Daily routine versus on-demand chest radiograph policy and practice in adult ICU patients- clinicians’ perspective

The scientific and ethical approval of the study was granted by the research committee of the Public Health College and King Abdullah International Medical Research Center (KAIMRC) with IRB approval # SP14/044. Administrative permission was also obtained from the necessary authorities of all participating hospitals, and participants’ informed consent was attached with the study survey. Most of the questionnaires were hand-delivered to the majority of the hospitals in Riyadh, while the remaining questionnaires were sent via email to designated persons in hospitals at other regions in the Kingdom of Saudi Arabia. An introductory letter informing the participants about the researcher and the purpose of the study was also attached. The distribution started in the middle of June 2014 and the last batch of the questionnaire was received during the third week of October 2014. Self-administered questionnaires were distributed to most of the tertiary hospitals of Kingdom of Saudi Arabia (KSA). The first part was about demographic characteristics questions such as gender, age, nationality, qualifications, work experience and hospital/ICU-setting questions, and the second part dealt with clinical questions concerning chest radiography and reasons for doing the procedure. At the end of the second part of the questionnaire, participants were asked if they would recommend changing their hospital policy from daily routine to on-demand chest radiograph in a ‘Yes’ or ‘No’ format (see Additional file 1). Demographic characteristics were included in order to analyze the relationship between responders' demographic characteristics and their opinion about the radiograph procedure. All clinicians working in the cardiac and pediatric ICUs, and all other healthcare providers whose primary assignment areas were not in the ICU were excluded from this study.

Data collected were entered into SPSS package version 20.0. Statistical analyses were done; descriptive and inferential analyses. The level of p ≤ 0.05 was used to indicate statistical significance. All variables were summarized and reported across the study subjects using descriptive statistics. The mean age of the participants was 35 ± 8 years. Categorical variables under (demographics, ICU characteristics, chest radiography data, and healthcare provider’s perception) were summarized and reported in terms of the frequency distribution (N (%). Chi-square / Fisher exact test was used to compare health care provider’s perceptions about on-demand chest radiography across categorical demographic characteristics. Results were reported in terms of count, percent, and p-value. For continuous variables (Age), normality assumption was first tested and verified using Kolmogorov-Smirnov test (P- < 0.0001), then independent sample T-test was used to compare between different perceptions, and the level of significance was set at p ≤ 0.05.

Out of 730 questionnaires sent to study respondents, we received 495 completed questionnaires representing 67.8% response rate of the anticipated participants responded to the survey. Table 1 shows the respondents and ICU characteristics. About two-thirds (66.6%) of respondents were females, the Saudi nationality was 14.5% and the rest were from more than six international regions, especially the Asians and they accounted for 59.3% and the remaining 40.7% clinicians were collectively from the Middle East, Europe, Africa and North America. Only 16.9% of the participants had a diploma and the rest of the participants had a bachelor degree or higher. Nearly 38.9% of the respondents worked at the Ministry of National Guard Health Affairs in Riyadh while the remaining participants were from different public and private hospitals across regions in Saudi Arabia. One-third of the respondents (35.5%) have more than 10 years working experience in the ICU, 36.1% have been working in the ICU for between 6 and10 years while 28.4% have work experience in the ICU between 1 and5 years. The majority of the respondents (92.3%) have been working in the ICU for more than 6 months. More than two-thirds (70.9%) of the ICU Units belong to academic hospitals compared to 24% who were from non-academic hospitals. About half (49.9%) are open format ICU (i.e. other doctors apart from intensivists can make treatment orders for ICU patients) and 46.9% are close format (only intensivists can make treatment orders). More than half of the respondents (51.2%) have been working in a hospital setting with more than 20 ICUs beds and only 3.2% were from the hospital with less than 5 ICU beds, while the remaining hospitals had ICUs with beds ranging between 5 and 20.

The top five Reasons for daily chest radiography at ICUs in Saudi Arabia were endotracheal intubation, insertion of a central venous line, chest tube, and tracheostomy and in ventilator deterioration (Fig. 1). A daily routine chest radiograph was performed in almost 96.8% of the participants ICUs as a daily routine. More than two-thirds (64%) of clinicians reported that daily routine chest radiograph was performed in their ICU for all admitted patients and 19.4% of clinicians said it was for an only specific patient group in comparison to 33.3% clinicians performing only for intubated and mechanically ventilated patients [Table 2]. As shown in Table 3, about one-third (35.9%) of the participants agreed that daily routine CXRs influences patient care, while similar percentage (35.2%) of the participants agreed to the same rating for the on-demand policy and the remaining in the both categories thought that CXRs influence was less than 60% for the ICU patient care management. Interestingly, the results of the question concerning chest radiograph essentiality for the judgment of the presence or absence of Acute Respiratory Distress Syndrome (ARDS) shows that 93.3% of the clinicians agreed on the importance of chest radiograph in the early detection and treatment which is very crucial to optimize treatment outcomes; because patients with ARDS are at a very high risk for complications such as pneumonia (92.3%) and pneumothorax (93.5%). (Table 3).

Table 3 also shows the health care provider standpoints towards changing to on-demand chest radiology instead of a routine chest x-ray. The majority (73.1%) of the surveyed clinicians agreed that the current daily routine chest radiograph policy for patients in the ICUs should be replaced by the on-demand chest radiograph policy compared to only 24% in favor of keeping the current daily routine policy. These findings showed a high percentage of support to replace the current policy across all the demographic characteristics in the study survey with no significant difference in the recommendation of on-demand chest radiography across subject’s demographic characteristics (P-value > 0.05). Table 4 shows the pattern of on-demand chest radiography practice in ICU. The majority of the clinicians (56.5%) disagreed that an on-demand chest radiograph is always performed after circulatory deterioration (Table 4).

Results also revealed that 78.9% of physicians working in the ICUs, 71.4% residents and 50% fellows agreed that the current policy should be replaced. A similar percentage was obtained from the ICU nurses, 75% in support of abandoning the current policy and 25% believed that the current policy should be retained. More than three-fourths (80%) of the respiratory therapists supported replacing the current policy with the on-demand chest radiograph protocol and only 20% disagreed with protocol changing (Table 5).

The selection of hospitals across the various regions in the Kingdom of Saudi Arabia and the good  percentage of response from the clinicians working in the Intensive care units were the main strengths of this study. On the other hand, our study has few limitations. Firstly, the majority of clinicians working in Saudi Arabia were expatriates from many different countries so the chest radiography protocol in practice in their home countries might have an influence on their perspective. Secondly, local factors related to how chest radiographs were ordered and obtained varies from different hospitals, and the level of technology in the medical imaging department was not the same across the board in Saudi Arabia. The obtained data in the current study were not validated against picture archiving and communication system. Moreover, we did not include the opinions of radiologists in the study, which is highly recommended to be surveyed in any future research .