nueva página del texto (beta)
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Introduction
Arterial hypertension is the main preventable cause of cardiovascular disease, which significantly decreases the life expectancy and quality of those who suffer from it1,2. In Peru, a 2018 study found an age-standardized prevalence of arterial hypertension of 20.6% with 40-50% of people with hypertension aware of their condition. Approximately 40% of patients aware of their condition reported receiving treatment, with treatment being effective about 40% of the time. This translates into only about one in 20 Peruvians with hypertension achieving good blood pressure control3,4. Therefore, arterial hypertension is a common condition that is not treated as well as it could in Peru. Getting more people treated requires a multidisciplinary approach that includes improving screening, access to medicine, treatment protocols, and encouraging patient medication adherence.
Given these low blood pressure control rates, investigating factors influencing adherence to hypertension therapies becomes crucial for improving patient outcomes. Poor adherence is a complex problem that has multiple causes that are related to sociodemographic factors, the type of therapy, patient attitudes, and the healthcare system5,6. One important factor in adherence is family functionality, defined as the ability of the family unit to maintain cohesive relationships, fulfill roles, manage problems, adapt to new patterns, and have effective communication among members7. When a family member has a chronic disease, a readaptation process occurs, which involves alterations in family functioning and relationships, which can impact the patient's health behaviors8-11.
To better understand how adherence is associated with family functioning and other sociodemographic factors such as age, sex, time since diagnosis, and type of treatment, we designed a cross-sectional study of patients seen in a public hospital in Peru. Given previous results3,4, we hypothesize that poor family functioning and other sociodemographic factors are associated with poor adherence.
Materials and methods
Design and setting
This is a cross-sectional study of a non-probabilistic convenience sample of 180 patients seen in the outpatient clinic of the cardiometabolic clinical unit of Hospital I Florencia de Mora EsSalud in Florencia de Mora, La Libertad, Peru. Patient recruitment, interviews, and blood pressure measurement occurred between December 2022 and February 2023.
Participants
Participants were included if they were at least 18 years old, had been diagnosed with arterial hypertension (systolic pressure > 140 mmHg, diastolic pressure > 90 mmHg)12, and had a medical prescription for antihypertensive drugs for at least 6 months. Patients with a diagnosis of cognitive impairment, neurodegenerative disease, stroke sequelae, were pregnant and/or had secondary arterial hypertension were excluded from the study. The authors evaluated medical records to determine eligibility and then contacted patients to ask them if they wanted to participate. All participants provided their informed written consent and were informed of their right to withdraw from the study at any time.
Variables
Social-clinical data collected during the interview included age, self-reported gender, educational level (none, primary, secondary, post-secondary), participants' living arrangement, time since the initial diagnosis of arterial hypertension, number of antihypertensive medications prescribed, and presence of other comorbidities.
Blood pressure measured at the time of the interview was classified as stage I hypertension (systolic: 140-159 mmHg and/or diastolic: 90-99 mmHg) or stage II hypertension (systolic: ≥ 160 mmHg and/or diastolic: ≥ 100 mmHg)12.
Adherence to treatment was measured using the 8-item Morisky medication adherence scale (MMAS-8). The questionnaire contained 7 yes/no questions and one question assessed on a Likert-like scale. Each question is assigned a point, and a score of < 8 indicates non-adherence. The test has an internal consistency with Cronbach's α of 0.8313,14.
Family functionality was evaluated with the family APGAR test, which was intended to measure the participant's perception of family functionality. This scale consists of five questions to evaluate family adaptability, participation, growth, affection, and resources. Each question was scored from 0 to 4 points, for a maximum of 20. A score of more than 17 points indicates the absence of family dysfunction. Scores between 13 and 16, 10 and 12, and < 9 indicated slight, moderate, or severe family dysfunction, respectively. In previous studies, the test showed a correlation coefficient of 0.86 and a Cronbach α of 0.8415,16.
Data collection
During the study period, a medical doctor supervised by the authors interviewed the patients to determine sociodemographic characteristics. The doctor also measured the blood pressure of the participants using standard procedures and equipment. The participants then completed two questionnaires to measure adherence to treatment and family functionality. The results of these questions were recorded and used to determine the appropriate scores for statistical analysis. Bias was addressed by using the same method to complete the validated questionnaires and to supervise the data collection.
Study size
Population and sample size were determined using the sample size for frequency in a population test on Open Epi version 3 (https://www.openepi.com/), assuming a 95% confidence level and a difference in proportions based on a Chinese study of hypertension and family functioning, which found that 31.1% of patients with family support adhered to their treatment17.
This resulted in a sample size estimate of 166 participants. The authors decided to recruit 180 participants to account for incomplete data collection or ineligibility for approximately 10% of the participants.
Statistical methods
A descriptive analysis was carried out showing the absolute and percentage frequency of the variables and their socioclinical characteristics. Chi-square statistical tests of independence with continuity correction were used to determine the association of study variables and covariates associated with adherence to treatment, considering that p ≤ 0.05 are significant. These covariates were analyzed to address potential sources of bias and confusion in a multivariate analysis. Furthermore, the prevalence odds ratio product (ORP) and its 95% confidence intervals (95% CI) were calculated. Analysis was performed using RStudio version 4.2.2 (https://cran.rstudio.com/).
Ethical considerations
This study was approved by the Bioethics Committee of the Universidad Privada Antenor Orrego with approval number N°001-2023-UPAO.
The data generated during and/or analyzed during the present study are not publicly available nor are they available on request due to ethical considerations of the participants, as the data are linked to their personal information.
Results
All 180 participants completed the data collection process, so there were no missing data. Of the participants, 57.8% were 70 years or older, 59.4% were women, 40.6% were men, and 40.6% had a post-secondary education level. Most of the participants live with relatives (75.0%), were diagnosed at least 10 years before the interview (54.4%), have stage I arterial hypertension (84.4%), were prescribed an antihypertensive drug (70.6%), and had comorbidities (68.9%) (Table 1).
Table 1 Socioclinical characteristics of the participants. The right column presents the number of observations and percent in parentheses
| Sociodemographic characteristic | Frequency | |
|---|---|---|
| Number | % | |
| Age, years | ||
| 40-49 | 5 | 2.8 |
| 50-59 | 20 | 11.1 |
| 60-69 | 51 | 28.3 |
| ≥ 70 | 104 | 57.8 |
| Sex | ||
| Male | 73 | 40.6 |
| Female | 107 | 59.4 |
| Education level | ||
| None | 5 | 2.8 |
| Primary | 48 | 26.7 |
| Secondary | 54 | 30.0 |
| University | 73 | 40.6 |
| Family dysfunction | ||
| None | 117 | 65.0 |
| Slight | 33 | 18.3 |
| Moderate | 19 | 10.6 |
| Severe | 11 | 6.1 |
| Treatment adherence | ||
| Low | 41 | 22.8 |
| Medium | 62 | 34.4 |
| High | 77 | 42.8 |
| Living arrangement | ||
| Alone without children | 8 | 4.4 |
| Alone with children nearby | 14 | 7.8 |
| With a spouse of similar age | 23 | 12.8 |
| With relatives | 135 | 75.0 |
| Time since diagnosis, year | ||
| < 5 | 51 | 28.3 |
| 5-9 | 34 | 18.9 |
| 10-19 | 61 | 33.9 |
| ≥ 20 | 34 | 18.9 |
| Hypertension stage | ||
| Stage I | 152 | 84.4 |
| Stage II | 28 | 15.6 |
| Number of antihypertensives prescribed | ||
| 1 | 127 | 70.6 |
| 2 | 45 | 25.0 |
| ≥ 3 | 8 | 4.4 |
| Other comorbidities | ||
| No | 56 | 31.1 |
| Yes | 124 | 68.9 |
In Table 2, participants were divided into groups as to whether they adhered to treatment according to the MMAS-8 questionnaire. In total, 42.8% of the participants adhered to treatment. In the group of patients who adhered to treatment, 68.8% were over 70 years old, 83.1% did not have family dysfunction, and 66.2% were diagnosed at least 10 years before the interview. The non-adherent group tended to be younger, had a higher frequency of family dysfunction, and was diagnosed less time before the interview. According to the Chi-square analysis, statistically significant (p < 0.05) differences between groups were observed for age groups, family functioning, and time since first diagnosis. No statistically significant differences were found between the adherent and non-adherent groups for sex, educational level, living arrangement, whether the participant had stage I or stage II hypertension, the number of antihypertensives prescribed, or the presence of comorbidities.
Table 2 Comparison of socioclinical characteristics and adherence to treatment according to the independent Chi-square test
| Sociodemographic characteristic | Adherence to treatment | ||||
|---|---|---|---|---|---|
| Yes n = 77 (42.8%) | No n = 103 (57.2%) | p | |||
| n | % | n | % | ||
| Age (years) | |||||
| 40-49 | 1 | 1.3 | 4 | 3.9 | 0.005 |
| 50-59 | 2 | 2.6 | 18 | 17.5 | |
| 60-69 | 21 | 27.3 | 30 | 29.1 | |
| ≥ 70 | 53 | 68.8 | 51 | 49.5 | |
| Sex | |||||
| Female | 40 | 51.9 | 67 | 65.0 | 0.106 |
| Male | 37 | 48.1 | 36 | 35.0 | |
| Education level | |||||
| None | 2 | 2.6 | 3 | 2.9 | 0.351 |
| Primary | 17 | 22.1 | 31 | 30.1 | |
| Secondary | 21 | 27.3 | 33 | 32.0 | |
| University | 37 | 48.1 | 36 | 35.0 | |
| Family dysfunction | |||||
| None | 64 | 83.1 | 53 | 51.5 | 0.000 |
| Slight | 5 | 6.5 | 28 | 27.2 | |
| Moderate | 5 | 6.5 | 14 | 13.6 | |
| Severe | 3 | 3.9 | 8 | 7.8 | |
| Living arrangement | |||||
| Alone without children | 4 | 5.2 | 4 | 3.9 | 0.918 |
| Alone with children nearby | 6 | 7.8 | 8 | 7.8 | |
| With a spouse of similar age | 11 | 14.3 | 12 | 11.7 | |
| With relatives | 56 | 72.7 | 79 | 76.7 | |
| Time since diagnosis (years) | |||||
| < 5 | 11 | 14.3 | 40 | 38.8 | 0.003 |
| 5-9 | 15 | 19.5 | 19 | 18.4 | |
| 10-19 | 32 | 41.6 | 29 | 28.2 | |
| ≥ 20 | 19 | 24.7 | 15 | 14.6 | |
| Hypertension stage | |||||
| Stage I | 63 | 81.8 | 89 | 86.4 | 0.527 |
| Stage II | 14 | 18.2 | 14 | 13.6 | |
| Number of antihypertensives prescribed | |||||
| 1 | 49 | 63.6 | 78 | 75.7 | 0.082 |
| 2 | 22 | 28.6 | 23 | 22.3 | |
| ≥ 3 | 6 | 7.8 | 2 | 1.9 | |
| Other comorbidities | |||||
| No | 22 | 28.6 | 34 | 33.0 | 0.636 |
| Yes | 55 | 71.4 | 69 | 67.0 | |
Table 3 shows the multivariate analysis of the variables found to be significant in table 2. The degree of family functionality and whether the patient experienced any degree of family dysfunction are conceptually identical; therefore, the degree of functionality was eliminated to avoid redundancy and multicollinearity in the analysis. We observed that the same three variables remained statistically associated with adherence. The ORP and the 95% CI of 4,39 (2,08-9,76) for the correlation between adherence and family functioning was slightly greater than the ORP for the time since diagnosis of more than 20 years. The ORP for other significant associations was lower.
Table 3 Multivariate analysis of the variables associated with adherence among study participants, where ORP is the prevalence odds ratio and CI 95% is the 95% confidence interval
| Sociodemographic characteristic | p | ORP | CI 95% |
|---|---|---|---|
| Family dysfunction | |||
| None | 0.000 | 4.39 | 2.08-9.76 |
| Present | Reference | ||
| Age (years) | |||
| 40-49 | 0.989 | 0.00 | N/A |
| 50-59 | 0.036 | 0.18 | 0.03-0.75 |
| 60-69 | 0.796 | 0.91 | 0.43-1.92 |
| ≥ 70 | Reference | ||
| Time since diagnosis (years) | |||
| < 5 | 0.030 | 3.20 | Reference |
| 5-9 | 0.003 | 3.88 | 1.14-9.41 |
| 10-19 | 0.006 | 4.30 | 1.61-9.84 |
| ≥ 20 | 1.14-9.41 |
ORP: odds ratio product; CI: confidence intervals.
Discussion
Our study focused on determining whether there is an association between selected sociodemographic factors and adherence to prescribed medication(s) in patients with arterial hypertension seen in the cardiometabolic clinical unit of the Hospital I Florencia de Mora EsSalud in Florencia de Mora, La Libertad, Peru. Adherence to treatment is a key component in controlling blood pressure and its complications. In this study, most of the participants were 70 years or older (57.8%) with a predominance of women (59.4%) and with a diagnosis time of at least 10 years (52.8%). This differs from the study by Asgedom et al., where the mean age was 55 ± 12.7 years, 53.2% were men and 57.9% had duration of < 5 years18. This could be due to patient sampling: the setting of the present study provides care mainly to older adult patients, which may be associated with a longer diagnosis time. On the other hand, our results are similar to a Mexican study by Becerra Partida and Villegas Galindo, where patients with high adherence had been diagnosed for more than 10 years before the study19. Our results also showed lower proportions of any family dysfunction than in three published studies18-20.
Patient adherence in our study was within 12% of three other recent Latin American studies, demonstrating a persistently low degree of adherence to treatment21-23. This level of adherence has been observed throughout the world6. These other studies conclude that older patients, women, monotherapy prescription and diagnosis at least 10 years before the study are associated with greater adherence. Of these associations, we observed that age and time since diagnosis were statistically associated with adherence, while the other variables often came close but did not reach statistical significance. Since age correlates with the time since diagnosis, both have a likely similar explanation. Frequent follow-ups with nursing staff, doctors, and nutritionists are likely to reinforce the importance of adherence to treatment over time.
We also found that family function is associated with adherence, with greater adherence observed among participants without family dysfunction. From the literature, this appears to be a cross-cultural phenomenon. Studies in Nigeria8 and China17 showed that family function or social support is associated with adherence to treatment. The Nigerian cross-sectional study calculated an odds ratio and a 95% CI of 2.6 (1.6-4.1), which overlaps with the 95% CI for family functioning determined here. An additional review from India suggested family involvement as one way to improve adherence5. Given the similarity of the results in different cultures, it is likely that family function is a more important determining factor than culture for adherence.
Interestingly, the number of medications prescribed came close but did not correlate significantly with adherence, which is in contrast to other research on the subject6,24. The proportions of adherent to non-adherent patients were almost identical when one or two medications were prescribed, but when three or more medications were prescribed, the proportion of adherence decreases to approximately 1 in 4. It is likely that a small sample number among patients with 3 or more prescribed drugs has made it difficult to draw clear conclusions in this study.
Education level has also been shown to be positively correlated with adherence25; the shifting proportions in the education level row of table 2 tend to corroborate this, with more highly educated participants having higher proportions of adherence. However, this did not reach statistical significance, likely again due to a small sample size. Therefore, to better measure the correlation between the number of medications prescribed and the level of education and adherence, a larger and more balanced sample size is recommended.
Limitations
Limitations of this study included the use of a non-probability sampling method, which could limit generalization. The cross-sectional nature of the study does not show causality between the variables. An additional problem is that the time since diagnosis is correlated with age. This could have overestimated the ORP calculated here. Furthermore, patient-recorded data on adherence or sociodemographic factors could have generated recall bias. Some variables, such as cardiovascular risk and the presence of clinical depression, have been shown to affect adherence26 but were not included in this study. Therefore, it would be advisable to conduct future research that includes a larger number of hypertensive patients from multiple health centers and include additional variables known to be associated with adherence. In addition, measuring adherence with a blood test could decrease recall bias and performing a matched study could reveal confounding effects.
Conclusion
The objective of this study was to find sociodemographic factors associated with antihypertensive medicine adherence within a study population of hypertensive patients from Northern Peru. This study found that increased age and time of diagnosis, as well as good family functioning associated with increased adherence. The factor most strongly correlated with adherence was family functioning. This could be because patients experiencing family dysfunction may have a lack of emotional support, inadequate communication, and interpersonal conflicts in their family unit, making adherence to a treatment protocol less important. In contrast, functioning families could encourage a patient to continue treatment and provide the necessary resources for adherence.
From this study and others, it is becoming clear that there are multiple factors associated with adherence to treatment. Given this body of data, it is reasonable to suggest screening for family functioning among hypertensive patients, as well as focusing adherence-related interventions in more recently diagnosed patients to improve adherence.
