A cross-sectional design was employed for analysis and consecutive sampling was carried out on 180 HF patients, consisting of both male (n: 108) and female (n: 72) at the outpatients unit. The respondents who agreed to participate were given informed consent. The respondents completed the data, including demographic characteristics, illness-related factors, the Minnesota Living with Heart Failure Questionnaire (MLHFQ), and they performed Five Times Sit-to-Stand Test (FTSST). This study has been reviewed and approved by the Hospital Institutional Review Board (HIRB) (PP04.03/XIX.3/3580/2021).

MLHFQ was used to measure HRQOL.13 Scores ranged from 0 to 105, with higher scores indicating lower HRQOL. MLHFQ used a Likert scale of 0–5, consisting of 21 items, and represented two dimensions, namely physical and emotional. Disease severity was assessed based on the New York Heart Association Classification (NYHA Class), which was a classification based on the level of physical activity associated with HF symptoms, the scale of the NYHA Class was I to IV, with higher classes indicating increased symptoms and decreased physical activity.14 However, only NYHA Classifications I–III were used in this study. HS was measured using the Self-Rated Health Subscale questionnaire. The Self-Rated Health Subscale consisted of four items, with scores ranging from 4 to 14, and higher scores indicated lower HS.15 The comorbidity status was measured using Charlson Comorbidity Index. The high score indicates the seriousness of the comorbidities.16

PP used the FTSST measuring instrument and the FTSST score was based on the amount of time (seconds) patients moved from sitting to standing and back to sitting five times. The lower the seconds to complete the test, the better the result.17 Furthermore, demographic characteristics consisted of age, gender, occupation, marital status, and place of residence. Disease-related factors included hemoglobin, body mass index, duration of HF diagnosis, comorbidity, and HF medications.

Before data analysis, all data were examined to check for errors, missing data values, distribution, normality of data on the main variables, homogeneity of variance, homoscedasticity, and linearity for each statistical procedure. Relationship between demographic characteristics, disease-related factors, PP, and HRQOL was explored using an independent t-test, one-way ANOVA, and Pearson correlation. Meanwhile, hierarchical multiple regressions explained the predictor variables on HRQOL. Dummy variables were performed on NYHA Classification and PP of the respondents in this study. Variables included in the regression model were those that were significant with HRQOL, based on a cut point value of P<.20 with HRQOL.

The demographic characteristics of the 180 patients with HF involved in this study showed that the mean age was 59.98 (11.86) years old, most of the respondents were male 60%, married 92.3%, and employed 60.6%. Approximately 45% of respondents had senior high school education and 57% had NYHA Class II. Based on illness-related factors, the mean values were as follows, body mass index 25.10 (4.56)kg/m2, length of diagnostic HF was (32.29±35.38) months, comorbidity status 2.87 (1.44), HF medication 1.47 (0.66), and the median Hbg was 13.35g/dL, with most respondents using diuretics (57.8%). Moreover, the mean HF of respondents was 10.15 (1.92), FTSST 9.56 (6.94), and the mean score of MLHFQ was 43.14 (20.74) (Table 1).

This study explained that MLHFQ of patients with HF had a significant correlation with HF medication (r=.16, P<.05), HS (r=.24, P<.01), and FTSST (r=.40, P<.01). The results showed that MLHFQ had a significant association with NYHA Classification (F=8.358, P<.001). After post hoc analysis, NYHA Class I had a lower MLHFQ score (37.98±18.79) than Class III (55.64±24.75) and Class II (41.32±18.68), while NYHA Class II had a lower MLHFQ score than Class III. However, MLHFQ did not correlate with age, gender, job, marital status, living arrangement, Hbg, body mass index, length of diagnostic HF, and comorbidity, as presented in Table 2 and Table 3.

Table 4 shows that four variables that significantly correlated with MLHFQ, namely HF medication, HS, NYHA Classification, and FTSST as well as hemoglobin and comorbidity based on cut point values were included in the multiple linear regression model. In the first model, the variables included were hemoglobin, comorbidity, HF medication, HS, and dummy variables for NYHA Classification, which accounted for 12.6% of MLHFQ variance. The significant variables with MLHFQ in this model were HF medication (β=3.57, P<.05) and NYHA Class III (β=11.04, P<.05). In the second model, PP variables using FTSST were included, and the results revealed that R2 increased by 31.1% (P<.001) from .095 to .283. The significant variables were HS (β=1.76, P<.05), NYHA Class III (β=−10.84, P<.05), and FTSST (β=1.75, P<.001).

The limitations of this study included the distribution of respondents, with less than 50% in NYHA Class I and III. Meanwhile, in Class I, some patients did not feel the effect of HF symptoms and refrained from seeking hospital treatment. Most of the respondents in NYHA Class III were more inclined to seek treatment or visit the emergency room due to the severity of their symptoms rather than receiving outpatients care due to symptoms experienced. However, with the use of good sampling methods and data analysis, the results of this study can be trusted for accuracy. Secondly, HRQOL and PP were measured cross-sectionally, which limited the ability to determine causal relationships. The data obtained can provide evidence that PP using FTSST plays an important role in HRQOL of patients with HF. Therefore, a longitudinal study is a recommendation to determine the causal relationship between HRQOL and PP for future studies.