Contraception 103 (2021) 310–315

Contents lists available at ScienceDirect

Contraception
journal homepage: www.elsevier.com/locate/contraception

Original Research Article

Development of the Person-Centered Contraceptive Counseling scale

(PCCC), a short form of the Interpersonal Quality of Family Planning
care scale
Christine Dehlendorf a,b,c,∗, Edith Fox a, Ilana A Silverstein a, Alexis Hoffman a,
María Paula Campora Pérez a, Kelsey Holt a, Reiley Reed a, Danielle Hessler a
a
Department of Family & Community Medicine, University of California, San Francisco, CA, United States
b
Department of Obstetrics, Gynecology & Reproductive Sciences, University of California, San Francisco, CA, United States
c
Department of Epidemiology & Biostatistics, University of California, San Francisco, CA, United States

a r t i c l e i n f o a b s t r a c t

Article history: Objective: Person-centeredness is a critical component of quality in family planning. We previously val-
Received 20 July 2020 idated an 11-item Interpersonal Quality of Family Planning (IQFP) scale. We sought to create a parsimo-
Received in revised form 13 January 2021
nious version of the scale in preparation for testing its appropriateness as a patient-reported outcome
Accepted 20 January 2021
performance measure.
Study design: To explore clarity and importance of each of the 11 items, we conducted English and
Keywords: Spanish cognitive interviews with patients who received contraceptive counseling (n = 33) at 3 publicly
Contraception funded California clinics. We triangulated these results with psychometric analysis of previously collected
Contraceptive counseling IQFP data (n = 1097) to assess validity and reliability of selected item combinations.
Patient-centered
Results: The 11-item IQFP scale was reduced to a 4-item scale (the Person-Centered Contraceptive Coun-
Patient-reported outcome measure (PROM)
seling scale, or PCCC) that includes items evaluating provider performance regarding respect for patients,
Patient-reported outcome performance
measure (PRO-PM) information provision, and eliciting and honoring patient preferences for birth control. Interview partic-
Person-centered ipants deemed the items included in the 4-item PCCC important and clear in both English and Spanish
versions of the instrument. The 4-item PCCC retained the 11-item IQFP’s psychometric properties, in-
cluding internal consistency (Cronbach’s alpha = 0.92 vs 0.97 for the PCCC and IQFP, respectively) and
a consistent single factor analysis solution (factor loadings = 0.86–0.92 and 0.81–0.91). The 4-item PCCC
additionally retained the construct and predictive validity of the IQFP.
Conclusions: The 4-item PCCC is a valid and reliable as a measure of person-centered contraceptive coun-
seling that reflects patients’ perspectives on contraceptive counseling.
Implications: Person-centered measures such as the 4-item PCCC can help inform efforts to improve
health care quality. Future work will investigate the validity and reliability of the 4-item PCCC as a per-
formance measure to determine the appropriateness of its use in the quality improvement context.
© 2021 Elsevier Inc. All rights reserved.

1. Introduction
The measurement of health service quality is a growing pri-
ority across health care, including in the field of family planning.
Defining quality in contraceptive care is a uniquely complex issue,
as patients’ decisions around pregnancy, childbearing, and use of
contraceptive methods are highly individualized and personal [1].
How best to measure quality across a broad range of patient expe-
riences, preferences for care, and preferred outcomes can therefore
be challenging to determine.
Designing quality measures focused on patient-centeredness—
which by definition takes into account patient preferences—is im-
portant in all aspects of health care, with patient-centeredness
recognized by the National Academies of Medicine as a core do-
main of health care quality [2]. Given the personal nature of fam-
ily planning care, these types of measures are of particular im-
portance in this context, including the need for measures that

∗
Corresponding author.
E-mail address: christine.dehlendorf@ucsf.edu (C. Dehlendorf).

https://doi.org/10.1016/j.contraception.2021.01.008
0010-7824/© 2021 Elsevier Inc. All rights reserved.
C. Dehlendorf et al.
can evaluate people’s experience of contraceptive care. Traditional
outcomes used to denote success among providers of reproduc-
tive health care, including the prevention of unintended preg-
nancy and the use of highly effective methods, may not in fact
reflect patient priorities. Further, the history in the United States
of reproductive injustices against poor women and women of
color, including forced sterilization [3] and racially biased promo-
tion of highly effective methods [4], highlight the need to cen-
ter patient voices in definitions of quality from a health equity
perspective.
While there has been substantial work to define quality in fam-
ily planning globally for the past 30 years [5,6], it is only in the
past decade that defining and measuring quality in family planning
care has gained attention in the United States. In 2014, the Office of
Population Affairs and Centers for Disease Control and Prevention
published Quality Family Planning, an evidence-based set of rec-
ommendations for providing comprehensive, patient-centered fam-
ily planning care [7]. In 2016, the National Quality Forum (NQF),
a nationally recognized nonprofit organization which reviews and
endorses evidence-based health care performance measures, en-
dorsed the first three quality measures specific to family planning,
based on the proportion of women using highly and moderately
effective contraception [8,9]. The goal of these measures is to moti-
vate improved access to the most effective methods for all women
[10]. At the same time, these measures’ focus on method effective-
ness raised concerns among reviewers and commenters in the NQF
review process about the potential for these measures to incen-
tivize counseling toward more effective methods, particularly long-
acting reversible contraception (LARC). This incentive could result
in neglect of patient preferences and patient-centered care [11].
Discussions of these concerns resulted in the recognition of the
importance of having a measure of person-centeredness in con-
traceptive counseling, both to monitor for any adverse impacts of
measures focusing on effectiveness, and to ensure attention to pa-
tient experience more generally. Indeed, in their recently published
report, The Care We Need, NQF identified person-centeredness as
key to value-based payment systems [12]. As health care organi-
zations, insurance payers, and government agencies look to NQF
as a source of rigorously tested and vetted performance measures
for use in accountability programs, including public reporting and
pay-for-performance programs [13], the need for an NQF endorsed
measure of person-centeredness that can act as a balancing mea-
sure of the newly endorsed measures of contraceptive provision is
imperative.
The Interpersonal Quality of Family Planning Care (IQFP) scale
is a validated 11-item scale of patient experience of contracep-
tive counseling [14,15]. The IQFP addresses 3 domains of contra-
ceptive counseling preferences elucidated though qualitative inter-
views with patients: interpersonal connection, decision support,
and adequate information [16]. Notably, information and interper-
sonal connection were highlighted as important elements of qual-
ity from the client’s perspective in the seminal framework pub-
lished by Judith Bruce in 1990 [5]. In line with more recent iter-
ations of this framework which have sought to expand the con-
struct of quality to include elicitation of client preferences to pro-
vide personalized counseling [6], the IQFP construct covers deci-
sion support as an additional critical element of quality. A reduced
version of the IQFP that is more feasible for use in clinical qual-
ity improvement has the potential to serve as a patient-reported
outcome measure (PROM) for person-centeredness in contracep-
tive counseling [17,18]. When aggregated, this PROM could be used
as a patient-reported outcome performance measure (PRO-PM) of
person-centeredness at various levels of analysis (e.g., provider-
level, facility-level, or payer-level). We sought to create a parsimo-
nious version of the IQFP for use as a PROM, with the ultimate goal
of endorsement as a PRO-PM from the NQF.
311
Contraception 103 (2021) 310–315
Fig. 1. Iterative process for Person-Centered Contraceptive Counseling (PCCC) scale
item selection: Data from cognitive interviews, a previous cohort study, and a pre-
vious randomized control trial (RCT) fed into an iterative item reduction process to
produce the final PCCC scale.
2. Material and methods
We conducted cognitive interviews with patients who had re-
cently received contraceptive counseling, and psychometric anal-
ysis of previously collected IQFP data. Qualitative (including item
importance, clarity and language equivalence) and quantitative re-
sults (item total correlations, item response theory [IRT]) were tri-
angulated in an interactive process (Fig. 1) in order to reduce the
IQFP scale to a shortened version (the Person-Centered Contracep-
tive Counseling scale, or PCCC). This study was approved by the
Institutional Review Board of the University of California, San Fran-
cisco.
2.1. Cognitive interviews
We conducted cognitive interviews in English and Spanish be-
tween October 2016 and February 2017 in order to inform selection
of PCCC items. We recruited patients from three publicly funded
clinics providing family planning services in California. Eligible pa-
tients were female, age 15 to 45, able to speak and read in En-
glish, Spanish, or both languages, and had visits involving contra-
ceptive counseling on the day of recruitment. We purposively sam-
pled bilingual English- and Spanish-speaking patients in order to
understand the equivalence of items across the 2 languages.
Research staff approached patients either before or after visits,
depending on clinic flow, and assessed patient interest in complet-
ing an interview about their care after their visit. Staff screened
interested patients for eligibility and asked for patients’ informed
consent to participate. Consenting patients were interviewed that
day or, if the patient did not have time, typically within 2 to 3
days, and within a maximum of 3 weeks postvisit in order to min-
imize recall bias.
Before interviews, participants completed a brief demographic
survey. During interviews, participants were presented with the
full 11-item IQFP and asked to respond to it in reference to their
visit (Table 2). Using a “think-aloud” technique for understanding
cognitive response [19,20], research staff asked participants to de-
scribe their rationale for their response to each item. Think-aloud
responses provided information on participants’ initial reactions to
items, the clarity of item language to participants, and their in-
terpretations of item meaning. During the think-aloud process, re-
search staff asked follow-up questions to probe for item clarity, po-
tential repetitiveness of items, and reasoning behind item scores.
After completing the survey and think-aloud process, research staff
asked a series of open-ended questions about the overall experi-
ence of taking the survey, participant understanding of the survey’s
C. Dehlendorf et al. Contraception 103 (2021) 310–315

purpose, any repetitiveness of items, and any important missing Table 1

Cognitive interview participant demographics.
constructs not addressed in the survey.
Participants were then asked to rank each item with regard to Characteristic n (%)
their importance, with 1 indicating the most important and 11 in- Language of interview
dicating the least important. Participants were asked to verbally Spanish 13 (39.4)
explain their rationale for item rankings. English 10 (30.3)
Bilingual participants completed an interview in the language Bilingual Spanish and English 10 (30.3)
Age
in which they were most comfortable seeking medical care. After
15–19 6 (18.2)
reviewing the IQFP in this language, they were shown the English 20–29 17 (51.5)
and Spanish IQFP side by side and asked to describe whether or 30–39 9 (27.3)
not they thought the overall survey and each item had equivalent 40–45 1 (3.0)
Race/ethnicitya
meanings as presented in each language.
American Indian/Alaska Native 3 (9.1)
We used content analysis [21] to understand participant re- Asian 2 (6.1)
sponses to each item and their overall experience of taking the Black/African American 3 (9.1)
survey. We calculated mean rankings of items to quantitatively White 4 (12.1)
understand ranked item importance. Qualitative and quantitative Hispanic or Latina 25 (75.8)
Other 1 (3.0)
analyses were conducted iteratively, with emerging combinations
Parental education
of items informing additional exploration through cognitive inter- 8th grade or less 12 (36.4)
views. High school graduate or some high school 9 (27.3)
4-year college graduate or some college 9 (27.3)
More than 4-year college graduate 3 (9.1)
Number of pregnancies
2.2. Psychometric analysis 0 10 (30.3)
1 8 (24.5)
We performed psychometric analysis to test the validity and re- ≥2 15 (45.5)
liability of potential item combinations. Analyses were performed Number of births
0 15 (45.5)
using survey data from 2 previous studies: a cohort study of 1 7 (21.2)
women’s experience of contraceptive counseling [22] (2009–2012; ≥2 11 (33.3)
n = 347) and a randomized control trial of a decision support Birth control method chosen at visit
tool for contraception [23] (2014–2016; n = 730) (total n for psy- Pill 9 (27.3)
Patch 1 (3.0)
chometric analysis = 1077). These studies included the IQFP in
Ring 0
postvisit surveys, as well as items about the experience of con- Shot 3 (9.1)
traceptive counseling and contraceptive use in immediate postvisit Hormonal IUD 4 (12.1)
and follow-up surveys. Before analyzing reduced item combina- Non-hormonal IUD 5 (15.2)
tions, we used a 2-parameter IRT model [24] with all 11 items to Implant 7 (21.2)
Condoms 1 (3.0)
examine whether substantial differences existed in items’ difficulty None 2 (6.1)
or discrimination levels. Did not decide 1 (3.0)
Combinations of 3 to 6 IQFP items were examined as potential Total 33 (100)
reduced measures based on cognitive interview findings and our a
Percentages add up to <100 as some participants re-
assessment that this was an acceptable length for a PRO-PM that ported more than one race/ethnicity.
could be feasibly implemented in the context of contraceptive care.
Combinations included items that were clear in meaning with no
concerns around language equivalence, rated at or above the scale 3. Results
mid-point of 6 (1–11 scale) of participant importance. We also pri-
oritized including items across three domains of patient-centered 3.1. Cognitive interviews
contraceptive counseling that had been identified in previous re-
search by our team (interpersonal connection, adequate informa- A total of 33 participants completed cognitive interviews. Ten
tion, and decision support) [16]. Internal consistency of reduced interviews were conducted primarily in English and 13 were con-
item combinations was tested using Cronbach’s alpha. Factor load- ducted primarily in Spanish, and 10 interviews were conducted
ings were used to test for consistency of a single factor analysis so- with bilingual participants, with substantial discussion in both
lution. External construct validity was tested by examining the as- English and Spanish. Participant demographics are depicted in
sociation of item combination responses with other items included Table 1. The mean age of participants was 26 years old. The ma-
in postvisit patient surveys. These included measures of global sat- jority of participants (75.8%) identified as Hispanic or Latina. Most
isfaction with the visit, satisfaction with the method selection pro- participants (70.7%) had at least 1 previous pregnancy, and 54.5%
cess, satisfaction with method choice, and recommendation of the had previously given birth. Nearly half of participants (48.5%) had
provider to a friend. In order to replicate the original IQFP measure chosen to use a LARC method at their most recent appointment.
validation, predictive validity was also tested by examining the as- Overall, participants accurately described that the 11-item sur-
sociation between item combination responses and follow-up data vey intended to measure patient experience and reported that it
of contraceptive use at 6 months, using the data from the cohort was easy to complete. No domains emerged as notably missing
study. Cronbach’s alpha and measures of external validity for each from the survey with regard to capturing person-centeredness of
item combination were compared to findings that validated the contraceptive counseling.
original IQFP, to understand how well each combination retained Think-aloud responses indicated that participants were gen-
the original’s properties [14,15]. The IQFP is scored dichotomously, erally able to respond to items and describe the reasoning for
as either a “top-box” score of 55 (where the participant marks 5, their responses, but that clarity of items varied. For example,
the highest score, for every item) vs less than 55. We calculated several participants had relative difficulty responding to Item 6
the sensitivity and specificity of top-box scores of reduced item (Table 2) because of varying interpretations of the phrase “per-
combinations for a top-box IQFP score. sonal situation.” Some participants understood “personal situation”

312
C. Dehlendorf et al.
Table 2
Participant rankings of the Interpersonal Quality of Family Planning (IQFP) scale item importance.
IQFP item
(1) Respecting me as a person
(2) Showing care and compassion
(3) Letting me say what mattered to me about my birth control method
(4) Giving me the opportunity to ask questions
(5) Taking my preferences about my birth control seriously
(6) Considering my personal situation when advising me about birth control
(7) Working out a plan for my birth control with me
(8) Giving me enough information to make the best decision about my birth control method
(9) Telling me how to take or use my birth control method most effectively
(10) Telling me the risks and benefits of the birth control method I chose
(11) Answering all my questions
to refer to someone’s demographic characteristics, such as age
or race/ethnicity, and expressed disapproval of these characteris-
tics influencing a provider’s advice on birth control. One Spanish-
speaking participant understood “personal” in this context to mean
the privacy of various methods. An English-speaking participant
expressed that “personal” was vague in this context, stating: “I
don’t really know what my personal situation is … are you talk-
ing about me being a grad student, a dancer, a person who is dat-
ing? When I read the question, I can’t really think of anything I’d
put under this.” Item 7 was also frequently cited as confusing. Sev-
eral participants said they did not think having a “plan” applied to
them because they would be using a LARC method requiring little
patient effort after insertion. Participants generally understood the
introductory instructions to the measure. However, the research
team noted that reading level for this text was higher than desired,
and revised this text for the resulting shortened measure.
When asked which of the 11 items might be repetitive, partic-
ipants often replied that Items 8 and 10, 5 and 6, and 4 and 11
were repetitive item pairs, and that there was overlap between 3
and 4. In making final item combination selections for quantitative
analysis, we avoided including item pairings described as repetitive
by interview participants, as well as those items most frequently
described as unclear. Participant rankings of item important are
depicted in Table 2, and highly ranked items were prioritized in
resulting item combinations.
Interviews with bilingual participants indicated general equiv-
alence of items across language. As emerging results indicated
which items would be appropriate for inclusion in the PCCC, we
explored improving the translations on these specific items. Several
bilingual and Spanish-monolingual participants provided recom-
mendations to improve the clarity of item 5, and bilingual partic-
ipants confirmed these changes improved the equivalence of item
meaning in English and Spanish.
3.2. Psychometric analysis
Full descriptions of the 2 samples used in analyses are de-
scribed elsewhere [22,23]. Two-parameter logistic IRT model re-
sults indicated that item responses did not vary substantially
with regard to difficulty (range 7.64–12.53) or discrimination lev-
els (range −0.72–−0.92), and therefore no items were excluded or
specifically selected based on the IRT results. Twelve combinations
of 3 to 6 of items were selected for initial psychometric testing on
the basis of items receiving adequate patient importance ratings
(items 7, 9, and 11 not considered given ratings >6), and adequate
representation across the three domains. Multiple psychometric in-
dicators remained consistently strong across the item groupings in-
cluding: consistent item-total correlations (all > 0.70), Cronbach’s
alpha (0.88–0.94), and factor loadings (all > 0.70). While associ-
ations with construct validity items were overall consistent across
313
Contraception 103 (2021) 310–315
Average ranking (1 = most important,
Domain 11 = least important)
Interpersonal 3.20
Interpersonal 5.55
Interpersonal 5.35
Decision-making 5.36
Decision-making 5.73
Decision-making 6.04
Decision-making 7.70
Information 5.33
Information 6.86
Information 5.00
Information 7.39
the combinations examined, predictive validity of the measure was
maintained by 5 of the item combinations. Given qualitative and
investigator concerns around clarity of items 4 and 6, combina-
tions including these items were removed from further consider-
ation. The 2 resulting options had nearly identical psychometric
properties and thus a final solution was selected on the basis of
parsimony of items.
3.3. Final measure
The PCCC (Table 4) includes 4 items (items 1, 3, 5, and 8) that
were indicated in both qualitative and quantitative analysis to re-
tain the IQFP’s measurement of 3 conceptual domains of person-
centeredness in contraceptive counseling (Interpersonal Connec-
tion, Information Sharing, and Decision Support) (Table 4). In cog-
nitive interviews, participants found these items to be evocative of
both their overall patient experience and specific exchanges and
aspects of their appointment that influenced that experience, with
included items scoring among those most important (all items re-
ceived a score of < 5.50). All included items were deemed under-
standable and equivalent between Spanish and English. This com-
bination retained all of the psychometric properties of the IQFP,
including strong internal consistency (Cronbach’s alpha = 0.92 vs
0.97 for the PCCC and IQFP, respectively), a consistent single fac-
tor analysis solution (factor loadings 0.86–0.92 and 0.81–0.91 for
the PCCC and IQFP, respectively), and high item-total correlations
(0.73–0.82 and 0.75–0.84 for the PCCC and IQFP, respectively). Di-
rectly comparing PCCC scored responses (dichotomous score of 20
vs <20) to the 11-item standard score (dichotomous score of 55
vs <55), yielded a specificity of 1.00 and sensitivity of 0.80 for
the PCCC. Validity of the PCCC was comparable to the IQFP, with
all previous significant associations between scored responses with
external validity items retained and similar in strength (Table 3).
Both the IQFP and PCCC are associated with measures of con-
struct validity (high global satisfaction with visit, satisfaction with
method selection process, p < 0.001 for IQFP and PCCC, satisfac-
tion with method choice, recommendation of provider to friend, p
< 0.01), and predictive validity, with high IQFP and PCCC scores
predictive of contraceptive continuation at 6 months (p < 0.05).
4. Discussion
Our iterative mixed-methods approach to the reduction of
the 11-item IQFP scale produced a brief measure of person-
centeredness in contraceptive care that retains the validity of the
original scale and reflects patient preferences for counseling. Com-
pared with the IQFP, which was developed for the context of re-
search, the 4-item PCCC scale is more feasible for use in quality
improvement, where it is critical to minimize the burden of data
collection on patients and administrators. Like the IQFP, the PCCC
C. Dehlendorf et al. Contraception 103 (2021) 310–315

Table 3
Construct and predictive validity of the 4-item Person-Centered Contraceptive Counseling (PCCC) scale compared to the 11-item Interpersonal Quality of Family Planning
(IQFP) scale.

IQFP score IQFP PCCC high PCCC

Validity type Description high (55) (<55) (20) (<20) IQFP adj OR (95% CI) PCCC adj OR (95% CI)

Construct validity % High global visit satisfaction 96.2% 61.8% 95.3% 55.0% 14.80 (9.29, 23.60)c 16.33 (10.71, 24.90)c
Construct validity % Completely satisfied with method selection process 83.8% 45.1% 80.9% 41.1% 8.35 (5.25, 13.27)c 7.73 (4.77, 12.54)c
Construct validity % Very/completely satisfied with method choice 69.9% 48.5% 68.2% 46.5% 1.97 (1.14, 3.38)a 2.02 (1.28, 3.19)b
Construct validity % Would recommend to friend 96.0% 63.6% 94.4% 55.7% 14.90 (5.95, 37.32)c 15.56 (6.10, 39.68)c
Predictive validity Contraception method continuation at 6 months 45.6% 36.1% 45.3% 35.0% 1.74 (1.17, 2.59)a 1.58 (1.01, 2.53)a

Adjusted logistic regression odds ratios accounting for clustering by provider (random effect) and clinic (fixed effect), age (continuous), race/ethnicity, federal poverty level.
a
p < 0.05.
b
p < 0.01.
c
p < 0.001.

Table 4
Final Person-Centered Contraceptive Counseling (PCCC) scale.

Think about your visit. How do you think [provider name] did? Please rate them on each of
the following by circling a number. Poor Fair Good Very good Excellent

Respecting me as a person 1 2 3 4 5
Letting me say what mattered to me about my birth control method 1 2 3 4 5
Taking my preferences about my birth control seriously 1 2 3 4 5
Giving me enough information to make the best decision about my birth control method 1 2 3 4 5

For printable versions of the survey in English and Spanish, see the appendices online or visit https://pcccmeasure.ucsf.edu/.

scale can be interpreted dichotomously, with respondents giving
either a top score of 5 on all 4 items, or less than this top-box
score. A simple metric for person-centeredness can be calculated
as the percentage of respondents who gave the top-box score on
the PCCC scale at a given level of aggregation (e.g., provider or
facility). As such, the four items of the PCCC scale constitute a
PROM for person-centeredness and, when aggregated, a PRO-PM.
As frameworks and recommendations for contraceptive care qual-
ity developed both in the United States and globally define person-
centeredness as a core dimension of contraceptive care quality, this
metric can provide actionable data to improve contraceptive care
services [5–7].
The PCCC is designed to be used to assess the quality of care
in any organization that offers contraceptive counseling to its pa-
tients in either a family-planning-specific or primary care setting.
These settings may include public health department clinics, inde-
pendent nonprofit care providers, Community Health Centers, pub-
lic or private hospitals, or other private entities. In addition to its
potential for use in research as a parsimonious measure of patient
experience, if found to be valid and reliable as a PRO-PM, the PCCC
can offer these organizations a unique opportunity to measure the
quality of their patients’ experience of contraceptive counseling.
The use of this measure in quality improvement efforts can include
both as a stand-alone measure of person-centeredness and along-
side contraceptive provision measures as means to track changes
in person-centeredness with changes in provision. This may be of
particular benefit in Community Health Centers, where there has
been increased focus on building capacity for family planning care
in the years following passage of the Affordable Care Act [25–
29]. Additionally, as many health care systems move toward value-
based payment systems, the PRO-PM could be tied to payment
for improvements to the quality of contraceptive counseling. The
implications for increased attention and value placed on person-
centeredness in contraceptive counseling has the potential to coun-
teract, or at least call attention to, the ongoing legacy of implicit
racist bias and coercion in contraceptive care [4].
Limitations of this work include that our cognitive interviews
informing item reduction were conducted with a relatively small
number of participants, and only 9% of these participants were
Black, limiting generalizability. By triangulating interview data
with a larger sample of quantitative IQFP data that was more rep-
resentative by race/ethnicity [15,23], we ensured that item selec-
tion reflected a large number of patient responses, while still col-
lecting rich qualitative data to guide the selection process. There
was also potential for social desirability bias in interviews if par-
ticipants perceived that the research team was affiliated with the
providers from whom they received care. While social desirabil-
ity may have affected participant responses to the items, it would
have less impact on participants’ item interpretations. Our quanti-
tative analyses used the same data source as the validation study
for the original IQFP, which serves as a strength in our ability to
compare shorter and longer versions of the measure directly, but
also a limitation in that our analysis here did not demonstrate re-
producibility.
The PCCC provides a valid and reliable approach to measur-
ing person-centeredness in contraceptive counseling that can be
implemented in a range of care settings, including primary care
and specialized reproductive health clinics. Development of PROMs
such as the PCCC provides an opportunity to ensure that patient
voices and preferences are prioritized in clinical and translational
research and in quality improvement.
Declaration of competing interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared to
influence the work reported in this paper.
Funding
The research reported in this publication was funded through
the Office of Population Affairs (grant number 1 FPRPA006062-01-
00) and an anonymous foundation.
Acknowledgments
This work would not have been possible without the input of
our Patient Stakeholder group, which provided feedback on inter-
view guides, IQFP items, survey design, and implementation work-
flows.

314
C. Dehlendorf et al. Contraception 103 (2021) 310–315

Appendix A. Printable version of the 4-item Person-Centered References

Contraceptive Counseling (PCCC) scale in English
[1] Gomez AM, Wapman M. Under (implicit) pressure: young Black and Latina
women’s perceptions of contraceptive care. Contraception 2017;96:221–6.
Appendices A and B [2] Institute of Medicine Committee on Quality of Health Care in A Crossing the
quality chasm: a new health system for the 21st century. Washington (DC):
This survey is about your recent visit. Your answers are pri- National Academies Press (US); 2001.
vate. Your health care team will not see your individual an- [3] Stern AM. Sterilized in the name of public health: race, immigration, and re-
swers. productive control in modern California. Am J Public Health. 2005;95:1128–38.
[4] Dehlendorf C, Anderson N, Vittinghoff E, Grumbach K, Levy K, Steinauer J.
Did you and a member of the health care team talk about your Quality and content of patient-provider communication about contraception:
birth control options during your visit? differences by race/ethnicity and socioeconomic status. Womens Health Issues
2017;27:530–8.
[5] Bruce J. Fundamental elements of the quality of care: a simple framework.
Yes No
Stud Fam Plan 1990;21:61–91.
[6] Jain AK, Hardee K. Revising the FP quality of care framework in the context of
Whether you answered ‘Yes’ or ‘No’ to the question above,
rights-based family planning. Stud Fam Plan 2018;49:171–9.
we want to hear from you! Please answer the rest of the survey [7] Gavin L, Moskosky S, Carter M, Curtis K, Glass E. Providing quality family plan-
as best as you can. It is okay to leave a question blank if you ning services — recommendations from CDC and the U.S. Office of Population
Affairs. MMWR Morb Mortal Wkly Rep 2014;63:1–29.
are unable to answer.
[8] Performance measures. https://www.hhs.gov/opa/performance-measures/
index.html: Office of Population Affairs.
Think about your visit. How do [9] Gavin L, Frederiksen B, Robbins C, Pazol K, Moskosky S. New clinical perfor-
you think the member of the mance measures for contraceptive care: their importance to healthcare quality.
health care team did? Please rate Contraception 2017;96:149–57.
them on each of the following by [10] Law A, Yu JS, Wang W, Lin J, Lynen R. Trends and regional variations in
circling a number. Poor Fair Good Very good Excellent provision of contraception methods in a commercially insured population
in the United States based on nationally proposed measures. Contraception.
Respecting me as a person 1 2 3 4 5 2017;96:175–82.
[11] Gomez AM, Fuentes L, Allina A. Women or LARC first? Reproductive autonomy
Letting me say what mattered 1 2 3 4 5 and the promotion of long-acting reversible contraceptive methods. Perspect
to me about my birth control Sex Reprod Health. 2014;46:171–5.
method [12] Force NQT. The care we need: driving better health outcomes for people and
communities. National Quality Forum; 2020.
Taking my preferences about 1 2 3 4 5 [13] Gavin LE, Ahrens KA, Dehlendorf C, Frederiksen BN, Decker E, Moskosky S. Fu-
my birth control seriously ture directions in performance measures for contraceptive care: a proposed
framework. Contraception. 2017;96:138–44.
Giving me enough information 1 2 3 4 5
[14] Dehlendorf C, Henderson JT, Vittinghoff E, Steinauer J, Hessler D. Development
to make the best decision about
of a patient-reported measure of the interpersonal quality of family planning
my birth control method care. Contraception. 2018;97:34–40.
[15] Dehlendorf C, Henderson JT, Vittinghoff E, et al. Association of the quality of
interpersonal care during family planning counseling with contraceptive use.
Am J Obstet Gynecol 2016;215:78 e1-9.
Appendix B. Printable version of the 4-item Person-Centered [16] Dehlendorf C, Levy K, Kelley A, Grumbach K, Steinauer R. Women’s pref-
erences for contraceptive counseling and decision making. Contraception.
Contraceptive Counseling (PCCC) scale in Spanish
2013;88:250–6.
[17] Cella D, Hahn EA, Jensen S, Butt Z, Nowinski C, Rothrock N. Methodological
Esta encuesta es sobre su visita reciente. Sus respuestas son issues in the selection, administration and use of patient-reported outcomes in
privadas. Su equipo de atención médica no verán sus respuestas performance measurement in health care settings. Washington, DC: Prepared
by Northwestern University for NQF; 2012.
individuales. [18] Deutsch A, Smith L, Gage B, Kelleher C, Garfinkel D. PRO-based performance
¿Usted y un miembro del equipo de atención médica hablaron measures for healthcare accountable entities. Washington, DC: Prepared by RTI
de sus opciones de métodos anticonceptivos durante su visita? International and Brookings Institution for NQF; 2012.
[19] Fonteyn ME, Kuipers B, Grobe SJ. A description of think aloud method and
Sí No protocol analysis. Qual Health Res 1993;3:430–41.
[20] van Someren MW, Barnard YF, Sandberg JAC. The think aloud method: a
practical approach to modelling cognitive processes. London: Academic Press;
Independientemente de si usted contestó “Sí” o “No” a la pre- 1994.
gunta anterior, a nosotros nos gustaría saber de usted. Por favor, [21] Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual
Health Res 2005;15:1277–88.
complete el resto de la encuesta lo mejor que pueda. Está bien de- [22] Dehlendorf C, Tharayil M, Anderson N, Gbenedio K, Wittman A, Steinauer J.
jar la pregunta en blanco si no puede responder. Counseling about IUDs: a mixed-methods analysis. Perspect Sex Reprod Health
2014;46:133–40.
[23] Dehlendorf C, Fitzpatrick J, Fox E, Holt K, Vittinghoff E, Reed R, et al. Cluster
Piense en su visita. ¿Cómo le fue randomized trial of a person-centered contraceptive decision support tool, My
la visita con el miembro de Birth Control. Am J Obstet Gynecol 2019;220:565 e1-.e12.
equipo de atención médica? Por [24] Harris D. Comparison of 1-, 2-, and 3-parameter IRT models. Educ Meas Issues
favor califíquelo(s) con respecto a Pract 1989;8:35–41.
las siguientes cualidades [25] Shah SD, Prine L, Waltermaurer E, Rubin SE. Feasibility study of family plan-
marcando con un círculo en el ning services screening as clinical decision support at an urban Federally Qual-
número que corresponda. Malo Regular Bueno Muy bueno Excelente ified Health Center network. Contraception. 2019;99:27–31.
[26] Pace LE, Dolan BM, Tishler LW, Gooding HC, Bartz D. Incorporating long-acting
Respetarme como persona 1 2 3 4 5 reversible contraception into primary care: a training and practice innovation.
Womens Health Issues 2016;26:131–4.
Permitirme decir lo que más 1 2 3 4 5 [27] Shin P, Sharac J, Rosenbaum S. Community health centers and Medicaid at 50:
me importa sobre mi método an enduring relationship essential for health system transformation. Health Aff
anticonceptivo 2015;34:1096–104.
[28] Goldberg DG, Wood SF, Johnson K, Mead KH, Beeson T, Lewis J, et al. The orga-
Tomar en cuenta mis 1 2 3 4 5
nization and delivery of family planning services in community health centers.
preferencias sobre los
Womens Health Issues 2015;25:202–8.
anticonceptivos [29] Darney BG, Jacob RL, Hoopes M, Rodriguez MI, Hatch B, Marino M, et al. Eval-
Entregarme la información 1 2 3 4 5 uation of Medicaid expansion under the affordable care act and contraceptive
care in US Community Health Centers. JAMA Netw Open 2020;3:e206874.
suficiente para tomar la mejor
decisión sobre mi método
anticonceptivo

315