Characteristics of Participants
Table 1 shows the participant characteristics. A total of 372 family caregivers participated in the study (12 participants in Phase I to investigate face validity, 120 participants in Phase II for instrument development, and 240 participants in Phase III for psychometric testing). Of all participating family caregivers, 89.0% (n = 331) were female, and 69.1% (n = 257) were > 30 years old. Most participants (n = 307; 82.5%) had a low educational background (below upper secondary school education), and 38.9% (n = 144) were unemployed. More than half (n = 266; 71.5%) of the caregivers were children of older adults.
Table 1
Characteristics of Participants
Characteristics of Participants
|
Participants (%)
|
|
Face validity (n = 12)
|
Phase II (n = 120)
|
Phase III (n = 240)
|
Total
(n = 372)
|
Gender
|
|
|
|
|
Male
|
3 (20.0)
|
15 (12.5)
|
23 (9.6)
|
41 (11.0)
|
Female
|
9 (80.0)
|
105 (87.5)
|
217 (90.4)
|
331(89.0)
|
Age category
|
|
|
|
|
< 20 years
|
0 (0.0)
|
7 (5.8)
|
14 (5.8)
|
21 (5.6)
|
20–30 years
|
3 (20.0)
|
28 (23.3)
|
63 (26.3)
|
94 (25.3)
|
31–40 years
|
6 (60.0)
|
33 (27.5)
|
54 (22.5)
|
93 (25.0)
|
> 40 years
|
3 (20.0)
|
52 (43.3)
|
109 (45.4)
|
164 (44.1)
|
Education
|
|
|
|
|
Primary education
|
3 (30.0)
|
82 (68.3)
|
147 (61.3)
|
232 (62.4)
|
Lower secondary education
|
2 (10.0)
|
15 (12.5)
|
58 (24.2)
|
75 (20.2)
|
Upper secondary education
|
7 (60.0)
|
16 (13.3)
|
27 (11.3)
|
50 (13.4)
|
Diploma
|
0 (0.0)
|
7 (5.8)
|
8 (3.3)
|
15 (4.0)
|
Occupation
|
|
|
|
|
Unemployed
|
4 (40.0)
|
48 (40.0)
|
92 (38.3)
|
144 (38.7)
|
Student
|
0 (0.0)
|
2 (1.7)
|
3 (1.3)
|
5 (1.3)
|
Employee
|
2 (20.0)
|
38 (31.7)
|
84 (35.0)
|
124 (33.3)
|
Self-employed
|
6(40.0)
|
32 (26.7)
|
61 (25.4)
|
99 (26.6)
|
Relationship with older adult
|
|
|
|
|
Spouse
|
0 (0.0)
|
17 (14.2)
|
28 (11.7)
|
45 (12.1)
|
Children
|
6 (50.0)
|
63 (52.5)
|
170 (70.8)
|
239 (64.2)
|
Other relatives
|
6 (50.0)
|
40 (33.3)
|
42 (17.5)
|
88 (23.7)
|
Knowledge domain
Table 2 shows the topics and items generated in the knowledge domain and their statistical analysis results at each stage. In the process of item development, four topics and 18 items were generated (items no. 1–18). In the process of content validation, of these 18 items, five items (item no. 2, 5, 9, 12, and 18) were deleted due to a Content Validity Index (CVI) of < 1 each, while one item was added to the supporting interface (item no. 19). Thus, 14 items remained content valid and therefore, were tested for construct validity in phase II.
Table 2
No
|
Domain and item generation
|
Phase I
|
Phase II
|
Phase III
|
I-CVI
|
Conclusion
|
Item difficulty
|
Discriminating index
|
Quality of response alternatives
|
Conclusion
|
Factor loading
|
“fit indices” of CFA
|
Cronbach’s Alpha
|
Topic: Definition and characteristic of older adult
|
|
Multifocal model with three factors.
CFI = 0.87
TLI = 0.82
RMSEA = 0.04
|
Overall : 0.83
F1 : 0.77
F2 : 0.78
F 3: 0.74
|
1
|
Older adults in Indonesia is defined as:
a. People aged ≥ 50 years
b. People aged ≥ 60 years*
c. People aged ≥ 70 years
|
1
|
Retained
|
0.40
|
0.09c
Cronbach’s Alpha if item deleted = 0.85
|
Options:
a = 0.25
c = 0.35
|
Deleted
|
Item deleted/
not tested
|
2
|
What is characteristic of older adults:
a. Have limited regenerative abilities*
b. Have limited activity daily living
c. Have limited interaction with people
|
< 1
|
Deleted
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
3
|
The normal changes that occur in the older adult’s skin are:
a. The skin becomes wrinkled and moist
b. The skin becomes wrinkled and gets wet easily
c. The skin becomes wrinkled and dries easily*
|
1
|
Retained
|
0.72
|
0.63
Cronbach’s Alpha if item deleted = 0.81
|
Options:
a = 0.19
b = 0.17
|
Retained
|
0.58
(factor 3)
|
Topic: Definition of PI and symptoms associated with PIs
|
|
|
|
|
|
|
|
4
|
A Pressure injury is:
a. An injury that occurs due to the use of diapers
b. An injury on the skin which usually occurs over a bony prominence as a result of pressure*
c. An injury that occurs due to pressed by tight clothes
|
1
|
Retained
|
0.30
|
0.50
Cronbach’s Alpha if item deleted = 0.82
|
Options:
a = 0.50
c = 0.20
|
Retained
|
0.47
(factor 3)
|
5
|
A pressure injury can also be defined as:
a. An injury on the skin over a bony prominence as a result of shear
b. An injury on the skin because of heat
c. An injury on the skin because of diabetic
|
< 1
|
Deleted
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
6
|
Symptom(s) of pressure injuries are:
a. The skin looks reddish
b. There is visible skin damage/wounds
c. Options A and B are correct*
|
1
|
Retained
|
0.72
|
0.68
Cronbach’s Alpha if item deleted = 0.80
|
Options:
a = 0.18
b = 0.10
|
Retained
|
0.42
(factor 3)
|
Topic: Cause and consequences of PIs
|
|
|
|
|
|
|
|
7
|
The cause of a pressure injury is:
a. Continuous pressure and shear against the skin*
b. Squeezed objects falling on the body
c. The pressure of clothes attached to the body
|
1
|
Retained
|
0.53
|
0.54
Cronbach’s Alpha if item deleted = 0.81
|
Options:
b = 0.13
c = 0.34
|
|
0.67
(factor 2)
|
8
|
Pressure injuries in older adults can cause:
a. Pain and infection*
b. Nausea and vomiting
c. Urinary incontinence
|
1
|
Retained
|
0.67
|
0.62
Cronbach’s Alpha if item deleted = 0.81
|
Options:
b = 0.14
c = 0.19
|
Retained
|
0.48
(factor 2)
|
9
|
Pressure injuries in older adults can make them:
a. Difficult to mobile*
b. Difficult to urinate
c. Difficult to concentrate
|
< 1
|
Deleted
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
10
|
What will happen if redness in the skin of older adults is left untreated?
a. It will develop into deep pressure ulcers*
b. It becomes blackish and then heals.
c. It will heal itself when the skin is dry.
|
1
|
Retained
|
0.08b
|
0.36c
Cronbach’s Alpha if item deleted = 0.83
|
Options:
b = 0.32
c = 0.60
|
Deleted
|
Item deleted/
not tested
|
Topic: Preventive strategies that family caregivers can perform to prevent PIs
|
|
|
|
|
|
|
|
11
|
What to do to prevent pressure injuries in older adults?
a. Wear loose clothes
b. Use footwear when leaving the house.
c. Prevent prolonged pressure on the skin*
|
1
|
Retained
|
0.38
|
0.49
Cronbach’s Alpha if item deleted = 0.82
|
Options:
a = 0.42
b = 0.20
|
Retained
|
0.44
(factor 1)
|
12
|
What to do to prevent redness on older adult’s skin?
a. Prevent shear on the skin
b. Prevent applying lotion on the skin
c. Prevent using hard mattress
|
< 1
|
Deleted
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
13
|
Pressure ulcers in older adults can also be prevented by:
a. Adequate feeding and drinking*
b. Sunbathing
c. Prevent stress on older adults
|
1
|
Retained
|
0.66
|
0.62
Cronbach’s Alpha if item deleted = 0.81
|
Options:
b = 0.13
c = 0.21
|
Retained
|
0.92
(factor 1)
|
14
|
For immobile/ bedridden older adults, what should be done to prevent pressure injuries?
a. Mobilization to the left and right sleeping position*
b. Positioning the older adults always sleeps on their back without any wedge
c. Let older adults sleep without being disturbed.
|
1
|
Retained
|
0.49
|
0.58
Cronbach’s Alpha if item deleted = 0.81
|
Options:
b = 0.31
c = 0.20
|
Retained
|
0.95
(factor 1)
|
15
|
A thing that should be done on older persons' dry skin to avoid pressure injuries is:
a. Apply powder to keep the skin dry
b. Moisturizes dry skin*
c. Cover the dry skin with a bandage
|
1
|
Retained
|
0.53
|
0.60
Cronbach’s Alpha if item deleted = 0.81
|
Options:
a = 0.23
c = 0.24
|
Retained
|
0.44
(factor 1)
|
16
|
A thing that should be done when an older persons' skin turns red is:
a. Let it dry itself
b. Release pressure and shear*
c. Give betadine or iodine.
|
1
|
Retained
|
0.75
|
0.70
Cronbach’s Alpha if item deleted = 0.81
|
Options:
a = 0.15
c = 0.10
|
Retained
|
0.45
(factor 1)
|
17
|
A thing that should be done if the skin of an older adult shows deep pressure injury is:
a. Take the older adults to health care services*
b. Treat using honey
c. Let it open
|
1
|
Retained
|
0.71
|
0.59
Cronbach’s Alpha if item deleted = 0.81
|
Options:
b = 0.17
c = 0.12
|
Retained
|
0.95
(factor 1)
|
18
|
What to do to prevent deep pressure injuries in older adults:
a. Consult the injuries to health care provider*
b. Apply a traditional medicine like honey or coffee
c. I do not know the answer
|
< 1
|
Deleted
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
Item deleted/
not tested
|
19
|
Using a special mattress for older adults can prevent pressure injuries. This statement is:
a. True*
b. False
c. I do not know
|
1
|
Added
|
0.49
|
0.58
Cronbach’s Alpha if item deleted = 0.81
|
Options:
b = 0.36
c = 0.15
|
Retained
|
0.92
(factor 1)
|
Note : |
* correct answers |
a Value of person correlation (r-value) smaller than r-table |
b Value of item difficulty smaller than 0.10 or larger than 0.90 |
c Value of discriminant index (D value) smaller than 0.40 |
In the data analysis of phase II, two items (items no.1 and no.10) were identified who had item-total correlation much lower than those of the other items. These two items did not contributed to the Cronbach’s alpha; means that if these two items were deleted, the Cronbach’s alpha was higher. Item no.10 also had a difficulty index < 0.10. As a result, item no.1 and 10 were removed from the instrument, leaving 12 items with a good difficulty index (mean = 0.57), good discriminating index (mean = 0.59) and good distractors of the multiple-choice alternatives. The dimensionality of all 12 items were tested using Principal Factor Analysis (PFA), generating three factors for the Eigenvalue greater than 1 (as shown in Fig. 2). Factor 1 represented topic about PI prevention (item no.6,7,8,9,10), factor 2 was about cause and consequences of PIs (item no.7 and 8), and factor 3 related to characteristic of older adults and PI (item no.1,2,3). These three factors together explained 64.4% of total variance. Factor loading of these 12 items ranged from 0.40–0.92.
A Confirmatory Factor Analysis with three factor was run for these 12 items in phase III among 240 family caregivers. The results showed that a model of three factor was accepted (CFI = 0.87; TLI = ; 0.82 and RMSEA = 0.04). The internal consistency of Cronbach's alpha was 0.83. All these results show that the KAP-PI instrument in the knowledge domain with 12 items can be used as an instrument to measure family caregivers' knowledge prevention among community-dwelling older adults.
Attitude domain
Table 3 shows the items generated in the item development and psychometric evaluation of the items in the attitude domain. At the beginning, ten items (items no.1–10) were generated. Based on the content validity, two (items no.2 and 6) of these items were deleted due to the CVI < 1 of each, while one item was added to the supporting interface (item no.11), leaving nine items to be included in phase II.
Table 3
No
|
Domain and item generation
|
Phase I
|
Phase II
|
Phase III
|
PFA (n = 120)
Factor Loading
|
Reliability
(Cronbach’s Alpha)
|
CFA (n = 240)
Factor Loading
|
Reliability
(Cronbach’s Alpha)
|
“fit indices” of CFA
|
1.
|
I am responsible for the health of the older relative in my house
|
Retained
(I-CVI = 1)
|
0.69
|
0.92
|
0.66
|
0.93
|
Tucker-Lewis Index/ TLI = 0.83
Comparative Fit Index/ CFI = 0.87
RMSEA = 0.02.
|
2.
|
I am responsible for pressure injury problems in the older relative in my house
|
Deleted
(I-CVI < 1)
|
Item deleted/
not tested
|
Item deleted/
not tested
|
3.
|
The personal hygiene of the older relative in my house must be cared for carefully
|
Retained
(I-CVI = 1)
|
0.71
|
0.68
|
4.
|
I have to pay attention to the skin moisture and hygiene of the older relative in my house.
|
Retained
(I-CVI = 1)
|
0.79
|
0.75
|
5.
|
It is important to pay attention to the food and drink of the older relative in my house
|
Retained
(I-CVI = 1)
|
0.88
|
0.83
|
6.
|
I am responsible for the nutritional problem in the older relative in my house
|
Deleted
(I-CVI < 1)
|
Item deleted/
not tested
|
Item deleted/
not tested
|
7.
|
Pressure injuries on the older relative in my house should be prevented
|
Retained
(I-CVI = 1)
|
0.87
|
0.86
|
8.
|
Helping the older relative in my house in their activities and movements is my responsibility
|
Retained
(I-CVI = 1)
|
0.79
|
0.81
|
9.
|
Immobile older relative in my house need to be helped in movement and positioning
|
Retained
(I-CVI = 1)
|
0.77
|
0.77
|
10.
|
The older relative in my house who experience pressure injuries need to be checked to health care service
|
Retained
(I-CVI = 1)
|
0.80
|
0.80
|
11.
|
The older relative in my house who are at risk of getting pressure injuries need a special mattress to prevent pressure injuries
|
Added
(I-CVI = 1)
|
0.78
|
0.75
|
A Principal Factor Analysis (PFA) with oblique rotation was run for phase II. All requirements for performing PFA were met (KMO MSA = 0.89; Bartletts Test of sphericity (Sig.) = 0.00). All nine items tested had factor loading > 0.40 of each (0.64–0.87), which means that no items had to be deleted. As shown in Figure. 3, these nine items constructed one factor for the Eigenvalue greater than 1 and explained 62.13% of the total variability, which is higher than the required 60% (44)).
In phase III, all nine items were tested among 240 family caregivers. A Confirmatory Factor Analysis was run resulting in “fit indices” including Tucker-Lewis Index/ TLI of 0.83; Comparative Fit Index/ CFI of 0.87; and RMSEA of 0.02. These three “fit indices” indicated that the unidimensional model which resulted from the PCA in phase II, was confirmed as the model fit. The internal consistency of the final versions of the KAP-PI instrument in the attitude domain had a Cronbach's alpha of 0.93, indicating high reliability. In conclusion, all validity and reliability test results indicate that the nine items can be trusted as a means of measuring family caregivers' attitudes toward pressure injury prevention in community-dwelling older adults.
Practice domain
Table 4 shows the items generated in the practice domain and its psychometric evaluation. Four topics and 13 items (items no.1–13) were developed in phase I. Of the 13 items, one item (item no.13) was deleted were deleted due to the CVI < 1 of each, while one item was added to the supporting interface (item no.14), leaving 13 items to be tested for phase II. A PFA with oblique rotation was run in phase II. All requirements for performing PFA were met (KMO MSA = 0.85; Bartletts Test of sphericity (Sig.) = 0.00). When looking to the factor loading of each item, item no.12 had a factor loading < 0.40 (0.14), indicating the item should be deleted from the instrument. After deleting item no.12, the PFA with oblique rotation was re-run and this second PFA resulted in a factor loading > 0.40 of each item (0.41–0.82). As shown in Fig. 4, all 12 items constructed three factors for the Eigenvalue greater than 1, i.e. factor 1 (item no. 6, 7, 8,and 9), factor 2 (item no.1, 2, 10, 11, and 12), and factor 3 (item no. 3, 4, and 5). These three factors explained 66.11% of the total variability (required not less than 60%(44)).
Table 4
No
|
Domain and item generation
|
Phase I
|
Phase II
|
Phase III
|
PFA (n = 120)
Factor Loading
|
Reliability
(Cronbach’s Alpha)
|
CFA (n = 240)
Factor Loading
|
Reliability
(Cronbach’s Alpha)
|
Fit indices of CFA
|
Factor 1
|
Factor 2
|
Factor 3
|
Factor 1
|
Factor 2
|
Factor 3
|
|
Topic: Activities performed to support older adults to meet
their nutritional and fluid needs
|
|
0.87
|
|
Overall:
0.89
Factor 1:
0.80
Factor 2:
0.73
Factor 3:
0.87
|
Tucker-Lewis Index/ TLI = 0.83
Comparative Fit Index/ CFI = 0.87
RMSEA = 0.06.
|
1.
|
Provide healthy food for the older relative in my house
|
Retained (I-CVI = 1)
|
|
0.74
|
|
|
0.51
|
|
2.
|
Provide mineral water for the older relative in my house at least 8 glasses in a day
|
Retained (I-CVI = 1)
|
|
0.53
|
|
|
0.55
|
|
|
Topic: Activities performed to support older adults in
mobilization and repositioning
|
|
|
3.
|
Helping the older relative in my house to do activities if they cannot do it him/herself
|
Retained (I-CVI = 1)
|
|
|
0.86
|
|
|
0.87
|
4.
|
Helping for the older relative in my house to move if they cannot do it him/herself
|
Retained (I-CVI = 1)
|
|
|
0.78
|
|
|
0.82
|
5.
|
Helping the bedridden older relative in my house to change their position (positioning) regularly if they cannot do it him/herself
|
Retained (I-CVI = 1)
|
|
|
0.61
|
|
|
0.87
|
|
Topic: Activities performed to support older adults in skin
hygiene and moisture care
|
|
|
6.
|
Prevent the older relative in my house from using damp and wet clothes, including changing diapers regularly (if they use diapers)
|
Retained (I-CVI = 1)
|
0.75
|
|
|
0.64
|
|
|
7.
|
Prevent long pressure on the body of the older relative in my house
|
Retained (I-CVI = 1)
|
0.85
|
|
|
0.87
|
|
|
8.
|
Moisturizing the skin of the older relative in my house by giving lotions/oils
|
Retained (I-CVI = 1)
|
0.76
|
|
|
0.85
|
|
|
9.
|
Check the entire skin of the older relative in my house for redness
|
Retained (I-CVI = 1)
|
0.71
|
|
|
0.67
|
|
|
|
Topic: Activities performed to support older adults to maintain their health and ensure access to health care services
|
|
|
|
10.
|
Maintain the environmental hygiene of for the older relative in my house
|
Retained (I-CVI = 1)
|
|
0.68
|
|
|
0.67
|
|
|
11.
|
Took the older relative in my house to health services if they suffer from wounds
|
Retained (I-CVI = 1)
|
|
0.61
|
|
|
0.74
|
|
12.
|
Took the older adult to health services if they had a health problem
|
Retained (I-CVI = 1)
|
MSA = 0.45 (< 0.50)
Item was not included in PCA.
|
|
Item deleted/
not tested
|
|
13.
|
Contact health care provider to get their suggestion according to condition of the older relative
|
Deleted (I-CVI < 1)
|
Item deleted/
not tested
|
|
Item deleted/
not tested
|
|
|
Topic: Support older adult to have a special mattress
|
|
|
|
|
14.
|
Provide a special mattress for a bedridden elderly relative in my house
|
Added (I-CVI = 1)
|
|
|
0.64
|
|
|
0.49
|
|
|
Only the highest factor loading is shown. |
In phase III, all 12 items were tested among 240 family caregivers. A Confirmatory Factor Analysis was run to check whether the three factors resulted by Principal Factor Analysis (PFA) was a good model or not. The CFA resulted in “fit indices”, i.e. Tucker-Lewis Index/ TLI of 0.83; Comparative Fit Index/ CFI of 0.87; and RMSEA of 0.06. These three “fit indices” indicated that the three factor model resulting from the PCA in phase II, was confirmed as the model fit for practice domain. The correlations between factors were > 0.60. The internal consistency a Cronbach's alpha of factor 1, 2 and 3 were 0.80, 0.73, and 0.87 respectively. The overall internal consistency of the final version of the KAP-PI instrument in the practice domain had a Cronbach's alpha of 0.89, indicating high reliability. In conclusion, all validity and reliability test results indicate that the 12 items can be trusted as a means of measuring family caregivers' practice toward pressure injury prevention in community-dwelling older adults.