Unintended pregnancy is an entrenched public health problem in the United States, with 45% of all pregnancies being unplanned.1 Unintended pregnancy is associated with health disparities and economic costs for the woman, her family, and society.2
Oral contraceptives (OC) are a popular birth control method in the United States, with nearly a third of all females relying on them for contraception.3 Ensuring convenient and timely access to OCs is important for contraceptive efficacy and continuation rates. Pharmacist prescription of OC is one strategy to improve access to contraception.
Graduating pharmacists are highly trained health professionals that practice in a variety of settings including hospital, ambulatory care, and community or retail pharmacies. Community pharmacies offer an alternative point of access for patients and are highly accessible with 90% of Americans living within 5 miles of a pharmacy.4 Most have more extended hours than a traditional ambulatory clinic and do not require an appointment. Previous research has shown that pharmacists can safely prescribe hormonal contraception in a community pharmacy setting, and that patients were satisfied with this type of service.5
On January 1, 2016, Oregon implemented a statewide expansion of the scope of pharmacists to directly prescribe short-acting methods of hormonal contraception without a visit to a clinician. Pharmacists in Oregon are required to undergo training, use a self-screening risk-assessment questionnaire and to follow a Standard Procedures Algorithm.6–8 The self-screening questionnaire and the Standard Procedures Algorithm for Oregon can be found in Appendices 1 and 2, available online at http://links.lww.com/AOG/B376.
We sought to understand who is being served by this policy change. We describe the demographics and utilization patterns of patients in Oregon's Medicaid program receiving a contraceptive prescription from a pharmacist in the first 2 years of program implementation.
We conducted a retrospective analysis of pharmacist-prescribed contraception in Oregon Medicaid patients within the first 2 years after implementation of the law. Our study objectives were to 1) characterize trends in pharmacist contraception prescribing over time from all claims within the study time period and 2) describe the population of patients receiving contraception from a pharmacist. We sought to examine whether patients accessing care from pharmacists were more likely to be new or continuing contraceptive users. We also summarized initial days' supply dispensed, whether patients with medical contraindications to estrogen were receiving appropriate therapy, and characteristics of participating pharmacies. The Institutional Review Board at Oregon Health & Science University approved this project and data use agreements were executed with the Oregon Health Authority.
Our analytic dataset was derived from Oregon Medicaid pharmacy claims, eligibility, medical diagnostic, and demographic data from January 1, 2016, to December 31, 2017. During the study time period, Oregon pharmacists were restricted to prescribing only oral and transdermal methods of hormonal contraception. Therefore, we included only contraceptive claims for products that pharmacists were eligible to prescribe, the combined hormonal pill and patch, and progestin-only pills. We determined claims for oral and transdermal contraceptive methods using prescription drug names and national drug codes. Although the focus of our analysis was on pharmacist prescribing, we also examined the number of new contraceptive prescriptions issued by nonpharmacists (eg, physicians, nurse practitioners) during the same period for context. We used descriptive statistics to report population characteristics within the claims. Results are presented as averages or percentages of qualifying characteristics of the overall claims or population.
We identified pharmacist-prescribed contraception as pharmacy claims where the prescribing provider had a National Provider Identifier from the list of pharmacists certified to prescribe contraception in the state. This list is maintained by the Oregon State University College of Pharmacy that has developed the training and certification program.9
We first analyzed monthly counts of contraception prescriptions prescribed by pharmacists during the study period. We further analyzed these trends by pharmacy type and location. We categorized pharmacies as retail chain, independent, or outpatient health care system-affiliated. We used the pharmacy's ZIP code to determine rural status according to the Oregon Office of Rural Health.10
For our second objective, to describe the population of patients accessing care from a pharmacist, we restricted our sample to patients who had 180 days of Medicaid enrollment before their initial (index) contraception prescription to ensure complete capture of antecedent utilization data. For these patients, we summarized demographic data including age, race–ethnicity, and urban or rural location. Medicaid and race–ethnicity are reported in approximately 60 categories, and members have the option to opt out of answering this question. We categorized race–ethnicity into three groups: white, non-white, and unknown. We categorized the contraceptive product for the index claim as combined hormonal transdermal, combined hormonal OC, or progestin-only OC. For each index fill, we summarized the intended duration of the first fill using the prescription days' supply.
New hormonal contraceptive users included any patient without a prescription for one in the prior 30 days. This could include patients who have never used this form of contraception, or those who used hormonal contraception only before our 30-day window. We used pharmacy claims data in the 30 days before the index claim to assess whether the woman was continuing contraception or whether the index claim was a new start. We defined contraceptive continuation to mean the woman had a prior contraceptive prescription that would have ended in the 30-day period before the index claim. We calculated end dates of prior contraceptive prescriptions by adding the days’ supply to the fill date of contraceptive claims dispensed in the prior 180 days, which would include refills on any existing prescriptions. When accounting for contraceptive coverage, we included all hormonal oral, transdermal, vaginal, and injectable contraceptive methods. As a sensitivity analysis, we also analyzed contraceptive prescriptions in the 180 days before their index claim, which is the time period of continuous Medicaid coverage for this population. We did not assess methods of contraception or conditions affecting pregnancy risk that are not found in pharmacy claims (barrier methods, long-acting reversible contraceptives, lactation amenorrhea, recent pregnancy, abortion, or miscarriage).
The Oregon pharmacist Standard Procedures Algorithm explicitly prohibits prescribing to individuals at risk for contraceptive-related adverse events (eg, coagulation disorder) or contraceptive failure (eg, seizure medications). We assessed compliance with the algorithm by analyzing International Classification of Diseases, 10th Revision (ICD-10) diagnostic codes present on medical claims in the 180 days before the index claim. Specifically, we characterized comorbidities provided by the US Medical Eligibility Criteria as a relative (category 3) or absolute (category 4) contraindication with combined OC.11 ICD-10 codes were used to identify medical claims for the following diagnoses, which are associated with increased risk of adverse events or failure of that method: bariatric surgery, breast cancer, cardiovascular disease (includes hypertension), coagulation disorder, diabetes (types 1 and 2), inflammatory conditions, liver disease, migraines with aura, sickle cell disease, stroke or transient ischemic attack, tobacco use, and 35 years or older, and pregnancy history or breastfeeding in the previous 6 weeks (from the index date). We also examined recent (within 30 days before the index date) prescriptions of drugs known to interact with contraceptives such as some antiseizure drugs and antiinfectives. Appendix 3 (available online at http://links.lww.com/AOG/B376) contains the full list of ICD-10 codes and interacting medications used.
Over the first 2 years of the program (January 2016–December 2017), a total of 3,614 patients in the Oregon Medicaid program received a new prescription for oral and transdermal methods from all providers. Within this time period, 162 pharmacists prescribed contraception for 367 patients, resulting in 1,313 fill claims. Among all patients using oral and transdermal hormonal contraception, 10% received their prescription from a pharmacist.
Trends and characteristics of pharmacist-prescribed contraception are shown in Figures 1 and 2. Within 4 months of implementation, pharmacist-prescribed contraception claims consistently totaled more than 40 claims per month among Medicaid enrollees. In the next 7 months, pharmacists filled an average of 61 contraceptive prescriptions per month as the prescriber. The highest number of claims in a month occurred in July and August of 2017, with 80 and 79 claims per month, respectively. The majority of claims originated from retail chain pharmacies (94%) in urban locations (71%).
Table 1 summarizes demographic and utilization data for the population of patients with 180 days of continuous enrollment preceding their index contraceptive prescription. This included 252 of the total 367 women who received their contraception from a pharmacist. The majority (82%) of patients were between 18 to 35 years, while only 7% were younger than 18 years. Because there is an option to opt out, race–ethnicity is unknown for approximately 38.5% of the population. Of the 155 (of 252) patients who reported race–ethnicity, 133 (86%) were white and 22 (14%) made up all other minority groups.
The most common method of contraception prescribed by pharmacists was the combined OC (90.5%), with only 5.6% being prescribed the progestin-only pill, and 3.2% prescribed the transdermal patch. Patients received an average of 65 days' supply of contraception on their index fill, with a range from 21 to 364 days. Most (59%) index claims were between 31 and 90 days' supply.
The majority of patients receiving a prescription from a pharmacist were new hormonal contraceptive users: 73.8% had no evidence of contraceptive coverage in the previous 30 days. When we increased the timeframe to look for contraceptive coverage before the index claim, 61.5% had not filled a contraception prescription that would end in the preceding 180 days.
Fewer than 5% (12) of patients had a diagnostic code indicating a possible contraindicating comorbidity. The most frequently identified conditions were cardiovascular disease (8), liver or gallbladder disease (5), and tobacco use in patients aged 35 or older (3), with some of the patients identified as having more than one of these conditions. There were no patients identified with a history of bariatric surgery, breast cancer, coagulation disorders, diabetes, inflammatory conditions, sickle cell disease, stroke or transient ischemic attack, or a pregnancy history or breastfeeding in the previous 6 weeks. Contraindicated drugs identified included two patients on lamotrigine (an antiseizure medication commonly used as a mood stabilizer). No patients were found to have interacting antiinfectives or cystic fibrosis drugs.
Identifying effective strategies to help reach new contraceptive users or improve method adherence is essential to meet national goals to reduce unintended pregnancy. Expanding the role for pharmacists to prescribe hormonal contraception is one strategy that has been implemented in Oregon. In the first 2 years of program implementation, we found evidence that pharmacists were safely reaching new contraceptive users.
Overall, 10% of new oral or transdermal hormonal contraceptive prescriptions were written by pharmacists. We found that, among Oregon's Medicaid enrollees, a majority (73.8%) of patients who received hormonal contraception from a pharmacist were new contraceptive users. Our finding held when we adopted a more conservative approach and examined contraceptive use over a longer time period. In the 180 days before receiving contraception from a pharmacist, 61.5% of patients had no contraceptive use, and were initiating hormonal contraceptive care in the pharmacy.
We also examined contraceptive safety by looking at whether patients with medical contraindications (Medical Eligibility Criteria Category 3 or 4) were receiving contraindicated methods. We found that overall adherence to the clinical algorithm for prescribing pharmacists was high. Only 12 (5%) patients were identified as having Medical Eligibility Criteria Category 3 or 4 medical conditions, and two (less than 1%) patients with medications contraindicating OC use received a prescription. This safety profile seen with pharmacists is on par with what is seen among clinicians prescribing contraception.12
Pharmacist prescription of contraception is a voluntary program for pharmacists that requires additional time and training. We found that the majority of prescriptions were written by pharmacists within retail chain pharmacies (94%) in urban locations (71%). In Oregon, 70% of community pharmacies are retail chain pharmacies and 61.8% are in urban locations (personal communication, F. Karbowicz, 2018). The higher percentage of retail chain pharmacies we found in our study may reflect encouragement from chain management to compensate pharmacists to become trained to offer these services. Oregon is unique in that Medicaid has reimbursed for both the cost of the contraceptive and the pharmacists’ time in counseling the patients since the program began.
Our study has limitations. As with all observational research, our analysis is limited by variables captured in our data source: claims to Medicaid. Lack of rich clinical data (eg, blood pressure, smoking status), input errors (eg, misclassification of diagnostic data), or omissions could have affected our results. Our sample size is relatively small. We focused our data on patients with an initial fill for a prescription of oral or transdermal contraception in the first 2 years after the policy started. The initial piece of legislation passed in Oregon only included oral and transdermal hormonal contraception as methods pharmacists could prescribe. This was amended to include the vaginal ring and injection in 2017 (with implementation in 2018) making it possible for pharmacists to prescribe all forms of short-acting hormonal contraception. As the program matures, and contracts with additional insurers are implemented at pharmacies, we expect the number of pharmacist prescriptions to increase. Importantly, we did not have access to the screening questionnaire or blood pressure readings obtained by pharmacists at the time of contraceptive prescription. All individuals receiving contraception from a pharmacist are required to complete a self-screening questionnaire with information on risk of current pregnancy, sexual activity, and other contraindicating comorbidities. We are unable to corroborate medical history self-reported by patients on the screening form with the comorbidities we captured from claims data. Future research should look at the safety of pharmacist-prescribed contraception with guidelines in the US Medical Eligibility Criteria as identified in the questionnaire, and compare this with prescribing patterns seen with other providers.
Our ability to characterize past contraceptive history may not be complete because we are not able to identify whether patients received contraception through Title X funding, those paid by other payers, or with cash. Additionally, because previous contraceptive use was identified with pharmacy claims, we do not know whether the patients in our population had a recent long-acting reversible contraceptive placed or removed, but we did gather information on recent pregnancy, along with other previous medical conditions. Our study was restricted to patients enrolled in Oregon's Medicaid, which affects the generalizability of our results.
Oregon has a relatively small population with significant demographic differences from other areas of the United States. We compared our study population with that seen in Oregon's Medicaid as a whole and noted a meaningful difference in race, ethnicity, or both. Reproductive-aged patients in Oregon who are Medicaid enrollees are 24.7% non-white.13 In contrast, among patients who received a pharmacist-written prescription for hormonal contraception, 14% were non-white. This observation may reflect our small sample size or be emblematic of known racial or ethnic disparities in contraceptive use.3
It is important to interpret our findings within the context that this represents claims from the first 2 years of policy implementation. Although pharmacists were overall accepting of the program, and a majority became certified, it took months for most pharmacies to be prepared to provide care14,15 (Figs. 1 and 2).
Ensuring equitable reimbursement for the counseling and services pharmacists provide is known to be essential to maintain contraceptive access in pharmacies.5,16 Oregon is unique among states that have expanded the role of pharmacists to include prescription of contraception. Medicaid Fee for Service has covered the costs of both the contraceptive drug and the pharmacists’ time for counseling since program inception. Oregon's Medicaid program is administered through 16 distinct coordinated care organizations with a minority of enrollees in open card, or Fee for Service (approximately 10%). Establishing the infrastructure and contracts for pharmacies to be able to bill and be reimbursed by payors is a lengthy and challenging process. Not all coordinated care organizations had established billing relationships with pharmacies during our study period. Oregon has been actively working to facilitate coordinated care organizations’ and private payors’ reimbursement of pharmacists’ time in line with other health care providers’ reimbursement for similar services. Future research should examine how increased participation by other payors affects utilization of pharmacies for contraceptive care. It is similarly important to understand who is accessing contraception from pharmacists, and why. An improved understanding of patients’ reasons for seeking care in pharmacies and their experience with the service may improve contraceptive care within pharmacies and clinics. Robust implementation research, qualitative and prospective cohort studies are needed to understand how these programs are being implemented, and to identify the effect they are having.
Pharmacist prescription of hormonal contraception has the potential to improve contraceptive utilization and continuation rates nationwide. Early data from Oregon demonstrates that pharmacists are reaching new contraceptive users who are at risk for unintended pregnancy.
1. Finer LB, Zolna MR. Declines in unintended pregnancy in the United States, 2008–2011. N Engl J Med 2016;374:843–52.
2. Dehlendorf C, Rodriguez MI, Levy K, Borrero S, Steinauer J. Disparities in family planning. Am J Obstet Gynecol 2010;202:212–13.
3. Jones J, Mosher W, Daniels K. Current contraceptive use in the United States, 2006–2010, and changes in pattern of use since 1995. Natl Health Stat Rep 2012;60:1–25.
4. National Association of Chain Drug Stores. Re: health care workshop, project no. P131207. Available at: http://www.nacds.org/ceo/2014/0508/supplemental_comments.pdf
. Retrieved February 18, 2019.
5. Gardner JS, Miller L, Downing DF, Le S, Blough D, Shotorbani S. Pharmacist prescribing of hormonal contraceptives: results of the Direct Access study. J Am Pharm Assoc 2008;48:212–21.
6. Oregon laws 2015. Oregon House Bill 2879. Available at: https://www.oregonlegislature.gov/bills_laws/lawsstatutes/2015orLaw0649.pdf
. Retrieved October 8, 2018.
7. Oregon laws 2017. Oregon House Bill 2527. Available at: https://olis.leg.state.or.us/liz/2017R1/Downloads/MeasureDocument/HB2527/Enrolled
. Retrieved October 8, 2018.
8. Oregon Board of Pharmacy rules related to pharmacist prescribing of contraceptives. OARs 855-019-0400 through 855-019-0435. Available at: https://secure.sos.state.or.us/oard/displayDivisionRules.action%3bJSESSIONID_OARD=fYcoTN30KnMGbPf9xZb5vZInE_8-NI5kVt1ivvDozP3XaAPUI-oZ%21479495115?selectedDivision=3967
. Retrieved October 8, 2018.
9. Oregon State University College of Pharmacy. Comprehensive contraceptive education for the Oregon pharmacist. Available at: https://pace.oregonstate.edu/catalog/comprehensive-contraceptive-education-oregon-pharmacist
. Retrieved October 8, 2018.
10. Oregon Office of Rural Health. ORH urban and rural areas. Available at: http://www.ohsu.edu/xd/outreach/oregon-ruralhealth/data/rural-definitions/upload/orh-rural-map.png
. Retrieved January 18, 2018.
11. Curtis KM, Tepper NK, Jatlaoui TC, Berry-Bibee E, Horton LG, Zapata LB, et al. U.S. medical eligibility criteria for contraceptive use, 2016. MMWR Recomm Rep 2016;65:1–103.
12. Grossman D, White K, Hopkins K, Amastae J, Shedlin M, Potter JE. Contraindications to combined oral contraceptives among over-the-counter compared with prescription users. Obstet Gynecol. March 2011;117:558–65.
13. Oregon Health Authority. Monthly Medicaid population report: gender/race-ethnicity by CCO/open card for physical plan type. Available at: https://www.oregon.gov/oha/HSD/OHP/DataReportsDocs/December%202017%20Physical%20Health%20Service%20Delivery%20by%20Gender%20and%20Race-Ethnicity.pdf
. Retrieved February 24, 2019.
14. Rodriguez MI, Biel FM, Swartz JJ, Anderson L, Edelman AB. Pharmacists' experience with prescribing hormonal contraception in Oregon. J Am Pharm Assoc 2018;58:608–613.
15. Rodriguez MI, McConnell K, Swartz J, Edelman AB. Pharmacist prescription of hormonal contraception in Oregon: Baseline knowledge and interest in provision. J Am Pharm Assoc. 2016;56:521–6.
16. Gomez AM. Availability of pharmacist-prescribed contraception in California, 2017. JAMA 2017;318:2253–4.