Correlation between classification in risk categories and clinical aspects and outcomes

ABSTRACT Objective: to correlate classification in risk categories with the clinical profiles, outcomes and origins of patients. Method: analytical cross-sectional study conducted with 697 medical forms of adult patients. The variables included: age, sex, origin, signs and symptoms, exams, personal antecedents, classification in risk categories, medical specialties, and outcome. The Chi-square and likelihood ratio tests were used to associate classifications in risk categories with origin, signs and symptoms, exams, personal antecedents, medical specialty, and outcome. Results: most patients were women with an average age of 44.5 years. Pain and dyspnea were the symptoms most frequently reported while hypertension and diabetes mellitus were the most common comorbidities. Classifications in the green and yellow categories were the most frequent and hospital discharge the most common outcome. Patients classified in the red category presented the highest percentage of ambulance origin due to surgical reasons. Those classified in the orange and red categories also presented the highest percentage of hospitalization and death. Conclusion: correlation between clinical aspects and outcomes indicate there is a relationship between the complexity of components in the categories with greater severity, evidenced by the highest percentage of hospitalization and death.


ontext, ERs a
e characterized as one of the main entrance doors into the health system, and cases not characterized as emergencies are the ones that most consume this type of service, due to its convenience and the difficulty individuals face accessing primary health care (PHC) services (3) .One recent South Korean study reports that on days with overcrowding, delay in providing complex care was associated with increased intra-hospital mortality (4) .Therefore, one of the consequences of overcrowding in ERs is the need to identify those who require immediate care among patients, because the time between medical assessment and treatment influences the pati nts' prognoses.

Risk classification (RC), a resource used in ERs and implemented by nurses, emerged as a tool intended to allow recognizing those patients that require care be provided in the shortest interval of time possible (5) .

In the 1990s, hospitals in various countries started adopting and improving RC scales to identify patients according to the severity of
n 2004, the Brazilian Ministry of Health (MH) devised the QualiSUS Program and the National Humanization Policy, called HumanizaSUS, and initiated the triage process in Brazil denoted Acolhimento com


Avaliação e Classificação de Risco [Reception with Risk

Assessment and Classification].The idea was that patients classified according to this device would receive care according to the severity of thei conditions and no one would be excluded from this process (6)(7) .

The MH expected some results after the Reception with Risk Assessment was implemented in emergency rooms, including a decrease in the risk of avoidable deaths, extinction of triage performed by non-qualified workers, giving priority to patients according to clinical criteria, shorter waiting times, and the detection of cases that may aggravate if care is postponed (6)(7) .

Given this context, and after approximately 10 years of RC in Brazil, we realize that research addressing this topic has assessed the relationship between components of classification and outcomes, seeking to meet some of the MH's objectives.In Brazil, the most frequently used and researched risk classification protocol is the Man hester system, which is implemented in most Brazilian states (8) .

Recently, Brazilian researchers concluded that the Manchester system shows that patients classified in categories of greater risk remain in hospitals longer when compared to low-risk categories, indicating that the Manchester is a good clinical predictor for length of hospitalization of patients with more severe conditions (9) .This is similar to another study that shows the Manchester system is a good indicator of risk of death for patien s classified as high-risk, as opposed to low-risk patients (10) .

Given this context, this study's guiding question was: What is the relationship existing between the components of clinical aspects, origin, and outcome with the classification of risk categories?Hence, this study's objectives were to correlate CR categ

ies wit
the clinical profile, outcomes and origin of patients.hours a day, seven days a week and the assessment is performed by a nurse, who provides a brief n

sing con
ultation in which the patient is asked about signs and symptoms, time of the onset of symptoms, personal history, medications, and allergies.Vital signs are taken and a color code, according to the category of risk, is assigned to the patient, who is then referred to a medical specialty.The RC protocol is institutional and was implemented in 2009 based on the Ministry of Health's guidelines (6)(7) and on the Manchester system.This protocol is composed of five categories, identified by colors, while each color refers to a waiting time: red (immediate care); orange (10 minutes), yellow (60 minutes), green (120 minutes), and blue (240 minutes).


Method

The study project was approved by the Institutional Review Board at the Federal University of São Paulo (UNIFESP) (CAAE 05739412910015505).The study population was composed of forms from patients 18 years old or older, classified by the RC sector.The forms Oliveira GN, Vancini-Campanharo CR, Lopes MCBT, Barbosa DA, Okuno MFP, Batista REA.

were

nually comp
eted and were digitally available in the facility's electronic system.Incomplete or illegible forms were excluded.Data were collected after sample size was calculated based on the Chi-square test (Effect Size), with a power of 80% and level of significance at 5%, so that a minimum of 531 electronic forms were established.

Data were accessed online using the institution's Categorical variables were descriptively (frequency and percentage) analyzed, while mean, standard deviation, median, minimum a maximum values were calculated for the continuous variables.Chisquare, or when necessary, the likelihood ratio test, was used to compare RC with origin, primary complaint, signs and symptoms, complementary exams, personal antecedents, medical specialty, and outcomes, with a level of significance at 5% (p-value < 0.05).


Results

The 696 forms showed a predominance of female patients (n = 418; 60.1%), aged 44.5 (±19.2) years old on average, while the origin of most patients was residence (n = 682; 9 .0%).The most frequent signs and symptoms were: pain (n = 304; 44.1%), dyspnea (n = 97; 14.1%), incapacity to move a body part due to musculoskeletal lesions (n = 89; 13.0%), and dermatological and infectious problems (n = 79; 11.4%).In regard to personal history, the most prevalent morbidities included systemic blood hypertension  2).The remaining signs and symptoms did not present statistically signifi ant differences in regard to the categories of risk.

In regard to the exams undertaken by the patients during their stay in the hospital, those classified in the yellow, orange, and red categories required more exams than the others.Electrocardiogram was the most frequent xam among patients classified in the red category, bearing in mind that those classified yellow, orange or red presented the highest percentage of laboratory and image exams (Table 3) Discussion

Overcrowding of ERs in recent years made RC scales a mandatory tool in these facil ties (5) .Some of the characteristics of the patients in this study, such as being mostly women (60.1%) with an average age of 44.5 years, are similar to those found in another study addressing forms from the R of the Odilon Behrens Municipal Hospital in Minas Gerais, Brazil (1) .

In terms of signs and symptoms, pain (44.1%) and dyspnea (14.1%) were the symptoms most frequently presented in the sample.The literature has shown that pain is one of the main complaints presented by individuals seeking care at an ER, with approximately 80% of the patients, even though p oper management of pain remains a challenge for emergency services (1,(11)(12)(13) .

A Brazilian study, conducted in the ER of a university hospital in the interior of the state of São Paulo, Brazil, and a study conducted in the United States found that systemic arterial hypertension and diabetes mellitus presented the highest incident in the populations under study, a result that corroborates this study's findings, as these were the comorbidities most frequently eported, 18.1% and 7.8%, respectively (11,14) .These diseases increased in recent years due to population aging, which when ssociated with sedentariness and obesity, impacted metabolic and cardiovascular diseases and became public health problems (15) .Medical expenditures from complications arising from these diseases are high and patients are increasingly seeking emergency rooms due to clinical decompensation (14)(15) .

In regard to RC categories, most patients (61.1%)

were classified in the green category, followed by yellow (21.9%), showing a tendency of low-risk patients to seek emergency services (1,11) ; that is, the profile of patients seeking ERs is of low complexity.The reasons are multifactorial and are linked to a culture of solving problem rapidly, of using technology to perform examinations, easily accessing a service provided 24 hours a day, and geographical location.Due to these factors, hospitals

Rev. Latino-Am.Enfermagem 2016;24:e2842 have become the main facility where health care is delivered to the population (3,16) .

In this study, the following specialties provided the most care: medical clinic (36.8%), orthopedics (16.8%) and surgery (13.2%).A more intense search for these specialties may be associated with a decompensation of chronic diseases, violence, and traffic accidents (2,12) .
Another reason is that, up to the present, there is no regulation in Brazil for the medical specialty of urgent and emergency care.Medical teams working in ERs are guided by the urgent and emergency care policies according to the level of complexity of the care delivered.

Hence, the nurse implementing RC takes into account the patient's compla nt and associates it with signs and symptoms, and r fers the patient to a medical specialty accordingly (2) .

The most frequently demanded diagnostic exams were imaging (40.4%) and laboratory exams (30.7%).

Exams play an important role in emergency services because they help to establish a medical diagnosis, though the wait for exam results extends the time of the patient in he sector, contributing to the overcrowding in these units.Nonetheless, in this context of high demand, the release of medical reports may be delayed, just as there may be failures in the production process of the imaging sector and such delays and failures may harm patients.Various diagnostic exams, such as laboratory exams, can be performed in PHC units, minimizing potential failures that are observed on overcrowded days (11,17) .

Hospital discharge (94.5%) was the most frequent o tcome, similar to what other studies describing the profile of patients seeking emergency services in the south and southeast of Brazil, have reported.This outcome is likely to be related to the low complexity of the patients' clinical conditions, indicating that these patients could have been cared for in a PHC service, as well as showing a preference on the part of the populati

to seek em
rgency services (9,11) .

Those classified in the red category presented the highest percentage of patients originating form SAMU and AMA, and mostly demanded care due to surgical reasons when compared to other risk categories.This may be associated with the hospital's urban locale, which is near major traffic routes, with a high incident of veh