Pregnancy screening and contraceptive counseling are essential components of treating female patients with psychiatric illness to avoid placing the mother or fetus at risk if pregnancy occurs during treatment. The primary objective of this study was to evaluate current practices for pregnancy and contraception screening in female veteran patients prescribed psychotropic medications at a Veterans Affairs/Department of Defense (VA/DoD) facility.

According to statistics reported by the American College of Obstetricians and Gynecologists (ACOG), approximately 500,000 pregnancies in the United States each year involve women with psychiatric illnesses. It is also estimated that “one third of all pregnant women are exposed to a psychotropic medication at some point during pregnancy.”1 These statistics are concerning, especially considering that there is little known about the human fetal effects of several psychotropic medications, particularly those that are categorized as pregnancy category B or C. In addition, some psychotropic medications are classified as pregnancy category D for which human fetal risk has been demonstrated, and the benefits must outweigh the risk to warrant use in pregnancy. Furthermore, a select number of psychotropic medications are labeled category X for which risk of use in pregnancy clearly outweighs the benefit due to positive evidence of human fetal abnormalities.2 Regardless of pregnancy category, the decision to use a psychiatric medication during pregnancy should involve an extensive discussion between patient, provider, and pharmacist and include a detailed evaluation of risk versus benefit.

The assessment of risk versus benefit to the mother and fetus is partially based on what is known about the specific medication(s) involved. All psychotropic medications carry a risk of exposure to the fetus as they are known to cross the placenta and are present in amniotic fluid.1 The effects of several psychotropic medications in pregnancy have been studied, primarily using retrospective databases and/or pregnancy registries.3–6 Please refer to the toolbox accompanying this issue for a summary of potential fetal and neonatal effects of specific psychotropic medications.

There are many possible risks to consider when deciding whether to utilize specific psychotropic agents during pregnancy. Despite the risks from exposure to psychotropic agents, the risk of a maternal psychiatric illness left pharmacologically untreated is also significant. For example, discontinuation of lithium in a pregnant patient with severe bipolar disorder could lead to high risk of relapse. Lithium discontinuation could also increase risk of harm to the mother and fetus if mania and/or psychosis is present, potentially causing impaired judgment in the mother.7 The decision on whether psychotropic medication(s) should be continued during pregnancy for a patient with pre-existing psychiatric illness should be an individualized one. The decision should occur before the pregnancy and should be shared, involving a discussion between the patient, obstetrician, primary care provider, and mental health provider.1 Pharmacists should also be active participants in the discussion as they can help interpret pregnancy categories and identify fetal risks associated with specific medications. Although ideal, this shared decision-making does not always occur, and there remains the possibility for female patients to become pregnant while taking psychotropic medications. Therefore, pregnancy screening and contraceptive counseling are essential components of treating female patients with psychiatric illness to avoid putting the mother or fetus at risk if pregnancy occurs during treatment.

Study objectives and endpoints

The primary study objective was to evaluate current practices for pregnancy and contraception screening in female veteran patients prescribed psychotropic medications at a Veterans Affairs/Department of Defense (VA/DoD) facility. The secondary objective was to identify any deficiencies or areas for improvement in the current screening process at this facility. The primary study endpoints were the differences in percentage of female patients of child-bearing age screened for pregnancy and contraception (evaluated separately) in a mental health clinic versus a women's health clinic. In order to evaluate hypothesis 2 (see section below), the primary endpoints were also compared for patients enrolled in either clinic alone versus those enrolled in both clinics. The secondary endpoints of the study included what types of contraceptive methods were used and whether contraceptive counseling was documented for patients enrolled in either or both clinics.

Hypothesis 1. Female patients of child-bearing age enrolled in the women's health clinic and receiving psychotropic medications would be screened more frequently for pregnancy and contraception compared to those enrolled in the mental health clinic.

Hypothesis 2. Female patients of child-bearing age enrolled in both clinics (women's health and mental health) and receiving psychotropic medications would be screened more frequently for pregnancy and contraception compared to those enrolled in either clinic alone.

Study design

This retrospective study was Institutional Review Board (IRB)-approved and involved a chart review of female patients prescribed psychotropic medications who were also enrolled in the mental health and/or women's health clinic(s) at a VA/DoD facility. For the primary endpoints, each patient's chart was reviewed for documentation of pregnancy screening and use of contraception (either an active prescription for a contraceptive agent or chart documentation of a birth control method) at least once during the study period. Documentation of pregnancy screening was defined as completion of the pregnancy clinical reminder in the electronic chart or a laboratory pregnancy test. The pregnancy clinical reminder is an alert that appears in the electronic medical record when patients ages 16–50 are seen in clinic. This alert asks the provider to answer two “yes” or “no” questions as to whether the patient is pregnant and/or lactating, which completes the reminder and records it in the electronic chart.

For the secondary endpoints, each patient's chart was reviewed for what types of contraceptive methods were used (if any) and whether contraceptive counseling was documented during the study time frame. The patients were divided into three groups as follows: 1. Patients enrolled in the mental health clinic. 2. Patients enrolled in the women's health clinic. 3. Patients enrolled in both clinics. Enrollment in a clinic was defined as the presence of at least one documented visit to that clinic during the specified study time period. The primary and secondary endpoints were compared among the three groups to evaluate the current screening practices for pregnancy and contraception in this female patient population.

Inclusion criteria

Patients met inclusion criteria for the study if they were female veterans and of child-bearing age (between ages 18–50). During the study time frame, patients had to be prescribed one or more psychotropic medications in the following categories: benzodiazepines, non-benzodiazepine sedative-hypnotics, mood stabilizers, antidepressants, and/or antipsychotics. Patients had to be enrolled in the facility's mental health clinic, women's health clinic, or both.

Exclusion criteria

Patients enrolled in the facility's mental health clinic, women's health clinic, or both were also excluded if they had no documented visits to a clinic during the study time frame (due to no-show, cancelling appointments, etc.). A documented history of hysterectomy and/or tubal ligation also led to exclusion from the study.

Subjects

A list of female patients (age 50 or under) who filled an active prescription for psychotropic medications during the specified study time frame was obtained. Based on this list, patient charts were individually screened to determine which subjects met the inclusion and exclusion criteria.

Data collection procedures

Electronic charts for study subjects meeting the inclusion/exclusion criteria were reviewed for mental health and/or women's health clinic visits that took place between June 1, 2010 and May 31, 2011. Progress notes from visits with mental health and/or women's health providers were further reviewed for the following information: documentation of pregnancy screening, documentation of screening for contraceptive use, type of contraception (if used, classified as either oral contraceptive or alternative method), documentation of counseling/patient education regarding contraception, psychiatric diagnoses, psychotropic medication(s) prescribed, and other demographic data (age and race). In addition to the information collected from the electronic chart, pregnancy category of psychotropic medication(s) prescribed was also documented (see  Appendix A for list of psychotropic medications by pregnancy category).

Statistical analysis

Fisher's exact tests were utilized to compare differences in the primary and secondary endpoints among the three groups. The secondary endpoint, type of contraception used, was reported as percentages of participants in one of two categories: Category 1. Use of oral contraceptives or Category 2. Use of another method (e.g., condoms, intrauterine device, estrogen patch, intramuscular medroxyprogersterone). Due to the use of this dichotomous dependent variable, if a patient used more than one type of contraceptive method but was still taking a form of oral contraception, she was considered to be part of category 1. For example, if a female patient used both condoms and oral contraception, she was included in category 1. Descriptive statistics for each study group were reported for demographic data as follows: psychiatric diagnoses, psychotropic medications and their pregnancy category, age, and race.

Results

Out of 416 patients screened, 180 patients met inclusion criteria. Reasons for exclusion of 236 patients were as follows: hysterectomy and/or tubal ligation, absence of outpatient visits in women's health and/or mental health clinic(s), non-veteran status (active duty military or military dependent), medications prescribed for non-psychiatric indications (e.g. pain or migraine), or post-menopausal status. The breakdown of patients in each study group was as follows: 41 patients in group 1 (patients enrolled in the mental health clinic), and 33 patients in group 2 (patients enrolled in the women's health clinic), and 106 patients group 3 (patients enrolled in both clinics).

Primary Endpoints

For each study group, the following percentages of patients were screened for pregnancy via clinical reminder or pregnancy laboratory test: 90.2% of patients in the mental health clinic, 100% of patients in the women's health clinic, and 100% of patients enrolled in both clinics. The differences in pregnancy screening frequency were significantly different between the three study groups (p=0.003), indicating that clinic enrollment (independent variable) was significantly associated with pregnancy screening. The following percentages of patients were screened for contraception in each study group: 14.6% of patients in the mental health clinic, 63.6% of patients in the women's health clinic, and 81.1% of patients enrolled in both clinics. The differences in contraception screening frequency were significantly different between the three study groups (p<0.001), indicating that clinic enrollment was significantly associated with contraception screening (see Table 1).

Table 1:

Frequency and Percentage of Patients Screened for Pregnancy and Contraception

Frequency and Percentage of Patients Screened for Pregnancy and Contraception
Frequency and Percentage of Patients Screened for Pregnancy and Contraception

Secondary Endpoints

The following percentages of patients were counseled about contraception in each study group: 2.4% of patients in the mental health clinic, 6.1% of patients in the women's health clinic, and 10.4% of patients enrolled in both clinics (see Table 2). The frequency of contraceptive counseling was not significantly different among the three study groups (p=0.285). The other secondary endpoint was type of contraception used. Out of six patients in the mental health clinic screened for contraception, only one patient was using oral contraception (16.7%), while the five remaining patients were using non-oral forms of contraception (83.3%). Out of 21 patients in the women's health clinic screened for contraception, type of contraception was documented for only 15 patients. Of the 15 patients for whom type of contraception was known, three were using oral forms of contraception (20%), while 12 patients were using non-oral contraception (80%). Of the 86 patients enrolled in both clinics who were screened for contraception, type of contraception was documented for 69 of these patients. Of the 69 patients in both clinics for whom type of contraception was known, 30 patients were using oral contraception (43.5%), while 39 patients were using non-oral forms of contraception (56.5%). Of note, if there was a documented history of contraceptive use (either oral or non-oral) but there was no indication in the chart that a patient was currently using contraception, this patient was not included in the count. There was no significant difference in frequencies of patients using oral versus non-oral contraception between the three study groups (p=0.169). This result further indicates that clinic enrollment was not significantly associated with type of contraception used.

Table 2:

Frequency and Percentage of Patients Counseled on Contraception

Frequency and Percentage of Patients Counseled on Contraception
Frequency and Percentage of Patients Counseled on Contraception

Demographic Data

The mean ages for each study group were as follows: 34.7 years for patients in the mental health clinic, 38.2 years for patients in the women's health clinic, and 36.3 years for patients enrolled in both clinics. The demographic data for race and psychiatric diagnoses by study group are presented in Tables 3 and 4. Since the majority of patients were prescribed more than one psychotropic medication, pregnancy category was evaluated collectively for all 180 patients. A total of 440 psychotropic medications were prescribed to a total of 180 patients. The breakdown of pregnancy categories for psychotropic medications prescribed was as follows: 35 medications in pregnancy category B, 302 medications in pregnancy category C, 96 medications in pregnancy category D, and 7 medications in pregnancy category X. For Table 4, also note that patients could have more than one psychiatric diagnosis. The data for diagnosis was collected from psychiatry progress notes under Axis I diagnoses. If no psychiatry progress note was available, then the patient's problem list or another progress note (primary care or women's health) was utilized for data extraction.

Table 3:

Patient Race by Study Group

Patient Race by Study Group
Patient Race by Study Group
Table 4:

Psychiatric Diagnoses by Study Group

Psychiatric Diagnoses by Study Group
Psychiatric Diagnoses by Study Group

Interpretation of Results

The frequency of pregnancy screening was significantly different among the three study groups. However, it should be noted that despite these statistical differences, the percentage of patients screened was greater than 90 percent in all three groups. Therefore, it appears that both the mental health clinic and women's health clinic are doing a thorough job of asking patients if they are pregnant, likely due to the clinical reminder that alerts providers monthly when patients ages 16–50 are seen in clinic. The mental health clinic should aim for a 100% pregnancy screening rate since the other two study groups had full compliance. There was also a significant difference between study groups in frequency of patients screened for use of contraception. Patients enrolled in both clinics had the highest percentage of contraception screening (~81%), followed by patients enrolled in the women's health clinic alone (~64%) and those enrolled in the mental health clinic alone (~15%). The primary outcome results are congruent with the study hypothesis that female patients enrolled in the women's health clinic would be screened more frequently for pregnancy and contraception compared to those enrolled in the mental health clinic. The results for contraception screening are also congruent with the hypothesis that female patients enrolled in both clinics would be screened more frequently compared to those enrolled in either clinic alone.

The percentages for contraception screening are not ideal, considering that just over half of patients in the study enrolled in the women's health clinic were screened and less than a quarter of patients enrolled in the mental health clinic were screened. A suggestion for improving frequency of contraception screening in both clinics would be to revise the clinical reminder to include a question about current use of contraception with a required section for comments regarding type of contraception. In this way, the provider would not only have to ask whether the patient was utilizing a contraceptive method but also identify what type was being used to document in the comments section.

For the secondary outcome of contraceptive counseling, there was not a significant difference in frequency among the three study groups. However, all three study groups had low percentages of patients counseled on contraception with the highest being patients enrolled in both clinics at 10.4%. These results indicate that the mental health clinic and the women's health clinic are greatly lacking in the task of counseling patients on contraception. This finding is concerning, especially considering that the majority of prescribed psychotropic medications are labeled as pregnancy category C and pregnancy category D. Ideally, contraceptive counseling would involve a discussion between the patient and provider on appropriate use of contraception to avoid unplanned pregnancy while using psychotropic medications (if the patient does not desire to conceive at that time). For the purposes of this study, contraceptive counseling was considered any chart documentation that the provider discussed contraceptive options with a patient, educated her on the proper use of a contraceptive agent, and/or counseled her on the importance of using contraception during psychotropic medication use to avoid unplanned pregnancy and potential harm to the fetus. Overall, this counseling was being done infrequently in both the mental health and women's health clinics. In order to increase the frequency of contraceptive counseling, one option would be to revise the clinical reminder further to include a section for documentation of contraceptive counseling. The provider would be prompted to answer “Yes” or “No” as to whether they counseled the patient with an additional section of required comments describing briefly what was discussed between patient and provider.

For the secondary outcome, type of contraception, there were no differences between the three study groups for use of oral versus non-oral contraception. In each study group, there was a higher percentage of patients using non-oral contraception (e.g. Mirena, Nuvaring, DepoProvera) compared to oral birth control. However, the percentage of patients receiving oral contraception was not statistically different among the three study groups, indicating that clinic enrollment was not correlated with the type of contraception used.

Limitations

Study limitations include reliance on electronic chart documentation for review of primary and secondary endpoints. Data collection relied on the assumption that providers were accurately documenting pregnancy screening and contraceptive counseling if these practices in fact took place. In addition, if patients were receiving contraceptive counseling and/or contraception outside the VA/DoD facility, there was a greater likelihood that it was not documented in the electronic chart. Therefore, there was a possibility of underestimating pregnancy screening and contraceptive counseling practices due to lack of electronic chart documentation. Another limitation was the lack of precise definition of the term “contraceptive counseling” in the study protocol. Since contraceptive counseling was left as an ambiguous term, contraceptive counseling was limited to the scenarios as described above. However, contraceptive counseling did not include situations where the provider educated the patient on the pregnancy category and potential harm of their psychotropic medications if the provider did not specifically mention the importance of utilizing contraception while taking these medications. Therefore, the frequency of patient education on use of psychotropic medications in pregnancy could have been underestimated based on this unclear definition of “contraceptive counseling.” Future studies should clearly define the meaning of “counseling” and outline the scenarios that qualify as such counseling.

Another limitation of the current study was that the list of psychotropic medications evaluated was not comprehensive. For example, the current study did not evaluate psychostimulants such as methylphenidate and dextroamphetamine. In addition, the list of mood stabilizers only included FDA-approved medications commonly prescribed at the VA/DoD facility: lithium, valproic acid, carbamazepine, and lamotrigine. Other medications sometimes used for mood stabilization (e.g. gabapentin and topiramate) were not included in the study since they were not listed in the study protocol. Future studies may consider expanding the number of psychotropic medications assessed to include psychostimulants and medications prescribed off-label for mood stabilization.

The potential for type II error should also be considered with regard to the non-significant outcomes for the secondary endpoints, contraceptive counseling and type of contraception. The smaller sample size of the study could have contributed to a lack of statistical difference for these secondary outcomes.

Despite the limitations of this retrospective study, the results identify the need for a better screening tool for pregnancy and contraception at the VA/DoD facility, which could possibly be expanded to other facilities with similar patient populations. Since the clinical reminder was being used consistently, it should be continued but expanded to include questions about the following: patients' plans to become pregnant, use of contraception and method, and whether contraceptive counseling was completed. The clinical reminder should also be revised to include required comments following each “Yes” or “No” question such that the provider will have to document patient education and discussion. Use of this revised clinical reminder could promote more consistent screening practices for pregnancy and contraception at VA/DoD medical centers and other similar facilities.

1.
ACOG Practice Bulletin: Clinical management guidelines for obstetrician-gynecologists number 92, April 2008 (replaces practice bulletin number 87, November 2007). Use of psychiatric medications during pregnancy and lactation
.
Obstet Gynecol
.
2008
;
111
(
4
):
1001
20
. .
2.
Drugs.com
.
FDA pregnancy categories
.
Available at: http://www.drugs.com/pregnancy-categories.html. Accessed June 23, 2013
.
3.
Holmes
LB
,
Wyszynski
DF.
North American antiepileptic drug pregnancy registry
.
Epilepsia
.
2004
;
45
(
11
):
1465
. .
4.
Alwan
S
,
Reefhuis
J
,
Rasmussen
SA
,
Olney
RS
,
Friedman
JM.
Use of selective serotonin-reuptake inhibitors in pregnancy and the risk of birth defects
.
N Engl J Med
.
2007
;
356
(
26
):
2684
92
.
DOI: 10.1056/NEJMoa066584. PubMed PMID: 17596602
.
5.
Louik
C
,
Lin
AE
,
Werler
MM
,
Hernández-Díaz
S
,
Mitchell
AA.
First-trimester use of selective serotonin-reuptake inhibitors and the risk of birth defects
.
N Engl J Med
.
2007
;
356
(
26
):
2675
83
.
DOI: 10.1056/NEJMoa067407. PubMed PMID: 17596601
.
6.
Chambers
CD
,
Hernandez-Diaz
S
,
Van Marter
LJ
,
Werler
MM
,
Louik
C
,
Jones
KL
et al
.
Selective serotonin-reuptake inhibitors and risk of persistent pulmonary hypertension of the newborn
.
N Engl J Med
.
2006
;
354
(
6
):
579
87
.
DOI: 10.1056/NEJMoa052744. PubMed PMID: 16467545
.
7.
Altshuler
LL
,
Cohen
L
,
Szuba
M
,
Burt
VK
,
Gitlin
M
,
Mintz
J.
Pharmacologic management of psychiatric illness during pregnancy: dilemmas and guidelines
.
Am J Psychiat
1996
;
153
(
5
):
592
606
.
8.
U.S. Food and Drug Administration
.
Valproate anti-seizure products: Drug safety communication – Contraindicated for pregnant women for prevention of migraine headaches
. .

Appendix A: List of Psychotropic Medications in Pregnancy Categories B-X1