Oxygen delivery and monitoring in neonatal intensive care units in Mexico in 2011 and in 2023: an observational longitudinal study
verfasst von:
Luz Consuelo Zepeda Romero, Daniel Perez Rulfo Ibarra, Juan Carlos Barrera De Leon, Valeria Alejandra Salas Cruz, David Blanco Bouzo, Jose Alfonso Gutierrez Padilla, Clare Gilbert
Retinopathy of prematurity (ROP) is a leading cause of avoidable blindness in children, particularly in Latin America, where hyperoxia is a significant risk factor. This study evaluated resource availability and use for administering and monitoring supplemental oxygen in Mexico.
Methods
In 2011, an observational study in which 32 government neonatal intensive care units (NICUs) across Mexico were visited. Data collected included occupancy, staffing levels, and equipment to deliver and monitor supplemental oxygen. Preterm infants receiving oxygen were observed. In 2023, 13 NICUs were revisited, and similar data collected. Staffing levels were benchmarked against Argentinian and US recommendations.
Results
In 2011, only 38% of NICUs had adequate medical and staffing levels to meet recommended cot-to-staff ratios for all shifts. Staffing ratios were worse during weekends and at night than during weekdays. Only 25.5% of cots had blenders, and 80.1% had saturation monitors. 153 infants were observed 87% of whom were being monitored. Upper and lower oxygen saturations were ≥ 96% in 53%, and ≤ 89% in 8%, respectively. Alarm settings were inadequate, as 38% and 32% of upper and lower alarms were switched off and 16% and 53% were incorrectly set, respectively. In the 13 NICUs with data from 2011 and 2023, cot-to-staff ratios deteriorated over time, and in 2023 no unit had recommended ratios for all shifts. Equipment provision did not change, with similar proportions of babies in oxygen being monitored (79% 2011; 75% 2023). Rates of hyperoxia decreased slightly from 54% in 2011 to 49% in 2023. More upper alarms were set (46% 2011; 75% 2023), but a higher proportion were incorrectly set (52% 2011; 68% 2023).
Conclusions
Between 2011 and 2023, cot-to-staff ratios worsened, and equipment for safe oxygen delivery and monitoring remained insufficient. Despite available monitoring equipment, oxygen saturations often exceeded recommended levels, and alarms were frequently not set or incorrectly configured. Urgent improvements are needed in healthcare workforce numbers and practices, along with ensuring adequate equipment for safe oxygen delivery.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
Survival of neonates has improved considerably over the last few decades in Mexico due to a combination of public health policies and programs to improve maternal and child care, focusing on prevention, maternal nutrition, and improvements in quality of life and education [1]. In 2007, the inclusion of neonatal intensive care in the Seguro Popular Insurance (a public health insurance scheme), was a significant factor. Neonatal survival of babies born at 27–29 week’s gestational age (GA) is now 74%, and 90% for babies born at 30–33 weeks. Unfortunately, this progress is associated with high rates of major disability, including blindness from retinopathy of prematurity (ROP), which affects 25% and 19% of survivors in these GA groups, respectively [2].
The quality of neonatal care tends to be higher in high-income countries than in middle-income countries like Mexico. This is primarily due to the greater availability of sophisticated technology, specialized staff, and robust healthcare infrastructure and systems in high-income settings. Many middle-income countries, including Mexico, face challenges that result in variation in the quality of neonatal care between neonatal intensive care units (NICUs) [3]. As a result, many preterm babies are exposed to potentially modifiable risk factors for ROP [4]. In the Latin American region, the proportion of preterm infants who develop Type 1 ROP is 14–20%, which is higher than in high-income settings [5]. Additionally, many of these infants have a GA of > 32 weeks and a birth weight (BW) of > 1500 g, resembling the characteristics of children who became blind from ROP during the first epidemic which occurred as a result of expose to unregulated 100% supplemental oxygen in the 1940s and 50s [6].
Anzeige
Optimum target oxygen saturations for preterm infants have been a matter of considerable debate, leading to a series of multicentre randomized clinical trials. The trials compared higher (91–95%) and lower (85–89%) target oxygen saturation levels in infants born at a GA of less than 28 weeks, with death or major morbidity as a composite outcome [7]. A meta-analysis by the NeOProM Collaboration of the 4,965 infants enrolled showed no significant difference between the lower and upper SpO2 target ranges on the primary composite outcome of death or major disability at a corrected age of 18–24 months [5, 6]. However, the lower saturation target was associated with lower rates of treatment requiring ROP but higher mortality rates.
Following these studies, most high resource settings now recommend an SpO2 target of 90–94% with the upper alarm set at 95% to avoid hyperoxia which increases the risk of bronchopulmonary dysplasia as well as ROP [8, 9]. However, it is unclear how the findings of the meta-analysis translate to middle- and low-income countries (LIMICs) where many babies with treatment-requiring ROP have a GA of more than 28 weeks, as there have only been a few observational studies from Latin America, for example [9‐12]. Nevertheless, the World Health Organization (WHO) recommends that the SpO2 should not be lower than 89% or higher than 94% for preterm babies with a GA of less than 32 weeks. This means that the lower alarm should be set at 88% and the upper at 95% [13].
The safe delivery of oxygen for preterm infants requires considerable resources in terms of equipment to mix, warm, humidify and deliver oxygen, and to measure and monitor oxygen saturations. Neonatal staff need to be aware of the potential harms of saturations that fall outside the target range (i.e. death, cerebral palsy and ROP) and require training and support to respond appropriately when the alarms are activated. The purpose of this study was to assess oxygen delivery and monitoring in NICUs in Mexico, focusing on public sector facilities in large cities in the largest States.
Methods
In this longitudinal observational study, five States in the Mexican Republic were purposively selected based on population size: Mexico State, Puebla, the Federal District where the capital city is located, Jalisco and Chiapas. (Veracruz State was not included due to flooding.) The five states selected account for 37% of the population (Fig. 1) and almost 40% of births. A list of NICUs in the public sector in each state was obtained from the Ministry of Health, and ranked according to the number of incubators (data available from the Ministry of Health website), as a proxy for the number of high dependency admissions [14].
Anzeige
In each state, NICUs in cities with the largest number of incubators were purposively selected. Thirty-five NICUs were selected, two of whom declined, and one was a short stay unit. The 32 NICUs were visited between June and August 2011.
Fig. 1
Map of Mexico showing States included in the study
×
After explaining the study and obtaining written informed consent, a senior member of the neonatal staff was interviewed in each NICU using a structured questionnaire developed by the research team (Additional file 1). The questions covered the characteristics of each unit, such as the total number of admissions in 2010 and bed occupancy, survival rates and complications among infants with BWs < 1,500 g, resuscitation practices, number and qualification of staff working on the NICU, as well as the availability of equipment to deliver and monitor oxygen. All the nurses on the unit at the time of the visit who were willing to participate were given a self-administered, anonymous questionnaire to complete. The questionnaire had three sections; the first focused on the nurses’ qualifications followed by an assessment of their knowledge of complications of prematurity, and oxygen delivery and saturation targets. Lastly, participants were asked about their training hours, courses they had attended, and opportunities for further professional development.
In each unit, preterm babies receiving supplemental oxygen were identified and observed. The equipment being used to deliver oxygen was recorded (blender, compressed air) and method of delivery (CPAP, ventilation etc.). Oxygen monitoring was assessed by observing whether a pulse oximeter was being used, and if so, the saturation levels and upper and lower alarm settings (i.e., not set, correctly or incorrectly set).
In January 2023, 13 of the larger NICUs were revisited to assess the impact of administrative changes within the Mexican healthcare system over the previous decade, after excluding three NICUs with adequate oxygen management in 2011. During the second visit, data collection focused on staffing levels, equipment for oxygen delivery and monitoring, and oxygen saturations and alarm settings in preterm infants receiving supplemental oxygen.
Bench-marking
Mexican norms for neonatal care do not specify the number or qualifications of staff; norms from Argentina on the regulation of neonatology services and US National Association of Neonatal Nurses were used [15‐17]. Cots: staff ratios for nursing and medical staff for shifts at different times of the day and week were calculated. The following shifts were used: weekday mornings, afternoons and nights, and weekend days and nights. Levels of staffing were scored as follows: at or above recommended levels (green) score = 3; below recommendations (yellow) score = 2 and well below recommendations (red) score = 1.
Data management
Data were entered into a database created in Epi-Info 7 (Center for Disease Control and Prevention, www.cdc.go/epiinfo) and transferred into STATA 11.1 [18] for descriptive analysis. For descriptive analysis, frequency and percentages were used. The Kolmogorov-Smirnov test was used to assess distribution curves. Means and standard deviations were used for normally distributed data, and medians and ranges for skewed data. Student’s t-tests were used to compare means, and Mann-Whitney U tests to compare medians. The Chi-square test was used to compare proportions. A significance level of p < 0.05 was considered statistically significant.
Results
Findings in 2011
Characteristics of the NICUS visited in 2011
Most of the 32 hospitals and NICUs were financed by the Secretary of Health and Assistance, and three were financed through other mechanisms. The NICUs were located in general hospitals (n = 18), maternity hospitals (n = 8) or pediatric hospitals (n = 6). Twelve NICUs were level II and 20 were level III (Table 1). The total number of high dependency cots was 369 (mean 11.5, range 4–22). Four NICUs had 50% or lower occupancy at the time of the visit, in 12 occupancy was 50–99%; in 13 it was 100% and exceeded 100% in three units. There was no difference in survival rates for preterm babies with a birthweight of < 1500 g between level II and level III units (level II: mean 69.4% (range 50–85), median 71% (SD ± 11.4%); level III mean 71.9% (range 48–93), median 72% (SD ± 12.4%), U Mann Whitney for medians p = 0.72).
Table 1
Characteristics of hospitals and NICUs and bed occupancy in 32 neonatal units in 2011. Shaded NICUs [13] were revisited in 2023
State code
NICU code
Health
provider
Type of hospital
Type of deliveries
Unit Level
Infants admitted
Cots
(n)
Occupancy (%)
State A
1
SSA
General
N/high risk
3
N/high risk
9
100
State A
2
ISSSTE
General
N/high risk
2
N/high risk
15
53
State A
3
SSA
Paediatric
None
3
High risk
8
100
State A
4
SSA
Maternity
N/high risk
3
N/high risk
14
100
State A
5
SSA
Maternity
N/high risk
3
N/high risk
14
100
State A
6
SSA
General
High risk
2
Normal
6
117
State A
7
SSA
General
N/high risk
2
N/high risk
10
60
State A
8
SSA
Maternity
High risk
3
Normal
19
53
State B
9
SSA
General
None
2
High risk
11
45
State B
10
SSA
Maternity
N/high risk
3
Normal
17
82
State B
11
SSA
General
N/high risk
2
N/high risk
20
45
State B
12
SSA
Paediatric
None
3
High risk
7
100
State C
13
SSA
Maternity
High risk
3
Normal
16
100
State C
14
SSA
Paediatric
None
3
High risk
11
92
State C
15
SSA
General
N/high risk
3
N/high risk
8
75
State C
16
SSA
Maternity
Normal
2
Normal
15
107
State C
17
SSA
Maternity
High risk
3
Normal
16
88
State C
18
SSA
General
Normal
2
Normal
8
100
State C
19
SSA
General
Normal
3
Normal
14
100
State C
20
SSA
Paediatric
None
3
High risk
4
50
State C
21
SSA
Paediatric
None
2
High risk
11
79
State D
22
SSA
General
None
3
High risk
22
82
State D
23
SSA
General
N/high risk
2
Normal
18
100
State D
24
SSA
General
N/high risk
3
Normal
10
100
State D
25
SSA
General
N/high risk
3
Normal
8
88
State D
26
SSA
Maternity
Normal
3
N/high risk
18
78
State E
27
IMSS
General
N/high risk
3
N/high risk
8
100
State E
28
ISSSTE
General
N/high risk
2
N/high risk
4
50
State E
29
SSA
General
Normal
2
Normal
6
100
State E
30
SSA
General
Normal
2
Normal
6
83
State E
31
SSA
General
N/high risk
3
N/high risk
8
163
State E
32
SSA
Paediatric
None
3
High risk
8
100
SSA = Secretaria de Salud y Asistencia (Secretary of Health and Assistance); IMSS = Instituto Mexicano del Seguro Social (Mexican Social Security Institute); ISSSTE = Instituto de Seguridad Social y Servicios para los trabajadores del Estado (Institute of Social Security and Services for State Workers); N/High risk = normal and high risk deliveries / infants. Bold font = neonatal units revisited in 2023
Staffing levels
Physicians
There were 160 shifts across the week in the 32 units. Staffing levels of physicians (pediatricians or neonatologists) were better than for nursing (Fig. 2).
Recommended staffing levels for physicians is 8 cots / physician. For all the shifts, 66% were at or above recommended levels, 19% were below and 16% were well below. Eleven NICUs (34%) had the required number of physicians for all shifts, and eight (25%) NICUs had ratios that were inadequate for three or more of the five shifts. Staffing levels did not differ between level II and III units (median 7.2 (6-10.5) and 8 (6.2–12.7) p = 0.526, respectively).
Fig. 2
Nursing and medical staffing levels (number of cots per member of staff) by type of unit (level II or III) in 2011
×
Average cots: physician ratios varied by shift. During weekday mornings the average was 6:1 (range 2–11:1) and 9:1 (range 3–20:1) during weekday nights. During weekends the ratio was 8:1 (range 2.6–22) during the day and 9:1 (range 3-22) during the night.
Nurses
Units were staffed by nurses trained in neonatology (specialist nurses), general and auxiliary nurses. Half (54%) of the shifts had overall nursing levels at or above recommended levels (i.e., 2 cots/nurse), 13% were below and 33% were well below recommendations. Only 11 (34%) NICUs had the recommended ratio for all shifts and in 15 (47%) ratios were inadequate for three or more of the five shifts (Fig. 2).
Although the cots: nurse ratios were often 2:1, as recommended, there was considerable variation by shift. The range during weekday mornings was 1-2.7 and during weekday nights the range was 0.7–3.3. The worst ratio, range 0.7-4, was during the night at weekends when 16 NICUs had ratios of one nurse to 2.5 or more cots. There was no significant difference in nursing levels between level II and III units; median 12 (interquartile range 8.25-15) and 9.50 (interquartile range 6.25-15), respectively (U Mann Whitney for medians, p = 0.60).
In most units less than 30% of nurses were specialized nurses, and 5 units did not have any trained nurses, all nursing staff being auxiliaries and general nurses. In four (13%) NICUs the proportion of neonatal nurses was below recommendations, and in 27 (84%) NICUs it was well below recommendations.
Nurses’ knowledge of ROP
Anonymous questionnaires were administered to 152 nurses. Half of the nurses had less than three years of experience in neonatal care; 53% were general nurses, 29% were certified or specialist nurses, 5% were auxiliary nurses, 7% did not specify and 6% were students. However, their salary scale did not always reflect their qualifications as 43% were paid as general nurses, 34% as auxiliary nurses and 10% as specialist nurses. Half (50%) of the nurses had attended less than two hours of formal teaching in the previous year and 28% had attended 20 or more hours of teaching. A third (35%) said they had not been given any opportunity for training; 49% said they would welcome additional professional development and 20% would value training in neonatal intensive care.
Over a third (37%) of the nurses knew that ROP was a major complication of prematurity and a third (32%) said that they were aware of the Mexican ROP technical guidelines. Most (87%) knew that high and low oxygen saturation levels were harmful, and 68% knew that monitors have an upper limit alarm. Additionally, 42% described the correct upper limit alarm for oxygen saturation, and 66% identified the lower limit.
Equipment to deliver and monitor supplemental oxygen
Over half of the units (18/32, 56%) did not have equipment to blend air and oxygen, and only six (19%) had enough blenders for all babies receiving supplemental oxygen. Over half the units (56%) had enough equipment to monitor oxygen saturation in all babies receiving oxygen; in eight units, up to 50% of babies could be monitored, and in the remaining six < 50% could be monitored. There was no significant difference between level II and level III units in terms of their ability to blend air and oxygen (median equipment score 3; range 0.8-4 for level II and III units), nor in equipment for monitoring oxygen saturation levels: level II (median equipment score 2.51 ± SD 0.95; range 0.8-4) vs. level III (median 3.0 ± SD 0.93; range 0.87-4) (p = 0.894). Protocols on safe oxygen saturation levels were present in four of the 12 level II NICUs and in five of the 20 level III units (27% overall)(p = 0.207).
Oxygen saturation levels
153 preterm babies aged less than 30 days receiving oxygen were observed, 20 (13%) of whom were not being monitored. Overall, only 39% of babies had oxygen saturations within the range recommended at the time (i.e., 89–94%), and in 53% saturations were ≥ 95% (Fig. 3A).
Fig. 3
Oxygen saturations in babies observed, by level of neonatal care A. All 32 units in 2011, B. 13 units revisited in 2023 and C. data for the same 13 units first visited in 2011
×
Level III units had a higher proportion of infants with saturations within the range recommended than level II units, but this was not statistically significant (Chi-sq test, p = 0.07).
In 7 of the 18 units with adequate equipment to monitor oxygen saturation, none of the infants had oxygen saturations of ≥ 95%. In 3 NICUs, all infants had oxygen saturations of ≥ 95%.
Alarm settings
Alarm settings and oxygen saturation levels were available for 133 babies being monitored (Fig. 4A). In half the units, both alarms were switched on, but in 11 (34%) both were switched off. If only one alarm was on, the lower alarm was more likely to be switched on than the upper alarm. In the majority of NICUs, the upper and lower alarms were not correctly set; both alarms were switched on and set correctly in only one level III unit. In 24 units, the upper alarm was set at 100%; in eight units, the lower alarm was set at 80%, and the majority of other lower alarms were set at 84% or 85%.
Fig. 4
Monitor alarms settings in babies observed A All 32 units in 2011, B 13 units revisited in 2023 and C data for the same 13 units first visited in 2011
×
Findings in 2023
Characteristics of the NICUS visited
The 13 NICUs revisited were financed by the Secretary of Health and Assistance (see bold font in Table 1). The NICUs were located in general hospitals (n = 6) and maternity wards (n = 7). Three NICUs were level II and ten were level III. The total number of high-dependency cots was 241 (mean 18.8, range 4–35). Two NICUs had 100% occupancy and four exceeded (range 123–150%) capacity.
Staffing levels
Cots: staff ratios were worse in 2023 than in 2011 for physicians and nurses for all five shifts across the week (Fig. 5).
Fig. 5
Cots: staff ratios in 13 NICUs in 2011 and 2023 for A physicians and B nurses
×
Oxygen delivering, saturations and alarm settings
In 2023, 145 babies were observed; 97 had been observed in the same 13 NICUs in 2011. The proportion of infants being monitored was similar (79% in 2011; 75% in 2023). Oxygen saturations were similar in terms of the proportion with saturations within the recommended range (39% in 2011; 44% in 2023)(Fig. 3B and C). The proportion of monitors with activated alarms had improved and the proportion with correctly set lower alarms increased from 0% in 2011 to 7% in 2023 (Fig. 4B and C).
Discussion
The prevention of Type 1 ROP requires high quality neonatal care from immediately after birth, which includes avoiding hyperoxia and fluctuating hyperoxia / hypoxia [19]. The safe delivering of oxygen requires supplies of compressed air and oxygen, adequate equipment which is well maintained, and an adequate number of staff who know about the risk of hyperoxia as well as hypoxia. Staff need to be skilled in using the equipment and setting the alarms, monitoring oxygen saturations and responding appropriately when the alarms are triggered.
Anzeige
In this study, the majority of neonatal nurses in some of the largest NICUs in Mexico were aware that hyperoxia as well as hypoxia are harmful, and although two thirds knew the lower saturation alarm limit, less than half knew the upper safe limit. There was a marked lack of equipment to safely deliver oxygen, which included a lack of air-oxygen blenders, as well as equipment to monitor oxygen saturations. When monitoring equipment was available, both alarms were switched off in one third of the infants observed. Both alarms were switched on and correctly set in only one level III unit. These findings may, in part, explain why 15% of the infants observed who were receiving supplemental oxygen were not being monitored at all, and in over half of those being monitored, oxygen saturations were ≥ 95%. We are not aware of any other studies which have observed oxygen saturation levels in preterm infants receiving supplemental oxygen, but a study in India also showed that upper alarms were switched off or were incorrectly set in 94% of infants; a higher proportion had lower alarms correctly set and switched on, as in our study [20]. These findings suggest that nursing staff are motivated to reduce the risk of hypoxia.
In 2023 oxygen saturations were not very different from those in 2011, but more alarms were switched off or were not correctly set. This finding is surprising, given that there was less equipment and graduate nurses in 2023 compared with 2011. One interpretation is that there was a better understanding of the need to control oxygen levels in 2023 compared with 2011.
Low nurse: baby ratios and training
Factors which are likely to have contributed to these findings included overcrowded units which would worsen cots: staff ratios. In 2011, the number of qualified and specialist nurses was also below recommended levels, and half of the nurses interviewed had received less than two hours of formal training in neonatal care in the previous year.
Low nurse: baby ratios in NICUs are not limited to low resource settings and have been reported in the United Kingdom [21] and the United States of America [22], for example. Low numbers of nurses and inadequate training are recognised bottlenecks in the delivery of neonatal care in low resource settings [23]. Under staffing has been associated with delayed or missed care practices [21], higher rates of nosocomial infection [22], inpatient mortality [24] and poorer patient care and counselling of parents [22]. Given the variation in case mix between level II and level III NICUs, as well as within level III units, there is considerable debate concerning optimal staffing levels and the skills mix required [21]. Overcrowding, as is often the case in low- and middle-income settings, increases demands on nursing staff and can also increase infection rates [25]. In low resource settings recruiting and retaining nursing staff is challenging, particularly in rural areas, which is compounded by a lack of competency-based training and refresher training, and low motivation [23]. In our study the number of graduate nurses was lower in 2023 than in 2011 in a significant proportion of NICUs. Although reasons for this were not explored, poor remuneration and hospital policies to rotate staff to different services, such as pediatric emergency care, may have been contributory factors.
Anzeige
In our study, less than one in three nurses knew that ROP was a potential complication of preterm birth. Similar findings have been reported from other low resource settings [20, 26]. For example, in a study in a government hospital in Kerala, India, only 48% of nurses had a diploma in nursing; >50% had up to 5 years of experience in paediatric care and 73% also worked in other areas in the hospital. Half of the nurses had some knowledge of ROP but only a quarter had a good level of knowledge and were able to identify optimal oxygen target saturations to prevent it [27].
Lack of equipment and protocols
In our study, lack of equipment limited the ability of staff to adequately deliver and monitor supplemental oxygen, a frequent finding in NICUs in low- and middle-income countries [20, 28, 29]. In addition, most units did not have written protocols on oxygen delivery. In 2023 the availability of equipment to deliver and monitor oxygen was worse than in 2011 in almost 40% of the units revisited.
Improving the quality of neonatal care is challenging in low- and middle-income countries, where there is less emphasis on teamwork and where perinatal services are less well organized than in high income countries. [30] However, several quality improvement initiatives have demonstrated that packages of evidence-based practices delivered primarily by nurses can reduce neonatal mortality, [31] improve the outcomes of neonatal care, [32] lead to a more rational use of supplemental oxygen, [28, 33] and less exposure to hyper- or hypoxia [34]. An initiative in Lima, Peru, which entailed workshops for neonatal nurses on oxygen management, led to a lower incidence of treatment requiring ROP in three of the six NICUs included [11]. Training nurses in the optimal use of oxygen in Mexico City also led to a reduction in aggressive ROP (personal communication).
ROP has a multifactorial etiology and although there has rightly been a focus on oxygen management, other aspects of perinatal care are also important. These include provision of antenatal steroids when preterm delivery is threatened; [35] prevention of hypothermia in labour wards, control of sepsis and improved nutrition including the provision of mothers’ breast milk, [36] reducing pain, and providing supportive care, including kangaroo care. Oral caffeine, which is usually given to prevent or treat apnea of prematurity, is one of the few therapies shown in high quality randomized controlled trails (RCTs) to reduce the incidence of severe ROP [37].
Anzeige
Some of the changes between 2011 and 2023 regarding staffing levels and equipment, and therefore oxygen management, are likely to reflect political and regulatory changes in Mexico during the intervening period. In 2003, Mexico instituted a national health insurance plan, Seguro Popular, which provided access to a package of comprehensive health services with financial protection for more than 50 million people [38]. Neonatal care was included, and by 2011, the infant mortality rate had declined to 15/1000 live births. By 2012, Mexico was on track to achieve universal health coverage [39]. However, in 2019, the new government cancelled Seguro Popular as part of an anti-corruption campaign, which led to sudden and drastic cuts of 30–40% in spending on government health services. The cuts included terminating all contracts with third parties, such as contracting agencies which provided staff to government facilities, and the suspension of payments to non-governmental organizations to provide services, such as HIV testing and management. Time will tell if the cuts lead to reforms that strengthen the health system, including the care of newborns.
Limitations of this study include the selection of large NICUs located in the most densely populated capitals of the country, potentially limiting generalizability of the findings to smaller NICUs and less populated states. In addition, nurses’ knowledge of oxygen management and ROP was not assessed in 2023.
Conclusions
Despite government guidelines, this study highlights considerable challenges related to the safe delivery of oxygen, including overcrowded units, low nursing levels and a lack of equipment. Nurses also demonstrated inadequate knowledge of safe oxygen management, which may contribute to poor practices.
Neonatologist and paediatricians need to assume team leadership roles, by ensuring they are up to date on best practices, that protocols for oxygen management are established and implemented, and that nursing staff receive ongoing professional development and support. Advocacy is required with health care providers and hospital administrators for better nurse training, with refresher training and clear career pathways, and to reduce the rotation of skilled neonatal staff to other services. Advocacy will also be needed to ensure that neonatal units are adequately equipped to deliver oxygen and to monitor oxygen saturations.
Preventing ROP by improving the quality of neonatal care would be cost-efficient, avoiding the costs of treatment and the lifelong expenses associated with blindness. Without sustained policies and practices which improve neonatal care, ROP will remain the main avoidable cause of blindness in children in Mexico.
Acknowledgements
We thank the hospitals that assisted us in data collection.
Declarations
Ethics approval and consent to participate
The study was approved by the ethics committee of the London School of Hygiene & Tropical Medicine and by the Ministry of Maternal and Child Health in Mexico. No names were collected, and a unique study number was allocated to each NICU, the heads of each NICU authorized the collection of data. In each of the NICUs a written informed consent form was obtained.
Consent for publication
Not applicable.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Oxygen delivery and monitoring in neonatal intensive care units in Mexico in 2011 and in 2023: an observational longitudinal study
verfasst von
Luz Consuelo Zepeda Romero Daniel Perez Rulfo Ibarra Juan Carlos Barrera De Leon Valeria Alejandra Salas Cruz David Blanco Bouzo Jose Alfonso Gutierrez Padilla Clare Gilbert