Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Effectiveness of a WHO Safe Childbirth Checklist Coaching-based intervention on the availability of Essential Birth Supplies in Uttar Pradesh, India

Effectiveness of a WHO Safe Childbirth Checklist Coaching-based intervention on the availability... Objective: Evaluate the impact of a World Health Organization Safe Childbirth Checklist coaching- based intervention (BetterBirth Program) on availability and procurement of essential childbirth- related supplies. Design: Matched pair, cluster-randomized controlled trial. Setting: Uttar Pradesh, India. Participants: 120 government-sector health facilities (60 interventions, 60 controls). Supply- availability surveys were conducted quarterly in all sites. Coaches collected supply procurement sources from intervention sites. Interventions: Coaching targeting implementation of Checklist with data feedback and action planning. Main Outcome Measures: Mean supply availability by study arm; change in procurement sources for intervention sites. Results: At baseline, 6 and 12 months, the intervention sites had a mean of 20.9 (95% confidence interval (CI): 20.2–21.5); 22.4 (95% CI: 21.8–22.9) and 22.1 (95% CI:21.4–22.8) items, respectively. Control sites had 20.8 (95% CI: 20.3–21.3); 20.9 (95% CI: 20.3–21.5) and 21.7 (95% CI: 20.8–22.6) items at the same time-points. There was a small but statistically significant higher availability in interven- tion sites at 6 months (difference-in-difference (DID) = 1.43, P < 0.001), which was not seen by 12 months (DID = 0.37, P = 0.53). Greater difference between intervention and control sites starting in the bottom quartile of supply availability was seen at 6 months (DID = 4.0, P = 0.0002), with no © The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. 769 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 770 Maisonneuve et al. significant difference by 12 months (DID = 1.5, P = 0.154). No change was seen in procurement sources with ~5% procured by patients with some rates as high as 29% (oxytocin). Conclusions: Implementation of the BetterBirth Program, incorporating supply availability, resulted in modest improvements with catch-up by control facilities by 12 months. Supply-chain coaching may be most beneficial in sites starting with lower supply availability. Efforts are needed to reduce reliance on patient-funding for some critical medications. Trial Registration: ClinicalTrials.gov #NCT02148952; Universal Trial Number: U1111-1131–5647 Key words: supply availability, maternal and newborn health, WHO Safe Childbirth Checklist, quality improvement Introduction conducted from October 2014 to January 2017 (Appendix A). The trial Reducing childbirth-related maternal and neonatal morbidity and included primary, community and first-referral health centers (60 inter- mortality requires improving patients’ access to high quality maternity vention and 60 control facilities). Detailed descriptions of the BetterBirth care [1–3]. Although rates of facility-based deliveries have increased, Program and trial methodology are described elsewhere [13, 21, 22]. this shift has not consistently translated into improved outcomes [4–7]. In 2012, the Neonatal Mortality Rate in Uttar Pradesh, India BetterBirth Program remained unacceptably high at 49 deaths per 1000 live births [8, 9]. Core to the BetterBirth intervention was a coaching team who worked In an effort to improve the quality and outcomes of maternity care to address non-adherence to essential birth practices, including supply- and in collaboration with the Government of India and State of Uttar related barriers, at multiple healthcare system levels. At each of the 60 Pradesh, a randomized controlled trial was undertaken to measure the intervention facilities, coaches (nurses) conducted 43 day-long visits impact of the BetterBirth Program, a coaching-based implementation over 8 months (twice per week tapering to monthly) during which of the World Health Organization (WHO) Safe Childbirth Checklist they worked with birth attendants to deliver Checklist-identified prac- [10–13]. The Checklist is a job aid designed to help birth attendants tices. Coaches provided feedback to birth attendants about their adhere to 28 essential birth practices known to save lives [14]. adherence, engaged them in identifying barriers and assisted with Adherence to the Checklist-identified practices requires 28 supplies action planning to resolve issues. Coach Team Leaders (physicians or (Table 1). However in India, there are significant gaps in the availabil- public health professionals) accompanied coaches on alternating visits, ity of medicines and equipment for maternity care in government- working with facility leadership to review data on Checklist adher- sector facilities [15–17]. For example, a survey in Uttar Pradesh found ence. They then worked with leadership to identify and resolve that only 53.8% of first-line referral facilities had injectable magne- facility-level barriers to Checklist adoption, including improving sium sulfate available [17]. When supplies are unavailable in facilities, supply availability. They also coached the Childbirth Quality patients either buy them or forego treatment. As a result, the unavail- Coordinator, a facility-based champion responsible for motivating ability of supplies is a serious barrier to the provision of high quality staff utilization of the Checklist, identifying and resolving supply issues care in facilities and, consequentially, a positive outcome for mothers and supporting the facility’s commitment to Checklist-use post-trial. and babies of all socioeconomic backgrounds [18]. In addition to one-on-one coaching, Coaches and Coach Team To avoid creating a non-sustainable supply source in the context of Leaders held data-sharing meetings at facilities fortnightly to feed- the study, the BetterBirth Program did not provide any supplies to facil- back data on Checklist adherence and supply availability and sup- ities [13]. Instead, the theory of change was that coaching at the point port action planning. To ensure district-level support, Coach Team of care, as well as at facility and district levels would improve supply Leaders facilitated bi-monthly progress meetings with district health availability by promoting behavior and system change through (a) officials and facility leadership to review adherence and supply- improved knowledge of the essential supplies for childbirth and advo- related data and help identify plans needed at the district level to cacy at the frontline to improve availability; (b) routine monitoring and improve childbirth care, including supply availability. The data driven communication of supply gaps to facility and district lea- BetterBirth Program did not provide supplies (except paper copies ders; accompanied by (c) coaching on action plans for solutions which and posters of the Checklist during the trial) or financial support. leveraged existing supply chains and facility and district resources. Results of the impact BetterBirth had on essential birth practice Although there have been efforts to identify interventions that uptake and health outcomes are described elsewhere [22, 23]. improve health systems’ supply-chains, further research is needed to understand how supply availability through strengthening existing systems can be effectively integrated into coaching-based quality Ethics improvement efforts [15, 19, 20]. Here, we describe the BetterBirth The study protocol received approval by the following ethical review Program’s impact on essential birth supply availability and the pro- boards at the Community Empowerment Lab, Jawaharlal Nehru curement source of available supplies. Medical College, Harvard T.H. Chan School of Public Health, Population Service International, the WHO and the Indian Council of Medical Research. Each facility and birth attendant formally Methods agreed to participate in the BetterBirth Program at the beginning of Study design and site characteristics the study. Coaches accompanied birth attendants during their work- The BetterBirth Trial was a matched pair, cluster-randomized controlled shift and documented practices during patient-care activities at the trial across 120 government health facilities in Uttar Pradesh, India facility. Coaches collected no patient identifiers. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 771 Table 1 Supplies associated with WHO Safe Childbirth Checklist essential birth practices, location of evaluated items, and whether their source of procurement was evaluated in the BetterBirth trial in Uttar Pradesh, India Components of checklist Associated essential birth Location evaluated for availability (1- admission, 2- duty room, Evaluated for supply score (28 items total) practice on the WHO Safe 3- labor and delivery (L&D) room, 4- post-partum room or procurement (N = 23) Childbirth Checklist anywhere in facility (Facility)) Pause Point 1 1. Thermometer Mother’s temperature All four locations X 2. Blood pressure cuff Mother’s blood pressure All four locations X 3. Stethoscope X 4. Fetoscope or Doppler Fetal heart rate Admission room, Duty room, L&D room X 5. Partograph Partograph started Admission room, Duty room, L&D room 6. Antibiotics (mother) Antibiotics for the mother Facility X 7. Urine dip sticks Magnesium sulfate for the Facility X 8. Magnesium sulfate mother All four locations X 9. HIV testing kit HIV status of the mother Facility 10. Nevirapine (mother) Administer nevirapine as Facility needed Pause Point 2 11. Clean gloves Confirm essential supplies for Admission room, Duty room, L&D room X mother are at bedside 12. [(Soap and Admission room, Duty room, L&D room X 12. Clean water) or Admission room, Duty room, L&D room X 12. (Alcohol rub)] Admission room, Duty room, L&D room 13. Sterile needle/syringe All four locations X 14. Oxytocin L&D room X 15. Clean pads Admission room, Duty room, L&D room X 16. Clean towel Confirm essential supplies for L&D room X 17. Clean blade/scissor baby are at beside L&D room X 18. Cord tie/clamp L&D room X 19. Mucus extractor/suction L&D room X 20. Bag-and-mask L&D room X 21. Vitamin K Facility X Pause Point 3 1. Thermometer Baby’s temperature All four locations Above 3. Stethoscope Baby’s respiratory rate All four locations Above 22. Baby scale Baby’s weight L&D room X 23. Intravenous fluid bag Start IV fluids (if mother Facility bleeding) 24. Antibiotics (baby) Antibiotics for the baby Facility X 25. Baby warmer Special care/monitoring for L&D room baby 26. Nevirapine (baby) Administer nevirapine as Facility needed Pause Point 4 27. BCG vaccine Administer vaccines Facility X 28. Polio vaccine Facility X BCG, Bacillus Calmette–Guérin Vaccine; L&D, Labor and Delivery. Item is deemed available if it is present in a designated area and functional (or not expired, as relevant). While certain items (e.g. stethoscope or thermometer) can be available within any room in the labor ward complex (e.g. availability in either admission or duty room, or labor and delivery room, or post-partum ward), other items were expected to be available in a designated area (e.g. neonatal bag-and-mask available in the labor and delivery room). Procurement is how essential medicines and medical supplies are sourced. The source of supply was defined by four categories; Official: (1) district or state government defined system; (2) from facility funds; Unofficial: (3) Patient or family: brought from home or purchased or (4) Any other means, e.g. staff purchased or donation. Denotes a checklist practice repeated across multiple Pause Points. For simplicity, these are listed in this table only at the first Pause Point at which they are performed. designated areas, it was coded as available. Coaches were not pre- Data sources sent on days when surveys were conducted and did not have access Facility essential birth supply availability survey to data. Data collectors conducted facility surveys at baseline and Study staff conducted the supply-availability survey (Appendix D) at quarterly over the study. Baseline for intervention sites was just all sites to measure availability of the 28 essential birth supplies prior to the beginning of the intervention. Baseline for control sites (Table 1). During unannounced visits, data collectors observed facil- was just prior to the first study-related visits, which occurred ~2 ity areas where supplies were supposed to be stocked. If an item was months after intervention start at their matched pair [22]. Surveys observed as functional, not expired, and in at least one of the Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 772 Maisonneuve et al. were collected from the 60 pairs at baseline through 6 months and baseline): 14 intervention sites (23%) and 14 control sites (23%). from 58 pairs at 9 and 12 months (two facilities closed and moved An additional model was fit that included a control for baseline sup- after completing their third survey at 6 months, resulting in exclu- ply availability interacted with study arm. The term would be sig- sion of these sites and their pairs from the 12 month analysis). For nificant if the intervention had a different effect among sites with simplicity, only data from baseline, 6 and 12 months are reported; fewer baseline supplies as compared to those with more. 3- and 9-month data are available upon request. Coach supply source survey Coach supply source survey The proportion of available supplies purchased by an unofficial pro- During facility visits, Coach Team Leaders conducted a supply curement method (defined as being purchased by the patient/family source survey (Appendix E) with the facility leader or Childbirth or ‘other source’) versus the official supply chain (procured by facil- Quality Coordinator to measure the availability and procurement ity or district) over time was examined. In cases where a supply was source of 23 of the 28 checklist supplies (Table 1). If an item was procured from both an unofficial and an official method, it was available, Coach Team Leaders would discuss with facility represen- coded as procured through the unofficial method. For this analysis, tatives how the supply was procured: (1) through the district or state coach supply source data was grouped into 1-month time periods government supply-system, (2) from facility funds, (3) from patient (except for the final months 8 and 9 which were grouped together) or family (brought from home or purchased) or (4) from any other and the mean proportion of all available items procured by each means (e.g. facility staff purchased or donation). The first survey source was calculated. was conducted on the day after the BetterBirth Program launched and was repeated approximately every 2 weeks and then monthly, Results as Coach Team Leader visits decreased in frequency. After their vis- it, Coach Team Leaders entered data into a mobile phone-based Facility essential birth supply availability CommCare application (Dimagi, Cambridge, MA), which visualized A total of 596 completed facility supply-availability surveys were avail- data over time as a heat map (Appendix F) to be shared with facility able for analysis. At baseline, the intervention sites had a mean of 20.9 staff and leadership to inform action planning [13]. items (95% CI: 20.2–21.5), which improved to 22.4 (95% CI: 21.8–22.9) at 6 months and 22.1 (95% CI: 21.4–22.8) at 12 months (Table 2). The control sites started with a mean of 20.8 items (95% CI: Analysis of surveys 20.3–21.3), which increased to 20.9 (95% CI: 20.3–21.5)at6months All statistical analyses were conducted using SAS v9.4 (SAS and 21.7 (95% CI: 20.8–22.6) at 12 months. There was a small but Institute, Cary, NC). statistically significant better improvement in the intervention versus control arm at 6 months (DID = 1.43 supplies, P < 0.001), which was Facility essential birth supply-availability survey no longer significant by 12 months (DID = 0.37 supplies, P = 0.53). The primary outcome was the difference in overall availability of The facilities with supply availability in the lowest quartile (<20 essential birth supplies after 6 months of coaching. Six-months was supplies) at baseline were spread across all five study-defined chosen because maximum impact was expected at this time while regions. While the average supply-availability score for intervention coaching was still underway and a decay was seen in adherence to and control sites in this subgroup increased over time, intervention Checklist practices in intervention sites at 12 months (4 months after sites had significantly better improvement by 6 months (interven- the intervention’s end) [22]. tion: 17.6–22.0 for intervention and 18.3–18.7 for controls at base- Essential birth supply availability was calculated as the count of line and 6 months, respectively, DID: 4.00, P = 0.0002) (Fig. 1 and the 28 Essential Birth Supplies available during a survey administra- Table 2). The difference in improvement between intervention and tion (Table 1). The mean percentage of sites with each supply avail- control was no longer statistically significant by 12 months (DID: able over time was calculated, in addition to the mean percentage of 1.51, P = 0.154). Modeling results found that the interaction term sites with all four medications considered critical to reducing mater- between baseline supply level and study arm was statistically signifi- nal and neonatal harm (oxytocin, magnesium sulfate, Vitamin K cant (P < 0.001, data not shown), confirming that the impact of the and antibiotics for mothers or babies). To avoid inflating the intervention was stronger among sites with lower baseline supplies. analysis-wide α (type I) error, we did not conduct significance tests Individual item supply availability was variable across sites. In on these differences. the analysis of all sites, 18 items were highly available (≥80% of Within each matched pair, facility surveys could occur up to 2 facilities) regardless of study arm at baseline and remained above months apart due to differences in baseline. Thus, models were con- that threshold throughout the study; five items had mid-range avail- structed to estimate differences in supply levels across study arms at ability (50–79% of sites) throughout the study; and five items (par- baseline, 6 and 12 months. Logistic regression models were fit using tograph, nevirapine for baby, nevirapine for mother, Doppler or the counting method to predict the number of 28 items available, fetoscope and Vitamin K) had low availability (presence in <50% of adjusting for matching and repeated measures over time [24]. facilities) at baseline (Table 3). While availability of Vitamin K and Models included nonlinear terms for time and an interaction fetoscope/Doppler increased to mid-range in intervention but not between study arm and time. From these models, the difference-in- control sites at month 6, the availability of partograph and nevira- difference (DID) was calculated for the number of supplies available pine for mother and baby remained low in both arms (Table 3). For from baseline to 6 months and, as a secondary outcome, from base- the bottom quartile intervention sites, eight items moved from mid- line to 12 months, with tests for statistical significance at α = 0.05. level availability at baseline to high availability at 6 months To explore facilities starting with fewer essential birth supplies, a (Table 4). In contrast to the overall intervention sites, in the bottom sub-analysis was conducted of facilities which fell into the bottom quartile intervention sites, no increase from the low availability cat- quartile of baseline supply availability (<20 of 28 supplies at egory was seen for the fetoscope/Doppler. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 773 Figure 1 Mean number of Essential Birth Supplies (N = 28) over time by study arm for sites starting at the bottom quartile of supply availability (<20 of the 28 supplies at baseline). Less than one-fifth of all sites had all four of the critical medica- tions (oxytocin, magnesium sulfate, Vitamin K, antibiotics) available at baseline (15% and 16.7% in intervention and control arms, respectively) with availability improving at 6 months (30.0% and 20.0% in intervention and control arms, respectively) and 12 months (41.4% and 22.4% in intervention and control arms, respectively; Table 3). Coach supply source survey Over the course of the study, Coach Team Leaders completed 964 supply source surveys across the 60 intervention sites. Each facility was expected to have a minimum of 14 surveys completed; Coach Team Leaders completed more at their discretion (mean 16, range 9–23). As expected due to Coach Team Leader visits decreasing in frequency over the intervention, the total number of surveys com- pleted per month declined from the first month of the intervention (N = 152) to the final months (N = 56). In the first month of the intervention, facilities procured 95% of available supplies from offi- cial sources (94% district, 1% facility) and 5% from unofficial sources (4% from patients and 1% from ‘other’ sources). No signifi- cant change was seen in the percentage of supplies sourced from official or unofficial sources over time (Appendix C). The top four supplies available due to purchase by patients were injectable oxyto- cin (29.1% purchased by patients), a sterile blade (28.4%), soap (23.1%) and Vitamin K (15.4%). Discussion The BetterBirth Program integrated a focus on supply readiness and use into coaching to leverage existing supply chains and improve availability at the facility and point of care. On average, facilities started with relatively high supply levels. The intervention resulted in a modest but statistically significant higher mean overall availabil- ity of the required 28 supplies at 6 months. This difference was largely driven by increased availability of 2 supplies: fetal heart rate monitors and vitamin K, which increased from availability in <50% of facilities at baseline to between 50–79% of facilities at 6 months. However, this difference was no longer seen at 12 months, in part due to no further improvement in intervention sites along with Table 2 Mean number of the 28 essential birth supplies (with 95% CI) by arm over time in (1) all sites and (2) in facilities starting in the bottom quartile of supply availability (<20 supplies at baseline) 0 Months (baseline) 6 Months 12 Months Difference-in-difference Intervention Control Intervention Control Intervention Control 0–6 Months 0–12 Months (N = 60) (N = 60) (N = 60) (N = 60) (N = 58) (N = 58) (1) All sites 20.9 (20.2–21.5) 20.8 (20.3–21.3) 22.4 (21.8–22.9) 20.9 (20.3–21.5) 22.1 (21.4–22.8) 21.7 (20.8–22.6) 1.43; P < 0.001 0.37; P = 0.53 0 Months (baseline) 6 Months 12 Months Difference-in-difference Intervention Control Intervention Control Intervention Control 0–6 Months 0–12 Months (N = 14) (N = 14) (N = 14) (N = 14) (N = 14) (N = 14) (2) Bottom 17.6 (16.5–18.6) 18.3 (17.4–19.2) 22.0 (21.0–23.1) 18.7 (17.6–19.9) 21.8 (20.6–23.1) 21.0 (19.6–22.5) 4.00 P = 0.0002 1.51 P = 0.154 quartile sites Note: At baseline and 6 months, all 60 intervention sites are included. At month 12, only 58 sites are included because two matched pairs were excluded from the study (due to closure) and month 12 surveys were not conducted. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 774 Maisonneuve et al. Table 3 Average availability of each of the 28 Essential Birth Supplies on the WHO Safe Childbirth Checklist across intervention and control sites over time Average percentage of sites with each item available over time Item 0 Months (baseline) 6 Months 12 Months Category based on change in % of Intervention sites with item available from baseline to 6 months Intervention Control Intervention Control Intervention Control n (%) n (%) n (%) n (%) n (%) n (%) (N = 60) (N = 60) (N = 60) (N = 60) (N = 58) (N = 58) Suction machine/mucus extractor 60 (100%) 60 (100%) 60 (100%) 59 (98.3%) 58 (100%) 57 (98.3%) High baseline availability across intervention Neonatal bag-and-mask 52 (86.7%) 55 (91.7%) 59 (98.3%) 56 (93.3%) 57 (98.3%) 54 (93.1%) facilities (≥80%) that remains high (≥80%) at 6 Sterile blade 53 (88.3%) 48 (80.0%) 55 (91.7%) 56 (93.3%) 54 (93.1%) 51 (87.9%) months Cord tie or clamp 54 (90.0%) 55 (91.7%) 55 (91.7%) 53 (88.3%) 52 (89.7%) 53 (91.4%) Hand hygiene supplies 54 (90.0%) 52 (86.7%) 55 (91.7%) 53 (88.3%) 53 (91.4%) 51 (87.9%) Gloves 58 (96.7%) 59 (98.3%) 58 (96.7%) 58 (96.7%) 58 (100%) 57 (98.3%) IV Fluid 56 (93.3%) 57 (95.0%) 57 (95.0%) 56 (93.3%) 57 (98.3%) 57 (98.3%) Baby scale 59 (98.3%) 57 (95.0%) 60 (100%) 60 (100%) 58 (100%) 58 (100%) Pads 58 (96.7%) 55 (91.7%) 57 (95.0%) 57 (95.0%) 53 (91.4%) 56 (96.6%) Blood pressure instrument 50 (83.3%) 51 (85.0%) 59 (98.3%) 49 (81.7%) 51 (87.9%) 52 (89.7%) Stethoscope 56 (93.3%) 54 (90.0%) 60 (100%) 54 (90.0%) 57 (98.3%) 55 (94.8%) Urine dip sticks 54 (90.0%) 54 (90.0%) 54 (90.0%) 56 (93.3%) 53 (91.4%) 50 (86.2%) Sterile needle-syringe 58 (96.7%) 60 (100%) 59 (98.3%) 56 (93.3%) 57 (98.3%) 56 (96.6%) Antibiotics mother 53 (88.3%) 59 (98.3%) 57 (95.0%) 56 (93.3%) 56 (96.6%) 54 (93.1%) BCG vaccine 60 (100%) 59 (98.3%) 59 (98.3%) 60 (100%) 57 (98.3%) 58 (100%) Polio vaccine 60 (100%) 60 (100%) 58 (96.7%) 58 (96.7%) 57 (98.3%) 58 (100%) Baby warmer 58 (96.7%) 53 (88.3%) 56 (93.3%) 52 (86.7%) 50 (86.2%) 51 (87.9%) Thermometer 48 (80.0%) 51 (85.0%) 60 (100%) 46 (76.7%) 55 (94.8%) 51 (87.9%) Antibiotics baby 39 (65.0%) 36 (60.0%) 52 (86.7%) 42 (70.0%) 45 (77.6%) 47 (81.0%) Mid-range availability that remains mid-range Oxytocin 43 (71.7%) 47 (78.3%) 36 (60.0%) 45 (75.0%) 41 (70.7%) 35 (60.3%) Magnesium Sulfate 37 (61.7%) 30 (50.0%) 40 (66.7%) 34 (56.7%) 39 (67.2%) 32 (55.2%) HIV testing kit 35 (58.3%) 43 (71.7%) 42 (70.0%) 40 (66.7%) 41 (70.7%) 37 (63.8%) Clean towel 33 (55.0%) 28 (46.7%) 38 (63.3%) 29 (48.3%) 37 (63.8%) 33 (56.9%) Partograph 10 (16.7%) 7 (11.7%) 22 (36.7%) 15 (25.0%) 18 (31.0%) 14 (24.1%) Low availability (<50%) that remains low Nevirapine baby 2 (3.3%) 1 (1.7%) 3 (5.0%) 2 (3.3%) 2 (3.4%) 0 (0%) Nevirapine mother 4 (6.7%) 1 (1.7%) 3 (5.0%) 1 (1.7%) 3 (5.2%) 1 (1.7%) Fetoscope or doppler 23 (38.3%) 23 (38.3%) 41 (68.3%) 23 (38.3%) 47 (81.0%) 31 (53.4%) Low availability (<50%) at baseline that moves to mid-range availability at 6 months Vitamin K 16 (26.7%) 16 (26.7%) 32 (53.3%) 26 (43.3%) 37 (63.8%) 28 (48.3%) Percentage of sites with four critical medicines available 9 (15.0%) 10 (16.7%) 18 (30.0%) 12 (20.0%) 24 (41.4%) 13 (22.4%) Note: At baseline and 6 months, all 60 intervention sites are included. However, at month 12, only 58 sites are included because two matched pairs were excluded from the study (due to closure) and month 12 surveys were not conducted. Presence of Hand hygiene supplies defined as either water and soap OR alcohol rub. Critical medications: Vitamin K, Magnesium Sulfate, Oxytocin, Antibiotics. For example, on average at baseline 26.7% of intervention sites and 26.7% control sites had Vitamin K available. At 6 months, 53.3% of intervention sites and 43.3% of control sites had Vitamin K available. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 775 Table 4 Categorization of individual items based on change in percentage of sites with item available from baseline to 6 months for all intervention sites (N = 60) and the intervention sites starting with the lowest quartile of supply availability (<20 of 28 supplies, N = 14) High availability (≥80%) that Mid-range availability (50–79%) Mid-range availability that Low availability (<50%) to mid- Low availability that remains low remains high to high availability remains mid-range range availability All Bottom quartile All Bottom quartile All Bottom quartile All Bottom quartile All Bottom quartile intervention intervention sites intervention intervention sites intervention intervention sites intervention intervention sites intervention intervention sites sites sites sites sites sites Suction machine X X Gloves X X Baby scale X X Pads X X Urine dip sticks X X Sterile needle-syringe X X BCG vaccine X X Polio vaccine X X Baby warmer X X Stethoscope X X Neonatal bag-and-mask X X Sterile blade X X Cord tie or clamp X X Hand hygiene supplies X X IV Fluid X X Blood pressure instrument X X Antibiotics mother X X Thermometer X X Antibiotics baby XX Oxytocin XX Magnesium Sulfate XX HIV testing kit XX Clean towel XX Vitamin K XX Fetoscope or doppler XX Partograph XX Nevirapine baby XX Nevirapine mother XX See Appendix B for actual percentages for each item. Note: At baseline and 6 months, all 60 intervention sites are included. However, at month 12, only 58 sites are included because two matched pairs were excluded from the study (due to closure) and month 12 surveys were not conducted. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 776 Maisonneuve et al. increases in supplies in the control facilities including vitamin K, In the intervention sites, while facilities procured an average of fetoscope/doppler and antibiotics for baby. 95% of items assessed from official sources, patients provided nearly In this setting, coaching with data feedback and action planning one-quarter of some important supplies including oxytocin, sterile at both district and facility levels was not enough to achieve a major blades and soap. This is consistent with other research demonstrat- and sustained impact on supply availability compared with temporal ing that although treatment at public facilities in India is supposed trends. Coaching was intended to address barriers in supply avail- to be free, this is not always the case [30]. Further research on sup- ability related to leadership engagement, birth attendant motivation ply availability should capture procurement methods in greater and data feedback linked with action planning (Semrau and detail to allow for better designed interventions that can improve Maisonneuve, personal communication). However, there were likely the supply chain rather than relying on patients. other barriers less amenable to coaching that prevented a larger While other studies have also noted variation in supply availabil- impact, such as those related to finances, policies, and district, state ity across Uttar Pradesh and India, the global literature is less clear and national supply chain performance [15]. on interventions to address supply gaps that do not require imple- The WHO Safe Childbirth Checklist has been implemented in other menting new supply chains or direct provision of resources. A sys- global settings and the availability of essential supplies has remained a tematic review analyzing the impact of interventions on medicine consistent barrier to performing the Essential Birth Practices [25]. availability at the primary healthcare level found a variable degree of Some coaching-based implementation models resolved supply barriers evidence and rigor [31]. Evidence was found that supervision visits by providing supplies to facilities [26, 27]. Another WHO Checklist in Zimbabwe strengthened primary health centers’ stock manage- implementation model in Namibia did not provide supplies and found ment though had limited effect on availability [32]. Comparatively, that availability improved over time with the coaching, however, avail- interventions focused on staff training programs on logistics manage- ability of some items declined during the lower intensity ‘maintenance ment systems in Nepal showed improvements in supply availability phase’ [28]. As noted, the BetterBirth Program did not provide supplies and stockouts [33]. In our search, we did not find further evidence to facilities as we did not want to create an unsustainable supply-chain on the effectiveness of coaching interventions targeting supplies. Our system in parallel to the state system. findings suggest while coaching was able to catalyze improvements Despite the intervention’s modest impact, the BetterBirth through changing some behaviors and actions in the existing system, Program did result in a greater difference in availability among sites additional supply chain strengthening interventions beyond knowl- which started with lower baseline supply availability. This suggests edge and behavior change are needed to effectively address gaps in that coaching may play a larger role in improving supply availability essential birth supplies and sustain improvement [15, 16]. in facilities with weaker supply chain management. Facilities with This study had several limitations. While both surveys utilized higher baseline supply availability may not have much room for drew from existing surveys and were pilot tested, they were not inde- improvement in those supplies and supply chain gaps responsive to pendently validated. However, we had formal data quality assurance coaching and local change. protocols and trainings to support data collectors for the facility sur- Further investigations are needed to understand why some sup- vey to ensure the quality of the data [34]. Additionally, we do not plies were responsive to coaching and why others were not. The have information on the quantity of stock available or if it was appro- intervention team noted in our discussions with them that supply- priately stored. Thus, we may have overestimated the supply availabil- availability improvements were often seen when specific barriers ity. Sources of procurement were also only by facility report and were amenable to district and facility coaching were resolved (for not confirmed. Finally, we did not measure change in attitudes and example, lack of knowledge of the importance of a supply or com- culture related to supply availability so we cannot completely explain munication gaps in identifying and addressing shortages). We also the successes and challenges encountered. Despite these limitations, found that for a single supply, there could be multiple site specific our study is an important first step towards identifying opportunities or broader barriers to availability. This may explain the heterogen- and challenges for coaching to help impact supply availability. eity seen in how the intervention improved availability of an item in some but not all sites (for example, Vitamin K). However, des- Conclusion pite 6 months of coaching, less than one-third of intervention sites had all four critical medicines. This persistent gap highlights the Integrating a focus on supply availability into coaching resulted ini- need for broader supply chain remedies beyond coaching. We also tially in a modest increase in overall supply availability, with lower found that supplies which were either infrequently needed (nevira- performing sites experiencing greater difference in availability after 6 pine in a low HIV seroprevalence setting) or remained at low months of the intervention. While coaching can play a role in uptake in intervention sites (partograph) did not respond to the strengthening supply availability, further research is needed to intervention. understand how it can be combined with broader supply-chain inter- We did see some improvement in supply availability in control ventions to contribute to and sustain improvement in essential birth sites. It is possible that, while data collection visits were unannounced, supply availability in frontline facilities in resource-limited settings. the quarterly assessment of supplies improved awareness of the importance of supply status in control sites. Benefits to control sites Supplementary material may also have spread from the intervention due to district-level advo- cacy meetings with the Chief Medical Officer as some districts con- Supplementary material is available at International Journal for Quality in tained both control and intervention sites. There were also targeted Health Care online. state-wide programs focusing on health-facility assessments during the study that may have contributed to improvements such as Kaya Kalp, Acknowledgements a National Rural Health Mission initiative focused on infection control [29]. These programs were incorporated at intervention and We recognize the Governments of India and Uttar Pradesh for collaboration control sites. and support to conduct this trial in public health facilities. We thank the Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 777 district and facility staff, women and their newborns for their participation in 16. Galvin G, Kalita T, Gupta S et al. Coaching to increase availability of the study. We are grateful to the members of the trials’ Scientific Advisory essential birth supplies for the WHO Safe Childbirth Checklist in Uttar Committee who contributed crucial guidance to the development of this study Pradesh, India. Proceedings of the Global Maternal and Newborn Health protocol: Himanshu Bhushan, Amit Kumar Ghosh, Zulfiqar Bhutta, Conference; 2015 Oct 18–21; Mexico City, Mexico. Waldemar Carlo, Vinita Das, Paul Francis, Amod Kumar, Matthews Mathai, 17. National Health Mission. Section VII: essential drugs and supplies. New Packirisamy Padmanbhan, Vinod Paul and Rajiv Tandon. We also thank the Delhi (India): Ministry of Health and Family Welfare, Government of past and current members of the BetterBirth study team in Boston and the India. Available at http://www.sifpsa.org/pdf/New/7-%20Essential% BetterBirth field team based in Uttar Pradesh for study implementation. 20Drugs%20and%20Supplies.pdf. Accessed [September 11 2017]. 18. Maiti R, Bhatia V, Padhy BM et al. Essential medicines: an Indian per- spective. Indian J Community Med 2015;40:223–32. 19. Patel S, Thumm M, Rahman J et al. Subnational procurement of maternal Funding health medicines: results from an assessment in Bangladesh. Arlington This work was generously funded through a grant from the Bill and Melinda (VA): Center for Pharmaceutical Management, Management Sciences for Gates Foundation. Health; 2014 May. Available at http://siapsprogram.org/publication/ subnational-procurement-of-maternal-health-medicines-results-from-an- assessment-in-bangladesh/. References 20. Mwencha M, Rosen JE, Spisak C et al. Upgrading supply chain manage- 1. Koblinsky M, Moyer CA, Calvert C et al. Quality maternity care for ment systems to improve availability of medicines in Tanzania: evaluation every woman, everywhere: a call to action. Lancet 2016;388:2307–20. of performance and cost effects. Glob Health Sci Pract 2017;5:399–411. 2. Kinney MV, Boldosser-Boesch A, McCallon B. Quality, equity, and dig- 21. Semrau KEA, Hirschhorn LR, Kodkany B et al. Effectiveness of the nity for women and babies. Lancet 2016;388:2066–8. WHO Safe Childbirth Checklist program in reducing severe maternal, 3. Sharma J, Leslie HH, Kundu F et al. Poor quality for poor women? fetal, and newborn harm in Uttar Pradesh, India: study protocol for a Inequities in the quality of antenatal and delivery care in Kenya. PLoS matched-pair, cluster-randomized controlled trial. Trials 2016;17:576. One 2017;12:e0171236. 22. Semrau KEA, Hirschhorn LR, Marx Delaney M et al. Outcomes of a 4. Mohanan M, Bauhoff S, La Forgia G et al. Effect of Chiranjeevi Yojana coaching-based WHO Safe Childbirth Checklist Program in India. N Engl on institutional deliveries and neonatal and maternal outcomes in J Med 2017;377:2313–24. Gujarat, India: a difference-in-differences analysis. Bull World Health 23. Marx Delaney M, Maji P, Kalita T et al. Improving adherence to essential Organ 2014;92:187–94. birth practices using the WHO Safe Childbirth Checklist with peer coach- 5. Randive B, Diwan V, De Costa A. India’s conditional cash transfer pro- ing: experience from 60 public health facilities in Uttar Pradesh, India. gramme (the JSY) to promote institutional birth: is there an association Glob Health Sci Pract 2017;5:217–31. between institutional birth proportion and maternal mortality? PLoS One 24. McCullagh P, Nelder JA. Generalized linear models, 2nd edn. Boca 2013;8:e67452. Raton (FL): CRC Press, 1989. 6. Powell-Jackson T, Mazumdar S, Mills A. Financial incentives in health: 25. Perry W, Bagheri Nejad S, Tuomisto K et al. Implementing the WHO new evidence from India’s Janani Suraksha Yojana. J Health Econ 2015; Safe Childbirth Checklist: lessons from a global collaboration. BMJ Glob 43:154–69. Health 2017;2:e000241. 7. Ministry of Health & Family Welfare, Department of Health and Family 26. Kumar S, Yadav V, Balasubramaniam S et al. Effectiveness of the WHO Welfare, India:. http://mohfw.gov.in/sites/default/files/9147562941489753121. SCC on improving adherence to essential practices during childbirth, in pdf Accessed [September 11 2017]. resource constrained settings. BMC Pregnancy Childbirth 2016;16:345. 8. Office of the Registrar General & Census Commissioner Vital Statistics 27. Nababan HY, Islam R, Mostari S et al. Improving quality of care for Division. Annual health survey 2012–13 fact sheet Uttar Pradesh. New maternal and newborn health: a pre-post evaluation of the Safe Delhi (India): Ministry of Home Affairs, Government of India; 2013. Childbirth Checklist at a hospital in Bangladesh. BMC Pregnancy 9. Office of the Registrar General & Census Commissioner. Sample registra- Childbirth 2017;17:402. tion system statistical report 2014. New Delhi (India): Government of 28. Kabongo L, Gass JD Jr., Kivondo B et al. Implementing the WHO Safe India; 2015. Childbirth Checklist: lessons learnt on a quality improvement initiative to 10. Sharma AK. National rural health mission: time to take stock. Indian J improve mother and newborn care at Gobabis District Hospital, Community Med 2009;34:175–82. Namibia. BMJ Open Qual 2017;6:e000145. 11. Programme Evaluation Organisation. Evaluation study of National Rural 29. National Health Mission. Award to public health facilities: Kayakalp. Health Mission (NRHM) in 7 states. 2011. New Delhi (India): New Delhi (India): Ministry of Health and Family Welfare, Government Government of India, Planning Commission; 2011 Feb. Available from: of India. Available at https://www.nhp.gov.in/kayakalp-swacchta- http://planningcommission.nic.in/reports/peoreport/peoevalu/peo_2807.pdf. guidelines-for-public-health-facilities_pg. Accessed [September 11 2017]. Accessed [September 11 2017]. 30. Issac A, Chatterjee S, Srivastava A et al. Out of pocket expenditure to 12. Harvard School of Public Health, World Health Organization, Population deliver at public health facilities in India: a cross sectional analysis. Services International. Better Birth: Rollout and evaluation of the WHO Reprod Health 2016;13:99. safe childbirth checklist. International Consultation Meeting; 2011 Dec 31. Nunan M, Duke T. Effectiveness of pharmacy interventions in improving 13–14; New Delhi, India. Geneva: World Health Organization; 2009. availability of essential medicines at the primary healthcare level. Trop 13. Kara N, Firestone R, Kalita T et al. The BetterBirth Program: pursuing Med Int Health 2011;16:647–58. effective adoption and sustained use of the WHO Safe Childbirth 32. Trap B, Todd CH, Moore H et al. The impact of supervision on stock Checklist through coaching-based implementation in Uttar Pradesh, management and adherence to treatment guidelines: a randomized con- India. Glob Health Sci Pract 2017;5:232–43. trolled trial. Health Policy Plan 2001;16:273–80. 14. Spector JM, Lashoher A, Agrawal P et al. Designing the WHO Safe 33. USAID Deliver Project, Task Order I. After receiving USAID/DELIVER Childbirth Checklist program to improve quality of care at childbirth. Int project logistics training, access to health products improves in Nepal. J Gynaecol Obstet 2013;122:164–8. Arlington (VA): Jon Snow Inc; 2009 July. Available at http://pdf.usaid. 15. Bahl KLC, Palmer A, Rosenberg T. Increasing access to essential maternal gov/pdf_docs/PA00KRRK.pdf. Accessed [September 11 2017]. health supplies: a scoping of market-based activities, gaps, and opportun- 34. Gass JD Jr., Misra A, Yadav MNS et al. Implementation and results of an ities. Washington DC: Results for Development Institute (R4D), integrated data quality assurance protocol in a randomized controlled Reproductive Health Supplies Coalition (RSC), 2016. March. trial in Uttar Pradesh, India. Trials 2017;18:418. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal for Quality in Health Care Oxford University Press

Effectiveness of a WHO Safe Childbirth Checklist Coaching-based intervention on the availability of Essential Birth Supplies in Uttar Pradesh, India

Download PDF
9 pages

Loading next page...
1
 
/lp/ou_press/effectiveness-of-a-who-safe-childbirth-checklist-coaching-based-8IXwUzlTx1

References (38)

Publisher
Oxford University Press
Copyright
Copyright © 2022 International Society for Quality in Health Care and Oxford University Press
ISSN
1353-4505
eISSN
1464-3677
DOI
10.1093/intqhc/mzy086
Publisher site
See Article on Publisher Site

Abstract

Objective: Evaluate the impact of a World Health Organization Safe Childbirth Checklist coaching- based intervention (BetterBirth Program) on availability and procurement of essential childbirth- related supplies. Design: Matched pair, cluster-randomized controlled trial. Setting: Uttar Pradesh, India. Participants: 120 government-sector health facilities (60 interventions, 60 controls). Supply- availability surveys were conducted quarterly in all sites. Coaches collected supply procurement sources from intervention sites. Interventions: Coaching targeting implementation of Checklist with data feedback and action planning. Main Outcome Measures: Mean supply availability by study arm; change in procurement sources for intervention sites. Results: At baseline, 6 and 12 months, the intervention sites had a mean of 20.9 (95% confidence interval (CI): 20.2–21.5); 22.4 (95% CI: 21.8–22.9) and 22.1 (95% CI:21.4–22.8) items, respectively. Control sites had 20.8 (95% CI: 20.3–21.3); 20.9 (95% CI: 20.3–21.5) and 21.7 (95% CI: 20.8–22.6) items at the same time-points. There was a small but statistically significant higher availability in interven- tion sites at 6 months (difference-in-difference (DID) = 1.43, P < 0.001), which was not seen by 12 months (DID = 0.37, P = 0.53). Greater difference between intervention and control sites starting in the bottom quartile of supply availability was seen at 6 months (DID = 4.0, P = 0.0002), with no © The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. 769 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 770 Maisonneuve et al. significant difference by 12 months (DID = 1.5, P = 0.154). No change was seen in procurement sources with ~5% procured by patients with some rates as high as 29% (oxytocin). Conclusions: Implementation of the BetterBirth Program, incorporating supply availability, resulted in modest improvements with catch-up by control facilities by 12 months. Supply-chain coaching may be most beneficial in sites starting with lower supply availability. Efforts are needed to reduce reliance on patient-funding for some critical medications. Trial Registration: ClinicalTrials.gov #NCT02148952; Universal Trial Number: U1111-1131–5647 Key words: supply availability, maternal and newborn health, WHO Safe Childbirth Checklist, quality improvement Introduction conducted from October 2014 to January 2017 (Appendix A). The trial Reducing childbirth-related maternal and neonatal morbidity and included primary, community and first-referral health centers (60 inter- mortality requires improving patients’ access to high quality maternity vention and 60 control facilities). Detailed descriptions of the BetterBirth care [1–3]. Although rates of facility-based deliveries have increased, Program and trial methodology are described elsewhere [13, 21, 22]. this shift has not consistently translated into improved outcomes [4–7]. In 2012, the Neonatal Mortality Rate in Uttar Pradesh, India BetterBirth Program remained unacceptably high at 49 deaths per 1000 live births [8, 9]. Core to the BetterBirth intervention was a coaching team who worked In an effort to improve the quality and outcomes of maternity care to address non-adherence to essential birth practices, including supply- and in collaboration with the Government of India and State of Uttar related barriers, at multiple healthcare system levels. At each of the 60 Pradesh, a randomized controlled trial was undertaken to measure the intervention facilities, coaches (nurses) conducted 43 day-long visits impact of the BetterBirth Program, a coaching-based implementation over 8 months (twice per week tapering to monthly) during which of the World Health Organization (WHO) Safe Childbirth Checklist they worked with birth attendants to deliver Checklist-identified prac- [10–13]. The Checklist is a job aid designed to help birth attendants tices. Coaches provided feedback to birth attendants about their adhere to 28 essential birth practices known to save lives [14]. adherence, engaged them in identifying barriers and assisted with Adherence to the Checklist-identified practices requires 28 supplies action planning to resolve issues. Coach Team Leaders (physicians or (Table 1). However in India, there are significant gaps in the availabil- public health professionals) accompanied coaches on alternating visits, ity of medicines and equipment for maternity care in government- working with facility leadership to review data on Checklist adher- sector facilities [15–17]. For example, a survey in Uttar Pradesh found ence. They then worked with leadership to identify and resolve that only 53.8% of first-line referral facilities had injectable magne- facility-level barriers to Checklist adoption, including improving sium sulfate available [17]. When supplies are unavailable in facilities, supply availability. They also coached the Childbirth Quality patients either buy them or forego treatment. As a result, the unavail- Coordinator, a facility-based champion responsible for motivating ability of supplies is a serious barrier to the provision of high quality staff utilization of the Checklist, identifying and resolving supply issues care in facilities and, consequentially, a positive outcome for mothers and supporting the facility’s commitment to Checklist-use post-trial. and babies of all socioeconomic backgrounds [18]. In addition to one-on-one coaching, Coaches and Coach Team To avoid creating a non-sustainable supply source in the context of Leaders held data-sharing meetings at facilities fortnightly to feed- the study, the BetterBirth Program did not provide any supplies to facil- back data on Checklist adherence and supply availability and sup- ities [13]. Instead, the theory of change was that coaching at the point port action planning. To ensure district-level support, Coach Team of care, as well as at facility and district levels would improve supply Leaders facilitated bi-monthly progress meetings with district health availability by promoting behavior and system change through (a) officials and facility leadership to review adherence and supply- improved knowledge of the essential supplies for childbirth and advo- related data and help identify plans needed at the district level to cacy at the frontline to improve availability; (b) routine monitoring and improve childbirth care, including supply availability. The data driven communication of supply gaps to facility and district lea- BetterBirth Program did not provide supplies (except paper copies ders; accompanied by (c) coaching on action plans for solutions which and posters of the Checklist during the trial) or financial support. leveraged existing supply chains and facility and district resources. Results of the impact BetterBirth had on essential birth practice Although there have been efforts to identify interventions that uptake and health outcomes are described elsewhere [22, 23]. improve health systems’ supply-chains, further research is needed to understand how supply availability through strengthening existing systems can be effectively integrated into coaching-based quality Ethics improvement efforts [15, 19, 20]. Here, we describe the BetterBirth The study protocol received approval by the following ethical review Program’s impact on essential birth supply availability and the pro- boards at the Community Empowerment Lab, Jawaharlal Nehru curement source of available supplies. Medical College, Harvard T.H. Chan School of Public Health, Population Service International, the WHO and the Indian Council of Medical Research. Each facility and birth attendant formally Methods agreed to participate in the BetterBirth Program at the beginning of Study design and site characteristics the study. Coaches accompanied birth attendants during their work- The BetterBirth Trial was a matched pair, cluster-randomized controlled shift and documented practices during patient-care activities at the trial across 120 government health facilities in Uttar Pradesh, India facility. Coaches collected no patient identifiers. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 771 Table 1 Supplies associated with WHO Safe Childbirth Checklist essential birth practices, location of evaluated items, and whether their source of procurement was evaluated in the BetterBirth trial in Uttar Pradesh, India Components of checklist Associated essential birth Location evaluated for availability (1- admission, 2- duty room, Evaluated for supply score (28 items total) practice on the WHO Safe 3- labor and delivery (L&D) room, 4- post-partum room or procurement (N = 23) Childbirth Checklist anywhere in facility (Facility)) Pause Point 1 1. Thermometer Mother’s temperature All four locations X 2. Blood pressure cuff Mother’s blood pressure All four locations X 3. Stethoscope X 4. Fetoscope or Doppler Fetal heart rate Admission room, Duty room, L&D room X 5. Partograph Partograph started Admission room, Duty room, L&D room 6. Antibiotics (mother) Antibiotics for the mother Facility X 7. Urine dip sticks Magnesium sulfate for the Facility X 8. Magnesium sulfate mother All four locations X 9. HIV testing kit HIV status of the mother Facility 10. Nevirapine (mother) Administer nevirapine as Facility needed Pause Point 2 11. Clean gloves Confirm essential supplies for Admission room, Duty room, L&D room X mother are at bedside 12. [(Soap and Admission room, Duty room, L&D room X 12. Clean water) or Admission room, Duty room, L&D room X 12. (Alcohol rub)] Admission room, Duty room, L&D room 13. Sterile needle/syringe All four locations X 14. Oxytocin L&D room X 15. Clean pads Admission room, Duty room, L&D room X 16. Clean towel Confirm essential supplies for L&D room X 17. Clean blade/scissor baby are at beside L&D room X 18. Cord tie/clamp L&D room X 19. Mucus extractor/suction L&D room X 20. Bag-and-mask L&D room X 21. Vitamin K Facility X Pause Point 3 1. Thermometer Baby’s temperature All four locations Above 3. Stethoscope Baby’s respiratory rate All four locations Above 22. Baby scale Baby’s weight L&D room X 23. Intravenous fluid bag Start IV fluids (if mother Facility bleeding) 24. Antibiotics (baby) Antibiotics for the baby Facility X 25. Baby warmer Special care/monitoring for L&D room baby 26. Nevirapine (baby) Administer nevirapine as Facility needed Pause Point 4 27. BCG vaccine Administer vaccines Facility X 28. Polio vaccine Facility X BCG, Bacillus Calmette–Guérin Vaccine; L&D, Labor and Delivery. Item is deemed available if it is present in a designated area and functional (or not expired, as relevant). While certain items (e.g. stethoscope or thermometer) can be available within any room in the labor ward complex (e.g. availability in either admission or duty room, or labor and delivery room, or post-partum ward), other items were expected to be available in a designated area (e.g. neonatal bag-and-mask available in the labor and delivery room). Procurement is how essential medicines and medical supplies are sourced. The source of supply was defined by four categories; Official: (1) district or state government defined system; (2) from facility funds; Unofficial: (3) Patient or family: brought from home or purchased or (4) Any other means, e.g. staff purchased or donation. Denotes a checklist practice repeated across multiple Pause Points. For simplicity, these are listed in this table only at the first Pause Point at which they are performed. designated areas, it was coded as available. Coaches were not pre- Data sources sent on days when surveys were conducted and did not have access Facility essential birth supply availability survey to data. Data collectors conducted facility surveys at baseline and Study staff conducted the supply-availability survey (Appendix D) at quarterly over the study. Baseline for intervention sites was just all sites to measure availability of the 28 essential birth supplies prior to the beginning of the intervention. Baseline for control sites (Table 1). During unannounced visits, data collectors observed facil- was just prior to the first study-related visits, which occurred ~2 ity areas where supplies were supposed to be stocked. If an item was months after intervention start at their matched pair [22]. Surveys observed as functional, not expired, and in at least one of the Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 772 Maisonneuve et al. were collected from the 60 pairs at baseline through 6 months and baseline): 14 intervention sites (23%) and 14 control sites (23%). from 58 pairs at 9 and 12 months (two facilities closed and moved An additional model was fit that included a control for baseline sup- after completing their third survey at 6 months, resulting in exclu- ply availability interacted with study arm. The term would be sig- sion of these sites and their pairs from the 12 month analysis). For nificant if the intervention had a different effect among sites with simplicity, only data from baseline, 6 and 12 months are reported; fewer baseline supplies as compared to those with more. 3- and 9-month data are available upon request. Coach supply source survey Coach supply source survey The proportion of available supplies purchased by an unofficial pro- During facility visits, Coach Team Leaders conducted a supply curement method (defined as being purchased by the patient/family source survey (Appendix E) with the facility leader or Childbirth or ‘other source’) versus the official supply chain (procured by facil- Quality Coordinator to measure the availability and procurement ity or district) over time was examined. In cases where a supply was source of 23 of the 28 checklist supplies (Table 1). If an item was procured from both an unofficial and an official method, it was available, Coach Team Leaders would discuss with facility represen- coded as procured through the unofficial method. For this analysis, tatives how the supply was procured: (1) through the district or state coach supply source data was grouped into 1-month time periods government supply-system, (2) from facility funds, (3) from patient (except for the final months 8 and 9 which were grouped together) or family (brought from home or purchased) or (4) from any other and the mean proportion of all available items procured by each means (e.g. facility staff purchased or donation). The first survey source was calculated. was conducted on the day after the BetterBirth Program launched and was repeated approximately every 2 weeks and then monthly, Results as Coach Team Leader visits decreased in frequency. After their vis- it, Coach Team Leaders entered data into a mobile phone-based Facility essential birth supply availability CommCare application (Dimagi, Cambridge, MA), which visualized A total of 596 completed facility supply-availability surveys were avail- data over time as a heat map (Appendix F) to be shared with facility able for analysis. At baseline, the intervention sites had a mean of 20.9 staff and leadership to inform action planning [13]. items (95% CI: 20.2–21.5), which improved to 22.4 (95% CI: 21.8–22.9) at 6 months and 22.1 (95% CI: 21.4–22.8) at 12 months (Table 2). The control sites started with a mean of 20.8 items (95% CI: Analysis of surveys 20.3–21.3), which increased to 20.9 (95% CI: 20.3–21.5)at6months All statistical analyses were conducted using SAS v9.4 (SAS and 21.7 (95% CI: 20.8–22.6) at 12 months. There was a small but Institute, Cary, NC). statistically significant better improvement in the intervention versus control arm at 6 months (DID = 1.43 supplies, P < 0.001), which was Facility essential birth supply-availability survey no longer significant by 12 months (DID = 0.37 supplies, P = 0.53). The primary outcome was the difference in overall availability of The facilities with supply availability in the lowest quartile (<20 essential birth supplies after 6 months of coaching. Six-months was supplies) at baseline were spread across all five study-defined chosen because maximum impact was expected at this time while regions. While the average supply-availability score for intervention coaching was still underway and a decay was seen in adherence to and control sites in this subgroup increased over time, intervention Checklist practices in intervention sites at 12 months (4 months after sites had significantly better improvement by 6 months (interven- the intervention’s end) [22]. tion: 17.6–22.0 for intervention and 18.3–18.7 for controls at base- Essential birth supply availability was calculated as the count of line and 6 months, respectively, DID: 4.00, P = 0.0002) (Fig. 1 and the 28 Essential Birth Supplies available during a survey administra- Table 2). The difference in improvement between intervention and tion (Table 1). The mean percentage of sites with each supply avail- control was no longer statistically significant by 12 months (DID: able over time was calculated, in addition to the mean percentage of 1.51, P = 0.154). Modeling results found that the interaction term sites with all four medications considered critical to reducing mater- between baseline supply level and study arm was statistically signifi- nal and neonatal harm (oxytocin, magnesium sulfate, Vitamin K cant (P < 0.001, data not shown), confirming that the impact of the and antibiotics for mothers or babies). To avoid inflating the intervention was stronger among sites with lower baseline supplies. analysis-wide α (type I) error, we did not conduct significance tests Individual item supply availability was variable across sites. In on these differences. the analysis of all sites, 18 items were highly available (≥80% of Within each matched pair, facility surveys could occur up to 2 facilities) regardless of study arm at baseline and remained above months apart due to differences in baseline. Thus, models were con- that threshold throughout the study; five items had mid-range avail- structed to estimate differences in supply levels across study arms at ability (50–79% of sites) throughout the study; and five items (par- baseline, 6 and 12 months. Logistic regression models were fit using tograph, nevirapine for baby, nevirapine for mother, Doppler or the counting method to predict the number of 28 items available, fetoscope and Vitamin K) had low availability (presence in <50% of adjusting for matching and repeated measures over time [24]. facilities) at baseline (Table 3). While availability of Vitamin K and Models included nonlinear terms for time and an interaction fetoscope/Doppler increased to mid-range in intervention but not between study arm and time. From these models, the difference-in- control sites at month 6, the availability of partograph and nevira- difference (DID) was calculated for the number of supplies available pine for mother and baby remained low in both arms (Table 3). For from baseline to 6 months and, as a secondary outcome, from base- the bottom quartile intervention sites, eight items moved from mid- line to 12 months, with tests for statistical significance at α = 0.05. level availability at baseline to high availability at 6 months To explore facilities starting with fewer essential birth supplies, a (Table 4). In contrast to the overall intervention sites, in the bottom sub-analysis was conducted of facilities which fell into the bottom quartile intervention sites, no increase from the low availability cat- quartile of baseline supply availability (<20 of 28 supplies at egory was seen for the fetoscope/Doppler. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 773 Figure 1 Mean number of Essential Birth Supplies (N = 28) over time by study arm for sites starting at the bottom quartile of supply availability (<20 of the 28 supplies at baseline). Less than one-fifth of all sites had all four of the critical medica- tions (oxytocin, magnesium sulfate, Vitamin K, antibiotics) available at baseline (15% and 16.7% in intervention and control arms, respectively) with availability improving at 6 months (30.0% and 20.0% in intervention and control arms, respectively) and 12 months (41.4% and 22.4% in intervention and control arms, respectively; Table 3). Coach supply source survey Over the course of the study, Coach Team Leaders completed 964 supply source surveys across the 60 intervention sites. Each facility was expected to have a minimum of 14 surveys completed; Coach Team Leaders completed more at their discretion (mean 16, range 9–23). As expected due to Coach Team Leader visits decreasing in frequency over the intervention, the total number of surveys com- pleted per month declined from the first month of the intervention (N = 152) to the final months (N = 56). In the first month of the intervention, facilities procured 95% of available supplies from offi- cial sources (94% district, 1% facility) and 5% from unofficial sources (4% from patients and 1% from ‘other’ sources). No signifi- cant change was seen in the percentage of supplies sourced from official or unofficial sources over time (Appendix C). The top four supplies available due to purchase by patients were injectable oxyto- cin (29.1% purchased by patients), a sterile blade (28.4%), soap (23.1%) and Vitamin K (15.4%). Discussion The BetterBirth Program integrated a focus on supply readiness and use into coaching to leverage existing supply chains and improve availability at the facility and point of care. On average, facilities started with relatively high supply levels. The intervention resulted in a modest but statistically significant higher mean overall availabil- ity of the required 28 supplies at 6 months. This difference was largely driven by increased availability of 2 supplies: fetal heart rate monitors and vitamin K, which increased from availability in <50% of facilities at baseline to between 50–79% of facilities at 6 months. However, this difference was no longer seen at 12 months, in part due to no further improvement in intervention sites along with Table 2 Mean number of the 28 essential birth supplies (with 95% CI) by arm over time in (1) all sites and (2) in facilities starting in the bottom quartile of supply availability (<20 supplies at baseline) 0 Months (baseline) 6 Months 12 Months Difference-in-difference Intervention Control Intervention Control Intervention Control 0–6 Months 0–12 Months (N = 60) (N = 60) (N = 60) (N = 60) (N = 58) (N = 58) (1) All sites 20.9 (20.2–21.5) 20.8 (20.3–21.3) 22.4 (21.8–22.9) 20.9 (20.3–21.5) 22.1 (21.4–22.8) 21.7 (20.8–22.6) 1.43; P < 0.001 0.37; P = 0.53 0 Months (baseline) 6 Months 12 Months Difference-in-difference Intervention Control Intervention Control Intervention Control 0–6 Months 0–12 Months (N = 14) (N = 14) (N = 14) (N = 14) (N = 14) (N = 14) (2) Bottom 17.6 (16.5–18.6) 18.3 (17.4–19.2) 22.0 (21.0–23.1) 18.7 (17.6–19.9) 21.8 (20.6–23.1) 21.0 (19.6–22.5) 4.00 P = 0.0002 1.51 P = 0.154 quartile sites Note: At baseline and 6 months, all 60 intervention sites are included. At month 12, only 58 sites are included because two matched pairs were excluded from the study (due to closure) and month 12 surveys were not conducted. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 774 Maisonneuve et al. Table 3 Average availability of each of the 28 Essential Birth Supplies on the WHO Safe Childbirth Checklist across intervention and control sites over time Average percentage of sites with each item available over time Item 0 Months (baseline) 6 Months 12 Months Category based on change in % of Intervention sites with item available from baseline to 6 months Intervention Control Intervention Control Intervention Control n (%) n (%) n (%) n (%) n (%) n (%) (N = 60) (N = 60) (N = 60) (N = 60) (N = 58) (N = 58) Suction machine/mucus extractor 60 (100%) 60 (100%) 60 (100%) 59 (98.3%) 58 (100%) 57 (98.3%) High baseline availability across intervention Neonatal bag-and-mask 52 (86.7%) 55 (91.7%) 59 (98.3%) 56 (93.3%) 57 (98.3%) 54 (93.1%) facilities (≥80%) that remains high (≥80%) at 6 Sterile blade 53 (88.3%) 48 (80.0%) 55 (91.7%) 56 (93.3%) 54 (93.1%) 51 (87.9%) months Cord tie or clamp 54 (90.0%) 55 (91.7%) 55 (91.7%) 53 (88.3%) 52 (89.7%) 53 (91.4%) Hand hygiene supplies 54 (90.0%) 52 (86.7%) 55 (91.7%) 53 (88.3%) 53 (91.4%) 51 (87.9%) Gloves 58 (96.7%) 59 (98.3%) 58 (96.7%) 58 (96.7%) 58 (100%) 57 (98.3%) IV Fluid 56 (93.3%) 57 (95.0%) 57 (95.0%) 56 (93.3%) 57 (98.3%) 57 (98.3%) Baby scale 59 (98.3%) 57 (95.0%) 60 (100%) 60 (100%) 58 (100%) 58 (100%) Pads 58 (96.7%) 55 (91.7%) 57 (95.0%) 57 (95.0%) 53 (91.4%) 56 (96.6%) Blood pressure instrument 50 (83.3%) 51 (85.0%) 59 (98.3%) 49 (81.7%) 51 (87.9%) 52 (89.7%) Stethoscope 56 (93.3%) 54 (90.0%) 60 (100%) 54 (90.0%) 57 (98.3%) 55 (94.8%) Urine dip sticks 54 (90.0%) 54 (90.0%) 54 (90.0%) 56 (93.3%) 53 (91.4%) 50 (86.2%) Sterile needle-syringe 58 (96.7%) 60 (100%) 59 (98.3%) 56 (93.3%) 57 (98.3%) 56 (96.6%) Antibiotics mother 53 (88.3%) 59 (98.3%) 57 (95.0%) 56 (93.3%) 56 (96.6%) 54 (93.1%) BCG vaccine 60 (100%) 59 (98.3%) 59 (98.3%) 60 (100%) 57 (98.3%) 58 (100%) Polio vaccine 60 (100%) 60 (100%) 58 (96.7%) 58 (96.7%) 57 (98.3%) 58 (100%) Baby warmer 58 (96.7%) 53 (88.3%) 56 (93.3%) 52 (86.7%) 50 (86.2%) 51 (87.9%) Thermometer 48 (80.0%) 51 (85.0%) 60 (100%) 46 (76.7%) 55 (94.8%) 51 (87.9%) Antibiotics baby 39 (65.0%) 36 (60.0%) 52 (86.7%) 42 (70.0%) 45 (77.6%) 47 (81.0%) Mid-range availability that remains mid-range Oxytocin 43 (71.7%) 47 (78.3%) 36 (60.0%) 45 (75.0%) 41 (70.7%) 35 (60.3%) Magnesium Sulfate 37 (61.7%) 30 (50.0%) 40 (66.7%) 34 (56.7%) 39 (67.2%) 32 (55.2%) HIV testing kit 35 (58.3%) 43 (71.7%) 42 (70.0%) 40 (66.7%) 41 (70.7%) 37 (63.8%) Clean towel 33 (55.0%) 28 (46.7%) 38 (63.3%) 29 (48.3%) 37 (63.8%) 33 (56.9%) Partograph 10 (16.7%) 7 (11.7%) 22 (36.7%) 15 (25.0%) 18 (31.0%) 14 (24.1%) Low availability (<50%) that remains low Nevirapine baby 2 (3.3%) 1 (1.7%) 3 (5.0%) 2 (3.3%) 2 (3.4%) 0 (0%) Nevirapine mother 4 (6.7%) 1 (1.7%) 3 (5.0%) 1 (1.7%) 3 (5.2%) 1 (1.7%) Fetoscope or doppler 23 (38.3%) 23 (38.3%) 41 (68.3%) 23 (38.3%) 47 (81.0%) 31 (53.4%) Low availability (<50%) at baseline that moves to mid-range availability at 6 months Vitamin K 16 (26.7%) 16 (26.7%) 32 (53.3%) 26 (43.3%) 37 (63.8%) 28 (48.3%) Percentage of sites with four critical medicines available 9 (15.0%) 10 (16.7%) 18 (30.0%) 12 (20.0%) 24 (41.4%) 13 (22.4%) Note: At baseline and 6 months, all 60 intervention sites are included. However, at month 12, only 58 sites are included because two matched pairs were excluded from the study (due to closure) and month 12 surveys were not conducted. Presence of Hand hygiene supplies defined as either water and soap OR alcohol rub. Critical medications: Vitamin K, Magnesium Sulfate, Oxytocin, Antibiotics. For example, on average at baseline 26.7% of intervention sites and 26.7% control sites had Vitamin K available. At 6 months, 53.3% of intervention sites and 43.3% of control sites had Vitamin K available. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 775 Table 4 Categorization of individual items based on change in percentage of sites with item available from baseline to 6 months for all intervention sites (N = 60) and the intervention sites starting with the lowest quartile of supply availability (<20 of 28 supplies, N = 14) High availability (≥80%) that Mid-range availability (50–79%) Mid-range availability that Low availability (<50%) to mid- Low availability that remains low remains high to high availability remains mid-range range availability All Bottom quartile All Bottom quartile All Bottom quartile All Bottom quartile All Bottom quartile intervention intervention sites intervention intervention sites intervention intervention sites intervention intervention sites intervention intervention sites sites sites sites sites sites Suction machine X X Gloves X X Baby scale X X Pads X X Urine dip sticks X X Sterile needle-syringe X X BCG vaccine X X Polio vaccine X X Baby warmer X X Stethoscope X X Neonatal bag-and-mask X X Sterile blade X X Cord tie or clamp X X Hand hygiene supplies X X IV Fluid X X Blood pressure instrument X X Antibiotics mother X X Thermometer X X Antibiotics baby XX Oxytocin XX Magnesium Sulfate XX HIV testing kit XX Clean towel XX Vitamin K XX Fetoscope or doppler XX Partograph XX Nevirapine baby XX Nevirapine mother XX See Appendix B for actual percentages for each item. Note: At baseline and 6 months, all 60 intervention sites are included. However, at month 12, only 58 sites are included because two matched pairs were excluded from the study (due to closure) and month 12 surveys were not conducted. Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 776 Maisonneuve et al. increases in supplies in the control facilities including vitamin K, In the intervention sites, while facilities procured an average of fetoscope/doppler and antibiotics for baby. 95% of items assessed from official sources, patients provided nearly In this setting, coaching with data feedback and action planning one-quarter of some important supplies including oxytocin, sterile at both district and facility levels was not enough to achieve a major blades and soap. This is consistent with other research demonstrat- and sustained impact on supply availability compared with temporal ing that although treatment at public facilities in India is supposed trends. Coaching was intended to address barriers in supply avail- to be free, this is not always the case [30]. Further research on sup- ability related to leadership engagement, birth attendant motivation ply availability should capture procurement methods in greater and data feedback linked with action planning (Semrau and detail to allow for better designed interventions that can improve Maisonneuve, personal communication). However, there were likely the supply chain rather than relying on patients. other barriers less amenable to coaching that prevented a larger While other studies have also noted variation in supply availabil- impact, such as those related to finances, policies, and district, state ity across Uttar Pradesh and India, the global literature is less clear and national supply chain performance [15]. on interventions to address supply gaps that do not require imple- The WHO Safe Childbirth Checklist has been implemented in other menting new supply chains or direct provision of resources. A sys- global settings and the availability of essential supplies has remained a tematic review analyzing the impact of interventions on medicine consistent barrier to performing the Essential Birth Practices [25]. availability at the primary healthcare level found a variable degree of Some coaching-based implementation models resolved supply barriers evidence and rigor [31]. Evidence was found that supervision visits by providing supplies to facilities [26, 27]. Another WHO Checklist in Zimbabwe strengthened primary health centers’ stock manage- implementation model in Namibia did not provide supplies and found ment though had limited effect on availability [32]. Comparatively, that availability improved over time with the coaching, however, avail- interventions focused on staff training programs on logistics manage- ability of some items declined during the lower intensity ‘maintenance ment systems in Nepal showed improvements in supply availability phase’ [28]. As noted, the BetterBirth Program did not provide supplies and stockouts [33]. In our search, we did not find further evidence to facilities as we did not want to create an unsustainable supply-chain on the effectiveness of coaching interventions targeting supplies. Our system in parallel to the state system. findings suggest while coaching was able to catalyze improvements Despite the intervention’s modest impact, the BetterBirth through changing some behaviors and actions in the existing system, Program did result in a greater difference in availability among sites additional supply chain strengthening interventions beyond knowl- which started with lower baseline supply availability. This suggests edge and behavior change are needed to effectively address gaps in that coaching may play a larger role in improving supply availability essential birth supplies and sustain improvement [15, 16]. in facilities with weaker supply chain management. Facilities with This study had several limitations. While both surveys utilized higher baseline supply availability may not have much room for drew from existing surveys and were pilot tested, they were not inde- improvement in those supplies and supply chain gaps responsive to pendently validated. However, we had formal data quality assurance coaching and local change. protocols and trainings to support data collectors for the facility sur- Further investigations are needed to understand why some sup- vey to ensure the quality of the data [34]. Additionally, we do not plies were responsive to coaching and why others were not. The have information on the quantity of stock available or if it was appro- intervention team noted in our discussions with them that supply- priately stored. Thus, we may have overestimated the supply availabil- availability improvements were often seen when specific barriers ity. Sources of procurement were also only by facility report and were amenable to district and facility coaching were resolved (for not confirmed. Finally, we did not measure change in attitudes and example, lack of knowledge of the importance of a supply or com- culture related to supply availability so we cannot completely explain munication gaps in identifying and addressing shortages). We also the successes and challenges encountered. Despite these limitations, found that for a single supply, there could be multiple site specific our study is an important first step towards identifying opportunities or broader barriers to availability. This may explain the heterogen- and challenges for coaching to help impact supply availability. eity seen in how the intervention improved availability of an item in some but not all sites (for example, Vitamin K). However, des- Conclusion pite 6 months of coaching, less than one-third of intervention sites had all four critical medicines. This persistent gap highlights the Integrating a focus on supply availability into coaching resulted ini- need for broader supply chain remedies beyond coaching. We also tially in a modest increase in overall supply availability, with lower found that supplies which were either infrequently needed (nevira- performing sites experiencing greater difference in availability after 6 pine in a low HIV seroprevalence setting) or remained at low months of the intervention. While coaching can play a role in uptake in intervention sites (partograph) did not respond to the strengthening supply availability, further research is needed to intervention. understand how it can be combined with broader supply-chain inter- We did see some improvement in supply availability in control ventions to contribute to and sustain improvement in essential birth sites. It is possible that, while data collection visits were unannounced, supply availability in frontline facilities in resource-limited settings. the quarterly assessment of supplies improved awareness of the importance of supply status in control sites. Benefits to control sites Supplementary material may also have spread from the intervention due to district-level advo- cacy meetings with the Chief Medical Officer as some districts con- Supplementary material is available at International Journal for Quality in tained both control and intervention sites. There were also targeted Health Care online. state-wide programs focusing on health-facility assessments during the study that may have contributed to improvements such as Kaya Kalp, Acknowledgements a National Rural Health Mission initiative focused on infection control [29]. These programs were incorporated at intervention and We recognize the Governments of India and Uttar Pradesh for collaboration control sites. and support to conduct this trial in public health facilities. We thank the Downloaded from https://academic.oup.com/intqhc/article/30/10/769/4990396 by DeepDyve user on 20 July 2022 Coaching and birth supplies � Quality improvement 777 district and facility staff, women and their newborns for their participation in 16. Galvin G, Kalita T, Gupta S et al. Coaching to increase availability of the study. We are grateful to the members of the trials’ Scientific Advisory essential birth supplies for the WHO Safe Childbirth Checklist in Uttar Committee who contributed crucial guidance to the development of this study Pradesh, India. Proceedings of the Global Maternal and Newborn Health protocol: Himanshu Bhushan, Amit Kumar Ghosh, Zulfiqar Bhutta, Conference; 2015 Oct 18–21; Mexico City, Mexico. Waldemar Carlo, Vinita Das, Paul Francis, Amod Kumar, Matthews Mathai, 17. National Health Mission. Section VII: essential drugs and supplies. New Packirisamy Padmanbhan, Vinod Paul and Rajiv Tandon. We also thank the Delhi (India): Ministry of Health and Family Welfare, Government of past and current members of the BetterBirth study team in Boston and the India. Available at http://www.sifpsa.org/pdf/New/7-%20Essential% BetterBirth field team based in Uttar Pradesh for study implementation. 20Drugs%20and%20Supplies.pdf. Accessed [September 11 2017]. 18. Maiti R, Bhatia V, Padhy BM et al. Essential medicines: an Indian per- spective. Indian J Community Med 2015;40:223–32. 19. Patel S, Thumm M, Rahman J et al. Subnational procurement of maternal Funding health medicines: results from an assessment in Bangladesh. Arlington This work was generously funded through a grant from the Bill and Melinda (VA): Center for Pharmaceutical Management, Management Sciences for Gates Foundation. Health; 2014 May. Available at http://siapsprogram.org/publication/ subnational-procurement-of-maternal-health-medicines-results-from-an- assessment-in-bangladesh/. References 20. Mwencha M, Rosen JE, Spisak C et al. Upgrading supply chain manage- 1. Koblinsky M, Moyer CA, Calvert C et al. Quality maternity care for ment systems to improve availability of medicines in Tanzania: evaluation every woman, everywhere: a call to action. Lancet 2016;388:2307–20. of performance and cost effects. Glob Health Sci Pract 2017;5:399–411. 2. Kinney MV, Boldosser-Boesch A, McCallon B. Quality, equity, and dig- 21. Semrau KEA, Hirschhorn LR, Kodkany B et al. Effectiveness of the nity for women and babies. Lancet 2016;388:2066–8. WHO Safe Childbirth Checklist program in reducing severe maternal, 3. Sharma J, Leslie HH, Kundu F et al. Poor quality for poor women? fetal, and newborn harm in Uttar Pradesh, India: study protocol for a Inequities in the quality of antenatal and delivery care in Kenya. PLoS matched-pair, cluster-randomized controlled trial. Trials 2016;17:576. One 2017;12:e0171236. 22. Semrau KEA, Hirschhorn LR, Marx Delaney M et al. Outcomes of a 4. Mohanan M, Bauhoff S, La Forgia G et al. Effect of Chiranjeevi Yojana coaching-based WHO Safe Childbirth Checklist Program in India. N Engl on institutional deliveries and neonatal and maternal outcomes in J Med 2017;377:2313–24. Gujarat, India: a difference-in-differences analysis. Bull World Health 23. Marx Delaney M, Maji P, Kalita T et al. Improving adherence to essential Organ 2014;92:187–94. birth practices using the WHO Safe Childbirth Checklist with peer coach- 5. Randive B, Diwan V, De Costa A. India’s conditional cash transfer pro- ing: experience from 60 public health facilities in Uttar Pradesh, India. gramme (the JSY) to promote institutional birth: is there an association Glob Health Sci Pract 2017;5:217–31. between institutional birth proportion and maternal mortality? PLoS One 24. McCullagh P, Nelder JA. Generalized linear models, 2nd edn. Boca 2013;8:e67452. Raton (FL): CRC Press, 1989. 6. Powell-Jackson T, Mazumdar S, Mills A. Financial incentives in health: 25. Perry W, Bagheri Nejad S, Tuomisto K et al. Implementing the WHO new evidence from India’s Janani Suraksha Yojana. J Health Econ 2015; Safe Childbirth Checklist: lessons from a global collaboration. BMJ Glob 43:154–69. Health 2017;2:e000241. 7. Ministry of Health & Family Welfare, Department of Health and Family 26. Kumar S, Yadav V, Balasubramaniam S et al. Effectiveness of the WHO Welfare, India:. http://mohfw.gov.in/sites/default/files/9147562941489753121. SCC on improving adherence to essential practices during childbirth, in pdf Accessed [September 11 2017]. resource constrained settings. BMC Pregnancy Childbirth 2016;16:345. 8. Office of the Registrar General & Census Commissioner Vital Statistics 27. Nababan HY, Islam R, Mostari S et al. Improving quality of care for Division. Annual health survey 2012–13 fact sheet Uttar Pradesh. New maternal and newborn health: a pre-post evaluation of the Safe Delhi (India): Ministry of Home Affairs, Government of India; 2013. Childbirth Checklist at a hospital in Bangladesh. BMC Pregnancy 9. Office of the Registrar General & Census Commissioner. Sample registra- Childbirth 2017;17:402. tion system statistical report 2014. New Delhi (India): Government of 28. Kabongo L, Gass JD Jr., Kivondo B et al. Implementing the WHO Safe India; 2015. Childbirth Checklist: lessons learnt on a quality improvement initiative to 10. Sharma AK. National rural health mission: time to take stock. Indian J improve mother and newborn care at Gobabis District Hospital, Community Med 2009;34:175–82. Namibia. BMJ Open Qual 2017;6:e000145. 11. Programme Evaluation Organisation. Evaluation study of National Rural 29. National Health Mission. Award to public health facilities: Kayakalp. Health Mission (NRHM) in 7 states. 2011. New Delhi (India): New Delhi (India): Ministry of Health and Family Welfare, Government Government of India, Planning Commission; 2011 Feb. Available from: of India. Available at https://www.nhp.gov.in/kayakalp-swacchta- http://planningcommission.nic.in/reports/peoreport/peoevalu/peo_2807.pdf. guidelines-for-public-health-facilities_pg. Accessed [September 11 2017]. Accessed [September 11 2017]. 30. Issac A, Chatterjee S, Srivastava A et al. Out of pocket expenditure to 12. Harvard School of Public Health, World Health Organization, Population deliver at public health facilities in India: a cross sectional analysis. Services International. Better Birth: Rollout and evaluation of the WHO Reprod Health 2016;13:99. safe childbirth checklist. International Consultation Meeting; 2011 Dec 31. Nunan M, Duke T. Effectiveness of pharmacy interventions in improving 13–14; New Delhi, India. Geneva: World Health Organization; 2009. availability of essential medicines at the primary healthcare level. Trop 13. Kara N, Firestone R, Kalita T et al. The BetterBirth Program: pursuing Med Int Health 2011;16:647–58. effective adoption and sustained use of the WHO Safe Childbirth 32. Trap B, Todd CH, Moore H et al. The impact of supervision on stock Checklist through coaching-based implementation in Uttar Pradesh, management and adherence to treatment guidelines: a randomized con- India. Glob Health Sci Pract 2017;5:232–43. trolled trial. Health Policy Plan 2001;16:273–80. 14. Spector JM, Lashoher A, Agrawal P et al. Designing the WHO Safe 33. USAID Deliver Project, Task Order I. After receiving USAID/DELIVER Childbirth Checklist program to improve quality of care at childbirth. Int project logistics training, access to health products improves in Nepal. J Gynaecol Obstet 2013;122:164–8. Arlington (VA): Jon Snow Inc; 2009 July. Available at http://pdf.usaid. 15. Bahl KLC, Palmer A, Rosenberg T. Increasing access to essential maternal gov/pdf_docs/PA00KRRK.pdf. Accessed [September 11 2017]. health supplies: a scoping of market-based activities, gaps, and opportun- 34. Gass JD Jr., Misra A, Yadav MNS et al. Implementation and results of an ities. Washington DC: Results for Development Institute (R4D), integrated data quality assurance protocol in a randomized controlled Reproductive Health Supplies Coalition (RSC), 2016. March. trial in Uttar Pradesh, India. Trials 2017;18:418.

Journal

International Journal for Quality in Health Care – Oxford University Press

Published: Dec 1, 2018

Keywords: india; labor; world health organization; birth; oxytocin; arm; newborn; mothers; government; feedback

There are no references for this article.