PEDIATRICS Vol. 122 No. 2 August 2008, pp. e411-e415 (doi:10.1542/peds.2008-0181)
ARTICLE |
Care of Children Isolated for Infection Control: A Prospective Observational Cohort Study
a Department of Pediatrics, University of Toronto
b Pediatric Outcomes Research Team, Child Health Evaluative Sciences, Research Institute, Hospital for Sick Children, Toronto, Ontario, Canada
c Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
d Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
BACKGROUND. Patients with community-acquired or nosocomial infections are often managed in a hospital with isolation precautions. Given the high prevalence and substantial inconvenience associated with implementation of isolation precautions in pediatric settings, we explored the impact of this intervention on the care provided to children and their families.
OBJECTIVE. The purpose of this work was to compare the quantity and quality of care received by isolated patients relative to nonisolated patients.
PATIENTS AND METHODS. Sixty-five consecutive newly admitted inpatients in private rooms (24 isolated, 41 nonisolated) were recruited from the general pediatric service at the Hospital for Sick Children. Interactions between a medical team with patients and their families were observed. All of the participants were blinded to the study objectives. The medical team was observed during its morning rounds, and data were collected on the quantity and quality of care. Quantity of care was determined by the amount of time that the medical team spent interacting with the patient and parents and the number of organ systems examined by the attending physician during morning rounds. Quality of care was determined by using parental completion of the Pediatric Family Satisfaction Questionnaire.
RESULTS. We found no significant difference in the average amount of time spent interacting with isolated compared with nonisolated patients (516 vs 480 seconds) or the number of organ systems examined in isolated compared with nonisolated patients (3 vs 4). Isolated and nonisolated groups gave high ratings to all of the items pertaining to the facility, doctors, and nurses.
CONCLUSIONS. No large differences in quality or quantity of care were observed between isolated and nonisolated patients in the first 2 days of admission to a pediatric ward.
Key Words: quality of care • isolation • infection control
Abbreviations: RIW—resource intensity rating • PFSQ—Pediatric Family Satisfaction Questionnaire
Nosocomial, or health care–associated, infections are the most common complication affecting hospitalized patients.1 Infection prevention and control measures have been instituted widely to prevent the transmission of these, as well as community-acquired infections in pediatric hospitals. All patients are placed under "standard precautions," which include hand hygiene to reduce the transmission of both recognized and unrecognized sources of infection. An important subset of patients with documented or suspected infection with highly transmittable pathogens of epidemiologic significance is assigned to additional precautions.2,3 These can be airborne, droplet, or contact precautions or any combination of the above. Operationally, this requires members of the health care team to wear a combination of gowns, gloves, and/or masks while in the patient's room.
Isolation precautions, like any health care intervention, require systematic evaluation of possible untoward effects. Currently, there are conflicting data on potential negative effects of these additional isolation precautions on patients with some studies showing deleterious effects of isolation precautions,1,4,5 whereas other studies show no adverse effects.6 In addition, few data are available on pediatric populations or on the impact of isolation precautions on health care workers' care of children and their families. The main objective of this study was to compare the quantity and quality of care received by isolated patients relative to nonisolated patients on a general pediatric inpatient ward. As a secondary objective, we also recorded the frequency that the staff washed their hands or used an alcohol-based antiseptic before direct contact with patients.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Setting
We conducted a blinded, prospective observational single-center cohort study on a 55-bed general pediatric inpatient unit at the Hospital for Sick Children between May 31 and August 30, 2006. This unit was predominantly made up of private rooms, with only a few multibed rooms reserved for those patients who need constant nursing observation. Each room was equipped with a hand-washing sink and a bottle of hand sanitizer outside the door.
A single observer, who was not a part of the health care team, obtained consent before morning rounds and made observations. Both the families and physicians involved in the study were aware that they were being observed but were blinded to the study objectives. The study protocol was approved by the hospital institutional review board.
The clinical team, composed of the attending physician, a senior resident, junior residents, clerks, and a pharmacist, conducted rounds every morning 7 days a week. Rounds took 
2 to 3 hours to complete. Although the number of patients seen varies, attending physicians must see and examine all of the newly admitted patients on morning rounds. A variable number of medical housestaff also examine the patients during these rounds.
Recruitment
All of the patients meeting the inclusion criteria were recruited for the study consecutively. The inclusion criteria for this study were as follows: admission to a private room from the emergency department or the ICU, parental ability to complete a survey in English, and families not having met the attending physician before morning rounds. Six different attending physicians were observed over the study period.
Members of the health care team who were not involved in the study assigned children for isolation precautions. At our hospital, any health care worker can isolate a patient, but this is most commonly done by a ward nurse. Patients are placed under isolation precautions according to established practice guidelines.3 For instance, isolation can be instituted for children with certain infectious symptoms (eg, fever and diarrhea), those known to be colonized with a multidrug-resistant organism, or children transferred from an institution with a high prevalence of multidrug-resistant organisms.
Data Collection
Baseline data, including age, gender, and family income, were abstracted from patient charts, hospital databases, and as a part of a survey handed out to parents on the second day of their stay.
The observer accompanied the health care team on morning rounds and made direct observations during this time. Time spent in rooms was recorded using a digital stopwatch. The observer also recorded the number of organ systems examined during the interaction, and which members of the health care team had not washed or sterilized their hands before patient contact. Recorded vital signs, discharge diagnosis, and length of stay were abstracted from the patient's charts after discharge.
To compare the potential confounders of acuity and complexity of illness between the 2 groups, data were collected on 2 proxy measures: GRASP scores (GRASP Systems, Inc, Ft Collins, CO)7,8 and resource intensity weighting (RIW).9,10 The GRASP system served as a validated assessment of nursing workload that is used routinely in hospitals. Nurses select, on a daily basis, the interventions that they have performed during their shift. The nursing care requirements (GRASP score), expressed in hours, were then determined for each patient. The RIW represented a measure of resource consumption that used case mix group methodology to describe the average use of resources by a patient in the hospital relative to the "average hospital case," which is assigned a value of 1.0.
Outcome Measures
The quantity of care provided to patients and families was measured using multiple methods. The primary outcome was the amount of time spent by the attending physician in the patient's room during morning rounds providing patient care, which was defined as the total time spent in the room less the time spent directly teaching residents and clerks without any interaction with the patient or family. Secondary outcomes included the extent of the physical examination, defined as the number of organ systems examined by the attending physician, and the proportion of nursing vital signs not recorded. The time spent with patients and the extent of the physical examination were directly observed; missed nursing vitals were abstracted from the patient charts. Missed vital signs were defined as ordered vitals that were not recorded at all. Partial vital signs were defined as recorded vitals where 
1 vital sign was missing from the patient record. Examples of this included, for instance, records containing heart rate and respiratory rate without blood pressure or temperature.
Quality of care was measured by the mean rating per item on the Pediatric Family Satisfaction Questionnaire (PFSQ),10,11 administered after 48 hours of hospitalization. The PFSQ is a 35-item instrument designed to assess parental satisfaction with pediatric inpatient hospital care in multiple domains, such as hospital services and accommodation, nursing care, and medical care. Secondary quality outcomes abstracted from hospital databases included the number of incident reports filled out while the patients were on the ward or the number of complaints made to the patient representative or to the hospital administration. The compliance of the medical staff with infection-control guidelines was measured by the percentage of times that they washed or sanitized their hands before making contact with their patients.
Statistical Analysis
Baseline characteristics of the 2 groups were collected and described by using means, medians, and proportions, as well as SDs, interquartile ranges, and 95% confidence intervals, as appropriate. Comparisons for baseline characteristics and outcomes were made between the isolated and nonisolated groups by using independent t tests for normally distributed data, the Mann-Whitney test for nonnormal data, and the 
2 test for categorical data. All of the P values reported were 2-tailed. Statistical analyses were performed with Minitab 14 (Minitab; State College, PA).
Sample-Size Calculation
The primary outcome for sample size calculation was the amount of time spent by the attending physician in the patient's room. Based on a retrospective review of 1 year of data from the inpatient setting, it was predicted that an unequal sample size would be obtained, with 38% of all of the admitted inpatients being placed under isolation precautions. Assuming the mean amount of time spent in each patient's room by the attending physician would be 15 minutes, an SD of 4 minutes, and a clinically important difference being a 20% difference between the groups, a sample size of 60 subjects (23 isolated and 37 nonisolated) would be required to provide 80% power to detect such a difference at an 
level of .05 (2-tailed).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
A total of 65 patients were recruited, of which 24 were under isolation precautions and 41 were under standard precautions (nonisolated). Baseline distributions between the 2 groups were similar in terms of age, gender, income, household size, and whether they had been previously hospitalized (Table 1). Overall, 20 (80%) of 24 isolated patients were discharged with a primary discharge diagnosis of an infectious disease compared with 11 (27%) of 41 nonisolated patients (P < .001).
|
Five patients were isolated with droplet precautions, 5 with contact precautions, and 1 with airborne precautions. Eleven isolated patients were placed under a combination of contact and droplet precautions, and 2 patients were isolated for a combination of contact, droplet, and airborne precautions. There was only 1 patient isolated for a suspicion of a multidrug-resistant organism. All of the patients were isolated for suspected, rather than confirmed, infectious diseases. Subsequent investigations revealed that 9 of these patients had laboratory-confirmed infectious diseases: 4 with skin infections (herpes simplex virus type 1 in 2 patients, Pseudomonas aeruginosa, and Staphylococcus aureus), 2 with meningitis (Neisseria meningitidis and enterovirus); 2 with complicated pneumonia (Mycoplasma pneumoniae and Streptococcus anginosus); and 1 with Mycobacterium tuberculosis.
On average, the attending physician spent 9 minutes in the patient's room, of which 8 minutes was directly in contact with the patient and/or the family (Table 2). The quantity of care was similar in both groups in terms of time spent in the room, organ systems examined, vital signs recorded, GRASP scores, and RIW. The groups differed in the length of stay, which was 4.5 days for isolated patients versus 2.0 days for nonisolated patients (P = .014).
|
Measures of quality of care are summarized in Table 3. The PFSQ was completed by the parents of 26 (63%) of non-isolated and 18 (75%) of the isolated patients (P = .32). There was no difference between the 2 groups in the average rating per item on the PFSQ (P = .209) or in the number of safety or incident reports completed (P = .109). Overall, 42 (41.6%) of the 101 medical staff who examined patients washed or sterilized their hands before touching the patients and were more likely to wash or sterilize their hands for nonisolated compared with isolated patients (37 of 61 [60.7%] vs 5 of 40 [12.5%]; P < .001). Among attending physicians, 32 (56.1%) of 57 patient interactions were preceded by hand hygiene precautions.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This blinded, prospective observational study found no significant difference in terms of time spent in the room by the attending physician or number of organ systems examined in isolated compared with nonisolated patients. If anything, there was a nonsignificant trend toward more time and more organ systems examined in isolated patients than in nonisolated patients. Few vital signs were missed or partially recorded in both groups. Both groups rated the quality of their care to be high, and we found no statistically significant difference in the number of safety or incident reports filled out between the 2 groups.
These results are consistent with a prospective, randomized trial of infection-control measures (gowning and gloving) in 70 mechanically ventilated children in an intensive care setting.6 These authors found no detrimental effects of isolation precautions; the frequency of contact with patients and their families was comparable between the 2 groups, and protective isolation was well accepted by patients and their families. The study also noted that isolated patients had fewer adverse events (4.4 vs 8.6 nosocomial infections per 100 days) and a lower death rate (9% vs 13%) compared with the control group. Our findings of comparable rates of adverse events and no mortality, in controls relative to isolated patients, are likely largely because of substantial differences in the study populations. Intensive care patients are likely to be more susceptible to the adverse effects of isolation-preventable nosocomial infections that prolong length of stay and increase the risk of death compared with general ward patients. Nevertheless, even on a general pediatric ward, where the potential benefit of isolation may be lower, we were unable to detect any important short-term consequences of isolation.
In contrast, a prospective observational study conducted in an adult ICU of 219 patient encounters5 found that health care workers were approximately half as likely to enter the room of patients in contact isolation compared with control subjects, although visits to the rooms of isolated patients tended to be longer. Another study of an adult medical ward reviewed a retrospective cohort of 78 patients isolated for methicillin-resistant S aureus and compared them with 2 matched control groups. The isolated patients in this study were more likely to have incomplete records of vital signs (14% vs 9%) or days with no vital signs recorded at all (5% vs 1%). The authors also found that isolated patients were twice as likely as control patients to experience preventable adverse events and were more likely to file complaints with public relations offices or have them recorded in their charts. The differences in results between previous research and ours may be because of the fact that the baseline rate of missed vital signs, complaints, and adverse events was much lower in our study, which may be attributable to important differences in the quality of care between adult patients isolated for prolonged periods for multidrug-resistant organisms and pediatric patients isolated usually for limited periods for other reasons.
Our findings of hand-hygiene compliance rates are similar to other work. A recent study using both a questionnaire and direct observation to evaluate hand-washing practices by health care personnel12 found that only 45.1% washed their hands before contact with a patient when directly observed, which is similar to our observations. Good hand-hygiene techniques by attending physician role models can have the additional benefit of reinforcing this behavior for trainees.13 Our finding that health care workers are ironically less likely to wash or sterilize their hands before contact with isolated patients is concerning. Even if health care workers wear gloves while caring for infectious patients, they may contaminate the surface of the gloves before patient contact and frequently contaminate their hands when they remove their gloves.14 This pattern implies that isolated patients may be an important population for targeted interventions to promote hand hygiene in health care workers.
There are several limitations to this study. The sample size was not powered to detect small differences in the time spent with patients, but small differences may not be clinically important. Direct observations were only made during morning rounds, and conceivably other clinically important activities throughout the day could have been missed. The study was designed to look primarily at early effects of isolation; data were only collected prospectively for the first 48 hours. Effects of prolonged isolation precautions were not assessed. These were teaching rounds and, therefore, may reflect better compliance with isolation precautions. The behaviors of those involved may have been modified, because they were aware that they were being observed. Although we tried to account for potential confounders that may have affected the outcome of the study, such as family demographics and resource intensity scores, because of the observational design, unknown confounders may not have been accounted for, which would have necessitated a randomized trial. This is not ethical given that infection-control precautions are proven to be effective and considered a standard of care.3 One potential confounder could have been acuity of illness. The increased length of stay in isolated patients may have been because the patients were sicker than the nonisolated patients and, thus, received greater attention to care as an explanation for the lack of a difference between the 2 groups. However, there were no differences between the groups in the other proxies for acuity or complexity of illness, such as GRASP scores and RIW.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This study found no large differences in the quality or quantity of care for patients on a general pediatric inpatient unit who were in isolation. Future prospective studies on long-term consequences of isolation in pediatric inpatient settings may be helpful.
|
ACKNOWLEDGMENTS |
|---|
Salary support for Janice Austin was provided by the Ronald and Annie MacDonnell Fund.
|
FOOTNOTES |
|---|
Accepted Mar 28, 2008.
Address correspondence to Eyal Cohen, MD, MSc, 555 University Ave, Toronto, Ontario, M5G 1X8 Canada. E-mail: eyal.cohen{at}sickkids.ca
The authors have indicated they have no financial relationships relevant to this article to disclose.
| What's Known on This Subject There may be deleterious effects of isolation precautions on the care of isolated patients, but few data are available on pediatric patients.
|
| What This Study Adds No large differences in the quality or quantity of care were observed between isolated and nonisolated patients in the first 2 days after admission to a general pediatric ward.
|
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Burke J. Infection control: a problem for patient safety.
N Engl J Med. 2003;348
(7):651
–656
[Free Full Text] - Garner J. Guidelines for isolation precautions in hospitals. Infect Control Hosp Epidemiol. 1995;16 (2):105 –113[Web of Science][Medline]
- Siegel J, Rhinehart E, Jackson M, Chiarello L, and the Healthcare Infection Control Practices Advisory Committee. 2007 guideline for isolation precautions: preventing transmission of infectious agents in healthcare settings. Available at: www.cdc.gov/ncidod/dhqp/pdf/guidelines/isolation2007.pdf. Accessed March 21, 2008
- Stelfox HT, Bates DW, Redelmeier DA. Safety of patients isolated for infection control.
JAMA. 2003;290
(14):1899
–1905
[Abstract/Free Full Text] - Kirkland KB, Weinstein JM. Adverse effects of contact isolation. Lancet. 1999;354 (9185):1177 –1178[CrossRef][Web of Science][Medline]
- Klein BS, Perloff WH, Maki DG. Reduction of nosocomial infection during pediatric intensive care by protective isolation. N Engl J Med. 1989;320 (26):1714 –1721[Abstract]
- Meyer D. GRASP: A Patient Information and Workload Management System. Morganton, NC: MCS; 1978
- Adams-Wendling L, Dedonder J, Tidwell S, Pimple C, Schmiot L, Okeson D. Budgeting nursing workload for required minimum data set assessments. J Nurs Manag. 2007;15 (4):442 –448[CrossRef][Medline]
- Preyra C, Sandor P. A Prospective Method of Complexity Adjustment to Case Mix Groups. Toronto, Canada: Finance and Information Management Branch, Ontario Ministry of Health and Long Term Care; 2002
- Bragadottir H, Reed D. Psychometric instrument evaluation: the pediatric family satisfaction questionnaire. Pediatr Nurs. 2002;28 (5):475 –482[Medline]
- Budreau G, Chase L. A family-centered approach to the development of a pediatric family satisfaction questionnaire. Pediatr Nurs. 1994;20 (6):604 –608[Medline]
- Sacar S, Turgut H, Kaleli I, et al. Poor hospital infection control practice in hand hygiene, glove utilization, and usage of tourniquets. Am J Infect Control. 2006;34 (9):606 –609[CrossRef][Web of Science][Medline]
- Pittet D, Simon A, Hugonnet S, Pessoa-Silva CL, Sauvan V, Perneger TV. Hand hygiene among physicians: performance, beliefs, and perceptions.
Ann Intern Med. 2004;141
(1):1
–8
[Abstract/Free Full Text] - Goldmann D. System failure versus personal accountability–the case for clean hands.
N Engl J Med. 2006;355
(2):121
–123
[Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  A. G. Matlow and S. K. Morris Control of antibiotic-resistant bacteria in the office and clinic Can. Med. Assoc. J., May 12, 2009; 180(10): 1021 - 1024. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
