Public Reporting of Hospital-Specific Breastfeeding Measures

Abstract

Background:

Establishing breastfeeding in the first days of an infant’s life is important for longer term success in breastfeeding. In 2009, New York State (NYS) was the second state to require maternity care facilities to collect infant feeding information and to publicly disseminate hospital-specific infant feeding statistics. Public reporting of these statistics as performance measures is a strategy to prompt hospitals to improve breastfeeding support.

Objective:

This qualitative study sought to explore how maternity care administrators and clinical staff responded to the mandate for publicly reported performance measures and whether they used this information to improve maternity care practices.

Methods:

This study used a stratified random sample of NYS hospitals with maternity care units. Participants were recruited by email and telephone calls. A total of 25 hospitals participated in the study, and 37 hospital administrators and staff completed in-depth interviews by telephone. The interviews were analyzed using an explanatory framework in NVivo 8.

Results:

Publicly reported hospital-specific breastfeeding measures increased attention to breastfeeding performance. Hospital administrators and staff reported comparing their relative rankings to other hospitals in the state. Some hospitals used publicly reported breastfeeding measures to monitor performance, whereas others were prompted to generate additional measures for more frequent monitoring. Hospitals with relatively low breastfeeding statistics took certain actions to improve their maternity care practices to support breastfeeding. Limitations of the usefulness of publicly reported measures were reported by interview participants.

Conclusion:

Publicly reported, hospital-specific breastfeeding measures may prompt hospitals to monitor and improve maternity care practices related to supporting breastfeeding.

Keywords

breastfeeding breastfeeding rates health policy quality improvement regulation

Well Established

Publicly reported measures of performance are increasingly being used to influence hospital practices and change patient outcomes. The use of such measures is associated with increased performance monitoring and practice improvement.

Newly Expressed

New York State is the second state to implement mandated public reporting of hospital-specific breastfeeding measures. There are no known publications that provide a description of staff and administrators’ perspectives on mandatory hospital reporting of breastfeeding measures. There are also no known qualitative studies that use a representative statewide sample of hospital staff and administrators to explore the influence of mandatory reporting policies on hospital practices.

Background

Public reporting of performance measures is a quality improvement strategy with the goals of prompting hospitals to improve their practices and influencing prospective patients’ decisions about where to seek care.^1-3 The Joint Commission and the Centers for Medicare and Medicaid have been publicly reporting hospital performance data on clinical indicators since 2004.^4,5 Public reporting of performance measures has been associated with hospital staff’s responding with internal practice changes,⁶ performance improvement,^3,7 and better patient outcomes.^3,8,9

The effect of a mother’s in-hospital experience on breastfeeding success cannot be underestimated,¹⁰ and exclusive breastfeeding while in the hospital is considered a measure of success. Healthy People 2020¹⁰ has set national goals for exclusive breastfeeding, but New York State (NYS) has not met these goals. With the aim to improve exclusive breastfeeding measures, NYS was the first known state to mandate public reporting of hospital-specific breastfeeding statistics in this state and the second state after California to publicly disseminate infant feeding and breastfeeding statistics to pregnant women.¹¹ Since 2008, hospitals in New York have been required to report infant feeding data (from birth until discharge or the first 5 days of life, whichever comes first) into the electronic Statewide Perinatal Data System.

In 2009, the NYS Health Commissioner required hospital-specific infant feeding and breastfeeding statistics to be shared with hospital leaders and reported on the NYS Department of Health (DOH) website.^11,12 These statistics also were mandated to be included in each hospital’s Maternity Informational Leaflet, which is required to be given to all pregnant women.^13,14 In 2010, efforts were undertaken to standardize how infant feeding data were collected and reported by NYS hospitals to improve quality and to be consistent with Joint Commission recommendations.^4,14,15 Details of the process of developing and implementing the mandate are described elsewhere.^15-17 This article explores how the NYS mandate for public reporting of hospital-specific breastfeeding measures influences hospital maternity care staff and administrator practices to improve breastfeeding.

Methods

Study Design

This study was conducted using in-depth, semistructured interviews. The qualitatively trained principal investigator and a trained graduate research assistant conducted the interviews with maternity care unit staff, including lactation consultants, and administrative staff, including the directors of maternity care units or chief nursing officers. The development of the interview guide was informed by findings from a small formative study of 10 NYS hospitals (n = 7, 70% response rate).¹⁷ The study identified preliminary themes, which guided the development of this study’s interview guide, including hospital clinical priorities, use of data, and maternity care clinical practices that supported breastfeeding (see the appendix for the interview guide). This study was reviewed and approved by the NYS DOH Institutional Review Board.

Sampling

A stratified random sampling method was used to obtain a representative sample of NYS hospitals. There were 127 hospitals with maternity care units in NYS at the time of the study. Ten hospitals selected for the formative study were excluded to avoid participant burden. The remaining 117 hospitals were stratified based on 3 variables related to 2011 breastfeeding outcomes: hospital breastfeeding policy score, number of live births, and the percentage of healthy infants who were exclusively breastfed. The hospital breastfeeding policy score reflected the number of components (maximum of 28) in compliance with NYS’s hospital regulations and the Breastfeeding Mothers’ Bill of Rights, which were included in the hospital’s written breastfeeding policy.^17-19

For each variable, hospitals were divided into 2 groups, using the median as the dividing point (below the median vs at or above the median). This created 8 strata of hospitals based on 2 levels of the number of births, exclusive breastfeeding percentage, and hospital breastfeeding policy scores. The 4 strata that included lower breastfeeding policy scores were oversampled (60% vs 40%), with the expectation that they would have a lower participation rate. With an expected overall 50% response rate from hospitals, 40 hospitals were selected, with the goal of obtaining 2 interviews with different types of staff from 20 hospitals.

Staff names, titles, and contact information for each hospital in the sample were obtained from NYS DOH personnel, who provided the mandated hospital-specific breastfeeding data to the hospitals. The sampling resulted in a total of 37 staff and administrator interviews, representing 25 hospitals. Table 1 describes the hospital sample and Table 2 describes the sample of hospital administrators and staff.

Table 1.

Hospital Sampling Stratification for Study Representativeness (n = 25).

Sampling Category	Hospitals in Sample, n (%)
Low breastfeeding rate, low total delivery, low policy score	4 (16)
Low breastfeeding rate, low total delivery, high policy score	1 (4)
Low breastfeeding rate, high total delivery, low policy score	2 (8)
Low breastfeeding rate, high total delivery, high policy score	5 (20)
High breastfeeding rate, low total delivery, low policy score	5 (20)
High breastfeeding rate, low total delivery, high policy score	3 (12)
High breastfeeding rate, high total delivery, low policy score	2 (8)
High breastfeeding rate, high total delivery, high policy score	3 (12)
Total	25 (100)

Table 2.

Description of Hospital Staff and Administrators’^a Job Titles (N = 37).

Category of Staff/Administrator	Types of Job Titles	Sample, n (%)
High-level administrator	Chief nursing officer, vice president of patient services	7^b (18.9)
Mid-level administrator	Patient care director, nursing service administrator, director of maternal child health & nursing education, director of women’s services	7 (18.9)
Staff manager	Clinical director, director of nursing, program director lactation services, nurse manager	10 (27.0)
Staff	Lactation consultant, staff nurse, parent coordinator, nurse midwife, breastfeeding coordinator	13 (35.1)
Total		37 (100)

It was very difficult to categorize administrators in different categories because of hospital variation in titles and roles. This categorization is intended to help situate perspectives but should not be considered for relational purposes.

One administrator interview was thrown out due to poor audio recording and therefore unusable data.

Data Analysis

Each interview was conducted over the telephone and was audio recorded, with verbal consent. The audio recordings were transcribed and imported into NVivo 8.²⁰ Each interview was read and coded by 1 of 2 trained coders, who coded for patterns based on the research questions. The questions were guided by the formative study and the literature. A second round of coding was conducted to identify patterns not previously identified. Descriptive codes were used for the coding process and were developed ex ante.^21,22 Examples of codes used in this study include EBF (exclusive breastfeeding) as PI (performance indicator), unknown data source, EBF rate share, share external, share internal, and EBF data limits.

Ten percent of the interviews were coded independently by a second research staff person.^21-23 The 2 coders met to discuss discrepancies in the coding and the need for additional codes. After 2 iterations of coding agreement discussions, coding definitions were revised.^22-25 The Kappa score for interrater agreement for all codes used in the analysis was 0.72.^25,26

Explanatory analysis was conducted to explore the themes identified in the preliminary study and the peer-reviewed literature on mandatory reporting.^24,27 Recurrent themes also were identified. These included themes that emerged from the text and those based on previous work.^22,27 Secondary analysis explored clusters of themes related to hospital use of mandatory reporting data and how the data may have influenced maternity care practice.^22,27 Differences in the use of breastfeeding measures and performance monitoring by hospitals based on hospital-specific breastfeeding measures were investigated. The results are clusters of themes or patterns of relationships identified through explanatory analysis.

Results

Participation

Potential hospital administrator and maternity staff with titles that indicated that they would be involved in maternity care were identified and contacted by e-mail and telephone. On average, it took 7.6 email or phone contact attempts per hospital to identify and recruit persons who met the staff criteria. All interviews were conducted by telephone for 10 to 30 minutes. The high-level administrator interviews were much shorter because those interviewed did not have detailed knowledge about how the mandatory breastfeeding rates influenced practice in maternity care units. These high-level administrators often referred us to mid-level (unit) administrators and staff for more information.

The stratified sampling strategy yielded a balanced study sample of 37 staff and administrators who represented 25 hospitals, of which 52% (n = 13) had breastfeeding policy scores below the median; 48% (n = 12) had exclusive breastfeeding measures below the median; and 52% (N = 13) had the number of births below the median. Within the sample of 25 hospitals, there was 1 staff member, generally a staff manager, interviewed for 14 hospitals. For the 10 hospitals sampled, 2 staff members each were interviewed; in 1 hospital, 3 staff members were interviewed. Table 2 provides a summary of the job titles of hospital staff and administrators who participated in the study. Analysis did not identify a relationship between the study question, hospital policy score, and hospital delivery rates.

Principal Findings

Dissemination of publicly reported infant feeding measures within hospitals varies across hospitals, as not all have maternity care unit staff members who are designated to review measures. Several staff members reported that this information is disseminated to maternity care committees, whereas other staff members reported that quality management units are responsible for reviewing the measures. This study did not identify a consistent top-down or bottom-up pattern of who receives, reviews, and monitors the publicly reported measures. Nevertheless, several themes emerged from the analysis, most of which focused on how the measures are used to monitor hospital performance.

Use of publicly reported breastfeeding measures

The dissemination of publicly reported breastfeeding measures raises awareness of breastfeeding as an important indicator of hospital performance. Thus, all interviewees were asked about the use of breastfeeding information as a performance measure. When administrators receive publicly reported performance measures, they are prompted to ask unit staff for explanations. A director of nursing explained, “Actually, the NYS data goes [sic] to the CEO, and then he comes to my office and [he] says, ‘Why do we do so poorly?’ We compare them [breastfeeding rates] to our consortium rates, and I explain why they are different.”

Administrators use the breastfeeding measures to raise awareness among staff. A nurse manager noted, “We use it to keep people [staff within hospital] aware that we are tracking it [breastfeeding data], that we are looking at it, that we care about whether an infant has been exclusively breastfed or not.” Hospital administrators use the breastfeeding measures and benchmarks from professional organizations to support efforts to improve hospital maternity care practices that support breastfeeding. According to a director of women’s services,

[Exclusive breastfeeding data] is [sic] very helpful to me because we want to improve our performance, and it shows us compared to all the rest of the hospitals, and it helps me make the case for the physicians and the nurses and the whole team that we have a ways to go, we can improve, and we need to improve and that this is the goal for Healthy People 2020, the American Academy of Pediatrics, the WHO [World Health Organization] and that we are falling below that, and so it helps make the case for why we need to make changes to improve our care.

Staff and administrators from nearly all hospitals reported the use of breastfeeding data for planning purposes. A vice president of patient care stated, “Several staff members and the lactation consultants review data and look at ways to improve our exclusive breastfeeding.” The data also are used for accountability, for performance monitoring. A breastfeeding coordinator explained, “We have an interdisciplinary breastfeeding committee that meets every 2 weeks, and we look at the stats monthly and see what’s going on. If it is dropping off, why, and discuss it.” Hospital staff reported that, when breastfeeding measures were monitored more frequently, staff members were able to respond more rapidly to negative changes in the data.

The use of breastfeeding measures data to advocate for hospital practice changes was noted more among hospitals with breastfeeding measures that were lower than the state average. According to a medical services director,

I have gone to our patient quality work group and presented on things when our CEO got that information [breastfeeding data from NYS DOH]. [We discussed] what we have done to get us where we were and what our plans were for the future.

Hospital administrators compared their hospital’s breastfeeding measures with those of hospitals in their consortium or peer hospitals. A nursing service administrator stated, “The NYS data is [sic] helpful because it gives us a comparison on how we compare to other hospitals.” In the case of 1 hospital that had a very high percentage of exclusive breastfeeding (in 2011, 72% of infants were exclusively breastfed), the comparison was leveraged for marketing purposes, as reported by the patient care director: “We put an article in the paper. . . advertising our [high] breastfeeding rates.”

Publicly reported measures as prompting regular performance monitoring

Prior to the 2010 mandate for public reporting of breastfeeding summary statistics on a yearly basis, there were no hospital-specific breastfeeding rates being calculated or disseminated by NYS DOH.²⁸ Most hospital records were paper records, and thus, it was time consuming and cumbersome to calculate breastfeeding measures. Hospitals also did not have access to the breastfeeding measures of other hospitals or the NYS average for comparison purposes. The dissemination of these breastfeeding measures, due to the mandate and the change from paper to electronic medical records, may have prompted some hospitals to collect infant feeding/breastfeeding data directly from their patients’ medical records. Most staff, especially those from hospitals that reported quality improvement efforts, reported that they regularly calculate infant feeding measures and tracking performance related to maternity care practices and breastfeeding support. A breastfeeding coordinator noted, “Initially, I report out my own data because I have that quicker, month to month.”

Confusion about Data Sources and Limitations of Publicly Reported Data

It became clear during the first few interviews that hospital staff were describing more current measures of infant breastfeeding, calculated directly from their patients’ medical records rather than the yearly measures reported by the NYS DOH. Thus, a question was added to the interview guide to clarify which information hospitals were using for monitoring performance. The use of the mandated breastfeeding data, compiled through the NYS Perinatal System, was more often regarded as a mechanism for comparison of performance to peer hospitals than used for internal monitoring. However, receiving publicly reported data annually may have prompted hospitals to regularly generate hospital-specific breastfeeding data for frequent internal performance monitoring.

Several hospital staff and administrators reported concern about the use of mandated publicly reported data for performance monitoring because the breastfeeding measures provided by NYS DOH had a delay in their release to the public, about 12 months after the reporting year ended. This is due to the largest region of the state’s submitting complete birth certificate data to NYS DOH up to 12 to 14 months after the end of the year. One interviewee, a nursing service administrator, explained, “While we look at the NYS data, it is [sic] essentially old data.” This delay limits the ability to monitor hospital performance in real time. Some staff and an administrator spoke about the concern that infant feeding information reported by their hospital on the birth certificates may not fully reflect the patient’s medical record. A chief nursing officer suggested that feeding status at discharge was a better measure of hospital performance related to the provision of breastfeeding support:

You probably have a higher number [rate of formula supplementation] on the birth certificate [data] than you would at discharge, and I know she gets the data on the birth certificates, too, but the reason that we have chosen to monitor [infant feeding status at] the discharge is because that truly shows if we have been able to do what we are supposed to do in facilitating and making sure they [mothers] are comfortable.

Staff could not articulate, however, why the data from mandatory reporting would be different; some did not know how the information was collected or who at their hospital reported the data. This reflects confusion about the type of infant feeding data being collected by NYS DOH. The data reported covers the entire period of time between birth and hospital discharge, or the first 5 days of life, whichever is sooner.¹⁹

External Influences on Data Use for Monitoring Breastfeeding

Breastfeeding education and quality improvement initiatives were being implemented in NYS during the same time as the implementation of the NYS public reporting mandate.^29,30 The use of breastfeeding measures in these initiatives to track and improve hospital breastfeeding performance further promoted utilization of these measures for quality improvement. Some hospitals were engaging in breastfeeding promotion and performance monitoring initiatives to become Baby-Friendly,^30,31 and 1 was already designated as such. According to a breastfeeding coordinator, “It’s [BF percentages] increasing because we are talking about Baby-Friendly, and everybody is becoming more aware. It is definitely increasing.” These hospitals reported being in the process of collecting additional data related to maternity care practices that support successful breastfeeding. These data collection processes are above and beyond the state-required collection of infant feeding data and public reporting of breastfeeding measures.

Conclusion

This qualitative study found a relationship between the mandatory reporting of breastfeeding measures and increased awareness of breastfeeding measures among hospital administrators, which may have prompted hospitals to use these measures to monitor hospital breastfeeding performance. Many hospital administrators and staff use the publicly reported measures to compare their hospital to peer institutions and to see their standing relative to the state average. Some hospitals in this study collect breastfeeding data directly from their medical records to calculate more frequent breastfeeding measures to be able to more regularly monitor their hospital’s breastfeeding statistics.

The heightened awareness documented in this study is consistent with findings reported in the earlier exploratory study that suggested that mandatory disclosure of hospital-specific breastfeeding measures may have provided performance feedback that led to more regular data collection for quality improvement.¹⁶ Not only does this study provide more depth and information, the rigorous sampling design and larger sample size allow for the findings to be better generalized to all NYS hospitals that provide maternity care. The current study is congruent with previous research on public disclosure of nationally standardized performance measures of acute myocardial infarction, heart disease, and pneumonia, which document a relationship between public reporting and performance improvement efforts at hospitals.^6,7,9,10,32

Additional education, training, and quality improvement efforts to improve breastfeeding support that occurred around the same time also may have encouraged individual hospitals to monitor infant feeding/breastfeeding measures. In 2009, when the mandate was announced, only 2 hospitals in NYS were Baby-Friendly-designated, and a third hospital became designated that year. Literature suggests that hospitals that comply with Baby-Friendly Hospital Initiatives have more extensive performance monitoring.³²

Research on data-driven performance monitoring suggests several key factors for effectiveness: It takes time for data to become valued and deemed credible, the source and timeliness of data are critical for perceived validity, and benchmarks improve the effectiveness of data feedback.^32-35 This study identified the limitation of a time delay for receiving summary statistics. Another potential limitation is that not all staff members have accepted these breastfeeding measures as valid. Some staff questioned the accuracy of breastfeeding/infant feeding data collected and reported by their hospital to the statewide perinatal data system, whereas others disagreed with the timeframe for assessing infant feeding. Both of these perceived limitations have been identified by studies of other clinical performance indicators,^2,6 including the quality and accuracy of birth certificate data.^36,37 Birth statistics are routinely used for performance monitoring, so it is important to ensure standardized processes for data collection and reporting to improve data quality.

Limitations

The study has some of the limitations of most qualitative research, but its ability to be generalized is strengthened by its rigorous sampling strategy. This study is retrospective with temporal bias, as it was conducted roughly 3 years after the implementation of the mandate. This study also did not involve interviews with patients; thus, it does not provide information about whether the public reporting of breastfeeding measures influenced prospective patients’ choice of hospital for their delivery. This is a common flaw among evaluations of mandates for public reporting of clinical indicators.^2,7,16

Summary

With the advent of electronic medical records, insurers and states are collecting more data. This will likely lead to increased use of publicly reported performance measures as a mechanism for improving health care quality. The provision of hospital-specific breastfeeding measures to hospitals appears to have influenced hospital staff and administrators to compare their hospital with others and to increase efforts to better support breastfeeding. This study was conducted within the first 4 years of passage and adoption of the NYS public reporting mandate. Organizational culture and staff behavior changes take time to show an effect on outcomes. Longitudinal research over a longer period of time that includes comparison states without such mandates may be needed to determine the full effect of public reporting mandates on hospital maternity care practices and breastfeeding measures.³² Future studies also should examine the effect of publicly reported breastfeeding measures on women’s selection of an obstetric provider and/or birthing hospital.

Footnotes

Appendix

Acknowledgements

The authors would like to thank Timothy Lytton and Deborah Gregg for their thoughtful contributions to the research process and early versions of the article. They appreciate the time that hospital staff and administrators spent completing the interview during their workday.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support for this project was provided by the Robert Wood Johnson Foundation’s Public Health Law Research program (Grant No. 12-069). These findings do not necessarily represent the views of the funders.

References

Davies

. Public release of performance data and quality improvement: internal responses to external data by US health care providers. Qual Health Care. 2001;10(2):104-110.

Fung

Graham

Weil

. Full Disclosure: The Perils and Promise of Transparency. New York, NY: Cambridge University Press; 2007.

Fung

Lim

Mattke

Damberg

Shekelle

. Systematic review: the evidence that publishing patient care performance data improves quality of care. Ann Intern Med. 2008;148(2):111-123.

The Joint Commission. Home page. https://www.jointcommission.org/. Accessed June 7, 2016.

Hibbard

Stockard

Tusler

. Does publicizing hospital performance stimulate quality improvement efforts? Health Affair. 2003;22(2):84-94.

Hafner

Williams

Koss

Schurtz

Schmaltz

Loeb

. The perceived impact of public reporting hospital performance data: interviews with hospital staff. Int J Qual Health Care. 2011;23(6):697-704.

Marshall

Shekelle

Leatherman

Brook

. The public release of performance data: what do we expect to gain? A review of the evidence. JAMA. 2000;283(14):1866-1874.

Werner

Konetzka

Stuart

Norton

Polsky

Park

. Impact of public reporting on quality of post-acute care. Health Serv Res. 2009;44(4):1169-1187.

Werner

Bradlow

. Public reporting on hospital process improvements is linked to better patient outcomes. Health Affair. 2010;29(7):1319-1324.

10.

Healthy People Maternal, Infant, and Child Health. HealthyPeople.gov. https://www.healthypeople.gov/2020/topics-objectives/topic/maternal-infant-and-child-health/objectives. Accessed May 23, 2016.

11.

New York State Department of Health. State health department issues call to action to promote breastfeeding in New York. http://www.health.ny.gov/press/releases/2009/2009-08-06_call_to_action_to_promote_breastfeeding_in_ny.htm. Published 2009. Accessed October 29, 2015.

12.

Justia. 2010 New York code. PBH—public health article 28 (2800-2822). Hospitals: 2803-J—Information for maternity patients. http://law.justia.com/codes/new-york/2010/pbh/article-28/2803-j/

13.

New York State Department of Health. Breastfeeding promotion program. https://www.health.ny.gov/community/pregnancy/breastfeeding/. Accessed July 31, 2015.

14.

Dennison

FitzPatrick

. On the frontline: New York State policies and initiatives to increase breastfeeding. Breastfeed Med. 2011;6(5):271-276.

15.

The Joint Commission. Core measures sets. http://www.jointcommission.org/core_measure_sets.aspx. Accessed January 2, 2016.

16.

New York State Department of Health. NYS prevention agenda dashboard—state level. https://apps.health.ny.gov/doh2/applinks/ebi/SASStoredProcess/guest?_program=/EBI/PHIG/apps/dashboard/pa_dashboard. Accessed May 2014.

17.

Lytton

Dennison

Nguyen

Jurkowski

. There is more to transparency than meets the eye: the impact of mandatory disclosure laws aimed at promoting breastfeeding. Am J Law Med. 2014;40:393-413.

18.

Dennison

Hawke

Ruberto

Gregg

. Implementing legislation to improve hospital support of breastfeeding, New York State, 2009-2013. Prev Chronic Dis. 2015;12:E122. doi:10.5888/pcd12.150121.

19.

Hawke

Dennison

Hisgen

. Improving hospital breastfeeding policies in New York State: development of the model hospital breastfeeding policy. Breastfeed Med. 2013;8(1):3-7.

20.

QSR International. Home page. http://www.qsrinternational.com/products_nvivo.aspx. Accessed February 22, 2014.

21.

Gläser

Laudel

. Life with and without coding: two methods for early-stage data analysis in qualitative research aiming at causal explanations. Qual Soc Res. 2013;14(2). http://www.qualitative-research.net/index.php/fqs/article/view/1886/3528#g33. Accessed May 18, 2014.

22.

Saldana

. The Coding Manual for Qualitative Researchers. Thousand Oaks, CA: SAGE Publications; 2012.

23.

Malterud

. Qualitative research: standards, challenges, and guidelines. The Lancet. 2001;358(9280):483-488.

24.

Pope

Ziebland

Mays

. Qualitative research in health care: analyzing qualitative data. BMJ. 2000;320(7227):114-116.

25.

Morse

Barrett

Mayan

Olson

Spiers

. Verification strategies for establishing reliability and validity in qualitative research. Int J Qual Methods. 2008;1(2):13-22.

26.

Hallgren

. Computing inter-rater reliability for observational data: an overview and tutorial. Tutor Quant Methods Psychol. 2012;8(1):23-34.

27.

Ritchie

Lewis

Nicholls

Ormston

, eds. Qualitative Research Practice: A Guide for Social Science Students and Researchers. Thousand Oaks, CA: SAGE Publications; 2013.

28.

Dennison

Nguyen

Fan

Jurkowski

. Impact of state legislation on hospital breastfeeding support in New York. J Pub Health Manage Pract. In press.

29.

Justia. 2012 New York consolidated laws. PBH—public health article 41—vital statistics, title 3—(4130-4138-D). Registration of births: 4130. http://law.justia.com/codes/new-york/2012/pbh/article-41/title-3/4130.

30.

Hisgen

Dennison

FitzPatrick

Waniewski

. Improving maternity care practices in New York: answering the surgeon general’s call to action. Breastfeed Med. 2012;7(5):337-342.

31.

Baby-Friendly USA. D2 development. https://www.babyfriendlyusa.org/get-started/d2-development. Published 2012.

32.

Merewood

Mehta

Chamberlain

Philipp

Bauchner

. Breastfeeding rates in US Baby-Friendly hospitals: results of a national survey. Pediatr. 2005;116(3):628-634.

33.

Strauch

Rohrer

Refaat

. Increased hospital documentation requirements may not increase breastfeeding among first-time mothers. J Eval Clin Pract. 2016;22(2):194-199. doi:10.1111/jep.12453.

34.

Smith

Wright

Quorum

Lamb

. Public reporting helped drive quality improvement in outpatient diabetes care among Wisconsin physician groups. Health Affair. 2012;31(3):570-577.

35.

Bradley

Holmboe

Mattera

Roumanis

Radford

Krumholz

. Data feedback efforts in quality improvement: lessons learned from US hospitals. Qual Saf Health Care. 2004;13(1):26-31.

36.

Bodnar

Abrams

Bertolet

. Validity of birth certificate-derived maternal weight data. Paediatr Perinat Epidemiol. 2014;28(3):203-212.

37.

Clayton

Sappenfield

Gulitz

. The Florida investigation of primary late preterm and cesarean delivery: the accuracy of the birth certificate and hospital discharge records. Matern Child Health J. 2013;17(5):869-878.