Until April 2021, there were 98 patients taking their medications at the HCFMRP-USP Ambulatory Pharmacy for the treatment of leprosy. We interviewed 42 patients initially, but only 38 remained until the end of the study. The others decided to leave the study due personal issues. The mean age found was 51.43 ± 16.15 years, with a minimum age of 22 years and a maximum of 83 years.
The sociodemographic characterization demonstrated an adult population, declared male (64.3%), brown (52.4%), with no or low education (up to elementary school) (71.5%), married (40.5%), retired (45.2%), who do not smoke (85.7%), do not use alcohol (76.2%) and do not perform physical activity (66.7%). Of the patients who reported having other types of diseases, other than leprosy, the prevalence was arterial hypertension (31%).
When asked about the use of Hansen’s disease medications, more than 75% of patients believe that medications improve their symptoms and that this occurs within 3 months after starting use (58.1%) (Table 1 – in the ending of the article).
Regarding the SALSA scale, most patients had mild or no limitation (69.1%). Regarding the risk awareness score, 31% of patients did not score, and the second highest percentage (19%) was associated with a score of 1 (very low), followed by 14.3% for 5 points.
Besides, this study wanted to assess the social participation of these patients before and after follow-up, to find out if there was a significant difference between the data, in order to find out if they felt more welcomed and less stigmatized with their diseases (Table 2).
To assess whether there was a statistically significant difference in adherence to treatment after the intervention, McNemar's exact test was performed for paired proportions in the responses obtained in MGT. It was found that there is a significant difference in the results, with the percentage of adherents after the intervention being significantly higher than the percentage of adherents before (78.4% > 35.1%), with p < 0.001*.
Unlike the MGT, when patients are asked for a single question, in HST, the percentage of adherents before treatment is already high in comparison with post intervention adherents.
The same McNemar test was performed for HST, in order to assess whether the intervention performed in the research interfered with the patients' adherence results. Regarding before and after, the result was very close to the level of significance (p = 0.07), with a tendency towards a significant difference between before and after (81.1% < 97.3%).
However, it is very clear that the HST and MGT tests assessed differently the adherence of the study patients. To assess the association between the HST and MGT, the McNemar test was performed and the Kappa coefficient of agreement was calculated. There is a significant difference (p < 0.001*), the percentage of adherence in MGT before the intervention is significantly lower than that of HST (37.5% < 82.5%) (Table 3).
After calculating the Kappa (0.23) with a confidence interval of 95% (0.05–0.40), a p = 0.03* was obtained, that is, the coefficient is significantly different from 0, but your interpretation would only be reasonable.
Assessing the association of the tests after the intervention, there is a significant difference (p = 0.04*), the percentage of adherents in MGT is significantly lower than that of HST (76.3% < 94.7%). After calculating Kappa (0.11) with a 95% confidence interval, it was found that the coefficient is not significantly different from 0 (p = 0.42), and its interpretation would be weak.
For the variable gender in association with adherence in MGT, we have X2 = 5.18 and p = 0.02*. In other words, there is a significant difference, the percentage of adherents is significantly higher in females (60.0% > 24.0%), with the Odds ratio = 4.75 and 95%CI (1.19–18.92), it can be inferred that females are 4.7 times more likely to be adherent, before any intervention, which could not be observed in HST (Table 4).
Other variables did not influence adherence before the intervention. However, the variable mental illness influenced the result of adherence to the HST: with p < 0.001*, there was a significant difference, with the percentage of adherents being significantly higher in those who did not have mental illness (96.8% > 33.3%). That is, those who have mental illness are 58 times more likely to be non-adherent (Odds ratio = 58.8), CI 95% (5.09–100.1).
In this study, there was no significant difference in the association of the color variable with adherence to MGT and HST before and after the intervention. Therefore, color does not influence treatment adherence. The same was true for the variable marital status and occupation, which did not affect patient adherence before the intervention.
We initially tested the normality (Shapiro-Wilks test) of the age variable for the MGT and HST. We found that only age did not reject the hypothesis of normality, so we applied Student's t test and treatment time for this variable. We applied the non-parametric Mann-Whitney test. There is no significant difference between ages and the results obtained in terms of adherence or not to treatment using MGT and HST.
In HST, it is verified that using the Exact X2 test for linear trend, there is a significant linear association (p = 0.03*). That is, as education increases, adherence significantly increases, even before the intervention.
To understand the impact of the intervention performed with the patients, a comparison of MGT variable between before x after (Time variable) of the intervention was performed. For this analysis we used the method known as Generalizer Estimating Equations (GEE), which is an extension of generalized linear models that allows the analysis of repeated measures or other correlated observations, in longitudinal studies.
We used the Binary Logistics function for the adjustment. The covariance matrix used was the “unstructured” type. In this way, we were able to assess the predictive factors for adherence, that is, the evolution of adherence. We can see, with the Wald Chi Square value equal to 12.752, that there is a significant difference (p < 0.001*) between before the intervention and after the intervention in the results of treatment adherence by the MGT: 77% > 37%.
When we evaluated the gender variable, this significant interaction (p ≤ 0.05) indicates that the evolution, the increase, in MGT rate evolved, increased in a significantly different way between the sexes. That is, sex influences the way the adherence rate evolves (graph 1).
The percentage of schooling was not different before (31% vs. 50% p = 1.00) or after (83% vs 60% p = 1.00). However, the educational level of the group of patients who have completed elementary school was 31–83% (p < 0.001*) while the other group who had at least high school, even if incomplete, was 50–60% (p = 1.00). We infer that practically all the increase in the rate from 33–77% is due to the low education group, including a higher final rate in males (83% vs 60%). However, it must be emphasized that, before the intervention, the adherence rate of those who had at least high school education been higher than that of the other group.
When we grouped the variables “no limitation” with “mild limitation” in SALSA scale against the other variables and compared them with the adherence data obtained by HST, we observed an X2 = 5.18 and p = 0.001*. Therefore, there is a significant difference and the percentage of adherents is significantly higher in the group without limitation with mild limitation (96.4% > 50.0%). The odds ratio obtained was 27.02, with I.C.95% (2.72–250). Thus, it can be noted that patients classified as "No limitation" or "mild limitation" are 27 times more likely to adhere to the treatment (Table 4). The same could not be observed in MGT.
As the other test of the research hypothesis was to demonstrate a possible variation in the level obtained by the participation scale, that is, the increase in the perception of oneself, socially, it was evaluated whether there was a relationship between the results obtained in the scale and adherence to the treatment. It was possible to verify that there is a significant difference (p < 0.001*), with the percentage of adherents obtained by the HST being significantly higher in patients classified as “Unrestricted” compared to “Extreme restriction” (100% > 20%).
When comparing the data before the intervention and after the intervention for the participation scale, using the McNemar exact test, we noticed that the value of p = 0.07 is very close to the adopted significance level (p ≤ 0.05) indicating a strong tendency to have a significant increase in the percentage of pooled patients who were classified as “no restriction and light restriction” (from 65.8–81.6%) after the intervention. This indicates a strong tendency to claim that intervention with pharmaceutical care improves patients' social perspective (Table 2).
When we grouped the variables “without restriction” with “slight restriction” against the other variables and compared them with the adherence data obtained by the Haynes test, we observed an X2 = 5.18 and p = 0.02*. Therefore, there is a significant difference and the percentage of adherents is significantly higher in the group that was classified as “no restriction” and “light restriction” (92.9% > 58.3%). The odds ratio obtained was 9.26, with I.C.95% (1.47–48.52). Thus, it can be noted that patients who obtained a classification of "No restriction" or who have "light restriction" are 9.2 times more likely to adhere to the treatment. The same could not be observed in MGT (Table 4).
All data generated or analyzed during this study are included in this article, therefore there is no supplementary material.