Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Outcome of children with different accessibility to tertiary pediatric intensive care in a developing country - a prospective cohort study.

Goh AYT, Abdel-Latif MEA, Lum LCS, Abu-Bakar MN.

Intensive Care Med 2003; 29: 97-102. [abstract]

Reviewed by Jose Cortes MD, Baylor College of Medicine and Texas Children's Hospital, Houston, TX

Review posted June 19, 2004

I. What is being studied?:

The study objective:

The authors compared patients' outcome and length of ICU stay depending on the patient access to tertiary critical care. Patients already in a tertiary care center requiring ICU admission were compared to those requiring transportation into the center for ICU admission. They also compared the number of patients who were denied ICU admission and their reason in both groups.

The study design:

Prospective cohort study is and control groups were assessed.

II. Are the results in the study valid?

Primary questions:

1. Was there a representative and well-defined sample of patients at a similar point in the course of the disease?

346 consecutive patients whom required PICU admission at a single children's hospital in Kuala Lumpur, Malaysia, (131 transferred from other community hospitals and 215 from in hospital), over 12 months were studied. There was a reasonable mix of diagnosis in both groups, although significantly more of the community hospital transfers had a pulmonary diagnosis requiring ICU admission. The cohort groups were very young, mean age of 1 month for the transported patients and a mean age of 6 months for the tertiary care hospital group.

The PRISM scores were diverse with a mean PRISM of 10 for the community transfer patients and 9 for the internal transfers, very similar to the average PRISM score of our ICU patients.

2. Was follow-up sufficiently long and complete?

Follow up was until the end of patient ICU stay and this was long enough to evaluate the effects of not having access to ICU immediately and the morbidity/mortality added due to the need of transport. All patients enrolled in the study were completely followed up.

They were looking for the effects of having immediate access to tertiary critical care versus the need to be transported from another facility for this care to be rendered. The ICU length of stay was sufficient to evaluate if a true difference existed.

Secondary questions:

3. Were objective and unbiased outcome criteria used?

Yes, they used PICU mortality and length of ICU stay.

4. Was there adjustment for important prognostic factors?

Yes. They adjusted for differences in accessibility to tertiary PICU, PRISM II scores, age, diagnosis and length of stay. They used multiple logistic regression to create a model that considered PRISM II, length of stay and age. Also, they did an adjustment for difference in mean age and distribution of diagnosis using a severity-of-illness case-mix adjustment tool.

III. What are the results?

1. How large is the likelihood of the outcome event(s) in a specified period of time?

In other words, did it matter whether they had direct access or got transferred in? No, even though the pre-admission PRISM score were statistically different between the two groups, with a mean of 6 in the community transfers and 10 in the internal transfers, the 24 hour PRISM scores were not different. The observed deaths and predicted deaths were not different between the groups. The length of stay was statistically different between the groups, with community transfers staying more than the internal transfers, but when analyzed as an independent predictor of poor outcome (death), length of stay was not statistically relevant.

In a multiple regression model (where PRISM score was accounted for), the odds ratio for death in the community hospital patients was 1.7 (compared to the reference of 1.0 for the in-hospital transfers), but the 95% confidence interval was 0.7 to 4.3, indicating a lack of significance of this variable. In this model, the only reliable predictor of poor outcome was PRISM score (OR 1.3, 95% CI 1.2-1.4).

2. How precise are the estimates of likelihood?

As noted above, the confidence intervals of community versus internal transfers for risk of death was wide and crossed 1, indicating a lack of precision or significance. PRISM score, on the other hand, had a CI 1.2-1.4 and an OR of 1.3.

IV. Will the results help me in caring for my patients?

1. Were the study patients and their management similar to my own?

This is difficult to determine. The authors report a broad range of reasons for admission, but in very general terms, e.g., respiratory, cardiovascular, etc. The actual diseases in rural Malaysia and in Kuala Lumpur may be very different than in suburban and urban American or European regions. The crude mortality rates are roughly three times higher than a typical US PICU; this may indicate a different disease distribution than what we see in a city like Houston.

2. Will the results lead directly to selecting or avoiding therapy?

No. The results did not show a benefit towards improving poor outcome from being admitted directly into the tertiary care center or being transferred from a community hospital.

3. Are the results useful for reassuring or counseling patients?

The evidence presented here suggests that the risk of mortality is not statistically different from being admitted into a tertiary hospital with PICU capability directly compared to being transported from a community hospital. Since most parents of critically ill children do not usually have a choice where they first enter the health care system, this question is not strictly applicable. If we ask, however, whether on the basis of this study we should counsel the public health system of Malaysia to change their practice of referring critically ill children to the tertiary care center (as opposed to upgrading local hospitals to have PICU's), the answer would be no.

 

Comments

Back to the EB Journal Club Index