 (<1)		104 (<1)
Other	1 (<1)	2 (<1)		1 (<1)		0 (0)		0 (0)		1 (<1)
Completed vaccine series *
No	1761 (35)	19998 (47)	<. 01	22341 (53)	<.0 1	1318 (36)	<. 01	13300 (41)	<. 01	14782 (47)
Yes	3212 (65)	22844 (53)	<. 01	19760 (47)	<.0 1	2362 (64)	<. 01	18816 (59)	<. 01	16779 (53)
JJ - External	0 (0)	6 (<1)		7 (<1)		0 (0)		1 (<1)		3 (<1)
JJ - Internal	0 (0)	3 (<1)		2 (<1)		0 (0)		3 (<1)		2 (<1)
JJ – Self-Reported	0 (0)	0 (0)		2 (<1)		0 (0)		1 (<1)		0 (0)
Moderna - External	586 (18)	4482 (20)		3930 (20)		402 (17)		3784 (20)		3326 (19)
Moderna - Internal	769 (24)	4770 (21)		4010 (20)		569 (24)		4100 (22)		3570 (21)
Moderna - Self-Reported	24 (<1)	100 (<1)		59 (<1)		18 (<1)		100 (<1)		60 (<1)
Pfizer - External	991 (25)	8013 (35)		7129 (36)		677 (29)		6186 (34)		5632 (34)
Pfizer - Internal	810 (25)	5292 (23)		4516 (23)		671 (28)		4500 (24)		4200 (25)
Pfizer – Self-Reported	31 (<1)	1(<1)		104 (<1)		25 (1)		140 (<1)		85 (<1)
Other	1 (<1)	1 (<1)		1 (<1)		0 (0)		1 (<1)		1 (<1)
Total	4,973	42,842		42,101		3,680		32,116		31,561

Abbreviations: TM = transmasculine; CF = cisgender females; CM = cisgender males

^*1 dose of J&J vaccine or 2 doses of Pfizer or Moderna

TABLE 3.

Rates of COVID-19 vaccine initiation and completion in transmasculine persons and transfeminine persons as compared with matched cisgender men and cisgender women

	Vaccination Initiationa				Vaccination Completionb
Participant Characteristics	Adjustedd		Adjustedd		Adjustedd		Adjustedd
	Incidence Ratec	HR 95% CI	Incidence Ratec	HR 95% CI	Incidence Ratec	HR 95% CI	Incidence Ratec	HR 95% CI
	TM vs. CW		TM vs. CM		TM vs. CW		TM vs. CM
Gender Identity
Cisgender	1.49	(ref)	1.26	(ref)	1.07	(ref)	1.30	(ref)
Transmasculine	1.96	1.35 (1.30–1.40)	1.96	1.78 (1.71–1.85)	1.73	1.34 (1.29–1.40)	1.73	1.81 (1.73–1.88)
Medicaid Insurance Indicator
No	1.29	(ref)	1.36	(ref)	1.09	(ref)	1.16	(ref)
Yes	0.94	0.73 (0.70–0.76)	1.06	0.81 (0.77–0.85)	0.73	0.72 (0.68–0.75)	0.84	0.81 (0.77–0.85)
Living in Poverty Area 5
No	1.29	(ref)	1.35	(ref)	1.12	(ref)	1.16	(ref)
Yes	0.89	0.82 (0.78–0.87)	0.98	0.80 (0.76–0.85)	0.74	0.81 (0.76–0.85)	0.86	0.79 (0.75–0.84)
Charlson Comorbidity Index
0	1.21	(ref)	1.27	(ref)	1.04	(ref)	1.09	(ref)
1+	1.45	1.10 (1.06–1.13)	1.46	1.11 (1.08–1.15)	1.22	1.10 (1.06–1.13)	1.27	1.13 (1.09–1.17)
	TF vs. CW		TF vs. CM		TF vs. CW		TF vs. CM
Gender Identity
Cisgender	1.50	(ref)	1.71	(ref)	1.52	(ref)	1.30	(ref)
Transfeminine	1.99	1.15 (1.10–1.20)	1.99	1.40 (1.34–1.46)	1.72	1.11 (1.06–1.17)	1.72	1.40 (1.33–1.46)
Medicaid Insurance Indicator
No	1.58	(ref)	1.53	(ref)	1.37	(ref)	1.33	(ref)
Yes	1.19	0.71 (0.67–0.75)	1.09	0.78 (0.73–0.83)	0.98	0.69 (0.65–0.73)	0.90	0.77 (0.72–0.82)
Living in Poverty Area e
No	1.58	(ref)	1.53	(ref)	1.37	(ref)	1.33	(ref)
Yes	1.17	0.86 (0.80–0.91)	1.09	0.81 (0.76–0.87)	0.99	0.84 (0.78–0.89)	0.91	0.80 (0.75–0.86)
Charlson Comorbidity Index
0	1.44	(ref)	1.39	(ref)	1.24	(ref)	1.20	(ref)
1+	1.90	1.08 (1.05–1.12)	1.87	1.16 (1.12–1.20)	1.67	1.08 (1.04–1.12)	1.65	1.16 (1.12–1.21)

^aVaccine initiation was defined as any COVID-19 vaccine date within the study period

^bVaccine completion was defined as one dose of Janssen/Johnson & Johnson (J&J), or two doses of Moderna, Pfizer-BioNTech, or other vaccines

^cRate per person-year

^dStratified by cluster ID to account for matching variables (age, race/ethnicity, site, and enrollment at the start of follow-up), other participant characteristics listed in the table are included in the model as covariates

^eDefined as a census tract with at least 20% of households living in poverty

Abbreviations: TM = transmasculine person, TF = transfeminine person, CM = cisgender men, CW = cisgender women

Discussion:

COVID-19 vaccination rates among TGD persons were higher compared to matched cisgender men and cisgender women in this cohort of insured TGD adults. These results are consistent with other reports^11,12 that found TGD persons do not have lower rates of COVID-19 vaccination relative to the general US population. Proportions of TGD or non-binary adults who initiated (75.7%) and completed (71.4%) COVID-19 vaccination did not differ from cisgender adults in the National Immunization Survey Adult COVID module.¹² Vaccination rates overall in TGD and cisgender persons, beyond COVID-19 vaccination, have also shown to be comparable in various settings. The Centers of Disease Control (CDC) National Immunization Survey-Teen data also demonstrated that vaccination rates in 203 TGD adolescents and age-matched cisgender peers at a large tertiary care center in New York State were generally comparable. Specifically, TGD patients had higher rates of human papillomavirus vaccination coverage at 76.4% compared to 60.5% among New York State adolescents.¹³ HPV vaccination initiation and completion prevalence were comparable among cisgender females and higher among cisgender males in adults 18 to 45 years in the Behavioral Risk Factor Surveillance System.¹⁴ TGD persons have been shown to be vaccinated against COVID-19 and other diseases at levels that were similar to those observed among their cisgender counterparts in multiple recent studies. Higher rates of COVID-19 vaccination among TGD adults have been associated with living in an urban area, low social vulnerability, and household income above the poverty line.¹² After adjusting for similar indicators, TGD persons still initiated and completed vaccination at higher rates than matched cisgender men and cisgender women in this cohort of insured transgender adults. Importantly, TGD individuals enrolled in integrated healthcare systems such KP represent a population with health insurance who have access to comprehensive, high-quality, gender-affirming care.¹⁵ Most TGD persons surveyed in the 2022 Transgender Survey reported health insurance (87%), but 25% of respondents indicated at least one problem with their insurance in the last 12 months due to their gender identity.¹⁶

Uninsurance or underinsurance have been repeatedly found to be associated with unmet health needs, including preventative services like vaccination.^17,18 We believe this study underscores the positive effect of reducing structural barriers to care for TGD persons as evidenced in better than expected vaccination uptake. TGD persons with well-developed gender affirming care and routine engagement in this type of care may be more likely to have multiple opportunities to consider vaccination. For this reason, we should be careful in inferring the results of this study to the broader TGD population. Gender non-binary or gender diverse persons were categorized into TM, and TF groups based on their sex assigned at birth because self-reported gender identity data were missing for most participants. Routine gender identity data collection will enable future analyses stratified by self-reported gender identity. Vaccination data completeness was not examined for external validity. Future studies linking vaccination data with state registries may confirm our study findings.

Conclusions:

In this cohort of insured adults, TGD persons initiated and completed COVID-19 vaccination at higher rates compared to matched cisgender men and cisgender women. These findings conform with previous reports that demonstrate sexual and gender minority groups show no difference in vaccination uptake compared to the general US population. These results should be contextualized considering the insured status of those included in this cohort. Addressing structural barriers to care such as lack of health insurance¹⁹, lack of competent providers,²⁰ or fear of stigma and discrimination⁷ may improve COVID-19 vaccine uptake among TGD persons across broader social-economic strata.