- Email: [email protected]
Vaccination Timeliness at Age 24 Months in Michigan Children Born 2006–2010
RESEARCH ARTICLE
Vaccination Timeliness at Age 24 Months in Michigan Children Born 2006–2010 Abram L. Wagner, PhD, MPH,1 Amanda M. Eccleston, MPH,1 Rachel C. Potter, DVM, MS,2 Robert G. Swanson, MPH,2 Matthew L. Boulton, MD, MPH1,3 Introduction: Delays in vaccination can stymie the development of herd immunity, and a large proportion of children in the U.S. are known not to receive vaccines on time. This study quantifies delays in vaccination, compares vaccination timeliness to the proportion of children vaccinated, and evaluates the impact of combination vaccine use and timely administration of hepatitis B vaccine birth dose on vaccine timeliness among Michigan children. Methods: This retrospective cohort study used data from the Michigan Care Improvement Registry
—the state immunization information system—for children born 2006–2010. Children aged 24 months as of December 31, 2012, were included. The proportion of children with timely administration of vaccine doses was calculated, and the mean days of vaccination delay with SD were reported.
Results: Among 620,592 Michigan children, 42.2% had received all vaccines, but only 13.2% were vaccinated on time by age 24 months. Children’s vaccinations were delayed an average of 59.2 (SD¼91.2) days by age 24 months for all recommended vaccine doses. Children who received a timely hepatitis B vaccine birth dose or who received a combination vaccine had less delay in vaccination.
Conclusions: Michigan children have high vaccination coverage based on standard measures but few receive these vaccines on time. Promoting use of combination vaccines may improve parental compliance with timely vaccination of children. Am J Prev Med 2018;54(1):96–102. & 2017 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.
INTRODUCTION
I
n the U.S., the Advisory Committee on Immunization Practices (ACIP) produces a recommended immunization schedule, which provides information on the age at which each dose should be administered and the allowable interval between doses; this schedule is approved by professional societies, such as the American Academy of Pediatrics.1 Vaccination programs have traditionally been assessed through measurements of vaccine coverage, the proportion of children who have received a vaccine without any consideration of the timeliness.2 Analyses of the Vaccine Safety Datalink, a consortium of ten healthcare delivery organizations, show that although overall vaccination coverage is generally high (490%) in the U.S., timely vaccination is not3,4; less than 50% of children born 96 Am J Prev Med 2018;54(1):96–102
between 2004 and 2008 within the Vaccine Safety Datalink had received all vaccine doses on time.3 Nationwide studies have also found a high rate of noncompliance with the pediatric immunization schedule; according to the 2012 National Immunization Survey, only 26% of children in the U.S. had received all vaccines at the recommended ages.5
From the 1Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, Michigan; 2Division of Immunizations, Michigan Department of Health and Human Services, Lansing, Michigan; and 3Department of Internal Medicine, Division of Infectious Diseases, University of Michigan Medical School, Ann Arbor, Michigan Address correspondence to: Abram L. Wagner, PhD, MPH, 1415 Washington Heights, Ann Arbor MI 48109. E-mail: [email protected]. 0749-3797/$36.00 https://doi.org/10.1016/j.amepre.2017.09.014
& 2017 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.
Wagner et al / Am J Prev Med 2018;54(1):96–102
Delaying vaccine receipt beyond ACIP-recommended ages places children at risk for potentially serious vaccine-preventable diseases (VPDs). A review of recent pertussis outbreaks have found that a large proportion of cases have been undervaccinated.6 During an outbreak of pertussis in 2012 in Oregon, undervaccinated infants had a quicker onset of disease compared with fully vaccinated infants and therefore potentiated an earlier spread of disease.7 Lack of vaccine timeliness extends beyond individual risk by delaying the establishment of herd immunity and the protection of infants too young to receive vaccinations or children contraindicated from receiving vaccines.8 Vaccine timeliness is an important area of research because of the potential health consequences of untimely vaccination and because it is a more precise methodology for quantifying immunization system performance. Using data from children born between 2006 and 2010 in Michigan’s immunization information system (IIS), this paper examines trends in vaccine timeliness, assesses vaccine delay, compares vaccine timeliness with the proportion of children vaccinated at 24 months, and evaluates factors associated with increased vaccine timeliness.
METHODS This retrospective cohort study used data obtained from the Michigan Care Improvement Registry (MCIR), Michigan’s IIS, which were analyzed in 2015. This comprehensive immunization database was created in 1998 and by law, all vaccinations administered to children aged o20 years must be reported.9 The database is populated by birth data from the state’s electronic birth certificate system. Children with and without private insurance are included in the system, and providers may create new records for children not born in Michigan. MCIR is one of six sentinel IISs in the U.S., selected on account of having ≥85% of providers actively using the IIS, ≥85% of children participating in the IIS, and ≥70% of vaccine doses submitted to the IIS within 30 days.10 The Michigan Department of Health and Human Services regularly assesses providers through the assessment, feedback, and incentives exchange program and that vaccinations reported to the IIS are the standard for this assessment. This study includes the immunization records from all Michigan children born January 1, 2006, through December 31, 2010. Children were only included if they had received one or more vaccine dose to exclude children who had moved out of state. Vaccine doses were valid if given within the ACIP-recommended age range. Doses were excluded if given before the ACIPrecommended age or if violating the minimum recommended interval between doses of one vaccine series. Vaccine doses were considered valid if they were given in the 4 days prior to the beginning of the ACIP-recommended age range or vaccine interval. To study biases resulting from children moving out of state, a sensitivity analysis was performed, limited to children who had January 2018
97
been vaccinated after their first birthday. This analysis gave substantively similar findings (results not shown). The vaccines assessed include four doses of the diphtheria, tetanus, acellular pertussis vaccine (DTaP); three doses of the inactivated polio vaccine (IPV); one dose of the measles, mumps, rubella vaccine; three (or four) doses of the Haemophilus influenzae type b vaccine (Hib); three doses of the hepatitis B (HepB) vaccine; one dose of the varicella vaccine; and four doses of the pneumococcal conjugate vaccine (PCV). Together, these vaccines make up the 4:3:1:3:3:1:4 series that are recommended before age 2 years. These vaccine series are required before enrolling at a school or licensed day care (although Hib and PCV are not required after 5 years).11,12 Rotavirus was not included because it should not be administered after age 8 months and is not required for school entry. The number of doses required for the Hib vaccine series depends on the vaccine manufacturer. Three Hib doses were assessed when the PRP-OMP vaccine was used for every dose; otherwise, a four-dose Hib schedule was assessed. The last dose of both schedules is a booster dose recommended at age 12–15 months; therefore, calculations regarding the final recommended vaccine dose in both schedules are marked as the last dose, whereas Hib “dose 3” only includes children on a four-dose Hib schedule. There was a nationwide shortage of Hib vaccine between December 2007 and July 2009, and so its requirement before school entry was suspended during this period. For this reason, summary statistics are calculated with and without Hib doses included. The outcomes of interest include a standard measure of the proportion of children vaccinated, in addition to vaccine timeliness and delayed vaccination. All measures were assessed when children were aged 24 months (i.e., the child’s second birthday), which corresponds to a common Healthcare Effectiveness Data and Information Set measure.13 A timely dose was administered within 1 month (ranging from 28 to 31 days depending on the length of the month) of the ACIP-recommended age range at vaccination. Delay in vaccination was calculated as the difference, in days, between actual vaccine dose administration and when the vaccine dose is recommended for administration based on the ACIP-recommended acceptable time range (up to 24 months). For example, the recommended age of administration for the first dose of DTaP is 2 months. Children were delayed if they had not yet received the first dose of DTaP at age 3 months. The average number of days delayed by age 24 months was calculated in a method similar to that used by Glanz et al.3 or Glauber.14 This value summed the number of days delayed for each vaccine dose by age 24 months, and divided this total number by the total number of vaccine doses that a child should have received according to the ACIP-recommended immunization schedule. Factors associated with vaccine timeliness were assessed, including the receipt of any combination vaccine doses. Delayed vaccination was evaluated in children who had received one or more combination vaccine doses compared with children who did not receive any combination vaccine doses. The receipt of the HepB birth dose on time was also evaluated as an early predictor of later vaccine timeliness. The HepB birth dose was considered on time if it was given within 3 days of birth, with the day of birth counted as day 0. The average number of days of delayed vaccinations by age 24 months was compared between children who had and had not received the HepB birth dose on time. All analyses were conducted using SAS, version 9.3, software.
98
Wagner et al / Am J Prev Med 2018;54(1):96–102
Table 1. Vaccine Coverage (%) at 24 Months in the Michigan Care Improvement Registry (n¼620,592) Vaccine Hepatitis B dose 1 Hepatitis B dose 2 Hepatitis B dose 3 DTaP dose 1 DTaP dose 2 DTaP dose 3 DTaP dose 4 Hib dose 1 Hib dose 2 Hib dose 3 Hib last dose IPV dose 1 IPV dose 2 IPV dose 3 PCV dose 1 PCV dose 2 PCV dose 3 PCV dose 4 MMR dose 1 Varicella dose 1 All doses All doses except Hib
All years
2006
2007
2008
2009
2010
97.7 95.4 89.3 97.7 95.6 92.4 77.8 90.0 83.1 80.0 50.1 97.2 95.0 89.3 96.4 93.7 89.5 76.3 87.2 86.3 42.2 65.9
97.9 96.0 89.9 97.5 95.6 92.6 78.1 88.9 80.8 70.9 33.6 97.2 95.2 88.5 95.7 93.1 88.8 75.0 88.1 85.9 26.6 64.5
97.7 95.4 88.6 97.4 95.2 92.0 76.8 81.0 64.4 60.4 19.0 96.9 94.7 87.6 95.9 93.2 88.9 75.4 86.2 85.4 14.9 63.4
97.5 95.0 88.8 97.6 95.4 92.2 76.9 91.2 88.0 82.7 59.8 97.0 94.7 88.8 96.4 93.6 89.3 76.9 86.1 85.7 50.4 65.5
97.7 95.2 89.7 98.0 95.9 92.9 78.9 96.5 93.9 90.0 74.9 97.4 95.3 91.0 97.0 94.4 90.2 77.3 87.8 87.4 64.5 68.2
97.8 95.2 89.9 98.2 95.8 92.6 78.7 93.6 90.4 86.7 68.0 97.5 95.1 90.7 97.3 94.5 90.4 77.4 87.8 87.5 58.8 68.5
DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
This study was reviewed by the Michigan Department of Health and Human Services IRB and determined to be exempt from human subjects research.
RESULTS A total of 636,998 Michigan children born 2006–2010 were selected into the study. After removing children who did not receive any valid vaccine doses during the study period (n¼16,406), the final study population contained 620,592 Michigan children. Overall, 42.2% of children had been vaccinated with the entire 4:3:1:3:3:1:4 schedule by age 24 months; excluding Hib doses, this figure was 65.9%. At age 36 months, 49.8% were vaccinated with all doses, and at age 48 months, 52.8% were (excluding Hib doses, these figures were 74.0% and 77.0%, respectively). At age 24 months, the proportion of children immunized varied widely by dose and vaccine, from 50.1% for the last dose of Hib to 97.7% for DTaP dose 1 (Table 1). Among these doses of vaccines whose overall coverage was o90%, coverage held steady for HepB dose 3, DTaP dose 4, IPV dose 3, PCV dose 4, and for the measles, mumps, rubella and varicella doses, whereas it increased between 2006 and 2010 for the Hib doses. Figure 1 depicts vaccination coverage as with increasing age, for children born in 2010. In contrast to the high proportion of children vaccinated at age 24 months, only 13.2% of all the children received all
4:3:1:3:3:1:4 doses (27.1%, if excluding Hib doses). Table 2 shows the timeliness of vaccine doses over the study period. Timeliness ranged from a low of 32.9% for Hib last dose to a high of 87.5% for HepB dose 3. Timeliness was increased across birth cohorts; excluding Hib doses, it moved from 12.6% to 30.0% between 2006 and 2010. By age 24 months, children had delayed each recommended vaccine for an average of 59.2 days (SD¼91.2). Across all birth cohorts, the vaccine doses characterized by the most prolonged periods of delay were all Hib doses, particularly Hib dose 2 with a mean delay of 149.9
Figure 1. Vaccine coverage in Michigan by age up to 24 months, for children born in 2010 (n¼114,051). DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
www.ajpmonline.org
99
Wagner et al / Am J Prev Med 2018;54(1):96–102
Table 2. Vaccine Timeliness (%) at 24 Months in the Michigan Care Improvement Registry (n¼620,592) Vaccine Hepatitis B dose 1 Hepatitis B dose 2 Hepatitis B dose 3 DTaP dose 1 DTaP dose 2 DTaP dose 3 DTaP dose 4 Hib dose 1 Hib dose 2 Hib dose 3 Hib last dose IPV dose 1 IPV dose 2 IPV dose 3 PCV dose 1 PCV dose 2 PCV dose 3 PCV dose 4 MMR dose 1 Varicella dose 1 All doses All doses except Hib
All years
2006
2007
2008
2009
2010
79.0 74.6 87.5 85.5 73.4 62.2 65.1 58.5 49.5 50.0 32.9 82.8 70.4 87.0 81.2 68.8 57.2 61.4 73.8 74.0 13.2 27.1
77.9 74.5 88.1 86.4 74.8 63.9 65.3 35.5 30.6 44.2 23.8 82.9 70.9 85.6 81.2 69.3 58.0 59.4 75.5 73.1 6.3 12.6
77.7 73.3 86.9 85.5 73.3 61.8 64.2 35.7 30.3 37.2 10.6 81.8 69.2 84.9 81.6 69.2 57.4 60.6 73.0 73.1 2.8 12.6
79.2 73.8 87.0 84.8 72.3 60.8 64.5 72.3 59.4 47.9 33.1 81.1 68.3 86.7 80.9 68.1 56.5 61.8 72.8 73.5 13.2 16.2
80.5 75.6 87.7 85.0 72.9 61.8 65.5 81.2 68.9 57.5 52.3 83.7 71.6 89.1 81.2 68.5 56.7 61.9 73.5 74.8 23.4 28.5
80.1 76.2 87.9 85.8 73.6 62.5 66.2 72.2 61.8 57.6 48.2 84.5 72.5 89.0 81.2 68.6 57.6 63.5 74.1 75.4 22.3 30.0
DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
days (SD¼206.5), and PCV dose 3 (mean 88.3, SD¼167.8). Substantive decreases in the degree of delay for both Hib and PCV doses at age 24 months were observed across all birth cohorts (Table 3). Children who received the HepB birth dose on time were less delayed for all vaccine doses (49.4, SD¼77.0, days for all doses; 33.5, SD¼71.4 days, excluding Hib doses) than children who did not receive the HepB birth dose on time (96.0, SD¼124.9, days for all doses; 83.4, SD¼124.3 days, excluding Hib doses). There were substantial improvements in vaccine timeliness in children who had received one or more combination vaccine doses compared with children who did not receive any combination vaccine doses. At age 24 months, children with one or more combination vaccine doses were delayed by an average of 51.6 (SD¼75.6) days for all vaccine doses (35.4, SD¼70.5 days, excluding Hib doses), whereas children who did not receive any combination vaccine doses were delayed by an average of 109.3 (SD¼150.6) days (100.9, SD¼147.9 days, excluding Hib doses).
DISCUSSION Children who receive vaccinations later than recommended are at unnecessary and prolonged risk for greater morbidity and mortality from VPDs and can contribute January 2018
to ongoing transmission, particularly among other young, susceptible children and among elderly household contacts.3,15,16 Nevertheless, a mere 13.2% of children in Michigan received all recommended vaccine doses on time during their first 24 months of life, with the sizeable remainder with delayed vaccinations at some point during the study period. Although there was a clear trend of increased vaccine timeliness across birth year cohorts, this was generally because of more timely vaccination with predominantly relatively newer vaccines such as PCV, whereas the dose timeliness for more established vaccines such as DTaP and IPV remained unchanged at approximately 80%. These findings are similar to those previously reported: Luman and colleagues17 found that timely receipt of all vaccine doses ranged from 2% in Mississippi to 26% in Massachusetts. Also similar to previous studies,3,4,18,19 this study shows high vaccine coverage of more than 90% for all vaccine doses except for the second, third, and fourth doses of Hib, and the fourth dose of PCV, but significantly lower timeliness of vaccine receipt. The findings from this and other studies highlight the potentially significant limitation associated with simply reporting overall vaccine coverage as the key indicator for VPD protection. Standard coverage measures can mask sometimes substantial periods of time during which children
100
Wagner et al / Am J Prev Med 2018;54(1):96–102
Table 3. Number of Days Undervaccinated at 24 Months in the Michigan Care Improvement Registry (n¼620,592) Vaccine Hepatitis B dose 1 Hepatitis B dose 2 Hepatitis B dose 3 DTaP dose 1 DTaP dose 2 DTaP dose 3 DTaP dose 4 Hib dose 1 Hib dose 2 Hib dose 3 Hib last dose IPV dose 1 IPV dose 2 IPV dose 3 PCV dose 1 PCV dose 2 PCV dose 3 PCV dose 4 MMR dose 1 Varicella dose 1 All doses All doses except Hib
All years
2006
2007
2008
2009
2010
28.4 ± 117.0 46.9 ± 143.0 17.4 ± 47.8 24.0 ± 105.5 43.8 ± 130.7 68.9 ± 149.7 40.6 ± 64.0 85.1 ± 193.2 149.9 ± 220.2 138.2 ± 206.5 135.1 ± 114.1 28.6 ± 114.6 49.5 ± 137.9 17.8 ± 48.0 35.7 ± 129.1 58.5 ± 151.2 88.3 ± 167.8 69.3 ± 103.4 42.2 ± 85.1 43.8 ± 87.1 59.2 ± 91.2 44.0 ± 87.6
27.4 ± 111.3 42.3 ± 134.1 16.6 ± 46.8 23.9 ± 107.2 42.1 ± 129.8 66.1 ± 148.3 40.3 ± 63.9 105.2 ± 198.6 215.2 ± 218.0 185.5 ± 227.9 169.0 ± 108.5 27.0 ± 112.2 46.5 ± 134.2 19.2 ± 49.6 40.4 ± 139.0 61.6 ± 156.9 91.1 ± 171.6 73.1 ± 105.2 39.2 ± 83.0 45.2 ± 88.3 64.4 ± 89.5 43.9 ± 87.5
29.7 ± 117.1 47.0 ± 142.1 18.4 ± 49.0 25.8 ± 111.1 45.8 ± 134.7 71.9 ± 153.3 42.1 ± 64.8 148.9 ± 243.7 267.4 ± 257.4 235.5 ± 242.0 205.7 ± 83.1 30.2 ± 118.6 51.7 ± 140.8 20.5 ± 51.0 38.1 ± 135.3 60.6 ± 155.6 90.3 ± 170.6 71.4 ± 104.5 44.3 ± 87.1 45.9 ± 89.0 75.1 ± 91.7 45.8 ± 90.3
30.1 ± 122.1 49.7 ± 147.5 18.2 ± 48.7 25.5 ± 108.6 45.6 ± 132.6 70.9 ± 151.0 41.9 ± 64.7 71.6 ± 185.2 98.9 ± 193.5 132.3 ± 197.5 119.4 ± 110.2 31.3 ± 118.9 52.9 ± 141.1 18.3 ± 48.7 36.1 ± 130.0 59.1 ± 151.7 88.5 ± 168.0 68.5 ± 102.8 44.7 ± 87.4 45.1 ± 88.3 57.3 ± 93.4 45.4 ± 89.5
27.7 ± 118.2 48.2 ± 145.8 17.0 ± 47.1 22.8 ± 100.7 42.8 ± 127.4 67.9 ± 147.4 39.3 ± 63.0 34.6 ± 127.0 57.0 ± 148.5 86.0 ± 164.7 77.8 ± 103.8 27.5 ± 111.7 48.0 ± 135.5 14.9 ± 44.4 32.3 ± 120.7 56.0 ± 145.9 87.1 ± 165.0 67.2 ± 102.0 41.6 ± 84.0 41.5 ± 84.8 46.8 ± 89.1 42.6 ± 85.4
27.0 ± 116.2 47.8 ± 146.0 16.7 ± 46.9 21.6 ± 98.0 42.6 ± 128.4 67.6 ± 148.2 39.4 ± 63.4 57.0 ± 162.9 94.0 ± 183.3 97.5 ± 180.3 92.5 ± 110.5 26.8 ± 110.8 48.3 ± 137.4 15.3 ± 45.2 30.8 ± 116.6 54.6 ± 144.1 83.7 ± 162.6 65.5 ± 102.2 40.9 ± 83.9 40.7 ± 84.5 50.0 ± 89.1 41.8 ± 84.9
DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
remain susceptible to these diseases because they are undervaccinated. Untimely and delayed vaccination were particularly common with the Hib series. This may have been due, at least in part, to a nationwide Hib vaccine shortage that began in December of 2007 with the recall of two Hib conjugate vaccines, and continued until the middle of 2009.20,21 The lowest coverage levels and greatest untimely receipt for the Hib series occurred in the 2007 birth cohort corresponding to the national Hib shortage. Following the resolution of the shortage, the proportion of children receiving timely Hib doses increased, although it remains below the levels of the DTaP, PCV, and IPV vaccines. Reasons for vaccine delay have been evaluated in several other studies.22–24 Luman and colleagues reported that factors associated with vaccine delay included more children in the household, mothers younger than age 30 years, use of public vaccine providers, and having multiple vaccine providers.22 In a survey on vaccine hesitancy in Utah, parents who had a child who was at least 6 months overdue for one or more vaccines reported confusion about the vaccine schedule and not knowing when to return, concerns about a child’s pain, crying, or anxiety during the immunization, and wanting the child to be older before receiving vaccines all as reasons for
vaccine delay.23 Additional reasons for vaccine delay include missed appointments, emergencies, safety concerns, and ill children.24 Missed opportunities for vaccination (i.e., children being at a vaccination provider but not receiving a vaccine that they were eligible to receive) also greatly contribute to vaccination delays.25 An analysis from the Oregon IIS found that the majority of children not up-to-date at age 24 to 35 months had previous encounters with providers.26 At an office visit, providers should be aware of all vaccines that the child is eligible to receive (e.g., through first accessing the IIS) and should use ethically acceptable practices like nudges to confidently inform parents about which vaccines their child should be receiving.27 Parental concerns about the number of vaccine injections a child receives in one visit is a common reason for vaccine delay,28 and therefore parents may prefer to use combination vaccines.29 A significantly shorter period of delayed vaccination was observed in children who received combination vaccines compared with children who did not. Similarly, in a study of children aged 2 years in the Georgia Medicaid program, children with three or more doses of the combination vaccine DTaP-HepB-IPV showed significantly shorter periods of delayed vaccination compared with children who did not (29.5 days vs 70.4 days of delay).30 Another www.ajpmonline.org
Wagner et al / Am J Prev Med 2018;54(1):96–102
study of children aged 2 years in the Georgia Medicaid program reported that children who received one dose of either Hib-HepB or DTaP-HepB-IPV had higher coverage of DTaP, HepB, and IPV compared with children who did not receive any combination vaccines.31 Using combination vaccines could result in improvements to vaccine timeliness by reducing the logistic difficulties of scheduling multiple visits for vaccine administration or possibly by assuaging parents’ concerns about immunizing children with multiple shots during a single visit. Children with timely vaccination of the HepB birth dose had a much shorter duration of delayed vaccination compared with children who did not receive the birth dose on time, which aligns with previous research.32 Early interventions aimed at educating the parents about the importance of vaccination at the time of the child’s birth by encouraging receipt of the birth dose of HepB may increase vaccine timeliness later in life. Future studies can investigate differences between parents who do or do not have their child receive the HepB vaccination at birth. Limiting delays in vaccination could mitigate economic and health consequences. A recent study by Curran et al.15 evaluated the implications of delayed pertussis vaccination in infants in the U.S. based on the 2010 National Immunization Survey. In a counterfactual model of strict adherence to the ACIP-recommended immunization schedule, 313 of 3,052 cases of pertussis, 112 hospitalizations and one death each year could be avoided by the timely receipt of all vaccine doses in all children, which amounts to a cost savings of $1.3 million annually.15
Limitations An important limitation was the lack of demographic information or attitudes toward vaccination in this data set, which would be important factors to consider when developing public health strategies to increase vaccine timeliness. Additionally, although few do, some families may opt out of including their child’s records in MCIR or may voluntarily refuse vaccination, and children who leave the state may still be included in the database even though their records are not being updated. Moreover, the authors’ interpretation of the association between combination vaccine use or HepB birth dose and delayed vaccination may be limited because delayed vaccination could be caused by other factors not evaluated, including vaccine shortages in some years. For example, certain types of providers, such as pediatricians or local health departments, have greater success in vaccinating children than other physicians, like family medicine.33 An important strength in this study is that the data set came from a comprehensive immunization surveillance system that January 2018
101
used only provider-verified data, minimizing chances for misclassification and incorrect vaccination status. The use of birth year cohorts also allowed the authors to evaluate whether vaccine timeliness and vaccine coverage are changing over time. Continued use of IISs to examine timeliness of measures is important because IISs already collect comprehensive data on vaccination dates, and may be less subject to response bias compared with studies, such as the National Immunization Survey, which has had increasing nonresponse rates over time.34
CONCLUSIONS The timeliness of vaccine administration can provide a more precise, albeit complementary, measure of vaccine coverage and protection from VPDs in children compared with standard vaccine coverage measures alone. In evaluating vaccine timeliness for a large group of Michigan children over a several-year period, few children had on-time administration of all recommended vaccines. Timely vaccination helps protect both the individual and others in the community by reducing the transmission of disease. Increasing timely vaccine receipt should be a greater public health priority given its critical role in protecting children from common but potentially serious VPDs.
ACKNOWLEDGMENTS The authors would like to acknowledge the healthcare workers who diligently submit vaccination records into the Michigan Care Improvement Registry. No funding support is reported for this project. ALW contributed to data analysis and interpretation, and reviewed and revised the manuscript for critical content. AME carried out the initial analyses and drafted the initial manuscript. RCP conceptualized and designed the study, and reviewed and revised the manuscript. RGS coordinates the surveillance system from which the data were drawn for this study and critically reviewed the manuscript. MLB contributed to data analysis and interpretation, and reviewed and revised the manuscript for critical content. All authors approved the final manuscript as submitted. No financial disclosures were reported by the authors of this paper. Dr. Boulton serves as the Editor-in-Chief for AJPM. He had no involvement in the peer review and decision-making processes for this paper.
REFERENCES 1. Kroger AT, Sumaya CV, Pickering LK, Atkinson WL. General recommendations on immunization; recommendations of the Advisory Committee on Immunizations Practices (ACIP). MMWR Morb Mortal Wkly Rep. 2011;60(2):3–61.
102
Wagner et al / Am J Prev Med 2018;54(1):96–102
2. Luman ET, Barker LE, Shaw KM, McCauley MM, Buehler JW, Pickering LK. Timeliness of childhood vaccinations in the United States: days undervaccinated and number of vaccines delayed. JAMA. 2005;293(10):1204–1211. https://doi.org/10.1001/jama.293.10.1204. 3. Glanz JM, Newcomer SR, Narwaney KJ, et al. A population-based cohort study of undervaccination in 8 managed care organizations across the United States. JAMA Pediatr. 2013;167(3):274–281. https: //doi.org/10.1001/jamapediatrics.2013.502. 4. McCarthy NL, Irving S, Donahue JG, et al. Vaccination coverage levels among children enrolled in the vaccine safety datalink. Vaccine. 2013;31(49):5822–5826. https://doi.org/10.1016/j.vaccine.2013.10.011. 5. Kurosky SK, Davis KL, Krishnarajah G. Completion and compliance of childhood vaccinations in the United States. Vaccine. 2016;34(3):387– 394. https://doi.org/10.1016/j.vaccine.2015.11.011. 6. Phadke VK, Bednarczyk RA, Salmon DA, Omer SB. Association between vaccine refusal and vaccine-preventable diseases in the United States. JAMA. 2016;315(11):1149–1158. https://doi.org/10.1001/jama.2016.1353. 7. Robison SG, Liko J. The timing of pertussis cases in unvaccinated children in an outbreak year: Oregon 2012. J Pediatr. 2017;183:159– 163. https://doi.org/10.1016/j.jpeds.2016.12.047. 8. Pelton SI, Bernadi PG, Block SL, Marshall GS, Middleton DB, Stevenson AM. Why Vaccination Timeliness Matters. Sanofi Pasteur Inc., 2011. 9. Michigan Care Improvement Registry. Law, administrative code and public acts related to MCIR. www.mcir.org/resource/laws/. Accessed September 12, 2017. 10. Centers for Disease Control and Prevention (CDC). Q & A about IIS sentinel sites. www.cdc.gov/vaccines/programs/iis/activities/sentinelsites.html. Published 2016. Accessed May 10, 2017. 11. Department of Community Health Bureau of Epidemiology. Communicable and related diseases. http://w3.lara.state.mi.us/orr/Files/Admin Code/1472_2014-073CH_AdminCode.pdf. Published 2014. Accessed July 27, 2017. 12. Michigan Department of Health and Human Services. Vaccines required for child care and preschool in Michigan. www.michigan. gov/documents/mdhhs/Child_Care_Reqs_parents_553545_7.pdf. Published 2017. Accessed August 9, 2017. 13. National Committee for Quality Assurance. Childhood immunization status. HEDIS measures. www.ncqa.org/portals/0/Childhood%20Immu nization%20Status.pdf. Published 2009. Accessed May 22, 2017. 14. Glauber JH. The immunization delivery effectiveness assessment score: a better immunization measure? Pediatrics. 2003;112(1):e39–e45. 15. Curran D, Terlinden A, Poirrier JE, Masseria C, Krishnarajah G. Vaccine timeliness: a cost analysis of the implications of delayed vaccination. Value Health. 2013;16(7):A346. https://doi.org/10.1016/j. jval.2013.08.143. 16. Winter K, Harriman K, Zipprich J, Schechter R, Talarico J. California 2010 Pertussis Epidemic [Abstract 2526]. 45th National Immunization Conference. Washington, DC: Public Health Foundation, 2011. 17. Luman ET, Barker LE, McCauley MM, Drews-Botsch C. Timeliness of childhood immunizations: a state-specific analysis. Am J Epidemiol. 2005;95(8):1367–1374. https://doi.org/10.2105/AJPH.2004.046284. 18. Lernout T, Theeten H, Hens N, et al. Timeliness of infant vaccination and factors related with delay in Flanders, Belgium. Vaccine. 2014;32 (2):284–289. https://doi.org/10.1016/j.vaccine.2013.10.084. 19. Fadnes LT, Nankabirwa V, Sommerfelt H, Tylleskär T, Tumwine JK, Engebretsen IMS. Is vaccination coverage a good indicator of
20.
21.
22.
23.
24.
25.
26.
27. 28.
29.
30.
31.
32.
33.
34.
age-appropriate vaccination? A prospective study from Uganda. Vaccine. 2011;29(19):3564–3570. https://doi.org/10.1016/j.vaccine.2011. 02.093. Centers for Disease Control and Prevention (CDC). Continued shortage of Haemophilus influenzae type b (Hib) conjugate vaccines and potential implications for Hib surveillance—United States, 2008. MMWR Morb Mortal Wkly Rep. 2008;57(46):1252–1255. Chamberlain AT, Wells K, Seib K, et al. Lessons learned from the 2007 to 2009 Haemophilus influenzae type b vaccine shortage: implications for future vaccine shortages and public health preparedness. J Public Health Manag Pract. 2012;18(3):E9–E16. https://doi.org/10.1097/ PHH.0b013e31821dce27. Luman ET, McCauley MM, Stokley S, Chu SY, Pickering LK. Timeliness of childhood immunizations. Pediatrics. 2002;110(5):935–939. https://doi.org/10.1542/peds.110.5.935. Luthy KE, Beckstrand RL, Peterson NE. Parental hesitation as a factor in delayed childhood immunization. J Pediatr Health Care. 2009;23 (6):388–393. https://doi.org/10.1016/j.pedhc.2008.09.006. Smith PJ, Humiston SG, Vannice KS. The association between intentional delay of vaccine administration and timely childhood vaccination coverage. Public Health Rep. 2010;125(4):534–541. https://doi.org/ 10.1177/003335491012500408. Fu LY, Zook K, Gingold J, et al. Frequent vaccination missed opportunities at primary care encounters contribute to underimmunization. J Pediatr. 2015;166(2):412–417. https://doi.org/10.1016/j. jpeds.2014.10.066. Robison SG, Kurosky SK, Young CM, Gallia CA, Arbor SA. Immunization milestones: a more comprehensive picture of age-appropriate vaccination. J Biomed Biotechnol. 2010;2010:916525. https://doi.org/ 10.1155/2010/916525. Navin MC. The ethics of vaccination nudges in pediatric practice. HEC Forum. 2017;29(1):43–57. https://doi.org/10.1007/s10730-016-9311-2. Offit PA, Quarles J, Gerber MA, et al. Addressing parents’ concerns: do multiple vaccines overwhelm or weaken the infant’s immune system? Pediatrics. 2002;109(1):124–129. https://doi.org/10.1542/peds.109.1.124. Gidengil C, Lieu TA, Payne K, Rusinak D, Messonnier M, Prosser LA. Parental and societal values for the risks and benefits of childhood combination vaccines. Vaccine. 2012;30(23):3445–3452. https://doi. org/10.1016/j.vaccine.2012.03.022. Happe LE, Lunacsek OE, Kruzikas DT, Marshall GS. Impact of a pentavalent combination vaccine on immunization timeliness in a state Medicaid population. Pediatr Infect Dis J. 2009;28(2):98–101. https://doi.org/10.1097/INF.0b013e318187d047. Marshall GS, Happe LE, Lunacsek OE, et al. Use of combination vaccines is associated with improved coverage rates. Pediatr Infect Dis J. 2007;26(6):496–500. https://doi.org/10.1097/INF.0b013e31805d7f17. Lauderdale DS, Oram RJ, Goldstein KP, Daum RS. Hepatitis B vaccination among children in inner-city public housing, 1991–1997. JAMA. 1999;282(18):1725–1730. https://doi.org/10.1001/jama.282.18.1725. Potter RC, DeVita SF, Vranesich PA, Boulton ML. Adolescent immunization coverage and implementation of new school requirements in Michigan, 2010. Am J Public Health. 2014;104(8):1526–1533. https://doi.org/10.2105/AJPH.2014.301910. The National Research Council. The growing problem of nonresponse. In: Tourangeau R, Plewes TJ, eds. Nonresponse in Social Science Surveys: A Research Agenda. Washington, DC: National Academies Press, 2013:7–39. https://doi.org/10.17226/18293.
www.ajpmonline.org
Vaccination Timeliness at Age 24 Months in Michigan Children Born 2006–2010 Abram L. Wagner, PhD, MPH,1 Amanda M. Eccleston, MPH,1 Rachel C. Potter, DVM, MS,2 Robert G. Swanson, MPH,2 Matthew L. Boulton, MD, MPH1,3 Introduction: Delays in vaccination can stymie the development of herd immunity, and a large proportion of children in the U.S. are known not to receive vaccines on time. This study quantifies delays in vaccination, compares vaccination timeliness to the proportion of children vaccinated, and evaluates the impact of combination vaccine use and timely administration of hepatitis B vaccine birth dose on vaccine timeliness among Michigan children. Methods: This retrospective cohort study used data from the Michigan Care Improvement Registry
—the state immunization information system—for children born 2006–2010. Children aged 24 months as of December 31, 2012, were included. The proportion of children with timely administration of vaccine doses was calculated, and the mean days of vaccination delay with SD were reported.
Results: Among 620,592 Michigan children, 42.2% had received all vaccines, but only 13.2% were vaccinated on time by age 24 months. Children’s vaccinations were delayed an average of 59.2 (SD¼91.2) days by age 24 months for all recommended vaccine doses. Children who received a timely hepatitis B vaccine birth dose or who received a combination vaccine had less delay in vaccination.
Conclusions: Michigan children have high vaccination coverage based on standard measures but few receive these vaccines on time. Promoting use of combination vaccines may improve parental compliance with timely vaccination of children. Am J Prev Med 2018;54(1):96–102. & 2017 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.
INTRODUCTION
I
n the U.S., the Advisory Committee on Immunization Practices (ACIP) produces a recommended immunization schedule, which provides information on the age at which each dose should be administered and the allowable interval between doses; this schedule is approved by professional societies, such as the American Academy of Pediatrics.1 Vaccination programs have traditionally been assessed through measurements of vaccine coverage, the proportion of children who have received a vaccine without any consideration of the timeliness.2 Analyses of the Vaccine Safety Datalink, a consortium of ten healthcare delivery organizations, show that although overall vaccination coverage is generally high (490%) in the U.S., timely vaccination is not3,4; less than 50% of children born 96 Am J Prev Med 2018;54(1):96–102
between 2004 and 2008 within the Vaccine Safety Datalink had received all vaccine doses on time.3 Nationwide studies have also found a high rate of noncompliance with the pediatric immunization schedule; according to the 2012 National Immunization Survey, only 26% of children in the U.S. had received all vaccines at the recommended ages.5
From the 1Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, Michigan; 2Division of Immunizations, Michigan Department of Health and Human Services, Lansing, Michigan; and 3Department of Internal Medicine, Division of Infectious Diseases, University of Michigan Medical School, Ann Arbor, Michigan Address correspondence to: Abram L. Wagner, PhD, MPH, 1415 Washington Heights, Ann Arbor MI 48109. E-mail: [email protected]. 0749-3797/$36.00 https://doi.org/10.1016/j.amepre.2017.09.014
& 2017 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.
Wagner et al / Am J Prev Med 2018;54(1):96–102
Delaying vaccine receipt beyond ACIP-recommended ages places children at risk for potentially serious vaccine-preventable diseases (VPDs). A review of recent pertussis outbreaks have found that a large proportion of cases have been undervaccinated.6 During an outbreak of pertussis in 2012 in Oregon, undervaccinated infants had a quicker onset of disease compared with fully vaccinated infants and therefore potentiated an earlier spread of disease.7 Lack of vaccine timeliness extends beyond individual risk by delaying the establishment of herd immunity and the protection of infants too young to receive vaccinations or children contraindicated from receiving vaccines.8 Vaccine timeliness is an important area of research because of the potential health consequences of untimely vaccination and because it is a more precise methodology for quantifying immunization system performance. Using data from children born between 2006 and 2010 in Michigan’s immunization information system (IIS), this paper examines trends in vaccine timeliness, assesses vaccine delay, compares vaccine timeliness with the proportion of children vaccinated at 24 months, and evaluates factors associated with increased vaccine timeliness.
METHODS This retrospective cohort study used data obtained from the Michigan Care Improvement Registry (MCIR), Michigan’s IIS, which were analyzed in 2015. This comprehensive immunization database was created in 1998 and by law, all vaccinations administered to children aged o20 years must be reported.9 The database is populated by birth data from the state’s electronic birth certificate system. Children with and without private insurance are included in the system, and providers may create new records for children not born in Michigan. MCIR is one of six sentinel IISs in the U.S., selected on account of having ≥85% of providers actively using the IIS, ≥85% of children participating in the IIS, and ≥70% of vaccine doses submitted to the IIS within 30 days.10 The Michigan Department of Health and Human Services regularly assesses providers through the assessment, feedback, and incentives exchange program and that vaccinations reported to the IIS are the standard for this assessment. This study includes the immunization records from all Michigan children born January 1, 2006, through December 31, 2010. Children were only included if they had received one or more vaccine dose to exclude children who had moved out of state. Vaccine doses were valid if given within the ACIP-recommended age range. Doses were excluded if given before the ACIPrecommended age or if violating the minimum recommended interval between doses of one vaccine series. Vaccine doses were considered valid if they were given in the 4 days prior to the beginning of the ACIP-recommended age range or vaccine interval. To study biases resulting from children moving out of state, a sensitivity analysis was performed, limited to children who had January 2018
97
been vaccinated after their first birthday. This analysis gave substantively similar findings (results not shown). The vaccines assessed include four doses of the diphtheria, tetanus, acellular pertussis vaccine (DTaP); three doses of the inactivated polio vaccine (IPV); one dose of the measles, mumps, rubella vaccine; three (or four) doses of the Haemophilus influenzae type b vaccine (Hib); three doses of the hepatitis B (HepB) vaccine; one dose of the varicella vaccine; and four doses of the pneumococcal conjugate vaccine (PCV). Together, these vaccines make up the 4:3:1:3:3:1:4 series that are recommended before age 2 years. These vaccine series are required before enrolling at a school or licensed day care (although Hib and PCV are not required after 5 years).11,12 Rotavirus was not included because it should not be administered after age 8 months and is not required for school entry. The number of doses required for the Hib vaccine series depends on the vaccine manufacturer. Three Hib doses were assessed when the PRP-OMP vaccine was used for every dose; otherwise, a four-dose Hib schedule was assessed. The last dose of both schedules is a booster dose recommended at age 12–15 months; therefore, calculations regarding the final recommended vaccine dose in both schedules are marked as the last dose, whereas Hib “dose 3” only includes children on a four-dose Hib schedule. There was a nationwide shortage of Hib vaccine between December 2007 and July 2009, and so its requirement before school entry was suspended during this period. For this reason, summary statistics are calculated with and without Hib doses included. The outcomes of interest include a standard measure of the proportion of children vaccinated, in addition to vaccine timeliness and delayed vaccination. All measures were assessed when children were aged 24 months (i.e., the child’s second birthday), which corresponds to a common Healthcare Effectiveness Data and Information Set measure.13 A timely dose was administered within 1 month (ranging from 28 to 31 days depending on the length of the month) of the ACIP-recommended age range at vaccination. Delay in vaccination was calculated as the difference, in days, between actual vaccine dose administration and when the vaccine dose is recommended for administration based on the ACIP-recommended acceptable time range (up to 24 months). For example, the recommended age of administration for the first dose of DTaP is 2 months. Children were delayed if they had not yet received the first dose of DTaP at age 3 months. The average number of days delayed by age 24 months was calculated in a method similar to that used by Glanz et al.3 or Glauber.14 This value summed the number of days delayed for each vaccine dose by age 24 months, and divided this total number by the total number of vaccine doses that a child should have received according to the ACIP-recommended immunization schedule. Factors associated with vaccine timeliness were assessed, including the receipt of any combination vaccine doses. Delayed vaccination was evaluated in children who had received one or more combination vaccine doses compared with children who did not receive any combination vaccine doses. The receipt of the HepB birth dose on time was also evaluated as an early predictor of later vaccine timeliness. The HepB birth dose was considered on time if it was given within 3 days of birth, with the day of birth counted as day 0. The average number of days of delayed vaccinations by age 24 months was compared between children who had and had not received the HepB birth dose on time. All analyses were conducted using SAS, version 9.3, software.
98
Wagner et al / Am J Prev Med 2018;54(1):96–102
Table 1. Vaccine Coverage (%) at 24 Months in the Michigan Care Improvement Registry (n¼620,592) Vaccine Hepatitis B dose 1 Hepatitis B dose 2 Hepatitis B dose 3 DTaP dose 1 DTaP dose 2 DTaP dose 3 DTaP dose 4 Hib dose 1 Hib dose 2 Hib dose 3 Hib last dose IPV dose 1 IPV dose 2 IPV dose 3 PCV dose 1 PCV dose 2 PCV dose 3 PCV dose 4 MMR dose 1 Varicella dose 1 All doses All doses except Hib
All years
2006
2007
2008
2009
2010
97.7 95.4 89.3 97.7 95.6 92.4 77.8 90.0 83.1 80.0 50.1 97.2 95.0 89.3 96.4 93.7 89.5 76.3 87.2 86.3 42.2 65.9
97.9 96.0 89.9 97.5 95.6 92.6 78.1 88.9 80.8 70.9 33.6 97.2 95.2 88.5 95.7 93.1 88.8 75.0 88.1 85.9 26.6 64.5
97.7 95.4 88.6 97.4 95.2 92.0 76.8 81.0 64.4 60.4 19.0 96.9 94.7 87.6 95.9 93.2 88.9 75.4 86.2 85.4 14.9 63.4
97.5 95.0 88.8 97.6 95.4 92.2 76.9 91.2 88.0 82.7 59.8 97.0 94.7 88.8 96.4 93.6 89.3 76.9 86.1 85.7 50.4 65.5
97.7 95.2 89.7 98.0 95.9 92.9 78.9 96.5 93.9 90.0 74.9 97.4 95.3 91.0 97.0 94.4 90.2 77.3 87.8 87.4 64.5 68.2
97.8 95.2 89.9 98.2 95.8 92.6 78.7 93.6 90.4 86.7 68.0 97.5 95.1 90.7 97.3 94.5 90.4 77.4 87.8 87.5 58.8 68.5
DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
This study was reviewed by the Michigan Department of Health and Human Services IRB and determined to be exempt from human subjects research.
RESULTS A total of 636,998 Michigan children born 2006–2010 were selected into the study. After removing children who did not receive any valid vaccine doses during the study period (n¼16,406), the final study population contained 620,592 Michigan children. Overall, 42.2% of children had been vaccinated with the entire 4:3:1:3:3:1:4 schedule by age 24 months; excluding Hib doses, this figure was 65.9%. At age 36 months, 49.8% were vaccinated with all doses, and at age 48 months, 52.8% were (excluding Hib doses, these figures were 74.0% and 77.0%, respectively). At age 24 months, the proportion of children immunized varied widely by dose and vaccine, from 50.1% for the last dose of Hib to 97.7% for DTaP dose 1 (Table 1). Among these doses of vaccines whose overall coverage was o90%, coverage held steady for HepB dose 3, DTaP dose 4, IPV dose 3, PCV dose 4, and for the measles, mumps, rubella and varicella doses, whereas it increased between 2006 and 2010 for the Hib doses. Figure 1 depicts vaccination coverage as with increasing age, for children born in 2010. In contrast to the high proportion of children vaccinated at age 24 months, only 13.2% of all the children received all
4:3:1:3:3:1:4 doses (27.1%, if excluding Hib doses). Table 2 shows the timeliness of vaccine doses over the study period. Timeliness ranged from a low of 32.9% for Hib last dose to a high of 87.5% for HepB dose 3. Timeliness was increased across birth cohorts; excluding Hib doses, it moved from 12.6% to 30.0% between 2006 and 2010. By age 24 months, children had delayed each recommended vaccine for an average of 59.2 days (SD¼91.2). Across all birth cohorts, the vaccine doses characterized by the most prolonged periods of delay were all Hib doses, particularly Hib dose 2 with a mean delay of 149.9
Figure 1. Vaccine coverage in Michigan by age up to 24 months, for children born in 2010 (n¼114,051). DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
www.ajpmonline.org
99
Wagner et al / Am J Prev Med 2018;54(1):96–102
Table 2. Vaccine Timeliness (%) at 24 Months in the Michigan Care Improvement Registry (n¼620,592) Vaccine Hepatitis B dose 1 Hepatitis B dose 2 Hepatitis B dose 3 DTaP dose 1 DTaP dose 2 DTaP dose 3 DTaP dose 4 Hib dose 1 Hib dose 2 Hib dose 3 Hib last dose IPV dose 1 IPV dose 2 IPV dose 3 PCV dose 1 PCV dose 2 PCV dose 3 PCV dose 4 MMR dose 1 Varicella dose 1 All doses All doses except Hib
All years
2006
2007
2008
2009
2010
79.0 74.6 87.5 85.5 73.4 62.2 65.1 58.5 49.5 50.0 32.9 82.8 70.4 87.0 81.2 68.8 57.2 61.4 73.8 74.0 13.2 27.1
77.9 74.5 88.1 86.4 74.8 63.9 65.3 35.5 30.6 44.2 23.8 82.9 70.9 85.6 81.2 69.3 58.0 59.4 75.5 73.1 6.3 12.6
77.7 73.3 86.9 85.5 73.3 61.8 64.2 35.7 30.3 37.2 10.6 81.8 69.2 84.9 81.6 69.2 57.4 60.6 73.0 73.1 2.8 12.6
79.2 73.8 87.0 84.8 72.3 60.8 64.5 72.3 59.4 47.9 33.1 81.1 68.3 86.7 80.9 68.1 56.5 61.8 72.8 73.5 13.2 16.2
80.5 75.6 87.7 85.0 72.9 61.8 65.5 81.2 68.9 57.5 52.3 83.7 71.6 89.1 81.2 68.5 56.7 61.9 73.5 74.8 23.4 28.5
80.1 76.2 87.9 85.8 73.6 62.5 66.2 72.2 61.8 57.6 48.2 84.5 72.5 89.0 81.2 68.6 57.6 63.5 74.1 75.4 22.3 30.0
DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
days (SD¼206.5), and PCV dose 3 (mean 88.3, SD¼167.8). Substantive decreases in the degree of delay for both Hib and PCV doses at age 24 months were observed across all birth cohorts (Table 3). Children who received the HepB birth dose on time were less delayed for all vaccine doses (49.4, SD¼77.0, days for all doses; 33.5, SD¼71.4 days, excluding Hib doses) than children who did not receive the HepB birth dose on time (96.0, SD¼124.9, days for all doses; 83.4, SD¼124.3 days, excluding Hib doses). There were substantial improvements in vaccine timeliness in children who had received one or more combination vaccine doses compared with children who did not receive any combination vaccine doses. At age 24 months, children with one or more combination vaccine doses were delayed by an average of 51.6 (SD¼75.6) days for all vaccine doses (35.4, SD¼70.5 days, excluding Hib doses), whereas children who did not receive any combination vaccine doses were delayed by an average of 109.3 (SD¼150.6) days (100.9, SD¼147.9 days, excluding Hib doses).
DISCUSSION Children who receive vaccinations later than recommended are at unnecessary and prolonged risk for greater morbidity and mortality from VPDs and can contribute January 2018
to ongoing transmission, particularly among other young, susceptible children and among elderly household contacts.3,15,16 Nevertheless, a mere 13.2% of children in Michigan received all recommended vaccine doses on time during their first 24 months of life, with the sizeable remainder with delayed vaccinations at some point during the study period. Although there was a clear trend of increased vaccine timeliness across birth year cohorts, this was generally because of more timely vaccination with predominantly relatively newer vaccines such as PCV, whereas the dose timeliness for more established vaccines such as DTaP and IPV remained unchanged at approximately 80%. These findings are similar to those previously reported: Luman and colleagues17 found that timely receipt of all vaccine doses ranged from 2% in Mississippi to 26% in Massachusetts. Also similar to previous studies,3,4,18,19 this study shows high vaccine coverage of more than 90% for all vaccine doses except for the second, third, and fourth doses of Hib, and the fourth dose of PCV, but significantly lower timeliness of vaccine receipt. The findings from this and other studies highlight the potentially significant limitation associated with simply reporting overall vaccine coverage as the key indicator for VPD protection. Standard coverage measures can mask sometimes substantial periods of time during which children
100
Wagner et al / Am J Prev Med 2018;54(1):96–102
Table 3. Number of Days Undervaccinated at 24 Months in the Michigan Care Improvement Registry (n¼620,592) Vaccine Hepatitis B dose 1 Hepatitis B dose 2 Hepatitis B dose 3 DTaP dose 1 DTaP dose 2 DTaP dose 3 DTaP dose 4 Hib dose 1 Hib dose 2 Hib dose 3 Hib last dose IPV dose 1 IPV dose 2 IPV dose 3 PCV dose 1 PCV dose 2 PCV dose 3 PCV dose 4 MMR dose 1 Varicella dose 1 All doses All doses except Hib
All years
2006
2007
2008
2009
2010
28.4 ± 117.0 46.9 ± 143.0 17.4 ± 47.8 24.0 ± 105.5 43.8 ± 130.7 68.9 ± 149.7 40.6 ± 64.0 85.1 ± 193.2 149.9 ± 220.2 138.2 ± 206.5 135.1 ± 114.1 28.6 ± 114.6 49.5 ± 137.9 17.8 ± 48.0 35.7 ± 129.1 58.5 ± 151.2 88.3 ± 167.8 69.3 ± 103.4 42.2 ± 85.1 43.8 ± 87.1 59.2 ± 91.2 44.0 ± 87.6
27.4 ± 111.3 42.3 ± 134.1 16.6 ± 46.8 23.9 ± 107.2 42.1 ± 129.8 66.1 ± 148.3 40.3 ± 63.9 105.2 ± 198.6 215.2 ± 218.0 185.5 ± 227.9 169.0 ± 108.5 27.0 ± 112.2 46.5 ± 134.2 19.2 ± 49.6 40.4 ± 139.0 61.6 ± 156.9 91.1 ± 171.6 73.1 ± 105.2 39.2 ± 83.0 45.2 ± 88.3 64.4 ± 89.5 43.9 ± 87.5
29.7 ± 117.1 47.0 ± 142.1 18.4 ± 49.0 25.8 ± 111.1 45.8 ± 134.7 71.9 ± 153.3 42.1 ± 64.8 148.9 ± 243.7 267.4 ± 257.4 235.5 ± 242.0 205.7 ± 83.1 30.2 ± 118.6 51.7 ± 140.8 20.5 ± 51.0 38.1 ± 135.3 60.6 ± 155.6 90.3 ± 170.6 71.4 ± 104.5 44.3 ± 87.1 45.9 ± 89.0 75.1 ± 91.7 45.8 ± 90.3
30.1 ± 122.1 49.7 ± 147.5 18.2 ± 48.7 25.5 ± 108.6 45.6 ± 132.6 70.9 ± 151.0 41.9 ± 64.7 71.6 ± 185.2 98.9 ± 193.5 132.3 ± 197.5 119.4 ± 110.2 31.3 ± 118.9 52.9 ± 141.1 18.3 ± 48.7 36.1 ± 130.0 59.1 ± 151.7 88.5 ± 168.0 68.5 ± 102.8 44.7 ± 87.4 45.1 ± 88.3 57.3 ± 93.4 45.4 ± 89.5
27.7 ± 118.2 48.2 ± 145.8 17.0 ± 47.1 22.8 ± 100.7 42.8 ± 127.4 67.9 ± 147.4 39.3 ± 63.0 34.6 ± 127.0 57.0 ± 148.5 86.0 ± 164.7 77.8 ± 103.8 27.5 ± 111.7 48.0 ± 135.5 14.9 ± 44.4 32.3 ± 120.7 56.0 ± 145.9 87.1 ± 165.0 67.2 ± 102.0 41.6 ± 84.0 41.5 ± 84.8 46.8 ± 89.1 42.6 ± 85.4
27.0 ± 116.2 47.8 ± 146.0 16.7 ± 46.9 21.6 ± 98.0 42.6 ± 128.4 67.6 ± 148.2 39.4 ± 63.4 57.0 ± 162.9 94.0 ± 183.3 97.5 ± 180.3 92.5 ± 110.5 26.8 ± 110.8 48.3 ± 137.4 15.3 ± 45.2 30.8 ± 116.6 54.6 ± 144.1 83.7 ± 162.6 65.5 ± 102.2 40.9 ± 83.9 40.7 ± 84.5 50.0 ± 89.1 41.8 ± 84.9
DTaP, diphtheria-tetanus-pertussis vaccine; Hib, Haemophilus influenzae type b vaccine; IPV, inactivated polio vaccine; MMR, measles, mumps, rubella vaccine; PCV, pneumococcal conjugate vaccine.
remain susceptible to these diseases because they are undervaccinated. Untimely and delayed vaccination were particularly common with the Hib series. This may have been due, at least in part, to a nationwide Hib vaccine shortage that began in December of 2007 with the recall of two Hib conjugate vaccines, and continued until the middle of 2009.20,21 The lowest coverage levels and greatest untimely receipt for the Hib series occurred in the 2007 birth cohort corresponding to the national Hib shortage. Following the resolution of the shortage, the proportion of children receiving timely Hib doses increased, although it remains below the levels of the DTaP, PCV, and IPV vaccines. Reasons for vaccine delay have been evaluated in several other studies.22–24 Luman and colleagues reported that factors associated with vaccine delay included more children in the household, mothers younger than age 30 years, use of public vaccine providers, and having multiple vaccine providers.22 In a survey on vaccine hesitancy in Utah, parents who had a child who was at least 6 months overdue for one or more vaccines reported confusion about the vaccine schedule and not knowing when to return, concerns about a child’s pain, crying, or anxiety during the immunization, and wanting the child to be older before receiving vaccines all as reasons for
vaccine delay.23 Additional reasons for vaccine delay include missed appointments, emergencies, safety concerns, and ill children.24 Missed opportunities for vaccination (i.e., children being at a vaccination provider but not receiving a vaccine that they were eligible to receive) also greatly contribute to vaccination delays.25 An analysis from the Oregon IIS found that the majority of children not up-to-date at age 24 to 35 months had previous encounters with providers.26 At an office visit, providers should be aware of all vaccines that the child is eligible to receive (e.g., through first accessing the IIS) and should use ethically acceptable practices like nudges to confidently inform parents about which vaccines their child should be receiving.27 Parental concerns about the number of vaccine injections a child receives in one visit is a common reason for vaccine delay,28 and therefore parents may prefer to use combination vaccines.29 A significantly shorter period of delayed vaccination was observed in children who received combination vaccines compared with children who did not. Similarly, in a study of children aged 2 years in the Georgia Medicaid program, children with three or more doses of the combination vaccine DTaP-HepB-IPV showed significantly shorter periods of delayed vaccination compared with children who did not (29.5 days vs 70.4 days of delay).30 Another www.ajpmonline.org
Wagner et al / Am J Prev Med 2018;54(1):96–102
study of children aged 2 years in the Georgia Medicaid program reported that children who received one dose of either Hib-HepB or DTaP-HepB-IPV had higher coverage of DTaP, HepB, and IPV compared with children who did not receive any combination vaccines.31 Using combination vaccines could result in improvements to vaccine timeliness by reducing the logistic difficulties of scheduling multiple visits for vaccine administration or possibly by assuaging parents’ concerns about immunizing children with multiple shots during a single visit. Children with timely vaccination of the HepB birth dose had a much shorter duration of delayed vaccination compared with children who did not receive the birth dose on time, which aligns with previous research.32 Early interventions aimed at educating the parents about the importance of vaccination at the time of the child’s birth by encouraging receipt of the birth dose of HepB may increase vaccine timeliness later in life. Future studies can investigate differences between parents who do or do not have their child receive the HepB vaccination at birth. Limiting delays in vaccination could mitigate economic and health consequences. A recent study by Curran et al.15 evaluated the implications of delayed pertussis vaccination in infants in the U.S. based on the 2010 National Immunization Survey. In a counterfactual model of strict adherence to the ACIP-recommended immunization schedule, 313 of 3,052 cases of pertussis, 112 hospitalizations and one death each year could be avoided by the timely receipt of all vaccine doses in all children, which amounts to a cost savings of $1.3 million annually.15
Limitations An important limitation was the lack of demographic information or attitudes toward vaccination in this data set, which would be important factors to consider when developing public health strategies to increase vaccine timeliness. Additionally, although few do, some families may opt out of including their child’s records in MCIR or may voluntarily refuse vaccination, and children who leave the state may still be included in the database even though their records are not being updated. Moreover, the authors’ interpretation of the association between combination vaccine use or HepB birth dose and delayed vaccination may be limited because delayed vaccination could be caused by other factors not evaluated, including vaccine shortages in some years. For example, certain types of providers, such as pediatricians or local health departments, have greater success in vaccinating children than other physicians, like family medicine.33 An important strength in this study is that the data set came from a comprehensive immunization surveillance system that January 2018
101
used only provider-verified data, minimizing chances for misclassification and incorrect vaccination status. The use of birth year cohorts also allowed the authors to evaluate whether vaccine timeliness and vaccine coverage are changing over time. Continued use of IISs to examine timeliness of measures is important because IISs already collect comprehensive data on vaccination dates, and may be less subject to response bias compared with studies, such as the National Immunization Survey, which has had increasing nonresponse rates over time.34
CONCLUSIONS The timeliness of vaccine administration can provide a more precise, albeit complementary, measure of vaccine coverage and protection from VPDs in children compared with standard vaccine coverage measures alone. In evaluating vaccine timeliness for a large group of Michigan children over a several-year period, few children had on-time administration of all recommended vaccines. Timely vaccination helps protect both the individual and others in the community by reducing the transmission of disease. Increasing timely vaccine receipt should be a greater public health priority given its critical role in protecting children from common but potentially serious VPDs.
ACKNOWLEDGMENTS The authors would like to acknowledge the healthcare workers who diligently submit vaccination records into the Michigan Care Improvement Registry. No funding support is reported for this project. ALW contributed to data analysis and interpretation, and reviewed and revised the manuscript for critical content. AME carried out the initial analyses and drafted the initial manuscript. RCP conceptualized and designed the study, and reviewed and revised the manuscript. RGS coordinates the surveillance system from which the data were drawn for this study and critically reviewed the manuscript. MLB contributed to data analysis and interpretation, and reviewed and revised the manuscript for critical content. All authors approved the final manuscript as submitted. No financial disclosures were reported by the authors of this paper. Dr. Boulton serves as the Editor-in-Chief for AJPM. He had no involvement in the peer review and decision-making processes for this paper.
REFERENCES 1. Kroger AT, Sumaya CV, Pickering LK, Atkinson WL. General recommendations on immunization; recommendations of the Advisory Committee on Immunizations Practices (ACIP). MMWR Morb Mortal Wkly Rep. 2011;60(2):3–61.
102
Wagner et al / Am J Prev Med 2018;54(1):96–102
2. Luman ET, Barker LE, Shaw KM, McCauley MM, Buehler JW, Pickering LK. Timeliness of childhood vaccinations in the United States: days undervaccinated and number of vaccines delayed. JAMA. 2005;293(10):1204–1211. https://doi.org/10.1001/jama.293.10.1204. 3. Glanz JM, Newcomer SR, Narwaney KJ, et al. A population-based cohort study of undervaccination in 8 managed care organizations across the United States. JAMA Pediatr. 2013;167(3):274–281. https: //doi.org/10.1001/jamapediatrics.2013.502. 4. McCarthy NL, Irving S, Donahue JG, et al. Vaccination coverage levels among children enrolled in the vaccine safety datalink. Vaccine. 2013;31(49):5822–5826. https://doi.org/10.1016/j.vaccine.2013.10.011. 5. Kurosky SK, Davis KL, Krishnarajah G. Completion and compliance of childhood vaccinations in the United States. Vaccine. 2016;34(3):387– 394. https://doi.org/10.1016/j.vaccine.2015.11.011. 6. Phadke VK, Bednarczyk RA, Salmon DA, Omer SB. Association between vaccine refusal and vaccine-preventable diseases in the United States. JAMA. 2016;315(11):1149–1158. https://doi.org/10.1001/jama.2016.1353. 7. Robison SG, Liko J. The timing of pertussis cases in unvaccinated children in an outbreak year: Oregon 2012. J Pediatr. 2017;183:159– 163. https://doi.org/10.1016/j.jpeds.2016.12.047. 8. Pelton SI, Bernadi PG, Block SL, Marshall GS, Middleton DB, Stevenson AM. Why Vaccination Timeliness Matters. Sanofi Pasteur Inc., 2011. 9. Michigan Care Improvement Registry. Law, administrative code and public acts related to MCIR. www.mcir.org/resource/laws/. Accessed September 12, 2017. 10. Centers for Disease Control and Prevention (CDC). Q & A about IIS sentinel sites. www.cdc.gov/vaccines/programs/iis/activities/sentinelsites.html. Published 2016. Accessed May 10, 2017. 11. Department of Community Health Bureau of Epidemiology. Communicable and related diseases. http://w3.lara.state.mi.us/orr/Files/Admin Code/1472_2014-073CH_AdminCode.pdf. Published 2014. Accessed July 27, 2017. 12. Michigan Department of Health and Human Services. Vaccines required for child care and preschool in Michigan. www.michigan. gov/documents/mdhhs/Child_Care_Reqs_parents_553545_7.pdf. Published 2017. Accessed August 9, 2017. 13. National Committee for Quality Assurance. Childhood immunization status. HEDIS measures. www.ncqa.org/portals/0/Childhood%20Immu nization%20Status.pdf. Published 2009. Accessed May 22, 2017. 14. Glauber JH. The immunization delivery effectiveness assessment score: a better immunization measure? Pediatrics. 2003;112(1):e39–e45. 15. Curran D, Terlinden A, Poirrier JE, Masseria C, Krishnarajah G. Vaccine timeliness: a cost analysis of the implications of delayed vaccination. Value Health. 2013;16(7):A346. https://doi.org/10.1016/j. jval.2013.08.143. 16. Winter K, Harriman K, Zipprich J, Schechter R, Talarico J. California 2010 Pertussis Epidemic [Abstract 2526]. 45th National Immunization Conference. Washington, DC: Public Health Foundation, 2011. 17. Luman ET, Barker LE, McCauley MM, Drews-Botsch C. Timeliness of childhood immunizations: a state-specific analysis. Am J Epidemiol. 2005;95(8):1367–1374. https://doi.org/10.2105/AJPH.2004.046284. 18. Lernout T, Theeten H, Hens N, et al. Timeliness of infant vaccination and factors related with delay in Flanders, Belgium. Vaccine. 2014;32 (2):284–289. https://doi.org/10.1016/j.vaccine.2013.10.084. 19. Fadnes LT, Nankabirwa V, Sommerfelt H, Tylleskär T, Tumwine JK, Engebretsen IMS. Is vaccination coverage a good indicator of
20.
21.
22.
23.
24.
25.
26.
27. 28.
29.
30.
31.
32.
33.
34.
age-appropriate vaccination? A prospective study from Uganda. Vaccine. 2011;29(19):3564–3570. https://doi.org/10.1016/j.vaccine.2011. 02.093. Centers for Disease Control and Prevention (CDC). Continued shortage of Haemophilus influenzae type b (Hib) conjugate vaccines and potential implications for Hib surveillance—United States, 2008. MMWR Morb Mortal Wkly Rep. 2008;57(46):1252–1255. Chamberlain AT, Wells K, Seib K, et al. Lessons learned from the 2007 to 2009 Haemophilus influenzae type b vaccine shortage: implications for future vaccine shortages and public health preparedness. J Public Health Manag Pract. 2012;18(3):E9–E16. https://doi.org/10.1097/ PHH.0b013e31821dce27. Luman ET, McCauley MM, Stokley S, Chu SY, Pickering LK. Timeliness of childhood immunizations. Pediatrics. 2002;110(5):935–939. https://doi.org/10.1542/peds.110.5.935. Luthy KE, Beckstrand RL, Peterson NE. Parental hesitation as a factor in delayed childhood immunization. J Pediatr Health Care. 2009;23 (6):388–393. https://doi.org/10.1016/j.pedhc.2008.09.006. Smith PJ, Humiston SG, Vannice KS. The association between intentional delay of vaccine administration and timely childhood vaccination coverage. Public Health Rep. 2010;125(4):534–541. https://doi.org/ 10.1177/003335491012500408. Fu LY, Zook K, Gingold J, et al. Frequent vaccination missed opportunities at primary care encounters contribute to underimmunization. J Pediatr. 2015;166(2):412–417. https://doi.org/10.1016/j. jpeds.2014.10.066. Robison SG, Kurosky SK, Young CM, Gallia CA, Arbor SA. Immunization milestones: a more comprehensive picture of age-appropriate vaccination. J Biomed Biotechnol. 2010;2010:916525. https://doi.org/ 10.1155/2010/916525. Navin MC. The ethics of vaccination nudges in pediatric practice. HEC Forum. 2017;29(1):43–57. https://doi.org/10.1007/s10730-016-9311-2. Offit PA, Quarles J, Gerber MA, et al. Addressing parents’ concerns: do multiple vaccines overwhelm or weaken the infant’s immune system? Pediatrics. 2002;109(1):124–129. https://doi.org/10.1542/peds.109.1.124. Gidengil C, Lieu TA, Payne K, Rusinak D, Messonnier M, Prosser LA. Parental and societal values for the risks and benefits of childhood combination vaccines. Vaccine. 2012;30(23):3445–3452. https://doi. org/10.1016/j.vaccine.2012.03.022. Happe LE, Lunacsek OE, Kruzikas DT, Marshall GS. Impact of a pentavalent combination vaccine on immunization timeliness in a state Medicaid population. Pediatr Infect Dis J. 2009;28(2):98–101. https://doi.org/10.1097/INF.0b013e318187d047. Marshall GS, Happe LE, Lunacsek OE, et al. Use of combination vaccines is associated with improved coverage rates. Pediatr Infect Dis J. 2007;26(6):496–500. https://doi.org/10.1097/INF.0b013e31805d7f17. Lauderdale DS, Oram RJ, Goldstein KP, Daum RS. Hepatitis B vaccination among children in inner-city public housing, 1991–1997. JAMA. 1999;282(18):1725–1730. https://doi.org/10.1001/jama.282.18.1725. Potter RC, DeVita SF, Vranesich PA, Boulton ML. Adolescent immunization coverage and implementation of new school requirements in Michigan, 2010. Am J Public Health. 2014;104(8):1526–1533. https://doi.org/10.2105/AJPH.2014.301910. The National Research Council. The growing problem of nonresponse. In: Tourangeau R, Plewes TJ, eds. Nonresponse in Social Science Surveys: A Research Agenda. Washington, DC: National Academies Press, 2013:7–39. https://doi.org/10.17226/18293.
www.ajpmonline.org