| Bolman et al [16] |
Higher habit strength is positively correlated with higher adherence. Habit mediates the relationship between self-efficacy and medication adherence. Social norms moderate the relationship between habit and adherence; in weak habit, a supportive norm of taking medicine was positively related to adherence, and in strong habit, supportive norm correlated with less adherence. Perceiving few negative consequences of taking medicine was associated with better adherence. Control variables of risk perception and asthma severity were positively correlated with adherence. Female gender was positively correlated with adherence. Control variable of internal locus of control negatively correlated with adherence. From the central concepts, perceiving more pros, social support, higher self-efficacy, and stronger habit was associated with more adherence. From the central concepts, habit strength and attitude pros had the strongest correlation with medication adherence. Social norm and modeling were not significantly associated with adherence. Social influence subscales were highly intercorrelated, as well as habit with risk perception, pros, social support, and self-efficacy. After hierarchical multiple regression, habit strength proved to be significantly related to adherence. Of the control variables, only severity remained significant; of the ASEa concepts, only the cons remained significant.
|
| Burns et al [14] |
Interaction between habit strength and depressive symptoms was observed. If habit strength was weak or average, depressive symptoms were negatively associated with adherence. However, if habit was strong, no association was observed. Same significant interaction pattern was observed for diabetes distress and habit strength as well as major depressive syndrome and habit strength. Habit strength moderates the association between poor mental health symptoms and medication adherence. After adjusting for covariates, results remained significant.
|
| Durand et al [19] |
Medication-taking habit strength was the strongest predictor of adherence (compared with pill burden, illness coherence, and treatment-related beliefs). Habit strength explained 19% incremental variance in adherence beyond treatment-related beliefs. Habit strength was more strongly associated with unintentional nonadherence than intentional. Associations among adherence measures were weak to moderate, indicating that multiple measures are necessary to accurately assess adherence. Neither treatment-related beliefs nor CSMb coherence predicted adherence, even for patients with weak habit strength. Pill burden was not associated with habit strength or adherence. There was no significant interaction between treatment-related beliefs, habit strength, and adherence.
|
| Guenette et al [8] |
Strong habit was significantly associated with adherence. Perceived behavioral control, older age, no perceived side effects, a longer period since T2Dc diagnosis, and a lower number of NAIDd daily doses were significantly associated with adherence. Sex, level of education, and income are not associated with adherence. Intention, insulin use, number and type of NIAD drugs prescribed, perceived cost of antidiabetes medications, and use of glucometer or weekly pill organizer were not associated with NIAD adherence. Depressed mood, anxiety, and mental health were not associated with adherence. Behavioral control was found to be significant, so the 26 underlying beliefs were analyzed, and 12 beliefs were found to be significant with adherence.
|
| Hooa et al [10] |
One unit increase in habit index was associated with a 0.3% increase in the subsequent week’s adherence after controlling for current adherence. Those with variable adherence displayed higher mean cross-correlation coefficients (0.45) compared with those with consistent adherence (0.20-0.40).
|
| Hoob et al [9] |
Higher adherers reported stronger habit compared with lower adherers. A 1-unit increase in habit strength was associated with a 31% increase in odds of being in the next higher adherence category. In a multiple ordinal regression model with both habit and concerns scores, only habit was associated with adherence. Higher adherers had lower prior year intravenous use, tended to have higher %FEVe at baseline, and reported lower concerns.
|
| Murphy et al [20] |
Stronger habit strength was associated with better adherence. Those who never miss an OCPf reported significantly higher habit strength than those who miss 2 or more per month. There was no difference between those who never miss an OCP and those who miss 1 OCP per month. Having a fixed time of day to take the OCP was associated with better habit strength and adherence. There is, however, no association between habit strength and taking OCP at different times of the day. Having a fixed place to store the OCP was associated with habit strength but not adherence.
|
| Phillips et al [7] |
Habit strength was the strongest predictor of medication adherence (compared with beliefs and experiences plus efficacy)—explains 6%-27% incremental variance in adherence to that explained by treatment-related beliefs. Habit strength was more strongly related to unintentional medication nonadherence than intentional nonadherence. Patients’ CS-SRMg coherence was more strongly associated with intentional nonadherence than unintentional adherence. Patients’ treatment-related beliefs were not more strongly associated with intentional nonadherence than unintentional nonadherence. Habit strength had the strongest association with MEMSh dose timing out of all the adherence measures. The interaction between treatment-related beliefs and habit was not significant for any of the adherence measures. Patients’ beliefs and experiences did not predict overall adherence, even for weaker adherence. Patient experience, however, did predict intentional nonadherence.
|
| Phillips et al [21] |
Habit strength consistently predicted incremental variance in measured outcomes, both self-reported and measured. Correlations, between habit strength and % of the doses taken on time vs between habit strength and % of the days when medications were taken, were not significantly different. Habit strength does not predict unintentional nonadherence better than intentional. Habit strength is not relatively more important for predicting medication adherence than physical activity.
|
| Thorneloe et al [22] |
Patients in the biological cohort were more likely to be male, have a younger age of onset of psoriasis, longer duration of disease, more likely to have a diagnosis of inflammatory arthritis, have lower quality of life scores at the start of therapy, have longer duration of systemic therapy, have stronger beliefs in the chronicity of their illness, stronger beliefs that systemic therapy is necessary, weaker concerns about therapy and medicine, greater coherence, and less symptoms of depression. Patients using self-administered systemic therapy had strong habit strength. Being on a conventional systemic therapy, having strong medication concerns, longer treatment duration, and younger age were factors associated with overall nonadherence. Being on a conventional therapy and strong medication concerns were also significant for intentional nonadherence. Being on a conventional systemic therapy, stronger perceptions of psoriasis being a chronic condition, younger age, and longer treatment duration were factors associated with unintentional nonadherence. Group 1 membership (strongest medication concerns) was associated with intentional nonadherence, and weaker medication-taking routine or habit strength was associated with unintentional nonadherence.
|
| Voils et al [23] |
Dual conceptualization (self-report with psychometric principles) of medication nonadherence has stronger validity and reliability than other forms that confound these 2 variables. Extent of adherence was highly correlated with self-efficacy, where lower adherence levels were associated with lower self-efficacy. In all, 3 items assessing the extent of nonadherence produced reliable scores. Correlations between the extent and harm subscales with habit strength were above 0.3. Correlations and comparison measures showed convergent and divergent validity. Predictive validity was evidenced by correlations between extent and BPi. Means of the reasons items were well below the scale midpoint, and several distributions were positively skewed and kurtosis. The Morisky scale did not measure a single underlying construct in this sample. The Morisky score was not correlated with BP.
|
