Preventive service use among Medicaid expansion adults, 2014–2019 Authors: Michelle Bronsard, Sarah Hirsch, Dawn Pepin, Terrika Barham, Angela T Estadt, Michelle Van Handel, Joshua A Salomon, Adrienne Sabety Published: 2026 ABSTRACT AbstractIntroductionIn 2014, the Affordable Care Act (ACA) mandated that newly eligible Medicaid expansion beneficiaries receive recommended preventive services at no cost. Yet, evidence regarding the Medicaid expansion population's use of preventive services is limited.MethodsMedicaid claims data from 2011 to 2019. We focused on 12 selected preventive services that received new, revised, or upheld A or B USPSTF ratings since 2010.ResultsWe find that receipt of preventive services among the Medicaid Expansion population increased from 22% in 2014 to 26% in 2019, with 41% of beneficiaries receiving at least one of the 12 examined services from 2014 to 2019. The most received services were screening tests for HIV infection, cervical cancer, chlamydia and gonorrhea, and hepatitis C virus infection. Receipt of at least one preventive service was highest among non-Hispanic Asian, non-Hispanic Black, and Hispanic beneficiaries.ConclusionFuture changes to Medicaid expansion eligibility or preventive services recommendations could potentially disrupt these utilization patterns, with possible disproportionate effects on people from racial and ethnic minority populations. Introduction Implemented in 2014, the Patient Protection and Affordable Care Act (ACA) incentivized states to expand Medicaid eligibility to nondisabled, nonpregnant adults with incomes up to 138% of the federal poverty level (“Medicaid expansion”). The ACA also mandated that Medicaid expansion beneficiaries receive services recommended by the US Preventive Services Task Force (USPSTF) with an A or B rating at zero out-of-pocket cost.1 Describing patterns of preventive service use among Medicaid expansion beneficiaries provides important context for the ACA's coverage and cost-sharing provisions, their potential impact on access to care, and how these provisions may impact the use of preventive care. Preventive services are beneficial and cost-effective for low-income adults, a population that faces a high burden of preventable chronic disease and premature mortality. For instance, increased use of preventive services reduces complications and premature deaths from cancer and cardiovascular disease.2,3 This may be especially true for populations identifying as racial and ethnic minorities who face disproportionate burdens of the conditions these preventive services address and have historically faced greater barriers to accessing preventive care.4 However, evidence on preventive service use among Medicaid expansion beneficiaries remains mixed, likely due to a reliance on small samples and survey data.5-7 We analyzed Medicaid claims data across the 31 states and the District of Columbia, which expanded Medicaid from 2014 to 2019. We focused on 12 selected preventive services receiving new, revised, or upheld A or B USPSTF ratings since 2010. We examined trends in the receipt of preventive services among Medicaid expansion beneficiaries by age, sex, region, and race and ethnicity. Study data and methods We used Medicaid claims data from 2011 to 2019, consisting of Medicaid Analytic eXtract Files (2011-2014) and Transformed Medicaid Statistical Information System Analytic Files (2015-2019), and analyzed the data in Redivis (RRID:SCR_023111).8 We aggregated individuals across geography and time using beneficiary identifiers. We categorized beneficiary age, state of residence, biological sex, self-reported race and ethnicity, annual enrollment status, and eligibility type using the demographic file. The study used deidentified secondary data, approved by Stanford University's institutional review board. Generative artificial intelligence tools were used in a limited manner to assist with editing for clarity and grammar. Study population The study population included Medicaid expansion beneficiaries aged 18-64 years in the 31 states and the District of Columbia, which expanded Medicaid during 2014-2019. We focused on nondisabled adults with incomes up to 138% of the federal poverty level who were enrolled for at least 1 month. During 2014-2015, we imputed eligibility based on beneficiaries’ subsequent status in late 2015 and early 2016 in the 12 states that misreported eligibility (Supplement A). We defined a beneficiaries’ race and ethnicity as the most frequently reported nonmissing value from 2011 to 2019 (Figure 1). Proportions of Medicaid expansion beneficiaries receiving selected preventive services by race and ethnicity, 2014-2019. Abbreviations: AI, American Indian; AN, Alaska Native; LTBI, latent tuberculosis infection; NH, Non-Hispanic; NHPI, Native Hawaiian and Other Pacific Islander. For each service, rates are calculated as the number of recipients receiving the selected preventive service over the Medicaid expansion population aged 18-64 years in the relevant race and ethnicity category. Outcomes We examined 12 preventive services receiving new, revised, or upheld A or B ratings from USPSTF since 2010: statins for primary prevention of cardiovascular disease and preexposure prophylaxis for HIV, and screening tests for breast cancer, lung cancer, colorectal cancer, cervical cancer, latent tuberculosis infection, hepatitis C virus infection, hepatitis B virus infection, HIV infection, chlamydia and gonorrhea, and syphilis infection. We identified claims for each service using Current Procedural Terminology and Healthcare Common Procedure Coding System codes from inpatient, outpatient, and prescription claims (Table S1 Current Procedural Terminology codes were used to identify persons receiving services. https://www.ama-assn.org/practice-management/cpt; Healthcare Common Procedure Coding System codes were used to identify persons receiving services. https://www.cms.gov/Medicare/Coding/MedHCPCSGenInfo. International Classifications of Diseases, Ninth Revision (ICD-9) and Tenth Revision (ICD-10) were used to identify persons with an exclusionary diagnosis. Statistical analysis We calculated the proportion of beneficiaries who received 1 or more preventive services during 2014-2019. We also calculated an annual service use proportion: the number of beneficiaries receiving 1 or more services in a year divided by the number of beneficiaries enrolled for at least 1 month that year. A beneficiary was defined as eligible for a service if they met the inclusion criteria specified by the USPSTF. A beneficiary was defined as receiving a service if (1) we observed a claim with a procedure code indicating the service and (2) the beneficiary satisfied the inclusion criteria specified by the USPSTF (see Tables S1 and S2). Limitations There are several limitations. First, we are unable to determine why beneficiaries receive preventive services. Guidelines for some services changed over the study period, which could have influenced whether beneficiaries received those services. Second, we developed the exclusion and inclusion criteria for the numerator to balance sensitivity and specificity in identifying service use; the approach does not precisely capture the proportion of individuals with up-to-date screenings. Third, while we categorize beneficiaries’ receipt of service, we do not determine whether beneficiaries received subsequent downstream care. Fourth, this is a descriptive study of service utilization patterns among Medicaid expansion beneficiaries. We do not make causal claims about the effect of Medicaid expansion itself, as we do not compare expansion states to non-expansion states or use quasi-experimental methods to isolate the impact of the policy change. Fifth, we cannot confirm what beneficiaries actually paid due to incomplete out-of-pocket data in the Medicaid claims. Sixth, the 9.8% of beneficiaries with missing race and ethnicity from 2011 to 2019 were directly categorized as “missing” race. While T-MSIS race and ethnicity data quality varied across states, we use the most frequently reported non-missing value from 2011 to 2019 to mitigate potential misclassification. Additionally, our large sample size provides robust estimates even after excluding those with missing data. Seventh, we use a beneficiary's prior claims to identify whether they met the inclusion criteria specified by the USPSTF. However, we do not observe pre-expansion claims for 63% of our study population and we assume a beneficiary meets our inclusion criteria unless we observe otherwise. Study results The study population included 29 600 366 Medicaid expansion beneficiaries from 2014 to 2019 (Table 1). Within Medicaid expansion beneficiaries, 15 437 253 were women (52%), and 13 220 392 self-identified as members of a racial or ethnic minority group (45%). Table 1Medicaid expansion beneficiaries by selected characteristics, 2014-2019.Year2014201520162017201820192014-2019a n % n % n % n % n % n % n %Unique beneficiaries8 490 99713 186 58015 455 03616 496 85116 317 53816 463 38429 600 366 By race and ethnicity Hispanic1 743 26420.52 696 30720.43 230 43320.93 476 21421.13 453 83521.23 399 13120.65 978 55320.2 NH Asian549 6506.5817 2966.2960 8796.2992 3036.0971 2006.0964 6735.91 723 7405.8 NH AI/AN102 0871.2154 2171.2197 9321.3218 9111.3225 6881.4231 0181.4380 1011.3 NH Black1 312 86015.52 031 74515.42 476 63616.02 765 01216.82 804 07617.22 919 16617.74 849 63916.4 NH NHPI107 3351.3147 3051.1147 9491.0130 1430.8123 5330.8124 3810.8288 3591.0 NH White3 859 21945.56 121 70246.47 093 39945.97 434 38545.17 271 52644.67 350 69544.613 252 96644.8 Missing736 7578.71 092 9058.31 274 6758.21 413 1048.61 408 2488.61 419 0058.62 912 9969.8 By sex Men4 200 39749.56 484 55949.27 628 18149.48 043 44848.87 937 53648.67 950 76748.314 196 41648.0 Women4 290 60050.56 702 02150.87 826 85550.68 453 40351.28 380 00251.48 512 61751.715 437 25352.2 By ageb 18—292 780 76632.74 552 95434.55 406 29335.05 827 90335.35 726 71235.15 681 72534.511 839 98540.0 30—391 692 35319.92 806 67021.33 373 21321.83 676 70022.33 711 23622.73 811 13823.16 061 67520.5 40—491 519 89917.92 269 91217.22 598 97416.82 742 38416.62 708 68816.62 749 02016.74 756 34716.1 50—591 820 07821.42 567 16819.52 909 54518.82 995 15418.22 907 46717.82 915 15617.74 933 28816.7 60—64677 9018.0989 8767.51 167 0117.61 254 7107.61 263 4357.71 306 3457.92 009 0716.8 By region Northeast1 088 41912.82 300 27217.42 625 26117.02 730 22916.62 735 81316.82 750 99816.75 178 55517.5 Midwest1 765 45420.82 845 77221.63 454 35922.44 014 08624.34 011 95824.63 854 76923.46 901 05023.3 South1 159 67413.71 744 95513.22 072 75713.42 328 50714.12 333 35614.32 790 24716.94 631 72815.6 West4 477 45052.76 295 58147.77 302 65947.37 424 02945.07 236 41144.37 067 37042.913 139 32444.4Abbreviations: AI, American Indian; AN, Alaska Native; NH, Non-Hispanic; NHPI, Native Hawaiian and Other Pacific Islander. aAll individuals enrolled in Medicaid expansion for at least 1 month between 2014 and 2019. bFor aggregate years, age is measured at first entry in the dataset. Medicaid expansion beneficiaries by selected characteristics, 2014-2019. Abbreviations: AI, American Indian; AN, Alaska Native; NH, Non-Hispanic; NHPI, Native Hawaiian and Other Pacific Islander. aAll individuals enrolled in Medicaid expansion for at least 1 month between 2014 and 2019. bFor aggregate years, age is measured at first entry in the dataset. Table 2 reports the numbers and proportions of Medicaid beneficiaries overall and within specific demographic strata receiving 1 or more services by year. A total of 12 230 477 beneficiaries received 1 or more examined preventive services during 2014-2019 (41% = 12 230 477/29 600 366), including 56% of women and 26% of men. Yearly proportions of beneficiaries receiving at least 1 examined service ranged from 1 850 806 in 2014 (22%) to 4 351 939 in 2019 (26%). Across 2014-2019, yearly proportions ranged from 32% to 38% among women and 11% to 14% among men. Table 2Medicaid expansion beneficiaries receiving 1 or more selected preventive services,a 2014-2019.Year2014201520162017201820192014-2019b n % n % n % n % n % n % n %Received ≥ 1 servicec1 850 80621.82 989 70522.73 646 73423.64 168 08725.34 221 08525.94 351 93926.412 230 47741.3 By race and ethnicity Hispanic384 94322.1652 80724.2827 90725.6947 55727.3974 75228.2985 18329.02 665 54644.6 NH Asian143 32426.1226 82027.8279 30029.1305 25630.8309 28731.8313 64732.5836 24748.5 NH AI/AN17 22016.927 85118.139 02419.745 64320.947 84521.249 75621.5140 61137.0 NH Black321 22524.5537 12126.4689 60227.8838 37930.3856 53930.5898 71030.82 300 20247.4 NH NHPI19 73418.430 75920.931 57021.330 74923.630 56424.730 83024.8108 55537.6 NH White809 56521.01 302 37021.31 531 99621.61 704 36822.91 695 30323.31 753 67723.95 207 50339.3 Missing135 47318.4203 51418.6240 63518.9287 37120.3297 79321.1311 01421.9927 90731.9 By sex Men474 36311.3738 62311.4913 97012.01 059 66113.21 093 78313.81 148 61914.43 656 73825.8 Women1 376 44332.12 251 08233.62 732 76434.93 108 42636.83 127 30237.33 203 32037.68 577 95755.6 By aged 18 to 29504 64918.1930 58820.41 172 88921.71 360 53223.31 386 60524.21 412 42424.94 303 80036.3 30 to 39285 48516.9485 83417.3618 07118.3734 08820.0766 18720.6816 01921.42 216 20536.6 40 to 49347 24422.8518 54822.8603 07423.2666 87824.3667 40024.6691 74625.22 008 38442.2 50 to 59523 60228.8764 29729.8898 15730.9993 96533.2974 84033.5986 74133.82 620 02753.1 60 to 64202 22529.8311 31831.5380 30332.6444 30235.4458 92336.3480 21236.81 082 06153.9 By region Northeast317 46329.2601 25826.1670 87725.6712 63126.1724 17126.5778 32428.32 192 02842.3 Midwest392 36122.2686 74324.1837 90724.31 053 75726.31 049 40326.21 036 65726.92 931 04042.5 South246 93421.3304 60017.5430 23320.8546 82123.5568 24324.4671 38624.11 663 80335.9 West894 04820.01 397 10422.21 707 71723.41 854 87825.01 879 26826.01 865 57226.45 492 47741.8 Received selected servicee Breast cancer screening tests414 3074.9622 2354.7723 4774.7780 2104.7771 2294.7784 8464.82 488 8648.4 Cervical cancer screening tests623 4167.3974 0957.41 163 0687.51 282 6757.81 237 7167.61 220 0447.44 854 30316.4 Colorectal cancer screening tests334 3073.9507 2363.9620 4164.0687 2674.2670 4704.1679 8094.12 308 9447.8 Hepatitis B virus infection screening tests282 2723.3448 1443.4568 2863.7677 3064.1702 8304.3740 0074.52 921 3759.9 Hepatitis C virus infection screening tests374 8574.4598 0924.5782 1225.11 007 7666.11 047 9576.41 089 4676.64 083 64413.8 HIV infection screening tests482 0595.7842 8416.41 118 0907.21 374 4218.31 475 8329.01 539 2339.35 251 42917.7 Lung cancer screening tests<160.08980.05 7550.013 0330.121 0670.128 7310.260 3620.2 Preexposure prophylaxis for HIV (PrEP)27200.047350.090240.113 2840.117 3150.118 5710.140 4240.1 Statin use for primary prevention of cardiovascular disease186 5132.2235 9971.8240 3921.6238 6891.4244 0271.5258 5391.6773 4532.6 Chlamydia and gonorrhea screening tests672 8407.91 216 6849.21 588 64010.31 902 68511.51 986 31512.22 082 21912.64 825 90716.3 Syphilis infection screening tests358 0624.2631 8094.8853 5515.51 063 3716.51 150 9777.01 262 3097.74 054 54213.7 Latent tuberculosis infection screening tests166 6472.0283 0762.1349 3332.3415 7242.5442 0232.7482 8442.91 785 7436.0Abbreviations: AI, American Indian; AN, Alaska Native; LTBI, latent tuberculosis infection; NH, Non-Hispanic; NHPI, Native Hawaiian and Other Pacific Islander; PrEP, preexposure prophylaxis. aServices included are statins, PrEP for HIV, and screening tests for breast cancer, lung cancer, colorectal cancer, cervical cancer, latent tuberculosis infection, hepatitis C virus infection, hepatitis B virus infection, HIV infection, chlamydia and gonorrhea, and syphilis infection. bFor aggregate years, proportions (%) are calculated as the number of beneficiaries receiving at least 1 service during any year between 2014 and 2019 divided by the number of unique Medicaid Expansion beneficiaries in the dataset over the period, overall or within specific demographic strata. cFor each demographic category (race and ethnicity, sex, age, region), proportions (%) are calculated as the number of service recipients in that category divided by the Medicaid expansion population in that category (reported in Table 1). dAge categories are assigned based on the age at first appearance in the dataset. eFor specific services in this section of the table, proportions (%) are calculated as the number of persons receiving that service divided by the total number of unique Medicaid expansion beneficiaries in the sample during that year (reported in Table 1, Row 1). Medicaid expansion beneficiaries receiving 1 or more selected preventive services,a 2014-2019. Abbreviations: AI, American Indian; AN, Alaska Native; LTBI, latent tuberculosis infection; NH, Non-Hispanic; NHPI, Native Hawaiian and Other Pacific Islander; PrEP, preexposure prophylaxis. aServices included are statins, PrEP for HIV, and screening tests for breast cancer, lung cancer, colorectal cancer, cervical cancer, latent tuberculosis infection, hepatitis C virus infection, hepatitis B virus infection, HIV infection, chlamydia and gonorrhea, and syphilis infection. bFor aggregate years, proportions (%) are calculated as the number of beneficiaries receiving at least 1 service during any year between 2014 and 2019 divided by the number of unique Medicaid Expansion beneficiaries in the dataset over the period, overall or within specific demographic strata. cFor each demographic category (race and ethnicity, sex, age, region), proportions (%) are calculated as the number of service recipients in that category divided by the Medicaid expansion population in that category (reported in Table 1). dAge categories are assigned based on the age at first appearance in the dataset. eFor specific services in this section of the table, proportions (%) are calculated as the number of persons receiving that service divided by the total number of unique Medicaid expansion beneficiaries in the sample during that year (reported in Table 1, Row 1). Table 2 shows that receipt of at least one preventive service was highest among non-Hispanic Asian (49%), non-Hispanic Black (47%), and Hispanic (45%) beneficiaries. During 2014-2019, the greatest relative changes in service receipt were among non-Hispanic Native Hawaiian and Other Pacific Islander (35%), Hispanic (31%), and non-Hispanic American Indian or Alaska Native (28%) beneficiaries. Beneficiaries received the highest proportions of screening tests for HIV infection (18%), cervical cancer (16%), chlamydia and gonorrhea (16%), and hepatitis C virus infection (14%) (Table 2). The receipt of specific services also varied by race and ethnicity (Figure 1 and Table 2). Compared with other groups, non-Hispanic Black and Hispanic beneficiaries received screening tests for HIV infection, syphilis infection, and chlamydia and gonorrhea in the highest proportions. Screening proportions for breast cancer, colorectal cancer, and hepatitis B virus infection were highest among non-Hispanic Asian beneficiaries. Non-Hispanic White beneficiaries had the highest lung cancer screening proportions. Discussion During 2014-2019, 41% of Medicaid expansion beneficiaries received at least one preventive service covered under the ACA preventive services requirement. Service receipt was highest among non-Hispanic Asian, non-Hispanic Black, and Hispanic beneficiaries, consistent with studies focused on HIV testing or using survey data.9,10 A recent legal challenge to the ACA preventive services requirement, Kennedy v. Braidwood Mgt., raised concerns about the continued protection of no-cost coverage of USPSTF-recommended services.11,12 Although the case eventually was resolved in favor of continuing the mandate, the potential for future challenges to the ACA remains. This issue is of particular concern because research shows that increases in out-of-pocket costs lead to a decline in the use of preventive services.13 For instance, a modest increase in the cost of preexposure prophylaxis for HIV was associated with patients abandoning their prescriptions.14 Together with our descriptive results, these findings suggest that changes to Medicaid and USPSTF guidance or coverage requirements could impact the use of preventive services, widening existing health inequities in the United States. Taken together, these findings indicate that a substantial proportion of Medicaid expansion beneficiaries received one of the 12 examined preventive services. Our observed trends show increasing use of these services from 2014 to 2019. While our descriptive study cannot establish causality, changes to insurance coverage or the requirement could potentially disrupt these utilization patterns, with possible disproportionate effects on people from racial and ethnic minority populations. Supplementary Material