Unintended pregnancy is endemic in the United States, with significant consequences for the woman, her family, and the community.1–3 Latest estimates indicate that 45% of all pregnancies in the United States are unintended.1 Reducing the proportion of pregnancies that are unintended and decreasing pregnancies resulting from contraceptive failure have been identified as two national health priorities by the Office of Disease Prevention and Health Promotion's Healthy People 2020 initiative.4 Contraception is highly effective at preventing unintended pregnancy, but barriers exist to effective and consistent use.5,6 Innovations in service delivery, including task sharing, are needed to improve access to and continuation of contraception, especially among vulnerable populations.1,7,8
Identifying cost-effective strategies to reduce unintended pregnancy is important to achieve national public health goals. Pharmacists in the community are increasingly relied on to expand their scope of practice to deliver clinical services such as providing vaccinations and prescribing naloxone. Pharmacist prescription of hormonal contraception has the potential to strengthen access to and enable easier continuation of contraception for women.
In 2016, Oregon became the first state to implement legislation allowing pharmacists to independently prescribe hormonal contraception, including the pill, patch, or ring, directly to the patient without a traditional clinic visit.9 Pharmacists complete a 5-hour training course to become certified to prescribe contraception, and follow evidence-based, clinical algorithms for prescribing and referring women.10 One year after passage of this legislation, 63% of ZIP codes across the state of Oregon now have a pharmacist certified to prescribe hormonal contraception.11 Among women in Oregon using combined hormonal contraception, 10% receive their prescription from a pharmacist.12 Oregon's Medicaid program reimburses for the cost of the contraception, and the pharmacists' time to prescribe. Efforts are underway to establish billing mechanisms with private payors.
Since the implementation of this policy in Oregon, five other states (California, Colorado, Hawaii, New Mexico, and Utah) have passed legislation to allow pharmacist prescription of hormonal contraception and implemented services.13 A sixth state, Washington, has previously had the possibility for pharmacists to prescribe under collaborative practice agreements, but it is not known to what extent, if any, the practice is currently occurring. Additional states are anticipated to follow (Maryland, Tennessee, and Washington, D.C.). The effect of these programs on unintended pregnancy and Medicaid cost are not known. We therefore sought to determine the cost effectiveness of pharmacist prescription of hormonal contraception by modeling unintended pregnancies averted by the policy.
We created a decision-analytic model using TreeAge Pro 2018. This model was designed to assess the cost-effectiveness of a Medicaid program of a policy expanding the scope of pharmacists to directly prescribe hormonal contraception (Fig. 1). Model inputs were obtained from Oregon Medicaid data and the literature.
Our primary outcome was unintended pregnancies averted. We calculated the number of unintended pregnancies averted by modeling unintended pregnancies under two different policy scenarios: with and without pharmacist prescribing of hormonal contraception. Secondary outcomes included costs and quality-adjusted life years (QALYs). Our study used a 1-year time horizon. Our study was reviewed and approved by the institutional review board at Oregon Health & Science University.
As shown in Figure 1, the initial decision node was the existence of a policy allowing pharmacist prescription of hormonal contraception (yes or no). Methods of contraceptives were collapsed into standard tiers based on contraceptive efficacy (tier 1: intrauterine device [IUD], implant; tier 2: pill, patch, ring, injection; tier 3: barrier methods; and no method). In our model, women used tier 1, tier 2, tier 3, or no method of contraception in rates based on the literature. Model inputs are summarized in Table 1. We accounted for method discontinuation and switching, assuming that women who switched or discontinued would do so at an average of 6 months. Women who experienced an unintended pregnancy could have an induced abortion, experience a spontaneous abortion, or continue the pregnancy to term.2,14 Women using a tier 2 method with access to pharmacist-prescribing had the option to return to their original provider for a prescription or visit a pharmacy to obtain the method. We assumed that all women who were continuing their contraceptive method and chose to visit a pharmacy would be able to receive her method during that single visit. Women not using a method of contraception with access to pharmacist-prescribing had the option of initiating contraception at the pharmacy or continuing with no method. If she was ineligible for hormonal methods available at the pharmacy, she was referred to her provider for a prescription. Alternatively, women without access to pharmacist-prescribing just visited their provider for contraception.
We modeled outcomes for the proportion of the Medicaid population at risk of unintended pregnancy in each state where pharmacists are eligible to prescribe hormonal contraception. The Guttmacher Institute provides estimates of the number of women in each state at risk for unintended pregnancy, and in need of publicly funded contraception.15 In Oregon, 198,110 women are in need of publicly funded family planning and at risk of unintended pregnancy.15 In the six other states (California, Colorado, Hawaii, New Mexico, Utah, and Washington) that have expanded the scope of pharmacists to directly prescribe hormonal contraception, an additional 3.3 million women are at risk of unintended pregnancy. We used estimates specific to each state to make these calculations.
All model inputs were derived from the literature (Table 1) or Oregon Medicaid data. A database was created of Medicaid claims from January 1, 2016, through December 31, 2017, representing the first 24 months of the policy change in Oregon. Contraceptives prescribed by pharmacists were identified by national provider identifier. This database was used to estimate the rate of women seeking contraception under the policy change. This database was also used to determine the probability that women seeking contraceptives from pharmacists were new contraceptive users or continuing users. A new contraceptive user was defined as a woman who had no contraceptive use in the 30 days before initiating hormonal contraception. All other probabilities were obtained from the literature (Table 1). The probability of current contraceptive use with a tier 1 or 2 method was obtained from reports from the Oregon Health Authority.16 We used typical-use contraceptive efficacy rates to account for pregnancies related to contraceptive failure.17–19 We identified the probability that a current contraceptive user would switch methods in the literature.20,21 We assumed women switching methods would select a method of the same or higher efficacy.
Cost data specific to Oregon's Medicaid program was obtained from the literature and from Oregon's Fee for Service Reimbursement schedule (Table 1). To provide a conservative estimate of costs, only the costs of direct medical care were included. As Medicaid reimburses for the cost of the provider's time to provide counseling, as well as the contraceptive supplies, the cost for a counseling visit was included regardless of where a woman received her contraception. As clinicians are typically providing a range of services, the base estimate for contraceptive care is higher in a clinic than a pharmacy (Table 1; $81 vs $27). For clinic visits, we used the average of reimbursement for new and established level 3 and 4 visits. Pharmacists bill using Medication Therapy Service codes or a level 2 visit. We used the average of the reimbursement for these services. All costs were adjusted for inflation to 2018 dollars using the medical component of the Consumer Price Index.22 Cost data from Oregon was used to estimate costs for other states with pharmacist prescription of contraception.
Quality-adjusted life years are a standard measure used in decision and cost-effectiveness analyses to assess the effect of a wide range of health outcomes on quality of life. Quality-adjusted life years are the product of both life expectancy and utility. Utility is a measure of satisfaction or value for a particular health state. By convention, utility values range from 0 to 1, with 0 representing death and 1 indicating perfect health. The utility for an unintended pregnancy was obtained from the literature.23 The length of a woman's life after a pregnancy was calculated by subtracting the average age at first child birth from the current estimated life expectancy of an American women.24
We calculated total costs and QALYs to determine the incremental cost effectiveness of the existence of a pharmacist-prescribing policy. The cost-effectiveness threshold was set at a standard $100,000 per QALY.25 We calculated incremental cost-effectiveness ratios that compared a policy expanding the scope of pharmacists to prescribe hormonal contraception with no policy. Incremental cost-effectiveness ratio is a measure of value that compares the differences between the costs and health outcomes of two competing interventions.26
Univariate and multivariate sensitivity analyses were performed to test how varying one or more model parameters may affect the results. Univariate sensitivity analysis was performed on all inputs. We varied each input from 50% to 200% of base estimates to identify whether any threshold values existed. A threshold value marks the point at which a change in a variable would alter the model's conclusion.27 A tornado diagram was made to determine which variables had the greatest effect on the model when varied individually between their extreme values (Appendix 1, available online at http://links.lww.com/AOG/B388). We also performed bivariate sensitivity analysis on variables with threshold values and other key inputs, including the probability of accessing contraception in pharmacies and contraceptive continuation rates.
Finally, we performed a Monte Carlo simulation using 10,000 trials to evaluate how simultaneous multivariable changes would affect outcomes. Beta distributions were used for probabilities and gamma distributions for costs. The Monte Carlo simulation enabled variation of all probability estimates simultaneously by sampling distributions around the baseline estimate. We used scatter plots to represent uncertainty in results, and a 95% CI was generated.
Over the first 2 years of the program (January 2016–December 2017), 248 pharmacists wrote 1,313 prescriptions for 367 women in the Oregon Medicaid program. Within this time period, a total of 3,614 women received hormonal contraceptive prescriptions from all providers.12 Among Oregon's Medicaid population at risk for unintended pregnancy, the policy to expand the scope of pharmacists to prescribe hormonal contraception averted an estimated 51 unintended pregnancies and saved $1.6 million dollars in the first 2 years. Women's quality of life was also improved, with 158 QALYs gained per 198,110 women. Pharmacist prescription of hormonal contraception is a dominant strategy: it improved health outcomes and reduced costs (Table 2).
We identified the model inputs that varied the most and subjected those to univariate and bivariate sensitivity analyses. The model inputs that had the most influence on outcomes when varied were: contraceptive continuation rates, the costs of pharmacist time for contraceptive counseling and costs of a provider visit. We examined how differences in contraceptive continuation rates between pharmacists and clinicians affected our findings. If fewer than 46% of women continue the contraception at 1 year (64% discontinue), it is no longer cost effective at a willingness-to-pay threshold of $100,000 per QALY. Similarly, if contraceptive continuation rates among women receiving care from a pharmacist is 10% less than women receiving care from a clinician, pharmacist prescription of hormonal contraception will not avert unintended pregnancies. For the baseline strategy of no access to pharmacist prescription of hormonal contraception to be the preferred strategy, the cost of a pharmacy visit would need to exceed $100 and the cost of a clinic visit could only be $60 (Fig. 2). The average range for a pharmacy visit is $10–52, and clinic visits range considerably.28
One-way sensitivity analysis indicated that across all other probabilities, pharmacist prescription of hormonal contraception prevents unintended pregnancies and reduces costs. We examined how the percentage of women relying on IUDs and implants might affect study findings. Pharmacist prescription of hormonal contraception is cost effective regardless of the probability that current tier 1 or 2 users select an IUD or implant. A 132% increase in tier 1 contraceptive method use among current nonusers of contraception would be needed for pharmacist prescription of hormonal contraception to not be cost effective.
During multivariate sensitivity analysis, we found that our findings were robust: regardless of how change was introduced across the distributions of variables, pharmacist prescription of hormonal contraception was the preferred (dominant) strategy 100% of the time using a willingness-to-pay threshold of $100,000 per QALY (Fig. 3).
We then considered the effect on the six additional states where the policy is currently in place: California, Colorado, Hawaii, New Mexico, Utah, and Washington. We estimated the female population enrolled in Medicaid in each state, and at risk for unintended pregnancy. Data from the Guttmacher Institute was used to estimate the population of women in each state in need of publicly funded family planning services.29 Assuming these states achieve the same level of implementation as Oregon (4% of nonusers initiate care), 862 unintended pregnancies would be averted, with a gain of 66,366 QALYs, and $26.9 million saved (Table 3).
Our study provides an estimate of the cost-effectiveness of pharmacist prescription of hormonal contraception. Our findings suggest a meaningful effect of pharmacist prescription of hormonal contraception on unintended pregnancy rates and associated costs. However, we believe our findings to be conservative given that our model was based on use 24 months after implementation. We expect over time that knowledge of and use of contraceptive access from pharmacists will increase. If the policy continues to decrease the rate of contraceptive nonusers, an even greater policy effect will be realized. It is anticipated that as pharmacies contract with additional insurers for coverage, the practice will increase.
Although reaching nonusers is one important means by which pharmacist prescription of hormonal contraception may reduce unintended pregnancies and associated public costs, we also need the effect of the policy on contraceptive continuation rates. We identified a key threshold value in our model; if contraceptive continuation rates among women receiving care from pharmacists are lower by 10% than continuation rates among women accessing contraception from a clinician, the policy is not cost-effective. Multiple factors may affect contraceptive continuation rates including: concern about side effects, supply dispensed, cost, and access issues.30,31 Oregon has passed legislation requiring insurers to cover a 12-month supply of contraceptives be dispensed in an effort to try to improve contraceptive continuation rates. No one knows whether this policy has been successfully implemented. The Oregon Board of Pharmacy's algorithm for pharmacist prescribing of contraception helps to support this policy, because it educates and advises pharmacists to prescribe up to 12 months’ supply.32 Oregon law requires insurers to cover a 12-month supply. It is possible that pharmacists may improve contraceptive continuation rates by decreasing barriers to contraceptive access (eg, no appointment required, extended hours) or by dispensing larger supplies of contraceptives. It is not known whether there is a difference in supply dispensed by prescriber type. Data on the effect of policy changes expanding the scope of pharmacists to provide and bill for influenza vaccinations has demonstrated improved receipt of recommended vaccines in a range of settings.33 Research to study the effect of pharmacist prescription of hormonal contraception on contraceptive continuation rates, via prospective data collection or health systems research, is needed.
Our study has limitations. As with all decision models, outcomes are dependent on the accuracy and availability of model inputs. We performed sensitivity analysis widely around all variables to address this limitation. For example, we tested our model assumptions, such as the referral rate from pharmacies or new users reached by the policy, and how they would affect our outcomes. Across all reasonable ranges for variables, our findings remained robust. We did not systematically examine how differences in counseling by provider type may inform results. No data exists to guide this estimate. We did, however, conduct sensitivity analyses to determine how different health outcomes would need to be by provider type to affect population health. We noted that, if long-acting reversible contraceptive (LARC) use increases by 132% among current nonusers of contraception, pharmacist prescription of hormonal contraception would not be cost effective. Nationally, LARC rates have increased markedly over the past decade, from 6% to 14% of all contraceptive users.34 The increase in LARC usage parallels a decrease in sterilization: the increase in LARC use is thought to be the result of current users of contraception switching methods. In Oregon, LARC rates have increased. We do not have available data on what proportion of nonusers of contraception are initiating LARC, but it is believed that a 132% increase in LARC in this population is unlikely.
Oregon-specific findings may not be generalizable to other states. Oregon is unique in that Medicaid has reimbursed for pharmacist time and contraceptives since policy inception. Pharmacists have rapidly become certified to prescribe contraception in Oregon.11 Ensuring equitable reimbursement for services provided by pharmacists has been key to the uptake of the practice. Data from California, where Medicaid has not reimbursed for pharmacists’ counseling and time, demonstrates low availability of services when reimbursement is not assured.35 Our perspective is that of Medicaid costs—we did not include in the model the costs for pharmacies to implement this service. Medicaid reimbursements vary by state; if states reimburse less for clinic visits than in Oregon, it is possible pharmacist provision of contraception would not be cost effective.
Unintended pregnancy remains an entrenched public health problem in the United States, with multigenerational consequences.7 Pharmacist prescription of hormonal contraception has the potential to improve contraceptive use and continuation rates nationwide. States appear interested in this approach with a rapid increase in legislation. As services are expanded, research is needed to both monitor the implementation and quality of care, as well as to identify the public health effect.
1. Finer LB, Henshak S. Disparities in rates of unintended pregnancy in the United States, 1994 and 2001. Perspect Sex Reprod Health 2006;38:90–6.
2. Finer LB, Zolna MR. Shifts in intended and unintended pregnancies in the United States, 2001–2008. Am J Public Health 2014;104(suppl 1):843–52.
3. Sonfield A, Kost K, Gold R, Finer LB. The public costs of births resulting from unintended pregnancies: national and state-level estimates. Perspect Sex Reprod Health 2011;43:94–102.
4. Potter J, McKinnon S, Hopkins K, Amastae J, Shedlin M, Powers D. Continuation of prescribed compared with over-the-counter oral contraceptives. Obstet Gynecol 2011;117:551–7.
5. Grossman D, Fuentes L. Over-the-counter access to oral contraceptives as a reproductive healthcare strategy. Curr Opin Obstet Gynecol 2013;25:500–5.
6. Centers of Disease Control and Prevention. Achievements in public health, 1900–1999: family planning. JAMA 2000;283:326–7.
7. Dehlendorf C, Rodriguez MI, Levy K, Borrero S, Steinauer J. Disparities in family planning. Am J Obstet Gynecol 2010;202:214–20.
8. Access to contraception. Committee Opinion No. 615. American College of Obstetricians and Gynecologists. Obstet Gynecol 2015;125:250–5.
9. Rodriquez M, 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.
10. Rodriguez MI, Anderson L, Edelman AB. Prescription of hormonal contraception by pharmacists in Oregon: implementation of house bill 2879. Obstet Gynecol 2016;128:168–70.
11. Rodriguez MI, Biel FM, Swartz JJ, Anderson L, Edelman AB. Pharmacists experience with prescribing hormonal contraception in Oregon. J Am Pharm Assoc (2003) 2018;58:608–13.
12. Anderson L, Hartung D, Middleton L, Rodriguez MI. Pharmacist provision of hormonal contraception in the Oregon Medicaid Population. Obstet Gynecol 2019;133:1231–7.
13. National Alliance of State Pharmacy Associations. Pharmacists allowed to prescribe birth control in more states. 2017. Available at: https://naspa.us/2017/05/pharmacists-authorized-prescribe-birth-control-states/
. Retreived March 12, 2019.
14. Flink-Bochacki R, Meyn LA, Chen BA, Achilles SL, Chang JC, Borrero S. Examining intendedness among pregnancies ending in spontaneous abortion. Contraception 2017;96:111–17.
16. Oregon Health Authority. Oregon's health system transformation: CCO metrics 2015 final report. Salem (OR): Oregon Health Authority; 2016.
17. Trussell J, Hatcher R. Contraceptive technology. 21st ed. Atlanta (GA): Managing Contraception LLC; 2018.
18. Kost K, Singh S, Vaughan B, Trussell J, Bankole A. Estimates of contraceptive failure from the 2002 national survey of family growth. Contraception 2008;77:10–21.
19. Hatcher RA. Contraceptive technology. 21st rev. ed. New York (NY): Ayer Company; 2018. p. 97–8.
20. Trussell BJ, Vaughan B. Failure, discontinuation and resumption of use: results from the 1995 National of Family Growth Survey. Fam Plann Perspect 2015;31:64–72.
21. O'Neil ME, Peipert JF, Zhao Q, Madden T, Secura G. Twenty-four-month continuation of reversible contraception. Obstet Gynecol 2013;122:1083–91.
22. Consumer Price Index for all urban consumers: medical care (CPIMEDSL). Available at: https://fred.stlouisfed.org/series/CPIMEDSL
. Retrieved December 3, 2018.
23. Burlone S, Edelman AB, Caughey AB, Trussell J, Dantas S, Rodriguez MI. Extending contraceptive coverage under the Affordable Care Act saves public funds. Contraception 2013;87:143–8.
24. Martin J, Hamilton B, Osterman M, Driscoll A, Drake P. Births: final data for 2016. Natl Vital Stat Rep 2018;67:1–55.
25. Owens DK. Interpretation of cost-effectiveness analyses. J Gen Intern Med 1998;13:716–17.
26. Sanders G, Neumann P, Basu A, Al E. Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: second panel on cost-effectiveness in health and medicine. JAMA 2016;316:1093–103.
27. Lilford R, Pauker S, Braunholtz D, Chard J. Decision analysis and the implementation of research findings. BMJ 1998;317:405–9.
28. Oregon Health Authority. Fee schedule reimbursement. Salem (OR): Oregon Health Authority. Available at: https://www.oregon.gov/oha/hsd/ohp/pages/fee-schedule.aspx
. Retrieved February 19, 2019.
29. Guttmacher Institute. Data center. Available at: https://data.guttmacher.org/states
. Retrieved January 1, 2019.
30. Westhoff CL, Heartwell S, Sharon E, Zieman M, Stuart G, Cwiak C, et al. Oral contraceptive discontinuation: do side effects matter? Am J Obstet Gynecol 2007;196:412.e1–6.
31. Foster DG, Hulett D, Bradsberry M, Darney P, Policar M. Number of oral contraceptive pill packages dispensed and subsequent unintended pregnancies. Obstet Gynecol 2011;117:566–72.
32. Oregon Board of Pharmacy. Standard procedures algorithm for prescribing of contraceptives. Available at: https://www.oregon.gov/pharmacy/Imports/ContraceptivePrescribing/OregonStandardProceduresAlgorithmforRPhPrescribing12.17.pdf
. Retrieved February 7, 2018.
33. Isenor JE, Edwards NT, Alia TA, Slayter KL, MacDougall DM, McNeil SA, et al. Impact of pharmacists as immunizers on vaccination rates: a systematic review and meta-analysis. Vaccine 2016;34:5708–23.
34. Kavanaugh ML, Jerman J. Contraceptive method use in the United States: trends and characteristics between 2008, 2012 and 2014. Contraception 2018;97:14–21.
35. Gomez AM. Availability of pharmacist-prescribed contraception in California, 2017. JAMA 2017;318:2253–4.