Cost-Effectiveness of Revised US Pneumococcal Vaccination Recommendations in Underserved Minority Adults <65-Years-Old

Introduction

Newly approved 15- and 20-valent pneumococcal conjugate vaccines (PCV15 and PCV20) were recently recommended for US adults, with options to give either PCV20 alone or PCV15 followed by 23-valent pneumococcal polysaccharide vaccine (PPSV23) one year later to high-risk adults aged <65-years-old and to all adults aged 65 years and older.[1] These vaccines provide additional protection from pneumococcal serotypes circulating in the US population compared to the previously recommended 13-valent pneumococcal conjugate vaccine (PCV13). In adults 19–64 years of age with conditions placing them at higher pneumococcal disease risk, such as immunocompromising conditions, or chronic heart or lung disease, PCV15 use increases the proportion of pneumococcal disease covered by vaccine-contained serotypes by an absolute 12–14% compared to PCV13 (depending on age and chronic health status), while PCV20 use increases serotype coverage by an absolute 28–29% compared to PCV13.[2]

Economic analyses performed by CDC found PCV20 or PCV15/PPSV23 use in high-risk adults aged <65-years-old to be cost-saving compared to the previous pneumococcal vaccination recommendation for this group,[3] however, some experts advocate vaccinating all 50-year-olds to further simplify vaccination recommendations and to help reduce racial disparities in disease burden.[1] Underserved minority populations <65 years of age are at higher risk for pneumococcal disease and are more likely to have undiagnosed conditions placing them at higher risk of pneumococcal disease.[4,5] In addition, underserved minorities develop high-risk conditions at younger ages and have greater early mortality than other populations.[6] Thus, many persons of minority background may die before they reach age 65 years. Due to these factors that increase their risk of pneumococcal disease at younger ages, the underserved could disproportionately obtain clinical and public health benefits from routine pneumococcal vaccination before age 65 years compared to other segments of the US population.

Prior work suggests that routine pneumococcal vaccination at age 50 could be advantageous for the US general population,[7] with even greater benefits in underserved populations.[6] Here, using the Black population as a proxy for underserved minorities, we compare the cost-effectiveness of the most recent (2022) and prior CDC pneumococcal vaccination recommendations to strategies for the general population that vaccinate all 50-year-olds in Black and non-Black cohorts.

Methods

A Markov decision analysis model estimated the incremental cost-effectiveness of general population pneumococcal vaccination strategies in a hypothetical 1-year US Black population cohort aged 50 years. Strategies were also separately compared in a 50-year-old US non-Black cohort. Both options in the new adult pneumococcal vaccination recommendation (either PCV20 or PCV13 followed 1 year later by PPSV23 in persons aged <65 years with high-risk conditions and in all persons aged ≥65 years) were compared to each other and to four other age-based strategies: 1) PCV20 for all 50-year-olds, 2) PCV20 at ages 50 and 65 years old, 3) PCV15 and PPSV23 for all 50-year-olds, and 4) PCV15 and PPSV23 at ages 50 and 65. The analysis also included a strategy incorporating the prior recommendations, where PPSV23 was given to 50-year-olds with high-risk non-immunocompromising chronic medical conditions (CMC) and those with immunocompromising conditions were given both PCV13 and PPSV23. CMC and immunocompromising conditions were defined using CDC criteria.[1]

US populations were segmented into Black and non-Black cohorts and further segmented by health status using age and race-specific population data from National Health Interview Survey (NHIS) and National Center for Health Statistics (NCHS) datasets. Health status segments included average risk, smokers with otherwise average risk, non-immunocompromising chronic medical conditions conferring high pneumococcal disease risk (referred to hereafter as CMC), and immunocompromising conditions. Individuals aged <65 years with these conditions were vaccinated in accordance with current and prior US pneumococcal recommendations when those modeled strategies were considered. As the cohort aged, new onset of a CMC or an immunocompromising condition could occur in the average risk or the average risk smoker health state, based on NHIS data. In this circumstance, indicated pneumococcal vaccination could occur in the current recommendation and prior recommendations strategies. In the other strategies, age-based vaccination could occur. We assumed that age-based or risk-based vaccination uptake at ages 50–64 years was equally likely (38.3%)[8] regardless of indication or race; this rate is higher than reported risk-based uptake at these ages,[9] which could potentially bias results toward CDC recommendation strategies. In this instance and in others, assumptions consistently favored CDC recommendation strategies to potentially assist interpretation of analysis results. Vaccine uptake at ≥65 years of age was race-specific, based on 2018 NHIS surveillance data.[9]

Vaccine effectiveness and waning of effectiveness over time was estimated using, for the most part, values used in CDC-sponsored analyses performed to assist deliberations regarding adult PCV15 and PCV20 use. For PPSV23 effectiveness against invasive pneumococcal disease (IPD), we applied values that we have used previously, which are higher than those used in CDC analysis and could potentially bias our results toward strategies using PPSV23.[10] To examine this possibility, a separate analysis was performed using CDC vaccine effectiveness values. In all analyses, we assumed all conjugate vaccines were equally effective in preventing disease caused by pneumococcal serotypes contained within each vaccine, except for serotype 3, where we assumed a relative effectiveness of 0.25 when compared to conjugate vaccine effectiveness against other vaccine serotypes and was varied from 0.15 to 0.50 in sensitivity analyses, consistent with values used in CDC analyses.[3,8]

The Markov model is depicted schematically in Figure 1. Vaccinated individuals had an attenuated risk of pneumococcal infection, based on age, health state and time since vaccination, however all who became ill, vaccinated or not, could have IPD or nonbacteremic pneumococcal pneumonia (NBP) with outcomes or recovery, disability, or death. Mortality and disability risk in hospitalized NBP were assumed to be 50% of that in IPD. IPD-associated disability risk was estimated based on pneumococcal meningitis incidence.

Figure 1. Markov model schematic for pneumococcal policy strategies.

The blue square denotes the decision among the 7 vaccination strategies considered in the model. The green circular chance nodes represent probabilities of being in a certain health state, probabilities of vaccination based on strategy and health state, and probabilities of differing pneumococcal disease outcomes based on vaccination and health status.

Immuno = immunocompromising conditions, IPD = invasive pneumococcal disease, NBP = nonbacteremic pneumococcal pneumonia, PCV = pneumococcal conjugate vaccine, PPSV23 = pneumococcal polysaccharide vaccine

Age-, race-, and health status-specific IPD incidence data were derived from 2017–2018 CDC Active Bacterial Core surveillance (ABCs) data, the 2 most recent years with available data when this analysis was begun. Hospitalized NBP rates were estimated as 3 times the bacteremic pneumonia rate.[11] Cohorts were followed over their lifetimes from a healthcare perspective with costs and effectiveness discounted at 3% per year. Pneumococcal disease costs were obtained from CDC-sponsored analyses.[8] CDC vaccine prices, private sector, were used to estimate vaccine acquisition costs.[12] All other costs from prior years were inflated using the US Consumer Price Index.

Cost-effectiveness analysis results are presented as incremental cost-effectiveness ratios, calculated as difference in cost divided by the difference in effectiveness between strategies ranked in cost from lowest to highest. Strategy effectiveness was determined using quality adjusted life years (QALYs), the product of health state-specific quality of life utility weights (ranging from 0 = death to 1 = perfect health) and health state duration.

One-way sensitivity analyses were performed, individually varying each parameter over ranges listed in Table 1. In addition, scenarios examined strategy-specific changes in vaccination rates that could occur based on the relative simplicity of the single-dose PCV20 strategy compared to the PCV15/PPSV23 strategy, where PPSV23 is given 1 year after PCV15 is received. In the underserved, simpler vaccination strategies will likely result in higher vaccination rates and greater vaccine protection. Other scenarios examined potential indirect (herd protection) effects from future childhood vaccination with PCV15 or PCV20. Finally, probabilistic sensitivity analyses varying all parameter values simultaneously over distributions were also performed. Parameter distributions were fitted to ranges shown in Table 1, with beta distributions used for probabilities and utilities and gamma distributions used for costs.

Table 1.

Model Input Parameter Values and Ranges

Parameter	Value	Range	Source
Probabilities
Receiving pneumococcal vaccination
50 – 64-year-olds	38.8%	23.4% – 54.6%	[9]
PCV15 serotype coverage
Black Population	35.1%	28.1% – 43.7%
Non-Black Population	41.8%	36.7% – 47.5%
PCV20 serotype coverage
Black Population	52.0%	42.7% – 63.1%
Non-Black Population	60.0%	53.4% – 67.4%
Serotype 3 likelihood
Black Population	10.1%	7.1% – 14.2%
Non-Black Population	17.3%	14.6% – 20.6%
PPSV23 serotype coverage
Black Population	63.7%	52.9% – 76.6%
Non-Black Population	73.3%	65.7% – 81.8%
Vaccine Effectiveness
PCV against IPD and NBP	Supplemental Table 1
PPSV23 against IPD and NBP	Supplemental Table 1
Serotype 3 Vaccine Effectiveness
PCV15 and PCV20 against IPD	Supplemental Table 1
PCV15 and PCV20 against NBP	Supplemental Table 1
IPD yearly risk per 100,000
Black Population	31.8	29.3 – 34.5	CDC ABCs data
Non-Black Population	14.7	13.9 – 15.5
Non-bacteremic pneumococcal pneumonia yearly risk per 100,000*
Inpatient
Black Population	71.0	67.2 – 75.0	CDC ABCs data [11,19]
Non-Black Population	34.8	33.8 – 36.0
Outpatient
All Populations	117.9	23.2 – 235.8	[19,20]
IPD case fatality
Black Population	7.8%	5.7% – 10.2%	CDC ABCs data
Non-Black Population	12.8%	11.2% – 14.6%
Disability post IPD
Black Population	7.4%	5.4% – 9.8%	CDC ABCs data
Non-Black Population	6.7%	5.5% – 8.1%
Costs
Vaccine
PPSV23	$110.45	$48.78 – $152.45	[12]
PCV13	$226.43	$130 – $346	[12]
PCV15	$216.09	$136 – $356	[12]
PCV20	$249	$127 – $358	[12]
Administration	$29.20	$21.33 – $31.18	[8,21]
Vaccine side effects (per occurrence)	$0.76	$0–$2	Ibuprofen cost, Walgreens
IPD	$55,408	$51,752 – $59,239	[8]
Pneumonia- discharged alive	$26,464	$25,997 – $26,979
Outpatient pneumonia	$709	$701 – $716	[8]
Initial treatment of symptoms for IPD	$5	$0 – $10	Amoxicillin and ibuprofen cost, Walgreens
Disability (per yr)	$14,239	$7,947 – $21,763	[22]
Utility weights
Disability	0.4	0.21 – 0.59	Estimate[23]
Hospitalization	0.2	0.11 – 0.30	[24,25]
Vaccine side effects	0.9	0.77 – 0.97	Estimate[23]
Average risk	0.83	0.78 – 0.88	Age adjusted[24]
High risk	0.72	0.67 – 0.77	Age adjusted[24]
Disutility- outpatient NBP	0.004	0.002 – 0.006	[19]
Illness durations (days)
IPD	34	17–51	[24,25]
Hospitalized pneumonia	34	17–51	[24,25]
Vaccine side effects	3	1–8	[26]

^*Inpatient nonbacteremic pneumonia risk estimated as 3 times the bacteremic pneumonia risk from ABCs data. Outpatient nonbacteremic pneumonia risk estimated using the methods of the cited references

Results

Public health effects for all strategies under baseline vaccine effectiveness assumptions are shown in Table 2. In both populations, giving PCV15 and PPSV23 at ages 50 and 65 years resulted in the fewest IPD and NBP cases and deaths overall, followed by PCV20 at ages 50/65. Compared to PCV15/PPSV23 at 50/65, use of either current CDC recommended strategy resulted in 1.2% more pneumococcal disease cases and 2.2–2.4% more pneumococcal disease deaths in the Black cohort. In the non-Black cohort, 1.0–1.1% more cases and 1.6–1.7% more deaths occurred with current recommendations.

Table 2.

Public health results for pneumococcal vaccination strategies in 50-year-old cohorts under base case assumptions

Strategy	IPD		NBP
	Cases	Deaths	Cases	Deaths
Black Population (n = 549,197)
PCV20 age 50 only	5,564	765	72,701	557
Recommended PCV20	5,462	751	73,062	549
PCV15/PPSV23 age 50 only	5,425	749	73,259	558
Prior recommendations	5,474	754	72,529	559
Recommended PCV15/PPSV23	5,400	745	73,069	553
PCV20 ages 50/65	5,324	733	72,461	542
PCV15/PPSV23 ages 50/65	5,221	722	72,349	548
Non-Black Population (n = 3,485,130)
PCV20 age 50 only	26,515	3,822	402,750	3,451
Recommended PCV20	25,207	3,636	404.131	3,383
PCV15/PPSV23 age 50 only	26,173	3,778	401,780	3,461
Prior recommendations	25,122	3,645	405,108	3,445
Recommended PCV15/PPSV23	24,866	3,605	404,175	3,408
PCV20 ages 50/65	24,780	3,581	401,176	3,343
PCV15/PPSV23 ages 50/65	24,270	3,529	400,487	3,374

ICER = Incremental cost-effectiveness ratio, QALY = quality adjusted life year, PCV15 = 15-valent pneumococcal conjugate vaccine, PCV20 = 20-valent pneumococcal conjugate vaccine, PPSV23 = 23-valent pneumococcal polysaccharide vaccine

In both Black and non-Black cohorts, PCV20 given only at age 50 years was the least expensive strategy in terms of per person total pneumococcal vaccination and disease cost, as shown in Table 3. Current recommendation strategies, using either PCV20 or PCV15/PPSV23 regimens, and the prior recommendation (using PCV13/PPSV23) were dominated (more expensive and less effective) in both cohorts when compared to other strategies. In the Black cohort, PCV15/PPSV23 given only at age 50 years cost $104,723 per quality adjusted life (QALY) gained, while PCV15/PPSV23 given at both ages 50/65 cost $240,952/QALY compared to PCV15/PPSV23 only at age 50. In non-Blacks, giving PCV15 and PPSV23 at ages 50/65 cost $306,017/QALY while giving the PCV15/PPSV23 combination only at age 50 cost $195,985/QALY gained. Giving PCV20 at ages 50/65 was unfavorable in both Black and non-Black populations under base case vaccine effectiveness assumptions, due to higher incremental cost-effectiveness ratio than more effective strategies (i.e., extended dominance).

Table 3.

Cost-effectiveness results – base case vaccine effectiveness assumptions

Strategy	Costa	Incremental Cost	Effectiveness QALY	Incremental Effectiveness	ICER $/QALY
Black Population
PCV20 age 50 only	$858	-	12.73953	-	-
Recommended PCV20	$879	$21	12.73916	−0.00037	Dominatedb
PCV15/PPSV23 age 50 only	$893	$35	12.73987	0.00034	$104,723
Prior recommendation	$894	$1	12.73934	−0.0005	Dominated
Recommended PCV15/PPSV23	$923	$30	12.73940	−0.00047	Dominated
PCV20 ages 50/65	$932	$38	12.73991	0.00005	Ext. Dom.c
PCV15/PPSV23 ages 50/65	$1,000	$106	12.74031	0.00044	$240,952
Non-Black Population
PCV20 age 50 only	$574	-	13.65066	-	-
Recommended PCV20	$601	$25	13.65058	−0.00008	Dominated
PCV15/PPSV23 age 50 only	$610	$36	13.65084	0.00018	$195,985
Prior recommendation	$631	$21	13.65067	−0.00017	Dominated
Recommended PCV15/PPSV23	$650	$40	13.65074	−0.00009	Dominated
PCV20 ages 50/65	$669	$59	13.65102	0.00018	Ext. Dom.
PCV15/PPSV23 ages 50/65	$745	$135	13.65128	0.0004	$306,017

^aTotal per person vaccination and illness cost

^bDominated strategies were more costly and less effective than other strategies

^cExtended dominated (Ext. Dom.) strategies had higher incremental cost-effectiveness ratios than more effective strategies.

ICER = Incremental cost-effectiveness ratio, QALY = quality adjusted life year, PCV15 = 15-valent pneumococcal conjugate vaccine, PCV20 = 20-valent pneumococcal conjugate vaccine, PPSV23 = 23-valent pneumococcal polysaccharide vaccine

However, when using CDC-based vaccine effectiveness assumptions, where PPSV23 effectiveness against IPD was less, PCV20 use at age 50/65 fared better, costing about $202,000/QALY gained in the Black cohort and about $270,000/QALY in non-Blacks (Table 4). Under these assumptions, PCV15/PPSV23 at 50/65 cost >$650,000/QALY gained in either population when compared to PCV20 at ages 50/65.

Table 4.

Cost-effectiveness results – CDC analysis-based vaccine effectiveness assumptions

Strategy	Costa	Incremental Cost	Effectiveness QALY	Incremental Effectiveness	ICER $/QALY
Black Population
PCV20 age 50 only	$862	-	12.73938	-	-
Recommended PCV20	$881	$19	12.73908	−0.00030	Dominatedb
Prior recommendation	$903	$41	12.73902	−0.00036	Dominated
PCV15/PPSV23 age 50 only	$904	$42	12.73949	0.00010	Ext. Dom.c
Recommended PCV15/PPSV23	$930	$26	12.73916	−0.00033	Dominated
PCV20 ages 50/65	$936	$74	12.73975	0.00037	$202,170
PCV15/PPSV23 ages 50/65	$1,012	$76	12.73986	0.00012	$658,553
Non-Black Population
PCV20 age 50 only	$576	−	13.65059	-	-
Recommended PCV20	$602	$26	13.65053	−0.00005	Dominated
PCV15/PPSV23 age 50 only	$616	$40	13.65063	0.00005	Ext. Dom.
Prior recommendation	$637	$20	13.65047	−0.00016	Dominated
Recommended PCV15/PPSV23	$655	$38	13.65059	−0.00004	Dominated
PCV20 ages 50/65	$671	$95	13.65094	0.00035	$270,281
PCV15/PPSV23 ages 50/65	$753	$81	13.65101	0.00007	$1,096,263

^aTotal per person vaccination and illness cost

^bDominated strategies were more costly and less effective than other strategies

^cExtended dominated (Ext. Dom.) strategies had higher incremental cost-effectiveness ratios than more effective strategies.

ICER = Incremental cost-effectiveness ratio, QALY = quality adjusted life year, PCV15 = 15-valent pneumococcal conjugate vaccine, PCV20 = 20-valent pneumococcal conjugate vaccine, PPSV23 = 23-valent pneumococcal polysaccharide vaccine

One-way sensitivity analyses found that individual variation of parameters over ranges listed in Table 1 did not materially change favored strategies, with either of the newly recommended strategies never favored with these variations; Supplemental Table 2 details these analyses. In scenarios examining increased vaccine uptake when using the simpler single-dose PCV20 strategy under base case vaccine effectiveness assumptions, PCV20 at ages 50/65 became more economically favorable in the Black cohort if absolute vaccination rate increased 15% or more with PCV20 use compared to PCV15/PPSV23 (Supplemental Table 3), with similar effects seen in non-Black cohorts; however, current CDC recommended strategy options remained unfavorable in this analysis. Scenarios examining potential indirect effects from future childhood use of PCV15 or PCV15 similarly demonstrated that current recommendations were more costly and less effective than other strategies regardless of the vaccine used or the magnitude of indirect effects on vaccine serotypes (0–100%) modeled.

In the probabilistic sensitivity analysis, all parameters were simultaneously varied over distributions under base case assumptions. Probabilistic sensitivity analysis results are shown as a cost-effectiveness acceptability curve (Figure 2), which displays the likelihood of each strategy being the most favored compared to other strategies over a range of willingness to pay thresholds in the Black cohort. In this analysis, PCV20 at age 50 years was most likely to be favored at willingness to pay thresholds less than $100,000/QALY gained. At thresholds from $100,000-$250,000/QALY gained, PCV15/PPSV23 at age 50 was most likely to be favored. In non-Blacks (not shown), PCV20 at age 50 was the favored strategy at thresholds less than $200,000/QALY gained. In either cohort, recommended strategies were never favored at thresholds from $50,000-$250,000/QALY gained.

Figure 2. Probabilistic sensitivity analysis of pneumococcal vaccine strategies in the 50-year-old Black cohort under base case assumptions.

Curves depict the likelihood of a strategy being favored vs. all other strategies (y-axis) over a range of willingness-to-pay (or acceptability) thresholds (x-axis) when all parameters are varied simultaneously. PCV20 at age 50 was favored at thresholds less than $100,000 per QALY gained. Between $100,000 and $250,000 per QALY thresholds, PCV15/PPSV23 at age 50 was favored.

PCV15 = 15-valent pneumococcal conjugate vaccine, PCV20 = 20-valent pneumococcal conjugate vaccine, PPSV23 = 23-valent pneumococcal polysaccharide vaccine, QALY = quality adjusted life year

Discussion

This analysis found that recent US adult pneumococcal vaccination recommendations, with risk-based vaccination of adults <65-years-old and age-based vaccination at age 65 years, were economically and clinically unfavorable compared to strategies that vaccinate all 50-year-olds. Among those age-based strategies, favorability of PCV20- or PCV15/PPSV23-based regimens depended on vaccine effectiveness assumptions and potentially increased vaccine uptake when less complex single-dose PCV20 strategies are used, which could disproportionally benefit US Black populations due to their higher risk of pneumococcal disease and increased mortality at younger ages. The unfavorability of newly recommended pneumococcal vaccination strategies in 50-year-old Black and non-Black cohorts was robust to variation of individual model parameters and to simultaneous variation of all parameters in probabilistic sensitivity analyses, and occurred despite model assumptions favoring current recommendation strategies.

The CDC’s decision to revise adult pneumococcal vaccination recommendations was supported by cost-effectiveness analyses that found the newly recommended strategies cost saving and more effective than the prior recommendation (PCV13/PPSV23 in adults aged <65 years with immunocompromising condition, PPSV23 in those adults with other high-risk conditions).[3,8] However, based on Advisory Committee on Immunization Practice (ACIP) guidance, those analyses did not compare the newly recommended strategies to each other. CDC-sponsored analyses also considered PCV20 use for all 50-year-olds with no pneumococcal vaccination thereafter, but did not consider strategies where pneumococcal vaccination could occur again (e.g., at age 65 years) or include PCV15/PPSV23 given at age 50. Their analysis of age-based PCV20 use at age 50 years found it to be health improving and cost saving in some scenarios.

Our analyses differ in several respects from those performed to inform recent CDC/ACIP deliberations. We directly compared strategies using PCV20 and PCV15/PPSV23, including the currently recommended vaccination options, and added consideration of age-based PCV15/PPSV23 use at age 50 years and of age-based vaccination with either newly recommended option at both ages 50/65. In addition, all strategies were considered simultaneously and incrementally in the analysis, the preferred method in most cost-effectiveness analyses.[13] Thus, differences in results from our analysis compared to the CDC-sponsored analysis and other prior results[3] can be attributed, in part, to the strategies compared and the methods used to compare them. Finally, our analyses concentrated on potential effects of US general population vaccination strategies on 50-year-old underserved minority adults, while CDC analyses considered the entire population.

The CDC factors health equity into their vaccination policy decision making and recommendations.[1] In deliberations that produced the revised adult pneumococcal vaccination recommendations, it was postulated that age-based PCV20 use at age 50 years, compared to risk-based strategies: 1) might reduce pneumococcal disease disparity in adults aged 50–64 years, 2) would be easier to implement than risk-based recommendations, and 3) might provide more opportunities to vaccinate adults before underlying high-risk conditions develop or render vaccination less effective. Despite these considerations, age-based PCV20 use at age 50 years was rejected by the ACIP in a split vote (4 vs. 11).[1] Age-based vaccination at both ages 50 and 65 years has not, as yet, been considered by the ACIP, which would require a new evidence review and voting based on ACIP/CDC procedures.

Countering consideration of age- or risk-based use of newer pneumococcal conjugate vaccines in adults is the possibility of herd protection if these newer vaccines are used in children. Direct protection afforded to older adults from PCV13 use, which began in 2014, rapidly decreased after that time due to universal childhood PCV13 use, which indirectly decreased PCV13 serotype disease in all age groups. These effects prompted reevaluation of adult PCV13 use and the subsequent scaling back of recommendations for routine PCV13 use at ages ≥65 years in 2019. Similar effects could occur after PCV15 and/or PCV20 use begins in children, which may occur in the next 1–2 years.[14] In a scenario examining potential indirect effects from childhood PCV15 or PCV20 use, we found that newly recommended strategies continued to be unfavorable in 50-year-old Black and non-Black cohorts compared to age-based vaccination at age 50 years.

Are strategies recommending pneumococcal vaccination for 50-year-olds feasible? Age-based adult pneumococcal vaccination has consistently begun at age 65 years, thus vaccinating at a younger age would require adjustments in implementation to successfully make this change. Strategies such as convenient vaccination services, enhanced medical office systems to facilitate immunization, and other measures have improved vaccination rates.[15] Vaccinating 50-year-olds has some precedent: recombinant herpes zoster vaccination of 50-year-olds is recommended and increasingly performed, albeit at still insufficient rates.[9,16] Recommendations for herpes zoster vaccination and pneumococcal vaccination at age 50 years may improve uptake of both. On the other hand, pneumococcal vaccination only at age 50 years with no revaccination at older ages could result in waning vaccination protection at age 65 years and older, when pneumococcal disease rates, morbidity, and mortality are substantially higher. Our pneumococcal case and mortality projections for strategies that vaccinate only 50-year-olds corroborate this, making those strategies that vaccinate only at age 50 years less viable from a public health standpoint.

Limitations in our analysis included uncertainty regarding key parameter values in our model. PPSV23 effectiveness in preventing nonbacteremic pneumococcal pneumonia has been uncertain and controversial, however, recent data have shown some effectiveness against NBP.[17] Serotype 3 effectiveness in the conjugate vaccines is another area of uncertainty, with PCV13 use in children and adults showing little or no effect on illness from this serotype seen in CDC ABCs data. It is unclear if either of the newer vaccines will provide greater protection against serotype 3 disease. We assumed the same relative decrease of 25% effectiveness against serotype 3 in all PCV vaccines, and variation of this parameter did not substantially change results. If there is differential serotype 3 effectiveness between the newer vaccines, then strategies using the more effective vaccine could be more favorable. We did not consider strategies to increase vaccine uptake, which could also mitigate illness disparities in the underserved. Future vaccines that could increase pneumococcal serotype coverage in adults were also not considered.[18] Additional work to address these limitations and empiric surveillance data documenting the epidemiologic effects of the revised adult pneumococcal vaccination recommendations will inform and assist vaccine policymaking in the future. We also did not consider the potential effects of the COVID pandemic and its aftermath in our analysis. We used pre-pandemic pneumococcal disease epidemiologic data and assume that low pandemic-related pneumococcal disease rates will return to pre-pandemic levels. If rates remain low, then all pneumococcal vaccination strategies will be less cost-effective.

Conclusion

Recent US adult pneumococcal vaccination recommendations, with risk-based vaccination of adults <65-years-old, were economically and clinically unfavorable compared to strategies that vaccinate all 50-year-olds with either PCV20 or PCV15 followed 1 year later by PPSV23 in both Black and non-Black cohorts. Among those strategies, favorability depends on vaccine effectiveness assumptions and the potential for increased vaccine uptake from less complex PCV20 strategies. Easier access to fuller and more prompt protection afforded through a one-vaccine strategy will likely benefit US underserved minority populations more than those facing fewer barriers to preventive health care and may help reduce pneumococcal disease burden inequities.