Design. An ecological design with retrospective information collection was used to compare before and after implementation of a telemedicine system for patients with obstetric emergencies between two hospitals in southwestern Colombia. The study period before implementation, called “Period 1”, was between March 1st, 2017, and July 31st, 2018; while implementation of “Period 2” (when the teleconsultation process began) was between August 1st, 2018, and December 31st, 2019. The study protocol was approved by the FVL IRB on January 22, 2020 (approval number 1560).
FVL is a non-profit university hospital and is a reference medical care center for patients from the Southwest of Colombia, located in Santiago de Cali. The HFPS is located in Santander de Quilichao, approximately 1·5 hours driving distance from FVL. The hospital is ready to attend patients with medium complexity diseases in the urban, rural, and dispersed rural areas of 14 municipalities. However, this municipality has a MMR of 129 deaths per 100,000 LB [Ministry of Health and Social Protection (MSPS), 2016], almost doubling the MMR of the country in the same period[65 per 100,000 LB - National Health Institute (INS)].12,13 The near-miss mortality ratio (NMMR) of the territorial entity to which the hospital belongs is also higher than the average reported in the Colombian territory (52 per 1000 LB vs. 38 per 1000 LB), ranking among the six entities with the highest MMR in Colombia.14
Intervention model based on telehealth and education
The institutional factors that directly impact the MMR and the preventable NMMR in up to 90% of the events include the availability of qualified human resources and the logistic conditions of care that allow complete and adequate management. 15,16The use of telemedicine between highly complex hospitals and rural areas lacking specialists and environments with limited infrastructure can improve diagnosis, management, and patient outcomes.17,18 The adoption will depend on the acceptability of equipment, effectiveness, feasibility, use of resources, and indications for equity, gender, and rights. 9
Implementation of the model began with visits by the FVL medical group to determine the installed capacity of the HFPS for obstetric emergencies and the adoption of a telemedicine service. Then, both institutions conducted a concerted improvement plan based on quality policies in the care processes arranged for Colombia. 19 This plan included structuring a program with on-site education supported by simulation strategies and continuous monitoring by chat using the WhatsApp platform. The education modules included workshops from 6 to 12 hours held inside the hospital with interactive lectures and low-fidelity simulation modules supported by the institutional FVL protocols, emphasizing the use of communication strategies and non-technical skills.
The intervention group was made up of nursing assistants, nurses, general practitioners, and obstetrician-gynecologists who received the training program, accompanied by counseling for changing the behavior of the teams in the presence of critical events in pregnant women and the adoption of telehealth strategies. The trainers were obstetrician-gynecologists specializing in intensive care and had experience in face-to-face and virtual educational processes. The objectives of the modules were for prompt identification and management of safe delivery care, postpartum hemorrhage, hypertensive disorders of pregnancy, maternal sepsis, maternal cardiorespiratory arrest, and early neonatal resuscitation. Also, training was conducted using the MEOWS scale (Modified Early Obstetric Warning System)13 to detect clinical signs of deterioration in pregnant women with critical illnesses. 20 In each module, the FVL teams evaluated the assistance with pre and post-tests to establish the change in knowledge. Three re-training workshops and four follow-up visits were held during the intervention. The researchers organized group-specific training teleconferences at least once every three months to discuss implementation, compliance, and any difficulties, especially for telemedicine.
The FVL developed the telemedicine platform called Liliconnect, which allowed the legal exchange of information, consigning the clinical history of each patient for both institutions. Licensed communication platforms were used for telecalling, initially WebEX and later Microsoft Teams. It was established that the calls were attended by a nursing assistant and then transferred to the obstetrician-gynecologist specialized in intensive care, providing coverage 24 hours a day, every day of the week. To provide the care, the patients signed a digitized informed consent guarded at FVL. The costs derived from this process were assumed by the health system as established by Colombian regulations. The transfer of patients to the corresponding level of complexity for care was conducted according to their clinical condition (urgency or emergency), the health insurers’ networks, and the government for this purpose.
Population. A total of 250 patients with obstetric emergencies from the HFPS were included in this study, 102 before and 148 after the implementation of the model-based in telehealth and education. The criteria for the inclusion of patients for “Period 1” were pregnant women with obstetric emergencies from the HFPS referred for management in FVL. For “Period 2”, an additional inclusion criterion was attention in HFPS commented through the telemedicine service, and the patients were referred to FVL after the telecall. For both periods, the exclusion criteria were patients with obstetric emergencies in HFPS referred to other institutions (28 patients). The maternal near-miss criteria were defined according to the guidelines of the Colombian Ministry of Health.21
Variables. Information of the sociodemographic and clinical characteristics was collected, as well as maternal and perinatal outcomes. The variables were defined as follows:
Postpartum hemorrhage: a cumulative blood loss of greater than or equal to 500 mL in vaginal delivery or 1,000 mL or blood loss accompanied by signs or symptoms of hemodynamic instability. 21
Major surgery: procedures other than childbirth or cesarean section for the management of an obstetric complication or any condition generated as a consequence of a serious commitment of the woman. 22
Eclampsia: new onset of seizures or coma in a pregnant woman with preeclampsia.23
Hypertensive crisis: persistent (lasting 15 min or more), acute-onset, severe hypertension, defined as systolic BP greater than 160 mmHg or diastolic BP > 110 mmHg in the setting of pre-eclampsia or eclampsia. 24
Sepsis: a life-threatening condition defined as an organ dysfunction caused by an infection during pregnancy, de- livery, puerperium, or after an abortion. 25
Maternal near-miss mortality: When a woman nearly dies but survives a complication during pregnancy, childbirth, or within 42 days of termination of pregnancy. The criteria were defined according to the guidelines of the Colombian Ministry of Health.22,26
Maternal mortality: female deaths from any cause related to or aggravated by the pregnancy or its management (excluding accidental or incidental causes) during pregnancy and childbirth or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy. 4
Perinatal mortality: number of fetal deaths past 22 completed weeks of pregnancy plus the number of deaths among live-born children up to 7 completed days of life per 1000 total births (live births and stillbirths). 27
MEOWS at admission: Modified Early Obstetric Warning System composed of physiological parameters with a predetermined threshold that determines evaluation, treatment, or intervention.
Medical conditions: Chronic conditions, such as hypertension, preexisting diabetes mellitus, rheumatologic diseases, renal disease, etc.28
The source of information was the FVL medical history, and the history registered in Liliconnect
Statistical analysis. The unit of analysis was the HFPS. The normality of the data was tested using the Shapiro-Wilk and Kolmogorov-Smirnov tests. A descriptive analysis of the variables was expressed by percentages and absolute frequencies for qualitative variables, means, and standard deviation for quantitative variables with normal distribution and medians and interquartile ranges (IQR) for those not normally distributed. For comparisons of sociodemographic, pregnancy, and clinical characteristics between the two periods, the Mann-Whitney U test and the chi-square or Fisher´s exact tests were conducted. A p-value of < 0·05 was considered statistically significant. Multiple logistic regression estimated the effect adjusted of telemedicine program on the outcome of maternal. This variable indicated if a woman's pregnancy presented PPH, needs for blood and blood components, needs of major surgery for PPH, eclampsia, hypertensive crisis, or maternal near-miss. A multivariate skewed logistic regression estimated OR adjusted of telemedicine program on perinatal mortality. For both regressions, the covariables were: type of insurance, area of origin, occupation, parity (< 3 or > 3), and medical conditions, specifically for perinatal mortality were added the MEOWS at admission FVL, Vaginal delivery, Cesarean section, and outcome maternal. The selection of variables did make with the backward method. We adjusted p values for multiplicity with the false discovery rate (FDR) method. The statistical package used was Stata v.14 (StataCorp LLC, College Station, Texas, USA).