Enhancing knowledge and attitudes through the ‘healthy nutrition’ education program: a Solomon four-group experimental study with older adults at a social life campus | BMC Geriatrics

The study was conducted using an experimental method to examine the effect of healthy nutrition education on the nutritional awareness and attitudes of older adults living in a the SLHO located within the İBSLC.

This study utilized a Solomon four-group design with two experimental groups and two control groups to measure the effectiveness of the Healthy Nutrition Education Program. This design is ideal for comparing the effects of the intervention between experimental and control groups with pretest and posttest measurements. The group structure is as follows: Experimental Group 1 (T1: Pretest & T2: Posttest, n = 14), Experimental Group 2 (T5: Posttest Only, n = 16), Control Group 1 (T3:Pretest & T4:Posttest, n = 20), and Control Group 2 (T6:Posttest Only, n = 19). This design helps to isolate the effects of the educational intervention and controls for time-dependent changes, so that before and after the intervention, measurements were taken in both experimental and control groups. Fig. 1 illustrates the distribution of participants and the intervention process (Fig. 1) [14, 15].

Research design flow chart

The study was conducted in a SLHO located within the İBSLC of the Izmir Metropolitan Municipality between 01 February 2023 and 01 June 2024.

The population of the study includes 400 older adults staying in the SLHO in İBSLC. The research sample consisted of individuals aged 60 years and older without psychiatric or neurological disease. A total of 69 participants, 30 in the experimental group and 39 in the control group, voluntarily participated in the study.

Healthy nutrition education practice

The training was organized in two sessions lasting 2 h in total, conducted on August 3, 2023. The first session was planned as a traditional presentation-based training and the second session was planned as an interactive question and answer session. The training was organised for the older adults in the experimental group, while the control groups were measured before and after receiving the intervention.

The didactic (first) session of the Healthy Nutrition Education Program focused on enhancing the foundational knowledge of older adults about healthy eating. The following topics were covered:

This structured content was designed to provide comprehensive knowledge tailored to the physiological and lifestyle needs of older adults. Visual aids with power point presentation and examples enriched the session, making the content clear and engaging for participants.

In the interactive second session, a physician and a nutrition and dietetics specialist facilitated a discussion. Initially, participants were asked questions to assess their understanding and boost their confidence. An interactive question-and-answer environment was created, allowing participants to ask questions and engage in multiprofessional learning. This approach fostered a collaborative and supportive atmosphere, encouraging active participation and practical learning.

Data collection was carried out in two stages:

• Sociodemographic Data Form: It was used to collect basic information such as age, gender, and educational status of the participants.

• Healthy Nutrition Attitude Scale (HNAS) and Nutrition Knowledge Scale (NKS): It was applied to measure nutritional knowledge levels and attitudes.

The HNAS and NKS used in the study served as important measurement tools to evaluate the differences between the experimental and control groups. The HNAS, developed and validated by Demir and Cicioğlu (2019), consists of 21 items across four subscales: Nutrition Knowledge (NK), Nutrition-Related Emotions (NRE), Positive Nutrition (PN), and Poor Nutrition (PNR). The scale uses a 5-point Likert system, with scores ranging from 21 to 105, where higher scores indicate a more positive attitude towards healthy nutrition [16].

The NKS, developed by Yılmaz et al. (2021), evaluates nutritional knowledge across three domains: food and nutrient knowledge, food preparation methods, and nutrition-health relationships. It consists of 28 items rated on a 5-point Likert scale, with reverse coding applied to 10 items. Scores range from 0 to 126, with higher scores reflecting greater nutrition knowledge. Categories are defined as low (< 79), medium (79–89), high (90–100), and very high (> 101) knowledge levels [17].

Both instruments are reliable and valid tools for assessing attitudes and knowledge about nutrition in adult populations, ensuring the robustness of the study’s data collection process. The Cronbach’s alpha for the NKS was reported as 0.85, HNAS was reported as 0.89, indicating high internal consistency reliability for the scales [16, 17].

Strategies to minimize bias

A randomized controlled group design was implemented to minimize systematic differences between the intervention and control groups. To ensure external validity and reduce potential pre-test sensitization effects, the Solomon four-group design was utilized. Participants were assigned to groups based on a block-based randomization approach, where four blocks of the SLHO were allocated to one of the study groups (EG1, EG2, CG1, CG2). Room assignments within the blocks are determined by the SLHO administration based on application order and availability, rather than demographic factors such as age, gender, or health conditions., ensuring an unbiased and balanced distribution of participants.

Data collection process

Surveys and scales were administered following standardized protocols to ensure objective and unbiased measurements. Pre-test assessments were conducted immediately before the intervention on August 1–2, 2023, for Experimental Group 1 (T1) and Control Group 1 (T3) to establish baseline data on participants’ nutritional attitudes and knowledge. Post-test assessments were conducted two weeks after the intervention, on August 16–17, 2023, for all groups (T2, T4, T5, T6), allowing participants sufficient time to process the educational content and reflect on their learning while minimizing memory bias.

Statistical analysis

IBM SPSS-25 and JASP 0.16.4.0 statistical package programs were used for data analysis. The suitability of the data for normal distribution was checked by Shapiro-Wilks test and skewness and kurtosis values. It is understood that the data obtained from the pre-application and post-application of the scales used in the study showed normal distribution. Since the Shapiro-Wilks p value was greater than 0.05 in all experimental and control groups, the data showed normal distribution. When the skewness and kurtosis values of the pre-test and post-test scores of the scales used (HNAS and NKS) are analysed, it is seen that the determined values are suitable for normal distribution [18].

The differences between the experimental and control groups were analyzed using dependent and independent sample t-tests. Hypothesis testing, based on Solomon’s experimental design, involved coding data for each group and performing a series of t-tests on the HNAS and NKS groups (T1-T6). With simultaneous testing, the error coefficient was automatically controlled. Four hypotheses were tested to assess internal and external validity. A significant difference between T1 and T2 would indicate the effectiveness of the training, with other factors controlled [15].

The following hypotheses guided the research:

Intervention effectiveness

Main Hypothesis (H₁₁): Training is effective; there is a significant increase in Healthy Nutrition Attitude Scale (HNAS) and Nutrition Knowledge Scale (NKS) scores for T1 post-training compared to T2, while no significant change is expected in HNAS and NKS scores between control groups T3 and T4.

Null Hypothesis (H₀₁): There is no significant difference in HNAS and NKS scores between T1 and T2 before and after training.

Pre-test effect

Main Hypothesis (H₁₂): Pre-test scores on HNAS and NKS differ significantly among groups, suggesting a potential pre-test effect.

Null Hypothesis (H₀₂): Pre-test scores on HNAS and NKS do not significantly differ among groups (T3 = T4; T4 = T6; T1 = T4), indicating no pre-test effect.

Time/Process effect

Main Hypothesis (H₁₃): HNAS and NKS scores change significantly over time across groups, indicating a time-based process effect.

Null Hypothesis (H₀₃): HNAS and NKS scores do not change over time across all groups (T3 = T6; T1 = T6), suggesting no time-based process effect.

Group homogeneity

Main Hypothesis (H₁₄): Baseline HNAS and NKS scores differ significantly among groups, suggesting initial group differences.

Null Hypothesis (H₀₄): Baseline HNAS and NKS scores are similar among groups (T3 = T6; T1 = T6; T1 = T3), indicating homogeneity and equal conditions across groups at the beginning of the study.

The study design and group assignments were structured to monitor changes in experimental and control groups, enabling assessment of training impact, pre-test influence, and time-dependent effects, while maintaining baseline similarity. Details are provided in Fig. 1.

Homogeneity of groups: T3 = T6; T1 = T6; T1 = T3 hypotheses were formulated.