This study is to validate the use of the modified early warning score (MEWS) in an adult oncology ward as a predictor of 7 day mortality or intensive care unit (ICU) admission. It is expected that either a MEWS score ≥4 or a change in score of ≥2 from the initial reading will be associated with an increased risk of death or ICU admission, and should be regarded as a MEWS Alarm. If MEWS is proof to be a reliable alternative tool, early diagnosis and aggressive management of life-threatening complications in oncology patients, results in dramatic improvements in overall survival rates.

This is a prospective, single centre, observational, cohort study. Objective of this study is to investigate the ability of the Modified Early Warning Score (MEWS) to predict 7 day mortality in adult clinical oncology ward patients.

The alternate hypothesis of this study is a significant difference in 7 day mortality between adult oncology ward patients with a MEWS Alarm and adult oncology ward patients without a MEWS alarm.

The null hypothesis is a no significant difference in 7 day mortality between adult oncology ward patients with a MEWS Alarm and adult oncology ward patients without a MEWS alarm.

The primary outcome measure in this study is the number of patients who die within 7 days of a MEWS Alarm (7 day mortality).

The secondary outcome measure in this study is the number of patients admitted to ICU within 7 days of a MEWS Alarm.

This study will be conducted in a 26-bedded clinical oncology ward, a university teaching hospital in the Hong Kong Island of Hong Kong.

The patients consecutively admitted to clinical oncology ward from 18 January to 7 February 2015 are included in the study, the patients was counted 7 day mortality or 7 day ICU admission from the on set of reaching the MEWS Alarm. The patients with incomplete MEWS or discharge data are excluded.

A pilot study was conducted from 13 July at the time of 0am to 18 July at the time of 0am, 2014, 41 patients were admitted consecutively to the clinical oncology ward during the study period. 10 of them were excluded because of their incomplete MEWS chart, leaving 31 patients in the pilot study. 18 of them are male and 13 of them are female. 1 patient was transfer to ICU on her second day from admission, as the MEWS score increased by 3 from initial. Six patients had 7 day mortality with MEWS alarm and one patient had 7 day mortality without MEWS alarm in the study period. According to the sample size formulas and the calculators, the estimated sampling is 171 patients from 18 January to 7 February 2015.

MEWS is a standard nursing routine measurement by calculating the sum of five physiological parameters: systolic blood pressure, pulse rate, respiratory rate, body temperature and level of consciousness. The calculated MEWS score is being charted on an observation chart sheet. Nursing staff routinely collect MEWS score on admission and then once every day during hospital stay. The date of admission to ICU, and date of death are also collected .

In detail, blood pressure and pulse rate are measured electronically (Dinamap Pro 200 Critikon Vital Signs Monitor, U.S.A). The respiratory rate is counted over one minute. The tympanic temperature is recorded (ThermoScan, Type 6014, Braun, Germany), and the level of consciousness is based on the AVPU score: A is 'alert', V is 'response to verbal', P is 'response to pain' and U for 'unresponsive').

The MEWS score is counterchecked by doctors after doctors' assessment on admission and at the doctors' round everyday.

An ethical approval will be acquired from the CUHK and Queen Mary Hospital Clinical Research Ethics Committee. All the participants involved in the study will not experience any harm. Potential participants will be given written information regarding the nature of the study and the assurances confidentiality for the data collected. Written Consent will be signed. All report copies collected will be kept confidential. Patient can choose whether or not to be in this study. If they volunteer to be in this study, they may withdraw at any time without consequences of any kind of loss or benefits.

For data analysis, the 7 day mortality with reaching MEWS Alarm will compare with the ones without reaching MEWS Alarm by Chi-Squared Test. The relationship between MEWS and the mortality is statistically significant if the P value is lower than 0.05. Besides, same method of data analysis is used on ICU admission with or without reaching MEWS Alarm.

Inclusion Criteria:

• The patients consecutively admitted to clinical oncology ward from 18 January to 7 February 2015 are included in the study.

Exclusion Criteria:

• The patients with incomplete MEWS or discharge data are excluded.

Clinical oncology ward adult patients