Psychometric Evaluation of Chronic Patients Using the Persian Version of Patient Activation Measure (PAM)

Abstract

The Patient Activation Measure (PAM) is a 13-item questionnaire that assesses patients’ knowledge, skills, and confidence in self-management. The current study aimed to translate the American version of the PAM-13 into Persian and test the psychometric properties of the Persian version among chronic patients. This cross-sectional study was conducted on 438 chronically ill patients in Rafsanjan, Iran from May to November 2019. The American version of the PAM-13 was translated into Persian using a standardized forward–backward translation method. Internal consistency, test-retest reliability, face and content validity, as well as construct validity (structural and convergent validity) were all assessed. The content validity index of the Patient Activation Measure-13 Persian (PAM-13-P) was 0.91. Exploratory and confirmatory factor analyses showed that the PAM-13-P had a meaningful structural validity. The PAM-13-P scores were negatively correlated with the Partner in Health Measure (PIH) (r = −0.29, p < 0.001). In addition, the PAM13-P scores were positively correlated with the Satisfaction with Life Scale (SWLS) (r = 0.31, p < 0.001). The internal consistency was 0.88, and the repeatability was excellent [Intraclass Correlation Coefficient (ICC):0.96 and confidence interval (CI): 0.94–0.98]. This study demonstrates that the PAM-13-P is a reliable and valid measure for assessing activation among chronically ill patients. The PAM-13-P scale assesses the level of self-management of chronic patients and identifies appropriate care strategies to meet their needs.

Keywords

patient engagement patient activation validation known-group analysis inpatient chronic patients

Introduction

Chronic diseases are one of the leading causes of mortality and disability worldwide (Rizzuto et al., 2017), and they undoubtedly pose a major challenge to global health. Chronic diseases are the leading causes of poor health, disability and death from diseases, as well as higher healthcare costs in the United States (Raghupathi & Raghupathi, 2018). Chronic diseases such as cancer, diabetes, and respiratory and cardiovascular diseases account for approximately 70% of the global mortality (Organization, 2017). The number of chronically ill patients is increasing. According to a 2012 report, 29% of the European men and 33% of the European women had one or more chronic diseases or long-term health problems (Eurostat, 2016). According to the World Health Organization (WHO), chronic diseases are defined as conditions that include one or more of the following characteristics: ongoing illness and disability due to irreversible pathological changes, patient’s special education for rehabilitation, or ongoing medical attention and care (Hajat & Stein, 2018). Complications of chronic disease are affected by the emergence and prevalence of chronic diseases, related risk factors, and demographic characteristics such as aging (Ramsey et al., 2017). Chronically ill patients require self-management because of their long course of disease and treatment. Self-management is a key concept among chronically ill patients, which has resulted in better functioning and less reliance on expensive healthcare services over time (Hibbard et al., 2015).

Behavioral self-management improves one’s life, health, development, and well-being (Allegrante et al., 2019). Self-management of chronic diseases represents drug use, changes in lifestyle and health behaviors in order to prevent long-term complications and increase adherence to treatment regimens (Newland et al., 2021). Therefore, self-management seems to be the most effective method in achieving behavioral self-management changes (Allegrante et al., 2019), and patients require self-management skills and activation (Newland et al., 2021). Self-management plays a positive role in chronic patient care, leading to improved health status (McCabe et al., 2018) and care (Hibbard et al., 2015) in chronically ill patients.

The Patient Activation Measure (PAM) (22 items), one of the most important and well-known tools for assessing patient self-management, was developed by Hibbard et al. in the United States, who defined the term “activation” or more effective health care actors in 2004 and used the short version of the PAM-13 in 2005 (Hibbard et al., 2004, 2005). Patient activation is an important part in caring for chronically ill patients because it ensures that a patient has acquired the necessary skills, knowledge, and motivation to effectively participate in the care team (Newland et al., 2021). The PAM has focused on knowledge, skills, confidence, and vital behaviors for self-management and coping with a chronic disease (Hibbard & Tusler, 2007). The PAM assesses how a patient can manage their own health conditions (Hibbard et al., 2004). Chronically ill patients must actively participate in their self-care and management to achieve positive clinical outcomes (Wu et al., 2020).

High patient activation level indicates transition from a passive recipient of care to a confident self-care manager, and patients become active in many health behaviors that occur at higher activation levels (Hibbard et al., 2004), which are strongly associated with self-care and other health-related behaviors. Numerous studies have shown a correlation between higher scores of activations and health-related behaviors (Kim et al., 2016; Yadav et al., 2020). Therefore, policies and interventions are required to strengthen the role of patients in self-care management, based on which patient activation should be measured (Greene et al., 2015).

The PAM can be used to improve patient satisfaction and health behaviors (Magnezi & Glasser, 2014). Studies in the United States have shown the effectiveness of the PAM-13 in measuring patient’s self-management, so it has been translated into several languages [Germany, Austria, Switzerland (Brenk-Franz et al., 2013), the Netherlands (Rademakers et al., 2012), Denmark (Maindal et al., 2009), Italy (Graffigna et al., 2015), and Spain (Alegría et al., 2009)] (Rademakers et al., 2016). By using Cronbach’s alpha, the internal consistency of the PAM was 0.84, 0.88, 0.89, 0.88, 0.88 in three German-speaking countries (Germany, Austria, and Switzerland) (Brenk-Franz et al., 2013), the Netherlands (Rademakers et al., 2012), Denmark (Maindal et al., 2009), Italy (Graffigna et al., 2015), and Spain (Alegría et al., 2009), respectively. However, a comprehensive tool is required to assess the activation of chronically ill patients and to further research and develop self-management in Iran. Chronically ill patients’ self-management must be measured in order to reduce the burden on the healthcare system. Therefore, this study aimed to translate and evaluate the psychometric properties of the “patient activation measure” among chronically ill patients in Iran.

Methods

Study Design and Setting

This cross-sectional and methodological study was performed in teaching hospitals affiliated with Rafsanjan University of Medical Sciences in southeastern Iran in 2019. Two public-educational hospitals provide all medical services (specialized and general) to patients in Rafsanjan. The study aimed to translate and evaluate the psychometric properties of the Persian version of the PAM-13 among chronically ill patients in Iran.

Sampling

Sample size

Different sample sizes were used in different parts of the study. The sample sizes for the face validity, the qualitative content validity, the quantitative content validity, and test-retest reliability phases were 20 individuals, 10 experts, 15 experts, and 50 individuals, respectively. The largest sample size was available in the construct validity phase. More than 250 individuals are required to ensure a stable and robust estimate of the parameters (Brenk-Franz et al., 2013). Therefore, 300 samples were chosen for exploratory factor analysis. In general, confirmatory factor analysis necessitates 5–20 samples per estimated parameter (Brenk-Franz et al., 2013; Lee, 2007). Therefore, 130 samples were considered for confirmatory factor analysis. Totally, 520 questionnaires were distributed, with 500 of them returned, so a response rate of 96.15% was obtained. Following a review of 500 questionnaires, 62 questionnaires were excluded from the study due to confounding information and missing values. Finally, 300 samples underwent exploratory factor analysis, and 138 samples underwent confirmatory factor analysis.

Convenience sampling was used for sampling. Inclusion criteria were female/male patients aged above 18 years with one or more chronic diseases (based on the physician’s diagnosis and medical record documentation) [including ischemic heart disease (IHD), diabetes mellitus (DM), hypertension, congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD), multiple sclerosis (MS), rheumatoid arthritis (RA), cancer, psychological disorders]. Patients with dementia, cognitive impairment and active psychological disorders were not eligible to participate in this study.

Translation and Adaptation

For the first time, we prepared the Persian version of the PAM-13 using a forward-backward method. We began to translate after obtaining the necessary permission from Hibbard, the creator of the questionnaire (Hibbard et al., 2005). A panel of methodologists, health professionals, and a Persian-speaking translator translated the American version of the PAM-13 into Persian. The panel members agreed on the initial draft of translation. Two translators completed the backward translation independently. The panel went over all of the translations and revised the original version, which was approved by Hibbard. Fifteen chronically ill patients reviewed the Persian version in terms of the questionnaire items and format, content, clarity, language, and any other issues, and the panel approved the final Persian version of the PAM-13 based on the results.

Measures

Demographic and Clinical Variables

Participants’ age, sex, marital status, occupation, level of education, income level, Body Mass Index (BMI), presence of other diseases, number of admissions, and type of diagnosis were considered in the study.

Patient Activation Measure-13

Hibbard et al. (2005) developed the American short form of the PAM-13 to examine self-management. The PAM assesses patients’ knowledge, skills, beliefs, and confidence in health management and care. This measure consists of 13 items on the Likert scale, ranging from one (strongly disagree) to four (strongly agree). This measure shows four levels of activation: level 1 (patients believe that their roles are important: items 1 and 2), level 2 (patients have the confidence and knowledge to take action: items 3–8), level 3 (taking action: items 9–11), and level 4 (staying on course under stress: items 12 and 13). Answers are calculated on a scale of 0–100 using a standard metric system (0 being the lowest level of activation and 100 being the highest level of activation) with higher scores reflecting higher levels of patient activation. Answers 1 and 2 show low patient activation levels, while answers 3 and 4 show high patient activation levels. Recommended the PAM score ranges for levels 1, 2, 3 and 4 are 0–47.0, 47.1–55.1, 55.2–72.4, and 72.5–100, respectively (McCabe et al., 2018) (Supplemental Appendix: the PAM-P13 measure English and Persian version).

Partner in Health Measure

Battersby et al. (2003) used the PIH to assess patient self-management. This questionnaire includes 11 items on a nine-point Likert scale ranging from eight (very bad) to zero (very good), with a higher score indicating poor self-management. The scores on this questionnaire range from 0 to 88. Zakeri et al. (2022) confirmed the validity and reliability of the PIH. The validity of this questionnaire in Iran was determined using content validity and its reliability was 0.86 based on Cronbach’s alpha coefficient (Zakeri et al., 2022). In the current study, the Cronbach’s alpha for PIH was 0.87.

Satisfaction With Life Scale

Diener et al. (1985) used this scale to measure life satisfaction and reported the Cronbach’s alpha coefficient to be 0.87 (Diener et al., 1985). This scale assesses cognitive judgements of satisfaction with one’s life and consists of five items on a seven-point Likert scale ranging from one (strongly disagree) to seven (strongly agree). The possible range of scores is 5–35. Nooripour et al. (2021) in Iran showed acceptable validity and reliability of this scale (Cronbach’s alpha coefficient = 0.84) (Nooripour et al., 2021). The Cronbach alpha for the SWLS in this study was 0.84.

Data Collection

The researcher referred to two public hospitals with the necessary permits and collected the required samples from CCU, internal medicine, oncology, subspecialty and dialysis departments during different shifts (morning, afternoon and night). Data collection lasted from May to November 2019. The self-report questionnaires were completed by interviewing when a patient was unable to complete the questionnaire or when there were illiterate patients. In addition, we contacted individuals who took part in the test reliability to gather information for the second time.

Ethics

The present study was derived from a research project approved by the Research Ethics Committee of Rafsanjan University of Medical Sciences. Insignia Health LLC granted the researcher license No. (IR.RUMS.REC.1397.109) to translate the PAM-13 into Persian. After ethical considerations were approved, the researcher presented the chronic patients with a consent form. The following information is included on this form: (1) the study purpose and objectives, (2) the information confidentiality and (3) anonymous participation and withdrawal from the study at any time. The researcher explained the consent form to uneducated participants who then signed informed consent forms. No specific ethical issues arose during the study and data collection, and the researchers appreciated the study participants.

Statistical Analysis

SPSS 22 (SPSS Inc., Chicago, Illinois, USA) and LISREL version 8.70 (International Scientific Software, Chicago, Illinois, USA) were used for all analyses. The significance level of 0.05 was used in this study.

Validity

Face Validity

The face validity of the Persian version of the PAM was evaluated qualitatively. Twenty chronically ill patients completed the Persian version of the PAM-13. Then, based on their feedback on the items’ appropriateness, problems, relevance, and ambiguity, the necessary corrections were made. The time it would take to complete the questionnaire was also calculated.

Content Validity

The content validity of the Persian version of the PAM-13 was evaluated both qualitatively and quantitatively. Ten eligible experts (six PhD-prepared nurses, two psychologists and two physicians) evaluated the Persian version of the PAM-13 for grammar, appropriate words and phrases, and scoring (Colton & Covert, 2007). The questionnaire was then revised in response to their feedback. The content validity ratio (CVR) and the content validity index (CVI) were used to evaluate the quantitative content validity of the Persian version of the PAM-13. Fifteen experts rated each item of the PAM-13 based on a three-point scale (1 = not necessary, 2 = useful, but not necessary, 3 = necessary) (Almanasreh et al., 2019). If the number obtained from the Lawshe table is greater than 0.49, a phrase with a significance level (p < 0.05) will be necessary and important in this tool (Lawshe, 1975). The CVR was calculated using the following formula

CVR = \frac{n_{e} - \frac{n}{2}}{\frac{n}{2}}

(Where n_e is the number of experts who rated the item as “necessary” and n is the total number of experts).

CVI was used to calculate the three criteria of “simplicity and fluency”, “relevance” and “clarity or transparency”, with scores ranging from one (the lowest score) to four (the highest score) (Waltz et al., 2010). Items with scores greater than 0.79 were kept in the questionnaire. If the CVI score was between 0.79 and 0.70, the phrase must be corrected and revised, otherwise the phrase must be deleted (Jay Lynn et al., 2006). CVI was calculated using the following formula: “CVI = number of people who rated the item as 3 or 4/total number of evaluators"

Construct Validity

For structural validity, first, hidden factors were extracted using exploratory factor analysis. The Kaiser Meyer-Olkin (KMO) and Bartlett tests were calculated. KMOs between 0.7 and 0.8 were considered good, while KMOs between 0.8 and 0.9 were considered excellent (Cureton & D’Agostino, 2013). The hidden factors were then extracted using both Principal Component Analysis (PCA) and Principal Axis Factoring (PAF), with varimax rotation. The number of factors was determined using the following criteria: Eigen values greater than one, scree plots, and items with loadings of 0.4 or higher on any one factor (Green et al., 2010).

The construct validity was further assessed by the CFA, which tested the goodness of the structural equation model and the correlation between variables (items) and their underlying structures (subscales). The adequacy of the model was assessed using the chi-squared test. The Goodness-of Fit Index (GFI), Adjusted Goodness-of-Fit Index (AGFI), Comparative Fit Index (CFI), Incremental Fit Index (IFI), Non-Normed Fit Index (NNFI), Root Mean Squared Error of Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMR) are the main indices used to determine the fit of the model. Acceptable fit of the model is indicated by χ2/d.f. <3.0, RMSEA <.08, and SRMR <.05. The values of GFI, AGFI, CFI, IFI, and NNFI indices were ≥0.9 (Colton & Covert, 2007).

For convergent validity, we used the Spearman correlation coefficient to examine the correlation between the scores of the PAM-13, PIH, and the Satisfaction with Life scale. AS a lower PIH score shows higher self-management, convergent validity would be true if there was a negative correlation between the PAM-13 and PIH (inverse correlation) and a positive correlation between the Persian version of the PAM-13 and satisfaction with life (direct correlation).

Reliability

The use of consistency alone is not sufficient to assess reliability because it provides no information about participants’ response consistency (Lawshe, 1975). Therefore, we used internal consistency and test-retest. Internal consistency refers to the measurement of scale items in the same test, and our study used Cronbach’s alpha in final total sample to examine it (<0.70) (n = 438). The Intraclass Correlation Coefficient (ICC) was used to assess the retest reliability at 2-week intervals and the repeatability of the PAM-13 among 50 chronically ill patients. To interpret the coefficients obtained, we considered poor reliability for values less than 0.4, relatively good reliability for values between 0.4 and 0.7, and excellent reliability for values above 0.7 (Jay Lynn et al., 2006). In addition, the Spearman Correlation Coefficient, Independent t test, Mann Whitney U test, Kruskal–Wallis test and ANOVA tests were used to examine the association between the PAM-13-P, quantitative and qualitative variables.

Results

Face Validity/Content Validity

All the PAM-13-P items had acceptable quantitative face validity based on an item impact cut point of 1.5. The CVR scores for all the PAM-13-P items were greater than 0.49, ranging from 0.70 to 1.00. Furthermore, the CVI-S was 0.91, while the CVI-I was between 0.63 and 1.00.

Construct Validity: Study Population

The mean age of the participants was 62.21 ± 13.39 years. The majority of the participants were male (55.7%), married (85.6%), illiterate (45.0%), and had ischemic heart disease (N = 188, 42.9%) (Table 1). In addition, the PAM-13-P score was 56.49 ± 15.12. The PAM13-P levels were low in 44% (n = 193) and high in 56% of the participants (n = 245).

Table 1.

Demographic and Clinical Information of the Participants (N = 438).

Variables	Mean (SD)	PAM13-P
Variables	Mean (SD)	Spearman Correlation Coefficient	p Value
Age (yr.)	62.21 (13.39)	−0.01	p = 0.85
BMI	22.26 (3.57)	0.065	p = 0.17
Partner in Health (PIH)	46.71 (12.75)	−0.29	p < 0.001
Satisfaction with life (SWLS)	22.20 (6.57)	0.31	p < 0.001
	N (%)	Statistical test	p Value
Gender
Male	244 (55.70)	t = −0.49	p = 0.63
Female	194 (44.30)	t = −0.49	p = 0.63
Marital status
Married	375 (85.60)	t = −1.60	p = 0.12
Unmarried + divorce	63 (14.40)	t = −1.60	p = 0.12
Educational level
Illiterate	197 (45.0)	F = 1.08	p = 0.36
< diploma	135 (30.8)
Diploma	71 (16.2)
Academic	35 (8.0)
Employment status
Employed	208 (47.5)	F = 0.29	p = 0.75
Unemployed	151 (34.5)
Retired	79 (18.0)
Income (million riyal)
<0.5	144 (32.9)	H = 1.01	p = 0.60
0.5–1	105 (24.0)
1–2	125 (28.5)
>2	64 (14.5)
History of hospital stay (no)
0	190 (43.4)	F = 1.58	p = 0.18
1	61 (13.9)
2	58 (13.2)
3	48 (11.0)
>4	81 (18.5)
Other illnesses
Yes	255 (58.2)	t = 0.75	p = 0.46
No	183 (41.8)	t = 0.75	p = 0.46
Disability
Yes	36 (8.2)	t = −1.31	p = 0.19
No	402 (91.8)	t = −1.31	p = 0.19
Substance use
Yes	152 (34.7)	Z = - 1.66	p = 0.10
No	286 (65.3)	Z = - 1.66	p = 0.10
Diagnosis
IHD	188 (42.9)	F = 2.04	p = 0.07
Diabetes	55 (12.6)
Hypertension	73 (16.7)
CHF	45 (10.3)
COPD	41 (9.4)
Other^†	36 (8.2)

Notes. Data were presented numerically (%).t = Independent t test; Z = Mann Whitney U test; H = Kruskal–Wallis test; F = ANOVA test. *p < 0.05. Ischemic heart disease (IHD); Congestive heart failure (CHF); Chronic obstructive pulmonary disease (COPD).

^†[Chronic kidney disease (CKD); Multiple sclerosis (MS); Rheumatoid arthritis (RA); cancer; psychological].

Construct Validity: Structural Validity, Description of the PAM-13-P

The percentage of missing responses, frequency of response options, mean, standard deviation, and median were all used to evaluate the data for each item (Table 2). In general, the items were well-accepted, with missing percentages per item ranging from 0.0% (item 8) to 2.3% (item 2). The response option frequencies showed a low use of “strongly disagree” (from 2.3% to 10.3%) and a moderate use of “disagree” (13.7–28.3%). The “agree” option was selected mostly (36.7–50.3%), “strongly agree” was selected moderately (17.3–33.7%), and “not applicable” was selected infrequently (1.3–5.3%). The mean scores ranged from 2.77 (item 9) to 3.16 (item 7). The last five items of the scale had a lower mean score in comparison with the items at the beginning of the questionnaire. Except for item 4, eight items of the scale had higher mean scores in comparison with the items at the beginning of the questionnaire. The standard deviations ranged from 0.797 (item 3) to 1.10 (item 4).

Table 2.

Data description of the PAM13-P.

Questions	N	Missing Values	Strongly disagree	Disagree	Agree	Strongly agree	Not applicable	Mean	SD
Pam 1	298	2 (0.7)	28 (9.3)	41 (13.7)	125 (41.7)	94 (31.3)	10 (3.3)	3.06	0.981
Pam 2	293	7 (2.3)	11 (3.7)	59 (19.7)	136 (45.3)	78 (26.0)	9 (3.1)	3.05	0.861
Pam 3	298	2 (0.7)	9 (3.0)	56 (18.7)	151 (50.3)	77 (25.7)	5 (1.7)	3.04	0.797
Pam 4	298	2 (0.7)	31 (10.3)	77 (25.7)	110 (36.7)	70 (23.3)	10 (3.3)	2.84	1.010
Pam 5	299	1 (0.3)	13 (4.3)	61 (20.3)	113 (37.7)	96 (32.0)	16 (5.3)	3.14	0.947
Pam 6	299	1 (0.3)	5 (1.7)	61 (20.3)	133 (44.3)	86 (28.7)	14 (4.7)	3.14	0.853
Pam 7	299	1 (0.3)	11 (3.7)	48 (16.0)	130 (43.3)	101 (33.7)	9 (3.0)	3.16	0.861
Pam 8	300	0 (0.0)	7 (2.3)	59 (19.7)	138 (46.0)	86 (28.7)	10 (3.3)	3.11	0.837
Pam 9	295	5 (1.7)	23 (7.7)	85 (28.3)	129 (43.0)	52 (17.3)	6 (2.0)	2.77	0.899
Pam 10	296	4 (1.3)	19 (6.3)	74 (24.7)	131 (43.7)	65 (21.7)	7 (2.3)	2.89	0.901
Pam 11	295	5 (1.7)	10 (3.3)	72 (24.0)	147 (49.0)	62 (20.7)	4 (1.3)	2.93	0.800
Pam 12	297	3 (1.0)	12 (4.0)	68 (22.7)	134 (44.7)	77 (25.7)	6 (2.0)	2.99	0.856
Pam 13	299	1 (0.3)	24 (8.0)	62 (20.7)	130 (43.3)	77 (25.7)	6 (2.0)	2.93	0.930

Construct Validity: Structural Validity, Exploratory Factor Analysis

Bartlett’s test of sphericity was used to determine if the sample size was appropriate for factor analysis and whether the data followed a normal distribution. This test was statistically significant (χ2 = 1265.85, df = 78, p < 0.001). The KMO coefficient in this study was 0.84, confirming the factorability of the correlation matrix of the PAM-13-P. Both PCA and PAF with varimax rotation were used, and a three-factor solution with an Eigen value >1 was obtained (Table 3). The three-factor solution with the PCA extraction method explained 57.95% of the data variance, while the three-factor solution with the PAF extraction method explained 46.41% of the variance. According to scree plots, a three-factor solution was reasonable.

Table 3.

Rotated factor matrix: The PAM13-P.

Questions	Principal Component Analysis (Varimax Rotation)			Principal Axis Factoring (Varimax Rotation)
Questions	Factor 1	Factor 2	Factor 3	Factor 1	Factor 2	Factor 3
Pam 1			0.73			0.59
Pam 2			0.83			0.81
Pam 3			0.74			0.62
Pam 4		0.49	0.43
Pam 5		0.80			0.64
Pam 6		0.55			0.47
Pam 7		0.62			0.55
Pam 8		0.69			0.63
Pam 9	0.45
Pam 10	0.66			0.56
Pam 11	0.76			0.66
Pam 12	0.76			0.63
Pam 13	0.79			0.71
Eigen value	2.85	2.48	2.20	2.22	2.02	1.79
Percentage of explained variance	21.94	19.06	16.96	17.12	15.55	13.74
Cumulative percentage of explained variance	57.95			46.41

Construct Validity: Structural Validity, Confirmatory Factor Analysis

Following the identification of a three-factor solution via EFA, CFA was used to further test the factor model emerged from EFA. The second-order confirmatory factor analyses were used, in which a higher-order factor (the PAM) was assumed for the relationships between the individual factors (the three factors extracted from PAF). Goodness-of-fit indices were used to determine the degree of fit between the data and the results of the hypothesized models. All of the factor loadings were significant (t values >1.96). The χ2-associated p-value was less than the significance level of 0.05 (χ2 = 74.54, d.f. = 41, and p = 0.001). Except for RMSEA and PNFI, all fit indices were acceptable (χ2/d.f. = 1.82, RMSEA = 0.077, SRMR = 0.055, GFI = 0.91, CFI = 0.97, IFI = 0.97, NNFI = 0.96, PNFI = 0.70). Consequently, we could use the model to confirm the structure resulting from the exploratory factor analysis.

Construct Validity: Convergent Validity

Table 1 shows a negative correlation between the PAM13-P score and the Partner in Health (PIH) score (r = −0.29, p < 0.001), as well as a positive correlation between the PAM13-P score and the SWLS score (r = 0.31, p < 0.001).

Reliability

The Cronbach’s alpha value for the PAM13-P was 0.88. The PAM-13-P item-total correlations ranged from 0.46 (Item 4) to 0.63 (Item 11). The item-total correlations for 10 items of the PAM-13-P were ≥0.50. The test-retest reliability of the PAM-13-P was excellent at a 2-week interval, with an ICC of 0.96 [CI: 0.94–0.98].

Patient Activation Among Chronically Ill Patients

The mean patient activation score in the present study was 56.49 ± 15.12. The PAM-13-P score had no significant correlation with quantitative demographic characteristics and was unaffected when qualitative demographic characteristics were taken into account (Table 1).

Discussion

The present study aimed to translate and investigate the psychometric properties of the PAM-13-P among admitted patients with chronic diseases. To the best of our knowledge, this is the first study in the Iranian population, which assesses the activation and validation of the PAM-13-P among chronically ill patients. The results indicate acceptable psychometric properties of the instrument. The results of this study confirmed the results of previous studies on chronic patients (Kosar & Besen, 2019; Rademakers et al., 2012).

Data Quality

A low number of missing values confirmed the data quality. In addition, the response rate was high, which could be attributed to the chronic condition of inpatients. Moljord et al. (2015) found that patients with a high response rate were present in community mental health centers (CMHC). The ‘not applicable’ option was used less than 5% in previous studies (Moljord et al., 2015). Given that chronically ill patients require medication, medical treatment, and lifestyle changes, the lack of a ‘not applicable’ option was unsurprising.

The floor effect was 5.2% (ranging from 2.3 to 10.3%), but the ceiling effect was 26.19% (ranging from 17.3 to 33.7%). The ceiling effect of our study was slightly higher than that of Zeng et al. (2019) (25%), but lower than that of Moljord et al. (2015) (35%), indicating that future studies should prevent this ceiling effect among chronically ill patients evaluated with the PAM-13-P.

Construct Validity

The PCA-based three-factor solution explained 57.95% of the data variance, while the PAF-based three-factor solution explained 46.41% of the variance. Scree plots showed that a three-factor solution was reasonable. Consequently, we used the model to confirm the structure resulting from the exploratory factor analysis. Overall, EFA was significant. In addition, the structure of the factor was confirmed by CFA.

Brenk-Franz at al. found that the PCA revealed only one component, which explained 34.5% of the data variance (Brenk-Franz et al., 2013). Kephart et al. (2019) showed that the one-component solution explained 37.94% of the variance. Zeng et al. (2019) showed the four-factor CFA model. Differences in the studied samples, such as chronic or acute illness, physical or mental disorders, different cultural backgrounds, etc. can lead to different structures of the questionnaire. In addition, in present study and the study conducted by (Brenk-Franz et al., 2013), the eigenvalue was used for the numbers of extracted factors. Lance et al. (2006) stated that as many statistical packages supported accessibility, the eigenvalue-greater-than-1.00 criterion was still popular despite its tendency to retain too many factors. They suggested that the parallel analysis (PA) was an accurate alternative (Lance et al., 2006).

Convergent Validity

The results showed a negative significant correlation between the PAM-13-P and PIH as well as a correlation between the PAM-13-P and PIH scores, with lower PIH scores indicating better self-management among participants. Kephart et al. (2019) found a significant correlation between the PAM-13-P and PIH scores (r = 0.76; CI: 0.72, 0.80) as well as a positive significant correlation between the PAM-13-P item and life satisfaction scores. Therefore, the convergence validity of the PAM-13-P questionnaire was confirmed.

Internal Consistency and Repeatability

The internal consistency of the PAM-13-P was satisfactory (0.88), and the item–total correlation range was 0.46–0.63. The results of some studies showed that the PAM13-P was similar to other versions of the PAM-13, including Italian (Cronbach’s α = 0.88) (Graffigna et al., 2015), Dutch (Cronbach’s α = 0.88) (Rademakers et al., 2012), Singaporean (Cronbach’s α = 0.86) (Ngooi et al., 2017), Brazilian (Cronbach’s α = 0.83) (Cunha et al., 2019), American (Cronbach’s α = 0.81) (Prey et al., 2016), Turkish (Cronbach’s α = 0.81) (Kosar & Besen, 2019), Malay (Cronbach’s α = 0.79) (Bahrom et al., 2020) and German versions (Cronbach’s α = 0.79) (Bomba et al., 2018). The Spanish (Cronbach’s α = 0.93) (DeCamp et al., 2016) and Chinese versions of the PAM-13 (Cronbach’s α = 0.92) (Zeng et al., 2019) had lower internal consistency than other versions.

The item–total correlation coefficients for the PAM-13-P ranged from 0.46 to 0.63. Items with a total correlation coefficient of 0.30 must be used to distinguish individuals (Bujang & Baharum, 2017). In the present study, the item-total correlation coefficient for the PAM-13-P is similar to that reported in the literature. The item-total correlation coefficients were 0.48–0.65, 0.32–0.71, and 0.38–0.66 in the Danish (Maindal et al., 2009), the Korean (Ahn et al., 2015) and the Turkish versions, respectively (Kosar & Besen, 2019). The PA13-P reliability was evaluated using a test-retest over a 2-week period and the results showed that the PAM-13-P had excellent repeatability (ICC = 0.96). The PA13-P reliability in the Turkish version was 0.98 for each item for patients with diabetes, hypertension, or rheumatoid arthritis (Kosar & Besen, 2019).

Patient Activation Among Chronically Ill Patients

The mean patient activation score in the present study was 56.49 ± 15.12. The Chinese PAM-13 score for chronically ill patients was 60.1 ± 15.4 (Zeng et al., 2019). This difference could be explained by Zeng et al. who examined patients with high blood pressure and/or type 2 diabetes in community health centers. Furthermore, outpatients awaiting mental health treatment had a mean activation score of 51.93 ± 14.21 (Moljord et al., 2015), which was significantly lower than that in the current study. This difference indicates that patients’ activation levels may fall while waiting for treatment. Furthermore, culture can affect the results of activation, as Brenk-Franz et al. (2013) showed that the mean patient activation score in Germany, Austria, and Switzerland was 68.3 ± 14.8.

Limitations

One of the strengths of this study is that it provides a useful tool for measuring self-management in Iran. However, this questionnaire should be reviewed in clinical settings and future research should look into different aspects of this questionnaire. Longitudinal studies should be conducted on the effects of patient activation and changes in patient self-management practices over time. In this study, we used convenience sampling to collect data and to ensure that this result truly represented the population.

On the other hand, some precautions must be taken. Although our sample size was large enough for the analysis and validation process, we believe that our findings would have been strengthened if we had used different hospitals and geographical areas in Iran. Our study focused on self-management in admitted patients. Hospital conditions can affect patient self-management, so the generalization of results to outpatients should be done with caution. Despite our efforts to examine all types of chronically ill patients, our study excluded patients with Crohn’s disease, cystic fibrosis, acquired immunodeficiency syndrome (AIDS), and Parkinson’s disease, that should be considered in future studies. In addition, the eclectic inclusion of chronic groups may not be the best means of recruitment. Therefore, this study should be repeated to confirm that this result truly represents each population of chronically ill patients.

Conclusion

The Persian Patient Activation Measure 13 (PAM-13-P) appears to be a valid and reliable questionnaire to assess activation among chronically ill patients in hospital settings. The good psychometric properties and strong reliability of the PAM-13-P questionnaire support its use as a tool for determining the level of self-management of chronically ill patients and may provide useful information for hospital managers and researchers about chronically ill patient’s care programs. The current study should be considered as a pilot study because it is the first of its kind in Iran. Further studies within the given scope are required to validate its use as a tool for determining the level of self-management of chronic patients.

Practice Implications

Given that the PAM-13-P is a reliable and valid criterion for measuring activation among hospitalized patients in the hospital environment, it can be used to facilitate communication between patient and care provider. In addition, by determining the patient’s activation levels, the hospital can plan to increase patient self-management. This type of planning has the potential to improve patients' clinical outcomes, reduce readmissions, and increase patient participation in the course of treatment. Furthermore, examining the psychometric features of a scale in a variety of cultural and linguistic contexts can provide valuable insight and evidence concerning the validity and usability of a scale. A single scale can assess activation among chronically ill patients from different cultures and languages, which allows researchers to make a more accurate transcultural comparison.

Supplemental Material

Supplemental Material - Psychometric Evaluation of Chronic Patients Using the Persian Version of Patient Activation Measure (PAM)

Supplemental Material for Psychometric Evaluation of Chronic Patients Using the Persian Version of Patient Activation Measure (PAM) by Mohammad Ali Zakeri, Ali Esmaeili Nadimi, Golamreza Bazmandegan, Maryam Zakeri, and Mahlagha Dehghan in Evaluation & the Health Professions

Footnotes

Acknowledgments

Professor Hibbard et al. as well as Insignia Health, have graciously allowed us to use the PAM13. We would like to thank all of the participants who completed the PAM13-P as well as the Non-Communicable Diseases Research Center at Ali-Ibn Abi-Talib Hospital, Rafsanjan University of Medical Science, Rafsanjan, Iran for their support and collaboration.

Authors’ Contributions

MAZ, GB and MD designed the study. MAZ and MD wrote the manuscript. MAZ and MZ collected data. MD and MAZ provided statistical analysis. AEN and GB contributed to the study design. They provided critical feedback on the study and inputted to the draft of this manuscript. All authors have read and approved the final manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics Approval

Ethical approval was obtained from the Ethics Committee of Rafsanjan University of Medical Sciences (IR.RUMS.REC.1397.109).

Informed Consent

Before being included in the study, all participants received written and oral information and signed an informed consent form.

Availability of Data and Materials

All data analyzed during this study were included in this published article and its additional files.

ORCID iDs

Mohammad Ali Zakeri

Mahlagha Dehghan

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Ahn

Y.-H.

C.-H.

Ham

O.-K.

Kim

B.-J.

(2015). Psychometric properties of the Korean version of the “patient activation measure 13”(PAM13-K) in patients with osteoarthritis. Evaluation & The Health Professions, 38(2), 255–264. https://doi.org/10.1177/0163278714540915

Alegría

Sribney

Perez

Laderman

Keefe

(2009). The role of patient activation on patient–provider communication and quality of care for US and foreign born latino patients. Journal of General Internal Medicine, 24(3), 534–541. https://doi.org/10.1007/s11606-11009-11074-x

Allegrante

J. P.

Wells

M. T.

Peterson

J. C.

(2019). Interventions to support behavioral self-management of chronic diseases. Annual Review of Public Health, 40(1), 127–146. https://doi.org/10.1146/annurev-publhealth-040218-044008

Almanasreh

Moles

Chen

T. F.

(2019). Evaluation of methods used for estimating content validity. Research in Social and Administrative Pharmacy, 15(2), 214–221. https://doi.org/10.1016/j.sapharm.2018.1003.1066

Bahrom

N. H.

Ramli

A. S.

Isa

M. R.

Baharudin

Badlishah-Sham

Mohamed-Yassin

Abdul-Hamid

(2020). Validity and reliability of the patient activation measure^®(PAM^®)-13 malay version among patients with metabolic syndrome in primary care. Malaysian Family Physician: The Official Journal of the Academy of Family Physicians of Malaysia, 15(3), 22–34. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7735884/pdf/MFP-15-22.pdf

Battersby

M. W.

Ask

Reece

M. M.

Markwick

M. J.

Collins

J. P.

(2003). The partners in health scale: The development and psychometric properties of a generic assessment scale for chronic condition self-management. Australian Journal of Primary Health, 9(3), 41–52. https://doi.org/10.1071/py03022

Bomba

Markwart

Mühlan

Menrath

Ernst

Thyen

Schmidt

(2018). Adaptation and validation of the German patient activation measure for adolescents with chronic conditions in transitional care: PAM^® 13 for adolescents. Research in Nursing & Health, 41(1), 78–87. https://doi.org/10.1002/nur.21831

Brenk-Franz

Hibbard

J. H.

Herrmann

W. J.

Freund

Szecsenyi

Djalali

Steurer-Stey

Sönnichsen

Tiesler

Storch

(2013). Validation of the German version of the patient activation measure 13 (PAM13-D) in an international multicentre study of primary care patients. Plos One, 8(9), e74786. https://doi.org/10.1371/journal.pone.0074786

Bujang

M. A.

Baharum

(2017). Guidelines of the minimum sample size requirements for kappa agreement test. Epidemiology, Biostatistics and Public Health, 14(2), e12267-10. https://doi.org/10.2427/12267

10.

Colton

Covert

R. W.

(2007). Designing and constructing instruments for social research and evaluation (1st ed.). John Wiley & Sons.

11.

Cunha

C. M.

da Cunha

D. C. P. T.

de Oliveira Manzato

Nepomuceno

da Silva

Dantas

R. A. S.

(2019). Validation of the Brazilian version of the patient activation measure 13. Journal of Nursing Measurement, 27(1), 97–113. https://doi.org/10.1891/1061-3749.27.1.97

12.

Cureton

E. E.

D'Agostino

R. B.

(2013). Factor analysis: An applied approach. Psychology press. https://doi.org/10.4324/9781315799476

13.

DeCamp

L. R.

Leifheit

Shah

Valenzuela-Araujo

Sloand

Polk

Cheng

T. L.

(2016). Cross-cultural validation of the parent-patient activation measure in low income Spanish-and English-speaking parents. Patient Education and Counseling, 99(12), 2055–2062. https://doi.org/10.1016/j.pec.2016.07.003

14.

Diener

Emmons

R. A.

Larsen

R. J.

Griffin

(1985). The satisfaction with life scale. Journal of Personality Assessment, 49(1), 71–75. https://doi.org/10.1207/s15327752jpa4901_13

15.

Eurostat (2016). Eurostat website. http://ec.europa.eu/eurostat/. (Accessed 10 October 2016).

16.

Graffigna

Barello

Bonanomi

Lozza

Hibbard

(2015). Measuring patient activation in Italy: Translation, adaptation and validation of the Italian version of the patient activation measure 13 (PAM13-I). BMC Medical Informatics and Decision Making, 15(1), 109. https://doi.org/10.1186/s12911-015-0232-9

17.

Greene

Hibbard

J. H.

Sacks

Overton

Parrotta

C. D.

(2015). When patient activation levels change, health outcomes and costs change, too. Health Affairs, 34(3), 431–437. https://doi.org/10.1377/hlthaff.2014.0452

18.

Green

C. A.

Perrin

N. A.

Polen

M. R.

Leo

M. C.

Hibbard

J. H.

Tusler

(2010). Development of the patient activation measure for mental health. Administration and Policy in Mental Health, 37(4), 327–333. https://doi.org/10.1007/s10488-10009-10239-10486

19.

Hajat

Stein

(2018). The global burden of multiple chronic conditions: A narrative review. Preventive medicine reports, 12(2018), 284–293. https://doi.org/10.1016/j.pmedr.2018.10.1008

20.

Hibbard

J. H.

Greene

Shi

Mittler

Scanlon

(2015). Taking the long view: How well do patient activation scores predict outcomes four years later? Medical Care Research and Review, 72(3), 324–337. https://doi.org/10.1177/1077558715573871

21.

Hibbard

J. H.

Mahoney

E. R.

Stockard

Tusler

(2005). Development and testing of a short form of the patient activation measure. Health Services Research, 40(6 Pt1), 1918–1930. https://doi.org/10.1111/j.1475-6773.2005.00438.x

22.

Hibbard

J. H.

Stockard

Mahoney

E. R.

Tusler

(2004). Development of the patient activation measure (PAM): Conceptualizing and measuring activation in patients and consumers. Health Services Research, 39(4 Pt 1), 1005–1026. https://doi.org/10.1111/j.1475-6773.2004.00269.x

23.

Hibbard

J. H.

Tusler

(2007). Assessing activation stage and employing a “next steps” approach to supporting patient self‐management. The Journal of Ambulatory Care Management, 30(1), 2–8. https://doi.org/10.1097/00004479-200701000-00002

24.

Jay Lynn

Surya Das

Hallquist

M. N.

Williams

J. C.

(2006). Mindfulness, acceptance, and hypnosis: Cognitive and clinical perspectives. International Journal of Clinical and Experimental Hypnosis, 54(2), 143–166. https://doi.org/10.1080/00207140500528240

25.

Kephart

Packer

T. L.

Audulv

Å.

Warner

(2019). The structural and convergent validity of three commonly used measures of self-management in persons with neurological conditions. Quality of Life Research, 28(2), 545–556. https://doi.org/10.1007/s11136-018-2036-8

26.

Kim

J. Y.

Wineinger

N. E.

Steinhubl

S. R.

(2016). The influence of wireless self-monitoring program on the relationship between patient activation and health behaviors, medication adherence, and blood pressure levels in hypertensive patients: A substudy of a randomized controlled trial. Journal of medical Internet research, 18(6), e116. https://doi.org/10.2196/jmir.5429

27.

Kosar

Besen

D. B.

(2019). Adaptation of a patient activatıon measure (PAM) into Turkish: Reliability and validity test. African Health Sciences, 19(1), 1811–1820. https://doi.org/10.4314/ahs.v19i1.58

28.

Lance

C. E.

Butts

M. M.

Michels

L. C.

(2006). The sources of four commonly reported cutoff criteria: what did they really say? Organizational Research Methods, 7(9), 202–220. https://doi.org/10.1177/1094428105284919

29.

Lawshe

C. H.

(1975). A quantitative approach to content validity 1. Personnel Psychology, 28(4), 563–575. https://doi.org/10.1111/j.1744-6570.1975.tb01393.x

30.

Lee

S.-Y.

(2007). Structural equation modeling: A Bayesian approach (Vol. 711). John Wiley & Sons.

31.

Magnezi

Glasser

(2014). Psychometric properties of the hebrew translation of the patient activation measure (PAM-13). Plos One, 9(11), e113391. https://doi.org/10.1371/journal.pone.0113391

32.

Maindal

H. T.

Sokolowski

Vedsted

(2009). Translation, adaptation and validation of the American short form patient activation measure (PAM13) in a Danish version. BMC Public Health, 9(1), 209. https://doi.org/10.1186/1471-2458-9-209

33.

McCabe

P. J.

Stuart-Mullen

L. G.

McLeod

C. J.

Byrne

T. O.

Schmidt

M. M.

Branda

M. E.

Griffin

J. M.

(2018). Patient activation for self-management is associated with health status in patients with atrial fibrillation. Patient preference and adherence, 12, 1907–1916, https://doi.org/10.2147/PPA.S172970

34.

Moljord

I. E. O.

Lara-Cabrera

M. L.

Perestelo-Pérez

Rivero-Santana

Eriksen

Linaker

O. M.

(2015). Psychometric properties of the patient activation measure-13 among out-patients waiting for mental health treatment: A validation study in Norway. Patient Education and Counseling, 98(11), 1410–1417. https://doi.org/10.1016/j.pec.2015.06.009

35.

Newland

Lorenz

Oliver

B. J.

(2021). Patient activation in adults with chronic conditions: A systematic review. Journal of Health Psychology, 26(1), 103–114. https://doi.org/10.1177/1359105320947790

36.

Ngooi

B. X.

Packer

T. L.

Kephart

Warner

Koh

K. W. L.

Wong

R. C. C.

Lim

S. P.

(2017). Validation of the patient activation measure (PAM-13) among adults with cardiac conditions in Singapore. Quality of Life Research, 26(4), 1071–1080. https://doi.org/10.1007/s11136-016-1412-5

37.

Nooripour

Hosseinian

Ghanbari

Haghighat

Matacotta

J. J.

Gasparri

M. L.

(2021). Validation of the persian version of the satisfaction with life scale (SWLS) in Iranian women with breast cancer. Current Psychology, (2021), 1–8. DOI:10.1007/s12144-021-01662-2. https://doi.org/10.1007/s12144-12021-01662-12142

38.

Organization

W. H.

(2017). World health organization. Assessing national capacity for the prevention and control of noncommunicable diseases: Report of the 2015 global survey. http://apps.who.int/iris/bitstream/10665/246223/1/9789241565363-eng.pdf?ua=1

39.

Prey

J. E.

Qian

Restaino

Hibbard

Bakken

Schnall

Rothenberg

Vawdrey

D. K.

Creber

R. M.

(2016). Reliability and validity of the patient activation measure in hospitalized patients. Patient Education and Counseling, 99(12), 2026–2033. https://doi.org/10.1016/j.pec.2016.06.029

40.

Rademakers

Maindal

H. T.

Steinsbekk

Gensichen

Brenk-Franz

Hendriks

(2016). Patient activation in Europe: An international comparison of psychometric properties and patients’ scores on the short form patient activation measure (PAM-13). BMC Health Services Research, 16(1), 570. https://doi.org/10.1186/s12913-016-1828-1

41.

Rademakers

Nijman

van der Hoek

Heijmans

Rijken

(2012). Measuring patient activation in The Netherlands: Translation and validation of the American short form patient activation measure (PAM13). BMC Public Health, 12(1), 577. https://doi.org/10.1186/1471-2458-12-577

42.

Raghupathi

(2018). An empirical study of chronic diseases in the United States: A visual analytics approach to public health. International Journal of Environmental Research and Public Health, 15(3), 431. https://doi.org/10.3390/ijerph15030431

43.

Ramsey

M. L.

Conwell

D. L.

Hart

P. A.

(2017). Complications of chronic pancreatitis. Digestive Diseases and Sciences, 62(7), 1745–1750. https://doi.org/10.1007/s10620-017-4518-x

44.

Rizzuto

Melis

R. J.

Angleman

Qiu

Marengoni

(2017). Effect of chronic diseases and multimorbidity on survival and functioning in elderly adults. Journal of the American Geriatrics Society, 65(5), 1056–1060. https://doi.org/10.1111/jgs.14868

45.

Waltz

C. F.

Strickland

O. L.

Lenz

E. R.

(2010). Measurement in nursing and health research. Springer Publishing Company.

46.

Tang

Zhang

Jiang

Sedeh

S. S.

(2020). Self‐care management importance in kidney illness: A comprehensive and systematic literature review. Network Modeling Analysis in Health Informatics and Bioinformatics, 9(1), 1–13. https://doi.org/10.1007/s13721-020-00256-5

47.

Yadav

U. N.

Lloyd

Hosseinzadeh

Baral

K. P.

Bhatta

Harris

M. F.

(2020). Self-management practice, associated factors and its relationship with health literacy and patient activation among multi-morbid COPD patients from rural Nepal. BMC Public Health, 20(1), 1–7. https://doi.org/10.1186/s12889-020-8404-7

48.

Zakeri

M. A.

Abazari

Khoshnood

Dehghan

(2022). The effect of nursing continuous care on self-management in patients with myocardial infarction. Scientific Journal of Medical Sciences (SJMS), 1(2), 54–65. http://sjiums.irshums.ac.ir/article-51-19-en.html

49.

Zeng

Jiang

Zhou

Tian

Zhang

(2019). Measuring patient activation in Chinese patients with hypertension and/or diabetes: Reliability and validity of the PAM13. Journal of International Medical Research, 47(12), 5967–5976. https://doi.org/10.1177/0300060519868327

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.20 MB