Development and Validation of the Questionnaire for Adaptive Hyperactivity and Goal Achievement (AHGA)

2.1. Description of the Sample

The investigation of the reliability and factor structure of the AHGA was carried out through its administration to a sample of 120 older adults selected for a previous randomized controlled trial (RCT) [16, 17], according to the Declaration of Helsinki and its revisions [18]. The Ethical Committee of the Institutional Review Board of the University Hospital of Cagliari, Italy, approved this RCT, which included subjects who provided written informed consent [16, 17]. Sample had previously been recruited by the Department of Medical Science and Public Health, University of Cagliari, Italy. The following inclusion criteria were implemented: elderly people aged 65 or older of both genders, living at home, and being able to provide informed consent without impairment for severe diseases (e.g., mild cases of hypertension or diabetes were admitted). Additional inclusion criteria included the absence of a lifetime history of diagnosed bipolar spectrum disorder conditions.

Exclusion criteria were: refusal to participate in the study (non-signing the informed consent for accepting participation in the study); participants with a diagnosis of bipolar spectrum disorder, and participants under 65 years of age. All included subjects provided written informed consent and were invited to fill out the Italian version of the Questionnaire for Adaptive Hyperactivity and Goal Achievement (AHGA).

2.2. Measures

Socio-demographic information was collected for the following variables: age, sex, socioeconomic status, and civil status. Measures of socioeconomic status were grouped into four categories: elementary school diploma, lower secondary diploma, upper secondary diploma, and university degree [19].

AHGA is a 12-item, five-point Likert scale (from 1 to 5), developed by a team of PhD students, psychotherapists, and psychiatrists with decades of expertise in bipolar spectrum disorders from the University of Cagliari. The final version was reached through a process of item development and selection followed by administering the tool to the same target population of subjects not directly enrolled in the validation phase. This stage was carried out carefully to identify potential issues or unclear terms or items in order to improve the quality and understandability of the questionnaire.

The questions mainly focus on two areas: lifelong adaptive hyperactivity and goal achievement. The tool investigates transfers, employment differences, and economic well-being compared to parents, as well as the perceived speed in doing things and thoughts flow. Other elements of the investigation are related to the usual number of people with whom they have conversations, the amount of weekly physical activity, the number of hours spent outside the home and the number of user applications or programs during the day. Elements such as how many times a person has been in love in their lifetime, without even being in an official relationship, along with the average number of hours slept during the night, are also AHGA components. The original Italian version is included as an appendix in the current paper.

Since this questionnaire has a role in measuring principally adaptive hyperactivity, the Biological Rhythms Interview of Assessment in Neuropsychiatry (BRIAN) serves to indicate particularly pathological hyperactivity and dysregulation of biorhythms, predisposing factors that could suggest the presence of a certain psychiatric disorder.

BRIAN is an 18-item interviewer administered-instrument that focuses on evaluating four main areas related to circadian rhythm disturbance: sleep, activities, social rhythms, and eating patterns. Items are estimated using a 4-point scale: (1)=no difficulty, (2)=mild difficulty, (3)=moderate difficulty, and (4)=severe difficulty. The BRIAN scores, therefore range from 1 to 72, where higher scores indicate serious circadian rhythm disturbances [20].

2.3. Statistics

Data were imputed in Excel and analysed using the Statistical Package for Social Sciences (SPSS) version 27. Additional analyses were carried out in R. All tests were two-tailed, with alpha set at p<0.05.

Means with standard deviation counts and percentages were used to describe the distribution of the data in the sample. An ANOVA, when appropriate, was used to determine whether continuous variables differed among groups of subjects. The Games-Howell test was used to test post-hoc differences in groups since it does not assume equal variances and sample size. Chi-square tests were used for categorical data. Pearson’s correlation coefficient was used for correlations. Reliability was measured as internal coherence using Cronbach’s alpha. For group comparisons, a rule of thumb assumes that reliability values of 0.70 are considered acceptable [21].

The data was preliminarily subjected to a principal component analysis (PCA) to establish the spontaneous distribution of the 12 core items of the AHGA into one or more separate dimensions. Parallel analysis was used to determine the optimal number of components. In a parallel analysis, the scree plot of the observed data was compared with that of a random matrix of the same size as the original. The best solution is based on the number of components with eigenvalues higher than those generated by the random data, either simulated or reassembled by permutation from the original data. The parallel analysis and the subsequent PCA were carried out with the psych package running in R [22, 23]. Both were applied to a matrix of polychoric correlations since the data were ordinal.

After establishing the factor structure of the Italian version of AHGA, the correlation of its measures with those of BRIAN was assessed to check the convergent validity of the scale and its factors.

3.1. Reliability

The reliability of the 12-item AHGA, measured as internal consistency, was good: Cronbach’s alpha in the sample was 0.713 (95%CI: 0.630 to 0.783).

3.2. Factor Analysis and Extraction of the Main Components

Bartlett’s test of sphericity was 288.7542 (p < 0.0001), Kaiser-Meyer-Olkin’s adequacy value was 0.78. The matrix, thus, can be factorized (Fig. 1).

The parallel analysis suggested a total of two principal components (Fig. 1). The main components were then analyzed by estimating two factors (Table 2).

The separation of the two factors with the varimax technique did not differ from the separation of the two factors with the Promax technique, which admits correlation between the two factors as probable based on their nature. Based on the direction of loading (assigned to the item only if loading is positive) and the relative weight, items 2, 8, 9, 10, 11, and 12 were assigned to the first factor or first component (goal achievement), while items 3, 5, 6, and 7 were assigned to the second factor or second component (hyperactivity), while items 1 and 4 were not clearly separated between the two factors (the two components) (Fig. 2).

Subscale 1, comprising items 2, 8, 9, 10, 11, and 12, was named “goal achievement” based on the thematic content items. Subscale 2, comprising items 3, 5, 6, and 7, was named “hyperactivity” based on the thematic content of the items. The score of the two subscales for each participant was obtained by adding the values of each item and dividing by the number of items to obtain a homogeneous interval from 1 to 5, with scores increasing as the expression of the variable measured by the subscale increases.

Fig. (3) summarizes the item distribution in the two components, as well as the variance fraction. The solution explained 41% of the total variance.

3.3. Relationship of the Main Components of the AHGA with Socio-demographic Indicators

The distribution of the score concerning subscale 1 (goal achievement) did not differ among men (mean ± SD: 3.6 ± 0.8) and women (3.4 ± 0.8); F[2;116]=, 683; p = 0,057. On the other hand, subscale 2 (hyperactivity) showed higher scores among men (3.1 ± 0.7) than women (2.8 ± 0.9); F[2;116]=,039; p = 0.014.

No relationship was observed in the distribution of the two subscales in relation to age: subscale 1 (Pearson r = 0.081; p = 0.380) and subscale 2 (Pearson r = 0.063; p = 0.493). Also, no difference was observed in the distribution of the scores of the two subscales concerning the civil status. Instead, there was observed an increase in the scores of the two subscales equivalent to the growth of the level of education. In particular, subscale 1 showed higher scores among those who had obtained upper secondary and university diplomas than those with lower educational levels. The subjects with university diploma seem to be more goal achievers, perhaps partly reflecting the chances/resources, but also the type of question (there is a question about the school level and one about income, which comes partly from the educational level) (Table 3a-3c).

3.4. Concurrent and Divergent Validity Analyses

We investigated the links between adaptive hyperactivity and goal achievement scale and a measure of the rhythms in bipolar disorder, the BRIAN. The analysis was done with Pearson’s correlation coefficient (r), using a threshold of p<0.05 for statistical significance. We found that the goal achievement subscale of the scale (factor 1) was negatively related to the BRIAN and its subscales, except for the rhythm (Table 4).

The hyperactivity subscale (factor 2) was less negatively related to the BRIAN and its subscales (Table 4). Essentially, greater propensity to goal achievement was related to greater disruption of sleep, general activity, sociality, and eating habits. Greater propensity to hyperactivity only disrupted general activity and sociality. The results support the convergent validity of this new scale, since the higher the propensity to display symptoms of bipolar disorder, as higher the impact on social rhythms.