Perceptions of positive and negative clinician communication in obstetrical visits from the perspectives of pregnant patients who use substances

Abstract

Objectives

To explore the perceptions of pregnant patients who use substances regarding positive or negative clinician communication during obstetrical care.

Methods

We analyzed qualitative data from 85 semi-structured interviews with pregnant patients who reported or tested positive for substance use, which explored their interaction with obstetric providers during their first prenatal visit. This analysis focuses on patients' perceptions of negative versus positive clinician communication behaviors.

Results

Eighty-five participants described clinician communication behaviors they felt affected their feelings about the clinician and their willingness to talk about prenatal substance use and other sensitive topics. Negative behaviors included clinicians (1) expressing judgment, (2) rushing through the consultation and providing limited information to patients, and (3) using statements or behaviors that made patients feel dehumanized. Positive behaviors included clinicians (1) explicitly expressing care for the patient, (2) creating rapport by soliciting patient stories and building relationships, and (3) demonstrating attentive listening.

Innovation

To our knowledge, our study is the first to explore clinician communication behavior with a focus on prenatal substance use from the perspective of pregnant people using substances.

Conclusion

Our findings highlight pregnant patients' perspectives on communication patterns that could improve patient-clinician interactions and, in turn, maternal health care and outcomes.

Highlights

• •Clinician communication is integral to a successful patient-clinician relationship.
• •Clinician communication determines patients' prenatal substance use disclosure.
• •Clinician communication affects the adoption of healthy behaviors.
• •Positive clinician communication leads to engagement in care and care maintenance.
• •Positive communication pattern improves healthcare utilization and health outcomes.

1. Introduction

Prenatal substance use is associated with numerous adverse outcomes, including prematurity, low birth weight, and fetal and neonatal deaths [[1], [2], [3]]. National 2020 survey data highlight that in one month, 8.3 % of pregnant people used one or more illicit drugs, 8.4 % smoked cigarettes, and 10.6 % drank alcohol [4]. Although this data noted a decrease in prenatal tobacco use from 2019 to 2020, the use of drugs remained steady, and the use of marijuana significantly increased [4]. These rates, however, may be underestimated as many pregnant people do not disclose substance use to their healthcare clinicians due to fear of bias, stigma, or legal/custody implications [[5], [6], [7], [8], [9], [10], [11]]. Lack of patient disclosure may, in turn, reduce opportunities to provide information and resources regarding the adverse effects of prenatal substance use. Studies have shown substance use-related outcomes improve with high-quality prenatal care and good patient-clinician relationship/interaction [[12], [13], [14]].

The quality of the patient-clinician relationship has been vital to effective healthcare encounters [[15], [16], [17], [18], [19], [20], [21]] and high-quality prenatal care [[22], [23], [24]]. Positive patient-clinician relationships are those with high trust, mutual regard and respect, and confidence in clinician medical knowledge and understanding of patient concerns and values [[15], [16], [17], [18], [19], [20], [21]]. Positive patient-clinician relationships are associated with patient satisfaction, medical advice and management adherence, and better health outcomes [[15], [16], [17], [18],20,[25], [26], [27], [28], [29], [30], [31], [32], [33]]. For all birthing people, especially those vulnerable to bias and stigma, factors facilitating and inhibiting positive patient-clinician interaction/relationships must be better understood. Researchers have explored pregnant people's views of their prenatal care experiences and desired clinician characteristics [23,[34], [35], [36], [37], [38], [39]]. However, these studies had racially homogenous participants and did not specifically examine the perspectives of pregnant patients who use substances. Additionally, few studies focused on clinician communication in obstetrical care visits. Although Epstein et al. highlighted the importance of knowing what patients notice, want, and need when communicating with their clinicians, [40] none of the studies included in this review focused on pregnant patients or those who used substances.

Much of the current literature regarding patient-clinician communication related to prenatal substance use focuses on specific screening questions and approaches such as use of questionnaires and tools rather than examining clinician communication behaviors more broadly as interactive, dynamic, relational behaviors [[41], [42], [43], [44], [45]]. Prior studies exploring the experiences and perspectives of pregnant patients who use substances highlighted concerns such as fear of legal involvement, loss of custody or stigma [6,46,47]. These studies did not specifically ask about clinician communication. Our study addresses this gap in knowledge by describing negative and positive clinician communication behaviors from the perspective of pregnant patients who use substances during pregnancy. Greater understanding of patients' perceptions of what they like and do not like related to how clinicians talk and interact with them during their prenatal visits will help guide efforts to improve communication training in addressing sensitive and stigmatizing issues in pregnancy care such as prenatal substance use.

2. Methods

The current analysis utilized data from an NIH-funded observational study of patient-clinician communication regarding substance use in first prenatal visits [48,49]. The parent study aimed to identify communication styles and processes that encourage pregnant people who use substances to disclose alcohol or illicit drug use and subsequently stimulate patient-clinician conversations promoting the adoption of positive behaviors. The parent study recruited obstetric clinicians and their English-speaking pregnant patients who presented for a first obstetric visit at five obstetrics-gynecology offices. The conversation between clinicians and patients during the first obstetric visit was audio-recorded, and the researchers administered an after-visit survey to patients. At the end of the survey, the researchers informed patient participants about a study focused on substance use and invited them to participate in additional study activities, including providing a urine sample for cotinine and toxicology testing. Patient participants who disclosed prenatal alcohol, marijuana, or drug use in their recorded conversation or their post-visit survey or whose study urine test was positive for drugs were invited to participate in semi-structured qualitative interviews within four weeks of the initial recorded visit. These qualitative interviews intended to gain a deeper understanding of how participants experienced and interpreted their clinical interaction during the recorded first obstetrical visit. We used a phenomenological approach to explore their perceptions and views of what was meant and conveyed. A phenomenology framework aims to gain understanding of the essence and meaning of our participants' experience from their perspectives and interpretations [50]. We choose this approach as we felt it important to highlight the opinions and perceptions of pregnant people who may face and anticipate stigma and judgment for their substance use. The [institution redacted] Institutional Review Board (IRB #PR008090530) approved the study, and all participants who provided urine or took part in the semi-structured interviews signed another written informed consent for those additional study activities prior to data collection and received additional compensation for each activity.

All interviews were conducted in person in private clinical research offices by research team members trained and experienced in qualitative interviewing. The interviews asked participants to reflect upon their perception of their visit experience, their perspectives and beliefs regarding prenatal substance use and disclosing their substance use, as well as their perceptions of the communication used by their obstetric clinician to address prenatal substance use. The researchers played the audio clips of the recorded conversation during which their clinician addressed substance use for the participants, after which they were asked to reflect on the communication behaviors used by their obstetric clinician. We asked participants to consider how this communication did or did not influence their willingness to disclose their substance use using two key prompt questions: “What did your clinician do well?” and “What could your clinician have done differently?” Although this portion of the interview was when our team specially asked about communication in a dichotomous fashion (i.e., what went well versus what could be improved), interview participants described experienced and desired communication behaviors throughout the interview and research team members would prompt participants to reflect on their opinion how they felt about that communication (e.g., “How did you feel about that?” “Would this be helpful?”).

This analysis uses the portions of these semi-structured patient interviews where patient participants described their opinions about obstetric clinicians' communication related to substance use. Based on our framing of the questions and prompts, participants' discussions regarding clinician communication behaviors fell into dichotomized categories of perceived negative/bad (e.g., what they did not like about their clinician; behaviors that served as barriers to substance use disclosure, hindered communication, made them uncomfortable; reasons they wanted to discontinue care with the clinician) and perceived positive/good, (e.g., what they liked about their clinician; behaviors that facilitated substance use disclosure, fostered communication, made them feel comfortable with their clinician; reasons why they would want to continue prenatal care with the clinician).

2.1. Data analysis

We coded the qualitative data using an iterative, constant-comparison method [51] with an inductive approach [52]. All codes and analyses were stored and managed using NVivo 12 [53]. Three study team members independently read the first two transcripts for codebook development and organized data into initial categories and concepts. We met to compare categories, themes, and observations noted, then developed representative codes. We repeated this process, independently reading another two transcripts and then adapting the codebook. Three members then independently coded four interviews with this codebook then met to compare codes, discussing any code application inconsistencies, and adapting codes as needed. This iterative process was repeated, with coders meeting periodically to discuss coding differences and then refine and expand codes in response to new emerging themes. This resulted in a final codebook with code definitions, rules for code application, and examples of codes. The final codebook was re-applied to all already coded transcripts and used to complete coding on all remaining transcripts.

The team members double-coded 30 % of the 85 transcripts of patient interviews to ensure coding consistency and benefit from investigator triangulation [54]. During double coding, we planned for a third coder to review and adjudicate any interpretative differences noted between coders, but we noted no need for adjudication. After coding all transcripts, we compiled and reviewed the coding, noted emerging and prominent concepts, and identified relationships and patterns between codes. Following this, we clustered codes and concepts with similar patterns into mutually exclusive categories and then utilized these categories to identify several key topical domains.

For this paper, we focused on portions of the interviews that related to patients' perceptions of clinician communication, including general communication behaviors, communication around substance use, and communication experiences during their current and previous prenatal visits.

3. Results

Of the 85 patients who participated in semi-structured interviews, 62 % identified as Black or African American, 28 % as White, and 9 % as Other Race. Participants' ages ranged from 19 to 39 years, averaging 25.1 years. Forty-five clinicians conducted the 85 visits with most identifying as female and White race. While there was high patient-clinician race concordance for White patients, it was opposite for Black patients; White clinicians conducted visits for 71 % of the White patients; Black clinicians conducted only one of the visits for Black patients. Data reflected experiences from first prenatal visits with nurse midwives, nurse practitioners, 1st - 4th-year residents, and attending physicians (Table 1).

Table 1.

Participants sociodemographic and clinical characteristics.

Variables	Category	N (%)
Patient characteristics (N = 85)		N (%)
Race/Ethnicity	Black White Other	53 (62 %) 24 (28 %) 8 (9 %)
Age, years (mean = 25.1, SD = 4.9, min/max = 19/39)	<20 20–29 30–39 40+	8 (9 %) 58 (68 %) 19 (22 %) 0 (0 %)
Marital status	Single Living with partner Married Separated Divorced Widowed	35 (41 %) 40 (47 %) 7 (8 %) 2 (2 %) 1 (1 %) 0 (0 %)
Highest level of education completed.	Grade school High school/GED Associates degree Some college Finished college Graduate school	18 (21 %) 33 (39 %) 7 (8 %) 25 (29 %) 2 (2 %) 0 (0 %)
Annual household income, $	0–4999 5000–9999 10,000–14,999 15,000–19,999 20,000 and higher Refused	37 (44 %) 17 (20 %) 15 (18 %) 6 (7 %) 10 (12 %) 0 (0 %)
Substance use disclosure pattern	Patient disclosed current prenatal substance use to the clinician. Patient did not disclose current prenatal substance use to the clinician. Missing data	34 (40 %) 50 (60 %) 1 (0.01 %)
Gravidity (number of pregnancies)	Mean = 3, SD = 2, Min = 1, Max = 9
Parity (number of births)	Mean = 1, SD = 1, Min = 0, Max = 5
Gestational age at new obstetric appointment, week	Mean = 13.4, SD = 8.0, Min = 5, Max = 39.3
Clinician characteristics (N = 45)		N (%)	Number and proportion of visits
Clinician type	1st-year resident 2nd-year resident 3rd-year resident 4th-year resident Nurse midwife Nurse practitioner Attending Physician	5 (11 %) 15 (33 %) 12 (27 %) 3 (7 %) 3 (7 %) 6 (13 %) 1 (2 %)	9 (11 %) 25 (29 %) 20 (24 %) 3 (4 %) 5 (6 %) 22 (26 %) 1 (0.01 %)
Clinician gender	Female Male	42 (93 %) 3 (7 %)	81 (95 %) 4 (5 %)
Clinician self-identified race	White Black Asian	36 (80 %) 2 (4 %) 7 (16 %)	72 (85 %) 2 (2 %) 11 (13 %)

We identified six major themes reflecting negative and positive clinician communication behaviors. These were cross-cutting themes in the study participants regardless of demographics and substance use disclosure patterns.

3.1. Negative clinician communication behavior

Participants described communication behavior as negative when such behaviors elicited feelings of discomfort, shame, disrespect, or devaluation. Participants included behaviors they had experienced either during this pregnancy or in prior care and behaviors they anticipated or feared.

3.1.1. Clinicians expressing judgment

Concern regarding clinician judgment was the most dominant theme among this sample of pregnant people who used substances. Participants described how judgment related to their substance use impacted their healthcare relationships and experiences of care. For some patients, these experiences were barriers to disclosing their substance use or continuing care. One participant described how she had to change practices due to her obstetrician's explicit statement indicating his negative bias: “I can't go to that one [doctor] because I missed one appointment, and he says he has ‘no tolerance for women on methadone.’“ (P7).

Another participant described how she felt a change in the clinician's tone and attitude that contributed to poor quality care:

I felt like I have gotten… worse quality care because somebody finds out that I've used drugs or because I'm in drug treatment or something like that. And even when I had to come, the day I missed dosing, you know, as soon as the doctor, you know, saw me, she just, the whole attitude flipped on me…It is not uncommon to encounter that, which is why I think a lot of people do lie about their use and they don't want to talk about it…. Doctors are supposed to remain objective; they are not supposed to let personal prejudices get in the way, but they do, you know. And I feel like a lot of times, addicts, in particular, receive sub-par care because of it. It is like doctors and nurses figure okay, well, if they are not going to take care of themselves, then why the hell should we? … (P21).

Another participant also described sensing the clinician's irritation and condescension through the clinician's attitude and words:

I went to clinic one time, and it just seemed like she [obstetrician] was just pissed off the whole day. Like she just did not make your day any better. Like she just well said, “Honey, you know you're having sex, so you should use this [condoms to prevent infection].” Well, honey, I didn't. So, help me and let me know everything. Don't just give it to me like that. That's not the way you…show somebody information. Like you shouldn't just judge me because you don't know me.... So, you shouldn't have given it to me like that, and you need to fix your face because it was totally unprofessional. (P31).

Patient participants also described other verbal and nonverbal ways clinicians expressed negative judgment, including changes in clinicians' tones or attitudes and eye-rolling.

Interviewer (I): Tell me, what tips you off that you feel that you are being judged by the doctor?

Participant (PT): I mean their attitude, their body language, like if they roll their eyes…

Or they look like they just keep their head down; they don't want to look at you eye to eye, like different things like that. Or like you tell them something (takes a deep breath and sighs) and the different things they do with their body - shrug their shoulders, roll their eyes, put their head down, different things like that. Or the way they talk at you, like their tone of voice…like they're looking at you like you're a piece of shit.…

They're going to be like all giddy with you at first, then you like start telling them something, and it's like, oh well. You know, you can just tell by their actions…

…like, they get all monotone. Like, hmmm, they are looking at you like, wow, you're worthless. You know what I mean, that's how I feel if like they were judging me; that's how I would feel. (P15).

3.1.2. Rushing through the consultation and providing limited information

Participants also described feeling unhappy with clinician communication that was rushed and left them with limited information or insights about risks associated with substance use. They were disheartened by short patient visits or interactions compared to the long wait times they experienced. Participants described that clinicians' rushed communication created the impression that they were not interested in getting to know them as individuals and did not want to hear their views, as described by this participant:

…I just felt like she was ready to be done [with the visit], and she was just, oh well, I'm going to just sit here and type. This is what she said if you listen to the recorder, she said, “I'm going to sit here at the computer, and I'm basically going to type everything that you are saying.” And she literally typed everything I was saying. And it's just like no, no, no. Like, and then she was like, put your legs up we're going to …… why are you, you're fast. You're moving fast. I'm not saying move slow, but you know just take it, take it easy. And I just really felt like she didn't want to be there like she was tired and wanted to go home or something. (P26).

Participants also described feeling unhappy with conversations that asked them to share details about various behaviors, including prenatal substance use but did not provide any reason for the questioning or education about implications or risks. Participants indicated that they wanted clinicians to create opportunities for discussion and learning. They emphasized the need for clinicians to use interactions to understand more about their histories and concerns related to prenatal substance use and provide education, insight, and options. The following excerpt describes one participant's wish that the provider's inquiry about smoking would have offered her an opportunity to discuss concerns about prenatal marijuana use for nausea:

…Well, she asked me about, did I smoke anything?…

I told her cigarettes, but she never got into it about marijuana or anything else. And I was going to tell her, but she just was focused on the cigarettes…

So, I didn't get a chance to tell her about the marijuana. So, um, that could be an option [of a question to ask] because I am trying to stop.…… Just ask me more questions…

Like… what my body is going to go through.…While I'm pregnant. [Tell me] What can I take for morning sicknesses or do for morning sicknesses?…

(P36)

3.1.3. Dehumanizing behaviors and communication

Participants also reported experiencing clinician statements or behaviors that elicited feelings that they were not being seen as a person. They described feeling labeled as a certain diagnosis or problematic behavior rather than an individual. Examples of such clinician behaviors included when clinicians spent no time on rapport building, skipped usual interactional elements such as introductions or exchanging pleasantries, or went immediately to the computer and spent more time engaging with the device than with them. They described that these aspects of the interaction made them feel like they were “just a number.” In the following example, this participant reflects on what made her feel that her clinician did not view her as a person:

Like, sometimes the way she was talking, like I said, it just felt like so unpersonable. Like you were trying to distance yourself from, finding out who I, who this pregnant person is…To you, I'm age 27, height 5″4” and weight 220 pounds.…

…Like I'm not [participant's name]. That's who I am. That's the person I am. I'm a mother of one, soon to be two…. Like, I mean, when you walk through the door, you're used to the person shaking your hand; ‘Hi, I'm doctor so and so.’ She came in, went to the desk, ‘Hi, I'm doctor X', and sat down. I mean, shake a hand or something, a friendly smile, a glance, or anything. Like you're so quick into wanting to get into the appointment that you didn't personalize yourself with the person you are having the appointment with. If I was a different person, I could've just shut down…. (P14).

Participants also discussed how clinicians seemed to be more concerned about their pregnancies and less concerned about them as people with health concerns and needs outside of pregnancy. They highlighted how the dehumanization of pregnant people overall might make patients feel less likely to discuss concerns:

…Just by being pregnant…you don't always get talked to. You're almost talked to as if you are a baby maker, like baby machine… It's like, they …talk and look at your stomach, like when they are talking to you. It's just like, you lose that sort of normal talking thing. (P6).

3.2. Positive clinician communication behavior

Participants also described verbal and non-verbal communication patterns they found helpful. These were behaviors to which they attributed feeling more comfortable with their obstetric provider and were behaviors they would endorse all clinicians to use in their interactions with pregnant patients.

3.2.1. Expressing care about the patient as an individual

Many patients discussed the importance of clinicians voicing specific statements indicating their care or concern about their well-being and health. They described that such statements helped them feel valued as human beings and to build rapport. They also indicated that when clinicians seemed genuinely concerned about their physical, social, and mental health, they were more likely to consider stopping or reducing their substance use and adhering to medical advice, as described below. One patient explained the impact of the clinician's communication of concern: When you find out somebody cares and they are telling you [about substance use] in such a way that you know they care, it makes you think twice about even thinking about even doing it again or even stopping, you know, depending on the situation. (P9).

3.2.2. Creating rapport

Patient participants in our study also indicated that rapport-building through social talk and informal questions helped reduce their anxiety and improve patient-clinician interactions. They described rapport-building as soliciting their stories to get to know them as a person or talking with them about their families, friends, and other vital aspects of their lives. One participant said:

…Like she asked about my son and how he was doing and all different types of stuff... She actually ended up remembering who I was from being with [another obstetric practice], so… We had a nice conversation in there. (P1).

Rapport building was facilitated by clinicians' friendly interaction styles, having bubbly personalities, and laughing and joking easily with patients and any family members or friends present at visits. Humanizing patients in these ways reduced clinicians' strictly biomedical communication patterns, often through simple methods as described below:

… She made me, well, she, you know, she introduced herself. Then when she came in um, we chatted a little bit. Just like, get to know …… basically, you don't find that often. And like I said, she made me feel comfortable throughout, you know my .…., because I really don't find myself being comfortable with a lot of people.… So, she made me feel comfortable, you know, I opened up.…and I, she definitely earned my trust. (P51).

The excerpt below describes a participant's perspective of her clinician's communication behavior, which highlighted the clinician's smiling and spending time to get to know and laugh with the patient:

She was just a really sweet lady (laughs). She was really nice; like she was very friendly and made me feel like you know …… because I am nervous about this [substance use] …. She made me feel really comfortable. … she didn't make me feel singled out or embarrassed.…Her question about drugs or alcohol were really routine…. She was…really smiley and really personable and always just laughing with us. She was just really nice; like she connected, you know.... Yeah, it didn't seem like she was out the door onto her next patient. She actually really had the time to treat me. (P2).

3.2.3. Attentive listening

Our participants also described that when clinicians demonstrated attentive listening, they felt respected, that they were cared about, and that the clinician was present, which in turn contributed to accurate history-sharing. Patients said they could tell clinicians were listening attentively to them when they were not just looking at the computer or typing on it throughout the consultation. Some participants noted that their clinicians paused between filling the electronic medical records on the computer to look at them:

She wasn't just looking at the computer when I spoke. Or when she was writing down answers or asking questions. She wasn't just focused on the papers; she was attentive to me and was listening to what I had to say, and you know, made sure I knew everything as feedback, you know. I knew everything that I needed to know. (P12).

Participants also discussed the importance of clinicians maintaining eye contact during consultations, with some saying that clinicians' body language, including eye contact with the patient, conveyed more than words. Specifically, they suggested eye contact was a sign their clinician was actively listening and interested in them as a person.

PT: I think she communicates really well.

I: What makes you feel that way?

PT: Um, just her manners, and she looks at you in the eye when she talks to you, you know. (P8).

I: What was she doing that made you realize that she was interested in what you were saying?

PT: She gave me eye contact.… She looked at me when I talked, she listened when I was done, she responded…and she didn't say anything negative.……It lets me know more; she's focused on my eyes, looking into my eyes and let me know that she isn't waving me off.… (P16).

4. Discussion and conclusion

4.1. Discussion

While our study is among the first to explore prenatal patient-clinician communication from the perspective of pregnant patients with histories of prenatal substance use, our findings corroborate favorable communication patterns noted in other studies, including expressing care and concern, attentive listening and eye contact, rapport building, and pausing from computer use during the appointment [12,[55], [56], [57]]. Further, our participants indicated these behaviors contributed to their willingness to talk about their prenatal substance use and consider behavior change.

Participants' worries about clinicians' judgment support other research describing healthcare workers' judgment of pregnant people using substances during pregnancy and birthing people's reports of clinicians making them feel like bad mothers [23,35,37,[58], [59], [60], [61], [62], [63]]. Most of these studies were conducted with African American/Black pregnant people and low-income people, and this aligns with our findings from a study population with a majority of patients who identified as Black and of low socioeconomic status based on their reported income. There is a possibility that the skin color and socioeconomic status of most of our study population impacted their experiences with their provider, their previous experiences within the healthcare system, and their perspective, as studies have shown that minoritized populations and people of low socioeconomic status are more likely to receive worse care as compared to other populations and their previous experiences of discrimination within the healthcare system affects their interactions in the healthcare setting and their satisfaction with care [35,58,60,64,65].

Medical visits can be challenging for patients who may feel vulnerable and as though they must relinquish control of their bodies to clinicians. Experiences of dehumanization from clinicians worsen patients' feelings of impaired agency [66]. Although one study posited that some clinicians might dehumanize patients to reduce burnout [67], our participants found this behavior alienating. Various studies have also described the phenomenon of pregnant people feeling dehumanized, invisible, and neglected when clinicians prioritize the fetus' life and how these attitudes are associated with a negative impact on the quality of care and healthcare utilization [[68], [69], [70]]. There is a critical need for the provision of respectful and relationship-centered care to pregnant patients that ensures they feel heard and being cared for as a person.

Participants also discussed the importance of attentive listening and eye contact, which could be negatively impacted by clinicians' focus on completing the electronic health record (EHR) and rushed visits. These findings align with other studies that found clinicians focusing on EHR negatively impacted patient-clinician communication with loss of eye contact, long periods of silence, and limited interpersonal engagement [71,72]. Training clinicians on EHR management skills and developing new EHR processes that facilitate positive patient-clinician communication is necessary.

Numerous studies have shown that pregnant patients who use substances have concerns about stigma and judgment that affect their willingness to seek care or discuss it with their clinician. [47,[73], [74], [75], [76], [77]]. Our participants' descriptions of clinicians' tone, words, and body language highlighted what cues they interpret that represent potential judgment or disdain. They also specified the impact of behaviors such as eye contact, smiling, and expressing concern in reducing this stigma. While none of these communication behaviors described are new, indeed, they are consistent with other recommended approaches to addressing other sensitive issues [78], as well as align with general recommendations for conveying humanism in clinical interactions [79]; they provide specific actions and considerations obstetrical care providers should use or avoid in optimizing conditions for improved partnership to address prenatal substance use. They also corroborate studies that have proposed using a healing-centered approach that incorporates harm reduction and motivational interviewing to combat stigma, build patients' trust, and improve self-efficacy and the skills needed to achieve optimal health outcomes for both the pregnant patient and baby [77,80].

Our study is not without limitations. Our study population consists of English-speaking, educated individuals, with many describing their race as Black and reporting a low annual household income. Other themes could have emerged with a sample with different characteristics. Additionally, our analytic approach focused on identifying themes across the entire sample regardless of the type of substance used. We had not designed a sampling approach that allowed us to explore or compare themes among participants who used specific types of substances. Also, we acknowledge that the phrasing of our communication perception questions was presented in a dichotomous “good/bad” fashion, which may have limited discussion of communication categories that may be continuous, combined, or contextual. Additionally, despite interview prompts asking participants to reflect on the recorded conversations, many participants expanded their responses and reflections of what they considered good versus bad communication behaviors to other conversation experiences which limits insights specific to what occurred during those visits and complicates future analyses triangulating the observational data with interview perspectives. Finally, most obstetric clinicians who saw our participants for their recorded first prenatal visit were female and of white race. Our study's lack of racial diversity among the participating clinicians and high proportion of race discordance among visits may have also influenced the themes that emerged from the interviews.

Despite these limitations, our findings contribute to explorations of clinician communication behaviors to better address prenatal substance use and reduce stigma for pregnant people who use substances. While many clinical guidelines have focused on whether obstetrics clinicians screen and counsel on prenatal substance use [81], our findings suggest addressing how clinicians communicate, particularly regarding sensitive issues such as prenatal substance use, may be impactful.

4.2. Innovation

To our knowledge, this is the first study to examine patient-clinician communication with a focus on prenatal substance use from the perspective of pregnant people who were using substances during pregnancy.

4.3. Conclusion

Our study adds to the understanding of clinician communication behaviors patients perceive as “good” versus “bad,” which increases insight in developing clinician communication training interventions. Researchers and communication experts have shown that clinicians can learn to improve their communication behaviors [82,83]. Our findings augment calls for adopting, training, and utilizing relationship-centered communication approaches [84] to reduce stigma and barriers to addressing prenatal substance use and improve maternal care quality and outcomes.

Funding source

National Institute of Drug Abuse (NIDA), a component of the National Institutes of Health (R56 DA040617, R01 DA026410, R01 DA026410-S), provided funding. Funding was also partly from a grant from the Pennsylvania Department of Health. The National Institutes of Health, through Grant Number UL1TR000005, also supported the project. The study sponsors had no role in the study design, collection, analysis, and interpretation of data, writing the report, and the decision to submit the report for publication. The content does not necessarily represent the official views of NIDA, the National Institutes of Health, or the Pennsylvania Department of Health.

CRediT authorship contribution statement

Abisola Olaniyan: Writing – review & editing, Writing – original draft, Formal analysis, Conceptualization. Mary Hawk: Writing – review & editing, Supervision, Conceptualization. Dara D. Mendez: Writing – review & editing, Supervision, Conceptualization. Steven M. Albert: Writing – review & editing, Supervision, Conceptualization. Natalie Stern: Writing – review & editing, Formal analysis. Sneha Patnaik: Writing – review & editing, Formal analysis. Judy C. Chang: Writing – review & editing, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Dr. Judy C. Chang reports financial support was provided by National Institute of Drug Abuse. Dr. Judy C. Chang reports financial support was provided by National Institutes of Health. Dr. Judy C. Chang reports financial support was provided by Pennsylvania Department of Health. Dr. Judy C. Chang reports a relationship with Academy of Communication in Healthcare (ACH) that includes: board membership. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.