Informing Parents About Expanded Newborn Screening: Influences on Provider Involvement WHAT’S KNOWN ON THIS SUBJECT: Expansion in NBS increases the need to ensure that parents are informed about NBS before sample collection. Parents report wanting to receive some information in the prenatal period; doing so improves comprehension and responsiveness to NBS results. WHAT THIS STUDY ADDS: Health care providers positioned to inform parents about NBS before sample collection report differing views on this role and the cognitive and material barriers faced. Targeted educational initiatives are required to bolster confidence and engage them in this responsibility.
CONTRIBUTORS: Robin Z. Hayeems, PhD,a Fiona A. Miller, PhD,a Julian Little, PhD,b June C. Carroll, MD,c Judith Allanson, MD,d,e Pranesh Chakraborty, MD,d,e Brenda J. Wilson, MBChB,b Jessica P. Bytautas, BA,a and Robert J. Christensen, BSca Departments of aHealth Policy, Management, and Evaluation and cFamily and Community Medicine, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada; Departments of bEpidemiology and Community Medicine and ePediatrics, University of Ottawa, Ottawa, Ontario, Canada; and dDepartment of Genetics, Children’s Hospital of Eastern Ontario, Ottawa, Ontario, Canada KEY WORDS public policy, newborn screening, survey design, professional attitudes
abstract OBJECTIVE: Expanded newborn screening (NBS) identifies some disorders for which clinical benefit is uncertain, as well as “incidental” findings (eg, carrier status), thus enhancing the need to inform parents about NBS before sample collection. METHODS: A self-complete survey was sent to a cross-sectional, stratified, random sample of 5 provider groups in Ontario (obstetricians, midwives, family physicians, pediatricians, and nurses). Univariate and multivariate analyses were used to investigate the effects of core beliefs, perceived barriers, and demographic characteristics on the reported frequency of informing parents about NBS before sample collection. RESULTS: Virtually all of the midwives and almost half of the nurses reported discussing NBS with parents, whereas less than one sixth of the physicians did so. Providers who perceived a responsibility to inform parents were 3 times more likely to report doing so than those who did not perceive this responsibility (odds ratio: 2.9 [95% confidence interval: 2.1– 4.1]). Those who lacked confidence to inform parents were 70% less likely to discuss NBS with parents compared with those who did not experience this cognitive barrier (odds ratio: 0.3 [95% confidence interval: 0.2– 0.4]). Controlling for these covariates, family physicians and obstetricians were more likely than pediatricians to inform parents. CONCLUSIONS: These results provide guidance for capacity building among providers who are positioned to inform parents about NBS before sample collection. Our findings call for targeted educational interventions that consider patterns of provider practice related to prenatal and NBS care, seek to redress confidence limitations, and engage key provider groups in the importance of this professional responsibility. Pediatrics 2009;124:950–958
950
HAYEEMS et al
ABBREVIATIONS NBS—newborn screening OR— odds ratio CI— confidence interval The study sponsor had no role in study design, data collection, analysis, interpretation, writing of the report, or in the decision to submit articles for publication. The sponsors’ support of this work should not imply endorsement of the conclusions, for which the authors retain sole responsibility. www.pediatrics.org/cgi/doi/10.1542/peds.2008-3148 doi:10.1542/peds.2008-3148 Accepted for publication Feb 23, 2009 Address correspondence to Fiona A. Miller, PhD, University of Toronto, Department of Health Policy, Management, and Evaluation, 155 College St, Toronto, Ontario, Canada M5T 3M6. E-mail: fi
[email protected] PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). Copyright © 2009 by the American Academy of Pediatrics FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
ARTICLES
Universal newborn screening (NBS) for phenylketonuria and congenital hypothyroidism has been commonplace since the 1960s and 1970s. Given the severity and treatability of these conditions, NBS has long operated under a mandatory- or implied-consent model,1–3 with limited attention paid to informing parents about NBS before obtaining a heel-prick sample. Currently, expanded NBS panels include disorders for which clinical benefit may be uncertain, as well as an array of “incidental” findings (eg, carrier results and benign variants). With this shift from a “public health emergency” to a “public health service” mandate,4 there is an increasing need to ensure that parents are provided with information before sample collection.5–9 In addition, evidence suggests the following: (1) parents want to receive some information, especially in the prenatal period10–14; (2) providing information is necessary in the context of NBS expansion research paradigms7; and (3) informing parents before birth can enhance comprehension of results and uptake of follow-up services.8,15–19 However, distributing this responsibility among relevant health care providers and equipping them to assume this role pose challenges.20 Although Kerruish et al21 have examined the role of providers in informing parents in New Zealand and some research in North America has explored the attitudes of health care providers toward NBS expansion generally,22,23 evidence on the actual practices of health care providers related to this task and their perceived responsibility for it is limited in the North American context. Within 1 year of the onset of the expanded NBS program in Ontario, we examined how providers positioned to inform parents about NBS perceived or acted on this responsibility and what factors influenced their practice. PEDIATRICS Volume 124, Number 3, September 2009
PARTICIPANTS AND METHODS We surveyed providers about their perceived roles and responsibilities related to general NBS care and about their attitudes toward NBS for sickle cell disorders (SCD), in particular. Because of the latter issue, we approached genetics professionals and hematologists, in addition to obstetricians, midwives, family physicians, and obstetric/postpartum nurses. Most births in Ontario are in the hospital, and the heel-prick sample is typically taken by the maternal/newborn nurse, a hospital technician, or a midwife. For the current analysis, we focused on providers involved in routine prenatal care (obstetricians, midwives, and family physicians) or care of infants in the first days of life before or during the heel prick (pediatricians and nurses) and, therefore, excluded data from genetics professionals and hematologists. Potential respondents were identified using MDSelect, the Canadian Medical Directory, and directories from the College of Midwives of Ontario, the College of Nurses of Ontario, and the Canadian Association of Genetic Counselors. Twenty-seven obstetricians who indicated “oncology” as a subspecialty in the MDSelect database were excluded from the sampling frame. A self-administered questionnaire was mailed to a stratified random sample of these provider groups and was accompanied by a cover letter, a $2 coffee-shop coupon, and a postage-paid reply envelope. Five mailings were completed over an 8-week period.24 The survey was developed by a multidisciplinary team based on a review of the literature and pilot testing among 2 to 3 members of each group who were surveyed to ensure face validity. The initial sections of the questionnaire (ⱕ15 items) were providergroup specific to assess involvement in prenatal care or infant care in the
first days of life and to gauge perceptions of barriers related to NBS care. The bulk of the questionnaire (69 items) was the same for each provider group. Using categorical responses and 5-point Likert scales measuring strength of agreement or frequency of practice, we assessed the following: (1) knowledge and confidence about NBS and sickle cell disorders; (2) involvement in NBS or care of persons with SCD; (3) practices and attitudes regarding informing parents about NBS before sample collection or caring for families with positive screening results; (4) core beliefs about NBS generally; and (5) demographics. Data were entered by using Snap Survey Software (Snap Surveys Ltd, Portsmouth, NH) and analyzed by using SAS 9.1 (SAS Institute Inc, Cary, NC). Data from family physicians, obstetricians, midwives, pediatricians, and nurses who reported that they were not involved in prenatal care or care of infants in the first days of life were excluded from the analyses in this article. Frequency counts and percentages were computed for all of the variables. We compared demographic characteristics, practices regarding information provision to parents, core beliefs, and perceived barriers between the provider groups using the 2 test. Because virtually all of the midwives reported that they consistently or usually discussed NBS with parents before sample collection, we excluded midwives from the remainder of the analysis. Next, we calculated the odds ratios (ORs) and 95% confidence intervals (CIs) of the frequency of the provider having a general discussion about NBS with parents as the frequency relates to provider group, core beliefs, material barriers, cognitive barriers, and demographics. The frequency of having a discussion was categorized as “rarely or never” (the reference 951
TABLE 1 Response Rates Variable
Total, Ineligible, Could Adjusted Completed Surveys, n Response Rate % n Not Be Located, (Total Minus (Completed/Adjusted) Involved in Not Involved n Ineligible), Prenatal in Prenatal n Care Carea
Obstetricians 498 Midwives 339 Family physicians 729 Pediatricians 569 Nurses 725 Total 2860 a
106 13 144 136 51 450
392 326 585 433 674 2410
168 250 209 157 460 1244
44 0 87 116 29 276
54.1 76.7 50.6 63.0 72.6 63.1
Data were included in response rate calculations but excluded from analyses.
category), “sometimes,” or “consistently or usually.” Ordinal logistic regression was used when the proportional odds assumption was met (ie, OR was the same irrespective of the cutoff point chosen between the dichotomized categories of the outcome variable). Where variables departed from this ordinal trend (eg, provider’s view regarding whether NBS should be mandatory and length of time in practice), multinomial logistic regression was used. Given the possibility that the factors associated with having a general discussion with parents were themselves
interrelated, we performed a multivariate analysis using forward selection with P ⬍ .05 as the criteria for significance. Because the proportional odds assumption was met for each independent variable in this analysis, we treated the outcome variable as ordinal. Forward selection, backward elimination, and stepwise selection procedures generated the same results. The final model includes 5 variables and is reported with adjusted ORs and 95% CIs. To verify our interpretation of the final model, we conducted a second multivariate analysis
restricted to physicians and a third multivariate analysis restricted to nurses.
RESULTS Of 2410 eligible surveys, the overall adjusted response rate was 63.1% (Table 1). The majority of providers worked in nonacademic centers (75.2%), used fee-for-service reimbursement methods (67.7%), and practiced in urban settings (63.2%). Even ignoring midwives and nurses, all of whom were women, the proportion of women differed significantly among provider groups (P ⫽ .002). Unlike the majority of provider groups that had been in practice for ⬎10 years, the majority of midwives had been in practice for ⬍10 years (P ⬍ .0001; Table 2). Providers differed markedly on how consistently they discussed NBS with parents before sample collection (P ⬍ .0001; Table 3). Almost all of the midwives and half of the nurses stated that they consistently or usually had
TABLE 2 Summary of Respondent Characteristics Characteristics
Obstetricians, n (%)
Midwives, n (%)
Family Physicians, n (%)
Pediatricians, n (%)
Nurses, n (%)
2P
Total no. of respondents Practice setting Academic Nonacademic Method of reimbursementd Fee for service Nonfee for service Geographic location Urban located Nonurban located Gendera Men Women Years in practice ⬎10 ⬍10
168
250
209
157
460
—
54 (32.5)a 112 (67.5)
26 (10.5)b 221 (89.5)
20 (9.8)c 184 (90.2)
114 (26.9)d 42 (73.1)
92 (20.0)d 367 (80.0)
⬍.0001 —
152 (91.0)d 15 (9.0)
—
106 (51.5)b 100 (48.5)
101 (64.3) 56 (35.7)
—
⬍.0001 —
123 (73.7) 45 (26.3)
127 (51.4)b 120 (48.6)
120 (58.5)f 85 (41.5)
122 (77.7) 35 (22.3)
287 (63.1)c 168 (38.7)
⬍.0001 —
93 (56.4) 72 (43.6)b
—
84 (40.8) 122 (59.2)b
90 (57.3) 67 (42.7)
—
— .002
98 (59.0)a 68 (61)
52f (21.4) 191 (78.6)
121 (59.9)h 81 (40.1)
99 (63.9)a 56 (26.1)
297 (67.5)i 143 (32.5)
e
g
— indicates no data. a Information is missing for 2 respondents. b Information is missing for 3 respondents. c Information is missing for 5 respondents. d Information is missing for 1 respondent. e Data were excluded because 100% of midwives and registered nurses were coded as nonfee for service. f Information is missing for 4 respondents. g Data were excluded because 100% of midwives and registered nurses were women. h Information is missing for 7 respondents. i Information is missing for 20 respondents.
952
HAYEEMS et al
e
g
⬍.0001 —
ARTICLES
TABLE 3 Summary of Respondent Involvement in Information Provision Before NBS Have a General Discussion About NBS With Parents Before Sample Collection
Obstetricians (N ⫽ 167), n (%)
Midwives (N ⫽ 244), n (%)
Family Physicians (N ⫽ 195), n (%)
Pediatricians (N ⫽ 155), n (%)
Nurses (N ⫽ 456), n (%)
2P
Consistently/usually Sometimes Rarely/never
26 (15.6) 30 (18.0) 111 (66.5)
240 (98.4) 2 (0.8) 2 (0.8)
33 (16.9) 35 (18.0) 127 (65.1)
14 (9.0) 40 (25.8) 101 (65.2)
212 (46.5) 116 (25.4) 128 (28.1)
⬍.0001 ⬍.0001 ⬍.0001
general discussions with parents before sample collection (98.4% and 46.5%, respectively), whereas an indistinguishable majority of obstetricians, family physicians, and pediatricians rarely or never did so (65.0%– 66.0%). Provider core beliefs about NBS and their perception of potential barriers to informing parents were significantly different (Table 4; P ⬍ .0001). Midwives (97.2%) were most likely to perceive a professional responsibility to inform parents (defined as “disagree/strongly disagree to the
absence of a responsibility”), whereas obstetricians were least likely (41.3%). Although the responsibility perceived by midwives corresponded with their actual involvement in informing parents, a majority of physicians rarely or never had discussions with parents about NBS, although 41.3% of obstetricians, 56.3% of pediatricians, and 72.2% of family physicians perceived a professional responsibility to do so. With the sole exception of midwives (15.8%), the majority of all of the provider groups agreed that NBS should be mandatory (73.4%).
Physician respondents reported experiencing more material barriers (insufficient time or insufficient compensation) than nonphysician respondents. The majority of obstetricians, family physicians, and pediatricians agreed (defined as agreed or strongly agreed) that they had insufficient time to provide information about NBS before sample collection (65.2%, 65.6%, and 58.6%, respectively), whereas minorities of nurses (37.9%) and midwives (17.3%) agreed that time was a barrier. Similarly, at least half of all of the physicians agreed that insufficient
TABLE 4 Summary of Respondent Beliefs and Barriers Related to NBS Core Beliefs and Barriers
Obstetricians
Midwives
Family Physicians
Pediatricians
Nurses
2P
Total no. of respondents Core beliefs, n (%) Perceive responsibility of provider group to inform parentsa NBS should be mandatoryf Material barriers Insufficient timej Insufficient compensationj Cognitive barriers Insufficient trainingj Not up to date on conditions included in NBS panelp Not confident to explain NBS to parentst
168
250
209
157
460
NA
69 (41.3)b 98 (59.8)g
240 (97.2)c 39 (15.8)c
138 (72.3)d 121 (59.3)h
80 (56.3)e 126 (80.8)b
350 (78.7)e 374 (82.9)i
⬍.0001 ⬍.0001
107 (65.2)g 119 (72.6)g
43 (17.3)b 62 (25.1)c
126 (65.6)k 112 (58.3)k
85 (58.6)l 69 (48.6)e
168 (37.9)k 81 (18.6)m
⬍.0001 ⬍.0001
106 (64.2)c 84 (53.2)q 82 (51.9)q
80 (32.3)n 91 (36.7)n 49 (19.8)n
120 (62.5)k 110 (54.5)r 111 (55.0)r
41 (29.1)o 31 (21.1)q 21 (14.3)q
221 (49.8)o 221 (50.5)s 171 (39.1)u
⬍.0001 ⬍.0001 ⬍0.0001
NA indicates not applicable. a Respondents disagree/strongly disagree to the absence of a responsibility. b Information is missing for 1 respondent. c Information is missing for 3 respondents. d Information is missing for 18 respondents. e Information is missing for 15 respondents. f Respondents agree/strongly agree that NBS should be mandatory. g Information is missing for 4 respondents. h Information is missing for 5 respondents. i Information is missing for 9 respondents. j Respondents agree/strongly agree that insufficient time/training/compensation was a barrier to informing parents about NBS before sample collection. k Information is missing for 17 respondents. l Information is missing for 20 respondents. m Information is missing for 24 respondents. n Information is missing for 2 respondents. o Information is missing for 16 respondents. p Respondents disagree/strongly disagree that they are up to date on conditions included in Ontario’s NBS program. q Information is missing for 10 respondents. r Information is missing for 7 respondents. s Information is missing for 22 respondents. t Respondents disagree/strongly disagree that they are confident in their ability to explain NBS to parents. u Information is missing for 23 respondents.
PEDIATRICS Volume 124, Number 3, September 2009
953
compensation was a barrier (obstetricians 72.6%, family physicians 58.3%, and pediatricians 48.6%), whereas minorities of nurses (18.6%) and midwives (25.1%) agreed. Majorities of obstetricians and family physicians agreed that insufficient training was a barrier (64.2% and 62.0%, respectively), that they were not up to date on the conditions being screened (53.2% and 54.5%, respectively), and that they lacked confidence in their ability to explain NBS to parents (51.9% and 55.0%, respectively). Midwives and pediatricians were least limited by training (32.3% and 29.1%, respectively), knowledge (36.4% and 21.1%, respectively), and confidence (19.8% and 14.3%, respectively). The cognitive challenges faced by nurses were intermediate (49.8%, 50.5%, and 39.1% agreed to being limited by training, knowledge, and confidence, respectively). Table 5 describes the effects of provider group, core beliefs, material or cognitive barriers, and demographics on discussing NBS before sample collection. Nurses were 5 times more likely than obstetricians to have these discussions. Respondents who felt that their own provider group was responsible to inform parents were 6 times more likely to report doing so than those who did not feel this responsibility, whereas those who believed that NBS should be mandatory were half as likely to report having these discussions as those who did not share this view. Respondents who experienced material or cognitive barriers related to NBS were 40% to 70% less likely to report informing parents before screening. Finally, nonacademic, nonurban, and recent-to-practice providers were 1.5 to 2.0 times more likely to report having these discussions. Next, we used multivariate analysis to identify key factors among provider group, core beliefs, perceived barriers, and demographics that are 954
HAYEEMS et al
TABLE 5 Effects of Provider Group, Beliefs, Barriers, and Demographic Characteristics on Reported Provision of Information About NBS to Parents Before Sample Collection Variable
Have a General Discussion, n (%) Constantly/ Sometimes Usually
Provider groupc Obstetricians Family physicians Pediatricians Nurses Core beliefs Do not feel responsible to inform Feel responsible to inform NBS should not be mandatory NBS should be mandatorye
Material barriers Sufficient time Insufficient time Sufficient compensation Insufficient compensation Cognitive barriers Sufficient training Insufficient training Up to date regarding NBS Not up to date regarding NBS Confident regarding NBS Not confident regarding NBS Demographics Academic Nonacademic Fee for service Nonfee for servicef Urban located Nonurban located Men Womenf ⬎10 y in practice ⬍10 y in practicee
Unadjusted OR (95% CI)b
Rarely/ Nevera
26 (15.6) 33 (16.9) 14 (9.0) 212 (46.5)
30 (18.0) 35 (17.95) 40 (25.8) 116 (25.4)
111 (66.5) 127 (65.1) 101 (65.2) 128 (28.1)
1.00 (ref) 1.07 (0.70–1.62) 0.96 (0.61–1.51) 5.08 (3.53–7.29)d
45 (14.5)
55 (17.7)
211 (67.8)
1.00 (ref)
472 (55.0) 246 (54.0)
157 (18.3) 55 (12.1)
230 (26.8) 155 (34.0)
6.15 (4.68–8.08)d 1.00 (ref)
272 (36.6)
165 (22.2)
306 (41.2) Constantly/usually to rarely/never: 0.56 (0.43–0.73)d; sometimes to rarely/never: 1.52 (1.06–2.18)d
373 (57.2) 145 (28.0) 373 (51.4) 137 (31.6)
110 (16.9) 102 (19.7) 137 (18.9) 74 (17.1)
169 (25.9) 272 (52.4) 216 (29.8) 222 (51.3)
1.00 (ref) 0.30 (0.24–0.38)d 1.00 (ref) 0.42 (0.33–0.53)d
334 (54.6) 183 (32.9) 319 (49.3) 187 (35.8)
104 (17.0) 109 (19.6) 109 (16.8) 101 (19.4)
174 (28.4) 264 (47.5) 219 (33.8) 234 (44.8)
1.00 (ref) 0.42 (0.34–0.53)d 1.00 (ref) 0.60 (0.48–0.75)d
395 (53.0) 112 (26.5)
132 (17.7) 77 (18.2)
218 (29.3) 234 (55.3)
1.00 (ref) 0.33 (0.26–0.41)d
80 (35.1) 442 (45.2) 48 (13.7) 25 (15.3) 29 (38.1) 230 (51.9) 37 (14.3) 36 (14.2) 228 (35.1) 281 (52.8)
43 (18.9) 178 (18.2) 73 (20.8) 32 (19.6) 141 (18.5) 81 (18.3) 50 (19.4) 54 (21.3) 143 (22.0) 75 (14.1)
105 (46.0) 1.00 (ref) 359 (36.7) 1.50 (1.14–1.96)d 230 (65.5) 1.00 (ref) 106 (65.0) 1.04 (0.71–1.53) 331 (43.4) 1.00 (ref) 132 (29.8) 1.77 (1.42–2.22)d 171 (66.3) 1.00 (ref) 164 (64.6) 1.06 (0.74–1.52) 278 (42.8) 1.00 (ref) 176 (33.1) Constantly/usually to rarely/never: 1.95 (1.51–2.52)d; sometimes to rarely/never: 0.83 (0.59–1.16)
ref indicates referent. a Rarely/never is a reference category for the outcome variable. b The outcome variable (frequency of reporting a general discussion) was dichotomized and treated as ordinal because the OR was the same irrespective of the cutoff point chosen between the categories. c Midwives were removed from all of the univariate analyses. d P ⬍ .05. e Because the distribution of these variables departed from an ordered trend, the outcome was treated as multinomial. f Nurses were removed from this univariate analysis.
associated with discussing NBS before sample collection. Model 1 (Table 6) identifies the variables that remained significant predictors of practice. Respondents who believed that their own provider group was responsible to inform parents were 3 times more likely
to report having discussions than those who did not perceive this responsibility. Those who lacked confidence to inform parents were 70% less likely to discuss NBS before sample collection compared with those who did not experience this barrier. Finally,
ARTICLES
TABLE 6 Factors Associated With Reported Provision of Information About NBS to Parents in Multivariate Analyses Adjusted OR (95% CI)a
Variable Model 1 Provider groupb Obstetricians Family physicians Pediatricians Nurses Beliefs Do not feel responsible Feel responsible NBS should not be mandatory NBS should be mandatory Barriers Confident Lack confidence Sufficient training Insufficient training Demographics Men Women Urban located Nonurban located
Model 2
should be mandatory were twice as likely to report informing parents than nurses who did not share this belief.
Model 3
DISCUSSION 1.00 — 0.56 (0.33–0.94) 3.08 (1.91–4.98)
1.00 — 0.44 (0.25–0.77) —
— — — —
1.00 2.92 (2.08–4.08) — —
1.00 3.81 (2.37–6.13) — —
1.00 2.36 (1.48–3.76) 1.00 1.98 (1.12–3.26)
1.00 0.33 (0.24–0.46) — —
1.00 0.33 (0.20–0.56) 1.00 0.56 (0.34–0.90)
1.00 0.41 (0.27–0.60) — —
1.00 1.522 (1.011–2.292) 1.00 1.494 (1.100–2.028)
1.00 1.760 (1.143–2.710) 1.00 1.927 (1.234–3.010)
— — — —
Model 1 included obstetricians, family practitioners, pediatricians, and nurses. Model 2 included obstetricians, family practitioners, and pediatricians. Model 3 included nurses only. — indicates no data. a ORs were calculated with outcome treated as the ordinal. b Given the unequivocally predictive effect of being a midwife, midwives were removed from multivariate analyses; provider group variable was removed in entirety from model 3.
providers located in nonurban areas were 1.5 times more likely to have general discussions than urban-located providers. Thus, holding beliefs, barriers, and demographics constant, family physicians and obstetricians were similarly likely to inform parents, and nurses were more likely to do so, being 3 times as likely to inform parents as obstetricians despite their reduced confidence. However, pediatricians were half as likely as obstetricians to do so despite their greater confidence. Given the differences in practice role and compensation structure among nurses and physicians, we verified our interpretation of model 1 by carrying out further multivariate analyses restricted to physicians (model 2) and nurses (model 3; Table 6). In model 2, physicians who believed that their provider group was responsible to inform parents were ⬃4 times more likely to report informing parents than those who did not perceive this responsibility. Physicians who lacked training and PEDIATRICS Volume 124, Number 3, September 2009
confidence to inform parents were 40% to 70% less likely to inform parents before sample collection compared with those who did not experience these barriers, and female and nonurban-located physicians were almost twice as likely to report informing parents than male and urban-located physicians, respectively. Finally, taking into account other factors associated with providing information to parents, pediatricians would inform parents half as often as obstetricians, possibly because of their limited access to parents at appropriate times. In model 3, nurses who believed that they were responsible to inform parents were twice as likely to report discussing NBS with parents than those who did not perceive this responsibility. Those who lacked confidence to inform parents were approximately half as likely to report discussing NBS before sample collection compared with those who did not experience this barrier. Finally, nurses who believed that NBS
Almost all of the midwives and nearly half of the nurses reported consistently or usually having general discussions with parents, but the majority of physicians reported rarely or never doing so. Because our sampling strategy deliberately solicited the views of nurses for whom NBS sample collection is likely to be routine, it is not surprising that, compared with physicians, nurses reported informing parents about NBS more frequently. There was also some discrepancy between the perceived responsibility to inform parents and actual practice, with elevated perceptions of responsibility for each provider group alongside reduced actual involvement among individual providers. Respondents may assume that the provider who obtains the sample (oftentimes a nurse) also fulfills the broad responsibility of informing parents. This speculation would be supported by the suggestion that a shared responsibility may indeed be diffused among involved provider groups.20 Although univariate analyses suggested that core beliefs, perceived barriers, provider group, and demographics significantly influenced practice, 3 separate multivariate models identified a smaller and consistent set of significant independent variables. A first finding is that, irrespective of provider group, those who perceived a professional responsibility to inform parents were significantly more likely to do so, and those who had reduced confidence in their ability to inform parents were significantly less likely to do so. In addition, female providers and those working in nonurban areas were more likely to inform parents about NBS before sample collection. 955
A second finding is that provider group membership was a significant predictor of practice. Indeed, being a midwife was so predictive of practice that we excluded midwives from our modeling. In addition, after accounting for other factors, nurses were more likely than obstetricians to report informing parents, whereas pediatricians were less likely than obstetricians to report doing so. These trends conform to what is known about provider practice patterns. Midwives are well positioned to inform parents, as are most maternal/ newborn nurses. Pediatricians are only sometimes involved in primary care in the first days of life and are generally not involved in prenatal care; presumably it is their elevated level of reported confidence that encourages them to provide information, notwithstanding. Obstetricians and family physicians may play similar prenatal care roles, although only family physicians are involved in newborn care; regardless, they are equally likely to inform parents before sample collection. A third finding is that beliefs, barriers, and demographics that were significant in univariate analyses were not retained in the multivariate models. Restricting analyses to physicians (model 2) and nurses (model 3) verified our interpretation of model 1. In particular, key material barriers (insufficient time and insufficient compensation) were not independently influential predictors. In addition, beliefs regarding the appropriateness of mandatory NBS did not significantly predict provider involvement in informing parents, a finding that highlights the distinction between information provision and consent.21 Finally, the apparently positive influences of practicing outside an academic setting, being compensated through nonfee-for-service mechanisms, or being new to practice were not retained in the multivariate analyses. 956
HAYEEMS et al
These results provide clear guidance for policy makers concerned with enhancing the involvement of providers in informing parents about expanded NBS before sample collection. Educational interventions are clearly warranted, both to redress confidence limitations and to engage relevant providers in considering this professional responsibility. Providing more time or financial investments to incent providers to assume this role may be secondary. Given the timing and nature of the role of pediatricians and the lack of cognitive barriers in this group, it may make little sense to invest resources to enhance their involvement in parental education initiatives before sample collection. Similarly, midwives face few self-reported cognitive barriers and provide a proportionally small amount of perinatal care (ie, only 6% of pregnancy care and 4% of birth care in Ontario25), reducing the need for educational interventions for this group. By contrast, because maternal/ newborn nurses are well positioned and willing to inform parents, their cognitive barriers should be addressed as a matter of some urgency to increase both the quantity and quality of their informational role. Obstetricians and family physicians are similarly positioned to inform parents in the prenatal period, but our data suggest that they have been largely uninvolved in doing so. The limited confidence of majorities of both groups was an important influence on reduced involvement and could be addressed through educational interventions. However, although a majority of family physicians perceived a responsibility to inform parents, only a minority of obstetricians did so—a core belief that may be very difficult to change. Nonetheless, the disproportionate contribution of obstetricians to primary prenatal care (80% of women received pregnancy care from
an obstetrician25) means that it will be essential to engage them if prospective parents are to be given the opportunity to learn about NBS at the optimal time. Because this will indeed be challenging, efforts to inform parents through alternate prenatal venues (ie, prenatal classes) and modalities (ie, information/decision aids) are most certainly warranted. These data reflect the reported practices of providers, not their actual practices or what specific content is integrated into “a general discussion.” Similarly, provider knowledge of NBS was self-reported and not independently measured. Although the survey mailing was timely in response to expanded NBS in Ontario, respondents may not have had sufficient time to learn of these new issues arising to integrate relevant changes into their practice. It would be informative to gauge their views and practices over a longer period of time.
CONCLUSIONS Expanded NBS enhances the longstanding need to inform parents before sample collection. Health care providers positioned to do so must be encouraged to engage this professional responsibility and equipped with the knowledge that they need to perform this task effectively.
ACKNOWLEDGMENTS We thank the Ontario Ministry of Health and Long-term Care for funding for this project (06358). Dr Hayeems is supported by a Capacity for Applied and Developmental Research and Evaluation in Health Services and Nursing Research Program Postdoctoral Fellowship from the Canadian Institutes of Health Research and the Canadian Health Services Research Foundation. Dr Miller is supported by a New Investigator Award from the Institute of Health Services and Policy Research of the Canadian Institutes of Health Research (80495). Dr Little holds a
ARTICLES
Canada Research Chair in Human Genome Epidemiology. Dr Carroll is sup-
ported by the Sydney G. Frankfort Chair in Family Medicine.
We thank Shafagh Fallah for assistance with the statistical analyses.
REFERENCES 1. Mandl KD, Shlomit F, Larson C, Kohane IS. Newborn screening program practices in the United States: notification, research, and consent. Pediatrics. 2002;109(2):269 –273 2. Hiller EH, Landenburger G, Natowicz MR. Public participation in medical policy-making and the status of consumer autonomy: the example of newborn-screening programs in the United States. Am J Public Health. 1997;87(8):1280 –1288 3. Avard D, Vallance H, Greenberg C, Laberge C, Kharaboyan L, Plant M. Variability in the storage and use of newborn dried bloodspots in canada: is it time for national standards? Genomics Soc Policy. 2006;2(3):80 –95 4. Grosse SD, Boyle CA, Kenneson A, Khoury MJ, Wilfond BS. From public health emergency to public health service: the implications of evolving criteria for newborn screening panels. Pediatrics. 2006;117(3):923–929 5. Department of Human Services V. Informed Parental Consent for Newborn Screening in Victoria. Final Report. Victoria, Australia: Health Issues Centre; 2005 6. UK Newborn Screening Programme Centre. Newborn bloodspot screening in the UK: health professional handbook. 2005. Available at: http://newbornbloodspot.screening.nhs.uk/getdata.php? id⫽10953. Accessed July 16, 2009 7. Atkinson K, Zuckerman B, Sharfstein JM, et al. A public health response to emerging technology: expansion of the Massachusetts newborn screening program. Public Health Rep. 2001;116(2): 122–131 8. American Academy of Pediatrics, Newborn Screening Task Force. Serving the family from birth to the medical home: newborn screening—a blueprint for the future: a call for a national agenda on state newborn screening programs. Pediatrics. 2000;106(2 pt 2):389 – 422 9. American College of Obstetrics and Gynecologists. ACOG committee opinion: newborn screening No. 287. Obstet Gynecol. 2003;102(4):887– 889 10. Hargreaves K, Stewart R, Oliver S. Informed choice and public health screening: the case of blood spot screening. Health Expectat. 2005;8(2):161–171 11. Campbell E, Ross LF. Parental attitudes regarding newborn screening of PKU and DMD. Am J Med Genet. 2003;120A(2):209 –214 12. Davis TC, Humiston SG, Arnold CL, et al. Recommendations for effective newborn screening communication: results of focus groups with parents. providers, and experts. Pediatrics. 2006; 117(5 pt 2):S326 –S340 13. Tluczek A, Koscik RL, Farrell PM, Rock MJ. Psychosocial risk associated with newborn screening for cystic fibrosis: parents’ experience while awaiting the sweat-test appointment. Pediatrics. 2005;115(6):1692–1703 14. Yang Y-M, Andrews S, Peterson R, Shah A, Cepeda M. Prenatal sickle cell screening education effect on the follow-up rates of infants with sickle cell trait. Patient Educ Couns. 2000;39(2–3): 185–189 15. Larsson A, Therrell BL. Newborn screening: the role of the obstetrician. Clin Obstet Gynecol. 2002;45(3):697–710, discussion 730 –732 16. Gabbe SG, Neibyl JR, Simpson JL. Obstetrics: Normal and Problem Pregnancies. 4th ed. New York, NY: Churchill Livingston; 2002 17. Pollitt RJ, Green A, McCabe CJ, et al. Neonatal screening for inform errors of metabolism: cost, yield, and outcome. Health Technol Assess. 1997;1:i–iv, 1–202 18. Green J, Hewison J, Bekker H, Bryant LD, Cuckle HS. Psychosocial aspects of genetic screening of pregnant women and newborns: a systematic review. Health Technol Assess. 2004;8(33):iii, ix–x, 1–109 19. Petticrew M, Sowden A, Lister-Sharp D, et al. False negative results in screening programmes: systematic review of impact and implications. Health Technol Assess. 2000;4:1– 60 20. Clayton EW. Talking with parents about newborn screening. J Pediatr. 2005;147(3 suppl):S26 –S29 21. Kerruish NJ, Webster D, Dickson N. Information and consent for newborn screening: practices and attitudes of service providers. J Med Ethics. 2008;34(9):648 – 652 22. Hiraki S, Ormond KE, Kim K, Ross LF. Attitudes of genetic counselors towards expanding newborn screening and offering predictive genetic testing to children. Am J Med Genet A. 2006;140(21): 2312–2319 23. Koopmans J, Hiraki S, Ross LF. Attitudes and beliefs of pediatricians and genetic counselors
PEDIATRICS Volume 124, Number 3, September 2009
957
regarding testing and screening for CF and G6PD: implications for policy. Am J Med Genet A. 2006;140(21):2305–2311 24. Dillman D. Mail and Internet Surveys: The Tailored Design Method. 2nd ed. New York, NY: John Wiley; 2000 25. Provincial Perinatal Surveillance System Committee. Tailoring Services to Pregnant Women and Their Babies in Ontario: 2006 Provincial Perinatal Report. Ottawa, Ontario: Ontario Perinatal Program Partnership Committee; 2006
958
HAYEEMS et al
