A Qualitative Analysis of Approaches to Contraceptive Counseling

Abstract

CONTEXT

Underuse and inconsistent use of contraceptives contribute to the continued high rate of unintended pregnancy in the United States. High-quality interaction between patient and provider about contraception is associated with improved contraceptive use, yet little is known about how providers support patients in the decision-making process.

METHODS

A random sample of 50 family planning visits by patients in the San Francisco Bay Area was selected from a larger sample of 342 audio-recorded visits to six clinics between 2009 and 2012. In qualitative analysis guided by grounded theory techniques, transcripts were assessed to determine counseling approaches and patterns in the use of these approaches.

RESULTS

Providers employed three counseling approaches: foreclosed (in 48% of visits), characterized by discussion of few contraceptive methods and method selection by the patient with no involvement from the provider; informed choice (30%), characterized by detailed description of multiple methods, but little or no interaction between the patient and the provider; and shared decision making (22%), characterized by the provider’s interactive and responsive participation with the patient in method selection. Use of these approaches varied by patient’s age: Women 25 or younger experienced the foreclosed approach more often than those in the two other age-groups, and patients older than 35 were far more likely than their younger counterparts to experience the shared decision-making approach.

CONCLUSIONS

Most visits did not include interactive engagement between the patient and the provider. Contraceptive counseling interventions should encourage providers to responsively engage with patients of all ages to better meet their contraceptive needs.

The high rate of unintended pregnancy in the United States¹ places a heavy burden on women, society and the health care system.² One factor contributing to this high rate is the underuse and inconsistent use of contraceptives. Not only do 11% of women at risk of unintended pregnancy not use any contraceptive method,³ but 25% of women using a reversible method rely on condoms, a method with relatively low efficacy.⁴ In addition, female contraceptive users report high rates of discontinuation and incorrect use of their chosen method.^5,6

Patient-provider communication during family planning visits is one influence on women’s contraceptive use. Health communication has been recognized as a marker of quality health care by the Institute of Medicine,⁷ and has been found to be correlated with patient outcomes, including patient satisfaction and medication adherence, across a range of health conditions.^8,9 In family planning care, patient assessment of the quality of interpersonal aspects of care has been linked to contraceptive use and continuation.^10–12

Despite this evidence of the importance of contraceptive counseling, little is known about what occurs during routine family planning counseling. Qualitative studies of U.S. women’s counseling experiences have found that women often report being dissatisfied with the counseling they receive,¹³ and that many are concerned that providers do not fully disclose the potential side effects of methods.¹⁴ Research that has directly documented the contraceptive counseling dynamic has been largely limited to studies in the developing world, and these studies have found that family planning counseling sessions are often dominated by the provider.^15–18 Even when the patient is the one ultimately making the decision about which method to use, there is often minimal provider engagement with patients to determine their preferences and concerns about methods and how these issues relate to available options.^19,20

Given the increasing recognition of the value of patient-centered care, in which there is a focus on patients’ preferences and experiences,²¹ interventions to enhance patient-provider engagement during contraceptive decision making could improve the counseling experience, which may positively influence contraceptive use.²² However, because it is unclear whether these findings are relevant in the United States, an investigation of contraceptive counseling would improve our understanding of the range of approaches used by U.S. family planning providers and potentially identify areas for improvement. Specifically, elucidating how providers guide conversations with women who are selecting a method could inform interventions to improve providers’ ability to assist women in choosing a method that they will be able to use correctly and consistently. This is of critical relevance, as previous counseling interventions in family planning have been largely unsuccessful.^23,24

In this article, we report results of a qualitative analysis of audio recordings of contraceptive counseling visits at clinics in the San Francisco Bay Area that serve racially and ethnically diverse patient populations, and describe providers’ approaches to contraceptive counseling and patterns in their use.

METHODS

Sample and Data Collection

This analysis uses a subset of audio recordings collected between August 2009 and January 2012 as part of the Patient-Provider Communication About Contraception study. In this study, 342 women visiting 42 providers at six family planning, primary care and general gynecology clinics in the San Francisco Bay Area had their contraceptive counseling visits audio-recorded; consent was obtained from both the patient and the provider. Inclusion criteria for clinics were serving a diverse patient population and having nurse practitioners, physician assistants, certified nurse-midwives or physicians (medical doctors or doctors of osteopathic medicine) conduct the contraceptive counseling, as opposed to using peer counselors. This last criterion was used because of the belief that the dynamics of peer counseling may be distinct from those of counseling by licensed health professionals. Inclusion criteria for patients were speaking English; wishing to discuss initiating a new contraceptive method with their provider; not being or wanting to be pregnant; and self-identifying as black, white or Latina. The last criterion allowed for the testing of counseling disparities in the parent study.

Patients were recruited after checking in but prior to seeing the provider, and all had access to contraceptive methods at no or minimal cost through either publicly funded programs or private insurance. Participation rates were not formally tracked, as in many cases it was not possible to assess eligibility before potential participants indicated they did not wish to hear more about the study. Demographic data (age, race and ethnicity, educational level, household income) were collected from patients through a previsit survey; information on age, race and ethnicity, and professional degree was collected from all providers via a survey. This study received approval from the Committee on Human Research at the University of California, San Francisco.

For the present analysis, a sample of 50 counseling sessions—the number anticipated to ensure saturation for the research questions—was randomly selected from the 342 sessions. The sample was stratified by clinic to ensure representation across sites.

Analysis

Guided by the principles of grounded theory to inductively identify concepts in the data,²⁵ the second author listened to 25 counseling sessions for overarching themes and developed a preliminary list of codes related to both process (e.g., method selection, method exclusion) and content. Content-related codes were derived from method characteristics that previous research^26,27 had identified as important to patients during method selection, including frequency and mode of administration, side effects and efficacy. The first and second authors discussed the proposed codes, eliminating some as redundant or of marginal interest to the research questions and identifying additional codes of interest from the literature. The two authors agreed on a first-round code list, which the second author used to code all 50 sampled transcripts in Atlas.ti 6. In keeping consistent with a grounded theory approach, this author wrote memos throughout the coding process to summarize the codes and their interrelationships, and through discussion with the first author, developed our preliminary results.

The first and second authors then discussed these findings with attention to emergent patterns in providers’ overall counseling approaches. These approaches were defined by providers’ behavior, and did not take into account the behaviors of women or their responses to these approaches. Three counseling approaches were identified—two that corresponded with approaches discussed in the literature and one emergent approach. These authors composed a definition of each approach, informed by both the literature and the inductive coding, and noted important axes of variation in how these approaches may be applied. The second author reviewed all 50 transcripts again and coded each with a primary approach. Because thematic saturation was reached using these transcripts, we did not add more to the sample.

We then assessed the prevalence of use of the identified counseling approaches in the overall sample, as well as the patterns of use in relation to patient demographic characteristics. Emergent patterns were described to offer a broad sense of the relative frequency of each approach and to inform future hypothesis generation and testing; no formal statistical analyses were performed.

As part of the parent study, all transcripts were coded separately for whether the patient expressed a method preference. A patient was considered to have a preference if she clearly indicated at some point in the visit a preexisting interest in one particular method.

RESULTS

Twenty-five of the 42 providers in the full study were represented across the sample of 50 sessions. Thirteen providers contributed a single session, five contributed two sessions, three each contributed three or four sessions, and one accounted for six sessions. The most common professional degree was nurse practitioner (16), followed by physician (six), physician assistant (two) and certified nurse-midwife (one). All providers in the sample were women, and they ranged in age from 35 to 74. Fifteen were white, seven were Asian or Pacific Islander, one was Latina and two were of mixed race or ethnicity.

Women ranged in age from 17 to 44, and a majority were in the 17–25 age-group (Table 1[JT1]). Nearly half were white, a third were black and a fifth were Latina. Thirty-eight percent had completed some college, and 32% were college graduates. This was a predominantly low-income population; 68% of patients lived in households at or below 200% of the federal poverty level. Sixty-two percent expressed a contraceptive method preference during their visit (not shown).

TABLE 1.

Percentage distribution of female patients receiving contraceptive counseling at family planning, primary care and general gynecology clinics, by selected characteristics, San Francisco Bay Area, 2009–2012

Characteristic	% (N=50)
Age
17–25	60
26–35	26
36–44	14
Race/ethnicity
White	48
Black	34
Latina	18
Education
High school/GED/other*	30
Some college/associate’s degree	38
≥college	32
% of federal poverty level
<100	40
101–200	28
>200	32
Total	100

^*Two patients were currently enrolled in high school, and two had dropped out.

We discerned three primary approaches that providers took to counseling women on contraception, which we term foreclosed, informed choice and shared decision making. Each was characterized by a particular pattern of method discussion (Table 2[JT2]), reflecting the extent to which the patient articulated her contraceptive desires and needs, how the provider responded to them and how much information the woman received about her options.

TABLE 2.

Definitions of the three contraceptive counseling approaches; and percentage distribution of sessions and number of methods mentioned, by approach

Approach	Definition	% of sessions (N=50)	No. of methods mentioned*
			Range	Mean
Foreclosed	Provider offers information about methods that patient explicitly mentions, but does not introduce methods, actively guide the conversation or take a role in decision making	48	1–5	2.2
Informed choice	Provider shares method information and may introduce methods into the conversation, but leaves all decision making to the patient	30	2–10	5.0
Shared decision making	Provider serves as a source of method information, introduces methods, and interactively and responsively participates with the patient in method selection	22	2–12	5.2
Total		100	na	na

^*Includes only methods proposed for regular contraceptive use, not as bridge methods. Condoms were counted only when proposed for pregnancy prevention.

Note: na=not applicable.

Foreclosed Approach

The most common approach (used in 48% of visits) was what we term a foreclosed approach, wherein providers gave information about methods that patients explicitly mentioned, but did not introduce methods and explicitly abdicated any role in decision making. In practice, this meant that discussion of options the patient might not have been familiar with was closed off. For example, when a 39-year-old patient explained that she had used the pill for 22 years and, tired of taking the pill, was interested in an IUD, her provider endorsed this interest, offering that “the IUD is a great form of birth control.” The provider did not explore why the patient was interested in the IUD or mention any other methods. This approach was especially common during visits with patients who expressed a method preference (68%), suggesting that it was, in many cases, a response to the provider’s perception that the patient had already made a decision.

Providers also used the foreclosed approach in 35% of visits in which patients did not present themselves as confident in their initial preferences. A 17-year-old patient, for example, requested birth control pills and mentioned previous experience using them, noting some difficulties she had had with the method: “I guess in the morning I probably got a little sick. I felt like a little sick, but that’s it.” Instead of investigating the extent of the patient’s discomfort or whether this discomfort contributed to her discontinuing the method in the past, the provider reflected the patient’s minimization of the side effects and offered to prescribe the pill again, saying, “But it didn’t bother you that much? So, do you want to go ahead and start back on the pill now?” In using a foreclosed approach, the provider did not probe aspects of the patient’s experience and needs that could have consequences for method satisfaction and adherence.

Counseling sessions characterized by the foreclosed approach had the lowest average number of methods discussed (2.2). Even when patients did not express a preference or mention specific methods, providers avoided introducing methods, often by asking about the patient’s contraceptive history and encouraging resumption of a previously used method. For example, one provider broached the topic of contraception by asking: “Have you been on anything before?” The 25-year-old patient listed brands of the pill and the ring, noting, “I was very consistent with those for a large amount of time.” Although she expressed neither satisfaction nor dissatisfaction with either method, the provider took her assurance of method adherence as tacit endorsement. The provider foreclosed discussion of other methods when she asked, “Which one [of these two methods] worked best for you?” The patient replied that the pill had worked best and continued, letting a tentative question trail off: “If you have any other suggestions then …” By answering the provider’s question while also leaving open the possibility of adopting a different method, the patient’s response suggests that she was unsure that her past experience was sufficient to plan her current and future contraceptive use. However, the provider did not recommend another method or participate in decision making, let alone engage in discussion about methods. Rather than taking up the patient’s uncompleted question, she responded to the patient’s statement that the pill had worked best and offered to prescribe a specific brand of pill. No further methods were discussed.

Notably, providers using a foreclosed approach did not review characteristics—such as side effects and efficacy—of every mentioned method, or even of every selected method. In some cases, this was a result of the patient’s resuming a previously used method. For example, one provider implied that discussion of side effects was unnecessary because of the patient’s previous use, asking her patient if she needed to “refresh with me regarding the side effects of pills?” In the 33% of foreclosed sessions in which efficacy was discussed, it was most often referred to in vague terms, such as “really effective,” “very effective” and even not ineffective, as in one provider’s explanation that “the low-dose [version of the pill] doesn’t mean it’s ineffective.” Across the foreclosed sessions, there was no clear indication of why these characteristics were not reviewed or, when discussed, were given only cursory review. When side effects were discussed—which they were in only 29% of the foreclosed sessions—they were brought up not as attributes of a method that should influence decision making, but rather as additional information after a method had been selected. For instance, one provider concluded her description of the side effects of the method her patient had already selected by saying, “So, just be aware of that.”

Informed Choice Approach

The second most common counseling approach (used in 30% of visits) followed the format of the informed choice approach,²⁸ wherein the provider shared information with the patient, even introducing methods into conversation, but left all decision making to the patient. For example, one provider explained, “I can remind you of other options,” but she was clear that the contraceptive decision was exclusively the patient’s: “It’s kind of what you are feeling like.” Providers used an informed choice approach 21% of the time when patients expressed a preference and 35% of the time when they did not.

In this model, patients’ expression of a method preference did not prevent providers from introducing additional methods into discussion. However, providers often proposed methods in ways that offered little rationale for their introduction. For instance, when a 25-year-old patient suggested she might resume use of the pill, her provider voiced support for her interest and then described other options: “If you want to take the pill, that’s easy, and that’s fine. Just so that you know, there are other options and variations on the theme of the pill.” She then introduced the patch, ring, injectable, IUD and implant, all without comment from the patient and, moreover, without framing these methods in relation to the patient’s expressed wants and needs.

As in this case, providers using the informed choice approach often introduced methods in a sort of laundry list, naming several methods in succession, sometimes also offering brief descriptions. Largely because of this format, providers in informed choice sessions mentioned an average of 5.0 methods, more than twice the number in foreclosed sessions. As was also typical, the provider in the preceding example did not solicit the patient’s method preferences. Instead, she justified her listing of methods as a means of informing the patient and preparing her to make a contraceptive decision: “At least you know of [your options]. And you can decide what you like.” Providers who used this approach shared contraceptive information, but did not frame it in ways that were tailored to the patient.

This emphasis on information provision meant providers sometimes made inaccurate assumptions about patients’ needs. In one session, a 24-year-old patient asked about switching from the pill to the IUD, but offered no explanation for her interest in changing methods. Her provider responded by conveying information about not only the IUD but also the implant and the ring, obliquely explaining the introduction of this unsolicited information by saying, “[These are] the other things to think about … if you have trouble remembering to take the pill.” The provider assumed that the patient’s desire to change methods stemmed from her difficulty with the pill as a daily method. However, as the patient soon clarified, her interest in switching actually stemmed from a desire to use a hormone-free method. She said, “Can I interrupt you for a minute? The only reason I’m curious about the IUD is that I like the idea of not being on hormones. Just, I don’t know.” Without interactive and probing discussion about the patient’s interests, preferences and dislikes, the provider made assumptions that turned out to be incorrect.

In visits using the informed choice approach, the information provided about the characteristics of different methods most often was cursory; only 33% of these visits included mention of side effects, and 40% included efficacy, usually using imprecise terms, as in foreclosed sessions. When providers offered more comprehensive information about a method, they did not discuss characteristics for every method mentioned or discuss methods in ways that were directly comparable. For example, as she introduced four methods, one provider described the hormonal makeup and adherence challenges of the first (the pill); the physical characteristics, use and other patients’ positive evaluations of the second (the ring); the physical characteristics, use, primary side effects, other patients’ positive and negative evaluations, and duration of effectiveness of the third (the implant); and the physical characteristics, primary side effects and duration of effectiveness of the fourth (the IUD). In highlighting different aspects of each method, the provider made it difficult to compare methods across individual attributes. Furthermore, when listing the attributes of each method, the provider did not use comparative language (e.g., did not state that one method’s side effect was “more common” than that of another). Notably, in contrast to the discussion in the foreclosed approach, in the informed choice approach the discussion of side effects, when it occurred, took place before method selection.

Shared Decision-Making Approach

Finally, some providers employed what the literature has termed a shared decision-making approach, wherein the provider not only served as a source of information, but also interactively and responsively participated with the patient in method exclusion and selection.^29,30 Providers used their acquired knowledge of patients’ needs and wants regarding method characteristics, including the mode and frequency of administration and potential side effects, to introduce and responsively frame a variety of contraceptive methods. In shared decision-making sessions, the average number of methods discussed was 5.2, similar to the average considered in informed choice sessions, although the conversational patterns of the two approaches were distinct. This was the least common approach overall (accounting for 22% of visits), and was used in a notably smaller proportion of sessions in which patients expressed a preference (11%) than in sessions where they did not (29%).

Unlike counseling sessions using the other approaches, sessions characterized by shared decision making generally began with the provider’s establishing rapport with the patient, regardless of whether a patient expressed a preference. The session with a 39-year-old patient is one example. Before initiating the contraceptive discussion, the provider discussed the patient’s new baby and time off from work. The provider made supportive exclamations about the ease of the patient’s labor and delivery, saying, “Wow!” and “Oh my gosh! Good for you!” and asked after the patient’s overall well-being and postpartum adjustment: “Are you getting some good time off of work and everything?” “Are you doing okay at home?” These rapport-building interactions gave the provider information that improved the subsequent conversation about method choices. For example, during the opening conversation described above, the patient mentioned the difficulty of a prior birth and her desire to have no more children. When she and the provider later discussed contraceptive options, the patient explained that she was interested in her husband’s getting a vasectomy, but “I think he would rather that I do the IUD.” The provider then explored whether the patient’s desire for her husband to get a vasectomy was associated with her wish not to have any more children or was motivated by another reason. As it turned out, the patient was interested in a vasectomy because she wanted to allow her body to be “in its natural state.” She said, “I’ve been through a lot already. I don’t want to go through more.” For this reason, the provider realized, an IUD would not meet the patient’s needs. Instead, she researched the process for referring patients for a vasectomy and sent the patient home with printed information on the procedure.

Providers using a shared decision-making approach regularly summarized patients’ comments, reflecting their understanding of patients’ needs and concerns in ways that also moved the decision-making process forward and created and maintained rapport. When a 31-year-old patient came in reporting both pain and irregular bleeding with her IUD, her provider first probed to fully understand her dissatisfaction with the method, soliciting the patient’s experience with questions such as “Tell me more about the cramping and how often you’re getting it.” Having identified the problem, the provider suggested alternatives, one method at a time, gauging the patient’s interest and answering any questions. As they moved through the options, they narrowed the possibilities, and the provider gently reminded the patient of the methods they had ruled in. Once they were down to two methods, they discussed both in greater detail. In other examples, providers connected to patient needs by articulating a shared goal. For instance, one provider framed her questions in terms of a joint goal of preventing unintended pregnancy. She said, “One of my concerns is, I don’t want you to get pregnant if you don’t want to be pregnant.”

Patients played an active role in guiding the discussion in the shared decision-making approach. For instance, when a 20-year-old patient vaguely volunteered, “I think I need to learn more about the pill first before I just start,” her provider, rather than respond with detailed information about the pill, asked her to be specific: “Do you have some questions I can answer?” The provider then responded to each question, not only giving factual information, but also reflecting what insight the questions themselves gave her into the patient’s needs. For example, after the patient asked whether the pill had to be taken every day, her provider answered affirmatively and then responded to what she understood was the patient’s concern: “If you’re worried about taking something every day, the other option is the patch or the ring.” In contrast to the earlier example of the 24-year-old patient who expressed interest in switching from the pill to the IUD, the provider framed her interpretation of the patient’s concern as tentative (“if”), inviting the patient to affirm or correct her interpretation. The patient, in turn, tacitly confirmed the provider’s assumption, stating that she might be interested in the ring.

As in other models of counseling, providers using shared decision making frequently failed to discuss method characteristics or discussed them only for some methods. Just 27% of these visits included mention of efficacy, usually in cursory ways, and 45% included side effects. Nonetheless, unlike when providers used other approaches, when they used this approach they discussed side effects and other method characteristics in the flow of their interactive conversation, in response to something the patient said. The shared decision-making approach was also distinguished from the other approaches in that providers who engaged in discussion with their patients spent less time after method selection on adherence issues, having generally covered these aspects during the decision-making process.

Variation by Patient And Provider Characteristics

Age was the only patient characteristic that showed a pattern regarding counseling approaches, and no variation was found by provider characteristics or across the six clinic sites. In fact, of the 12 providers who appeared multiple times in our sample, nine used more than one approach across sessions; the three who used only a single approach appeared just two times each in the sample.

Strikingly, patients aged 25 or younger experienced counseling with a foreclosed approach 60% of the time, while patients aged 26–35 and those older than 35 experienced this approach 38% and 14% of the time, respectively (Table 3[JT3]). This pattern by age was not explained by women’s rates of expressing a method preference, as the proportions of patients doing so were similar across age-groups (from youngest to oldest, 40%, 38% and 29%). Providers also used an informed choice approach frequently with patients who were 35 or younger (33–38% of the time), with the result that foreclosed and informed choice approaches dominated the counseling of these patients. Hence, providers played little role in contraceptive decision making among women in the two youngest age-groups. In contrast, all but one patient who was older than 35 was counseled with a shared decision-making approach.

TABLE 3.

Percentage distribution of patients, by contraceptive counseling approach, according to age-group

Approach	17–25 (N=30)	26–35 (N=13)	>35 (N=7)
Foreclosed	60	38	14
Informed choice	33	38	0
Shared decision making	7	23	86
Total	100	100	100

Note: Percentages may not total 100 because of rounding.

DISCUSSION

Our investigation of contraceptive counseling revealed three main communication approaches: foreclosed, informed choice and shared decision making. Women who were counseled using foreclosed or informed choice approaches experienced little discussion about what they valued in a method and received inconsistent information about method characteristics, and their providers did not participate in the decision-making process. In contrast, visits characterized by shared decision making—which were the least common—featured an emphasis on collaborative decision making, including rapport building and active facilitation of the decision-making process to identify a method that was best suited to the woman’s expressed needs.

The process of decision making in contraceptive counseling can be understood in the context of the types of decision making described in the medical literature. These are commonly defined as including paternalistic decision making on one extreme, shared decision making as the middle ground and informed (also known as consumerist) decision making on the other extreme.³¹ Studies of medical decision making have found that the provider-dominated approach is common.^32–34 In contrast, our findings indicate that in contraceptive counseling, decision making occurs largely on the patient-driven end of the spectrum.

Our finding of a lack of provider involvement in or facilitation of decision making in the majority of family planning visits is similar to findings in the developing world^15–18 and is consistent with research on the philosophy of counseling in family planning programs, which emphasizes an autonomous model of decision making because of the personal nature of the reproductive health context.^19,28,35 While we were not able to determine the intentions of the providers in our sample, both the informed choice and the foreclosed models may be motivated by a desire to prioritize patient autonomy in the decision-making process. By deferring the decision entirely to the patient, the provider can avoid any perception of coercion. In addition, there is some evidence that this approach may enhance contraceptive continuation.³⁶

Other research addressing health communication both generally and in family planning specifically has raised questions, however, about whether this degree of provider detachment is desirable. In the general health literature, shared decision making has been increasingly recognized as a desirable, patient-centered approach to clinical communication.^37,38 The basis for this recognition is that each party brings unique expertise to the interaction: Medical providers have technical knowledge of the available options for a particular health care decision, whereas patients are experts in their own values and preferences.²⁹ When these two perspectives are brought together, the choice that best reflects the integration of medical evidence and patient preferences can be selected.

While it can be argued that the choice of a contraceptive method is a unique medical decision—because of the sensitive issues related to sexuality and reproduction—one study of women’s preferences for contraceptive counseling suggests that shared decision making is a desirable approach.¹⁴ This raises the question of whether our findings indicate that this approach is being underutilized. Importantly, counseling based on shared decision making remains focused on providing patients with their preferred method, and therefore would not be expected to result in lower rates of method continuation.

Another notable finding was a lack of counseling designed to promote the use of highly effective methods, or even to draw attention to differences in efficacy between methods, as would be consistent with a paternalistic approach. Given the low rate of overall contraceptive use and low rates of use of highly effective methods, some researchers have argued that directive counseling should play a role in this context.³⁹ In accordance with this perspective, several studies have tried, unsuccessfully, to employ motivational interviewing—a patient-centered, directive counseling approach—to improve contraceptive use.^40,41 Our results suggest that in the clinical context, providers do not take an active role in conducting behavior change counseling designed to increase the use of highly effective methods. The extent to which inclusion of this type of counseling in family planning care is appropriate or possible, especially given the documented association between perceived pressure to use a specific method or failure to receive a preferred method and method discontinuation,^36,42 is an important area of future research.

Our finding that the type of counseling varied by the patient’s age merits further research. We speculate that this pattern may be related to providers being more comfortable engaging in interactive discussion with patients with whom they are closer in age. Alternatively, this pattern may be attributed to provider concern about unduly influencing the decisions of younger women, who may be more impressionable. Research examining how and why providers make counseling decisions is an important area for future investigation, especially as younger patients may be most in need of active facilitation of the decision-making process, given that they likely have had less experience with contraceptive methods and less interaction with family planning providers.

Limitations

This study has several limitations, one of which is that the sample was restricted to a single metropolitan area. It is unknown to what extent counseling practices in the San Francisco Bay Area can be generalized to other parts of the United States. However, the consistency of our findings with those of studies performed in the international setting suggests that lack of engagement between the patient and the provider in the decision-making process is pervasive. Our sample was also restricted to sessions with licensed health professionals and may not be generalizable to peer counseling contexts. In addition, our focus on providers’ behaviors may have neglected the influence of patient behaviors (including affective tone and body language) on how providers relate to and communicate with patients. Furthermore, our analysis does not address whether patients found the counseling approach their provider used helpful in their contraceptive decision making. Finally, we acknowledge that having the coding completed by one author can be viewed as a limitation. However, steps were taken to ensure the validity—and, consistent with the qualitative model, trustworthiness—of the analyses, including an iterative coding process conducted with regular communication with coauthors.

Conclusions

To the extent that the continued high rate of unintended pregnancy in the United States is a consequence of the underuse and inconsistent use of effective contraceptives, our findings suggest that providers’ counseling approaches represent an unleveraged opportunity to meet patients’ contraceptive needs and facilitate appropriate use and continuation. The frequent lack of involvement of family planning providers in patients’ contraceptive method decision making suggests that research on counseling interventions should examine how providers can engage, and be taught to engage, with patients during this process. Such engagement must attend to the balance between respecting patient autonomy and facilitating ongoing use of effective contraceptive methods.

Biography

Christine Dehlendorf is associate professor in residence, Departments of Family and Community Medicine; Obstetrics, Gynecology and Reproductive Sciences; and Epidemiology and Biostatistics—University of California, San Francisco (UCSF). Katrina Kimport is assistant professor, Department of Obstetrics, Gynecology and Reproductive Sciences, and research sociologist, Advancing New Standards in Reproductive Health; Kira Levy is a medical student; and Jody Steinauer is associate professor, Department of Obstetrics, Gynecology and Reproductive Sciences, and Bixby Center for Global Reproductive Health—all at UCSF.