Lack of provision of social and emotional information about Down syndrome associated with negative prenatal diagnosis experiences

Abstract

To determine how physician adherence to recommended practices for discussing Down syndrome (DS) impacts patient experiences, and which of these recommendations most correlate with positive prenatal patient experiences. Online surveys were distributed to mothers of children with DS born between 2016–2021. The descriptions of prenatal experiences were assigned sentiment analysis scores: positive, negative, or neutral. The scores were then compared against the adherence of clinicians to recommended practices for delivering prenatal screening results. Of the 167 patients in this study, over 50% described a negative experience. The odds of having a neutral/positive diagnosis experience were about 18 times greater for those patients whose physicians adhered to all seven social recommendations. The odds of having a positive/neutral screening experience were 11.4 times greater for those patients whose physicians adhered to both emotional recommendations (not saying “I’m sorry” or conveying the diagnosis as bad news”). Most of the variance in the patient diagnosis experience was attributed to adherence to emotional recommendations and social recommendations. To meet the needs of patients, obstetricians need to address emotional, social, and informational needs by providing 1) access to balanced and accurate information about disabilities and 2) being better trained on sensitive communications.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12687-025-00801-7.

Introduction

For over 20 years, researchers in both the United States and Europe have repeatedly highlighted the negative experiences of parents when medical professionals deliver the news about a prenatal diagnosis of Down syndrome (DS) (Artal et al. 2024; Hedov et al. 2002; May et al. 2020; Meredith et al. 2023; Nelson Goff et al. 2013; Skotko 2005; Williams et al. 2002). Research by Nelson Goff et al. about the initial DS Diagnosis experience showed that among parents who received a prenatal diagnosis of DS, negative experiences with medical professionals were three times more likely than positive experiences (2013). Research exploring the psychological impact of a DS diagnosis by May et al. shows that these negative experiences can lead to lasting emotional harm given that diagnosis experiences qualify as flashbulb memories that parents remember lucidly for decades (2020). Moreover, negative diagnosis experiences can harm relationships between providers and the parents of medically vulnerable children with DS, particularly Hispanic parents of children with DS (Chung et al. 2023).

Multiple studies also point to the same causes for negative diagnosis experiences: the perception of pressure about pregnancy decisions, specifically bias toward terminating the pregnancy; the perpetuation of negative stereotypes about DS; perceived lack of sensitivity and compassion; and the lack of provision of information about DS (Artal et al. 2024; May et al. 2020; Meredith et al. 2023; Nelson Goff et al. 2013; Skotko 2005). Parents have repeatedly indicated that they want information that covers medical issues, life outcomes, local and national DS organization contact information, and supports and services that are particularly important as social determinants of health for people with disabilities at the first point on the life course (Hippman et al. 2012; May et al. 2020; Meredith et al. 2023; Sheets et al. 2011a, b).

Most obstetric care providers do not receive training about how to discuss DS (or other disabilities) when delivering a prenatal screening results, and bias can impact their provision of information (Driscoll et al. 2009; Meredith et al. 2023), some guidance from researchers and national obstetric and genetic organizations is available to help clinicians navigate those complex conversations (American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins—Obstetrics et al. 2020; Bull et al. 2022; Sheets et al. 2011a, b; Skotko et al. 2009). Following a literature review by the Down Syndrome Diagnosis Study Group led by Dr. Brian Skotko, Down syndrome medical experts recommend that clinicians provide accurate, balanced, and up-to-date information about the condition and contact information for parent support groups (Skotko et al. 2009) and also center individual counseling on the personal values and preferences of the patient (Crombag et al. 2020). Additionally, recommendations published by the American College of Obstetricians and Gynecologists (ACOG) (American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins—Obstetrics et al. 2020), the National Society of Genetic Counselors (NSGC) (Sheets et al. 2011a, b), the American Academy of Pediatrics (AAP) (Bull et al. 2022), the American College of Medical Genetics and Genomics (ACMG) (Gregg et al. 2016), and the International Society for Prenatal Diagnosis (ISPD) (Hui et al. 2023) add that the information about DS should cover medical care, developmental and educational information, supports and services, and psychosocial life outcomes.

However, research has yet to quantify the impact of obstetrician adherence to recommended practices on patient diagnosis experiences or the most significant types of information during those first conversations about DS. The present study investigated the extent to which adherence to the recommended practices impacts patients'self-described diagnosis experiences by assessing the data to answer these two research questions:

1. How does physician adherence to the recommended practices impact patient diagnosis experiences?

2. Which of the three adherence predictors examined in this study (e.g., medical, social, and emotionally oriented recommendations) is relatively more important than the others in accounting for patient diagnosis experiences?

Methods

The present study is an analysis of an unpublished portion of the data set originally collected by the authors which included parents of children with DS born between 2016 and 2021 (Ijaz et al. 2024; Meredith et al. 2023). Eligible participants were 18 years or older, US residents, and either received screening results suggesting a likely DS diagnosis or received a confirmed result via diagnostic testing. Online surveys were shared with members of the national Down Syndrome Diagnosis Network (DSDN) and 12 local DS organizations reflecting a cross-section of rural, suburban, and urban areas (see Table 1) (Ijaz et al. 2024; Meredith et al. 2023).

Table 1.

Local Down Syndrome Organization Study Partners^5,18

Local DS Organizations	State	Type
Montgomery County Down Syndrome Interest Group	PA	Suburban
Down Syndrome Association of Central NJ	NJ	Suburban
Green River Area Down Syndrome Association	KY	Rural
Down Syndrome Association of Houston	TX	Urban and suburban
Down Syndrome Association of Greater Cincinnati	OH	Urban and suburban
Down Syndrome Association of Middle Tennessee	TN	Urban and suburban
21 Connect	NY	Urban and suburban
Down Syndrome Connection of the Bay Area	CA	Urban and suburban
Down Syndrome Connection of Northwest Arkansas	AR	Suburban and rural
Down Syndrome Association of Brevard	FL	Suburban and rural
Down Syndrome Connections of the Northwest	ID	Rural
Down Syndrome Network of West Virginia	WV	State-wide

As reported in Table 2 (demographic data) and Table 3 (CHERRIES Checklist) (Ijaz et al. 2024; Meredith et al. 2023), the survey included demographic questions, background clinical information, qualitative questions about prenatal diagnosis experiences, and 14 yes/no quantitative questions about whether obstetricians (OBs) followed recommended practices for delivering a suspected or confirmed diagnosis. (See Table 4.) The study was approved for human subjects research by the University of Kentucky IRB (55,360), and participants completed IRB-approved online consent forms.

Table 2.

Demographical and clinical background of participants

Variable	Total Sample (n = 167)	% Occurrence
Year of Diagnosis
2016	8	4.8%
2017	19	11.4%
2018	21	12.6%
2019	32	19.1%
2020	65	38.9%
2021	20	12.0%
Unspecified	2	1.2%
Ethnicity*
Hispanic or Latino or Spanish Origin	18	10.7%
Not Hispanic or Latino or Spanish Origin	127	76.0%
Unspecified	22	13.2%
Race*
American Indian or Alaska Native	1	0.5%
Asian	1	0.5%
Black or African American	4	2.1%
Native Hawaiian or Other Pacific Islander	1	0.5%
White	127	82.0%
Unspecified	22	13.2%
Clinicians Providing Treatment**
Obstetrician/Gynecologist	163	97.6%
Maternal Fetal Medicine Specialist	148	88.6%
Genetic Counselor	104	62.3%
Geneticist	30	18.0%
Types of Test**
Cell-free DNA test or non-invasive prenatal screening	118	71.0%
Traditional maternal-serum screening (like the quad screen)	45	26.9%
Ultrasound	103	61.7%
Chorionic villus sampling (CVS)	10	6.0%
Amniocentesis	49	29.3%
Study Referrals
DSDN	110	65.9%
Local DS Org	53	31.7%
Unspecified	4	2.4%
At what stage(s) do you think it's most critical for medical professionals to give accurate, balanced information about Down syndrome or other genetic conditions? **
Prior to screening		31.1%
After screening		88.0%
After diagnostic testing		52.1%
What types of informational resources about conditions do patients prefer?**
Printed	113	67.7%
Online	106	63.5%
Other: Peer mentor, local organization, online support group	41	24.6%
Composite Sentiment Score (n = 167)
Negative	85	50.9%
Neutral	31	18.6%
Positive	51	30.5%
OB Sentiment Score (n = 98)
Negative	53	54.1%
Neutral	36	36.7%
Positive	9	9.2%
MFM Sentiment Score (n = 54)
Negative	28	51.9%
Neutral	21	38.9%
Positive	5	9.3%

* The questions about where respondents learned about the survey, state of residence, and race were not added to the survey until 1/27/21

** Participants could select multiple options

Table 3.

Checklist for reporting results of internet E-surveys (CHERRIES)^5,18

Design	Target population	Parents of children with DS born between January 1, 2016 and December 31, 2021
	Sample frame	Parents of children with DS referred by the Down Syndrome Diagnosis Network or 12 local DS Organizations (Table 1)
	Type of sample	Convenience sample
IRB	IRB Approval	Consent was received
	Informed Consent	Potential survey participants were provided information about length of time of the survey, which data were stored and how data would be used, who the investigator was, and the purpose of the study. They were then asked to confirm whether or not they wanted to participate See Appendix B
	Data protection	Collected data in university password-protected Qualtrics online database
Development and pre-testing	Development and testing	The test was reviewed prior to distribution by representatives of 12 local organizations and 3 internal university staff. Usability and technical functionality also assessed with Qualtrics internal tool
Recruitment process and description of the sample having access to the questionnaire	Open survey versus closed survey	Limited access survey: custom link only shared with invited participants via email or private or semi-private social media parent groups
	Contact Mode	Shared via private and semi-private advocacy organization social media groups and emails from local advocacy organization directly to new parents. Social media and emails shared quarterly. Search engines were blocked from indexing the survey
	Advertising the survey	Advertising in private and semi-private advocacy organization social media groups. See Appendix B
Survey Administration	Web/E-mail	E-survey not posted on a website. Shared via private and semi-private advocacy organization social media groups and emails to potential participants
	Context	DSDN shared survey invitations on private Facebook groups divided by birth year. The local organizations sent emails to parents of children 5 and under and shared on semi-private organizational social media pages
	Mandatory/Voluntary	Voluntary
	Incentives	No incentives
	Time/Date	Local organizations: 4/29/2020–12/31/2021 DSDN: 1/27/2021–12/31/2021
	Randomization of items or questionnaires	No randomization used
	Adaptive questioning	Adaptive questions appeared depending on the types of providers seen by patients
	Number of items	17 (2 of the 17 questions had 16 sub-items for a total of 47 items)
	Number of screens (pages)	3
	Completeness check	Checked for completeness after questionnaire was submitted. Completion determined by whether patient completed Question 8 or 9
	Review step	Questions 8–17 appear on page 3 which allows for scrolling to change answers
Response Rate	Unique site visitor	Each user was identified with cookies
	View rate	Because the survey did not have a link tracked on a website, view rates are not available
	Participation rate	99.8% of participants agreed to complete the survey 423/424. 25 participants were ineligible (7 received a postnatal diagnosis, 9 were outside the US, and 9 were outside the study time frame for a prenatal diagnosis)
	Completion rate	61% of the eligible participants completed the survey 243/399
Preventing multiple entries from the same individual	Cookies used and IP Check	Cookies and duplicate IP addresses were used to prevent multiple responses. Selected Qualtrics setting to prevent respondents from taking the survey multiple times
	Registration	No registration required
Analysis	Handling of Incomplete Questions	Surveys were included as long as they completed at least through question 8 and/or 9

Table 4.

Frequencies and Pearson Chi-square tests for guideline adherence indicators (N = 167)

Variable	Category	Negative		Neutral/Positive			χ2	df	p-value	Cramer's V
		N	%	N	%	Total N
Medical Guideline
My doctor set up a phone or face-to-face appointment to share the prenatal screening/test results with me so that I could process the information	Disagree	42	61.8%	26	38.2%	68	6.23	1	0.013	0.20
	Agree	38	41.8%	53	58.2%	91
My doctor presented me with all pregnancy options	Disagree	24	61.5%	15	38.5%	39	1.32	1	0.251	0.09
	Agree	57	50.9%	55	49.1%	112
My doctor provided more comprehensive prenatal care following my diagnosis/screenin g results to improve the health outcomes and viability of my baby	Disagree	20	74.1%	7	25.9%	27	6.94	1	0.008	0.21
	Agree	63	46.3%	73	53.7%	136
My doctor gave me resources about medical problems that people with Down syndrome may have	Disagree	62	56.9%	47	43.1%	109	3.77	1	0.052	0.15
	Agree	22	40.7%	32	59.3%	54
My doctor talked to me about medical problems that people with Down syndrome may have	Disagree	40	58.0%	29	42.0%	69	1.99	1	0.159	0.11
	Agree	44	46.8%	50	53.2%	94
Social Guideline
My doctor gave me information about a local Down syndrome support group	Disagree	73	63.5%	42	36.5%	115	24.64	1	< 0.001	0.39
	Agree	10	20.8%	38	79.2%	48
My doctor gave me information about a national Down syndrome advocacy organization	Disagree	69	56.6%	53	43.4%	122	5.60	1	0.018	0.19
	Agree	14	35.0%	26	65.0%	40
My doctor gave me accurate, up-to-date, and balanced resources about Down syndrome at the moment of diagnosis in language I could understand	Disagree	76	67.3%	37	32.7%	113	39.45	1	< 0.001	0.50
	Agree	6	12.8%	41	87.2%	47
My doctor gave me resources about the improved social and life outcomes for people with Down syndrome	Disagree	75	60.0%	50	40.0%	125	18.40	1	< 0.001	0.34
	Agree	7	19.4%	29	80.6%	36
My doctor talked to me about the improved social and life outcomes for people with Down syndrome	Disagree	74	69.8%	32	30.2%	106	41.42	1	< 0.001	0.51
	Agree	9	16.4%	46	83.6%	55
My doctor gave me resources about supports and services for people with Down syndrome (like Early Intervention)	Disagree	73	57.5%	54	42.5%	127	10.32	1	0.001	0.25
	Agree	9	26.5%	25	73.5%	34
My doctor talked to me about supports and services for people with Down syndrome (like Early Intervention)	Disagree	73	60.8%	47	39.2%	120	17.64	1	< 0.001	0.33
	Agree	9	22.5%	31	77.5%	40
Emotional Guideline
My doctor said"I'm sorry"when telling me my baby had Down syndrome	Agree	53	74.6%	18	25.4%	71	32.38	1	< 0.001	0.46
	Disagree	23	28.4%	58	71.6%	81
My doctor acted like a diagnosis of Down syndrome was bad news	Agree	64	73.6%	23	26.4%	87	39.91	1	< 0.001	0.50
	Disagree	16	22.9%	54	77.1%	70

^*Statistically significant values are highlighted in gray

Qualitative Data Analyses

When assessing the answers to the survey questions describing the patient diagnosis experiences, two reviewers (a bioethicist and a rhetorician/disability health specialist) performed a sentiment analysis (SA). (See Fig. 1). First, they used a scale of 1–5 to assess patient experiences with each type of medical provider mentioned in each response (Obstetrician, genetic counselor, midwife, geneticist, and maternal–fetal medicine specialist) with a defined rubric based on Pang and Lee’s standards used for software opinion mining and SA (Pang and Lee 2008). The complexity of the language nuance in the responses required the gold standard of expert human interpretation (van Atteveldt et al. 2021). The two primary reviewers determined the coding rubric after creating a proposed scale, reading the first 20 responses, then refining the scale after discussion. After the two reviewers independently assigned sentiment scores for each type of provider mentioned in each response, they met to discuss and resolve any differences in coding.

Fig. 1.

Sentiment Analysis Methodology Flowchart (Lucid)

Subsequently, the reviewers assigned an overall score for each respondent determined by the general description of the experience and a composite of the scores for each provider (maternal–fetal medicine specialists, obstetricians, genetic counselors, and midwives) using the following composite scoring guide for the overall experience:

• 1: Negative: Negative scores present for at least one provider. No positive descriptors used or positive score for any provider.

  • o
    Example: “They can mail you a bill but can’t mail you a brochure or anything! Infuriating!”
• 2: Neutral: Only neutral scores given for all providers for each respondent. No negative descriptors or negative score for any provider.

  • o
    Example: “My OB called me to tell me the ultrasound result and that he was referring me to a pediatric cardiologist.”

• 3: Mixed: Both positive and negative scores given for providers for each respondent. No example provided because these scores were omitted.

• 4: Positive: Positive scores present for at least one provider for each respondent. No negative descriptors or negative score for any provider.

  • o
    Example: “My OB gave me wonderful and comprehensive care … She provided me a lot of support.”

The 11 (0.05%) unresolved ratings were sent to a third rater, a developmental disabilities expert, for independent scoring; differences were resolved and final scores determined by consensus between all three raters. Eighteen (10.8%) respondents generally described the experience but did not mention a specific provider. Therefore, the composite score was used as the most inclusive SA score for conducting the statistical analysis. For this analysis, we only considered negative, neutral, and positive composite SA scores, since mixed scores (where patients provided both positive and negative descriptions) could not be directly correlated with the extent to which the recommended practices impacted a positive or negative experience. The study sample thus consisted of 167 out of 242 total participants who 1) indicated they had seen an MFM or OB; 2) provided answers to the 14 OB/MFM questions; and 3) answered the qualitative survey items (“Please describe your diagnosis experience”), which was classified as a positive, neutral, or negative patient diagnosis experience. Finally, we formed two contrasting groups (negative vs non-negative experiences) as we were more concerned about why some patients had negative experiences; the positive and neutral groups were combined, giving us two SA groups: negative and neutral/positive. Moreover, combining the neutral/positive SA scores isolated the definition of neutral/positive to be: no negative verbiage used to describe the experience, and isolated the definition of negative to be: only negative verbiage used to describe the experience. We used this method to control for differences in broader multidisciplinary team approaches because a negative score meant that no positive descriptors were used by a participant to describe any individual providers the entire experience. In contrast, among those who received a neutral or positive score, no negative descriptors were used by a participant to describe an individual provider or the entire experience. Therefore, no providers on the team changed the sentiment of the experience. Moreover, patients in previous studies have indicated that the information included in our survey needs to be provided right away by obstetric care providers upon first learning about a likely diagnosis to improve outcomes (Artal et al. 2025; Levis et al. 2012).

Quantitative data analyses

To assess physician adherence to recommended practices, 14 total recommendations were used, among which two were reverse coded before the analysis (See Table 4). These adherence indicators were dichotomous variables with two categorical choices (0 = Disagree and 1 = Agree). Among these items, the research team grouped together five items about physician adherence to medical recommendations, seven about adherence to social recommendations, and two about adherence to emotional recommendations. For each of these sub-scales, the items were averaged to form a composite variable with higher scores indicating better or more adherence. Kuder-Richardson Formula 20 (KR-20), a measure appropriate for the scales based on dichotomous variables, was used to assess the internal consistency reliability for the three sub-scales (Kuder and Richardson 1937). The reliability coefficients were 0.55, 0.88, and 0.72 for the medical, social and emotional sub-scales, respectively.

Statistical analyses

For Research Question 1, a logistic regression analysis was conducted using the statistical package SPSS 28 (IBM Corp 2021) to examine how the independent variables of adherence to medical, social, and emotional recommendations, evaluated by patients, are related to the dependent variable of patient diagnosis experience. The dependent variable was a dichotomous variable with 0 representing negative experience and 1 representing a neutral/positive experience. Methodologically, the composite independent variables of adherence to medical, social, and emotional recommendations are continuous. However, adherence to emotional recommendations has a limited number and range of values, and caution is warranted when interpreting its coefficient to avoid the issue of extrapolation.

For Research Question 2, a relative weight analysis for logistic regression (RWA) (J. W. Johnson 2000; Tonidandel and LeBreton 2015) was conducted to examine the relative importance of the three independent variables (i.e., adherence to medical, social, and emotional recommendations) in their explanatory power for the dependent variable (i.e., patient diagnosis experience). Confidence intervals and corresponding significance tests for the relative weight of individual independent variables were calculated and performed using bootstrapping with 5,000 replications; also, the ordinary least square R² statistic was adopted in the logistic relative weight analysis (Petscher et al. 2013). The RWA in this study was performed in R (R Core Team 2024; Tonidandel and LeBreton 2015). The alpha significance level for all the hypothesis tests in this study was set at 0.05.

Results

Based on assessments of diagnosis experiences through sentiment analysis, about 49.1% of the 167 respondents from 41 different states received a neutral or positive score, and about 50.9% received a negative score. Before conducting the logistic regression for Research Question 1, descriptive statistics demonstrated how the recommendations and the composite variables were associated with patient diagnosis experiences. Table 4 displays the frequencies for the 14 recommended practices and the chi-squared test of independence. Only three of 14 adherence indicators, which were focused on medical issues, were not significantly associated with patient diagnosis experiences. The remaining 11 adherence indicators focused on social and emotional factors were significantly associated with patient diagnosis experiences (all p-values ≤ 0.001). A pattern of the relationships emerged from the significant independence tests: patients who found the physicians adhered to all the diagnosis recommendations were more likely to report neutral/positive experience. The effect size of the association, as measured by Cramer’s V, for most of the significant independence tests was moderate or large by according to Cohen’s (1988) commonly used guidelines (i.e., 0.1 = small, 0.3 = moderate, and 0.5 = large).

Table 5 displays the means and standard deviations of the three composite adherence variables (i.e., adherence to medical, social, and emotional recommendations) by the two patient diagnosis experience outcomes (negative vs. neutral/positive) and the independent sample t-test results. The three composite variables were all significantly related to patient diagnosis experiences. Patients reporting neutral/positive diagnosis experiences on average had higher evaluation of their physician’s adherence to the medical, social, and emotional recommendations than their counterparts reporting negative experiences (all p-values < 0.001).

Table 5.

Descriptive statistics and independent t-tests for composite guideline adherence variables (N = 167)

	Patient Diagnosis Experience
Variable	Negative		Neutral/Positive		t	df	p
	N	M (SD)	N	M (SD)
Medical	84	0.54 (0.27)	80	0.68 (0.26)	−3.45	162	< 0.001
Social	84	0.11 (0.21)	80	0.43 (0.37)	−6.76*	123.84	< 0.001
Emotional	81	0.26 (0.37)	80	0.72 (0.40)	−7.59	159	< 0.001

*Welch’s t-test to address unequal variances

For Research Question 1, the logistic regression analysis is summarized in Table 6. The Wald chi-squared test statistic provides an inferential test of each logistic regression coefficient, and the odds ratio (OR) coefficient provides information related to the change in the odds of having a neutral/positive experience that would be associated with a one-unit change in each independent variable. Adherence to medical recommendations was not significantly related to patient diagnosis experience (Wald χ² = 1.32, df = 1, p = 0.25; OR = 2.51).

Table 6.

Logistic regression results for the outcome variable of patient diagnosis experience (N = 167)

Variable	Coef	S.E	Wald	df	p	OR
Medical	0.920	0.800	1.322	1	0.250	2.510
Social	2.874	0.803	12.824	1	<.001	17.712
Emotional	2.431	0.478	25.875	1	<.001	11.368
Constant	−2.391	0.557	18.403	1	<.001	0.092

However, adherence to social recommendations was significantly related to patient diagnosis experience (Wald χ² = 12.82, df = 1, p < 0.001; OR = 17.71); for each increase in 1 unit of adherence to social recommendations, the estimated odds of having a neutral/positive diagnosis experience would increase by a factor of 17.7 if the other two adherence variables are held constant in the model. In the context of this study, the odds of having a neutral/positive diagnosis experience were about 18 times greater for those patients who found their physicians adhered to all seven social recommendations than for those who found their physicians did not adhere to any social recommendations, if controlling for all the other variables in the model.

Similarly, adherence to emotional recommendations was significantly related to patient diagnosis experience (Wald χ² = 25.88, df = 1, p < 0.001; OR = 11.37); for each increase in 1 unit of adherence to emotional recommendations, the estimated odds of having a neutral/positive diagnosis experience would increase by a factor of 11.4 if the other two adherence variables are held constant in the model. In the context of this study, the odds of having a neutral/positive diagnosis experience were 11.4 times greater for those patients who found their physicians adhered to both emotional recommendations (not saying “I’m sorry” and not conveying the diagnosis as bad news) than for those who found their physicians did not adhere to either emotional guideline, if controlling for all the other variables in the model.

For Research Question 2, a relative weight analysis for logistic regression was conducted. The ordinary least square R² statistic was adopted in the logistic relative weight analysis as a proxy for the total variance in patient diagnosis experience that was accounted for by the three adherence variables (Petscher et al. 2013). Table 7 displays the raw variance (i.e., relative weight) explained by each adherence variable, the 95% confidence interval for the individual relative weights, and the rescaled relative weights that were obtained by dividing each relative weight by R². The analysis results indicate that the three adherence variables explained 41% of the variance in patient diagnosis experience (R² = 0.41). The relative weights of adherence to medical, social, and emotional recommendations were 0.04, 0.17, and 0.20, respectively. In other words, the medical, social, and emotional adherence variables accounted for 4%, 17%, and 20% of the variance in patient diagnosis experience, respectively. All the three adherence variables explained a statistically significant amount of the variance in patient diagnosis experience as the 95% CIs for the relative weights did not contain zero. The rescaled relative weights exhibit the relative importance of the three adherence variables more clearly. Most of the explained variance in patient diagnosis experiences was attributed to adherence to emotional recommendations (48% of R²) and adherence to social recommendations (42% of R²); the rest of the explained variance (10% of R²) was attributed to adherence to medical recommendations.

Table 7.

Summary of logistic regression relative weight analysis

Variable	RW	CI-L	CI-U	RS-RW (%)
Medical	0.0401	0.0088	0.0958	9.77%
Social	0.1717	0.0746	0.2674	41.77%
Emotional	0.1992	0.1077	0.3084	48.46%

RW Relative Weight; CI-L Confidence Interval around Relative Weight – Lower Bound; CI-U Confidence Interval around Relative Weight – Upper Bound; RS-RW Rescaled-Relative Weight

Comment

Principal findings

This study contributes to the literature by assessing whether physician adherence to professional recommendations for delivering prenatal screening/diagnosis results impact patient diagnosis experiences, and if so, which subset of these recommendations (medical, social, or emotional) most correlate with positive patient experiences. For our first research question, we found that adherence to both social and emotional recommendations was significantly related to neutral/positive patient diagnosis experiences, with an effect size of moderate to large for the majority of those 11 practices. Moreover, we found that adherence to emotional and social recommendations explained relatively more variance in patient diagnosis experiences than adherence to medical recommendations.

Our findings are consistent with previous literature in demonstrating the high percentage of negative experiences and the importance of psychosocial information and support (May et al. 2020; Meredith et al. 2023; Nelson Goff et al. 2013). However, this is the first study that has found the extent to which physicians follow the emotional and social recommendations can result in strikingly different patient perceptions of their diagnosis experiences as neutral/positive vs. negative.

Clinical implications

Given that over half of patients described their diagnosis experiences as negative due to the absence of social and emotional information and supports, the clinical implications are that obstetricians and gynecologists should recognize that saying “I’m sorry” or assuming that a potential diagnosis of DS is “bad news” can project a value-based judgement about life with DS and negatively influence the patient experience. Clinicians can still convey empathy and understanding for some patients who may feel emotional distress upon learning about a potential DS diagnosis or related comorbidities. To do so effectively, clinicians can follow best practices as outlined in the UK Consensus Guidelines for Communication in Obstetric Ultrasound and our previously published Recommendations to initially deliver the potential DS diagnosis with neutral and unbiased language and then provide emotional support and reflective listening as needed based on patient reactions (Johnson et al. 2020; Meredith et al. 2022). In addition, clinicians need to be sure to provide accurate, balanced, and up-to-date information about the life outcomes for people with DS and talk to their patients, in depth, about the improved social and life outcomes for people with DS that have been developed and achieved in recent years. Systemically, professional medical organizations can offer more explicit guidelines about how to provide social and emotional support and training for clinicians to navigate delicate conversations such as the Brighter Tomorrows training modules (Kleinert et al. 2009). An investment in these strategies could improve the diagnosis experience for patients, improve the emotional health of pregnant patients, and strengthen relationships of trust between providers and patients, particularly for those with intersectional identities (Carroll et al. 2018; Chung et al. 2023).

Research implications

Further research needs to determine if patients learning about other prenatal diagnoses have similar needs and if the provision of information about social determinants of health for people with disabilities at this early stage can potentially improve the health of medically vulnerable children (Kennedy and Wood 2020). Moreover, additional research needs to examine the specific needs of patients who terminate pregnancies following a diagnosis.

Strengths and limitations

This study is the largest study to date of those who received a prenatal diagnosis of Down syndrome and continued pregnancy and is very recent with a statistically rigorous analysis.

Several limitations should be noted within this study (Ijaz et al. 2024; Meredith et al. 2023):

1. Like most of the previous studies on patient diagnosis experiences, our sample included an over-representation of White respondents. This limitation is especially important given disparities in health care and outcomes for individuals with DS in minority populations, as well as cultural understandings of disability (Chung et al. 2023; Krell et al. 2023).

2. When our survey participants completed the quantitative questions about what their physician did or did not do, we did not ask them to specify if they were referring to their OB or MFM specialist (for those patients who had both). To minimize this limitation, we did not include the mixed responses in our SA scores, in which multiple providers were rated differently by the same patient.

3. Participant recall bias can be a factor; however, our respondents’ experiences were relatively recent (2016–2021) and the reception of a possible prenatal diagnosis of DS has been documented as a “flashbulb memory” (May et al. 2020).

4. Our findings may be positively skewed since advocacy organization participants lived in regions with medical outreach programs; thus, these results might not be generalizable to expectant couples without those supports.

5. We did not control for any background variable(s) in the analysis and therefore do not know to what extent patient diversity may have impacted our results, nor can we be certain that our results would generalize to a more international study.

Conclusions

Given the profound impact of emotional, social, and informational recommendations on the diagnosis experiences of patients receiving prenatal screening results, OBs can significantly improve patient experiences and support their psychological and emotional needs by providing 1) access to accurate, balanced, and up-to-date information about disabilities and 2) being better trained on disability awareness and delivery of prenatal screening results.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

S.M.: conceptualization, data curation, formal analysis (sentiment analysis), funding acquisition, investigation, methodology, project administration, resources, supervision, visualization, writing-original draft, writing-review and editing H.K.: formal analysis (sentiment analysis), methodology, project administration, writing-original draft, writing-review and editing J.L.: formal analysis (statistics), writing-original draft, writing-review and editing, and visualization S.W.: data curation, formal analysis (sentiment analysis), methodology, writing-review and editing J.D.: formal analysis, writing-review and editing.

Data availability

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Declarations

Ethics statement

The study was approved for human subjects on 4/6/2020 research by the University of Kentucky IRB (55360).

Patient consent statement

The study only included participants who were 18 or older, and participants completed IRB-approved online consent forms.

Competing interests

Dr. Harold Kleinert, Jian Li, Sierra Weiss, and Jonathan Drummond have no conflicts of interest to declare. Dr. Stephanie Meredith serves on the advisory board of the National Down Syndrome Congress in a volunteer capacity and the Trisomy Collaborative as a paid consultant and is the parent of an adult with Down syndrome. She has also received remuneration from Down syndrome non-profit organizations and non-profit medical/genetics organizations/programs (National Society of Genetic Counselors, American College of Osteopathic Obstetricians and Gynecologists, American College of Medical Genetics and Genomics, and American Society of Human Genetics) for speaking engagements and associated travel expenses.