Women's experiences with person-centered family planning care: Differences by sociodemographic characteristics.

Objective: Person-centered contraceptive care is associated with positive reproductive health outcomes. Our objective was to analyze patients' ratings on the newly developed Person-Centered Contraceptive Counseling scale (PCCC) to provide distributions for a nationally representative population and to assess differences by sociodemographic characteristics. Study design: Using data from 2017 to 2019 National Survey of Family Growth (NSFG), we analyzed ratings across the four PCCC items among 2242 women who received contraceptive counseling in the past year. Items measured patients' reports of how providers respected them, let them describe their contraceptive preferences, took their preferences seriously, and adequately informed them about their options. We studied each PCCC item individually as well as the combined scale, distinguishing between ratings of "excellent" versus lower ratings. Bivariate and multivariate logistic regression models assessed how patients' characteristics (age, race/ethnicity and English proficiency, sexual orientation, income, and parity) and provider type were associated with the likelihood of experiencing person-centered care.
Results: The majority of women (59%-69%) reported that their family planning provider was "excellent" across the four PCCC items and just over half (51%) reported "excellent" on all items. In multivariate analyses, having a lower income, Black race, non-heterosexual identity, and Hispanic ethnicity combined with low English proficiency were associated with lower PCCC ratings. Conclusions: In a nationally representative sample, the PCCC captured variation in women's experiences with person-centered family planning care by sociodemographic characteristics. Findings highlight the need for contraceptive counseling that centers on clients' preferences and experiences, particularly for patients who belong to groups experiencing health inequities. Implications: Person-centered care is a key component of high-quality family planning services. This analysis highlights sociodemographic disparities in person-centered care by analyzing PCCC ratings. Findings show the value of this new health care performance measure and affirm the need for family planning care that centers individuals' preferences and lived experiences.

Introduction

Measures of successful contraceptive care have historically concentrated on reducing barriers to contraceptive access, the use of highly effective methods, and prevention of unintended pregnancy with less focus on the quality of care [1,2]. Person-centered care (also referred to as patient-centered care), which focuses on patients’ values and preferences, has recently become considered a key aspect of high-quality family planning care [3] and healthcare more generally [4]. Person-centered family planning care may promote greater patient autonomy, trust, and satisfaction [2,5,6] and is associated with continued contraceptive use [7].

Previous research has found sociodemographic disparities in the quality and patient-centeredness of care that family planning clients receive. Much of this research is based on small, qualitative studies and centers on patients’ race/ethnicity. One study found that low-income women of color were more likely than middle-class White women to report that their provider advised them to limit their childbearing or to not have children, and another found that Black women were more likely than White women to report having been pressured by a clinician to use contraceptives [8,9]. A study utilizing patient videos found providers were more likely to recommend long-acting reversible contraceptives to low socioeconomic status Latina and Black women than to low socioeconomic status White women, suggesting providers may have biases when providing counseling [10]. Studies have found both English-speaking and Spanish-speaking Latina women are less likely than White women to give their most recent reproductive health visit the highest rating on patient-centeredness [8], and Spanish-speaking, Hispanic foreign-born, and lower socioeconomic status women are less likely than other women to report receiving high-quality postpartum contraceptive counseling [11].

There is also evidence of disparities based on sexual orientation and age. Individuals who are lesbian, gay, bisexual, transgender, queer, questioning, or additional identities (LGBTQIA) often receive lower quality family planning care than straight cisgender patients, in part due to provider discrimination and lack of cultural competency [12,13]. In a survey conducted by the Kaiser Family Foundation about healthcare utilization more generally, younger women were more likely to report negative experiences than older women [14].

For this research, we conducted analyses of women's responses to the Person-Centered Contraceptive Counseling scale (PCCC) to assess disparities in the quality of interpersonal care and patient-centeredness. The four-item PCCC scale was added to the National Survey of Family Growth (NSFG) in the 2017–2019 survey cycle [15], allowing us to understand how this performance measure operates in a nationally representative population and to assess the person-centeredness of family planning care for this sample of women. Rich background data in the NSFG and the oversampling of Black and Hispanic women allow us to examine differences by race/ethnicity and English proficiency, income, sexual orientation, age, parity, and provider type.

Materials and methods

Data

Our sample was drawn from the NSFG 2017–2019 [15]. The NSFG, conducted by the National Center for Health Statistics (NCHS), gathers information on family formation behavior and the reproductive health of men and women aged 15 to 49. Our analytic sample included 2242 women who reported receiving contraceptive counseling and/or a contraceptive method during a family planning visit in the past year and thus received the PCCC rating questions.

Measures

Person-centered contraceptive counseling ratings

The PCCC, developed at the University of California, San Francisco, asked patients to report the extent to which their providers respected them, let them describe their contraceptive preferences, took their preferences seriously, and adequately informed them about their options [1]. Scores were associated with patient's overall satisfaction with the provider visit, satisfaction with method choice, and contraceptive method continuation 6 months after the visit. NSFG respondents who had received a method of birth control or contraceptive counseling in the past 12 months were asked to rate their most recent experience with their family planning provider on a five-point scale from “poor” to “excellent” on the four items from the PCCC scale 1) “How did this provider rate on respecting you as a person?”; 2) “How did this provider rate with respect to letting you say what mattered most to you about your birth control method?”; 3) “How did this provider rate on taking your preferences about birth control seriously?”; 4) “How did this provider rate on giving you enough information to make the best decision about your birth control method?” Based on the distribution of responses, for each question, we created a binary measure identifying respondents who answered “excellent” versus “very good,” “good,” “fair,” or “poor.” We also created a summary measure coded as 1 if women answered “excellent” to all four questions and 0 otherwise, a measure referred to as “overall person-centered care.”

While PCCC scale creators have noted that the summary measure (“excellent” across all four items) can be used as patient-reported outcome performance measure for person-centeredness in contraceptive counseling [1], the inclusion of these measures in a national survey provided a sufficiently large and diverse sample to explore associations between women's sociodemographic characteristics and their overall rating of their family planning experiences as well as their ratings on specific dimensions of person-centered care.

Sociodemographics and provider type

Women's characteristics of interest included a combination measure of race/ethnicity and self-reported English-speaking proficiency (non-Hispanic White, non-Hispanic Black, Hispanic-low English proficiency, Hispanic-high English proficiency). English-speaking proficiency was based on the question “How well do you speak English?” with “very well” coded as high English proficiency and all other responses (“not at all,” “not well,” “well”) coded as low English proficiency. This cut point matches the U.S. Census Bureau's designation of “limited English speaking households” [16]. Other measures included sexual orientation (women who identified as heterosexual/straight vs. women who identified as lesbian, gay, bisexual or “something else”), age (15–19, 20–29, 30+), whether the respondent ever had a live birth, income level based on the federal poverty line (FPL) (less than 100% FPL, 101–250% FPL, >250% FPL). We also examined family planning provider type (publicly funded vs. private doctor/HMO).

Analysis

We first calculated the percentage of women reporting overall person-centered care and the percentage reporting “excellent” on the individual PCCC measures, by each sociodemographic characteristic. We then used univariate logistic regression models to test for significant bivariate differences in women's ratings by sociodemographic characteristics. To understand the association between each sociodemographic measure and women's experience with person-centered counseling, controlling for the other measured characteristics, we ran multivariate logistic regression models. We did this in a stepwise fashion, the first model included demographic characteristics and the second model added women's income level and provider type. The stepwise modeling allowed us to assess whether associations between demographics (race/ethnicity and English proficiency, a continuous measure of age, parity, and sexual orientation) and person-centered care were attenuated by socioeconomic status and provider type. We also ran sensitivity checks in which we ran two additional versions of the second model (1) with only income added, and (2) with only provider type added. All analyses were run in Stata 16 and accounted for NSFG's complex survey design.1

Results

Sample characteristics and PCCC ratings

Approximately half the sample was White (58%), 14% were Black, 6% were Hispanic with low English proficiency, and 13% were Hispanic with high English proficiency (Table 1). Fifteen percent were adolescents (age 15–19), 42% were 20–29 and 44% were age 30 to 49. Almost half the sample ever had a live birth (45%) and 14% reported their sexual orientation as lesbian, gay, bisexual, or “something else” (LGBQ).2 Approximately one in five (21%) reported family income at or below 100% FPL and 28% had a family income between 101 and 250% FPL. More than three-quarters received family planning services from a private doctor.

Table 1

Sample characteristics, women who received a birth control method or counseling in the past year, National Survey of Family Growth 2017–2019.

	Weighted%	Unweighted N
Race/ethnicity & English proficiency
White	58%	1042
Hispanic, high English proficiency	13%	393
Hispanic, low English proficiency	6%	173
Black	14%	442
Othera	10%	192
Age
15–19	15%	365
20–29	42%	922
30–49	44%	955
Ever had a live birth
Yes	45%	1086
No	55%	1156
Sexual orientation
LGBQ	14%	315
Heterosexual	86%	1911
Income level
0-100% FPL	21%	555
101–250% FPL	28%	695
>250% FPL	51%	992
Provider type
Private doctor	79%	1683
Public clinic	17%	466
Othera	4%	93
Sample total		2242

Data note: The National Survey of Family Growth (NSFG), conducted by the National Center for Health Statistics (NCHS), gathers information on the reproductive health and family formation behaviors of U.S. men and women aged 15 to 49.

We do not report on family planning experiences for women in the Other race/ethnic category and for women receiving services from providers classified as Other.

The majority of women (59%−69%) rated their provider “excellent” across the four PCCC items and few rated their provider “poor” or “fair” (2%–4%) (Table 2). Overall, women were least likely to give their provider the highest rating on the information sharing item. Approximately half (51%) of women reported receiving overall person-centered care (“excellent” on all four items, not shown).

Table 2

Women's ratings of their family planning provider on the PCCC, National Survey of Family Growth, weighted%s.

	Respecting you as a person	Letting you say what mattered most to you about your birth control method	Taking your preferences about birth control seriously	Giving you enough information to make the best decision about your birth control method
	n = 2242	n = 2231	n = 2233	n = 2235
Poor	0%	1%	1%	1%
Fair	2%	3%	3%	4%
Good	10%	12%	11%	13%
Very Good	19%	20%	18%	22%
Excellent	69%	65%	68%	59%

Data note: The National Survey of Family Growth (NSFG), conducted by the National Center for Health Statistics (NCHS), gathers information on the reproductive health and family formation behaviors of U.S. men and women aged 15 to 49.

Bivariate findings

Bivariate analyses revealed differences in women's ratings on the PCCC by sociodemographics (Table 3). Across all items, Hispanic women with low English proficiency and Black women were the least likely to rate their provider as excellent while White women and Hispanic women with high English proficiency were the most likely. Younger age was associated with lower PCCC scores, with fewer adolescents rating their provider as excellent. Approximately half (51%–58%) of LGBQ women rated their provider as excellent across the PCCC items while more than 60% (61%–71%) of heterosexual women did so. Women who ever had a live birth were more likely than those who have not to report overall person-centered care but the only individual PCCC item to reach significance (p < 0.05) was the information sharing item. Income was strongly related to higher ratings on the PCCC. Women with incomes below the FPL were the least likely to give their providers an excellent rating while women with incomes above 250% FPL were the most likely. Receiving services from a public provider was associated with a lower likelihood of reporting overall person-centered care; however, the only individual item to reach significance was the respecting the patient as a person item.

Table 3

Percentage of women reporting that their provider was "Excellent" on each PCCC item and percentage reporting overall person-centered care, by sociodemographic characteristics and provider type, National Survey of Family Growth, weighted%s.

	Respecting you as a person		Letting you say what mattered most to you about your birth control method		Taking your preferences about birth control seriously		Giving you enough information to make the best decision about your birth control method		Overall person-centered care
	n = 2242		n = 2231		n = 2233		n = 2235		n = 2235
Race/ethnicity & English proficiency
White	72%		69%		71%		62%		54%
Hispanic, high English proficiency	74%		64%		68%		59%		51%
Hispanic, low English proficiency	51%	a	43%	a,b,c	47%	a,b,c	46%	a	35%	a,b
Black	60%	a,b	63%	a	63%	a	56%		46%	a
Age
15–19	62%		57%		61%		56%		41%
20–29	68%		64%		66%		55%		48%
30+	72%	d	69%	d	71%	d	64%	e	56%	d
Sexual orientation		*		*		*		*		*
LGBQ	58%		51%		55%		51%		38%
Heterosexual	71%		68%		70%		61%		53%
Ever had a live birth								*		*
No	67%		63%		65%		55%		46%
Yes	71%		68%		70%		64%		56%
Income
0–100% FPL	58%	f,g	53%	f,g	54%	f,g	49%	f,g	39%	f,g
101–250% FPL	68%	f	63%	f	64%	f	57%	f	48%	f
>250% FPL	74%		72%		75%		64%		57%
Provider type		*								*
Private	71%		67%		70%		60%		53%
Public	61%		60%		61%		56%		44%

Data note: The National Survey of Family Growth (NSFG), conducted by the National Center for Health Statistics (NCHS), gathers information on the reproductive health and family formation behaviors of U.S. men and women aged 15 to 49.

Significant difference (p<.05).

Different from White.

Different from Hispanic high English proficiency.

Different from Black.

Different from 15 to 19.

Different from 20 to 29.

Different from >250% FPL.

Different from 101 - 250% FPL.

Multivariate findings

Tables 4 and 5 show the results from multivariate logistic regression models regressing PCCC ratings on women's demographics (Model 1) and on women's demographics plus income level and provider type (Model 2). Findings for Model 1 were similar to the bivariates; Hispanic women with low English proficiency had lower odds of reporting “excellent” on the PCCC items (and overall person-centered care) than White women and Hispanic women with high English proficiency. Black women had lower odds of reporting “excellent” on all the PCCC items (and overall person-centered care) than White women and lower odds than Hispanic women with high English proficiency for the item related to the provider respecting them as a person. Women with older age had higher odds of reporting “excellent” across three of the four PCCC items and overall person-centered care. LGBQ women had lower odds of reporting “excellent” on the three PCCC items related to interpersonal communication as well overall person-centered care. In multivariate Model 1, having a live birth was not associated with overall person-centered care or with any individual item.

Table 4

Sociodemographic characteristics associated with PCCC rating on individual items, odds ratios from multivariate logistic regression models, National Survey of Family Growth 2017–2019.

	Respecting you as a person		Letting you say what mattered most to you about your birth control method		Taking your preferences about birth control seriously		Giving you enough information to make the best decision about your birth control method
	n = 2242		n = 2231		n = 2233		n = 2235
	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2
Race/ethnicity & English proficiency
Ref: White	ref	ref	ref	ref	ref	ref	ref	ref
Hispanic, high English proficiency	1.05 (0.68, 1.62)	1.12 (0.72, 1.74)	0.77 (0.52, 1.15)	0.81 (0.54, 1.21)	0.87 (0.60, 1.28)	0.94 (0.64, 1.37)	0.85 (0.58, 1.23)	0.88 (0.60, 1.28)
Hispanic, low English proficiency	0.34a (0.19, 0.62)	0.46a (0.26, 0.80)	0.28a (0.16, 0.48)	0.35a (0.21, 0.58)	0.30a (0.18, 0.51)	0.41a (0.25, 0.69)	0.42a (0.27, 0.66)	0.48a (0.30, 0.78)
Black	0.56a (0.39, 0.82)	0.63a (0.43, 0.94)	0.73 (0.54, 0.97)	0.81 (0.59, 1.13)	0.69 (0.51, 0.93)	0.80 (0.63, 1.29)	0.73 (0.53, 0.99)	0.79 (0.56, 1.10)
Age (continuous)	1.01 (1.00, 1.03)	1.00 (0.99, 1.02)	1.02 (1.00, 1.04)	1.01 (0.99, 1.03)	1.02 (1.00, 1.04)	1.00 (0.99, 1.03)	1.02 (1.00, 1.04)	1.01 (0.99, 1.03)
LGBQ	0.60 (0.38, 0.97)	0.65 (0.41, 1.03)	0.53 (0.37, 0.74)	0.56 (0.40, 0.78)	0.57 (0.39, 0.84)	0.61 (0.41, 0.91)	0.75 (0.54, 1.02)	0.77 (0.56, 1.06)
Ever had a live birth	1.12 (0.79, 1.59)	1.24 (0.87, 1.78)	1.10 (0.80, 1.51)	1.25 (0.91, 1.71)	1.16 (0.84, 1.60)	1.35 (0.98, 1.87)	1.29 (0.95, 1.73)	1.43 (1.06, 1.92)
Income
0–100% FPL		0.57 (0.41, 0.79)		0.53 (0.37, 0.76)		0.47 (0.32, 0.70)		0.59 (0.42, 0.83)
101–250%FPL		0.81 (0.60, 1.11)		0.74 (0.54, 1.02)		0.68 (0.50, 0.92)		0.76 (0.57, 1.00)
(Ref: >250% FPL)		ref		ref		ref		ref
Public provider		0.83 (0.54, 1.27)		1.04 (0.65, 1.66)		0.94 (0.58, 1.51)		1.12 (0.76, 1.63)

Note: Boldfaced odds ratios indicate p < 0.05.

Data note: The National Survey of Family Growth (NSFG), conducted by the National Center for Health Statistics (NCHS), gathers information on the reproductive health and family formation behaviors of U.S. men and women aged 15 to 49.

significantly different from Hispanic, high English proficiency (p < 0.05) with an odds ratio less than 1.

Table 5

Sociodemographic characteristics associated with PCCC rating of overall person-centered care, odds ratios from multivariate logistic regression models, National Survey of Family Growth 2017–2019.

	Overall person-centered care
	n = 2235
	Model 1	Model 2
Race/ethnicity & English proficiency
Ref: White	ref	ref
Hispanic, high English proficiency	0.87 (0.58, 1.28)	0.92 (0.61, 1.37)
Hispanic, low English proficiency	0.37a (0.22, 0.63)	0.48a (0.29, 0.80)
Black	0.69 (0.50, 0.93)	0.77 (0.56, 1.07)
Age (continuous)	1.02 (1.00, 1.04)	1.01 (0.99, 1.03)
LGBQ	0.62 (0.44, 0.88)	0.66 (0.47, 0.93)
Ever had a live birth	1.27 (0.94, 1.71)	1.43 (1.06, 1.93)
Income
0–100% FPL		0.54 (0.39, 0.75)
101–250% FPL		0.74 (0.57, 0.96)
(Ref: >250% FPL)		ref
Public provider		0.95 (0.65, 1.38)

Note: Boldfaced odds ratios indicate p < 0.05.

Data note: The National Survey of Family Growth (NSFG), conducted by the National Center for Health Statistics (NCHS), gathers information on the reproductive health and family formation behaviors of U.S. men and women aged 15 to 49.

significantly different from Hispanic, high English proficiency (p < 0.05) with an odds ratio less than 1.

In Model 2 we added income level and provider type to the model. For women with incomes below the poverty line, their odds of reporting “excellent” on the individual PCCC items and overall person-centered care were approximately half those of women with incomes >250% FPL (OR = 0.47–0.59). In multivariate models, we found that provider type was not associated with PCCC ratings; this was true in Model 2 and in the sensitivity checks in which income was not in the model. In Model 2, some of the significant associations between demographics and PCCC ratings were attenuated. The association between Black race and lower PCCC ratings was no longer significant for overall person-centered care and three of the four individual items. However, Black women continued to have lower odds than White women and Hispanic women with high English proficiency of rating their provider “excellent” in respecting them as a person. Additionally, the positive association between PCCC ratings and age was no longer significant. The associations between LGBQ identity and PCCC lower ratings stayed consistent, with the exception of negative association between LGBQ identity and ratings on respecting them as a person, which no longer reached significance. In Model 2, however, women who have had a live birth had higher odds of reporting their provider was excellent in providing information and overall person-centered care (matching the bivariate findings). In the sensitivity checks, we found that only the model that included income attenuated the associations between sociodemographic characteristics and PCCC ratings.

Discussion

This research contributes to the field by utilizing new nationally representative data to examine women's ratings of their family planning experiences using the recently developed PCCC. Our work illustrates how the scale functions in a large national sample and examines associations between sociodemographics and person-centered care across multiple characteristics. In multivariate analyses we found that women with lower incomes, LGBQ women, and Hispanic women who self-report low English-speaking proficiency gave their providers lower scores across the individual PCCC items and were less likely to report overall person-centered care, while having had a live birth was associated with higher PCCC scores. Our findings align with a variety of studies that use mostly qualitative data.

The negative association between LGBQ identification and ratings of person-centered care may reflect provider discrimination and lack of cultural competency, such as heteronormative protocols for sexual health counseling found in other research [12,13,17]. The lack of LGBTQ-inclusive care may be part of the reason sexual minority women are less likely than their straight counterparts to receive needed contraceptive counseling and gynecological care [18,19], highlighting the need for accessible and appropriate care for LGBTQIA populations [20,21]. Qualitative research has found LGBTQIA patients have more positive family planning visits when clinicians have greater knowledge about LGBTQIA experiences and ensure confidentiality of services [22].

The lower PCCC ratings among Hispanic women with low-English proficiency compared to both White women and Hispanic women with high-English proficiency highlight the importance of language concordance between providers and patients to achieve person-centered care. Higher Spanish proficiency among providers is linked to improved patient satisfaction, greater perceived choice of medical care and treatment decisions, and better perceived connection between physicians and Spanish-speaking patients [23,24]. Additionally, anti-immigration policies and discrimination [25,26] may foster medical mistrust among Hispanic women who are less proficient in English, contributing to barriers to care and lower quality experiences when they do receive services [27]. Findings highlight the need for family planning clinicians that are not only culturally competent, which has been outlined in previous research [28], but who also understand the structural factors that can impact the health outcomes of patients of color [29].

Black women rated their provider lower on respecting them as a person, which supports other research finding that clinician implicit racial bias can negatively impact the quality of interpersonal care for Black patients [30]. Lower patient ratings among Black women may also reflect greater mistrust of family planning providers due to racism in reproductive health care [31]. Patient-physician racial/ethnic discordance may also play a role. Though provider race is not measured in the NSFG, existing research shows that less than one-quarter of Black patients have racial concordance with their usual health care provider [32] and that Black patients often have more positive experiences with Black medical professionals [33,34], emphasizing the importance of having clinicians with similar backgrounds to the client population.

Our analyses found a strong negative association between lower income and all components of patient-centered care. Research assessing patients’ experiences with general health care finds that patients with lower incomes report greater difficulty accessing care, poor communication with providers, less shared decision-making, and lower satisfaction with care than higher earners [35]. Patients with lower incomes often receive care at clinics and hospitals with fewer health resources, which could reduce their ratings of care [36]. Our analyses, however, did not find an association between provider type (private doctor vs. a publicly funded clinic) and patient-centered care, net of sociodemographic characteristics, possibly due to the lack of detail on the family planning setting and individual clinician. Finally, we found that women who had a live birth gave their provider higher rates on information sharing and were more likely to report overall person-centered care. Parity and family planning experience has not been well-studied, but research in Mexico found a similar association. Authors posit that women who have previous experience with pregnancy and birth may have clearer expectations about contraceptive counseling or more self-efficacy to navigate their visit [37], which could positively influence their satisfaction.

Taken together, our findings point to a need for a patient-centered, shared decision-making approach to contraceptive counseling for all women seeking family planning services, particularly women who belong to groups that experience healthcare inequities. A person-centered approach is a primary component of high-quality family planning service delivery among federally funded providers [3]. Recommendations for achieving person-centered care include respecting the patient's primary purpose for the visit and delivering services in a culturally competent manner to meet the needs of all clients [3]. Improving the patient experience can help ensure that individuals access the specific family planning services they need, obtain contraceptive methods that align with their family planning desires, and continue to use needed services.

A limitation of our analysis is that we can only examine person-centered care among women who received a birth control method and/or contraceptive counseling in the last 12 months. Women's prior experiences with family planning care that may have discouraged them from seeking services more recently would not be captured, and we cannot study those who are unable or unwilling to access care at all (who often belong to the groups most marginalized by the healthcare system) [38]. A second limitation is the lack of detail on the individual clinicians and the family planning setting, factors that can play a role in patients’ overall care experience and may contribute to person-centered care [34,36]. Future research could better differentiate between these characteristics and person-centered counseling. Despite these limitations, this paper contributes to the field by providing findings on person-centered family planning from a national sample using a newly created clinical performance measure and highlighting disparities in women's experiences by sociodemographic characteristics.

Disclosures

Conflicts of interest: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding: This publication was made possible by Grant Number 1 FPRPA006070-01-00 from the Office of Population Affairs (OPA), U.S. Department of Health and Human Services (HHS). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of OPA or HHS