The unpredictable nature of arrhythmias can cause a profoundly negative impact on a person’s daily life, leading to impaired health-related quality of life (HRQOL).1–3 Some patients have vague and unspecific symptoms during arrhythmia episodes, whereas others experience more disabling and handicapping symptoms such as pronounced tiredness, dizziness, near syncope, and sometimes even complete syncope, irrespective of type of arrhythmias. In patients with a bothersome symptoms, there can be negative effects on social as well as physical and mental domains.2,4–6 Never knowing when the next episode will occur can lead to feelings of insecurity and self-imposed restrictions in daily life.6,7 Living with recurrent arrhythmia episodes often leads to insecurity with a need to find coping strategies to manage the episodes.8
Health-related quality of life refers to function and well-being in connection with a disease or treatment and includes evaluation of well-being with regard to physical, mental, and social domains of life.9,10 Health-related quality of life is a multidimensional concept and is a measurement of a person’s subjective perception of health and how certain diseases and treatments impact on their health status.11–13
Assessment of HRQOL can provide valuable information about healthcare interventions, support the choice of treatment strategies, and assess adverse treatment reactions.14,15 Generic and disease-specific HRQOL scales can be used separately or in combination, where the questionnaires need to be properly evaluated to ensure that assessments are reliable and valid.11,16 Generic scales make it possible to compare patients with different diagnoses. Disease-specific scales make it possible to assess to what extent a certain disease influences a patient’s daily life situation.11,17 As generic scales are limited in their ability to detect important clinical changes, they should be supplemented with disease-specific scales.18 To avoid increasing respondent burden, short scales with sound psychometric properties are needed.
One problem concerning disease-specific scales for patients with arrhythmias is that it is not always known what type of arrhythmia the patient experiences and some have more than 1 arrhythmia diagnosis.19 Scores from disease-specific scales that target only 1 arrhythmia diagnosis may therefore be difficult to interpret. To avoid this problem, a disease-specific scale targeting different forms of arrhythmia is needed. This type of scale should also make it possible to compare HRQOL among patients with different forms of arrhythmias. Most existing disease-specific scales in arrhythmia patients only describe symptom burden, or assess HRQOL, are developed for atrial fibrillation (AF), and do not include other forms of arrhythmias or mix the symptoms with the HRQOL concerns within the same scale.3,20–25 In clinical practice and research, it would be valuable to have a validated arrhythmia-specific HRQOL scale suitable for different forms of arrhythmias. To the best of our knowledge, no such scale is available and validated today.
The aim of the study was to develop and validate a disease-specific scale to evaluate HRQOL, suitable for patients with different forms of arrhythmias.
The overall Arrhythmia-Specific questionnaire in Tachycardia and Arrhythmia (ASTA) was developed to assess the impact of arrhythmias on a patient’s symptom burden and HRQOL. The development of the ASTA symptom burden scale is described elsewhere,5 whereas the study reported in this article describes the development and validation of the ASTA HRQOL scale.
The Regional Ethical Review Board at the Faculty of Health Sciences, Linköping, Sweden, approved the protocol and gave permission for obtaining oral informed consent from participants (study number: M170-08 T111-08). Study participation was documented in the patients’ medical records. The study complies with the Declaration of Helsinki.26,27
The Development Phase
The aim during the development phase was to develop a scale that included both physical and mental aspects of HRQOL. To reach content validity and identify representative and relevant items, a literature review was conducted in Medline, searching for quality of life, health-related quality of life, arrhythmias, supraventricular tachycardia, and paroxysmal supraventricular tachycardia. In addition, results from a previous interview study including 300 arrhythmia patients before radiofrequency catheter ablation (RFA) treatment were used to describe how the arrhythmias influenced patients’ daily life situations.6
The research group, consisting of experienced cardiac nurses knowledgeable in psychometrics and an electrophysiologist, created the initial pool of items from the literature review and the patient interviews. The items were thereafter evaluated by an expert panel consisting of cardiologists, nurses, and arrhythmia patients referred for RFA treatment. The expert panel was asked to suggest additional items for the item pool if they found any domains or concerns that were not covered. The patients were encouraged to suggest new questions if they thought there were concerns missing at the time of data collection. During the initial and validation phases, the patients were also repeatedly asked for comments on the questionnaires’ readability and understandability. The expert panel was instructed to judge the appropriateness of the items for patients with different forms of arrhythmia.
One question about intimacy and sexual life was added to the pool of items. The expert panel also gave recommendations for changes to improve clarity and the logical order of the items.
A common response scale was created for all items. The response scale included the alternatives: no (0), yes, to a certain extent (1), yes, quite a lot (2), yes, a lot (3), or I don’t know. The last category was included as a result of patient recommendations, so that those who were not sure if the item was related to arrhythmia would still be able to respond. This category was not scored. Furthermore, the patients were instructed to rate how much the arrhythmia affected their daily life (“due to your arrhythmia”), but no timeframe was given. If the ASTA HRQOL scale is used separately (apart from its associated burden scale), we recommend that a timeframe be given in addition to the initial questions in part I (ASTA descriptive part). All patients start with part I, where they are asked about their most recent arrhythmia episode and current medication. For those using the whole ASTA questionnaire, including the symptom burden scale, there are questions about the number of episodes during the last 3 months and the duration of the episodes.5
Focusing on content validity, the ASTA HRQOL scale was then tested in a sample of 240 patients with different forms of arrhythmia. The initial test included patients with AF and atrial flutter (AFL) planned for DC-conversion and patients treated with RFA because of atrioventricular nodal reentry tachycardia, Wolff-Parkinson-White syndrome, focal atrial tachycardia (FAT), atrial macro reentry, AFL, and AF, as well as patients with ventricular arrhythmias. Patients were encouraged to comment and suggest questions that could be added if they thought that something was missing. One item concerning spending time with relatives, friends, and acquaintances was divided into 2 items, 1 for friends and relatives and 1 for acquaintances, as a result of patient recommendations. The patients distinguished between acquaintances and relatives/friends, but not between relatives and friends. Two items concerning whether patients tried to hide arrhythmia occurrence from others or if they felt neglected while having arrhythmia were tested initially but were considered not relevant for measuring HRQOL by the research group.
The development process resulted in a 13-item HRQOL scale, with 7 items in a hypothesized physical domain and 6 items in a mental domain (Table 1). As a final step, the content validity of the 13-item scale was evaluated using the content validity index (CVI). A panel consisting of 5 experienced cardiologists and 2 cardiac nurses rated the relevance of each item and the adequacy of the total scale. A 4-point scale ranging from 1 to 4 (not relevant to highly relevant) was used. The CVI was computed for each item (I-CVI) and for the 13-item scale (S-CVI). The I-CVI was computed as the percentage of experts rating 3 or higher for each of the item, and the average method of S-CVI was used to assess the CVI of the 13-item scale. According to recommendations in the literature, a level of 0.78 or greater for I-CVI and 0.80 or greater for S-CVI was used as a criterion for demonstrating content validity.28,29
The items can be presented separately or with sum scores. The scores for the total scale range between 0 and 39, with higher scores reflecting more negative and burdensome effects on HRQOL because of the heart rhythm disturbance. The possible score range for the physical and mental domains range between 0 and 21 and between 0 and 18, respectively. The time to complete the 13-item scale was a few minutes.
The Validation Phase
Patient Population and Inclusion Criteria
The study inclusion period lasted from May 2009 until December 2009. Patients eligible for study participation included those referred for RFA treatment because of different forms of arrhythmia, as well as atrioventricular-nodal reentry tachycardia, Wolff-Parkinson-White syndrome, FAT, atrial macro reentry, AFL, AF, and ventricular arrhythmia patients (n = 215) at a university hospital. Another 55 patients seeking acute care because of AF at 2 county hospitals in Sweden were also included. Patients were included if they were referred for RFA treatment because of arrhythmia or were seeking emergency care because of AF and were 18 years or older, with sufficient knowledge of the Swedish language and physical and mental ability to fill in the study questionnaires. During the inclusion period, another 44 patients planned for RFA were not included. The reasons were unwillingness to participate (n = 17), insufficient knowledge of Swedish (n = 5), not capable to participate (n = 4), and missing/not approached (n = 18). There are no data available for missing patients in the emergency care group.
The RFA patients were consecutively asked about participation and the outpatients were asked at the time of the acute care visit. All RFA patients received study information by mail before their RFA. They then received oral and written information at the time of admission to hospital, before the planned intervention. Most participating patients filled out the questionnaires the day before treatment. The patients seeking acute care received oral and written information at the time of their visit and filled out the questionnaires during the acute care visit.
Health-Related Quality of Life Questionnaires
A combination of a generic and a disease-specific questionnaire was used for the validation process. Together with the ASTA HRQOL scale, the 36-Item Short-Form Health Survey (SF-36 version 1.0) was used to assess general health. The SF-36 consists of 8 health scales representing physical functioning, role limitations due to physical health problems, bodily pain, general health, vitality (energy/fatigue), social functioning, role limitations due to emotional problems, and mental health (psychological distress and psychological well-being) and 2 summary measures, physical and mental component summary (PCS/MCS). In the present study, only PCS and MCS were used for the psychometric testing. The scores for PCS and MCS can range from 0 (worst possible health) to 100 (best possible health). The scoring of the SF-36 data was done as described by Ware and colleagues.9,30–32 The SF-36 has been used in several arrhythmia studies assessing HRQOL and has been shown to have good validity and reliability in this patient population.1–3,6,33
Means, standard deviations, and frequencies were used to describe the characteristics of the sample and patient-reported general and disease-specific HRQOL. Comparisons were made using the χ2 test for nominal variables. The Student unpaired t test was used to compare mean scores between the patients included in the psychometric evaluations and patients without complete data. The ASTA 13-item HRQOL scale was psychometrically evaluated regarding data quality, construct validity, and internal consistency reliability.
Data quality was evaluated regarding score distribution and missing data pattern. Frequencies were used to evaluate the distribution of item responses, whereas the Kolmogorov-Smirnov test was used to test if the scale scores deviated from a normal distribution. Frequencies were used to describe the missing data pattern and to describe floor and ceiling effects. Floor and ceiling effects were defined if values of the observed variables fell at the minimum or maximum possible score, that is, if most of the scores were distributed at either end of the response scale.34
As a first step in the evaluation of construct validity, item-total correlations adjusted for overlaps were calculated for each item and the ASTA HRQOL total scale. An acceptable level was set to 0.30 or greater.34 In the next step, confirmatory factor analysis (CFA) was used to evaluate the hypothesized factor structure, including a physical and mental subscale. The first model (I) was specified as a 1-factor model, and the second model (II) was specified as a 2-factor model. In model II, the factors were allowed to correlate freely. In this first step, no residual variances were allowed to correlate. Both models (I and II) were thereafter respecified. Based on modification indices, residual variances were allowed to correlate. Other models, including cross-loadings between the 2 factors and the indicator variables, were also tested. As these models did not improve the fit between the model and data, they are not reported in this study. A robust weighted least square estimator using a diagonal weight matrix (WLSMV) was used in all CFAs as the assumption of multivariate normality was violated and as the indicator variables were categorical.35,36 To evaluate the goodness of fit between the models and data, we used fit indices suitable for WLSMV, including the χ2 goodness-of-fit, weighted root-mean-square residual (WRMR), the root-mean-square error of approximation (RMSEA), close fit for RMSEA (CFit), comparative fit index (CFI), and Tucker-Lewis index (TLI). A sufficient model fit was defined as nonsignificant χ2 goodness-of-fit and CFit, WRMR of less than 1.0, RMSEA of 0.05 or less, and CFI and TLI of 0.95 or greater. As traditional χ2difference test is not appropriate for WLSMV, we used the DIFFTEST command in Mplus to make it possible to evaluate differences between the revival models.36 All those goodness-of-fit indices were used to evaluate if the supposed measurement model fit the data.
In a final step, construct validity was evaluated regarding convergent and discriminant validity.37 For this purpose, the physical and mental subscales in ASTA were correlated with the PCS and MCS in the SF-36. To support convergent validity, the strongest correlation should be demonstrated between the ASTA physical scale and SF-36 PCS and between the ASTA mental scale and SF-36 MCS (homotrait-heteromethod correlations). To support discriminant validity, weaker correlations should be demonstrated between the ASTA physical scale and SF-36 MCS and between the ASTA mental scale and SF-36 PCS (heterotrait-heteromethod correlations). For this purpose, Spearman correlation coefficients were used.
Internal consistency reliability in the ASTA HRQOL items was evaluated using Cronbach’s α coefficients.38 An α coefficient of 0.70 or greater was considered sufficient.34 In addition, a lower bound confidence interval (95%) for Cronbach’s α was calculated.39 A P value of <.05 was considered significant. Statistical analyses were conducted using Stata 12.1 for Mac (Stata Corporation, College Station, Texas) and Mplus 7.0 for Mac (Muthén & Muthén, Los Angeles, California).
The validation study included 270 patients, of whom 147 were available for a complete psychometric evaluation, that is, had questionnaires without missing data. Two hundred fifteen of the patients were treated with RFA and 55 patients sought emergency care because of AF. Most patients were men. There were no differences in age between men and women in the study (P = .875). Sixty-three percent had an educational level of upper secondary school certificate or more, including a university degree (Table 2). In the group of patients with complete data, that is, who participated in the psychometric evaluation, an educational level of upper secondary school certificate and/or a college/university degree was more common (χ23 = 11.55, P = .009), as was treatment with class I antidysrhythmic medication (χ21 = 6.69, P = .014).
There were significant differences between the RFA patients and the group of AF patients seeking emergency care. The RFA patients were significantly younger compared with the emergency care group (Δmean, 7.4 ± 2.5; t145 = 2.98, P = .003). However, the groups did not differ regarding gender (χ21 = 0.01, P = .934), education (χ21 = 0.75, P = .861), ASTA total score (t145 = 1.24, P < .217), ASTA physical score (t145 = 0.82, P < .416), or ASTA mental score (t145 = 1.62, P < .107).
Of the 270 patients included, 19 patients did not complete any of the 13 items and another 14 patients had missing values in 1 to 7 items. Because of the difficulties in interpreting and scoring the answer “I don’t know,” this alternative, used by 90 patients, was regarded as a missing value in this validation study. This resulted in a total of 147 patients completing the forms, who were available for a complete psychometric evaluation.
The ASTA HRQOL 13-item scale had a mean score of 13.5 ± 8.1. The mean score for the physical and mental subscales were 8.0 ± 5.2 and 5.5 ± 3.6, respectively. The total scale and the physical subscale did not deviate from a normal distribution (P = .359 and .188 respectively). In contrast, the mental subscale demonstrated a positive skew distribution (P = .004). The distribution of the 13 items in the ASTA HRQOL scale showed that items 2, 3, 10, and 11 met the criteria for floor effect. None of the ASTA HRQOL items showed ceiling effects (Table 3). The SF-36 mean scores for the SF-36 PCS and MCS were 42.4 ± 10.6 and 43.5 ± 11.1, respectively.
The separate items and the total ASTA HRQOL scale showed good content validity. The I-CVI ranged from 0.86 to 1.0 and the S-CVI was computed to 0.99. All items except for item 3, “Do you spend less time with acquaintances (people you do not know that well) than you would like to due to your arrhythmia?” was computed to I-CVI 1.0.
All 13 items in the ASTA HRQOL scale reached the expected level of item-total correlations (r ≥ 0.30), ranging from r = 0.48 for item 11 to r = 0.78 for item 12 (Table 3).
Confirmatory Factor Analysis
The factor loadings for the 1-factor model ranged between 0.56 and 0.94, all significant at a level of P < .001 (Table 4). Despite satisfactory high factor loadings, the model fit was partly unsatisfactory for the 1-factor model. Whereas CFI and TLI exceeded the critical level of 0.95, χ2 goodness-of-fit, WRMR, and RMSEA indicated unsatisfactory fit between model and data. A respecified model with correlated residual variances was therefore evaluated. This increased the model fit significantly according to the χ2difference test (χ24 = 73.92, P < .001). After this refinement, only the χ2 goodness-of-fit did not reach the criteria for acceptable fit between model and data. The RMSEA was close to 0.05, and the confidence interval showed that RMSEA did not deviate significantly from this critical level (Table 5).
In the 2-factor model, the factor loadings ranged between 0.68 and 0.94 in the physical subscale and between 0.61 and 0.86 in the mental subscale. The factor correlation between the physical and mental subscales was strong (0.84). Both the factor loadings and factor correlation were significant at a level of P < .001 (Table 4). The initial 2-factor model also showed unsatisfactory model fit regarding χ2 goodness-of-fit, WRMR, and RMSEA. After residual variances were allowed to correlate in the respecified 2-factor model, the model fit increased significantly according to the χ2difference test (χ25 = 67.05, P < .001), and all fit indices demonstrated satisfactory agreement, except for χ2 goodness-of-fit. In this model as well, RMSEA was close to 0.05 and did not statistically deviate from this critical level (Table 5). A comparison between the 1- and 2-factor models, without correlated residual variances, showed that the latter had significantly better model fit according to the χ2difference test (χ21 =27.07, P < .001).
Convergent and Discriminant Validity
Convergent validity was confirmed with the correlation analyses between the ASTA HRQOL subscales and SF-36 PCS and MCS dimensions. The strongest correlations (homotrait-heteromethod) were found between the ASTA physical subscale and the SF-36 PCS (rs = −0.61) and between and the ASTA mental subscale and the SF-36 MCS (rs = −0.62). Discriminant validity was supported with lower correlations (heterotrait-heteromethods) between the ASTA physical subscale and the SF-36 MCS (rs = −0.44) and between the ASTA mental subscale and the SF-36 PCS (rs = −0.32). In contrast, the correlation (heterotrait-homomethod) between the ASTA physical and mental scale (rs = 0.68) did not support discriminant validity.
Internal Consistency Reliability
The internal consistency of the ASTA HRQOL 13-item scale was satisfactory (α = .91), as were the physical and the mental subscales (α = .89 and .79, respectively). Internal consistency was also sufficient according to a 95% lower bound confidence interval, total scale α = .89, physical subscale α = .87, and mental subscale α = .75. Deleting an item from the analysis would not substantially improve the overall Cronbach’s α for any scale.
The present study aimed to develop and validate a disease-specific scale evaluating HRQOL in patients with different forms of arrhythmia treated according to national and international guidelines.40–42 In terms of data quality, construct validity, and internal consistency, this first evaluation showed that the ASTA HRQOL scale is a promising instrument for assessing HRQOL in the target population.
The disease-specific questionnaire available at the time of the development of ASTA was the arrhythmia-specific Symptom Checklist, Frequency, and Severity Scale. The Symptom Checklist, Frequency, and Severity Scale is a checklist evaluating symptoms in arrhythmia patients, mostly used in patients with AF, but it has also been used in patients with other tachycardias.3,23,24 Questionnaires regarding HRQOL in arrhythmia patients are often developed solely for patients with AF or patients with other forms of supraventricular tachycardia. Some questionnaires mix symptoms and HRQOL in the same scale or evaluate the daily life situation very briefly.3,20–25 We found it important to develop an HRQOL scale that focuses on the consequences that arrhythmias impose on patients’ daily lives.
To cover the sufficient content of the influence of arrhythmias on HRQOL, items were chosen from patient interviews, suggestions by experienced healthcare professionals, and from a literature review. In addition, we found it most important that the ASTA HRQOL scale be inspired by patients themselves. During the development and testing phase, patients were repeatedly asked to provide feedback on the coverage of arrhythmia affected domains and the readability of the questions. Therefore, the strength of the study lies in the repeated evaluations in the research group and in the expert panel including arrhythmia patients who were encouraged to give their opinions on the chosen symptoms and HRQOL concerns.
We used a qualitative approach with interviews combined with a pretest in a large share of patients and established the content validity of the final scale by using the CVI method.29 This is a common method used to secure content validity.28 The index was sufficient for all the ingoing items as well as for the total scale. However, the CVI can be criticized for not adjusting for chance agreement.43
In some of the items, it can be difficult to define if they belong to the physical or mental dimension. In the ASTA HRQOL scale, sleep disturbances and sexual life interferences are examples of such items. When making the final decision, we were guided by items’ placing in previous scales, such as the Minnesota Living With Heart Failure and SF-36. There are some similarities with the subscales in ASTA HRQOL and the disease-specific Minnesota Living With Heart Failure subscales.44 In both, sleep disturbances, daily activities, and spending time with others belong to the physical scale, whereas feeling low spirited or sad, worry, and ability to concentrate belong to the mental scale. According to the modification index, our CFAs showed no signs of substantial improvement by allowing these items to cross-load on both the physical and mental factors. To be certain, we allowed sleep disturbance and sexual life interference to cross-load to both factors. As this CFA did not improve the goodness of fit between model and data, the findings support that these items are appropriate indicator variables for the factors to which they were assigned.
Overall, data quality was good. There was a minor problem with skewed distribution of data for both items and scales. Although 3 of 4 items demonstrating a floor effect were related to the physical subscale, neither the total scale nor the physical subscale deviated from a normal distribution. Even if the mental subscale showed a skewed distribution, the scoring range was acceptable as all response categories were used. This indicates that the ASTA HRQOL scale has the ability to discriminate between different groups.34 The generally low frequency of missing data indicates that the ASTA HRQOL scale is relevant and not too burdensome or difficult to complete. There was no clear pattern found for missing data, which indicates that the ASTA HRQOL scale does not produce systematic missing data. One-third of the patients had chosen to answer “I don’t know” in at least 1 of the 13 items, most frequently in the item concerning the arrhythmia’s interference with sexual life. Possible reasons for the great number of patients who responded “I don’t know” are that these patients did not have distinct or bothersome arrhythmia symptoms or that some could not judge the arrhythmia to be the reason for their daily life impairment. Difficulties in handling the “I don’t know” responses resulted in a decision to exclude this alternative in the ASTA HRQOL scale.
This first evaluation of the ASTA HRQOL scale shows emerging evidence of validity and reliability. The item-total correlations for the total scale were high. This indicates that the items measure the same concept and that none of them fulfilled the criteria for removal. The findings were partly confirmed by the CFAs. The initial CFA analyses showed that the 1- and 2-factor model had similar properties, with strong factor loadings and satisfactory goodness-of-fit values according to the CFI and TLI. For both the 1- and 2-factor models, a good fit between model and data was reached only after residual variances were allowed to correlate. According to the χ2difference test and the other fit indices, the 2-factor model seems to be preferred. However, as model fit is only 1 aspect of model evaluation, no strong conclusion regarding the factor structure can be drawn.35,45 From this perspective, both models showed similar model fit and demonstrated strong and significant factor loadings. We also found a strong correlation between the physical and mental factor in the CFA analysis. The findings indicate a possible higher-order factor structure, that is, that both the total and subscale scores can be used in a meaningful way.
Similar findings were also shown in the evaluation of convergent and discriminant validity. There was a problem regarding discriminant validity because the strongest correlation was found between the physical and mental subscales of the ASTA HRQOL scale. This finding corresponds with the problem with a strong factor correlation. It is difficult to assess whether some domains are mostly physical or mental concerns (as mentioned earlier in the CFA discussion). This can be demonstrated in the present study by the arrhythmia’s interference with the patient’s sexual life or by sleeping disturbances. The high correlation between the subscales may imply that they measure the same concept. Difficulties with strong correlations between the physical and mental dimensions of HRQOL have also been demonstrated in other disease-specific scales, for instance, the Minnesota Living With Heart Failure Questionnaire.44 Considering the findings from the CFA and the evaluation of convergent and discriminant validity, our suggestion is to report both the total and subscale scores until there is more evidence regarding the factor structure. In particular, a higher-order factor model should be considered.
All ASTA HRQOL scales demonstrated satisfactory internal consistency reliability. As the Cronbach’s α is sample dependent,37 we calculated a lower bound confidence interval for the α coefficient. As both the total scale and the subscales demonstrated lower bound confidence intervals for Cronbach’s α values above the critical level of .70, the ASTA HRQOL scale can be expected to have satisfactory internal consistency reliability also in the population. It should be mentioned that Cronbach’s α is an appropriate measure of internal consistency only for unidimensional scales.34 As the factor structure needs to be evaluated further, the Cronbach’s α should be interpreted with this in mind.
The sample size in the present study was somewhat limited, even though all the arrhythmia diagnoses we aimed to include were represented. Reasons for the reduced number of patients were both the missing data in the ASTA HRQOL scale and the fact that a large share of the patients chose the response alternative “I don’t know.” A common solution for treating missing data is to impute data with estimated values, if the missing values are not too high.46 In this development and validation phase of the ASTA HRQOL scale, we have chosen to treat both missing values and the alternative “I don’t know” as missing data, with a reduced sample size as a consequence. Although this procedure reduced the sample size to 147 patients, we estimated this to be a sufficiently large sample as most of our analyses were used for description rather than inference.34
A related limitation is the variance in timing and location of data collection between emergency care and the RFA patients. This limitation is mostly related to the psychometric evaluation. Therefore, we compared the 2 groups with each other. As only the mean age differed between the groups, this sampling variance is probably of minor importance. Another limitation was that some patient groups were represented by only a few patients, that is, patients with FAT, atrial macro reentry, and ventricular arrhythmias. It is noteworthy that these patient groups were represented in the interviews during the development phase and included in the initial testing of the HRQOL scale. In addition, as the sample included patients of different genders, ages, and arrhythmias, the sample corresponded well to the target population of the ASTA HRQOL scale.
As the sample was somewhat limited, and as the scale was developed for different types of arrhythmias, no groups were excluded. In further studies, the factor structure needs to be evaluated regarding invariance across groups of different arrhythmia diagnoses. The ASTA HRQOL scale also needs to be further evaluated regarding known-groups validity, stability, and responsiveness.
Assessment of HRQOL is of importance for evaluating arrhythmia patients’ subjective experiences of their daily life situation. Generic instruments are often not sensitive enough to catch the effects of a certain disease, and therefore, there is a need for a suitable validated disease-specific questionnaire. The ASTA HRQOL scale was developed together with the experts themselves, that is, patients with different forms of arrhythmias. In terms of validity and reliability, the ASTA HRQOL scale showed to be a promising tool. The scale can be a valuable contribution to the assessment of HRQOL in both clinical practice and in research. Until there is more evidence regarding validity and reliability, in particular with regard to the factor structure, using both the total and subscale scores is recommended.
What’s New and Important
- A—ASTA is a newly validated arrhythmia-specific HRQOL questionnaire
- S—Suitable for patients with different forms of arrhythmias
- T—To catch the patient’s perspective. Subjective assessment of HRQOL is important in the evaluation of arrhythmia patients’ daily life situation.
- A—ASTA HRQOL questionnaire can be a valuable contribution to clinical practice and for research use for the assessment of HRQOL in patients with arrhythmia
Special thanks to all the participating patients, who made development of ASTA possible. We thank Associate Professor Håkan Walfridsson for his valuable contribution to the development of the ASTA questionnaire. We also thank the nurses and physicians working at the University Hospital in Linköping and the County Hospitals in Jönköping and Kalmar who were involved in data collection and those involved in the expert panel. We thank the foundation support represented by Carldavid Jönsson Foundation, Medical Research Council of Southeast Sweden (FORSS), County Council of Östergötland, Stina and Birger Johanssons Foundation, and Siv Olssons Foundation.
1. Thrall G, Lane D, Carroll D, Lip GY. Quality of life in patients with atrial fibrillation: a systematic review. Am J Med. 2006; 119: 448.e1–448.e19.
2. Walfridsson U, Stromberg A, Janzon M, Walfridsson H. Wolff-Parkinson-White syndrome and atrioventricular nodal re-entry tachycardia in a Swedish population: consequences on health-related quality of life
. Pacing Clin Electrophysiol. 2009; 32: 1299–1306.
3. Bubien RS, Knotts-Dolson SM, Plumb VJ, Kay GN. Effect of radiofrequency catheter ablation on health-related quality of life
and activities of daily living in patients with recurrent arrhythmias
. Circulation. 1996; 94: 1585–1591.
4. Brembilla-Perrot B, Beurrier D, Houriez P, Claudon O, Wertheimer J. Incidence and mechanism of presyncope and/or syncope associated with paroxysmal junctional tachycardia. Am J Cardiol. 2001; 88: 134–138.
5. Walfridsson U, Arestedt K, Stromberg A. Development and validation of a new Arrhythmia-Specific questionnaire in Tachycardia and Arrhythmia (ASTA) with focus on symptom burden. Health Qual Life Outcomes. 2012; 10: 44.
6. Walfridsson U, Walfridsson H. The impact of supraventricular tachycardias on driving ability in patients referred for radiofrequency catheter ablation. Pacing Clin Electrophysiol. 2005; 28: 191–195.
7. Dhala A, Bremner S, Blanck Z, et al. Impairment of driving abilities in patients with supraventricular tachycardias. Am J Cardiol. 1995; 75: 516–518.
8. Wood KA, Wiener CL, Kayser-Jones J. Supraventricular tachycardia and the struggle to be believed. Eur J Cardiovasc Nurs. 2007; 6: 293–302.
9. Sullivan M, Karlsson J. Hälsoenkät. Svensk manual och tolkningsguide. (Swedish Manual and Interpretation Guide). Gothenburg, Sweden: Health Care Research Unit, Medical Faculty, Gothenburg University and Sahlgrenska Hospital; 1994.
10. Valter L, Noorlind Brage H. Östgötens hälsa. vol. Rapport. Linköping, Sweden: Centre for Public Health Science, Linköping University Hospital; 2006: 8.
11. Calvert MJ, Freemantle N. Use of health-related quality of life
in prescribing research, part 1: why evaluate health-related quality of life
? J Clin Pharm Ther. 2003; 28: 513–521.
12. Darbar D, Roden DM. Symptomatic burden as an endpoint to evaluate interventions in patients with atrial fibrillation. Heart Rhythm. 2005; 2: 544–549.
13. Pinter A, Dorian P. New approaches to atrial fibrillation management: treat the patient, not the ECG. Cardiovasc Ther. 2010; 28: 302–310.
14. Rumsfeld JS, Alexander KP, Goff DC Jr, et al. Cardiovascular health: the importance of measuring patient-reported health status: a scientific statement from the American Heart Association. Circulation. 2013; 127: 2233–2249.
15. Spertus JA. Evolving applications for patient-centered health status measures. Circulation. 2008; 118: 2103–2110.
16. Anderson KL, Burckhardt CS. Conceptualization and measurement of quality of life as an outcome variable for health care intervention and research. J Adv Nurs. 1999; 29: 298–306.
17. Calvert MJ, Freemantle N. Use of health-related quality of life
in prescribing research, part 2: methodological considerations for the assessment of health-related quality of life
in clinical trials. J Clin Pharm Ther. 2004; 29: 85–94.
18. Bowling A. Measuring Disease: A Review of Disease Specific Quality of Life Measurement Scales. Buckingham, United Kingdom: Open University Press; 2001.
19. Peyrol M, Sbragia P, Bonello L, Levy S, Paganelli F. Characteristics of isolated atrial flutter versus atrial flutter combined with atrial fibrillation. Arch Cardiovasc Dis. 2011; 104: 530–535.
20. Badia X, Arribas F, Ormaetxe JM, Peinado R, de Los Terreros MS. Development of a questionnaire to measure health-related quality of life
(HRQoL) in patients with atrial fibrillation (AF-QoL). Health Qual Life Outcomes. 2007; 5: 37.
21. Braganca EO, Filho BL, Maria VH, Levy D, de Paola AA. Validating a new quality of life questionnaire for atrial fibrillation patients. Int J Cardiol. 2010;143(3):391–398.
22. Harden M, Nystrom B, Kulich K, Carlsson J, Bengtson A, Edvardsson N. Validity and reliability of a new, short symptom rating scale in patients with persistent atrial fibrillation. Health Qual Life Outcomes. 2009; 7: 65.
23. Jenkins LS, Brodsky M, Schron E, et al. Quality of life in atrial fibrillation: the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study. Am Heart J. 2005; 149: 112–120.
24. Kay GN, Ellenbogen KA, Giudici M, et al. The Ablate and Pace Trial: a prospective study of catheter ablation of the AV conduction system and permanent pacemaker implantation for treatment of atrial fibrillation. APT Investigators. J Interv Card Electrophysiol. 1998; 2: 121–135.
25. Wood KA, Stewart AL, Drew BJ, Scheinman MM, Frolicher ES. Development and initial psychometric evaluation of the Patient Perspective of Arrhythmia Questionnaire. Res Nurs Health. 2009; 32: 504–516.
27. Rickham PP. Human experimentation. Code of ethics of the World Medical Association. Declaration of Helsinki. Br Med J. 1964; 2: 177.
28. Polit DF, Beck CT. The content validity index: are you sure you know what’s being reported? Critique and recommendations. Res Nurs Health. 2006; 29: 489–497.
29. Polit DF, Beck CT, Owen SV. Is the CVI an acceptable indicator of content validity? Appraisal and recommendations. Res Nurs Health. 2007; 30: 459–467.
30. Ware JE Jr, Sherbourne CD. The MOS 36-item Short-Form health survey (SF-36), I: conceptual framework and item selection. Med Care. 1992; 30: 473–483.
31. Ware JE Jr, Snow KK, Kosinski M, et al. SF-36 Health Survey Manual and Interpretation Guide. Boston, MA: The Health Institute, New England Medical Center; 1993.
32. Ware JE Jr, Kosinski M. SF-36 Physical & Mental Health Summary Scales: A Manual for Users of Version 1. 2nd ed. Lincoln, RI: QualityMetric Incorporated; 2001.
33. Goldberg AS, Bathina MN, Mickelsen S, Nawman R, West G, Kusumoto FM. Long-term outcomes on quality-of-life and health care costs in patients with supraventricular tachycardia (radiofrequency catheter ablation versus medical therapy). Am J Cardiol. 2002; 89: 1120–1123.
34. Nunnally JC, Bernstein IH. Psychometric Theory. 3rd ed. New York, NY: McGraw-Hill; 1994.
35. Flora DB, Curran PJ. An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychol Methods. 2004; 9: 466–491.
36. Muthén LK, Muthén BO. Mplus User’s Guide. 7th ed. Los Angeles, CA: Muthén & Muthén; 2012.
37. Streiner DL, Norman GR. Health Measurement Scales; A Practical Guide to Their Development and Use. 2nd ed. Oxford, UK: Oxford University Press; 2002.
38. Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika. 1951; 16: 297–334.
39. Kristoff W. The statistical theory of stepped-up reliability coefficients when a test has been divided into several equivalent parts. Psychometrika. 1963; 28: 221–238.
40. Camm AJ, Lip GY, De Caterina R, et al. Guidelines ESC: 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J. 2012; 33: 2719–2747.
41. Kirchhof P, Auricchio A, Bax J, et al. Outcome parameters for trials in atrial fibrillation: executive summary. Eur Heart J. 2007; 28: 2803–2817.
42. Socialstyrelsen: The Swedish National Board of Health and Welfare’s; arrhythmias
scientific basics for National Guidelines for Cardiac Care 2008. Stockholm. 2008: 46–48.
43. Wynd CA, Schmidt B, Schaefer MA. Two quantitative approaches for estimating content validity. West J Nurs Res. 2003; 25: 508–518.
44. Rector TS, Cohn JN. Assessment of patient outcome with the Minnesota Living With Heart Failure questionnaire: reliability and validity during a randomized, double-blind, placebo-controlled trial of pimobendan. Pimobendan Multicenter Research Group. Am Heart J. 1992; 124: 1017–1025.
45. Brown TA. Confirmatory Factor Analyses for Applied Research. New York, NY: Guilford Press; 2006.
46. Fayers PM, Machin D. Quality of Life. The Assessment, Analysis and Interpretation of Patient-Reported Outcomes. Chichester, West Sussex, England: John Wiley & Sons, Ltd; 2008.
Keywords:Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved
arrhythmias; disease-specific questionnaire; health-related quality of life; validation studies