Cleaning work, particularly in hospital settings, is an occupation that demands high physical load on both the musculoskeletal and cardiovascular systems. Hospital cleaning workers are at high risk for musculoskeletal disorders (MSDs) due to prolonged standing, heavy lifting, bending, pushing-pulling, and repetitive movements. This situation reduces employees' quality of life and leads to workforce loss and economic burden on the health system.

Despite this, the risk levels of cleaning workers are often not adequately assessed and preventive interventions remain limited. Therefore, the aim of this study is to examine the effect of the Posture Protection Program (POSKOP), developed for hospital cleaning workers, on musculoskeletal discomfort, ergonomic risk levels, and postural awareness.

The study is planned as a quasi-experimental design to be conducted with 60 cleaning staff working in two public hospitals. Ergonomic risk assessment will be performed using the Ovako Working Posture Analysis System (OWAS) and will be objectively analyzed through developed software.

The intervention program consists of multi-component elements such as training, individual observation, instant feedback, exercise practices, visual materials (posters and pocket cards), and digital reminders (WhatsApp messages). The intervention is expected to reduce employees' musculoskeletal discomfort levels and increase their ergonomic awareness.

AIM / RATIONALE:

The aim of this study is to examine the effect of the posture protection program on musculoskeletal disorders, ergonomic risk levels, and postural awareness among hospital cleaning workers.

The hypotheses of the study are as follows:

H1-1: Following the posture protection program, the intervention group will show a reduction in musculoskeletal disorders compared to the control group.

H1-2: Following the posture protection program, the intervention group will show a reduction in ergonomic risk levels compared to the control group.

H1-3: Following the posture protection program, the intervention group will show an increase in postural awareness compared to the control group.

SUBJECT, SCOPE AND LITERATURE REVIEW:

Cleaning work is physically demanding and exhausting for both the musculoskeletal and cardiovascular systems. Musculoskeletal disorders (MSDs) are a prevalent health problem among hospital cleaning workers due to prolonged standing, heavy lifting, bending, repetitive movements, and inappropriate body posture.

Studies conducted in different countries have shown that a high proportion of hospital cleaning workers experience musculoskeletal symptoms, particularly in the lower back, shoulders, knees, and neck. These problems are commonly associated with repetitive movements, prolonged standing, frequent bending, heavy lifting, and insufficient rest periods. In addition, postural problems such as forward head posture have been reported to be associated with factors such as high body mass index and long working hours. These findings demonstrate a strong relationship between ergonomic risk factors and musculoskeletal disorders among cleaning workers.

Various intervention strategies have been developed to prevent MSDs caused by physical workload and non-ergonomic working conditions. The literature indicates that adapting equipment and tasks to suit the worker reduces ergonomic risks, and that ergonomics training helps decrease physical strain and prevent occupational health problems. Interventions focusing on improving postural habits and promoting physical activity have been shown to significantly improve workers' posture-related behaviors. Improvements are particularly observed in correct bending techniques, maintaining spinal alignment, alternating arm use, and using support when necessary. Additionally, structured exercise and stretching programs have been associated with reduced pain levels and increased motivation among workers.

These findings highlight the importance of multi-component interventions that combine education, physical exercise, ergonomic adjustments, and continuous feedback. Such approaches are considered more effective in achieving long-term improvements in posture and reducing the risk of musculoskeletal disorders. However, existing studies in Turkey are largely descriptive, and there is a lack of experimental research evaluating comprehensive posture protection programs specifically designed for hospital cleaning staff. Therefore, this study was planned to determine the effect of a posture protection program on musculoskeletal disorders, ergonomic risk levels, and postural awareness among hospital cleaning workers.

METHOD:

1.1. PHASE I In the first phase of the study, it is planned to develop the educational content and materials of the "Posture Protection Program" and to test the validity of its scope and quality. In addition, a software program will be developed to determine the ergonomic risks of the participants using the Ovako Working Posture Analysis System (OWAS).

1.1. Development of the Posture Protection Program

1.2. In the first phase of the study, the educational content will be developed by the researchers through a review of documents from the International Labour Organization (ILO), the United States Occupational Safety and Health Administration (OSHA), the Republic of Turkey Ministry of Labour and Social Security, and the Occupational Health and Safety Centre (İSGÜM), along with relevant literature, laws, and regulations pertaining to posture protection.

Posture Protection Program - Training Content

Topic Sub-topics

1. Definition and importance of ergonomics

   · Definition of ergonomics . Importance of ergonomics for cleaning staff

2. Musculoskeletal disorders

   • Causes and risk factors
   • Ergonomic risk factors in cleaning work

3. Importance of correct posture

   • Correct standing posture
   • Key considerations during bending, turning, and reaching movements
   • Correct heavy lifting technique
   • Examples of incorrect posture and associated risks

4. Ways to prevent musculoskeletal disorders

   • Use of ergonomic equipment
   • Ergonomic cleaning principles
   • Daily exercise and breaks
   • Postural awareness and self-assessment

Development of OWAS-Compatible Software for Ergonomic Risk Assessment

Within the scope of this study, an artificial intelligence-supported software program is planned to be developed for the objective assessment of participants' ergonomic risk scores in accordance with the Ovako Working Posture Analysis System (OWAS) method. Support for the software will be provided by Ahmet İnci, who serves as an instructor at Juventus Technical College and works as a software engineer and software architecture trainer at Digicomp.

Using the developed software, video recordings obtained from two cameras during participants' cleaning activities will be analyzed at specific time intervals (e.g., every 0.25-0.5 seconds) and divided into image frames. The resulting frames will be processed through a custom-developed software incorporating MediaPipe-based pose estimation algorithms. The software will automatically detect body segments and joint points in each frame, analyze back, arm, leg, and load-handling positions, and encode these postures according to the OWAS classification system. Within the OWAS framework, each posture will be assigned to ergonomic risk categories ranging from 1 (low risk) to 4 (very high risk).

Through this method, the risk profile of postures exhibited by workers throughout their tasks will be determined, the frequency and timing of high-risk movements will be identified, and video-based OWAS risk assessments will be conducted. This process provides an objective and reproducible measurement infrastructure that minimizes observer error. In addition, the accuracy of the developed pose estimation model will be evaluated using the MPII human pose dataset, and the similarity between model outputs and ground truth labels will be calculated using the Object Keypoint Similarity (OKS) metric. This evaluation process will be conducted based on a structure similar to the COCO evaluation framework. In addition to automated validation, the images obtained in the field will also be manually reviewed by experts, and the visual accuracy of the joint predictions produced by the system will be verified. In this way, particularly for complex positions where the automated system has limitations, model outputs will be qualitatively supported, and the consistency of the software will be tested.

PHASE II In the second phase of the study, a quasi-experimental study will be conducted to determine the effect of the posture protection program on musculoskeletal disorders, ergonomic risk levels, and postural awareness among hospital cleaning workers.

Study Design This study is planned as a non-randomized experimental study with a pre-test post-test control group design.

Study Population and Sample Selection The study population consists of a total of 163 cleaning workers employed in the inpatient wards of two public hospitals in the province of Isparta (65 at the university hospital and 98 at the city hospital). A power analysis was conducted using the G-Power program to determine the sample size. For this purpose, the effect size of 1.92, representing the reduction in participants' low back pain rates following health education, based on a previously published study, was used. With 95% power and a 5% margin of error, a sample of 27 participants per group was found to be sufficient. However, to account for the possibility of data loss, it was decided to increase this number by approximately 10%, resulting in a total sample of 60 participants, with 30 assigned to each group.

Data Collection Tools

A descriptive information form developed through a review of the literature, the Cornell Musculoskeletal Discomfort Questionnaire to assess musculoskeletal complaints, and the Ovako Working Posture Analysis System (OWAS) to determine participants' ergonomic risk levels will be used for data collection. In addition, the Postural Habits and Awareness Scale (PHAS) will be applied to assess participants' knowledge and awareness regarding ergonomic posture, and a Weekly Self-Assessment Form will be used to collect feedback on the intervention process.

Data Collection

Pre-test data collection for the study is planned for December 2026, following the receipt of necessary permissions and the establishment of the required infrastructure. Pre-test data will be collected on weekdays during working hours at times deemed appropriate by hospital management. The Descriptive Information Form, the Cornell Musculoskeletal Discomfort Questionnaire, and the Ergonomic Awareness Questionnaire will be obtained based on participants' self-reports. For ergonomic risk assessment according to OWAS using the non-participant observation method, video recordings of participants during cleaning activities will be taken and risk levels will be determined with software support. Following the implementation of the Posture Protection Program with the intervention group, post-tests are planned to be administered in the same manner. After the post-tests are completed, posture protection training will also be provided to the control group.

Research Intervention Protocol (Posture Protection Program)

The Posture Protection Program is a four-week posture-focused program comprising a two-session training on musculoskeletal disorders, the importance of ergonomics, and ergonomic cleaning guidelines, along with visual reminders, video demonstrations, spontaneous training sessions, stretching exercises, and a self-assessment process.

Data Analysis

The data obtained as a result of the study are planned to be analyzed using the SPSS 23.0 (Statistical Package for Social Sciences) software program. In the video recordings, OWAS scores will be calculated by observing each worker for a minimum of 15 minutes, and video recordings taken during participants' cleaning activities will be analyzed at specific time intervals (e.g., every 0.25-0.5 seconds) and divided into image frames. The risk profile of postures exhibited by workers throughout their tasks will be determined, the frequency and timing of high-risk movements will be identified, and a video-based OWAS risk distribution will thus be obtained for objective ergonomic analysis. A normality analysis of the data will be conducted, and within-group and between-group comparisons will be performed according to the type of data. A significance level of 95% (or an error margin of α = 0.05) will be used to determine differences in the analyses.

Inclusion Criteria:

Employed as an active hospital cleaning staff member for at least 12 months Working in inpatient hospital units Willing to participate in the study and providing written informed consent voluntarily Working in daytime shifts during the study period

Exclusion Criteria:

Having received training on the prevention of musculoskeletal disorders within the last 12 months Having congenital or acquired physical disabilities Having a previously diagnosed musculoskeletal disorder (MSD) Not performing routine cleaning duties (e.g., supervisors or administrative staff)

Plan Description

Individual participant data (IPD) will not be shared due to confidentiality and privacy considerations. The study includes occupational video recordings and health-related assessments of hospital cleaning workers; therefore, protecting participant anonymity and maintaining data security is essential. Access to the dataset will be restricted to the study investigators and authorized personnel only.

Blinding and Bias Prevention Since the intervention and control groups will be drawn from different hospitals, participants will be unaware of their group assignment and will thus be blinded. However, the principal investigator of the study will not be able to be blinded due to the nature of the study. To minimize the risk of bias in the results, video recordings of participants will be taken to assess their MSD risk and will be objectively evaluated through the software to be developed; this is also important in terms of preventing subjectivity and error in the assessment of these recordings. In addition, a code will be assigned to the study groups and data analysis is planned to be carried out by an independent statistician who is unaware of the group names, thereby also aiming to prevent bias in the interpretation of the results.