Patterns, procedures, and indications for pediatric surgery in a Tanzanian Refugee Camp: a 20-year experience.

Background: There are 103 million displaced people worldwide, 41% of whom are children. Data on the provision of surgery in humanitarian settings are limited. Even scarcer is literature on pediatric surgery performed in humanitarian settings, particularly protracted humanitarian settings. Methods: We reviewed patterns, procedures, and indications for pediatric surgery among children in Nyarugusu Refugee Camp using a 20-year retrospective dataset. Results: A total of 1221 pediatric surgical procedures were performed over the study period. Teenagers between the ages of 12 and 17 years were the most common age group undergoing surgery (n=991; 81%). A quarter of the procedures were performed on local Tanzanian children seeking care in the camp (n=301; 25%). The most common procedures performed were cesarean sections (n=858; 70%), herniorrhaphies (n=197; 16%), and exploratory laparotomies (n=55; 5%). Refugees were more likely to undergo exploratory laparotomy (n=47; 5%) than Tanzanian children (n=7; 2%; p=0.032). The most common indications for exploratory laparotomy were acute abdomen (n=24; 44%), intestinal obstruction (n=10; 18%), and peritonitis (n=9; 16%). Conclusions: There is a significant volume of basic pediatric general surgery performed in the Nyarugusu Camp. Services are used by both refugees and local Tanzanians. We hope this research will inspire further advocacy and research on pediatric surgical services in humanitarian settings worldwide and illuminate the need for including pediatric refugee surgery within the growing global surgery movement.

Patterns of referral for refugees in western Tanzania: a retrospective review.

Introduction: access to essential secondary and tertiary healthcare, including surgery and medical sub-specialties, is a challenge in low-and-middle income countries (LMICs), especially for displaced populations. Referrals from refugee camps are highly regulated and may pose barriers to accessing essential secondary healthcare in a timely manner. Refugee referral systems and the ways they interact with national systems are poorly understood. Such information is necessary for resource allocation and prioritization, optimizing patient outcomes, national-level planning, and investment in capacity-building. Methods: a retrospective review of referrals from Nyarugusu Refugee Camp in Tanzania to Kabanga Hospital between January 2016-May 2017 was conducted. Data was collected from logbooks on patient demographics, diagnosis, and reason for referral. Diagnoses and reasons for referral were further coded by organ system and specific referral codes, respectively. Results: there were 751 entries in the referral logbook between January 2016 and May 2017. Of these, 79 (10.5%) were excluded as they were caretakers or missing both diagnoses and reason for referral resulting in 672 (89.5%) total entries for analysis.The most common organ system of diagnosis was musculoskeletal (171, 25.5%) followed by head, ear, eye, nose and throat (n=164, 24.4%) and infectious disease (n=92, 13.7%). The most common reason for referral was imaging (n=250, 37.2%) followed by need for a specialist (n=214, 31.9%) and further management (n=116, 17.3%). X-ray comprised the majority of imaging referred (n=249, 99.6%). The most common specialties referred to were ophthalmology (n=104, 48.6%) followed by surgery (n=63, 29.4%), and otolaryngology (ENT) (n=17, 7.9%). Conclusion: given a large burden of referral for refugee patients and sharing of in and out-of-camp healthcare facilities with nationals, refugees should be included in national health care plans and have clear referral processes. Epidemiological data that include these intertwined referral patterns are necessary to promote efficient resource allocation, reduce unnecessary referrals, and prevent delays in care that could affect patient outcomes. International agencies, NGOs, and governments should conduct cost analyses to explore innovative capacity-building projects for secondary care in camp-based facilities.

Humanitarian surgical service utilization by a host country population: comparing surgery patterns between refugees and Tanzanians using an interrupted time-series analysis.

BACKGROUND: While current estimates suggest that up to three million additional surgical procedures are needed to meet the needs of forcibly displaced populations, literature on surgical care for refugee or forced migrant populations has often focused on acute phase and war-related trauma or violence with insufficient attention to non-war related pathologies. To our knowledge, no study has compared refugee versus host population utilization of surgical services in a refugee camp-based hospital over such an extended period of twenty years. The aim of this paper is to first describe the patterns of surgical care by comparing refugee and host population utilization of surgical services in Nyarugusu refugee camp between 2000 and 2020, then evaluate the impact of a large influx of refugees in 2015 on refugee and host population utilization. METHODS: The study was based on a retrospective review of surgical logbooks in Nyarugusu refugee camp (Kigoma, Tanzania) between 2000 and 2020. We utilized descriptive statistics and multiple group, interrupted time series methodology to assess baseline utilization of surgical services by a host population (Tanzanians) compared to refugees and trends in utilization before and after a large influx of Burundian refugees in 2015. RESULTS: A total of 10,489 operations were performed in Nyarugusu refugee camp between 2000 and 2020. Refugees underwent the majority of procedures in this dataset (n = 7,767, 74.0%) versus Tanzanians (n = 2,722, 26.0%). The number of surgeries increased over time for both groups. The top five procedures for both groups included caesarean section, bilateral tubal ligation, herniorrhaphy, exploratory laparotomy and hysterectomy. In our time series model, refugees had 3.21 times the number of surgeries per quarter at baseline when compared to Tanzanians. The large influx of Burundian refugees in 2015 impacted surgical output significantly with a 38% decrease (IRR = 0.62, 95% CI 0.46-0.84) in surgeries in the Tanzanian group and a non-significant 20% increase in the refugee group (IRR = 1.20, 95% CI 0.99-1.46). The IRR for the difference-in-difference (ratio of ratios of post versus pre-intervention slopes between refugees and Tanzanians) was 1.04 (95% CI 1.00-1.07), and this result was significant (p=0.028). CONCLUSIONS: Surgical care in conflict and post-conflict settings is not limited to war or violence related trauma but instead includes a large burden of obstetrical and general surgical pathology. Host population utilization of surgical services in Nyarugusu camp accounted for over 25% of all surgeries performed, suggesting some host population benefit of the protracted refugee situation in western Tanzania. Host population utilization of surgical services was apparently different after a large influx of refugees from Burundi in 2015.

Informing prehospital care planning using pilot trauma registry data in Yaoundé, Cameroon.

INTRODUCTION: About 54% of deaths in low- and middle-income countries (LMICs) are attributable to lack of prehospital care. The single largest contributor to the disability-adjusted life years due to poor prehospital care is injury. Despite having disproportionately high injury burdens, most LMIC trauma systems have little prehospital organization. An understanding of existing prehospital care patterns in LMICs is warranted as a precursor to strengthening prehospital systems. METHODS: In this retrospective pilot study, we collected demographic and injury characteristics, therapeutic itinerary, and transport data of patients that were captured by the trauma registry at the Central Hospital of Yaoundé (CHY) from April 15, 2009 to October 15, 2009. Bivariate and multivariate regression analyses were used to explore relationships between care-seeking behavior, method of transport, and predictor variables. RESULTS: The mean age was 30.2 years (95% CI [29.7, 30.7]) and 73% were male. Therapeutic itinerary was available for 97.5% of patients (N = 2855). Nearly 18.7% of patients sought care elsewhere before CHY and 82% of such visits were at district hospitals or health clinics. Moderately (OR 1.336, p = 0.009) and severely (OR 1.605, p = 0.007) injured patients were more likely to seek care elsewhere before CHY and were less likely to be discharged home after their emergency ward visit as opposed to being admitted to the hospital for further treatment (OR 0.462, p < 0.001). Commercial vehicles provided most prehospital transport (65%), while police or ambulance transported few injured patients (7%). CONCLUSIONS: Possible areas for prehospital trauma care strengthening include training lay commercial vehicle drivers in trauma care and formalizing triage, referral, and communication protocols for prehospital care to optimize timely transfer and care while minimizing secondary injury to patients.

Self-Directed Interactive Video-Based Instruction Versus Instructor-Led Teaching for Myanmar House Surgeons: A Randomized, Noninferiority Trial.

OBJECTIVE: To compare self-directed interactive video-based instruction (IVBI) with instructor-led teaching in the acquisition of basic surgical skills by House Surgeons at University of Medicine 1, Yangon. DESIGN: A prospective, 1:1 randomized controlled trial was conducted. Participants were randomized into 2 teaching arms: (1) self-directed IVBI or (2) instructor-led teaching. Self-directed IVBI participants were provided with a portable DVD player that could play, fast forward, rewind, and skip through skills modules. Participants in the instructor-led teaching group were taught in small groups by standardized instructors. Pretesting and posttesting of 1-handed knot tie, 2-handed knot tie, vertical mattress suture, and instrument tie was performed using the Objective Structured Assessment of Technical Skills (OSATS). Students randomized to self-directed IVBI completed an exit survey to assess satisfaction. Demographic data were collected of all participants. SETTING: University of Medicine 1, Yangon, Myanmar. PARTICIPANTS: Fifty participants were randomly selected from 78 eligible House Surgeons who were enrolled in their basic surgery rotation. RESULTS: Demographic characteristics and baseline skills were comparable in participants randomized to IVBI and instructor-led teaching. Mean OSATS score increased from pretest to posttest in both groups (p < 0.001). The mean posttest OSATS score of the IVBI group was 0.72 points below that of the instructor-led teaching group (90% CI: -3.8 to 5.2), with the 90% CI falling below the a priori noninferiority margin, satisfying criteria for noninferiority. More than 90% of students marked either "agree" or "strongly agree" to the following statements on the exit survey: further expansion of IVBI into other skills modules and integration of IVBI into training curriculum. CONCLUSION: IVBI is noninferior to instructor-led teaching of surgical skills based on OSATS scores. House Surgeons highly rated self-directed IVBI. Self-directed IVBI has the potential to significantly reduce the personnel required for skills teaching and may serve as a long-term learning adjunct in low-resource settings.

Exploring data sources for road traffic injury in Cameroon: Collection and completeness of police records, newspaper reports, and a hospital trauma registry.

BACKGROUND: Road traffic injury surveillance systems are a cornerstone of organized efforts at injury control. Although high-income countries rely on established trauma registries and police databases, in low- and middle-income countries, the data source that provides the best collection of road traffic injury events in specific low- and middle-income country contexts without mature surveillance systems is unclear. The objective of this study was to compare the information available on road traffic injuries in 3 data sources used for surveillance in the sub-Saharan African country of Cameroon, providing potential insight on data sources for road traffic injury surveillance in low- and middle-income countries. We assessed the number of events captured and the information available in Yaoundé, Cameroon, from 3 separate sources of data on road traffic injuries: trauma registry, police records, and newspapers. METHODS: Data were collected from a single-hospital trauma registry, police records, and the 6 most widely circulated newspapers in Yaoundé during a 6-month period in 2009. The number of road traffic injury events, mortality, and other variables included commonly in injury surveillance systems were recorded. We compared these sources using descriptive analysis. RESULTS: Hospital, police, and newspaper sources recorded 1,686, 273, and 480 road traffic injuries, respectively. The trauma registry provided the most complete data for the majority of variables explored; however, the newspaper data source captured 2, mass casualty, train crash events unrecorded in the other sources. Police data provided the most complete information on first responders to the scene, missing in only 7%. CONCLUSION: Investing in the hospital-based trauma registry may yield the best surveillance for road traffic injuries in some low- and middle-income countries, such as Yaoundé, Cameroon; however, police and newspaper reports may serve as alternative data sources when specific information is needed.

A cross-sectional survey of emergency and essential surgical care capacity among hospitals with high trauma burden in a Central African country.

BACKGROUND: As the overwhelming surgical burden of injury and disease steadily increases, disproportionately affecting low- and middle-income countries, adequate surgical and trauma care systems are essential. Yet, little is known about the emergency and essential surgical care (EESC) capacity of facilities in many African countries. The objective of this study was to assess the EESC capacity in different types of hospitals across Cameroon. METHODS: This cross-sectional survey used the WHO Tool for Situational Analysis to Assess EESC, investigating four key areas: infrastructure, human resources, interventions, and equipment and supplies. Twelve hospitals were surveyed between August and September 2009. Facilities were conveniently sampled based on proximity to road traffic and sociodemographic composition of population served in four regions of Cameroon. To complete the survey, investigators interviewed heads of facilities, medical advisors, and nursing officers and consulted hospital records and statistics at each facility. RESULTS: Seven district hospitals, two regional hospitals, two general hospitals, and one missionary hospital completed the survey. Infrastructure for EESC was generally inadequate with the largest gaps in availability of oxygen concentrator supply, an on-site blood bank, and pain relief management guidelines. Human resources were scarce with a combined total of six qualified surgeons, seven qualified obstetrician/gynecologists, and no anesthesiologists at district, regional, and missionary hospitals. Of 35 surgical interventions, 16 were provided by all hospitals. District hospitals reported referring patients for 22 interventions. Only nine of the 67 pieces of equipment were available at all hospitals for all patients all of the time. CONCLUSIONS: Severe shortages highlighted by this survey demonstrate the significant gaps in capacity of hospitals to deliver EESC and effectively address the increasing surgical burden of disease and injury in Cameroon. This data provides a foundation for evidence-based decision-making surrounding appropriate allocation and provision of resources for adequate EESC in the country.

Analysis of prospective trauma registry data in Francophone Africa: a pilot study from Cameroon.

INTRODUCTION: Injury rates in sub-Saharan Africa are among the highest in the world, but prospective, registry-based reports from Cameroon are limited. We aimed to create a prospective trauma registry to expand the data elements collected on injury at a busy tertiary center in Yaoundé Cameroon. METHODS: Details of the injury context, presentation, care, cost, and disposition from the emergency department (ED) were gathered over a 6-month period, by trained research assistants using a structured questionnaire. Bivariate and multivariate models were built to explore variable relationships and outcomes. RESULTS: There were 2,855 injured patients in 6 months, comprising almost half of all ED visits. Mean age was 30 years; 73 % were male. Injury mechanism was road traffic injury in 59 %, fall in 7 %, penetrating trauma in 6 %, and animal bites in 4 %. Of these, 1,974 (69 %) were discharged home, 517 (18 %) taken to the operating room, and 14 (1 %) to the intensive care unit. The body areas most severely injured were pelvis and extremity in 43 %, head in 30 %, chest in 4 %, and abdomen in 3 %. The estimated injury severity score (eISS) was <9 in 60 %, 9-24 in 35 %, and >25 in 2 %. Mortality was 0.7 %. In the multivariate analysis, independent predictors of mortality were eISS ≥9 and Glasgow Coma Score ≤12. Road traffic injury was an independent predictor for the need to have surgery. Trauma registry results were presented to the Ministry of Health in Cameroon, prompting the formation of a National Injury Committee. CONCLUSIONS: Injuries comprise a significant proportion of ED visits and utilization of surgical services in Yaoundé. A prospective approach allows for more extensive information. Thorough data from a prospective trauma registry can be used successfully to advocate for policy towards prevention and treatment of injuries.

Benchmarking of trauma care worldwide: the potential value of an International Trauma Data Bank (ITDB).

BACKGROUND: National trauma registries have helped improve patient outcomes across the world. Recently, the idea of an International Trauma Data Bank (ITDB) has been suggested to establish global comparative assessments of trauma outcomes. The objective of this study was to determine whether global trauma data could be combined to perform international outcomes benchmarking. METHODS: We used observed/expected (O/E) mortality ratios to compare two trauma centers [European high-income country (HIC) and Asian lower-middle income country (LMIC)] with centers in the North American National Trauma Data Bank (NTDB). Patients (≥16 years) with blunt/penetrating injuries were included. Multivariable logistic regression, adjusting for known predictors of trauma mortality, was performed. Estimates were used to predict the expected deaths at each center and to calculate O/E mortality ratios for benchmarking. RESULTS: A total of 375,433 patients from 301 centers were included from the NTDB (2002-2010). The LMIC trauma center had 806 patients (2002-2010), whereas the HIC reported 1,003 patients (2002-2004). The most important known predictors of trauma mortality were adequately recorded in all datasets. Mortality benchmarking revealed that the HIC center performed similarly to the NTDB centers [O/E = 1.11 (95% confidence interval (CI) 0.92-1.35)], whereas the LMIC center showed significantly worse survival [O/E = 1.52 (1.23-1.88)]. Subset analyses of patients with blunt or penetrating injury showed similar results. CONCLUSIONS: Using only a few key covariates, aggregated global trauma data can be used to adequately perform international trauma center benchmarking. The creation of the ITDB is feasible and recommended as it may be a pivotal step towards improving global trauma outcomes.

Is the Kampala trauma score an effective predictor of mortality in low-resource settings? A comparison of multiple trauma severity scores.

BACKGROUND: In the developed world, multiple injury severity scores have been used for trauma patient evaluation and study. However, few studies have supported the effectiveness of different trauma scoring methods in the developing world. The Kampala Trauma Score (KTS) was developed for use in resource-limited settings and has been shown to be a robust predictor of death. This study evaluates the ability of KTS to predict the mortality of trauma patients compared to other trauma scoring systems. METHODS: Data were collected on injured patients presenting to Central Hospital of Yaoundé, Cameroon from April 15 to October 15, 2009. The KTS, Injury Severity Score, Revised Trauma Score, Glasgow Coma Scale, and Trauma Injury Severity Score were calculated for each patient. Scores were evaluated as predictors of mortality using logistic regression models. Areas under receiver operating characteristic (ROC) curves were compared. RESULTS: Altogether, 2855 patients were evaluated with a mortality rate of 6 per 1000. Each score analyzed was a statistically significant predictor of mortality. The area under the ROC for KTS as a predictor of mortality was 0.7748 (95% CI 0.6285-0.9212). There were no statistically significant pairwise differences between ROC areas of KTS and other scores. Similar results were found when the analysis was limited to severe injuries. CONCLUSIONS: This comparison of KTS to other trauma scores supports the adoption of KTS for injury surveillance and triage in resource-limited settings. We show that the KTS is as effective as other scoring systems for predicting patient mortality.

The effectiveness of prophylactic inferior vena cava filters in trauma patients: a systematic review and meta-analysis.

IMPORTANCE: Trauma is known to be one of the strongest risk factors for pulmonary embolism (PE). Current guidelines recommend low-molecular-weight heparin therapy for prevention of PE, but trauma places some patients at risk of excess bleeding. Experts are divided on the role of prophylactic inferior vena cava (IVC) filters to prevent PE. OBJECTIVE: To perform a systematic review and meta-analysis examining the comparative effectiveness of prophylactic IVC filters in trauma patients, particularly in preventing PE, fatal PE, and mortality. DATA SOURCES: We searched the following databases for primary studies: MEDLINE, EMBASE, Scopus, CINAHL, International Pharmaceutical Abstracts, clinicaltrial.gov, and the Cochrane Library (all through July 31, 2012). We developed a search strategy using medical subject headings terms and text words of key articles that we identified a priori. We reviewed the references of all included articles, relevant review articles, and related systematic reviews to identify articles the database searches might have missed. STUDY SELECTION: We reviewed titles followed by abstracts to identify randomized clinical trials or observational studies with comparison groups reporting on the effectiveness and/or safety of IVC filters for prevention of venous thromboembolism in trauma patients. DATA EXTRACTION AND SYNTHESIS: Two investigators independently reviewed abstracts and abstracted data. For studies amenable to pooling with meta-analysis, we pooled using the random-effects model to analyze the relative risks. We graded the quantity, quality, and consistency of the evidence by adapting an evidence-grading scheme recommended by the Agency for Healthcare Research and Quality. RESULTS: Eight controlled studies compared the effectiveness of no IVC filter vs IVC filter on PE, fatal PE, deep vein thrombosis, and/or mortality in trauma patients. Evidence showed a consistent reduction of PE (relative risk, 0.20 [95% CI, 0.06-0.70]; I(2)=0%) and fatal PE (0.09 [0.01-0.81]; I(2)=0%) with IVC filter placement, without any statistical heterogeneity. We found no significant difference in the incidence of deep vein thrombosis (relative risk, 1.76 [95% CI, 0.50-6.19]; P=.38; I(2)=56.8%) or mortality (0.70 [0.40-1.23]; I(2)=6.7%). The number needed to treat to prevent 1 additional PE with IVC filters is estimated to range from 109 (95% CI, 93-190) to 962 (819-2565), depending on the baseline PE risk. CONCLUSIONS AND RELEVANCE: The strength of evidence is low but supports the association of IVC filter placement with a lower incidence of PE and fatal PE in trauma patients. Which patients experience benefit enough to outweigh the harms associated with IVC filter placement remains unclear. Additional well-designed observational or prospective cohort studies may be informative.

Influencing medical student education via a voluntary shadowing program for trauma and acute care surgery.

Decreasing application into trauma surgery may be attributed to decreased exposure during medical school. We instituted a voluntary trauma call program for students to provide exposure to the field. After 3 years, participants completed a survey to gauge their experience. Of 126 students who participated, 68 completed the survey (54%). Interest in trauma surgery among students not previously planning on a career in surgery (n = 48) improved after the experience (4.4-5.3 of 10 points; P < .005). Operative experience, exposure to a higher number of trauma cases, and time with residents were associated with increased interest in trauma surgery. Witnessing patient death for the first time was associated with decreased interest in trauma surgery. A voluntary overnight shadowing program improves medical students' perceptions of trauma surgery and increases their reported likelihood to apply into a surgical residency.

Should the IDC-9 Trauma Mortality Prediction Model become the new paradigm for benchmarking trauma outcomes?

BACKGROUND: Optimum quantification of injury severity remains an imprecise science with a need for improvement. The accuracy of the criterion standard Injury Severity Score (ISS) worsens as a patient's injury severity increases, especially among patients with penetrating trauma. The objective of this study was to comprehensively compare the mortality prediction ability of three anatomic injury severity indices: the ISS, the New ISS (NISS), and the DRG International Classification of Diseases-9th Rev.-Trauma Mortality Prediction Model (TMPM-ICD-9), a recently developed contemporary injury assessment model. METHODS: Retrospective analysis of patients in the National Trauma Data Bank from 2007 to 2008. The TMPM-ICD-9 values were computed and compared with the ISS and NISS for each patient using in-hospital mortality after trauma as the outcome measure. Discrimination and calibration were compared using the area under the receiver operator characteristic curve. Subgroup analysis was performed to compare each score across varying ranges of injury severity and across different types of injury. RESULTS: A total of 533,898 patients were identified with a crude mortality rate of 4.7%. The ISS and NISS performed equally in the groups with minor (ISS, 1-8) and moderate (ISS, 9-15) injuries, regardless of the injury type. However, in the populations with severe (ISS, 16-24) and very severe (ISS, ≥ 25) injuries for all injury types, the NISS predicted mortality better than the ISS did. The TMPM-ICD-9 outperformed both the NISS and ISS almost consistently. CONCLUSION: The NISS and TMPM-ICD-9 are both superior predictors of mortality as compared with the ISS. The immediate adoption of NISS for evaluating trauma outcomes using trauma registry data is recommended. The TMPM-ICD-9 may be an even better measure of human injury, and its use in administrative or nonregistry data is suggested. Further research on its attributes is recommended because it has the potential to become the basis for benchmarking trauma outcomes. LEVEL OF EVIDENCE: Prognostic study, level III.

Patterns of pediatric injury in South Africa: an analysis of hospital data between 1997 and 2006.

BACKGROUND: Pediatric injuries are associated with significant morbidity and mortality, especially in low- and middle-income countries. Despite the magnitude of this burden, there is lack of data to characterize the etiology and risk factors associated with childhood injuries, especially in low- and middle-income countries. The aim of this article is to describe the demographics, mechanisms, and severity of injuries during a 10-year time period using hospital-based data in Cape Town, South Africa. METHODS: Data from Childsafe South Africa's registry were used to study injured children younger than 13 years who presented with either intentional or unintentional injuries to the Trauma Unit of the Red Cross War Memorial Children's Hospital's (RCH) Causality Department between 1996 and 2007. Univariate and bivariate analyses were performed for demographic characteristics and injury mechanisms. Poisson regression analysis was used to analyze the age-adjusted annual incidence of injury presenting to RCH. RESULTS: Between 1997 and 2006, 62,782 children with a total of 68,883 injuries presented to RCH. The mean age was 5.4 years (standard deviation ± 3.5 years) and 61.7% were male. Mechanism of injury included falls (39.8%), road traffic injuries (15.7%), burns (8.8%), and assault (7.4%). The majority of injuries occurred in and around the home. Abbreviated injury severity scoring showed 60.2% of injuries were minor, 36.6% were moderate, and 3.2% were severe. Sixty-six deaths occurred in the trauma casualty department. Thirty-one percent of patients were admitted to the hospital; children who suffered burn and head injuries were more likely to require admission. CONCLUSION: Age, gender, mechanism, and severity of injury in pediatric populations have not been described elsewhere in South African national or sub-Saharan regional data. This retrospective, observational study uses Level II evidence to suggest the need for targeted interventions to address risk factors for pediatric injuries, emphasizing the importance of pediatric surveillance systems as a tool to study injuries in developing countries. LEVEL OF EVIDENCE: Prognostic study, level II.