Grossman, Daniel MD; White, Kari MA, MPH; Hopkins, Kristine PhD; Amastae, Jon PhD; Shedlin, Michele PhD; Potter, Joseph E. PhD
Combined oral contraceptives are the most widely used method of family planning in the United States.1 In the 50 years since combined oral contraceptives were first approved for sale, they have been demonstrated to be very safe.2 However, certain conditions are considered contraindications to combined oral contraceptive use because they place the user at increased risk for complications such as myocardial infarction or stroke. The World Health Organization's (WHO) Medical Eligibility Criteria for Contraceptive Use lists contraindications to use, including hypertension, smoking while older than age 34, migraine with aura, and others.3,4
Little research has evaluated the prevalence of contraindications in the general population. One study estimated that approximately 16% of U.S. women of reproductive age had at least one contraindication to combined oral contraceptive use, and these conditions were more prevalent among older women and those with public insurance or uninsured women.5 Another study from Texas found that 39% of reproductive aged women had at least one contraindication to combined oral contraceptive use.6
In recent years, there has been growing interest in the possibility of moving oral contraceptives over the counter (OTC), although some have expressed concern about whether women can accurately self-screen for contraindications without a clinician. Two studies in Mexico, where combined oral contraceptives are available without a prescription, found that the prevalence of contraindications was not significantly higher among those obtaining pills without a prescription compared with those who obtained the method from a clinician.7,8 One study in the United States found that women seeking contraception could accurately use a checklist to identify contraindications.9 In another study among the general population, women also were found to accurately self-screen for contraindications, although 6% thought they were eligible for use when they were truly contraindicated, largely because of unrecognized hypertension.6
The objective of this study was to understand how well women self-screen for combined oral contraceptive contraindications when they obtain this method OTC in a real-life situation. We took advantage of a natural experiment in which U.S. women living along the Texas–Mexico border have the opportunity to obtain combined oral contraceptives in Mexican pharmacies without a prescription.10 We compared the estimated proportion of contraindications to combined oral contraceptives between women living in El Paso, Texas, who obtained combined oral contraceptives in U.S. public clinics, compared with women who obtained combined oral contraceptives OTC in Mexican pharmacies.
MATERIALS AND METHODS
Between December 2006 and February 2008, bilingual (English–Spanish) female interviewers recruited El Paso resident oral contraceptive pills (OCP) users into a study on women's access to OCPs on the United States–Mexico border. Women who were between ages 18 and 44 reported receiving their last pack of OCPs either from a family planning clinic in El Paso or at a pharmacy OTC in Ciudad Juarez, Mexico, and were willing to complete a series of four interviews were eligible to participate. The target population for this study was 500 El Paso clinic users and 500 women who accessed OCPs OTC at pharmacies in Mexico; at the end of enrollment, we recruited 532 clinic users and 514 OTC users. Interviewers recruited some of the clinic users from the major family planning providers in the El Paso area. The remaining proportion of the clinic user sample, as well as the entire OTC sample, was recruited using flyers, announcements, presentations at local community centers, and referrals from other participants. Our recruitment strategies reached women across the El Paso metropolitan area, with participants residing in 46 different zip codes; 16 of these zip codes each had 20 or more participants.
Women who agreed to participate in the study were asked to complete four interviews at 3-month intervals: a face-to-face baseline interview, telephone interviews at 3 and 6 months after baseline, and a final face-to-face interview approximately 9 months after the baseline interview. Additionally, blood pressure, height, and weight were measured at the final interview. Data collection was completed in December 2008, and 941 women completed the 9-month follow-up interview.
Interview data were collected using standardized questionnaires. At baseline, interviewers asked participants about basic sociodemographic information and their current OCP use, including pill brand. Participants were also read a list of medical conditions considered contraindications to combined oral contraceptive use and asked to report if they have ever had them. At each follow-up interview, participants were asked whether they had any new medical conditions develop that would be considered contraindications to combined oral contraceptive use; they were also asked to report their current contraceptive method. Because of the fact that a high proportion of participants reported migraine with aura at the baseline interview, during the final interview participants were asked a more detailed series of questions to identify migraine with aura symptoms.
At the final interview, interviewers measured and recorded participants' blood pressure using an automated Omron HEM 705CP blood pressure monitor (Omron Healthcare, Bannockburn, IL). Interviewers underwent training in the proper measurement of blood pressure with these monitors. Blood pressure was measured twice, and the mean systolic and diastolic measurements were recorded. In addition, participants' weight was measured using a Seca Clara 803 digital scale and their height was measured using a Seca Road Rod 214 portable stadiometer (Seca Corporation). We calculated women's body mass index (BMI, calculated as weight (kg)/[height (m)]2) using the measurement of participants' height and weight at the final interview. We used women's self-reported height and weight from the baseline interview to compute BMI for participants who were missing these data from the final interview (n=181); several participants (n=38) were missing both self-reported height and weight and final interview measurements. Participants were categorized as normal weight (less than 25.0), overweight (25.0–29.9), and obese (30.0 or more).
Participants were asked to report on WHO Medical Eligibility Criteria (third edition) relative and absolute contraindications (category 3 and category 4 contraindications, respectively; Table 1).3 The study was performed before the U.S. Medical Eligibility Criteria were released.11 All contraindications were based on participants' self-report from the baseline interview, with the exception of migraine with aura and hypertension; we assessed these two contraindications with composite measures from the baseline and final interviews. Women were considered contraindicated because of migraine headaches with aura symptoms if they met the criteria of using combined oral contraceptives at their final interview and reported migraine headaches (ie, headaches on one side of the head with symptoms of nausea or vomiting, sensitivity to light, or both and that interfere somewhat or a lot with normal activities) that were also accompanied by numbness, weakness, or difficulty seeing or a combination of these.12 If they did not complete a final interview (n=105) or if they were not using combined oral contraceptives at the final interview (n=145), then we used reported migraine headaches with numbness, weakness, or difficulty seeing at baseline (n=27). We also assessed when these headaches began relative to current pill use, based on the reported date of starting combined oral contraceptives and the month and year the headaches began. In this analysis, we did not include contraindications to continuing combined oral contraceptives according to the WHO Medical Eligibility Criteria, such as the development of migraine without aura after starting combined oral contraceptives (reported by five clinic users and one OTC user).
Women's blood pressure was classified according to the WHO Medical Eligibility Criteria based on the average of the two blood pressure measurements taken at their final interview. Women with category 3 contraindications also included those participants who completed a final interview, had a mean systolic blood pressure measurement less than 140 mm Hg and mean diastolic pressure less than 90 mm Hg, but who reported using antihypertensive medication (n=13). Those who did not complete the final interview, did not have their blood pressure measured at the final interview (n=54), or who were not using OCPs at the final interview were considered to have hypertension if they reported at baseline that they had hypertension or had high blood pressure when it was last measured (n=5).
Participants provided written informed consent before completing the initial interview. If a participant reported contraindications to OCP use during any of the interviews or if a mean systolic/diastolic blood pressure measurement of 140/90 mm Hg or greater was recorded at the final interview, then she received a referral to a healthcare provider. Women received gift cards valued between $10 and $35 for each of the interviews they completed. This study was approved by the Institutional Review Boards at the University of Texas at Austin and University of Texas at El Paso.
Questionnaire data were entered into an EpiData (EpiData Association, Odense, Denmark) database, and results were analyzed using Stata 10.0 (StataCorp LP, College Station, TX). Participants who were using progestin-only pills (n=17), who did not report a pill brand (n=13), or who were missing relevant sociodemographic information (n=1) were excluded from analysis; this resulted in a sample of 1,015 participants, 514 clinic and 501 OTC combined oral contraceptive users.
Sociodemographic characteristics and the estimated proportion of contraindications to combined oral contraceptive use were calculated for clinic and OTC users separately. The statistical significance of differences in proportions between these groups was determined using χ2 tests, and the Wilcoxon-Mann–Whitney test was used to compare medians. Crude and multivariable-adjusted logistic regressions were used to model the association between having at least one category 3 or 4 contraindication and a woman's source of combined oral contraceptives (clinic compared with OTC), age, parity, education, country in which she was born and completed her last year of schooling, U.S. health insurance coverage, and BMI. These variables were included to assess whether an association between contraindications and source of combined oral contraceptives was attributable to characteristics common among women who obtain their pills OTC in Mexico,10 correlates of access to health care, or identifiable risk factors for contraindications.
Based on our target of recruiting 500 clinic and 500 OTC users, a sample size that was based on another outcome, and a 28% prevalence of contraindications among current OCP users in this population,13 assuming a two-tailed type 1 error of 5%, we had 80% statistical power to detect an absolute difference of 8 percentage points in the estimated proportion of contraindications between clinic and OTC users.
Of the 1,015 residents of El Paso who were combined oral contraceptive users, 514 women had obtained their last pill pack before the baseline interview from a family planning clinic in El Paso, and 501 women had obtained their most recent pill pack OTC at a pharmacy in Ciudad Juarez, Mexico. The characteristics of participants by source of combined oral contraceptives at baseline are presented in Table 2. Over-the-counter users were older than clinic users and less likely to be nulliparous and have a high school education. They were also more likely to have been born in and completed their last year of schooling in Mexico. U.S. health insurance coverage was low for both groups and less common among OTC users (11.0% compared with 23.9% of clinic users, P<.001). Body mass index was high among participants in this sample, with more than 60% of women in both groups in the overweight or obese categories. Finally, 18.8% of OTC users and 14.2% of clinic users reported being current smokers (P=.047).
Among all combined oral contraceptive users, hypertension was the most common contraindication (7.7%), followed by smoking at age 35 or older (4.9%) and migraine with aura (4.1%); for each of the remaining contraindications, less than 2% of participants reported that they had or currently have the condition. Although the estimated proportion of contraindications was not significantly different for OTC and clinic users for the majority of the individual contraindications, OTC users were significantly more likely than clinic users to have category 3 hypertension or to be 35 years and older and smoke (Table 3).
Category 3 contraindications were reported by 11.2% of all combined oral contraceptive users in the sample, and category 4 contraindications were reported by 6.3%. The estimated proportion of category 3 and category 4 contraindications among OTC users was 13.4% and 7.4%, respectively, compared with 8.6% and 5.3% among clinic users. At least one category 3 or 4 contraindication was reported by 21.4% of OTC users and 13.8% of clinic users (P=.002). Whereas the estimated proportion of any category 3 contraindication was significantly different between the groups, the proportion of category 4 contraindications was not different.
In the crude logistic regression model, OTC users had significantly higher odds of having at least one category 3 or 4 contraindication to combined oral contraceptive use compared with clinic users (odds ratio 1.69, 95% confidence interval 1.22–2.36). Results were similar after multivariable adjustment (Table 4). In addition, women between ages 35 and 44 were significantly more likely to have contraindications relative to younger women, and those born and educated in Mexico were less likely to have contraindications compared with women born and educated in the United States. Women with a BMI 30 or more also had significantly higher odds of having a category 3 or 4 contraindication (odds ratio 2.24, 95% confidence interval 1.40–3.56). Parity, education, and U.S. health insurance coverage were not significantly associated with being contraindicated to combined oral contraceptive use.
We found a surprisingly high proportion of women in this population contraindicated to combined oral contraceptives, a finding we previously reported in El Paso.6 In addition to being obese, this population has limited access to health maintenance screening, which might contribute to the high prevalence of contraindications.6 Nineteen percent of women who had obtained pills from a clinic had a category 3 or 4 contraindication, suggesting that clinician screening is not perfect in this setting. It is also possible that in these cases, clinicians weighed the risks and benefits of all contraceptive options and determined that combined oral contraceptives were the best option. Alternatively, some contraindications, such as hypertension, might have developed after a woman was initially screened by her prescribing clinician. A study using data from the National Health and Nutrition Examination Survey found that approximately 6% of OCP users in the United States were contraindicated, although not all contraindications were identifiable in the database.5
The estimated proportion of contraindications was significantly higher among women obtaining combined oral contraceptives in Mexico without a prescription compared with clinic users. These results are contrary to the findings reported in studies of Mexican women that found no difference in contraindications between those who obtained pills from a clinician or at a pharmacy.7,8 It may be that some women in El Paso who were denied pills at a clinic because of a contraindication later obtained the method at a pharmacy in Mexico, possibly because it was the most effective contraceptive they could easily obtain. We are currently exploring this through additional in-depth interviews with hypertensive women in this sample.
In the multivariable-adjusted analysis, we found that in addition to obtaining pills OTC in Mexico, older age and obesity were the strongest predictors associated with having a category 3 or 4 contraindication. Although BMI 30 or more itself is not a contraindication to combined oral contraceptives, obese women are at risk for contraindicated comorbidities such as hypertension, hypercholesterolemia, and diabetes.3,4,11,14 In settings in which combined oral contraceptives are available without a prescription and blood pressure measurement is not easily accessible, BMI could be used as a proxy measure of contraindicated status. Future research should examine the accuracy and feasibility of a simple self-screening tool including BMI, as well as a more detailed checklist,6 in this context.
Our findings raise the question of whether the criteria for contraindications might be too strict. In this setting where access to other forms of contraception, such as depot medroxyprogesterone acetate, intrauterine devices, implants, and sterilization are limited, women with contraindications have few options. Given that unintended pregnancy also conveys risk among women with contraindications, using combined oral contraceptives may be the lesser evil for some women. The majority of studies that identified hypertension as a risk factor for myocardial infarction and stroke while using combined oral contraceptives were case-control studies and relied on participant report of hypertension rather than measured blood pressure before using pills.15,16 Even the few prospective cohort studies of combined oral contraceptive users did not report the blood pressure measurements of women considered to have hypertension who had complications develop.17,18 Given the increasing prevalence of blood pressure 140/90 mm Hg or higher in recent years,19 as well as the association between obesity and hypertension, it might be worth reconsidering the blood pressure cutoff for the category 3 contraindication.
Our study has several limitations. We used a convenience sample of combined oral contraceptive users, which may have introduced bias; however, it is important to note that no other sampling technique was possible with OTC users. Women selected their source of combined oral contraceptives, and there may be important differences in the two populations that were not controlled for here. It is also likely that we overestimated the proportion of women with hypertension, because some participants might have had “white coat” hypertension and blood pressure was measured on a single day. Screening for contraindications was not performed by a licensed clinician and was more like self-screening. This may have introduced some measurement error, although the self-screening questions had been previously validated and found to correlate well with a clinician's assessment.6 The diagnosis of migraine with aura is particularly difficult with a small number of screening questions. In addition, in the period since we collected data for this study, both the WHO and U.S. Medical Eligibility Criteria added several category 3 and 4 contraindications, such as a history of malabsorptive bariatric surgery or use of drugs like ritonavir-boosted protease inhibitors.4,11 We do not know how common these contraindications are in this population, although we suspect they are rare. Finally, our findings are from one specific population and may not be reflective of what would happen if combined oral contraceptives were made available OTC throughout the United States.
The focus of this study was on contraindications to combined oral contraceptives among current users. However, women's assessment of the risks of the pill also can be overestimated, creating a self-imposed barrier to use.13 More efforts are needed to educate women about the contraindications to this method as well as the safety and noncontraceptive benefits of combined oral contraceptives.
Regarding the possibility of OTC access to OCPs in the United States, our findings present something of a cautionary tale. Although the evidence is clear that women can accurately self-screen for contraindications using simple checklists,6,9 if they are not given such tools, as was the case here, some women will miss contraindications. It may be that the population that chooses the OTC option lacks access to medical screening and may be at higher risk for contraindications to begin with. However, this study is somewhat reassuring in that at least absolute contraindications are not more common among women obtaining combined oral contraceptives OTC. An actual-use study of OTC provision of combined oral contraceptives is needed to demonstrate whether self-screening for contraindications using a checklist on the label is accurate and whether women would actually use it. The checklist could be combined with a blood pressure assessment using an automatic kiosk on-site. Alternatively, progestin-only pills might be a better candidate for the first OTC oral contraceptive, given the much fewer, and rarer, contraindications for this method.
1. Mosher WD, Jones J. Use of contraception in the United States, 1982–2008. Vital Health Statistics 2010;(23):29.
2. Schwingl PJ, Ory HW, Visness CM. Estimates of the risk of cardiovascular death attributable to low-dose oral contraceptives in the United States. Am J Obstet Gynecol 1999;180:241–9.
3. World Health Organization. Medical eligibility criteria for contraceptive use. 3rd ed. Geneva: World Health Organization; 2004.
4. World Health Organization. Medical eligibility criteria for contraceptive use. 4th ed. Geneva: World Health Organization; 2009.
5. Shortridge E, Miller K. Contraindications to oral contraceptive use among women in the United States, 1999–2001. Contraception 2007;75:355–60.
6. Grossman D, Fernandez L, Hopkins K, Amastae J, Garcia SG, Potter JE. Accuracy of self-screening for contraindications to combined oral contraceptive use. Obstet Gynecol 2008;112:572–8.
7. Zavala AS, Perez-Gonzales M, Miller P, Welsh M, Wilkens LR, Potts M. Reproductive risks in a community-based distribution program of oral contraceptives, Matamoros, Mexico. Stud Fam Plann 1987;18:284–90.
8. Yeatman SE, Potter JE, Grossman DA. Over-the-counter access, changing WHO guidelines, and contraindicated oral contraceptive use in Mexico. Stud Fam Plann 2006;37:197–204.
9. Shotorbani S, Miller L, Blough DK, Gardner J. Agreement between women's and providers' assessment of hormonal contraceptive risk factors. Contraception 2006;73:501–6.
10. Potter JE, White K, Hopkins K, Amastae J, Grossman D. Clinic compared with over-the-counter access to oral contraception: choices women make in El Paso, Texas. Am J Public Health 2010;100:1130–6.
11. Centers for Disease Control and Prevention. U.S. medical eligibility criteria for contraceptive use. MMWR Morb Mortal Wkly Rep 2010;59.
12. Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders. Cephalalgia 2004;24(Suppl 1):1–150.
13. Grossman D, Fernández L, Hopkins K, Amastae J, Potter JE. Perceptions of the safety of oral contraceptives among a predominately Latina population in Texas. Contraception 2010;81:254–60.
14. Society of Family Planning, Higginbotham S. Contraceptive considerations in obese women: release date 1 September 2009, SFP Guideline 20091. Contraception 2009;80:583–90.
15. Gillum LA, Mamidipudi SK, Johnston SC. Ischemic stroke risk with oral contraceptives: a meta-analysis. JAMA 2000;284:72–8.
16. Khader YS, Rice J, John L, Abueita O. Oral contraceptive use and the risk of myocardial infarction: a meta-analysis. Contraception 2003;68:11–7.
17. Croft P, Hannaford PC. Risk factors for acute myocardial infarction in women: evidence from the Royal College of General Practitioners' oral contraception Study. BMJ 1989;298:165–8.
18. Hannaford PC, Croft PR, Kay CR. Oral contraception and stroke. Evidence from the Royal College of General Practitioners' Oral Contraception Study. Stroke 1994;25:935–42.
19. Grebla RC, Rodriguez CJ, Borrell LN, Pickering TG. Prevalence and determinants of isolated systolic hypertension among young adults: the 1999–2004 US National Health and Nutrition Examination Survey. J Hypertens 2010;28:15–23.