- Email: [email protected]
Pharmacist impact on pediatric vaccination errors and missed opportunities in the setting of clinical decision support
SCIENCE AND PRACTICE Journal of the American Pharmacists Association xxx (2017) 1e6
Contents lists available at ScienceDirect
Journal of the American Pharmacists Association journal homepage: www.japha.org
RESEARCH
Pharmacist impact on pediatric vaccination errors and missed opportunities in the setting of clinical decision support Kelly A. Wise*, Sonya J. Sebastian, Anna C. Haas-Gehres, Melissa D. Moore-Clingenpeel, Kristen E. Lamberjack a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 June 2016 Accepted 7 February 2017
Objective: To determine pharmacist impact on vaccination errors and missed opportunities in the pediatric primary care setting with the presence of clinical decision support (CDS) by comparing a clinic with a pharmacist and CDS to a clinic with CDS alone. Design: A retrospective chart review of patients' electronic medical records compared vaccination errors and missed opportunities between 2 pediatric primary care clinics. Setting: Two urban, pediatric primary care clinics were selected for the study. Participants: Encounters were included in the analysis for children presenting for any visit over a 3-month period. Intervention: The intervention clinic had a full-time clinical pharmacist and CDS. The comparison clinic had CDS alone. Main outcome measures: Vaccination errors were defined as follows: doses administered before minimum recommended age, doses administered before minimum recommended dosing interval, unnecessary doses, and invalid doses for a combination of these reasons. Missed opportunities were defined as vaccine doses due at the date of encounter but not administered, without documented reason for vaccination delay or refusal by provider or patient. The likelihood of missing an opportunity was also assessed for patient age, visit type, and provider type. Results: One thousand and twenty patient encounters were randomly selected and reviewed. The vaccination error rate was 0.4% in the comparison group and 0% in the intervention group (P ¼ 0.4995). The number of encounters with a missed opportunity was significantly higher in the comparison group compared with the intervention group (51 vs. 30 encounters with missed opportunities; P ¼ 0.015; adjusted odds ratio, 2.14 [95% CI 1.3-35]). Conclusion: Although the use of CDS results in a low rate of vaccination errors, technology cannot be solely relied on for vaccination recommendations in the pediatric population because of the rigidity of CDS configuration. Pharmacists continue to play a vital role to ensure that children are appropriately vaccinated in the primary care setting. © 2017 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Vaccines have had a tremendous impact on reducing preventable diseases in the United States. Polio, diphtheria, and smallpox are nearly eradicated. Although still prevalent, other diseases, such as measles, mumps, rubella, tetanus, and pertussis, have decreased drastically since the 1900s.1 Overall, vaccines are among the most cost-effective strategies to prevent infectious diseases, with high returns on investment for childhood vaccine administration. For each birth cohort
Disclosure: The authors declare no conflicts of interest or financial interests in any product or service mentioned in this article. * Correspondence: Kelly A. Wise, PharmD, PGY-1 Pharmacy Resident, Nationwide Children's Hospital, 700 Children's Dr., Columbus, OH 43205. E-mail address: [email protected] (K.A. Wise).
vaccinated with the routine immunization schedule, society saves 33,000 lives, prevents 14 million cases of disease, reduces direct health care costs by $9.9 billion, and saves $33.4 billion in indirect costs.2 In December 2010, the U.S. Department of Health and Human Services launched Healthy People 2020. Healthy People 2020 is an initiative that includes 42 areas of interest that represent important health topics. One important area of focus is increasing immunization rates and reducing preventable diseases.2,3 Practitioners of all levels are expected to stay current on vaccine recommendations published by the Centers for Disease Control and Prevention (CDC) on an annual basis.4 All vaccine series have minimum recommended age and dosing intervals. Most vaccines have relevant contraindications and
http://dx.doi.org/10.1016/j.japh.2017.02.025 1544-3191/© 2017 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
SCIENCE AND PRACTICE K.A. Wise et al. / Journal of the American Pharmacists Association xxx (2017) 1e6
Key Points Background: The complexity of the pediatric vaccination schedules creates a risk of vaccination errors and missed opportunities to vaccinate children properly. Previous literature supports the impact of pharmacists on reducing vaccination errors and missed opportunities in the pediatric primary care setting. Clinical decision support is a tool within the electronic medical record that has the functionality to evaluate the current vaccine status of a patient and to alert the provider of vaccination opportunities. To our knowledge, no previous studies have evaluated pharmacist impact on vaccination errors and missed opportunities in the presence of clinical decision support. Findings: The use of clinical decision support results in low vaccination error rates. There is potential to miss opportunities to vaccinate children properly because of the rigidity of clinical decision support configuration. Pharmacist integration into a primary care clinic remains valuable despite the introduction of vaccine forecasting technology in the ambulatory setting.
precautions that must be considered before administration. In addition, there are high-risk populations that may benefit from additional doses of Haemophilus influenzae type b, pneumococcal, or meningococcal vaccines.4 The complexity of pediatric vaccination schedules creates a risk of vaccination errors and missed opportunities to vaccinate children properly.5,6 Vaccination errors lead to negative outcomes such as inadequate immunologic protection, possible injury, increased cost, inconvenience, and reduced confidence in the health care delivery system.7,8 The U.S. Vaccine Adverse Event Reporting System identified 21,843 errors that occurred from 2001 through 2013. Of these errors, 5947 (27%) occurred because of inappropriate scheduling, 4983 (23%) occurred because of storage and dispensing errors, and 3372 (15%) occurred because of administration of the wrong vaccine. In the 0-18eyear-old group, 53% of the errors caused by inappropriate scheduling involved children less than 1 year of age. Vaccines with age restrictions, such as rotavirus (RV) and human papillomavirus (HPV), were associated with inappropriate scheduling. Common vaccines that were associated with the wrong vaccine administration were primarily vaccines with similar names or acronyms, such as diphtheria, tetanus, and acellular pertussis (DTaP) versus tetanus, diphtheria, and acellular pertussis (Tdap) or pneumococcal conjugate (PCV13) versus polysaccharide (PPSV23).8 Pediatric vaccination rates have the potential to be positively affected by the integration of a pharmacist within the primary care setting. As medication experts, pharmacists are
2
knowledgeable about routine and catch-up vaccine schedules, high-risk populations, and relevant contraindications to vaccines. Pharmacists also have the opportunity to develop relationships with families and to discuss the risks of not vaccinating children while dispelling myths that surround the use of vaccines. Finally, pharmacists work directly with providers to recommend appropriate vaccines based on patients' medical histories and to review final vaccine orders from the providers. There have been numerous studies demonstrating a positive relationship between pharmacist involvement and improved vaccination rates. Most of these studies are conducted in the adult population and involve pharmacists administering vaccines in the community setting.9,10 A study conducted by Haas-Gehres et al.11 at our institution assessed the impact of ambulatory clinical pharmacists on vaccination errors and missed opportunities. This study compared the rate of vaccination errors and the number of missed opportunities between a pediatric primary care clinic with a pharmacist to a pediatric primary care clinic without a pharmacist. The error rate was found to be 2.8% in the clinic without a pharmacist compared to 0.28% in the clinic with a pharmacist (P ¼ 0.0021). The number of clinic visits with at least one missed opportunity also showed a statistically significant difference between the clinic without a pharmacist compared to the clinic with a pharmacist (29.3% vs. 10.2%; P <0.0001). The authors concluded that pharmacists play a key role in the pediatric primary care team to improve the use of vaccines.11 Since the publication of this study, our institution has implemented clinical decision support (CDS) that is designed to minimize vaccination errors and missed opportunities. Vaccination CDS is a tool within the electronic medical record that has the functionality to evaluate the current vaccine status of a patient and to alert the provider of vaccination opportunities. The alerts provided by CDS can be configured with the minimum age for vaccine administration, number of doses, dosing interval, precautions, and contraindications of vaccines. Functionality of CDS can vary among institutions.12 There is limited research assessing the impact of CDS on improving pediatric vaccination rates.13 To our knowledge, there have been no studies assessing pharmacist influence on vaccination errors and missed opportunities in the setting of CDS. We hypothesized that a setting with pharmacist support and CDS would result in fewer vaccine errors and missed opportunities compared with a setting with CDS alone. Objective The primary objective of the study was to determine the effect of pharmacist involvement on vaccination errors and missed opportunities in the setting of CDS by comparing a pediatric primary care clinic with a pharmacist and CDS to a clinic with CDS alone. The secondary objective was to determine whether the likelihood of a missed opportunity varies according to patient age, visit type, and provider type. Methods The study was performed at Nationwide Children's Hospital in Columbus, Ohio. As an academic medical center,
SCIENCE AND PRACTICE Pharmacist impact on vaccines in the setting of CDS
attending physicians and medical residents provide care to patients. The health system's primary care network consists of 13 primary care centers in Central Ohio serving 80,000 patients per year. The institutional review board considered this study exempt from review. Two clinics in the primary care network were selected for comparison. The clinics selected are located in the same ZIP code separated by a distance of 1 mile and routinely care for primarily urban patient populations. Medical residents, attending physicians, and staff members were consistently present in each clinic for the duration of the study. Medical residents spent a half day in the clinic each week and remained in the same clinic throughout the duration of their residency. The comparison clinic had an additional attending physician on their schedule compared with the intervention clinic, leading to an increased overall number of patients seen during the study period (Figure 1). The types of visits were similar between the 2 clinics. The intervention clinic had a full-time, board-certified ambulatory clinical pharmacist integrated into the health care team. The pharmacist reviewed charts as patients presented to the clinic to assess if patients were eligible for vaccination opportunities. The pharmacist developed a specific vaccination plan for eligible patients and communicated the plan to the provider by verbal communication or documentation in the electronic medical record. This task was performed in addition to regular duties, such as counseling patients, assisting in insurance concerns, performing medication reconciliation, and recommending proper treatment options. In contrast, the comparison clinic did not have any support from a pharmacist to assist staff members in determining appropriate vaccination options. Both clinics used the same electronic medical record with CDS to alert for vaccination opportunities based on the routine immunization schedule published by CDC. Considerations not included in the health system's CDS are high-risk populations, drug interactions, precautions, contraindications, and a 4-day grace period. Children 18 years old and younger who were seen in the primary care clinic for well visits, sick visits, or immunization
6 encounters excluded 2 erroneous encounters
visits were included in the study. Patients were excluded if there was an absent vaccine history, an erroneous encounter, or restricted privacy access to the patient's electronic medical record. All office visits from April 2015 through June 2015 were gathered, and encounters for review were randomly selected using computer-generated randomization. To achieve statistical power of 80% to detect a difference of 2.42% in error rate and 19.1% in missed opportunity rate, a sample size of 495 charts per clinic group was required for analysis.11 Individual vaccination records and clinic visit summaries were assessed in the health system's electronic medical record. Vaccines of interest included the following: hepatitis A (HepA); hepatitis B; inactivated polio; measles, mumps, and rubella; meningococcal conjugate; tetanus and diphtheria; varicella; DTaP; HPV; PCV13; PPSV23; RV; and Tdap. The 2015 recommended immunization schedules published by CDC were used to assess vaccination errors and missed opportunities. For each encounter, the following data was evaluated: visit type, provider type, number and type of vaccines administered, vaccination errors on encounter date, and missed opportunities on encounter date. An unblinded reviewer was assigned to each clinic to assess vaccination errors and missed opportunities. Reviewers evaluated encounters in which they were not part of the patient's care. A third reviewer who reviewed all errors independently was consulted as needed to ensure consistency in the evaluation process. There was 100% consensus between reviewers on vaccination errors. Vaccination errors were defined as follows: doses administered before minimum recommended age, doses administered before minimum recommended dosing interval, unnecessary doses, and invalid doses for a combination of these reasons. Vaccination error rate was calculated by assessing number of vaccination errors per number of total encounters for each clinic. Missed opportunities were defined as vaccine doses due at date of encounter but not administered. Missed opportunity rate was calculated by assessing the number of missed opportunities per total number of vaccination opportunities (vaccines given plus vaccines missed). Lack of vaccination was not considered a missed opportunity if
2591 encounters in comparison clinic
1534 encounters in intervention clinic
510 encounters selected for review in comparison clinic
510 encounters selected for review in intervention clinic
504 encounters included in data analysis for comparison clinic
504 encounters included in data analysis for intervention clinic
3 absent vaccine records 1 restricted privacy
6 encounters excluded 4 erroneous encounters 2 absent vaccine records
Figure 1. Patient selection.
3
SCIENCE AND PRACTICE K.A. Wise et al. / Journal of the American Pharmacists Association xxx (2017) 1e6
there was a documented reason (i.e., illness) for vaccination delay; refusal by provider, patient, or caregiver; or documentation of an active inquiry of vaccination history (i.e., release of information to another provider). All analyses were performed using SAS 9.3 (SAS Institute, Cary NC) with 2-sided P <0.05 considered statistically significant. Comparisons between clinics were assessed using chisquare or Fisher exact tests. Tests were performed for each vaccination type because children could receive more than 1 vaccine. Multivariable logistic regression was used to assess which factors were independently associated with having any missed opportunity; patient age, provider type, visit type, and clinic type (comparison vs. intervention) were considered as potential predictors. Variables with P <0.1 were retained in the final model. In addition, tests for an interaction between clinic type and patient age, visit type, and provider type were assessed to determine whether the relative benefit of the intervention clinic varies with patient age or visit characteristics. Results During the 3-month study period, there were 4125 clinic encounters (Figure 1). There were 510 encounters randomly selected for review from each clinic. Six encounters were excluded from the comparison clinic, and 6 encounters were excluded from the intervention clinic, allowing 504 encounters to be included in the data analysis for each clinic. Baseline characteristics were similar between the 2 groups, with the exception of provider type (Table 1). Vaccination errors There were 2 vaccination errors in the comparison clinic (error rate ¼ 0.4%) and no errors in the intervention clinic (error rate ¼ 0%; P ¼ 0.4995). A multivariable analysis could not be performed on such low event numbers. The 2 errors in the comparison clinic consisted of 1 unnecessary dose of RV and 1 unnecessary dose of Tdap. Missed opportunities
Table 1 Demographic information Demographics Patient age (years) 0e2 3e6 7e10 11e18 Visit type Well visit Sick visit Immunization visit Provider type Medical resident Attending physician Vaccine administrationsb Any DTaP Hib HepA HepB HPV MMR MCV PCV13/PPSV23 IPV RV Tdap Td VAR
4
Intervention clinic, n (%)
214 108 89 93
234 77 96 97
P value 0.0922
(42.5) (21.0) (17.6) (18.5)
(46.4) (15.2) (19.0) (19.2) 0.159
274 (54.4) 217 (43.1) 13 (2.6)
244 (48.4) 247 (49) 13 (2.6)
190 (38.7) 301 (61.3)
349 (71.1) 142 (28.9)
175 95 85 49 67 42 37 16 85 80 36 9 0 38
181 111 92 31 75 52 30 15 88 83 42 11 0 31
<0.0001a
(35.1) (18.9) (16.9) (9.7) (13.3) (8.3) (7.3) (3.2) (16.9) (15.9) (7.1) (1.8) (0) (7.5)
(35.9) (22.0) (18.3) (6.2) (14.9) (10.3) (6.0) (3.0) (17.5) (16.5) (8.3) (2.2) (0) (6.2)
0.7804 0.2114 0.5623 0.036a 0.4767 0.2787 0.3761 0.8552 0.8021 0.7975 0.4794 0.6515 d 0.3826
Abbreviations used: DTaP, diphtheria, tetanus, and acellular pertussis; Hib, Haemophilus influenzae type b; HepA, hepatitis A; HepB, hepatitis B; HPV, human papillomavirus; IPV, inactivated polio; MMR, measles, mumps, and rubella; MCV, meningococcal conjugate; PCV13, pneumococcal conjugate; PPSV23, pneumococcal polysaccharide; RV, rotavirus; Tdap, tetanus, diphtheria, and acellular pertussis; Td, tetanus and diphtheria; VAR, varicella. a Statistical significance. b P values are reported for each individual vaccine because patients could receive more than 1.
Table 2 Missed opportunities by vaccine Vaccine
There were 80 missed opportunities in the comparison clinic and 42 in the intervention clinic (missed opportunity rate ¼ 11.1% vs. 6.4%). There were significantly more patients with at least 1 missed opportunity in the comparison clinic than in the intervention clinic (10% vs. 6%; P ¼ 0.015; Table 2). When broken down by vaccine series, patients in the comparison clinic were significantly more likely to have a missed opportunity for DTaP, HepA, and PCV13/PPSV23 (Table 2). In multivariable assessment, clinic type, age of patient, and visit type were significantly associated with the odds of having at least 1 missed opportunity (Table 3). Provider type was not significantly associated with odds of missed opportunity and was, therefore, removed from the model. After adjusting for patient age and visit type, patients in the comparison clinic were still significantly more likely to have a missed opportunity (adjusted OR ¼ 2.14 [95% CI 1.3-3.5]). Regarding age, patients in the 7-10eyear age group were most likely to have a missed opportunity, followed by those in the 11-18-year age group. In addition, patients being seen for a sick visit were significantly
Comparison clinic, n (%)
Any DTaP Hib HepA HepB HPV MMR MCV PCV13/PPSV23 IPV RV Tdap Td VAR
Comparison clinic, n (%)
Intervention clinic, n (%)
P value
51 12 5 7 2 30 4 1 8 2 1 2 2 5
30 4 1 1 1 24 1 2 1 2 1 2 2 1
0.015a 0.0438a 0.1014 0.0332a >0.9999 0.4013 0.3738 >0.9999 0.0382a >0.9999 0.9999 0.9999 0.9999 0.2174
(10) (2) (1) (1) (0) (6) (1) (0) (2) (0) (0) (0) (2) (1)
(6) (1) (0) (0) (0) (5) (0) (0) (0) (0) (0) (0) (2) (0)
Abbreviations used: DTaP, diphtheria, tetanus, and acellular pertussis; Hib, Haemophilus influenzae type b; HepA, hepatitis A; HepB, hepatitis B; HPV, human papillomavirus; IPV, inactivated polio; MMR, measles, mumps, and rubella; MCV, meningococcal conjugate; PCV13, pneumococcal conjugate; PPSV23, pneumococcal polysaccharide; RV, rotavirus; Tdap, tetanus, diphtheria, and acellular pertussis; Td, tetanus and diphtheria; VAR, varicella. a Statistical significance.
SCIENCE AND PRACTICE Pharmacist impact on vaccines in the setting of CDS
Table 3 Adjusted odds of missed opportunity Variable Comparison clinic Intervention clinic 0e2 years old 3e6 years old 7e10 years old 11e18 years old Immunization visit Sick visit Well visit
% with MO
aOR
95% CI
P value
10.12 5.95 3.57 3.24 20.54 11.05 7.69 12.07 4.44
2.143 Reference Reference 0.673 5.65 2.75 1.827 2.467 Reference
1.309e3.509 d d 0.255e1.772 3.005e10.623 1.379e5.483 0.392e8.503 1.451e4.194 d
0.0025a d d 0.4224 <0.0001a 0.0041a 0.4425 0.0009a d
Abbreviations used: MO, missed opportunity; aOR, adjusted odds ratio. a Statistical significance.
more likely to have a missed opportunity compared with those being seen for a well visit (adjusted OR ¼ 2.5 [95% CI 1.5-4.2]). There were no statistically significant interaction effects between clinic type and patient age (P ¼ 0.5804), visit type (P ¼ 0.0632), or provider type (P ¼ 0.2367). Discussion A reduction in vaccine errors and missed opportunities has important positive outcomes, including improved public health, decreased preventable harm, reduced child discomfort, and potential cost savings. The use of CDS results in a low vaccination error rate; however, opportunities to vaccinate children may be missed because of the rigidity of CDS configuration. In a 3-month period, there were 175 vaccine administrations in the comparison clinic and 181 vaccine administrations in the intervention clinic. In addition, there were 51 encounters (10%) with at least 1 missed opportunity in the comparison clinic and 30 encounters (6%) with at least 1 missed opportunity in the intervention clinic. The significant difference in missed opportunities between the clinics remained after accounting for age of the patient, visit type, or provider type; even after accounting for these risk factors, patients in the comparison clinic had more than a 2-fold increased chance of having a missed opportunity. Although CDS is an aid in forecasting vaccine recommendations, the alerts can be overlooked easily by providers because of “alert fatigue,” especially when vaccines are not the primary reason for visit. In addition, the health system's CDS is not equipped to alert for high-risk populations, drug interactions, precautions, contraindications, and a 4-day grace period. Following categorization by vaccine series, patients in the comparison clinic were significantly more likely to have a missed opportunity for DTaP, HepA, and PCV13/PPSV23 compared with patients in the intervention clinic. These vaccines were also among those with the greatest number of missed opportunities overall. A factor that may have affected the number of missed opportunities in the DTaP series is the fourth dose of DTaP. The fourth dose should be given 6 months after the third dose, which is routinely given at 6 months of age. Therefore, administration of the fourth dose at 12 months of age is appropriate if there is a proper dosing interval from the previous dose; however, CDS does not alert for the fourth dose of DTaP until 15 months of age, creating the potential to miss an opportunity to vaccinate a child. HepA was added to the routine immunization schedule by CDC in 2006. It is
recommended to vaccinate any patient that did not receive HepA prior to 2006. Therefore, older patients are at risk for not being up to date for HepA. High-risk patient populations (e.g., sickle cell disease, chronic renal failure) were a factor affecting missed opportunities for the pneumococcal vaccines. There were high numbers of missed opportunities of the HPV vaccine in both clinics, although this was not statistically significantly associated with clinic type. In 2015, the health system introduced the recommendation to administer the first dose of HPV at 9 years of age. This alert was manually built into CDS and went live simultaneously in both clinics involved in the study in June 2015. Finally, it was determined in the study that the 7-10eyear age group had the highest likelihood of having a missed opportunity, followed by patients in the 11-18eyear age group. Historically, patients in the 7-10eyear age group were not scheduled for routine vaccinations, but they can often be candidates for catch-up vaccines. Because of the inconsistent need for vaccines in these age groups and the relatively new adjustment in recommended age for HPV initiation, patients aged 7-18 years are more vulnerable to missed opportunities. To our knowledge, this is the first study to assess the impact of pharmacists on vaccination errors and missed opportunities in the setting of CDS. The research question and findings of this study are an important safety consideration in the pediatric ambulatory arena, which is a modest but growing segment of pharmacy practice. It is evident that ambulatory pharmacists play a key role in improving vaccination rates with the aim of meeting the goals of Healthy People 2020.2,3 Limitations There are limitations to this study. First, it was a retrospective review that assessed vaccine orders; therefore, vaccination errors that occurred during administration were not assessed. Second, the review relied on documentation. Lack of documentation for a vaccine refusal was considered a missed opportunity. Although both clinics relied on the same vaccine schedules and primary care protocols, provider preference may have had an impact on results. Finally, staff adjustment to the change in recommended initial age of HPV may have influenced missed opportunities for this specific vaccine series. High-risk patients were included in population selection by randomization; however, it was not the primary focus of this research. Further research into the impact of pharmacist involvement on vaccination errors and missed opportunities in a setting focusing on high-risk patient populations is necessary; in particular, a larger sample size will be needed to have a large enough number of vaccination errors to facilitate multivariable analysis. The economic implication that is associated with pharmacist integration into a primary care clinic is another area that requires further investigation. Conclusion The results of this study indicate that pharmacist integration into a primary care clinic remains valuable despite the introduction of vaccine forecasting technology in the clinic setting. Although the use of CDS results in a low vaccination error rate, opportunities to vaccinate children may be missed 5
SCIENCE AND PRACTICE K.A. Wise et al. / Journal of the American Pharmacists Association xxx (2017) 1e6
because of the rigidity of the CDS configuration. Institutions are faced with the challenge of customizing technology to meet the needs of diverse patient populations, which requires additional resources. Pharmacist involvement in the primary care clinic appears to have a significant impact on reducing missed opportunities. Pharmacists are able to focus on vaccine opportunities, provide education and recommendations, and improve overall vaccination rates in the primary care setting. Acknowledgments We thank Han Yin, MS, Biostatistician and Yongjie Miao, Biostatistician for their contributions to the statistical analysis of the study. References 1. Centers for Disease Control and Prevention. Impact of vaccines in the 20th and 21st centuries; 2011. Available at: https://www.cdc.gov/vaccines/ pubs/pinkbook/downloads/appendices/e/impact.pdf. Accessed May 18, 2016. 2. U.S. Department of Health and Human Services. Healthy People 2020 topics and objectives: immunization and infectious diseases. Available at: https://www.healthypeople.gov/2020/topics-objectives/topic/immuniza tion-and-infectious-diseases. Accessed May 18, 2016. 3. U.S. Department of Health and Human Services. Healthy People 2020. Available at: http://www.cdc.gov/nchs/healthy_people/hp2020.htm. Accessed November 1, 2016. 4. U.S. Department of Health and Human Services. Recommended immunization schedules for persons aged 0 through 18 years; 2015. Available at: http:// www.cdc.gov/vaccines/schedules/hcp/child-adolescent.html. Accessed November 20, 2015. 5. Bundy DG, Shore AD, Morlock LL, et al. Pediatric vaccination errors: application of the “5 Rights” framework to a national error reporting database. Vaccine. 2009;27(29):3890e3896.
6
6. Weltermann BM, Markic M, Thielmann A, et al. Vaccination management and vaccination errors: a representative online survey among primary care physicians. PLoS One. 2014;9(8):e105199. 7. Rees P, Edwards A, Powell C, et al. Pediatric immunization-related safety incidents in primary care: a mixed methods analysis of a national database. Vaccine. 2015;33(32):3873e3880. 8. Hibbs BF, Moro PL, Lewis P, et al. Vaccination errors reported to the Vaccine Adverse Event Reporting System, (VAERS) United States, 20002013. Vaccine. 2015;33(28):3171e3178. 9. Taitel M, Cohen E, Duncan I, et al. Pharmacists as providers: targeting pneumococcal vaccinations to high-risk populations. Vaccine. 2011;29(45): 8073e8076. 10. Baroy J, Ching D, Frisch R, et al. The impact of pharmacist immunization programs on adult immunization rates: a systematic review and metaanalysis. J Am Pharm Assoc. 2016;56(4):418e426. 11. Haas-Gehres A, Sebastian S, Lamberjack K. Impact of pharmacist integration into a pediatric primary care clinic on vaccination errors: a retrospective review. J Am Pharm Assoc. 2014;54(4):415e418. 12. U.S. Department of Health and Human Services. Clinical Decision Support for Immunizations (CDSi). Available at: http://www.cdc.gov/ vaccines/programs/iis/interop-proj/downloads/cdsi-brochure.pdf. Accessed November 20, 2015. 13. Bundy DG, Persing NM, Solomon BS, et al. The ImmProve project: leveraging electronic health record data to promote immunization delivery. Acad Pediatr. 2013;13(5):458e465. Kelly A. Wise, PharmD, PGY-1 Pharmacy Resident, Nationwide Children's Hospital, Columbus, OH Sonya J. Sebastian, PharmD, BCACP, Clinical Pharmacist, Nationwide Children's Hospital, Columbus, OH Anna C. Haas-Gehres, PharmD, BCACP, Clinical Pharmacist, Nationwide Children's Hospital, and Assistant Professor, The Ohio State University, Columbus, OH Melissa D. Moore-Clingenpeel, MA, MAS, Biostatistics Core, Research Institute at Nationwide Children's Hospital, Columbus OH Kristen E. Lamberjack, PharmD, BCACP, Clinical Pharmacist, Nationwide Children's Hospital, Columbus, OH
Contents lists available at ScienceDirect
Journal of the American Pharmacists Association journal homepage: www.japha.org
RESEARCH
Pharmacist impact on pediatric vaccination errors and missed opportunities in the setting of clinical decision support Kelly A. Wise*, Sonya J. Sebastian, Anna C. Haas-Gehres, Melissa D. Moore-Clingenpeel, Kristen E. Lamberjack a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 June 2016 Accepted 7 February 2017
Objective: To determine pharmacist impact on vaccination errors and missed opportunities in the pediatric primary care setting with the presence of clinical decision support (CDS) by comparing a clinic with a pharmacist and CDS to a clinic with CDS alone. Design: A retrospective chart review of patients' electronic medical records compared vaccination errors and missed opportunities between 2 pediatric primary care clinics. Setting: Two urban, pediatric primary care clinics were selected for the study. Participants: Encounters were included in the analysis for children presenting for any visit over a 3-month period. Intervention: The intervention clinic had a full-time clinical pharmacist and CDS. The comparison clinic had CDS alone. Main outcome measures: Vaccination errors were defined as follows: doses administered before minimum recommended age, doses administered before minimum recommended dosing interval, unnecessary doses, and invalid doses for a combination of these reasons. Missed opportunities were defined as vaccine doses due at the date of encounter but not administered, without documented reason for vaccination delay or refusal by provider or patient. The likelihood of missing an opportunity was also assessed for patient age, visit type, and provider type. Results: One thousand and twenty patient encounters were randomly selected and reviewed. The vaccination error rate was 0.4% in the comparison group and 0% in the intervention group (P ¼ 0.4995). The number of encounters with a missed opportunity was significantly higher in the comparison group compared with the intervention group (51 vs. 30 encounters with missed opportunities; P ¼ 0.015; adjusted odds ratio, 2.14 [95% CI 1.3-35]). Conclusion: Although the use of CDS results in a low rate of vaccination errors, technology cannot be solely relied on for vaccination recommendations in the pediatric population because of the rigidity of CDS configuration. Pharmacists continue to play a vital role to ensure that children are appropriately vaccinated in the primary care setting. © 2017 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Vaccines have had a tremendous impact on reducing preventable diseases in the United States. Polio, diphtheria, and smallpox are nearly eradicated. Although still prevalent, other diseases, such as measles, mumps, rubella, tetanus, and pertussis, have decreased drastically since the 1900s.1 Overall, vaccines are among the most cost-effective strategies to prevent infectious diseases, with high returns on investment for childhood vaccine administration. For each birth cohort
Disclosure: The authors declare no conflicts of interest or financial interests in any product or service mentioned in this article. * Correspondence: Kelly A. Wise, PharmD, PGY-1 Pharmacy Resident, Nationwide Children's Hospital, 700 Children's Dr., Columbus, OH 43205. E-mail address: [email protected] (K.A. Wise).
vaccinated with the routine immunization schedule, society saves 33,000 lives, prevents 14 million cases of disease, reduces direct health care costs by $9.9 billion, and saves $33.4 billion in indirect costs.2 In December 2010, the U.S. Department of Health and Human Services launched Healthy People 2020. Healthy People 2020 is an initiative that includes 42 areas of interest that represent important health topics. One important area of focus is increasing immunization rates and reducing preventable diseases.2,3 Practitioners of all levels are expected to stay current on vaccine recommendations published by the Centers for Disease Control and Prevention (CDC) on an annual basis.4 All vaccine series have minimum recommended age and dosing intervals. Most vaccines have relevant contraindications and
http://dx.doi.org/10.1016/j.japh.2017.02.025 1544-3191/© 2017 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
SCIENCE AND PRACTICE K.A. Wise et al. / Journal of the American Pharmacists Association xxx (2017) 1e6
Key Points Background: The complexity of the pediatric vaccination schedules creates a risk of vaccination errors and missed opportunities to vaccinate children properly. Previous literature supports the impact of pharmacists on reducing vaccination errors and missed opportunities in the pediatric primary care setting. Clinical decision support is a tool within the electronic medical record that has the functionality to evaluate the current vaccine status of a patient and to alert the provider of vaccination opportunities. To our knowledge, no previous studies have evaluated pharmacist impact on vaccination errors and missed opportunities in the presence of clinical decision support. Findings: The use of clinical decision support results in low vaccination error rates. There is potential to miss opportunities to vaccinate children properly because of the rigidity of clinical decision support configuration. Pharmacist integration into a primary care clinic remains valuable despite the introduction of vaccine forecasting technology in the ambulatory setting.
precautions that must be considered before administration. In addition, there are high-risk populations that may benefit from additional doses of Haemophilus influenzae type b, pneumococcal, or meningococcal vaccines.4 The complexity of pediatric vaccination schedules creates a risk of vaccination errors and missed opportunities to vaccinate children properly.5,6 Vaccination errors lead to negative outcomes such as inadequate immunologic protection, possible injury, increased cost, inconvenience, and reduced confidence in the health care delivery system.7,8 The U.S. Vaccine Adverse Event Reporting System identified 21,843 errors that occurred from 2001 through 2013. Of these errors, 5947 (27%) occurred because of inappropriate scheduling, 4983 (23%) occurred because of storage and dispensing errors, and 3372 (15%) occurred because of administration of the wrong vaccine. In the 0-18eyear-old group, 53% of the errors caused by inappropriate scheduling involved children less than 1 year of age. Vaccines with age restrictions, such as rotavirus (RV) and human papillomavirus (HPV), were associated with inappropriate scheduling. Common vaccines that were associated with the wrong vaccine administration were primarily vaccines with similar names or acronyms, such as diphtheria, tetanus, and acellular pertussis (DTaP) versus tetanus, diphtheria, and acellular pertussis (Tdap) or pneumococcal conjugate (PCV13) versus polysaccharide (PPSV23).8 Pediatric vaccination rates have the potential to be positively affected by the integration of a pharmacist within the primary care setting. As medication experts, pharmacists are
2
knowledgeable about routine and catch-up vaccine schedules, high-risk populations, and relevant contraindications to vaccines. Pharmacists also have the opportunity to develop relationships with families and to discuss the risks of not vaccinating children while dispelling myths that surround the use of vaccines. Finally, pharmacists work directly with providers to recommend appropriate vaccines based on patients' medical histories and to review final vaccine orders from the providers. There have been numerous studies demonstrating a positive relationship between pharmacist involvement and improved vaccination rates. Most of these studies are conducted in the adult population and involve pharmacists administering vaccines in the community setting.9,10 A study conducted by Haas-Gehres et al.11 at our institution assessed the impact of ambulatory clinical pharmacists on vaccination errors and missed opportunities. This study compared the rate of vaccination errors and the number of missed opportunities between a pediatric primary care clinic with a pharmacist to a pediatric primary care clinic without a pharmacist. The error rate was found to be 2.8% in the clinic without a pharmacist compared to 0.28% in the clinic with a pharmacist (P ¼ 0.0021). The number of clinic visits with at least one missed opportunity also showed a statistically significant difference between the clinic without a pharmacist compared to the clinic with a pharmacist (29.3% vs. 10.2%; P <0.0001). The authors concluded that pharmacists play a key role in the pediatric primary care team to improve the use of vaccines.11 Since the publication of this study, our institution has implemented clinical decision support (CDS) that is designed to minimize vaccination errors and missed opportunities. Vaccination CDS is a tool within the electronic medical record that has the functionality to evaluate the current vaccine status of a patient and to alert the provider of vaccination opportunities. The alerts provided by CDS can be configured with the minimum age for vaccine administration, number of doses, dosing interval, precautions, and contraindications of vaccines. Functionality of CDS can vary among institutions.12 There is limited research assessing the impact of CDS on improving pediatric vaccination rates.13 To our knowledge, there have been no studies assessing pharmacist influence on vaccination errors and missed opportunities in the setting of CDS. We hypothesized that a setting with pharmacist support and CDS would result in fewer vaccine errors and missed opportunities compared with a setting with CDS alone. Objective The primary objective of the study was to determine the effect of pharmacist involvement on vaccination errors and missed opportunities in the setting of CDS by comparing a pediatric primary care clinic with a pharmacist and CDS to a clinic with CDS alone. The secondary objective was to determine whether the likelihood of a missed opportunity varies according to patient age, visit type, and provider type. Methods The study was performed at Nationwide Children's Hospital in Columbus, Ohio. As an academic medical center,
SCIENCE AND PRACTICE Pharmacist impact on vaccines in the setting of CDS
attending physicians and medical residents provide care to patients. The health system's primary care network consists of 13 primary care centers in Central Ohio serving 80,000 patients per year. The institutional review board considered this study exempt from review. Two clinics in the primary care network were selected for comparison. The clinics selected are located in the same ZIP code separated by a distance of 1 mile and routinely care for primarily urban patient populations. Medical residents, attending physicians, and staff members were consistently present in each clinic for the duration of the study. Medical residents spent a half day in the clinic each week and remained in the same clinic throughout the duration of their residency. The comparison clinic had an additional attending physician on their schedule compared with the intervention clinic, leading to an increased overall number of patients seen during the study period (Figure 1). The types of visits were similar between the 2 clinics. The intervention clinic had a full-time, board-certified ambulatory clinical pharmacist integrated into the health care team. The pharmacist reviewed charts as patients presented to the clinic to assess if patients were eligible for vaccination opportunities. The pharmacist developed a specific vaccination plan for eligible patients and communicated the plan to the provider by verbal communication or documentation in the electronic medical record. This task was performed in addition to regular duties, such as counseling patients, assisting in insurance concerns, performing medication reconciliation, and recommending proper treatment options. In contrast, the comparison clinic did not have any support from a pharmacist to assist staff members in determining appropriate vaccination options. Both clinics used the same electronic medical record with CDS to alert for vaccination opportunities based on the routine immunization schedule published by CDC. Considerations not included in the health system's CDS are high-risk populations, drug interactions, precautions, contraindications, and a 4-day grace period. Children 18 years old and younger who were seen in the primary care clinic for well visits, sick visits, or immunization
6 encounters excluded 2 erroneous encounters
visits were included in the study. Patients were excluded if there was an absent vaccine history, an erroneous encounter, or restricted privacy access to the patient's electronic medical record. All office visits from April 2015 through June 2015 were gathered, and encounters for review were randomly selected using computer-generated randomization. To achieve statistical power of 80% to detect a difference of 2.42% in error rate and 19.1% in missed opportunity rate, a sample size of 495 charts per clinic group was required for analysis.11 Individual vaccination records and clinic visit summaries were assessed in the health system's electronic medical record. Vaccines of interest included the following: hepatitis A (HepA); hepatitis B; inactivated polio; measles, mumps, and rubella; meningococcal conjugate; tetanus and diphtheria; varicella; DTaP; HPV; PCV13; PPSV23; RV; and Tdap. The 2015 recommended immunization schedules published by CDC were used to assess vaccination errors and missed opportunities. For each encounter, the following data was evaluated: visit type, provider type, number and type of vaccines administered, vaccination errors on encounter date, and missed opportunities on encounter date. An unblinded reviewer was assigned to each clinic to assess vaccination errors and missed opportunities. Reviewers evaluated encounters in which they were not part of the patient's care. A third reviewer who reviewed all errors independently was consulted as needed to ensure consistency in the evaluation process. There was 100% consensus between reviewers on vaccination errors. Vaccination errors were defined as follows: doses administered before minimum recommended age, doses administered before minimum recommended dosing interval, unnecessary doses, and invalid doses for a combination of these reasons. Vaccination error rate was calculated by assessing number of vaccination errors per number of total encounters for each clinic. Missed opportunities were defined as vaccine doses due at date of encounter but not administered. Missed opportunity rate was calculated by assessing the number of missed opportunities per total number of vaccination opportunities (vaccines given plus vaccines missed). Lack of vaccination was not considered a missed opportunity if
2591 encounters in comparison clinic
1534 encounters in intervention clinic
510 encounters selected for review in comparison clinic
510 encounters selected for review in intervention clinic
504 encounters included in data analysis for comparison clinic
504 encounters included in data analysis for intervention clinic
3 absent vaccine records 1 restricted privacy
6 encounters excluded 4 erroneous encounters 2 absent vaccine records
Figure 1. Patient selection.
3
SCIENCE AND PRACTICE K.A. Wise et al. / Journal of the American Pharmacists Association xxx (2017) 1e6
there was a documented reason (i.e., illness) for vaccination delay; refusal by provider, patient, or caregiver; or documentation of an active inquiry of vaccination history (i.e., release of information to another provider). All analyses were performed using SAS 9.3 (SAS Institute, Cary NC) with 2-sided P <0.05 considered statistically significant. Comparisons between clinics were assessed using chisquare or Fisher exact tests. Tests were performed for each vaccination type because children could receive more than 1 vaccine. Multivariable logistic regression was used to assess which factors were independently associated with having any missed opportunity; patient age, provider type, visit type, and clinic type (comparison vs. intervention) were considered as potential predictors. Variables with P <0.1 were retained in the final model. In addition, tests for an interaction between clinic type and patient age, visit type, and provider type were assessed to determine whether the relative benefit of the intervention clinic varies with patient age or visit characteristics. Results During the 3-month study period, there were 4125 clinic encounters (Figure 1). There were 510 encounters randomly selected for review from each clinic. Six encounters were excluded from the comparison clinic, and 6 encounters were excluded from the intervention clinic, allowing 504 encounters to be included in the data analysis for each clinic. Baseline characteristics were similar between the 2 groups, with the exception of provider type (Table 1). Vaccination errors There were 2 vaccination errors in the comparison clinic (error rate ¼ 0.4%) and no errors in the intervention clinic (error rate ¼ 0%; P ¼ 0.4995). A multivariable analysis could not be performed on such low event numbers. The 2 errors in the comparison clinic consisted of 1 unnecessary dose of RV and 1 unnecessary dose of Tdap. Missed opportunities
Table 1 Demographic information Demographics Patient age (years) 0e2 3e6 7e10 11e18 Visit type Well visit Sick visit Immunization visit Provider type Medical resident Attending physician Vaccine administrationsb Any DTaP Hib HepA HepB HPV MMR MCV PCV13/PPSV23 IPV RV Tdap Td VAR
4
Intervention clinic, n (%)
214 108 89 93
234 77 96 97
P value 0.0922
(42.5) (21.0) (17.6) (18.5)
(46.4) (15.2) (19.0) (19.2) 0.159
274 (54.4) 217 (43.1) 13 (2.6)
244 (48.4) 247 (49) 13 (2.6)
190 (38.7) 301 (61.3)
349 (71.1) 142 (28.9)
175 95 85 49 67 42 37 16 85 80 36 9 0 38
181 111 92 31 75 52 30 15 88 83 42 11 0 31
<0.0001a
(35.1) (18.9) (16.9) (9.7) (13.3) (8.3) (7.3) (3.2) (16.9) (15.9) (7.1) (1.8) (0) (7.5)
(35.9) (22.0) (18.3) (6.2) (14.9) (10.3) (6.0) (3.0) (17.5) (16.5) (8.3) (2.2) (0) (6.2)
0.7804 0.2114 0.5623 0.036a 0.4767 0.2787 0.3761 0.8552 0.8021 0.7975 0.4794 0.6515 d 0.3826
Abbreviations used: DTaP, diphtheria, tetanus, and acellular pertussis; Hib, Haemophilus influenzae type b; HepA, hepatitis A; HepB, hepatitis B; HPV, human papillomavirus; IPV, inactivated polio; MMR, measles, mumps, and rubella; MCV, meningococcal conjugate; PCV13, pneumococcal conjugate; PPSV23, pneumococcal polysaccharide; RV, rotavirus; Tdap, tetanus, diphtheria, and acellular pertussis; Td, tetanus and diphtheria; VAR, varicella. a Statistical significance. b P values are reported for each individual vaccine because patients could receive more than 1.
Table 2 Missed opportunities by vaccine Vaccine
There were 80 missed opportunities in the comparison clinic and 42 in the intervention clinic (missed opportunity rate ¼ 11.1% vs. 6.4%). There were significantly more patients with at least 1 missed opportunity in the comparison clinic than in the intervention clinic (10% vs. 6%; P ¼ 0.015; Table 2). When broken down by vaccine series, patients in the comparison clinic were significantly more likely to have a missed opportunity for DTaP, HepA, and PCV13/PPSV23 (Table 2). In multivariable assessment, clinic type, age of patient, and visit type were significantly associated with the odds of having at least 1 missed opportunity (Table 3). Provider type was not significantly associated with odds of missed opportunity and was, therefore, removed from the model. After adjusting for patient age and visit type, patients in the comparison clinic were still significantly more likely to have a missed opportunity (adjusted OR ¼ 2.14 [95% CI 1.3-3.5]). Regarding age, patients in the 7-10eyear age group were most likely to have a missed opportunity, followed by those in the 11-18-year age group. In addition, patients being seen for a sick visit were significantly
Comparison clinic, n (%)
Any DTaP Hib HepA HepB HPV MMR MCV PCV13/PPSV23 IPV RV Tdap Td VAR
Comparison clinic, n (%)
Intervention clinic, n (%)
P value
51 12 5 7 2 30 4 1 8 2 1 2 2 5
30 4 1 1 1 24 1 2 1 2 1 2 2 1
0.015a 0.0438a 0.1014 0.0332a >0.9999 0.4013 0.3738 >0.9999 0.0382a >0.9999 0.9999 0.9999 0.9999 0.2174
(10) (2) (1) (1) (0) (6) (1) (0) (2) (0) (0) (0) (2) (1)
(6) (1) (0) (0) (0) (5) (0) (0) (0) (0) (0) (0) (2) (0)
Abbreviations used: DTaP, diphtheria, tetanus, and acellular pertussis; Hib, Haemophilus influenzae type b; HepA, hepatitis A; HepB, hepatitis B; HPV, human papillomavirus; IPV, inactivated polio; MMR, measles, mumps, and rubella; MCV, meningococcal conjugate; PCV13, pneumococcal conjugate; PPSV23, pneumococcal polysaccharide; RV, rotavirus; Tdap, tetanus, diphtheria, and acellular pertussis; Td, tetanus and diphtheria; VAR, varicella. a Statistical significance.
SCIENCE AND PRACTICE Pharmacist impact on vaccines in the setting of CDS
Table 3 Adjusted odds of missed opportunity Variable Comparison clinic Intervention clinic 0e2 years old 3e6 years old 7e10 years old 11e18 years old Immunization visit Sick visit Well visit
% with MO
aOR
95% CI
P value
10.12 5.95 3.57 3.24 20.54 11.05 7.69 12.07 4.44
2.143 Reference Reference 0.673 5.65 2.75 1.827 2.467 Reference
1.309e3.509 d d 0.255e1.772 3.005e10.623 1.379e5.483 0.392e8.503 1.451e4.194 d
0.0025a d d 0.4224 <0.0001a 0.0041a 0.4425 0.0009a d
Abbreviations used: MO, missed opportunity; aOR, adjusted odds ratio. a Statistical significance.
more likely to have a missed opportunity compared with those being seen for a well visit (adjusted OR ¼ 2.5 [95% CI 1.5-4.2]). There were no statistically significant interaction effects between clinic type and patient age (P ¼ 0.5804), visit type (P ¼ 0.0632), or provider type (P ¼ 0.2367). Discussion A reduction in vaccine errors and missed opportunities has important positive outcomes, including improved public health, decreased preventable harm, reduced child discomfort, and potential cost savings. The use of CDS results in a low vaccination error rate; however, opportunities to vaccinate children may be missed because of the rigidity of CDS configuration. In a 3-month period, there were 175 vaccine administrations in the comparison clinic and 181 vaccine administrations in the intervention clinic. In addition, there were 51 encounters (10%) with at least 1 missed opportunity in the comparison clinic and 30 encounters (6%) with at least 1 missed opportunity in the intervention clinic. The significant difference in missed opportunities between the clinics remained after accounting for age of the patient, visit type, or provider type; even after accounting for these risk factors, patients in the comparison clinic had more than a 2-fold increased chance of having a missed opportunity. Although CDS is an aid in forecasting vaccine recommendations, the alerts can be overlooked easily by providers because of “alert fatigue,” especially when vaccines are not the primary reason for visit. In addition, the health system's CDS is not equipped to alert for high-risk populations, drug interactions, precautions, contraindications, and a 4-day grace period. Following categorization by vaccine series, patients in the comparison clinic were significantly more likely to have a missed opportunity for DTaP, HepA, and PCV13/PPSV23 compared with patients in the intervention clinic. These vaccines were also among those with the greatest number of missed opportunities overall. A factor that may have affected the number of missed opportunities in the DTaP series is the fourth dose of DTaP. The fourth dose should be given 6 months after the third dose, which is routinely given at 6 months of age. Therefore, administration of the fourth dose at 12 months of age is appropriate if there is a proper dosing interval from the previous dose; however, CDS does not alert for the fourth dose of DTaP until 15 months of age, creating the potential to miss an opportunity to vaccinate a child. HepA was added to the routine immunization schedule by CDC in 2006. It is
recommended to vaccinate any patient that did not receive HepA prior to 2006. Therefore, older patients are at risk for not being up to date for HepA. High-risk patient populations (e.g., sickle cell disease, chronic renal failure) were a factor affecting missed opportunities for the pneumococcal vaccines. There were high numbers of missed opportunities of the HPV vaccine in both clinics, although this was not statistically significantly associated with clinic type. In 2015, the health system introduced the recommendation to administer the first dose of HPV at 9 years of age. This alert was manually built into CDS and went live simultaneously in both clinics involved in the study in June 2015. Finally, it was determined in the study that the 7-10eyear age group had the highest likelihood of having a missed opportunity, followed by patients in the 11-18eyear age group. Historically, patients in the 7-10eyear age group were not scheduled for routine vaccinations, but they can often be candidates for catch-up vaccines. Because of the inconsistent need for vaccines in these age groups and the relatively new adjustment in recommended age for HPV initiation, patients aged 7-18 years are more vulnerable to missed opportunities. To our knowledge, this is the first study to assess the impact of pharmacists on vaccination errors and missed opportunities in the setting of CDS. The research question and findings of this study are an important safety consideration in the pediatric ambulatory arena, which is a modest but growing segment of pharmacy practice. It is evident that ambulatory pharmacists play a key role in improving vaccination rates with the aim of meeting the goals of Healthy People 2020.2,3 Limitations There are limitations to this study. First, it was a retrospective review that assessed vaccine orders; therefore, vaccination errors that occurred during administration were not assessed. Second, the review relied on documentation. Lack of documentation for a vaccine refusal was considered a missed opportunity. Although both clinics relied on the same vaccine schedules and primary care protocols, provider preference may have had an impact on results. Finally, staff adjustment to the change in recommended initial age of HPV may have influenced missed opportunities for this specific vaccine series. High-risk patients were included in population selection by randomization; however, it was not the primary focus of this research. Further research into the impact of pharmacist involvement on vaccination errors and missed opportunities in a setting focusing on high-risk patient populations is necessary; in particular, a larger sample size will be needed to have a large enough number of vaccination errors to facilitate multivariable analysis. The economic implication that is associated with pharmacist integration into a primary care clinic is another area that requires further investigation. Conclusion The results of this study indicate that pharmacist integration into a primary care clinic remains valuable despite the introduction of vaccine forecasting technology in the clinic setting. Although the use of CDS results in a low vaccination error rate, opportunities to vaccinate children may be missed 5
SCIENCE AND PRACTICE K.A. Wise et al. / Journal of the American Pharmacists Association xxx (2017) 1e6
because of the rigidity of the CDS configuration. Institutions are faced with the challenge of customizing technology to meet the needs of diverse patient populations, which requires additional resources. Pharmacist involvement in the primary care clinic appears to have a significant impact on reducing missed opportunities. Pharmacists are able to focus on vaccine opportunities, provide education and recommendations, and improve overall vaccination rates in the primary care setting. Acknowledgments We thank Han Yin, MS, Biostatistician and Yongjie Miao, Biostatistician for their contributions to the statistical analysis of the study. References 1. Centers for Disease Control and Prevention. Impact of vaccines in the 20th and 21st centuries; 2011. Available at: https://www.cdc.gov/vaccines/ pubs/pinkbook/downloads/appendices/e/impact.pdf. Accessed May 18, 2016. 2. U.S. Department of Health and Human Services. Healthy People 2020 topics and objectives: immunization and infectious diseases. Available at: https://www.healthypeople.gov/2020/topics-objectives/topic/immuniza tion-and-infectious-diseases. Accessed May 18, 2016. 3. U.S. Department of Health and Human Services. Healthy People 2020. Available at: http://www.cdc.gov/nchs/healthy_people/hp2020.htm. Accessed November 1, 2016. 4. U.S. Department of Health and Human Services. Recommended immunization schedules for persons aged 0 through 18 years; 2015. Available at: http:// www.cdc.gov/vaccines/schedules/hcp/child-adolescent.html. Accessed November 20, 2015. 5. Bundy DG, Shore AD, Morlock LL, et al. Pediatric vaccination errors: application of the “5 Rights” framework to a national error reporting database. Vaccine. 2009;27(29):3890e3896.
6
6. Weltermann BM, Markic M, Thielmann A, et al. Vaccination management and vaccination errors: a representative online survey among primary care physicians. PLoS One. 2014;9(8):e105199. 7. Rees P, Edwards A, Powell C, et al. Pediatric immunization-related safety incidents in primary care: a mixed methods analysis of a national database. Vaccine. 2015;33(32):3873e3880. 8. Hibbs BF, Moro PL, Lewis P, et al. Vaccination errors reported to the Vaccine Adverse Event Reporting System, (VAERS) United States, 20002013. Vaccine. 2015;33(28):3171e3178. 9. Taitel M, Cohen E, Duncan I, et al. Pharmacists as providers: targeting pneumococcal vaccinations to high-risk populations. Vaccine. 2011;29(45): 8073e8076. 10. Baroy J, Ching D, Frisch R, et al. The impact of pharmacist immunization programs on adult immunization rates: a systematic review and metaanalysis. J Am Pharm Assoc. 2016;56(4):418e426. 11. Haas-Gehres A, Sebastian S, Lamberjack K. Impact of pharmacist integration into a pediatric primary care clinic on vaccination errors: a retrospective review. J Am Pharm Assoc. 2014;54(4):415e418. 12. U.S. Department of Health and Human Services. Clinical Decision Support for Immunizations (CDSi). Available at: http://www.cdc.gov/ vaccines/programs/iis/interop-proj/downloads/cdsi-brochure.pdf. Accessed November 20, 2015. 13. Bundy DG, Persing NM, Solomon BS, et al. The ImmProve project: leveraging electronic health record data to promote immunization delivery. Acad Pediatr. 2013;13(5):458e465. Kelly A. Wise, PharmD, PGY-1 Pharmacy Resident, Nationwide Children's Hospital, Columbus, OH Sonya J. Sebastian, PharmD, BCACP, Clinical Pharmacist, Nationwide Children's Hospital, Columbus, OH Anna C. Haas-Gehres, PharmD, BCACP, Clinical Pharmacist, Nationwide Children's Hospital, and Assistant Professor, The Ohio State University, Columbus, OH Melissa D. Moore-Clingenpeel, MA, MAS, Biostatistics Core, Research Institute at Nationwide Children's Hospital, Columbus OH Kristen E. Lamberjack, PharmD, BCACP, Clinical Pharmacist, Nationwide Children's Hospital, Columbus, OH