ABSTRACT
Title
Adherence to anti-osteoporosis therapy in an Italian population
Authors
M. Casula1, E. Tragni1, A. Filippi2, F. Decè3, L. Defendi3, L. Gandolfi3, L. Perego4, R. Piccinelli3, A.L. Catapano1
1Epidemiology and Preventive Pharmacology Centre (SEFAP), Department of Pharmacological Sciences, University of Milan, Italy
2Italian Society of General Medicine (SIMG) , Florence, Italy
3Pharmacoeconomics units, LHU of Bergamo, Italy
4Primary care department, LHU of Bergamo, Italy
1Epidemiology and Preventive Pharmacology Centre (SEFAP), Department of Pharmacological Sciences, University of Milan, Italy
2Italian Society of General Medicine (SIMG) , Florence, Italy
3Pharmacoeconomics units, LHU of Bergamo, Italy
4Primary care department, LHU of Bergamo, Italy
Abstract
BackgroundAlthough effective drug therapies are available for treatment of osteoporosis, adherence to medications is currently low, with poor health outcomes.
ObjectiveTo analyze adherence (in terms of persistence and compliance) of anti-osteoporosis drug use in the Local Health Unit of Bergamo (Northern Italy) and to assess influence of patient-related and drug-related factors.
Design and setting An observational, retrospective study was conducted between 2006 and 2008: data on prescriptions of anti-osteoporosis drugs (AOD) were retrieved from the Pharmaceutical Service of Bergamo, an administrative database that covers all the out-of-hospital prescriptions in the Bergamo district. Key measurements were compliance (medication possession ratio, MPR) and persistence (all gaps between the end of a prescription and the next refill ≤30 days) at one year. We estimated the impact of some anagraphic and clinical variables on MPR using multivariate logistic regression analysis; the dependent variable to be explained was reaching an MPR of at least 80%. As persistence can change over time, we use a Cox proportional hazards model.
ResultsIn the population observed, 4304 in 2007 were new users of anti-osteoporosis drugs. 87.9% were women, with a mean ± SD of age of 70.83 ± 11.25 years, slightly more than men (70.26 ± 12.34 years; non significant difference). The most commonly prescribed drug was alendronic acid (with or without colecalciferol, 55.2% W and 64.1% M), followed by risedronic acid (20.5% W and 20.9 % M) and strontium ranelate (16.3% W and 9.4% M). Anti-osteoporosis therapy was primarily weekly (97.6% of all prescriptions of only alendronic acid and the 97.4% of all prescription of risedronic acid). Generics were used by 16.0% and 17.7% of women and men. In the total cohort, switching to another drug was more frequent than changing administration frequency (10.1% vs 6.1%). Overall, mean ± SD of MPR was 58.55% ± 36.57% for women and 45.49% ± 37.15% for men, with 63.6% of subjects with MPR under the cut off for optimal compliance. More than 50% of subjects had already stopped their treatment at 6 months. At one year, persistent women and men were 22.4% and 16.5%, respectively. Among this group, mean compliance ± SD was about 101% ± 13%. In the female and male subsets of non-persistent subjects, 80.1% and 86.0% had compliance <80%, respectively; these proportions included a large number of subjects with only one prescription (about one third of non-persistent women and half of non-persistent men). On the other hand, 48.9% and 31.0% of non-persistent women and men showed at least one prescription after the interruption, respectively. In regression analyses, age and frequency of administration were stronger associated with inadequate adherence. Age was significantly associated to poor compliance (Odds Ratio [IC 95%] <50 ys: 2.437 [1.600-3.712] for women and 3.505 [1.145-10.729] for men; ≥75 ys: 1.207 [1.051-1.387] for women), while weekly (OR [IC 95%]: 0.233 [0.098-0.555] for women) and monthly (OR [IC 95%]: 0.158 [0.064-0.389] for women) regimens vs daily administration resulted in a reduced risk. Similar results were obtained for non-persistence (Hazard Ratio [IC 95%] age <50 ys: 1.645 [1.336-1.981] for women and 1.596 [1.082-2.352] for men; age ≥75 ys: 1.120 [1.040-1.207] for women; weekly regimen: 0.531 [0.385-0.733] for women and 0.533 [0.325-0.873] for men; monthly regimen: 0.547 [0.386-0.775] for women and 0.516 [0.267-0.997] for men).
ConclusionIn our study, compliance to AOD was optimal only in half of treated patients and persistence declined significantly over time, as four out five stopped therapy at 1 year. The likelihood of poor adherence was significantly lower among those who used a less frequent administration regimen. According to our data, poor adherence and its consequences on fracture risk remains a main concern and other interventions to improve adherence need to be identified and implemented.
ObjectiveTo analyze adherence (in terms of persistence and compliance) of anti-osteoporosis drug use in the Local Health Unit of Bergamo (Northern Italy) and to assess influence of patient-related and drug-related factors.
Design and setting An observational, retrospective study was conducted between 2006 and 2008: data on prescriptions of anti-osteoporosis drugs (AOD) were retrieved from the Pharmaceutical Service of Bergamo, an administrative database that covers all the out-of-hospital prescriptions in the Bergamo district. Key measurements were compliance (medication possession ratio, MPR) and persistence (all gaps between the end of a prescription and the next refill ≤30 days) at one year. We estimated the impact of some anagraphic and clinical variables on MPR using multivariate logistic regression analysis; the dependent variable to be explained was reaching an MPR of at least 80%. As persistence can change over time, we use a Cox proportional hazards model.
ResultsIn the population observed, 4304 in 2007 were new users of anti-osteoporosis drugs. 87.9% were women, with a mean ± SD of age of 70.83 ± 11.25 years, slightly more than men (70.26 ± 12.34 years; non significant difference). The most commonly prescribed drug was alendronic acid (with or without colecalciferol, 55.2% W and 64.1% M), followed by risedronic acid (20.5% W and 20.9 % M) and strontium ranelate (16.3% W and 9.4% M). Anti-osteoporosis therapy was primarily weekly (97.6% of all prescriptions of only alendronic acid and the 97.4% of all prescription of risedronic acid). Generics were used by 16.0% and 17.7% of women and men. In the total cohort, switching to another drug was more frequent than changing administration frequency (10.1% vs 6.1%). Overall, mean ± SD of MPR was 58.55% ± 36.57% for women and 45.49% ± 37.15% for men, with 63.6% of subjects with MPR under the cut off for optimal compliance. More than 50% of subjects had already stopped their treatment at 6 months. At one year, persistent women and men were 22.4% and 16.5%, respectively. Among this group, mean compliance ± SD was about 101% ± 13%. In the female and male subsets of non-persistent subjects, 80.1% and 86.0% had compliance <80%, respectively; these proportions included a large number of subjects with only one prescription (about one third of non-persistent women and half of non-persistent men). On the other hand, 48.9% and 31.0% of non-persistent women and men showed at least one prescription after the interruption, respectively. In regression analyses, age and frequency of administration were stronger associated with inadequate adherence. Age was significantly associated to poor compliance (Odds Ratio [IC 95%] <50 ys: 2.437 [1.600-3.712] for women and 3.505 [1.145-10.729] for men; ≥75 ys: 1.207 [1.051-1.387] for women), while weekly (OR [IC 95%]: 0.233 [0.098-0.555] for women) and monthly (OR [IC 95%]: 0.158 [0.064-0.389] for women) regimens vs daily administration resulted in a reduced risk. Similar results were obtained for non-persistence (Hazard Ratio [IC 95%] age <50 ys: 1.645 [1.336-1.981] for women and 1.596 [1.082-2.352] for men; age ≥75 ys: 1.120 [1.040-1.207] for women; weekly regimen: 0.531 [0.385-0.733] for women and 0.533 [0.325-0.873] for men; monthly regimen: 0.547 [0.386-0.775] for women and 0.516 [0.267-0.997] for men).
ConclusionIn our study, compliance to AOD was optimal only in half of treated patients and persistence declined significantly over time, as four out five stopped therapy at 1 year. The likelihood of poor adherence was significantly lower among those who used a less frequent administration regimen. According to our data, poor adherence and its consequences on fracture risk remains a main concern and other interventions to improve adherence need to be identified and implemented.