This was a prospective, comparative observational study conducted at Shanxi Bethune Hospital between March 2020 and April 2023. A total of 86 elderly patients meeting the inclusion criteria were enrolled and divided into two groups: the AIDET group (n = 43) and the Conventional group (n = 43). Group assignment was performed using a simple alternation approach based on admission sequence or availability of personnel trained in the AIDET communication protocol. No formal randomization or allocation concealment was applied. The study did not involve blinding due to the nature of the intervention.

Due to the pragmatic and exploratory nature of this prospective observational study, a formal a priori sample size calculation or statistical power analysis was not performed. The final sample size of 86 patients (43 per group) was determined based on patient availability over the 3-year enrollment period and logistical feasibility within the study setting, using convenient sampling technique. Although this limited sample size may restrict the generalizability of findings and the detection of small effect sizes, statistically significant differences were observed across key outcomes, including oxygenation index, inflammatory biomarkers (HIF-1α, sTREM-1, hs-CRP), and weaning success.

Both groups received standard care, including ECG monitoring, phlegm reduction, anti-infection measures, and symptomatic treatment.

The Conventional group received standard invasive mechanical ventilation in the prone position. The ventilation settings were as follows: tidal volume of 8–10 mL/kg, pressure support of 8–40 cmH2O, and a respiratory rate of 8–12 breaths per minute. Patients were assisted into the prone position with their head down and feet elevated. Feeding and eating were stopped 30 min before ventilation to avoid reflux. During treatment, the patient’s arm, ankle, knee, head, face, iliac, and shoulder positions were adjusted to ensure comfort. Soft pads were placed as needed, and position changes were conducted every 1 to 2 h. Upper limbs were restrained when necessary to prevent accidental extubation.

AIDET communication and alternating prone ventilation treatment were used. (1) Establish an intervention team: Establish a multidisciplinary collaborative team for prone position ventilation treatment. The six core members of the multidisciplinary collaborative team include a critical care physician, head nurse, quality control team leader, accelerated recovery specialist nurse, and respiratory therapy specialist nurse, rehabilitation therapist, and other team members are the doctor in charge and the responsible nurse. Core members are responsible for program formulation, implementation and quality control. The doctor in charge and the rehabilitation therapist are mainly responsible for evaluating and determining the specific plan for prone position ventilation for patients; the head nurse and specialist nurses are responsible for quality control, feedback, and improvement of the implementation of the plan; and the responsible nurse is responsible for the implementation of the plan. (2) Formulate an alternating prone position ventilation plan: refer to the guidelines or consensus literature key points of RAOOF et al. (20), and based on evidence and combined with the clinical characteristics of elderly patients with severe pneumonia, formulate an AIDET communication model suitable for alternating prone position ventilation. The specific content is: A (Acknowledge): Greet the patient cordially, check the patient’s information, smile, and have a good attitude; considering that elderly patients may have hearing loss or cognitive function decline, they should speak slower and more clearly. Greet in the appropriate language and make sure there is adequate eye contact. And evaluate whether the patient has worries, uneasiness, anxiety and other emotions through language, behavior, expression, etc., and provide timely comfort and encouragement. I (Introduce): Elderly patients may not be familiar with medical terminology, so they need to use simpler, more life-like language when introducing, and repeat key information multiple times to enhance understanding. Explain to the patient in simple and easy-to-understand language the purpose and expected effects of alternating prone ventilation, as well as the matters that the patient needs to cooperate with. He also introduced in detail the clinical experience and professional experience of his team’s doctors and respiratory therapists to build patients’ confidence in cooperating with the medical team for treatment. D (Duration): Elderly patients have reduced physiological reserves and may have poor tolerance for prone position ventilation. A more detailed assessment of the patient’s overall condition, including cardiopulmonary function, skin condition, psychological status, nutritional status, and catheters and lines, is needed to ensure the safety of position changes during prone position ventilation. Assess the patient’s vital signs, oxygen therapy method, and whether the respiratory tract is unobstructed, and promptly remove secretions from the mouth, nose, and airways. Assess the skin at the pressure area, use foam accessories locally to reduce pressure, and assess the direction of turning in the prone position. E (Explanation): Explain the steps and precautions for prone position ventilation to the patient in detail, including how to cooperate with medical staff to flip the position. Make preparations before turning over, sort out various wires, drainage tubes, and infusion channels, clamp or remove non-emergency lines, and secure each catheter properly to prevent it from falling off. Align the direction of each catheter with the longitudinal axis of the body, and leave enough length for easy flipping. After confirming that the patient is able to perform the glute bridge exercise, the buttocks are lifted off the bed. Instruct older patients to understand and remember information related to the prone ventilation position through repetition and reinforcement. Use illustrations, demonstrations and other auxiliary means to help them understand the process and importance of prone position ventilation, and provide patients with pictures showing the prone position ventilation posture, see Figure 1 for details. Patients can freely choose 3 positions (prone position, chest and knee position, side prone position) for alternating prone position ventilation. Remove the electrode pads before prone positioning, and attach the electrode pads to the back while lying prone. Keep your head high and your feet low when lying prone to reduce head and face edema. Carry out prone position ventilation according to the specific measures in the plan, exclude eating and going out for examination and treatment, and extend the prone position ventilation time as much as possible. During the position change process, a rehabilitation therapist or senior nurse serves as the team leader, standing on the side of the patient’s head, responsible for issuing instructions and observing the patient’s reaction during the turning process, listening to the chief complaint, and evaluating various indicators. The other two people stood on both sides of the bed. One person is responsible for assessing whether the patient’s wires, pipes, infusion channels, etc. are unobstructed and in functional positions; one person is responsible for assisting the patient in prone positioning and placement of decompression tools. The team leader instructs the patient to turn to one side to complete the posture, and observes the patient at the bedside for 15 min to ensure that the patient’s position is comfortable. During the evaluation period, lung buckling or lung concussion decontamination treatment is performed. Patients who are obese or have weak mobility can be turned over by 4 to 5 people. Actively ask the patient if he or she has any doubts, encourage them to express them and patiently answer them. T (Thank you): After each treatment, thank the patient for his cooperation and efforts. Elderly patients may be worried about their health, and thanking them for their courage and efforts can boost their self-esteem and confidence. Prone ventilation was stopped when the patient failed to improve the oxygenation index or developed serious complications; the oxygenation index decreased compared with that in the flat position; the respiratory rate was greater than 40 breaths/min; the haemodynamics was unstable, very uncomfortable, and could not tolerate the prone position. The withdrawal operation was started when the patient met the conditions for withdrawal.

To ensure standardized implementation of the AIDET communication model, all healthcare personnel involved in the intervention—critical care physicians, nurses, and respiratory therapists—underwent structured training before the study commenced. The training program included a 4-h workshop led by certified clinical educators and a senior nurse educator, covering the AIDET components (Acknowledge, Introduce, Duration, Explanation, Thank you), communication role-play, and case-based simulations with elderly patients. Staff were required to demonstrate competency through supervised practice sessions and a checklist-based evaluation. Throughout the study, intervention fidelity was monitored via direct observations and weekly audits by the head nurse and quality control officer. Adherence to the AIDET protocol was assessed using a standardized compliance checklist (covering verbal communication structure, patient engagement, and explanation quality). Inter-rater reliability for fidelity scoring was established between two independent observers (Cohen’s kappa = 0.81). Feedback was provided in real time to staff to ensure consistent protocol delivery across all shifts. Deviations were logged and addressed through targeted retraining sessions when needed. These measures aimed to minimize variability in the execution of the AIDET communication model and strengthen internal validity.

The weaning time and success of weaning were collected from the two groups of patients. The observation time was set to 72 h, and the weaning time was from the time when the patients ventilation started to the time of successful weaning. Successful weaning follows the patient’s recovery of spontaneous breathing, good oxygenation, and no significant increase in PaCO₂, decrease in PaO₂, increased respiratory rate, increased heart rate, extreme fatigue, or excessive sternocleidomastoid muscle movement within 48 h after weaning from the ventilator, heart failure, arrhythmia, shock or worsening of consciousness and other symptoms.

GraphPad Prism 10 software was used for data processing, x ¯ ± s descriptions were applied to the measurement data, independent samples t-test was used for comparison between groups, paired samples t-test was used for comparison within groups at different times, non-normally distributed data with variance chi-square was used for non-parametric rank sum test, qualitative data was expressed as n (%), and the chi-square test was applied with the correction level of α = 0.05. Survival curve analysis was used to determine the success of the machine withdrawal in the two groups. Time difference.

Table 1 presents the baseline demographic and clinical characteristics of patients in the AIDET and Conventional groups. No statistically significant differences were observed between the two groups in terms of age, sex distribution, BMI, APACHE II score, or CPIS score (p > 0.05), indicating baseline comparability.

As seen in Figure 2A, there was no statistical significance in the comparison of the comfort level of patients in the two groups before ventilation and 1 h of ventilation, and the comfort level of patients in 6 h of ventilation (T6) and 12 h of ventilation (T12) increased, and the AIDET group was higher than the Conventional group, and the difference was statistically significant (p < 0.05). As seen in Figure 2B, the AIDET group was located above the Conventional group at all 3 time points after ventilation.

As seen in Figure 3, there was no statistical significance in the comparison of PaO₂/FiO₂ between the two groups before ventilation (T0), and the two groups of prone position ventilation for 24 h (T24) and ventilation for 72 h (T72) improved compared with the preventive period, but the AIDET group’s PaO₂/FiO₂ was significantly higher than the Conventional group, and the difference was statistically significant (p < 0.05).

There was no statistically significant difference in HIF-1α, sTREM-1, and hs-CRP levels between the two groups before ventilation (p > 0.05); the levels of HIF-1α, sTREM-1, and hs-CRP in the AIDET group at 72 h of ventilation (T72) were decreased compared with those in the Conventional group and were lower than those before ventilation in this group, and the difference was statistically significant (p < 0.05). See Figure 4 for details.

The withdrawal success rate in the AIDET group was 95.35% (41/43), significantly higher than the 76.74% (33/43) observed in the Conventional group, and the difference was statistically significant (Chi-squared = 13.285, p = 0.000) (Figure 5).

Despite the promising findings, this study has several limitations. First, the relatively small sample size and single-center design limit the generalizability of the results and may reduce statistical power to detect smaller but clinically relevant differences. Second, the absence of formal randomization and allocation concealment introduces the possibility of selection bias, and no stratification was performed based on baseline severity or comorbidities. Third, although baseline characteristics appeared comparable between groups, residual confounding from unmeasured variables (e.g., medication history, nutritional status, or underlying frailty) cannot be ruled out. Fourth, adherence to the prone ventilation protocol and fidelity to the AIDET communication framework may have varied among staff members, which could affect intervention consistency. Fifth, the study focused on short-term outcomes (72 h), and longer-term effects on survival, readmission, or lung function were not assessed. Finally, the study lacked blinding for patients and clinicians, which may have influenced subjective measures such as comfort scores. These limitations highlight the need for larger, multicenter randomized trials to validate the present findings and explore their long-term implications. Although fidelity monitoring and staff training were performed, complete standardization of human communication is inherently difficult. Minor variability in communication tone, experience level, and adherence may have influenced outcomes despite control measures.

Another important limitation of this study is its bundled intervention design. The AIDET communication model and alternating prone ventilation were introduced simultaneously in the intervention group, without separate control arms for each component. As a result, the individual contributions of AIDET communication and prone positioning to the observed improvements in oxygenation, comfort, inflammatory markers, and weaning success could not be disentangled. Future studies should consider a factorial or multi-arm randomized design to isolate and compare the specific effects of communication models and physical interventions such as prone positioning.

Another limitation of this study is the restricted temporal resolution of biomarker measurement. Inflammatory markers (HIF-1α, sTREM-1, and hs-CRP) were measured only at baseline (T0) and at 72 h (T72). This sampling strategy was chosen based on logistical feasibility and the assumption that 72 h would reflect early inflammatory response to the intervention. However, the lack of intermediate time points (e.g., 24 h or 48 h) prevents us from understanding the precise kinetics of biomarker changes—whether they declined steadily, peaked earlier, or were influenced by short-term clinical fluctuations.

The use of subjective measures such as self-reported comfort scores introduces the potential for reporting bias, particularly as no geriatric-specific validation of the analog scale was performed. Additionally, while “weaning success” was operationally defined using physiological criteria, interpretation may vary by clinical assessor, and the absence of assessor blinding could have introduced bias.