Effect of a multi-modal intervention on immunization rates in obstetrics and gynecology clinics




Background


There is increasing attention on immunizations by obstetrician-gynecologists and a need to improve vaccination rates for all women.


Objective


To evaluate the effect of a multimodal intervention on rates of immunization with tetanus, diphtheria, and acellular pertussis (Tdap); human papillomavirus (HPV); and influenza in outpatient obstetrics and gynecology clinics.


Study Design


Immunization rates at 2 clinics were compared pre- and post-implementation of multiple interventions at a public integrated health-care system. Study interventions began on June 6, 2012 and concluded on May 31, 2014; the preimplementation time period used was June 6, 2010 to June 5, 2012. Interventions included stocking of immunizations in clinics, revision and expansion of standing orders, creation of a reminder/recall program, identification of an immunization champion to give direct provider feedback, expansion of a payment assistance program, and staff education. All women aged 15 and older who made a clinic visit during influenza season were included in the influenza cohort; women who delivered an infant during the study time period and had at least 1 prenatal visit within 9 months preceding delivery were included in the Tdap cohort; each clinic visit by a nonpregnant woman aged 15–26 years was assessed and included in the HPV analysis as an eligible visit if the patient was lacking any of the 3 HPV vaccines in the series. The primary outcome was receipt of influenza and Tdap vaccine per current American College of Obstetricians and Gynecologists guidelines and receipt of HPV vaccine during eligible visits. Influenza and Tdap were assessed with overall coverage rates at the institutional level, and HPV was assessed at the visit level by captured opportunities. All analyses included generalized estimating equations and the primary outcome was assessed with time as a covariate in all models.


Results


A total of 19,409 observations were included in the influenza cohort (10,231 pre- and 9178 post-intervention), 2741 in the Tdap cohort (1248 pre- and 1493 post-intervention), and 12,443 in the HPV cohort (7966 pre- and 4477 post-intervention). Our population was largely Hispanic, English-speaking, and publicly insured. The rate of influenza vaccination increased from 35.4% pre-intervention to 46.0% post-intervention ( P < .001). The overall rate for Tdap vaccination increased from 87.6% pre-intervention to 94.5% post-intervention until the recommendation to vaccinate during each pregnancy was implemented (z = 4.58, P < .0001). The average Tdap up-to-date rate after that recommendation was 75.0% (z = -5.77, P < .0001). The overall rate of HPV vaccination with an eligible visit increased from 7.1% before to 23.7% after the intervention.


Conclusion


Using evidence-based practices largely established in other settings, our intervention was associated with increased rates of influenza, Tdap, and HPV vaccination in outpatient underserved obstetrics and gynecology clinics. Integrating such evidence-based practices into routine obstetrics and gynecology care could positively impact preventive health for many women.


There is increasing attention on immunizations in the obstetrics and gynecology setting. The American College of Obstetricians and Gynecologists (ACOG) has strengthened its efforts to improve immunization practices in the field and states that “obstetrician-gynecologists should embrace immunizations as an integral part of their women’s health care practice.” Pregnant women are a priority population for vaccines, and obstetrician-gynecologists (ob-gyns) are uniquely positioned to provide immunizations for all women. Approximately 20% of women making an ambulatory visit to an ob-gyn office identify their ob-gyn as their primary care provider, and ob-gyns provide more office-based well-woman care than any other type of provider. As such, ob-gyns play a vital role in the provision of immunizations as part of preventive care for women.


There is particular attention paid to 3 specific vaccines for young, healthy women: tetanus, diphtheria, and acellular pertussis (Tdap); human papillomavirus (HPV); and influenza. The Advisory Committee on Immunization Practices (ACIP) recommends the influenza vaccine for everyone aged 6 months and older as a strategy to prevent flu-related illness in all people, and pregnant women are a priority population. Pertussis incidence has significantly increased in recent years, with rates approaching the prevaccine era. Infants under 1 year of age are at the highest risk of pertussis morbidity and mortality. As a strategy to protect infants via passive immunization, both ACIP and ACOG now recommend that women receive a Tdap vaccine with each pregnancy. HPV is the most common sexually transmitted infection in the United States. In addition to the more than 1 million abnormal cervical cancer screening results that require evaluation, approximately 4000 women die from HPV-related cervical cancer annually. To reduce both benign and malignant HPV-associated disease, the HPV vaccine is recommended for all girls aged 11–12 with catch-up through age 26.


Adult vaccination rates remain low in the United States. The Centers for Disease Control and other authorities have called for improvements in adult vaccination to reduce the health consequences of vaccine-preventable diseases among adults and to prevent infectious morbidity and mortality in their infants. Best practices that have been shown to improve immunization rates include reminder/recall systems, efforts to remove administrative and financial barriers to vaccination, use of standing order programs for vaccination, and assessment of practice-level vaccination rates coupled with feedback to staff members.


The majority of data used to establish these best practices come from pediatric or inpatient adult populations, although investigators have used novel interventions such as text-messaging, automated reminders and computer alerts, and decision support to improve influenza, Tdap, and HPV vaccination in outpatient adult women. We aimed to increase the immunization rates of influenza, Tdap, and HPV in underserved outpatient ob-gyn clinics by using established evidence-based strategies in a multimodal intervention, and in this report we present our evaluation of the intervention.


Materials and Methods


Study setting


We performed a pre–post assessment of a multimodal intervention in 2 community-based ob-gyn clinics. Both clinics are part of Denver Health and Hospitals (DHH), an integrated safety-net health-care system with 8 community health centers and a central urban 500-bed public hospital, all using a common electronic medical record (EMR). An electronic clinical decision support tool including standing orders for immunizations was implemented in the DHH adult primary care clinics in 2007; however, this tool was not routinely used in DHH ob-gyn clinics until 2011. Obstetrics and gynecology care at our institution is provided by ob-gyn faculty and residents, family medicine faculty and residents, nurse practitioners, or certified nurse midwives at each health center and the hospital. Two women’s care clinics staffed primarily by certified nurse midwives with ob-gyn faculty oversight were included in this trial. These 2 clinics have approximately 7950 and 12,600 visits per year; the majority of patients are publicly insured or uninsured. This study was approved by the Colorado Multiple Institutional Review Board.


Study period


The study interventions began on June 6, 2012 and concluded on May 31, 2014. During the planning and implementation of the study, there were significant changes in ACIP and ACOG recommendations for Tdap vaccination in pregnant women. In March 2012, ACOG recommended the Tdap vaccine during pregnancy rather than immediately postpartum if not vaccinated previously (recommendation 1). This recommendation was approved by DHH and implemented in the study sites in June 2012. In June 2013, ACOG endorsed the ACIP updated guidelines to recommend that Tdap be administered during each pregnancy, irrespective of the patient’s prior Tdap vaccine history (recommendation 2). This change was approved and implemented at our institution in August 2013. No other institutional systematic changes pertinent to either clinic’s processes were implemented during the study time period. The 2 years preceding study interventions (June 6, 2010 through June 5, 2012) were used as the preintervention time period.


Study interventions


Study interventions are summarized in Table 1 . Most activities were aimed at improving standard processes for all immunizations in the clinics and were derived from evidence-based practices shown to increase vaccination rates. First, we provided education for nonprovider medical staff to raise awareness and empower them in their vital role in provision of immunizations. All medical staff at both clinics were required to attend 2 sessions, 1 on HPV and the other on Tdap in pregnancy. Standing orders for vaccines already in place were revised or expanded depending on the vaccine. The electronic clinical decision support tool is queried for each patient as they are triaged for a clinic visit for vaccine standing orders. Prior to the intervention, this tool would only provide a standing order for a vaccine if the vaccine was covered by insurance. This was revised to provide an order for all indicated vaccines regardless of insurance. The standing orders were also expanded to include influenza in the outpatient setting, as previous to the intervention there was only a standing order for influenza vaccine on the inpatient wards. Billing was not identified as an obstacle to immunization and was not addressed in this intervention. A single immunization champion was identified for both clinics. This was a registered nurse who performed periodic chart reviews and gave immediate feedback to providers when missed opportunities were identified. Additionally, at each clinic, a medical assistant who had the highest rates of vaccination prior to the intervention was given recognition at her clinic by receiving institutional awards. At the conclusion of the first year, 10 staff members were interviewed about the new processes and this midpoint qualitative feedback was used to improve and expand interventions.



Table 1

Interventions by immunization a















All Standing order revised and expanded
Clinic immunization champion identified
Immunization education sessions for nonprovider staff
Periodic chart reviews with immediate provider feedback
HPV Merck Assistance Program expanded
Reminder/recall program implemented
Standing order revised to include uninsured
Influenza Antenatal paper chart forms revised to prompt 3 refusals
Tdap Immunization stocked in clinic
Staff trainings regarding new recommendations
Antenatal paper chart forms revised to include standing order and prompt 3 refusals
Patient and family handouts created

HPV , human papillomavirus; Tdap , tetanus, diphtheria, and acellular pertussis.

Mazzoni et al. Immunization in ob-gyn clinics. Am J Obstet Gynecol 2016 .

a Interventions derived from Community Preventive Services Task Force.



The practice changes regarding Tdap required additional interventions. First, each clinic began stocking and administering Tdap, as this was the only of the 3 vaccines not available in ob-gyn clinics prior to the intervention. Additional staff trainings including providers were held with each recommendation change. Patient handouts were created and routinely given out at each first prenatal visit and at ultrasound visits. Although DHH uses an EMR, each obstetrics patient also has a paper chart, which is then scanned into the EMR. The antenatal paper chart forms were revised to include a standing order for Tdap, as the electronic support tool cannot recognize pregnancy and does not function in this setting. The revised form also provided a space for documentation of vaccine discussions and enforced the intervention requirement of 3 documented Tdap and/or influenza vaccine refusals among patients who declined vaccination.


Lack of insurance was identified as a barrier, especially for HPV vaccine; women without any form of insurance are required to pay the complete out-of-pocket cost for the vaccine, while women with insurance are not required to pay anything additional. Not only is HPV the most costly of the 3 vaccines, but nonpregnant women eligible for the vaccine are more likely to be uninsured compared to the pregnant population included in the Tdap and influenza cohorts, as pregnancy is always covered by some type of insurance in our state. Prior to the study interventions, a program through which uninsured patients can apply to receive the vaccine for free was available at only 1 of the clinics. This program was expanded to include both sites, the paperwork for the assistance program was streamlined, and, as mentioned above, the electronic standing order was revised to include uninsured patients. A reminder/recall program was also instituted to notify patients when second and third doses were due. Upon receipt of their first dose of HPV, patients were asked if they preferred to be contacted by telephone or mail and correct contact information was confirmed. The immunization champion nurse would then contact the patients up to 3 times when their next doses were due. It was not recorded how often the nurse would successfully contact the patient, or how many attempts at contact were made if fewer than 3.


Study population


Inclusion criteria were different for each immunization cohort. All women aged 15 years and older who made a visit to either clinic during influenza vaccination season (defined as August 1 through March 31) were included in the influenza cohort regardless of pregnancy. Women who delivered an infant from June 6, 2010 to June 5, 2014 at Denver Health Hospital and had at least 1 prenatal visit in either clinic within 9 months preceding delivery were included in the Tdap cohort. The HPV outcome was assessed per visit rather than per patient. As such, each clinic visit by a nonpregnant woman aged 15–26 years was assessed and included in the HPV analysis as an eligible visit if the patient was lacking any of the 3 HPV vaccines in the series.


Analytic methods


The primary outcome of our evaluation was receipt of Tdap and influenza immunization per current ACOG guidelines and HPV vaccine per eligible visit. All data were obtained retrospectively through use of the internal data warehouse. Of note, race/ethnicity and language are collected at registration based on patient self-report. Since patients could contribute data to the analysis on multiple occasions (eg, visits and pregnancies), all analyses included generalized estimating equations (GEE) to account for the within-subject correlation of repeated measures within individual patients. Immunization receipt was the primary dependent variable. For each model, we explored 3 approaches: (1) constant rates over time for each study period with a shift (increase/decrease) at the time of the intervention initiation, (2) spline models with different slopes for each period and a common intercept at the time the intervention began, and (3) a combination of different slopes for each period, along with a shift at the time of intervention initiation, an approach found to be useful in other studies. Using quasi-likelihood under the independence model criterion (QIC) scores, models were chosen based on best fit. The best-fitting model for Tdap was approach 1. The best-fitting model for HPV was approach 3.


For Tdap, the primary outcome was modeled at the individual level as a function of the period (preintervention, intervention following recommendation 1, and intervention following recommendation 2) and sociodemographic covariates. For HPV, the primary outcome was modeled at the visit level as a function of time and intervention period (preintervention slope, shift at time of intervention, and postintervention slope). For influenza, the primary outcome was modeled at the individual level, comparing the preintervention to the intervention period while adjusting for the sociodemographic covariates. Models for HPV and Tdap were tested for nonzero slope during each intervention period, as well as for shifts in slope at times of recommendation changes. Subsequently, for HPV we assessed whether the intervention significantly changed vaccination rates both with and without changing the slope found in the linear trend; a change in rates without change in the slope indicates that the effect is based on a shift rather than a change in slope. The same analyses were performed for Tdap, but since there were 2 recommendation changes these analyses compared the preintervention period both with the period between recommendations and after the second recommendation. For both Tdap and HPV, models assessing change in rate based on a shift were chosen based on better fit according to QIC scores. Time, patient age, race/ethnicity, payer, and language were included as covariates in all analyses. Demographics for each vaccine cohort before and after the intervention were compared using χ 2 or t test where appropriate. Significance was assessed at a P < .05 level. All analyses were performed using SAS Enterprise Guide software version 9.3 (SAS, Cary, NC).




Results


Study population


A total of 12,717 unique women were included in the influenza cohort, for a total of 19,409 observations (10,231 before and 9,178 after the intervention); 2650 women in the Tdap cohort, contributing 2741 observations (1248 before and 1493 after the intervention); and 4869 women in the HPV cohort, contributing 12,443 observations (7966 before and 4477 after the intervention). Patient demographics both before and after the intervention are presented in Table 2 . Our patients were largely Hispanic, English-speaking, and publicly insured. There were small differences in all demographics before and after the intervention, with an overall modest trend in increasing age and share of public insurance and a decreasing proportion of Hispanic ethnicity and Spanish-speaking patients.



Table 2

Demographics per vaccine cohort before and after the intervention













































































































































































Characteristic Influenza Tdap HPV
Before (n = 8354) After (n = 7525) P Before (n = 1244) After (n = 1466) P Before (n = 3626) After (n = 2334) P
Age 29.6 ± 10.5 30.3 ± 10.4 <.01 27.4 ± 6.0 27.0 ± 6.3 .09 21.9 ± 2.9 22.5 ± 2.6 <.01
Race/ethnicity
Hispanic 5842 (69.9) 5017 (66.7) 1018 (81.8) 1109 (75.7) 2577 (71.1) 1585 (67.9)
White, non-Hispanic 942 (11.3) 917 (12.2) <.01 92 (7.4) 149 (10.2) .01 361 (10.0) 233 (10.0) .01
Black 1325 (15.9) 1342 (17.8) 87 (7.0) 138 (9.4) 605 (16.7) 452 (19.4)
Other 245 (2.9) 249 (3.3) 47 (3.8) 70 (4.8) 83 (2.3) 64 (2.7)
Language
English 4964 (59.4) 4797 (63.8) 562 (45.2) 803 (54.8) 2494 (68.8) 1658 (71.0)
Spanish 3223 (38.6) 2552 (33.9) <.01 654 (52.6) 621 (42.4) <.01 1087 (30.0) 644 (27.6) .06
Other 167 (2.0) 176 (2.3) 28 (2.3) 42 (2.9) 45 (1.2) 32 (1.4)
Insurance
Public 5149 (61.6) 4865 (64.6) 757 (60.9) 961 (65.6) 2559 (71.7) 1694 (72.6)
Private 273 (3.3.) 338 (4.5) <.01 41 (3.3) 66 (4.5) .01 110 (3.0) 108 (4.6) .01
None 2932 (35.1) 2322 (30.9) 446 (35.9) 439 (30.0) 917 (25.3) 532 (22.8)

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

May 4, 2017 | Posted by in GYNECOLOGY | Comments Off on Effect of a multi-modal intervention on immunization rates in obstetrics and gynecology clinics

Full access? Get Clinical Tree

Get Clinical Tree app for offline access