How accurately do we estimate patients' weight in ... - Europe PMC

0 downloads 0 Views 1MB Size Report
Estimates out by more than 15% were made by 11% of nurses, 16% of ... d'erreur de plus de 15% ont et faites par 11% des infirmieres, 16% des mede-.
How accurately do we estimate patients' weight in emergency departments? Christopher M.B. Fernandes, MD Steven Clark Ann Price, RN Grant Innes, MD OBJECTIVE To assess the accuracy of estimates of patients' weight made by physicians, nurses, and patients themselves in emergency departments. DESIGN Observational prospective study. SETTING Tertiary referral centre in Vancouver, BC. PARTICIPANTS Eleven attending physicians, 26 nurses, and a convenience sample of 117 patients. INTERVENTIONS Patients themselves, attending physicians, and nurses independently estimated the weight of 117 patients. An investigator weighed each patient. MAIN OUTCOME MEASURES Mean error was determined by subtracting actual weight from estimated weight and dividing by actual weight; 95% confidence intervals (CI) were calculated. RESULTS Mean error in estimates was 3.1% (95% CI 2.7 to 3.5) for patients, 8.4% for nurses (CI 7.6 to 9.2), and 8.1% (CI 7.1 to 9.1) for physicians. Weight was estimated within 5% of actual weight by 32% of nurses, 39% of physicians, and 82% of patients. Weight was estimated within 10% of actual weight by 66% of nurses, 66% of physicians, and 97% of patients. Estimates out by more than 15% were made by 11% of nurses, 16% of physicians, and 1% of patients. CONCLUSIONS Patient estimates were most accurate. Physician and nurse estimates were unreliable. OBJECTIF Evaluer l'exactitude des estimations du poids des patients faites par les medecins, les infirmieres et les patients eux-memes dans les services d'urgence. CONCEPTION Une etude prospective par observation. CONTEXTE Un centre de soins tertiaires 'a Vancouver, en Colombie-Britannique. PARTICIPANTS Onze me'decins traitants, 26 infirmieres et un echantillon de commodite de 117 patients. INTERVENTIONS Les patients eux-memes, les medecins traitants et les infirmieres ont individuellement estime le poids de 117 patients. Une personne chargee de l'etude pesait chaque patient. PRINCIPALES MESURES DES RESULTATS L'erreur moyenne etait dkterminee en soustrayant le poids reel du poids estime puis en divisant le solde par le poids reel; les intervalles de confiance (IC) 'a 95% etaient calcules. RESULTATS L'erreur moyenne etait de 3,1% (IC 'a 95% de 2,7 'a 3,5) par les patients, de 8,4% (IC de 7,6 'a 9,2) par les infirmieres et de 8,1% (IC de 7,1 'a 9,1) par les medecins. Le poids etait estime avec une marge d'erreur de 5% par rappct au poids reel par 32% des infirmieres, 39% des medecins et 82% des patients. Le poids etait estime avec une marge d'erreur de 10% par 66% des infirmieres, 66% des medecins et 97% des patients. Des estimations avec,une marge d'erreur de plus de 15% ont et faites par 11% des infirmieres, 16% des medecins et 16 des patients. CONCLUSIONSk L'estimation des patients est plus exacte. Les estimations par les medecins et les infirmieres n'etaient pas fiables.

This article has been peer reviewed. Cet article a fait l'objet d'une evaluation externe. Can Fam Physician 1999;45:2373-2376. -k-

FOR PRESCRIBING INFORMATION SEE PAGE 2484

VOL45: OCIOBER * OCrOBRE 1999, Canadian Family Physician Le Medecin defamille canadien 2373

RESEARCH How accurately do we estimate patients' weight in emergency departments?

n a health care system with strained likely to make mistakes. We treat patients more quickly in order to achieve a faster turnaround time and are at risk of providing lower quality care. Underuse, overuse, and misuse of medications have been identified as important problems in health care.1 One simple but potentially important variable is the estimation of patients' weight for calculating doses of drugs, such as gentamicin or heparin. Often, physicians and nurses estimate a patient's weight before giving a drug. The literature carries reports that, in certain circumstances, inaccurate weight estimation can lead to hazardous drug dosing. Suboptimal heparin dosing, for instance, leads to risk of recurrent thromboembolism.23 Higher doses can lead to a 5% to 20% risk of hemorrhagic complications.3 Very little literature reports on this important question. Leffler and Hayes4 showed the inaccuracy of a standard formula for estimating children's weights in the emergency department (ED). They also showed that parents' estimates were closer to measured weight. One recent study has suggested use of a nomogram for estimating elderly patients' body weight.5 This nomogram is not in common use, however, nor has it been studied in other populations. The criterion standard is a measured weight, which is not always possible or practical to achieve in an ED. Estimates of patients' weight by patients themselves, nurses, or physicians are thus most commonly used. The objective of this study was to assess the accuracy of physician, nurse, and patient estimates of patients' weight in the ED. resources, staff are more

METHODS The study was conducted at St Paul's Hospital in Vancouver, BC, an academic emergency department with approximately 55 000 patient visits annually. The site sponsors an emergency medicine residency program and serves an inner-city population. We enrolled a convenience sample of 117 patients who presented to the ED between August and October 1997. We excluded admitted patients, nonambulatory patients, patients younger than 18 years, and patients unable to give informed consent. We obtained Drs Fernandes and hInes, Mr Clark, and Ms Price practise emergency medicine in the Department of

Emergency Medicine at St Paul's Hospital in Vancouver, BC.

informed consent from each patient after he or she was triaged. The patient, his or her attending physician, and a bedside nurse were independently asked to estimate the patient's weight while he or she was clothed (without jacket, jersey, or shoes) and ambulatory. Weight was recorded to the nearest kilogram. Then patients were all weighed on the same scale, which is regularly calibrated. Each estimator was blinded to other estimations. Investigators ensured blinding, recorded all weight estimates, and then weighed each patient. For each weight estimate, the percentage error was defined as estimated weight minus actual weight divided by actual weight times 100. Mean error was calculated for patients, physicians, and nurses, and 95% confidence intervals were determined for derived mean values. The ethics committees of St Paul's Hospital and the University of British Columbia approved this study. The study was conducted over a 3-month period.

RESULTS Eleven physicians, 26 nurses, and the patients themselves estimated the weights of 117 adult patients whose weight ranged from 42.8 to 120.5 kg (mean 70.9 kg, standard deviation 14.6). Estimates in all three groups were normally distributed (Figure 1). Distribution of-error estimates for the three study groups is shown in Table 1.

DISCUSSION This study demonstrates that physicians' and nurses' estimates of patients' weight in the ED are unreliable and that patients' own estimates are most accurate. This is important in managing life-threatening situations where weight estimations are required. Estimates that deviate more than 10% from actual weight could make treatment itself life threatening. At what point does an error in estimated weight become critical? This point varies, depending on the drug and the situation. In some cases, such as for anticoagulants or thrombolytic agents, a 10% error rate could be crucial, but in other cases, even a large error might not be important. Many factors influence weight estimation in the ED. One study suggested that socioeconomic differences accounted for some misclassifications of height, weight, and body mass index.6 A random sample of 3208 Swedish adults provided weight

2374 Canadian Family Physician . Le Medecin defamille canadien * VOL45: OCIOBER * OCIOBRE 1999

RESEARCH How accurately do we estimate patients' weight in emergency departments? .

-- i.

-'.'

.;

--

..

,,

lczu~~~.

-w.v l.if

.7

-

a

%*,,OR

st osthins

Nyde edluuu

estimations on a questionnaire and were later weighed during a health examination. The study showed that nonmanual workers underestimated their weight more than manual workers. Another factor that could influence the accuracy of weight estimations is the urgency of the situation, though this has not been studied. For instance, a patient requiring resuscitation might have his or her weight estimated by a physician or nurse who is already distracted by other issues in the patient's care. Thus, physicians' and nurses' weight estimates might be

Table 1. Distribution of error in the three study groups' estimates: Percentage error (95% confidence intervals). STUDY GROUP