Advertisement

Development of the person-centered prenatal care scale for people of color

Open AccessPublished:April 13, 2021DOI:https://doi.org/10.1016/j.ajog.2021.04.216

      Background

      Given the stark disparities in maternal mortality and adverse birth outcomes among Black, indigenous, and other people of color, there is a need to better understand and measure how individuals from these communities experience their care during pregnancy.

      Objective

      This study aimed to develop and validate a tool that can be used to measure person-centered prenatal care that reflects the experiences of people of color.

      Study Design

      We followed standard procedures for scale development—integrated with community-based participatory approaches—to adapt a person-centered maternity care scale that was initially developed and validated for intrapartum care in low-resource countries to reflect the needs and prenatal care experiences of people of color in the United States. The adaptation process included expert reviews with a Community Advisory Board, consisting of community members, community-based health workers, and social service providers from San Francisco, Oakland, and Fresno, to assess content validity. We conducted cognitive interviews with potential respondents to assess the clarity, appropriateness, and relevance of the questions, which were then refined and administered in an online survey to people in California who had given birth in the past year. Data from 293 respondents (84% of whom identified as Black) who received prenatal care were used in psychometric analysis to assess construct and criterion validity and reliability.

      Results

      Exploratory factor analysis yielded 3 factors with eigenvalues of >1, but with 1 dominant factor. A 34-item version of the person-centered prenatal care scale was developed based on factor analyses and recommendations from the Community Advisory Board. We also developed a 26-item version using stricter criteria for relevance, factor loadings, and uniqueness. Items were grouped into 3 conceptual domains representing subscales for “dignity and respect,” “communication and autonomy,” and “responsive and supportive care.” The Cronbach alphas for the 34-item and the 26-item versions and for the subscales were >0.8. Scores based on the sum of responses for the 2 person-centered prenatal care scale versions and all subscales were standardized to range from 0 to 100, where higher scores indicate more person-centered prenatal care. These scores were correlated with global measures of prenatal care satisfaction suggesting good criterion validity.

      Conclusion

      We present 2 versions of the person-centered prenatal care scale: a 34-item and a 26-item version. Both versions have high validity and reliability in a sample made up predominantly of Black women. This scale will facilitate measurement to improve person-centered prenatal care for people of color and could contribute to reducing disparities in birth outcomes. The similarity with the original scale also suggests that the person-centered prenatal care may be applicable across different contexts. However, validation with more diverse samples in additional settings is needed.

      Key words

      Introduction

      Adverse birth outcomes, including preterm birth and low birthweight, disproportionately affect Black, indigenous, and other people of color, often irrespective of income or education.
      • Braveman P.
      • Heck K.
      • Egerter S.
      • et al.
      Worry about racial discrimination: a missing piece of the puzzle of black-white disparities in preterm birth?.
      ,
      • Vines A.I.
      • Baird D.D.
      • McNeilly M.
      • Hertz-Picciotto I.
      • Light K.C.
      • Stevens J.
      Social correlates of the chronic stress of perceived racism among black women.
      Black people, in particular, have a 2-fold higher risk of preterm birth and a 3-fold higher risk of maternal mortality than their White counterparts.,
      • Neggers Y.H.
      Trends in maternal mortality in the United States.
      Socially determined factors such as racism and discrimination across the life course have been shown to adversely affect women and their pregnancies.
      • Slaughter-Acey J.C.
      • Caldwell C.H.
      • Misra D.P.
      The influence of personal and group racism on entry into prenatal care among African American women.
      • Alhusen J.L.
      • Bower K.M.
      • Epstein E.
      • Sharps P.
      Racial discrimination and adverse birth outcomes: an integrative review.
      • Alio A.P.
      • Richman A.R.
      • Clayton H.B.
      • Jeffers D.F.
      • Wathington D.J.
      • Salihu H.M.
      An ecological approach to understanding black-white disparities in perinatal mortality.
      • Dominguez T.P.
      Adverse birth outcomes in African American women: the social context of persistent reproductive disadvantage.
      • Mehra R.
      • Boyd L.M.
      • Ickovics J.R.
      Racial residential segregation and adverse birth outcomes: a systematic review and meta-analysis.
      • Nuru-Jeter A.
      • Dominguez T.P.
      • Hammond W.P.
      • et al.
      "It’s the skin you’re in”: African-American women talk about their experiences of racism. an exploratory study to develop measures of racism for birth outcome studies.
      Moreover, emerging evidence suggests that issues such as disrespect, abuse, and discrimination within the healthcare system play major roles in how people of color experience and therefore access care during pregnancy, during birth, and after delivery, subsequently influencing outcomes for mothers and babies.
      • Attanasio L.
      • Kozhimannil K.B.
      Patient-reported communication quality and perceived discrimination in maternity care.
      • Ruiz R.L.
      • Shah M.K.
      • Lewis M.L.
      • Theall K.P.
      Perceived access to health services and provider information and adverse birth outcomes: findings from LaPRAMS, 2007-2008.
      • Salm Ward T.C.
      • Mazul M.
      • Ngui E.M.
      • Bridgewater F.D.
      • Harley A.E.
      “You learn to go last”: perceptions of prenatal care experiences among African-American women with limited incomes.
      • Vedam S.
      • Stoll K.
      • Taiwo T.K.
      • et al.
      The Giving Voice to Mothers study: inequity and mistreatment during pregnancy and childbirth in the United States.

       Why was this study conducted?

      This study aimed to adapt and validate a scale to measure person-centered prenatal care (PCPC) that adequately captures the experiences of people of color in the United States.

       Key findings

      Two valid and reliable versions of a PCPC scale were developed—a 34-item and a 26-item version—with subscales for “dignity and respect,” “communication and autonomy,” and “responsive and supportive care.” Both versions have high validity and reliability in a sample of predominantly Black women.

       What does this add to what is known?

      Our scale extends ongoing efforts to measure and improve pregnancy-related healthcare experiences of people of color, which can contribute to reducing disparities in birth outcomes.
      People of color have repeatedly described care experiences that are disrespectful and result in feelings of loss of autonomy and self-determination. Qualitative studies have found that they feel powerless to make decisions and desire a more active role in their care,
      • Ebert L.
      • Bellchambers H.
      • Ferguson A.
      • Browne J.
      Socially disadvantaged women’s views of barriers to feeling safe to engage in decision-making in maternity care.
      ,
      • Altman M.R.
      • Oseguera T.
      • McLemore M.R.
      • Kantrowitz-Gordon I.
      • Franck L.S.
      • Lyndon A.
      Information and power: women of color’s experiences interacting with health care providers in pregnancy and birth.
      highlighting the need for more person-centered care—care that is respectful and responsive to patients’ preferences, needs, and values.
      Institute of Medicine
      Crossing the Quality Chasm a New Health System for the 21st Century.
      Studies have described what people of color desire in their healthcare providers, outlining how person-centered care expectations vary based on cultural and societal factors.
      • Lori J.R.
      • Yi C.H.
      • Martyn K.K.
      Provider characteristics desired by African American women in prenatal care.
      • Cuevas A.G.
      • O’Brien K.
      • Saha S.
      What is the key to culturally competent care: reducing bias or cultural tailoring?.
      • Altman M.R.
      • McLemore M.R.
      • Oseguera T.
      • Lyndon A.
      • Franck L.S.
      Listening to women: recommendations from women of color to improve experiences in pregnancy and birth care.
      Although qualitative descriptions of these types of interactions are essential, standardized quantitative measures would allow for monitoring experiences during prenatal care across different populations and contexts and over time.
      • Afulani P.A.
      • Phillips B.
      • Aborigo R.A.
      • Moyer C.A.
      Person-centred maternity care in low-income and middle-income countries: analysis of data from Kenya, Ghana, and India.
      Standardized tools also would facilitate measurement of patient experiences as predictors of birth outcomes in observational studies and as outcomes in clinical and pragmatic trials and would enable investigation of maternal equity issues alongside of health outcome assessments.
      Although validated scales exist for measuring different aspects of person-centered prenatal care (PCPC), most are narrow in scope and do not adequately capture the nuances of disrespectful care, provider bias, and racism experienced by people of color.
      • Peters R.M.
      • Benkert R.
      • Templin T.N.
      • Cassidy-Bushrow A.E.
      Measuring African American women’s trust in provider during pregnancy.
      • Vedam S.
      • Stoll K.
      • Martin K.
      • et al.
      The Mother’s Autonomy in Decision Making (MADM) scale: patient-led development and psychometric testing of a new instrument to evaluate experience of maternity care.
      • Alfaro Blazquez R.
      • Corchon S.
      • Ferrer Ferrandiz E.
      Validity of instruments for measuring the satisfaction of a woman and her partner with care received during labour and childbirth: systematic review.
      • Nilvér H.
      • Begley C.
      • Berg M.
      Measuring women’s childbirth experiences: a systematic review for identification and analysis of validated instruments.
      • Nápoles-Springer A.M.
      • Santoyo-Olsson J.
      • O’Brien H.
      • Stewart A.L.
      Using cognitive interviews to develop surveys in diverse populations.
      • Davis D.-A.
      • Scott K.
      Translating obstetric racism into a patient-reported experience measure.
      There is a need for a comprehensive measurement tool that includes all aspects of person-centered care,
      • Shakibazadeh E.
      • Namadian M.
      • Bohren M.A.
      • et al.
      Respectful care during childbirth in health facilities globally: a qualitative evidence synthesis.
      tailored for people of color whose unique experiences are often not captured by existing measures.
      Furthermore, very few measurement tools are created using community-based participatory approaches, missing a crucial opportunity for capturing the issues most salient for affected communities.
      • Nilvér H.
      • Begley C.
      • Berg M.
      Measuring women’s childbirth experiences: a systematic review for identification and analysis of validated instruments.
      We sought to develop and validate a measurement tool that can accurately ascertain the effect of interventions on perceptions of high-quality, person-centered care from the standpoint of people at highest risk of preterm birth.
      • Nuru-Jeter A.
      • Dominguez T.P.
      • Hammond W.P.
      • et al.
      "It’s the skin you’re in”: African-American women talk about their experiences of racism. an exploratory study to develop measures of racism for birth outcome studies.
      We used community-based participatory approaches to adapt a person-centered maternity care (PCMC) scale that had been validated for use in assessing intrapartum care in low-resource countries
      • Afulani P.A.
      • Diamond-Smith N.
      • Golub G.
      • Sudhinaraset M.
      Development of a tool to measure person-centered maternity care in developing settings: validation in a rural and urban Kenyan population.
      ,
      • Afulani P.A.
      • Diamond-Smith N.
      • Phillips B.
      • Singhal S.
      • Sudhinaraset M.
      Validation of the person-centered maternity care scale in India.
      to use in assessing prenatal care in the United States—which we named the PCPC scale—with particular attention to addressing the needs and experiences of people of color.

      Materials and Methods

      We followed standard procedures for scale development, integrated with community involvement, to ensure validity and reliability.
      • DeVellis R.F.
      Scale development: theory and applications.
      We engaged community members during the entire process of scale adaptation. The PCMC scale, developed by Afulani and colleagues,
      • Afulani P.A.
      • Diamond-Smith N.
      • Golub G.
      • Sudhinaraset M.
      Development of a tool to measure person-centered maternity care in developing settings: validation in a rural and urban Kenyan population.
      ,
      • Afulani P.A.
      • Diamond-Smith N.
      • Phillips B.
      • Singhal S.
      • Sudhinaraset M.
      Validation of the person-centered maternity care scale in India.
      served as the original scale used for the adaptation. Initial review of that scale helped to develop a definition for the construct and identify related domains and items.
      • Afulani P.A.
      • Diamond-Smith N.
      • Golub G.
      • Sudhinaraset M.
      Development of a tool to measure person-centered maternity care in developing settings: validation in a rural and urban Kenyan population.
      ,
      • Afulani P.A.
      • Diamond-Smith N.
      • Phillips B.
      • Singhal S.
      • Sudhinaraset M.
      Validation of the person-centered maternity care scale in India.
      This process was followed by expert reviews to adjust existing items to the new context and to optimize content validity and cognitive interviews with qualified participants with lived experience to assess clarity, appropriateness, and relevance of the questions (Appendix A).

       Expert reviews

      In the project’s first stage, research team members, including clinicians, faculty researchers, and staff, reviewed the original PCMC items and removed items not applicable to the United States setting. We then conducted 2 formal expert review sessions, one with 10 members of the California Preterm Birth Initiative’s Community Advisory Board (CAB), University of California San Francisco (UCSF), and one with 20 people representing researchers and community members. The CAB includes people of color (mostly identifying as Black and Latina) who have had preterm births, community-based health workers, and social service providers from San Francisco, Oakland, and Fresno, all having direct service and lived experiences that provide insight into what PCPC means to the communities of focus. Their expert reviews informed revisions and additional questions that reflected the needs of people of color.

       Cognitive interviews

      Cognitive interviews were conducted by the first 3 authors and another research staff member. Eligibility criteria included being ≥28 weeks pregnant or having given birth within the past year. Notably, 15 women who identified as Black (11), Latina (2), Asian (1), or White (2) participated in the cognitive interviews. The cognitive interview guide included questions on relevance, wording, and possible ambiguities. The team met after interviewing half the participants and made relevant changes. They then completed the remaining interviews and met again to make additional changes. Participants were remunerated with a $50 gift card. After completing the cognitive interviews, we had 35 candidate items for the PCPC scale (Table 1), which were integrated into a study questionnaire that included other items addressing pregnancy and birth and sociodemographic characteristics and piloted among 8 women.
      Table 1Person-centered care questions
      No.Question in original PCMC scaleFinal question PCPC scale
      All questions have response options (0, No, never; 1, Yes, a few times; 2, Yes, most of the time; 3, Yes, all the time), except the following: wait time and time with provider options (0, It was just right; 1, It was somewhat long; 2, It was very long; 3, It was extremely long), introduction options (0, No, none of them; 1, Yes, a few of them; 2, Yes, most of them; 3, Yes, all of them), and neglect, verbal, and physical abuse options (0, No, never; 1, Yes, once; 2, Yes, a few times; 3, Yes, many times)
      Label
      1.How did you feel about the amount of time you waited? Would you say it was very short, somewhat short, somewhat long, or very long?How did you feel about the amount of time you had to wait to be seen by a healthcare provider during prenatal visits?Wait time
      2.
      Blank implies not being part of the original scale and added as part of the adaptation process
      How did you feel about the amount of time the provider spent with you? (ie, was it rushed or did they take their time with you)Time with provider
      3.During your time in the health facility did the doctors, nurses, or other healthcare providers introduce themselves to you when they first came to see you?During your pregnancy did your providers introduce themselves to you when they first came to see you? (If you were seen by only 1 provider and they introduced themselves, you can select yes, all of them)Introduction
      4.Did the doctors, nurses, or other healthcare providers call you by your name?Did your providers call you by your preferred name?Called preferred name
      5.Did the doctors, nurses, or other staff at the facility treat you with respect?Did your providers treat you with respect?Treat you with respect
      6.Did you feel your experience and knowledge were valued?Knowledge valued
      7.Did you feel heard and listened to by your providers?Heard and listened to
      8.Did providers knock on your room's door and wait for response before entering?Privacy-knock
      9.During examinations in the labor room, were you covered up with a cloth or blanket or screened with a curtain so that you did not feel exposed?During exams (like abdominal and pelvic exams) were you covered up with a cloth or blanket or screened with a curtain so that you did not feel exposed?Privacy-not exposed
      10.Do you feel like your health information was or will be kept confidential at this facility?Did you feel your health information was kept confidential and private by providers and staff?Information confidentiality
      11.Did you feel like the doctors, nurses, or other staff at the facility involved you in decisions about your care?Did your providers involve you in decisions about your care?Involved in decisions
      12.Did the doctors, nurses, or other staff at the facility ask your permission or consent before doing procedures and examinations on you?Did providers or other staff ask your permission/consent before touching or doing procedures or examinations on you?Consent
      13.Did the doctors and nurses explain to you why they were doing examinations or procedures on you?Did your providers explain to you why they were doing examinations or procedures on you?Explain procedures
      14.Did the doctors and nurses explain to you why they were giving you any medicine?Did your providers explain to you why they were giving you any medicine?Explain medications
      15.Did you feel you could ask the doctors, nurses, or other staff at the facility any questions you had?Did you feel you could ask your providers any questions you had?Could ask any questions
      16.Did providers encourage you to ask questions?Encourage you to ask questions
      17.Did providers check that you understood information that was given to you?Check you understood information
      18.Do you feel your questions were answered when you did ask?Questions were answered
      19.Did the doctors, nurses, and other staff at the facility show they cared for you?Did providers give you information in a way that showed they cared about you?Information showed they cared
      20.Did you feel coerced or pressured into a decision by providers?Coerced
      21.Did you hold back on asking questions for any reason?Hold back on asking questions
      22.Did your providers ask about your birth preferences or birth plan?Birth plan
      23.Did the doctors and nurses at the facility talk to you about how you were feeling?Did your providers ask about your emotional well-being?Emotional well-being
      24.Did your providers provide you with resources to help with your emotional well-being if you needed it?Resources for emotional well-being
      25.Did providers respect your family or companions who were with you?Respect your family
      26.When you needed help, did you feel the doctors, nurses, or other staff at the facility paid attention?Did you feel your providers avoided, ignored, or otherwise neglected you?Neglected
      27.Did you feel the doctors, nurses, or other health providers shouted at you, scolded, insulted, threatened, or talked to you rudely?Did you feel your providers shouted at you, scolded, insulted, threatened, or talked to you rudely?Verbal abuse
      28.Did you feel like you were treated roughly like pushed, beaten, slapped, pinched, physically restrained, or gagged?Did you feel like your providers handled you roughly, held you down, or physically restrained you?Physical abuse
      29.Did you feel the doctors, nurses, or other staff at the facility took the best care of you?Did you feel your providers took the best care of you?Best care
      30.Did you feel that you had to advocate for yourself during any aspect of your prenatal care?Advocacy
      31.Did you feel you could completely trust the doctors, nurses, or other staff at the facility with regard to your care?Did you feel you could completely trust your providers with regard to your care?Trust
      32.During your time in the health facility, would you say you were treated differently because of any personal attribute, like your age, marital status, number of children, your education, wealth, your connections with the facility, or something like that?
      Fell out after factor analysis of the original scale. Questions in the original PCMC scale excluded from PCPC scale: • Did the doctors and nurses ask how much pain you were in?• Do you feel the doctors or nurses did everything they could to help control your pain?• Do you feel the doctors or nurses did everything they could to help control your pain?• Were you allowed to have someone you wanted to stay with you during labor?• Were you allowed to have someone you wanted to stay with you during delivery?• Thinking about the wards, washrooms, and the general environment of the health facility, will you say the facility was very clean, clean, dirty, or very dirty?• Thinking about the labor and postnatal wards, did you feel the health facility was crowded?• Was there electricity in the facility?
      Would you say you were discriminated against because of your race, ethnicity, culture, sex, gender, sexual orientation, language, immigration status, religion, income, education, age, marital status, number of children, insurance status, or other attributes?Discrimination
      33.In general, did you feel safe in the health facility?In general, did you feel physically safe in or around your clinic(s)?Safe
      34.Were you able to go to your preferred clinic for prenatal care?Preferred clinic
      35.Were you able to see your preferred provider for prenatal care?Preferred provider
      PCMC, person-centered maternity care; PCPC, person-centered prenatal care.
      Afulani et al. The person-centered prenatal care scale. Am J Obstet Gynecol 2021.
      a All questions have response options (0, No, never; 1, Yes, a few times; 2, Yes, most of the time; 3, Yes, all the time), except the following: wait time and time with provider options (0, It was just right; 1, It was somewhat long; 2, It was very long; 3, It was extremely long), introduction options (0, No, none of them; 1, Yes, a few of them; 2, Yes, most of them; 3, Yes, all of them), and neglect, verbal, and physical abuse options (0, No, never; 1, Yes, once; 2, Yes, a few times; 3, Yes, many times)
      b Blank implies not being part of the original scale and added as part of the adaptation process
      c Fell out after factor analysis of the original scale.Questions in the original PCMC scale excluded from PCPC scale: • Did the doctors and nurses ask how much pain you were in?• Do you feel the doctors or nurses did everything they could to help control your pain?• Do you feel the doctors or nurses did everything they could to help control your pain?• Were you allowed to have someone you wanted to stay with you during labor?• Were you allowed to have someone you wanted to stay with you during delivery?• Thinking about the wards, washrooms, and the general environment of the health facility, will you say the facility was very clean, clean, dirty, or very dirty?• Thinking about the labor and postnatal wards, did you feel the health facility was crowded?• Was there electricity in the facility?

       Online survey

      The final questionnaire was administered using an online survey created on the Research Electronic Database Capture platform.
      • Harris P.A.
      • Taylor R.
      • Thielke R.
      • Payne J.
      • Gonzalez N.
      • Conde J.G.
      Research Electronic Data Capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support.
      A total of 312 participants, who were recruited through community-based organizations and advertising on social media platforms, completed the survey between January and September 2020. Eligibility criteria included being ≤1 year postpartum and age of ≥15 years. The survey advertisements noted that we were particularly interested in the experiences of people of color. A personalized survey link was emailed to interested participants; on completion, they received a $40 electronic gift card. Participants were provided with study information on the link’s landing page and clicked a button to indicate informed consent before the survey opened up. The study was approved by UCSF’s Institutional Review Board.

       Psychometric analyses

      Construct validity was assessed using exploratory factor analysis, criterion validity using hypothesis testing, and internal consistency reliability using Cronbach alpha.
      • DeVellis R.F.
      Scale development: theory and applications.
      ,
      • Hinkin T.R.
      • Tracey J.B.
      • Enz C.A.
      Scale construction: developing reliable and valid measurement instruments.
      ,
      • Afifi A.
      • May S.
      • Clark V.A.
      Computer-aided multivariate analysis.
      We used the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy to assess whether the variables were suitable for factor analysis and conducted iterative factor analysis using both Kaiser’s rule of retaining only factors with eigenvalues exceeding unity and the “break” in the scree plot to decide on how many factors to retain.
      • DeVellis R.F.
      Scale development: theory and applications.
      ,
      • Hinkin T.R.
      • Tracey J.B.
      • Enz C.A.
      Scale construction: developing reliable and valid measurement instruments.
      ,
      • Afifi A.
      • May S.
      • Clark V.A.
      Computer-aided multivariate analysis.
      Because the person-centered care domains are theoretically related, we used oblique rotations, which allow for correlation between the rotated factors.
      • Hinkin T.R.
      • Tracey J.B.
      • Enz C.A.
      Scale construction: developing reliable and valid measurement instruments.
      ,
      • Collins D.
      Pretesting survey instruments: an overview of cognitive methods.
      We examined factor loadings and uniqueness to select items to exclude.
      • Afifi A.
      • May S.
      • Clark V.A.
      Computer-aided multivariate analysis.
      ,
      • Spector P.E.
      Input was sought from the CAB on the importance of items marked for exclusion, influencing decisions on which items were ultimately retained. Once the final items were selected, we identified subscales using factor loadings and conceptual groupings. We then generated summative scores for both full-scale versions and the subscales and examined whether they were associated with related measures (satisfaction, perceived quality of care, future prenatal care, and scores on the Mothers Autonomy in Decision Making [MADM] scale and the Mothers on Respect index [MORi]) in theoretically predictable ways to assess criterion validity.
      • Vedam S.
      • Stoll K.
      • Martin K.
      • et al.
      The Mother’s Autonomy in Decision Making (MADM) scale: patient-led development and psychometric testing of a new instrument to evaluate experience of maternity care.
      ,
      • Vedam S.
      • Stoll K.
      • Rubashkin N.
      • et al.
      The Mothers on Respect (MOR) index: measuring quality, safety, and human rights in childbirth.

      Results

      We excluded data from 13 respondents who reported not receiving prenatal care and 6 respondents who did not respond to all the PCPC items, resulting in an analytical sample of 293.
      Most of the sample self-identified as Black (84%) and were married or living with their partner (90%) and primiparous (77%). Approximately half were 29 to 32 years old (43%) and college graduates (53%) (Table 2).
      Table 2Characteristics of survey respondents (N=293)
      n%
      Age, y
       17–289833.4
       29–3212743.3
       33–456823.2
      Parity
       122576.8
       >16823.2
      Time since birth
       <3 mo10435.5
       3–4 mo10535.8
       5–12 mo8428.7
      Married26490.1
      Race or ethnicity
      Respondents could select multiple options
       Black or African American24684
       White or Caucasian279.2
       Asian62
       Hawaiian or Pacific Islander10.3
       Latina or Hispanic175.8
       American Indian or Alaska Native41.4
       Other51.7
       Multiracial82.7
       Prefer not to answer10.3
      Education attainment
       High school or less3712.6
       Some college10034.1
       College graduate15653.2
      Average yearly income
       <$50,0005619.1
       $50,000–$100,00018563.1
       >$100,0005217.7
      Employment status
       Full time15452.6
       Part time3411.6
       Not employed
      Includes 1 “Prefer not to answer”
      10535.8
      Insurance type
       Private or employer-provided insurance24382.9
       Other: Medicaid, Medicare, or Tricare4013.7
       No insurance
      Includes 2 “Prefer not to answer.”
      103.4
      First prenatal visit in the first trimester25286
      8 or more prenatal visits20770.6
      Setting where prenatal care was received
      Respondents could select multiple options
       Hospital-based clinic20971.3
       Community clinic8529.0
       Other72.4
      Most frequent prenatal care provider
       Obstetrician-gynecologist9833.4
       Family practice physician13947.4
       Nurse or midwife5619.1
      Gave birth before March 202016957.7
      Afulani et al. The person-centered prenatal care scale. Am J Obstet Gynecol 2021.
      a Respondents could select multiple options
      b Includes 1 “Prefer not to answer”
      c Includes 2 “Prefer not to answer.”
      Distributions for individual PCPC items are presented in Appendix B. There was good correlation among most items, ranging from 0.2 to 0.7. Of note was the question on advocating for oneself (reverse coded to indicate needing to advocate for oneself as lower PCPC), which was negatively correlated with most of the other items. The KMO values ranged from 0.60 to 0.98, with an overall KMO of 0.92, indicating that the variables were satisfactory for factor analysis.
      The initial factor analysis yielded 3 factors with eigenvalues of >1, accounting for 82% of the cumulative variance. However, 1 dominant factor with an eigenvalue of 12.0 accounted for 66% of the cumulative variance (Figure). The eigenvalues for the other factors were <2 (1.8 and 1.2). All items had factor loadings of >0.3 on at least 1 of the 3 factors except the advocacy question. Uniqueness for all items was <0.9 except for advocacy and explaining medications. The uniqueness for questions on coercion, holding back from asking questions, preferred name, and birth plan were between 0.8 and 0.9. When the analysis was restricted to the single-factor structure, all items had loadings of >0.3 except the items on advocacy, explaining medications, coercion, and holding back from asking questions, which had loadings of <0.2 and uniqueness of >0.9. The questions on being called by preferred name, birth plan, privacy, and wait time had loadings between 0.2 and 0.3 and uniqueness between 0.8 and 0.9.
      Figure thumbnail gr1
      FigureScree plot of eigenvalues
      Scree plot after factor analysis with 35 person-centered prenatal care items.
      Afulani et al. The person-centered prenatal care scale. Am J Obstet Gynecol 2021.
      After informing the CAB that items that had low loadings and high uniqueness could be dropped, they strongly recommended keeping all items, viewing them as crucial to the prenatal experience and capturing different aspects of PCPC. They also felt different items may be more important depending on the goals of a study. Thus, to be responsive to the CAB recommendations and ensure construct validity, we decide to drop only the advocacy question, which had loading of <0.3 and uniqueness of >0.9 on the 3 factors.
      The resultant factor analysis with 34 items was similar to the first iteration (Table 3) and yielded a Cronbach alpha of 0.94. Therefore, we recommend a 34-item PCPC scale that can be used to generate summative PCPC scores. The advocacy question was deemed very important by the CAB members, so although it is not included in the scale because of its very poor psychometrics, we recommend its inclusion as a separate question and that its relationship with the PCPC score be examined.
      Table 3Results of exploratory factor analysis for 34-item person-centered prenatal care scale
      3-factor structureSingle-factor structureLoading on individual subscales
      Factor1Factor2Factor3UniquenessFactor1UniquenessCommunication and autonomyDignity and respectResponsive and supportiveUniqueness
      • 1.
        Introduction
      0.410.350.690.490.760.460.79
      • 2.
        Called preferred name
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.480.820.290.920.210.95
      • 3.
        Explain procedures
      0.620.460.700.510.660.56
      • 4.
        Explain medications
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.350.900.240.940.190.96
      • 5.
        Consent
      0.570.600.540.710.540.71
      • 6.
        Could ask any questions
      0.300.600.670.550.630.61
      • 7.
        Hold back on asking questions
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.580.740.200.960.180.97
      • 8.
        Encourage you to ask questions
      0.370.410.780.390.770.41
      • 9.
        Check that you understood information
      0.570.400.710.500.750.44
      • 10.
        Questions were answered
      0.420.590.560.690.620.61
      • 11.
        Heard and listened to
      0.380.450.750.440.730.46
      • 12.
        Involved in decisions
      0.440.500.640.590.670.56
      • 13.
        Birth plan
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.430.830.250.940.310.90
      • 14.
        Coerced
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.380.900.071.000.090.99
      • 15.
        Treat you with respect
      0.370.460.740.460.730.47
      • 16.
        Respect your family
      0.420.590.610.620.630.60
      • 17.
        Information confidentiality
      0.730.510.470.780.510.74
      • 18.
        Privacy-knock
      0.450.590.600.640.580.66
      • 19.
        Privacy-not exposed
      0.510.700.380.850.370.86
      • 20.
        Verbal abuse
      0.650.430.680.540.660.56
      • 21.
        Physical abuse
      0.570.520.710.500.620.62
      • 22.
        Discrimination
      0.660.270.840.290.790.37
      • 23.
        Neglected
      0.440.510.670.550.670.55
      • 24.
        Knowledge valued
      0.360.340.780.390.810.34
      • 25.
        Wait time
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.530.760.280.920.280.92
      • 26.
        Time with provider
      0.520.680.460.790.470.78
      • 27.
        Information showed they cared
      0.530.340.800.370.800.36
      • 28.
        Best care
      0.300.360.470.700.510.720.48
      • 29.
        Emotional well-being
      0.730.490.630.600.600.64
      • 30.
        Resources for emotional well-being
      0.880.360.670.550.620.62
      • 31.
        Trust
      0.450.460.730.460.720.48
      • 32.
        Safe
      0.600.560.470.770.560.68
      • 33.
        Preferred clinic
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.410.600.580.660.570.68
      • 34.
        Preferred provider
        Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      0.720.400.740.450.650.58
      Afulani et al. The person-centered prenatal care scale. Am J Obstet Gynecol 2021.
      a Removed from the 26-item version. Subscales were based on the factor analysis and conceptual groupings.
      Given potential interest in a shorter scale, we ran various iterations of factor analysis, dropping items that might be more conceptually different from the others and using a stricter criterion for retention to obtain a parsimonious scale with good psychometric properties. For example, we dropped the items on preferred clinic and provider based on their conceptual difference—they indicated access, which although very important to person-centered care may be beyond the control of a particular clinic. We also dropped items that had loadings of <0.3 on the single dominant factor and uniqueness of >0.8, yielding a 26-item version of the PCPC scale (Appendix C), with a Cronbach alpha of 0.95, and strong correlation with the 34-item version (r=0.98).
      Moreover, 3 subscales were then created (“dignity and respect,” “communication and autonomy,” and “responsive and supportive care”) through an iterative process based on factor loading and conceptual groupings from the World Health Organization experience of care domains.
      • Tunçalp Ӧ.
      • Were W.M.
      • MacLennan C.
      • et al.
      Quality of care for pregnant women and newborns-the WHO vision.
      Factor analysis of items in each subscale yielded a single factor with all items loading at >0.4 on the single factor and with uniqueness of <0.8, except for the items excluded in the 26-item version (Table 3). The subscales all have Cronbach alphas of >0.8 (Table 4).
      Table 4Person-centered prenatal care scale and subscale properties and distribution of scores (N=293)
      Scale and subscaleNumber of itemsCronbach alphaRaw scoreStandardized score
      MeanStandard deviationMinMaxMeanStandard deviationMinMax
      34-item PCPC scale340.9487.412.825.0102.085.712.624.5100.0
      26-item PCPC scale260.9568.510.816.078.087.813.920.5100.0
      Communication and autonomy subscale140.8134.25.512.042.081.413.228.6100.0
      Dignity and respect subscale100.8727.63.78.030.091.912.226.7100.0
      Responsive and supportive care subscale100.8525.74.65.030.085.615.216.7100.0
      PCPC, person-centered prenatal care.
      Afulani et al. The person-centered prenatal care scale. Am J Obstet Gynecol 2021.
      Item responses were then added to create summative scores for both PCPC full-scale versions and the subscales. Scores are standardized by dividing the mean score by the maximum possible scores (eg, for the 34-item version, the maximum score is 102 [34×3], and for the 26-item version, it is 78 [26×3]) and then multiplying by 100. This creates a standardized score that ranges from 0 to 100 for both versions, where 0 is the worst PCPC and 100 is the best. The results show an average standardized score of >80 based on all the measures (Table 4).
      Correlations between both PCPC scale versions and other measures provided support for criterion validity, with correlation coefficients of >0.7 between the PCPC scores and the MORi score and >0.6 for the MADM score. Regressing each of the subscales and both versions of the full scale on patients’ ratings of satisfaction with services, general quality ratings, and whether they would receive prenatal care in the same facility if they were to be pregnant again showed that increasing PCPC is associated with higher ratings of satisfaction, quality of care, and intent to receive prenatal care in the same facility, suggesting high criterion validity (Table 5).
      Table 5Bivariate logistic and linear regression of scale scores on related measures to assess criterion validity (N=293)
      Scale and subscaleSatisfied with careWill receive prenatal care in the same place againRated quality of care as very goodMADM scoreMORi score
      OR95% CIOR95% CIOR95% CIβ95% CIβ95% CI
      34-item PCPC scale1.17
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.11–1.23)1.09
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.06–1.11)1.11
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.08–1.14)0.32
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.27–0.36)0.21
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.18–0.24)
      26-item PCPC scale1.15
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.10–1.21)1.08
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.05–1.11)1.10
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.07–1.12)0.29
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.25–0.33)0.19
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.16–0.22)
      Communication and autonomy subscale1.14
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.09–1.20)1.07
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.05–1.10)1.09
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.06–1.12)0.27
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.24–0.31)0.17
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.14–0.19)
      Dignity and respect subscale1.13
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.09–1.18)1.07
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.04–1.09)1.08
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.06–1.11)0.26
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.22–0.31)0.17
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.14–0.20)
      Responsive and supportive care subscale1.14
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.09–1.20)1.09
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.06–1.12)1.12
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (1.08–1.16)0.29
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.24–0.33)0.21
      P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      (0.18–0.24)
      CI, confidence interval; MADM, Mothers Autonomy in Decision Making scale; MORi, Mothers on Respect index; OR, odds ratio; PCPC, person-centered prenatal care.
      Afulani et al. The person-centered prenatal care scale. Am J Obstet Gynecol 2021.
      a P<.001. Each row is a different bivariate model. N=290 for MADM Model.
      Sensitivity analysis using various subsamples (only Black women and by level of education and timing of birth) were generally similar to those obtained with the full sample (Appendix D).

      Discussion

       Principal findings

      Using rigorous, community-engaged scale development approaches, we successfully adapted and validated the PCPC scale for use in evaluating prenatal care in the United States among pregnant people of color, with a focus on Black women. We created a 34-item and a 26-item version of the PCPC scale, each of which has 3 subscales (dignity and respect, communication and autonomy, responsive and supportive care). Both versions and all subscales have good content, construct, and criterion validity with Cronbach alphas of >0.9 for the main scales and > 0.8 for the subscales.

       Results

      In general, the PCPC scale performed similarly to the original PCMC scale.
      • Afulani P.A.
      • Diamond-Smith N.
      • Golub G.
      • Sudhinaraset M.
      Development of a tool to measure person-centered maternity care in developing settings: validation in a rural and urban Kenyan population.
      ,
      • Afulani P.A.
      • Diamond-Smith N.
      • Phillips B.
      • Singhal S.
      • Sudhinaraset M.
      Validation of the person-centered maternity care scale in India.
      The PCPC scale differs from other scales such as the MORi and MADM scales,
      • Vedam S.
      • Stoll K.
      • Martin K.
      • et al.
      The Mother’s Autonomy in Decision Making (MADM) scale: patient-led development and psychometric testing of a new instrument to evaluate experience of maternity care.
      ,
      • Vedam S.
      • Stoll K.
      • Rubashkin N.
      • et al.
      The Mothers on Respect (MOR) index: measuring quality, safety, and human rights in childbirth.
      because it takes a more comprehensive view of person-centered care. However, the PCPC subscales are aligned with these scales—with the dignity and respect subscale having the strongest correlation (r=0.71) with the MORi scale, which also measures respect. In addition, the frequency response format used in the PCMC scale provides more variability in responses than binary yes or no responses and is less prone to acquiescence bias than the agree or disagree format used in other scales.
      • Holbrook A.
      Acquiescence response bias.
      Unique to the PCPC scale is the level of community input in the adaptation process, leading to inclusion of items most relevant to the experiences of Black women (eg, feelings of not being listened to and knowledge and experiences not being valued).
      • Altman M.R.
      • Oseguera T.
      • McLemore M.R.
      • Kantrowitz-Gordon I.
      • Franck L.S.
      • Lyndon A.
      Information and power: women of color’s experiences interacting with health care providers in pregnancy and birth.
      ,
      • Altman M.R.
      • McLemore M.R.
      • Oseguera T.
      • Lyndon A.
      • Franck L.S.
      Listening to women: recommendations from women of color to improve experiences in pregnancy and birth care.
      Sustained partnership with the CAB provided an opportunity for community input in a process not usually considered amenable to community involvement—leading to a scale that may be more relevant and meaningful to communities of color.

       Clinical implications

      As recommended by our CAB, all items included in the PCPC scale have value in assessing high-quality PCPC, although the relevance of some may vary depending on program goals. Thus, we have provided multiple versions of the scale for different purposes. We recommend the 34-item version for comprehensive PCPC measurement and the 26-item version when PCPC is one of many outcomes and there are limitations on the number of items that can be included. The mix of questions to capture people’s subjective perceptions and more objective and actionable items in the scale can inform quality improvement.
      • Afulani P.A.
      • Aborigo R.A.
      • Walker D.
      • Moyer C.A.
      • Cohen S.
      • Williams J.
      Can an integrated obstetric emergency simulation training improve respectful maternity care? Results from a pilot study in Ghana.

       Research implications

      We have developed a robust PCPC scale that can be used in studies of interventions aimed at improving prenatal care experiences of people of color, which could reduce disparities in prenatal care and improve maternal and birth outcomes. The advocacy variable, which was added through community recommendation, performed poorly in psychometric analysis. This is likely caused by advocacy having the potential to be interpreted both positively (“I was able to advocate for myself”) and negatively (“I had to advocate for myself”). However, the CAB felt strongly that the question should remain, so we recommend that it be considered as an individual question outside of the validated scale. Future studies can examine both negative and positive framings of the advocacy questions and assess how these are associated with various dimensions of PCPC.
      Other questions addressing issues, such as coercion, holding back from asking questions, and having a birth plan (which were added through community recommendation) and receiving explanations for medications and being called by one’s preferred name (which were part of the original scale), performed worse than expected in the psychometric analysis. However, given the strong CAB recommendation to retain these items and the fact that excluding them does not substantially improve the reliability of the scale, they are included in the 34-item version of the scale. Future validations should assess the performance of these items.

       Strengths and limitations

      Using participatory approaches embedded in standard instrument development methods is a major strength of this study, particularly in assuring that the final scales were relevant and important to the target communities. In addition, starting with a validated tool created a rigorous, theory-based foundation from which to work. However, the final sample for psychometric analysis had higher educational attainment and income than the general population, potentially limiting the scale’s generalizability. One reason for the low representation of non-Black persons of color was that the survey was only administered in English. We plan to validate the scale in a Spanish-speaking population in a subsequent study. Other drawbacks include the length of the scale and associated participant burden, which we attempted to mitigate by creating an abbreviated scale and subscales.

       Conclusions

      We have presented 2 versions of a validated scale that can be used to measure the major domains of PCPC in the United States context among people of color. Both versions have high validity and reliability in a sample comprised predominantly of Black women. Our process highlights the importance and feasibility of community involvement in development of measures. Because experience of care is increasingly being examined as a major factor contributing to disparities in adverse outcomes such as preterm birth, we anticipate a growing need for measurement tools to assess the quality of care provided within the prenatal care setting. The PCPC scale can be used as a predictor of outcomes in observational studies aimed at understanding the reasons for disparities in birth outcomes. In addition, given the need to develop interventions that improve the quality and experience of care among people of color, the PCPC scale can be used as an outcome measure in the context of clinical trials and comparative effectiveness and implementation studies to determine the effect of interventions on pregnant people’s experience of care. The continued disparities in pregnancy and birth outcomes affecting people of color, particularly Black women, in the United States underscore the need for new ways of assessing quality of care; incorporating these measures into intervention studies is a crucial step toward reducing the risk placed on these communities. Future validations in other groups, such as indigenous people, religious minority groups, immigrants, and lesbian, gay, bisexual, transgender, and queer (or questioning) people, in both urban and rural settings will help ensure that the scale captures the unique experiences of these people.

      Acknowledgments

      We thank all members of the California Preterm Birth Initiative CAB, everyone who provided feedback at various phases of the scale development, and all the people who participated in the various stages of the project. We are grateful to Nadia Diamond-Smith and Nicholas Rubashkin at UCSF for their support on the project.

      Supplementary Data

      Appendix A

      Additional details on Methods.

       Expert reviews

      In the first stage of the project, members of the research team, made up of clinicians, faculty researchers, and research staff, reviewed the original validated person-centered maternity care (PCMC) items and removed items that were obviously not applicable to the United States setting. Questions removed from the original scale included those addressing the availability of water and electricity. We then conducted 2 formal expert review sessions.
      The first session was with 10 members of the University of California San Francisco (UCSF), California Preterm Birth Initiative’s Community Advisory Board (CAB), which consists of mothers who have had preterm births, community-based health workers, and social service providers from San Francisco, Oakland, and Fresno. All the 10 CAB members were women and self-identified as Black (7), Latina (4), or White (1). These women all have direct service and lived experiences that can provide insight into what PCMC means to the communities of focus and thus are considered community experts.
      After the initial expert review session, we incorporated the suggestions made by CAB members, primarily consisting of adding new questions they recommended. Questions that were added during the first round of expert reviews included items relating to communication and privacy or confidentiality such as “Did you feel heard and listened to?”, “Did providers encourage you to ask questions?”, and “Did providers knock on your room’s door and wait for response before entering?” In addition, we implemented suggestions made by the CAB members for rewording questions, including suggesting that the word “facility” be dropped from nearly all questions.
      The second session consisted of 20 people, including 2 community health workers, 4 CAB members, and 14 faculty members (researchers and clinicians). We included a variety of faculty members because we wanted to also capture the views of people with measurement expertise and with experience in research and in clinical work. These expert reviewers included 18 women (7 who self-identified as Black, 1 as Latina, 3 as Asian, and 7 as White) and 2 men (both White). The expert review sessions were moderated by the first 2 authors, who have experience moderating expert review sessions and qualitative interviews. Again, recommendations were made regarding the wording of existing questions and adding new questions. The new questions focused on infant feeding choices and feeding goals and emotional well-being and coping strategies. It was also at this meeting that it was suggested that the questions be split by prenatal and intrapartum phases of care. (This paper focuses on the scale we created to address prenatal care, which we call the person-centered prenatal care [PCPC] scale.) CAB members were paid $40/hour for their participation in these expert review sessions.

       Cognitive interviews

      Cognitive interviews were conducted by the first 3 authors and another research staff member (all female; 1 Black, 1 Latina, and 2 White). The first author, who has extensive measurement expertise, trained the team for the cognitive interviews. Participants were recruited from local community organizations and from a UCSF database comprised women who previously participated in pregnancy-related research and expressed interest in being contacted about future studies. People were eligible to participate if they were currently in their third trimester of pregnancy or up to 1 year after delivery. Notably, 15 women who identified as Black (11), Latina (2), Asian (1), or White (2) participated in the cognitive interviews (numbers do not add up to 15 because people could select more than 1 option). Although we had a diverse group conducting the cognitive interviews, race concordance was not always achieved between the interviewer and the respondent. Interviews occurred in a private room within the university and in other spaces convenient to the participants such as a local café and a health center in Oakland.
      The cognitive interview guide included questions on relevance, wording, and any possible ambiguities. The interviewers took notes and audio recorded the interviews. Researchers met after the first set of interviews and made changes to the scale items and field guide based on the initial interviews before proceeding with the final set. Most of the changes to the scale items suggested during cognitive interviews involved rephrasing of questions or substituting words. For example, many participants confused the meaning of the word “empathy” (which had been recommended during the expert reviews) for sympathy. Consequently, we replaced the word “empathy” with the phrase “showed they cared about you” (wording used in the original scale) to avoid ambiguity. In addition, participants were asked if any items on the scale made them feel uncomfortable and, if so, whether the question should be excluded and any issues that were important to them that they felt had not been captured. The team met again after the final set of cognitive interviews to make relevant changes to the questions. All women who participated in a cognitive interview were given a $50 gift card for their time. At the end of the cognitive interviews, we had 35 items for a PCPC scale.

       Pretesting

      After the cognitive interviews, the PCPC items were finalized and integrated into the study questionnaire, which included additional items regarding pregnancy and birth history and sociodemographic characteristics. The full questionnaire was piloted with 8 women (4 women who were recruited from a community-based organization for postpartum women and 4 CAB members) who were asked to review the tool one final time. Notably, 4 of these women were emailed the survey and asked to complete it at home, whereas the other 4 took the survey at a community-based organization on iPads provided by the study team. In addition to reviewing the questions, all 8 women reviewed the survey for correct logic and branching. Women were eligible to participate in the pretesting phase if they were up to 1 year postpartum. All participants in the pretesting phase received a $20 gift card for their time.

       Online survey

      The final questionnaire was administered using an online survey created on the Research Electronic Database Capture platform,
      • Harris P.A.
      • Taylor R.
      • Thielke R.
      • Payne J.
      • Gonzalez N.
      • Conde J.G.
      Research Electronic Data Capture (REDCap)--A metadata-driven methodology and workflow process for providing translational research informatics support.
      to provide data for the psychometric analysis. A sample size of 300 was determined as appropriate for the psychometric based on the rule of thumb of having 5 to 10 people per each scale item up to approximately 300.
      • DeVellis R.F.
      Scale development: theory and applications.
      Thus, we estimated a sample size of 300 based on the 30 items in initial scale that was adapted. This sample size is also supported by evidence that 200 to 300 is appropriate for factor analysis regardless of the number of items.
      • DeVellis R.F.
      Scale development: theory and applications.
      ,
      • Boateng G.O.
      • Neilands T.B.
      • Frongillo E.A.
      • Melgar-Quiñonez H.R.
      • Young S.L.
      Best practices for developing and validating scales for health, social, and behavioral research: a primer.
      Between January and September 2020, 312 participants, who were recruited through community-based organizations and advertising on social media platforms, completed the survey. Eligibility criteria included being ≤1 year postpartum and age of ≥15 years. Although not restricted to people of color, the survey advertisements noted that we were particularly interested in the experiences of these people. A personalized survey link was emailed to interested participants; on completion, they received a $40 electronic gift card. Participants were provided with study information on the link’s landing page and clicked a button to indicate informed consent before the survey opened up. The study was approved by UCSF’s Institutional Review Board.

       Psychometric analysis

      Before conducting the factor analysis, we examined the distributions of all the items and recoded some response options to obtain a uniform scale. To ensure that all response options ranged from 0 to 3, we recoded items that had a “not applicable” category (4) to the upper middle category (2, most of the time). We also reverse coded negative items so that all responses reflected a scale of 0 as the lowest level to 3 as the highest level. We then constructed a correlation matrix to examine the correlations among the items.
      We used the Kaiser-Meyer-Olkin measure of sampling adequacy to assess whether the variables were suitable for factor analysis and conducted iterative factor analysis using both Kaiser’s rule of retaining only factors with eigenvalues (the amount of information captured by a factor) exceeding unity and the “break” in the scree plot (plots of eigenvalues) to decide on how many factors to retain.
      • DeVellis R.F.
      Scale development: theory and applications.
      ,
      • Afifi A.
      • May S.
      • Clark V.A.
      Computer-aided multivariate analysis.
      ,
      • Hinkin T.R.
      • Tracey J.B.
      • Enz C.A.
      Scale construction: developing reliable and valid measurement instruments.
      Because the person-centered care domains are theoretically related, we used oblique rotations, which allows for correlation between the rotated factors.
      • Hinkin T.R.
      • Tracey J.B.
      • Enz C.A.
      Scale construction: developing reliable and valid measurement instruments.
      We examined the factor loadings (the degrees to which the original item scores correlate with the components) to determine which items to retain and which to delete. We decided to exclude items that did not have a loading of 0.3 or higher on any of the extracted factors, unless there was a strong theoretical rationale for including them. We also examined the uniqueness of each item (the variance that is “unique” to the variable) to identify items to exclude.
      Input was sought from the CAB on whether items marked for exclusion were important or relevant to include based on a theoretical rationale, influencing final decisions on which items were ultimately retained. Once the final items were selected, we identified subscales using the loadings on each factor and conceptual groupings. We then generated summative scores for the scale and subscales and examined whether they were related to other measures or outcomes in theoretically predictable ways to assess criterion validity.
      • DeVellis R.F.
      Scale development: theory and applications.
      ,
      • Crosby R.A.
      • DiClemente S.L.F.
      Research methods in Health Promotion.
      To assess criterion validity, we hypothesized that the PCPC scale scores would be correlated with global measures such as women’s ratings of their satisfaction with the services, their perceptions of the quality of care they received, and whether she would use the same service in the future.
      • Afulani P.A.
      • Diamond-Smith N.
      • Golub G.
      • Sudhinaraset M.
      Development of a tool to measure person-centered maternity care in developing settings: validation in a rural and urban Kenyan population.
      ,
      • Afulani P.A.
      • Diamond-Smith N.
      • Phillips B.
      • Singhal S.
      • Sudhinaraset M.
      Validation of the person-centered maternity care scale in India.
      In addition, we hypothesized that subscale scores will be associated with related measures of women’s experiences of care such as the Mothers Autonomy in Decision Making scale and the Mothers on Respect index.
      • Vedam S.
      • Stoll K.
      • Martin K.
      • et al.
      The Mother’s Autonomy in Decision Making (MADM) scale: patient-led development and psychometric testing of a new instrument to evaluate experience of maternity care.
      ,
      • Vedam S.
      • Stoll K.
      • Rubashkin N.
      • et al.
      The Mothers on Respect (MOR) index: measuring quality, safety, and human rights in childbirth.
      We tested these hypotheses through correlations and regression analysis. We used Stata version 15 (StataCorp LLC, College Station, TX) for all analyses.

      References

        • Braveman P.
        • Heck K.
        • Egerter S.
        • et al.
        Worry about racial discrimination: a missing piece of the puzzle of black-white disparities in preterm birth?.
        PLoS One. 2017; 12e0186151
        • Vines A.I.
        • Baird D.D.
        • McNeilly M.
        • Hertz-Picciotto I.
        • Light K.C.
        • Stevens J.
        Social correlates of the chronic stress of perceived racism among black women.
        Ethn Dis. 2006; 16: 101-107
      1. March of Dimes.
        Quick facts: preterm birth. 2020; (2020. Available at:)
        • Neggers Y.H.
        Trends in maternal mortality in the United States.
        Reprod Toxicol. 2016; 64: 72-76
        • Alhusen J.L.
        • Bower K.M.
        • Epstein E.
        • Sharps P.
        Racial discrimination and adverse birth outcomes: an integrative review.
        J Midwifery Womens Health. 2016; 61: 707-720
        • Alio A.P.
        • Richman A.R.
        • Clayton H.B.
        • Jeffers D.F.
        • Wathington D.J.
        • Salihu H.M.
        An ecological approach to understanding black-white disparities in perinatal mortality.
        Matern Child Health J. 2010; 14: 557-566
        • Dominguez T.P.
        Adverse birth outcomes in African American women: the social context of persistent reproductive disadvantage.
        Soc Work Public Health. 2011; 26: 3-16
        • Mehra R.
        • Boyd L.M.
        • Ickovics J.R.
        Racial residential segregation and adverse birth outcomes: a systematic review and meta-analysis.
        Soc Sci Med. 2017; 191: 237-250
        • Nuru-Jeter A.
        • Dominguez T.P.
        • Hammond W.P.
        • et al.
        "It’s the skin you’re in”: African-American women talk about their experiences of racism. an exploratory study to develop measures of racism for birth outcome studies.
        Matern Child Health J. 2009; 13: 29-39
        • Slaughter-Acey J.C.
        • Caldwell C.H.
        • Misra D.P.
        The influence of personal and group racism on entry into prenatal care among African American women.
        Womens Health Issues. 2013; 23: e381-e387
        • Attanasio L.
        • Kozhimannil K.B.
        Patient-reported communication quality and perceived discrimination in maternity care.
        Med Care. 2015; 53: 863-871
        • Ruiz R.L.
        • Shah M.K.
        • Lewis M.L.
        • Theall K.P.
        Perceived access to health services and provider information and adverse birth outcomes: findings from LaPRAMS, 2007-2008.
        South Med J. 2014; 107: 137-143
        • Salm Ward T.C.
        • Mazul M.
        • Ngui E.M.
        • Bridgewater F.D.
        • Harley A.E.
        “You learn to go last”: perceptions of prenatal care experiences among African-American women with limited incomes.
        Matern Child Health J. 2013; 17: 1753-1759
        • Vedam S.
        • Stoll K.
        • Taiwo T.K.
        • et al.
        The Giving Voice to Mothers study: inequity and mistreatment during pregnancy and childbirth in the United States.
        Reprod Health. 2019; 16: 77
        • Ebert L.
        • Bellchambers H.
        • Ferguson A.
        • Browne J.
        Socially disadvantaged women’s views of barriers to feeling safe to engage in decision-making in maternity care.
        Women Birth. 2014; 27: 132-137
        • Altman M.R.
        • Oseguera T.
        • McLemore M.R.
        • Kantrowitz-Gordon I.
        • Franck L.S.
        • Lyndon A.
        Information and power: women of color’s experiences interacting with health care providers in pregnancy and birth.
        Soc Sci Med. 2019; 238: 112491
        • Institute of Medicine
        Crossing the Quality Chasm a New Health System for the 21st Century.
        National Academy Press, 2001
        • Lori J.R.
        • Yi C.H.
        • Martyn K.K.
        Provider characteristics desired by African American women in prenatal care.
        J Transcult Nurs. 2011; 22: 71-76
        • Cuevas A.G.
        • O’Brien K.
        • Saha S.
        What is the key to culturally competent care: reducing bias or cultural tailoring?.
        Psychol Health. 2017; 32: 493-507
        • Altman M.R.
        • McLemore M.R.
        • Oseguera T.
        • Lyndon A.
        • Franck L.S.
        Listening to women: recommendations from women of color to improve experiences in pregnancy and birth care.
        J Midwifery Womens Health. 2020; 65: 466-473
        • Afulani P.A.
        • Phillips B.
        • Aborigo R.A.
        • Moyer C.A.
        Person-centred maternity care in low-income and middle-income countries: analysis of data from Kenya, Ghana, and India.
        Lancet Glob Health. 2019; 7: e96-e109
        • Peters R.M.
        • Benkert R.
        • Templin T.N.
        • Cassidy-Bushrow A.E.
        Measuring African American women’s trust in provider during pregnancy.
        Res Nurs Health. 2014; 37: 144-154
        • Vedam S.
        • Stoll K.
        • Martin K.
        • et al.
        The Mother’s Autonomy in Decision Making (MADM) scale: patient-led development and psychometric testing of a new instrument to evaluate experience of maternity care.
        PLoS One. 2017; 12e0171804
        • Alfaro Blazquez R.
        • Corchon S.
        • Ferrer Ferrandiz E.
        Validity of instruments for measuring the satisfaction of a woman and her partner with care received during labour and childbirth: systematic review.
        Midwifery. 2017; 55: 103-112
        • Nilvér H.
        • Begley C.
        • Berg M.
        Measuring women’s childbirth experiences: a systematic review for identification and analysis of validated instruments.
        BMC Pregnancy Childbirth. 2017; 17: 203
        • Nápoles-Springer A.M.
        • Santoyo-Olsson J.
        • O’Brien H.
        • Stewart A.L.
        Using cognitive interviews to develop surveys in diverse populations.
        Med Care. 2006; 44: S21-S30
        • Davis D.-A.
        • Scott K.
        Translating obstetric racism into a patient-reported experience measure.
        University of California, San Francisco2020 (Available at:)
        • Shakibazadeh E.
        • Namadian M.
        • Bohren M.A.
        • et al.
        Respectful care during childbirth in health facilities globally: a qualitative evidence synthesis.
        BJOG. 2018; 125: 932-942
        • Afulani P.A.
        • Diamond-Smith N.
        • Golub G.
        • Sudhinaraset M.
        Development of a tool to measure person-centered maternity care in developing settings: validation in a rural and urban Kenyan population.
        Reprod Health. 2017; 14: 118
        • Afulani P.A.
        • Diamond-Smith N.
        • Phillips B.
        • Singhal S.
        • Sudhinaraset M.
        Validation of the person-centered maternity care scale in India.
        Reprod Health. 2018; 15: 147
        • DeVellis R.F.
        Scale development: theory and applications.
        4th ed. Sage Publications, Inc, Los Angeles2016
        • Harris P.A.
        • Taylor R.
        • Thielke R.
        • Payne J.
        • Gonzalez N.
        • Conde J.G.
        Research Electronic Data Capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support.
        J Biomed Inform. 2009; 42: 377-381
        • Hinkin T.R.
        • Tracey J.B.
        • Enz C.A.
        Scale construction: developing reliable and valid measurement instruments.
        J Hosp Tour Res. 1997; 21: 100-120
        • Afifi A.
        • May S.
        • Clark V.A.
        Computer-aided multivariate analysis.
        4th ed. CRC Press, Boca Raton2003
        • Collins D.
        Pretesting survey instruments: an overview of cognitive methods.
        Qual Life Res. 2003; 12: 229-238
        • Spector P.E.
        Summated rating scale construction: an introduction. vol. 82. Sage, 1991
        • Vedam S.
        • Stoll K.
        • Rubashkin N.
        • et al.
        The Mothers on Respect (MOR) index: measuring quality, safety, and human rights in childbirth.
        SSM Popul Health. 2017; 3: 201-210
        • Tunçalp Ӧ.
        • Were W.M.
        • MacLennan C.
        • et al.
        Quality of care for pregnant women and newborns-the WHO vision.
        BJOG. 2015; 122: 1045-1049
        • Holbrook A.
        Acquiescence response bias.
        in: Encyclopedia of survey research methods. Sage Publications, Inc., Thousand Oaks, CA2008
        • Afulani P.A.
        • Aborigo R.A.
        • Walker D.
        • Moyer C.A.
        • Cohen S.
        • Williams J.
        Can an integrated obstetric emergency simulation training improve respectful maternity care? Results from a pilot study in Ghana.
        Birth. 2019; 46: 523-532