Integration of the Opportunity‐Ability‐Motivation behavior change framework into a coaching‐based WHO Safe Childbirth Checklist program in India

Abstract Objective To evaluate whether integration of the Opportunity‐Ability‐Motivation plus Supplies (OAMS) framework into coaching improved the delivery of essential birth practices in a low‐resource setting. Methods This prospective mixed‐methods study used routine coaching visit data obtained from the first eight intervention facilities of the BetterBirth trial in Uttar Pradesh, India, between December 19, 2014, and October 21, 2015. The 8‐month intervention was peer coaching that integrated the OAMS framework to support uptake of the WHO Safe Childbirth Checklist. Descriptive statistics were used to measure nonadherence to essential birth practices. The frequency and accuracy of coaches’ coding of barriers and the appropriateness of chosen resolution strategies to measure feasibility, acceptability, and fidelity of using OAMS, were assessed. Results Coaches observed 666 deliveries, including 12 602 practices. Overall, essential practice nonadherence decreased from 15.6% (262/1675 practices observed) to 4.5% (4/88 practices) (P<0.001). Of the 1048 barriers identified, opportunity (556 [53.1%]) and motivation (287 [27.4%]) were the most frequently reported categories; the frequency of both decreased over time (P=0.003 and P<0.001, respectively). The coaches appropriately categorized 930 (99.8%) of 932 barriers and provided an appropriate strategy for 800 (85.8%). The commonest reason for unaddressed barriers was lack of coaching opportunities. Conclusion Successful integration of OAMS framework into delivery attendant coaching enabled coaches to rapidly diagnose barriers to practice adherence and develop responsive strategies. ClinicalTrials.gov NCT2148952 (WHO Universal Trial Number: U11111‐1315‐647).


| INTRODUCTION
Childbirth-related mortality remains a major cause of suffering, globally, with 350 000 maternal and 3.1 million neonatal deaths annually. 1,2 Essential birth practices (EBPs) reduce maternal and neonatal morbidity and mortality; however, care providers do not employ these practices widely and consistently. 3 Although policy efforts have improved women's access to facility-based delivery, poor quality of care remains problematic in many resource-constrained settings. 4,5 To address the quality gap in maternal and neonatal care during facility-based delivery, WHO and other stakeholders created the WHO Safe Childbirth Checklist (SCC), a 28-item tool consisting of EBPs associated with improved maternal, fetal, and neonatal outcomes. [6][7][8] The SCC is organized to drive change at four critical moments (or pause points): on admission, before delivery, within 1 hour after delivery, and before discharge. Initial studies have demonstrated an association between SCC use and improved adherence to EBPs. [9][10][11] Evidence from quality-improvement initiatives has shown the importance of integrated interventions to change both provider behavior and the healthcare system. When done well, supportive supervision, clinical mentorship, and coaching can be effective in changing provider behavior in a variety of settings, increasing the rate of skill transfer or adoption and generating more sustained improvement in performance than training alone. [12][13][14] To maximize the impact of the SCC, a coaching-based implementation program (the BetterBirth program 15 ) was designed, and-based on behavior change literature from multiple fields-the Opportunity-Ability-Motivation (OAM) framework was integrated into this coaching strategy. 16 The OAM framework, initially developed for understanding individual consumer behavior, 17 postulates that barriers to and facilitators of behavior change operate within three domains: opportunity, ability, and motivation. Researchers in a number of fields including public health have adopted the OAM framework. 17,18 Given the prevalence of challenges associated with supplies and equipment in many resource-constrained settings, in the present study the OAM framework was adapted by dividing opportunity into supply-related and other opportunity-related barriers (Opportunity-Ability-Motivation-

Supplies [OAMS]).
During the BetterBirth trial, routine coach-reported data were collected to study whether coaches correctly and effectively applied the OAMS framework in diagnosing and addressing barriers to EBP performance among delivery attendants. The present study used data obtained from the first eight intervention facilities to evaluate whether integration of the OAMS framework into the BetterBirth coaching approach was feasible and acceptable; this was measured by the uptake and correct application by the coaches to rapidly diagnose barriers to practice adherence and develop responsive strategies.

| MATERIALS AND METHODS
The present study was a prospective mixed-methods study leveraging data collected by coaches as part of their work in the BetterBirth trial-a cluster-randomized controlled trial that was designed to test the effectiveness of a coaching-based implementation of the WHO SCC in Uttar Pradesh, India's most populous state. Uttar Pradesh has persistently high maternal and neonatal mortality rates. 19 The present study included all data collected by coaches between December 19, 2014, andOctober 21, 2015, in the first eight intervention sites. The facilities included in the BetterBirth trial provided labor and delivery services 24 hours a day on 7 days each week, had a minimum of 1000 deliveries per year, and employed at least three delivery attendants. 15 The study protocol was approved by the ethics review committees of In late February 2015, 2 months after initiation of the BetterBirth trial at the first study site, the coaches began using a coach support tool to record a brief narrative description of new barriers prioritized during a visit and any unresolved barriers that had been prioritized in earlier visits. The coaches recorded an OAMS category for each barrier and at least one specific coaching strategy they applied to resolve the barrier. These strategies could include a direct intervention, escalating to management to address system barriers such as facility stock-outs, deferring to the next visit if the delivery attendant was too busy or no patient was available for observing any behavior change, continuing interventions into the next coaching visit if no change was seen, or abandoning if the delivery attendant would no longer be working in labor and delivery. The coach support tool could have multiple entries if the coach prioritized more than one EBP challenge to address at a given visit or a challenge persisted over time. The quotes presented in this paper are written verbatim from the coaches' notes with abbreviations explained where needed.
In a quantitative analysis, eligible OTIS data were used to calculate the rates of nonadherence to EBPs, the application of the framework activities. The coaches collected no patient identifiers and all delivery attendant identifiers were removed before analysis for the present paper. The patients enrolled in the trial for follow-up were not the same patients who were observed during coaching.

| RESULTS
Across the eight intervention facilities, 46 delivery attendants received coaching from a median of 2.5 coaches per site (10 individual coaches).
During the 8 months, the coaches observed 666 deliveries at one or more pause points, documenting 12 602 EBPs across 1352 SCC pause points (see Table S1 for more details). Overall, the nonadherence rate for the EBPs documented in OTIS was 7.9% (997/12 602), with variation in nonadherence by pause point: at facility admission, 10.1% (268/2664) of EBPs were missed; 10.8% (414/3848) were missed just prior to delivery; 3.4% (43/1270) were missed at the time of delivery, 6.4% (196/3040) were missed within the first hour postpartum, and 7.1% (127/1780) were missed at discharge (Fig. S1). The rates of nonadherence to specific EBPs ranged from 0.0% (evaluation of the neonate's breathing) to higher rates such as 39.5% (taking the mother's temperature before delivery) (data not shown).
For the 997 EBPs that were not completed, the coaches showed high acceptability of the framework and recorded 1048 barriers.
Individual-level barriers (motivation and ability) and system-level barriers (supplies and opportunity) accounted for 32.7% (n=343) and According to OTIS data, the nonadherence rate decreased from 15.6% (262/1675) in the beginning of coaching to 4.5% (4/88) in the final 2 weeks of coaching (P<0.001) (Fig. 1). The rate of EBP nonadherence attributable to opportunity barriers decreased by an average of 3.3% relative to each previous 2-week period (P=0.003), and the rate of EBP nonadherence attributable to motivation barriers decreased by an average of 9.1% (P<0.001). The rates of EBP nonadherence attributable to ability and supplies barriers did not change significantly over time.
The coach support tool captured qualitative coaching data on 29 delivery attendants across the eight sites. From a total of 955 reports of EBP nonadherence, four were excluded because the barrier was resolved prior to the next visit, 14 because no mother was present at the next visit (and therefore a previously identified barrier could not be evaluated), and five because written descriptions of the barrier were missing. For the remaining 932 EBPs not performed, 130 (13.9%) had an opportunity barrier identified, 578 (62.0%) an ability barrier, 308 (33.0%) a motivation barrier, and 97 (10.4%) a supplies barrier (Table 1). Nearly all (930 [99.8%]) the barriers not performed had been categorized appropriately by the coaches, reflecting high feasibility and fidelity. For 800 (85.8%) of the 932 EBPs not performed, the coaches implemented strategies that corresponded to at least one of the barrier categories recorded. Thirty cases involved strategies that addressed two barrier categories. If the barrier was not addressed, the coaches most commonly cited a lack of coaching opportunity as the reason ( Table 2).
The coaches recorded many different strategies to help delivery attendants resolve barriers to performing EBPs (Table 3). These strategies ranged from telling a story to motivate an individual delivery attendant to involving a facility administrator to address a supply stock-out (Box 1). By way of example, in one busy facility, delivery attendants explained that they lacked sufficient time to prepare a delivery tray for each mother. The coaches suggested assigning a worker who was not a delivery attendant to prepare the trays. Implementation of this strategy ensured the completion of EBPs related to delivery supplies and gave delivery attendants the opportunity to focus on other EBPs.
Other examples included the coach using the SCC and other motivation techniques such as storytelling to encourage delivery attendants to integrate the EBPs to meet national standards and save lives. When delivery attendants successfully overcame a barrier, the coaches also F I G U R E 1 Rate of nonadherence to essential birth practices stratified by barrier type during an 8-month coaching period (based on coach observation data collected with the Observation Tool to Inform Support).
T A B L E 1 Classification of, and response to, barriers among 932 non-completed essential birth practices documented in the coach support tool.

Coach-coded barrier Frequency a Example
Opportunity 130 (13.9) "Because there is a lot of work, BA says is not able to take BP at PP4." System level (opportunity and/or supply) 223 (23.9) "No water supply in labor room. BA said when water supply will be fixed she will be able to wash hand."

| DISCUSSION
The OAMS framework was a feasible and acceptable structure for the coaching-based implementation of the WHO SCC. The coaches were able to categorize barriers to EBP adherence using the framework with high fidelity and develop coaching strategies that appropriately reflected and addressed these underlying barriers. This coaching approach, incorporating the OAMS framework, was associated with an increase in adherence to the observed EBPs over 8 months, although there was no change in morbidity and mortality. 21,22 Using the OAMS framework also enabled tracking change in In conclusion, integration of the OAMS behavior change framework into the coaching-based implementation of the WHO SCC was acceptable, feasible and facilitated coaches' correct categorization of barriers and their development of appropriately responsive strategies to address these barriers. The use of OAMS-informed coaching was associated with an increase in adherence to EBPs. 21 By contrast, supervision-as currently delivered in some settings-is not always associated with higher quality of care. 23 The present findings support the potential for coaching informed by the OAMS framework in conjunction with the WHO SCC to inspire behavior change in front-line providers and encourage them to use the skills they have gained through pre-and in-service training. These findings make this framework-based coaching an important tool to consider for programs that aim to strengthen the quality of care through the performance of evidence-based practices.

AUTHOR CONTRIBUTIONS
LRH, MK, and KEAS contributed to the design and planning of the study, data analysis, and writing and revising the manuscript.
JM contributed to conducting the study and data collection, data T A B L E 2 Response to barriers among 932 non-completed essential birth practices documented in the coach support tool. "It is not in the BA's habits to take BP yet so coaching is still necessary." "BA also does not know the importance of BP." "BA took BP." analysis, and writing and revising the manuscript. NK contributed to the design and planning of the study, conducting the study and data collection, data analysis, and revising the manuscript. TK contributed to conducting the study and data collection, and revising the manuscript. NH contributed to data analysis and revising the manuscript. DR, PM, MMD, NS, and VK contributed to conducting the study, data collection, and revising the manuscript. RF contributed to the design and planning of the study, conducting the study and data collection, and revising the manuscript. AAG contributed to the design and planning of the study, and revising the manuscript.

ACKNOWLEDGMENTS
The present trial was funded by the Bill & Melinda Gates Foundation, which reviewed the study design and the sample size calculations.
The funders did not have any input on the collection, management, analysis, or interpretation of the data. Further, they did not have any authority over the writing of the reports or the decision to submit the findings for publication. The authors wish to specifically thank Claire Stokes and Manisha Tripathi, Populations Services International, for their contributions on framework development and training.

CONFLICTS OF INTEREST
AAG has received royalties for books and essays, including publications on improving the quality and delivery of health care using check- lists. The authors have no other conflicts of interest.

Supporting Information
Additional supporting information may be found online in the Supporting Information section at the end of the article.