Adherence to Selective Serotonin and Serotonin-Norepinephrine Reuptake Inhibitor Prescriptions Affects Overall Medication Adherence in Older Persons: Evidence From the Italian Nationwide OsMed Health-DB Database

Adherence to Selective Serotonin and Serotonin-Norepinephrine Reuptake Inhibitor Prescriptions Affects Overall Medication Adherence in Older Persons:

Evidence From the Italian Nationwide OsMed Health-DB Database

Alessandra Marengoni, MD, PhDa,’ ¡,*; Graziano Onder, MD, PhDb,’ ¡; Luca Degli Esposti, EconDc,’ ¡; Pierluigi Russo, MD, PhDd,’ ¡; Diego Sangiorgi, MStatc; Stefano Buda, ElEngc; Massimo Fini, MDe,’ ¡; Niccolײ Marchionni, MDf,’ ¡; Stefano Bonassi, PhDg,’ ¡; Federica Mammarella, MDd,’ ¡; Walter Marrocco, MDd,’ ¡; Giuseppe Pozzi, MDh,’ ¡; Katie Palmer, PhDd; Alessandro Monaco, PhDd,’ ¡; Sergio Pecorelli, MD, PhDi; and Luca Pani, MDd; on behalf of the Geriatrics Steering Committee of the Italian Medicines Agency on behalf of the OsMed Health-DB Networkj

Poor adherence to treatment of chronic diseases is a worldwide problem of great magnitude. Adherence to long-term therapy for chronic illnesses in developed countries averages 50% and is even poorer in developing countries,1 and poor adherence has a negative impact on health outcomes, quality of life, and health care costs. Data from the United States estimated that more than $100 billion is spent each year on avoidable hospitalizations due to poor adherence.2 The origin of poor adherence is multifactorial, including social and economic factors, health systems organization, and behavior of professionals, as well as specific characteristics of diseases, individual patients, and therapies.3 Until now, interventions aimed at improving adherence have shown a limited efficacy.4

Though nonadherence is a general public health problem, it is particularly relevant in older adults, since they are more often prescribed several treatments simultaneously as a consequence of frequent chronic multimorbidity. The resulting polypharmacy is a well-recognized problem for many elderly patients, as it implies an increased risk of both adverse drug reactions and reduced adherence.5 Moreover, some barriers to adherence are more common in older patients and warrant particular attention in the clinical practice. For example, although forgetting one’s medication may occur at any age, in older patients memory difficulties may be exacerbated by other medications or by concomitant cognitive impairment or depression.

Depression is very frequent in older persons. A recent meta-analysis6 of individuals aged 50 years and older found a rate of 3.2% for current and 16.5% for lifetime major depression and an even higher (19.4%) prevalence of depressive symptoms. The estimated prevalence of major depression in persons aged 65 years and older is 9.8%.7 Symptoms may vary in severity, ranging from an isolated and mild episode through recurrent episodes of moderate severity to chronic and persistent severe illness. For patients with a definite diagnosis of depression, guidelines advocate that drug treatment should continue for at least 6 months following remission of symptoms. Furthermore, long-term preventive treatment is suggested for patients with 2 or more episodes of clinically significant depression within 5 years.8 The level of adherence to antidepressant medications is positively correlated with treatment outcomes in depressed patients, independent of type of drug used.9 Adherence to recommended antidepressants is reported to range from 40% to 70%,10 with approximately 25% of patients not informing their physician of drug withdrawal.11

In the Italian Medicines Agency, the National Observatory on the Use of Medicines (OsMed) is responsible for national monitoring of pharmaceutical consumption and expenditure. In this context, OsMed Health-DB is the largest Italian database that prospectively collects information on prescription drugs. This study aims to evaluate the prevalence of prescription of and adherence to selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) in persons aged 65 years of age or older and to analyze whether adherence to antidepressants affects overall adherence to medications.

METHODS

The Nationwide OsMed Health-DB Database was used. This database includes 35 Italian Local Health Units (LHU) and about 18 million subjects (30% of the Italian population), with a uniform distribution in the North (28.7%), Central (33.7%), and South (29.8%) regions of Italy and a historical series from 2009. The mean age in the sample is 43.5 years versus 43.3 years in the overall Italian population, and the percentage of men is 48.5%, in accordance with the national data.

The Italian Medicines Agency (AIFA—Agenzia Italiana del Farmaco) is the national competent authority responsible for the regulation of medicine use. Inside AIFA, the OsMed monitors national use of medicines in Italy and, since 2012, has started to prospectively analyze data collected per patient with the technical support of CliCon. CliCon was delegated by AIFA for data warehousing and data analysis activities of information voluntarily provided by LHUs and stored in their own local administrative databases. In order to comply with privacy legislation (Legislative Decree n. 196 of 2003, and following amendments), patient ID code was anonymized by LHUs. Using this database, the OsMed report on drug use in Italy (www.agenziafarmaco.gov.it/it/content/rapporti-osmed-luso-dei-farmaci-italia) is published yearly by the AIFA with the following objectives: (a) to describe drug consumption at a national level, (b) to examine changes in drug use over time, (c) to benchmark drug consumption across different Italian regions, and (d) to evaluate appropriate use of drugs and adherence to treatment.

For each LHU, the following current administrative databases have been captured: demographic database (including deaths), outpatient pharmaceutical prescriptions database, inpatient pharmaceutical prescriptions database, “per conto” (distribution on behalf) pharmaceutical prescription database, and hospital discharge database. Data included are gender, birth and death date, Anatomical Therapeutic Chemical (ATC) code, prescription date, number of prescribed packages, price per package (both the community and the direct distribution price), admission and discharge date, type of admission (eg, ordinary, day hospital), discharge data (eg, discharged, moved), principal diagnosis (International Classification of Diseases, 9th Revision [ICD-9]), secondary diagnosis (ICD-9), main procedure (ICD-9), secondary procedures (ICD-9), and reimbursement per hospital stay (hospital discharge records).

• Poor adherence to long-term therapy for chronic diseases is the single most important modifiable factor that compromises treatment outcome, but not all potential factors affecting adherence were considered in previous studies.
• In older persons, depression is frequent and commonly coexists with other chronic diseases. In this part of the population, adherence to antidepressant drugs is associated with higher overall adherence to medications for other chronic illnesses.

Italian Pharmaceutical Reimbursement System

The Italian pharmaceutical reimbursement system covers the treatment of all relevant diseases, providing pharmaceutical coverage to the whole population. The general conditions of the reimbursement system are established by the Italian Medicines Agency at the national level. Reimbursed drugs (known as Class A) include essential drugs and drugs for serious, acute, and chronic diseases (eg, antihypertensive drugs, antibiotics, hypoglycemic agents, antibiotics, antidepressants, antiplatelet agents, anticoagulants, antiparkinson drugs). Nonreimbursed drugs (known as Class C) include drugs for less relevant disease or minor ailments, drugs whose use is discouraged, and drugs not requiring a medical prescription (eg, antispasmodics, topical treatments). The present analysis is limited to reimbursed medications (Class A).

Therapeutic Categories

Information on each drug package, identified via ATC classification and ID package unique identifier codes, was tracked at the individual level. In the ATC system, antidepressants are coded N06A; in particular, SSRIs and SNRIs were searched with their specific coding (N06AB for SSRIs and N06AX16 and N06AX21 for SNRIs) for inclusion in the present analysis. The other therapeutic categories analyzed were antihypertensive drugs (ATC classification C02, C03, C07, C08, and C09), statins (C10AA and C10BA02), other psychiatric drugs (N03A and N05A), antidiabetics (A10), antiplatelets (B01AC), anticoagulants (B01A), drugs for chronic obstructive pulmonary disease (COPD) (R03), and antiosteoporotics (M05BA, M05BB, H05AA, H05BA, M05BX03, and G03XC01). Only medications prescribed and dispensed by a pharmacy are considered in the present analysis.

The total number of drugs was calculated as the number of different prescribed ATC codes independent of the number of doses.

Adherence Assessment

To calculate adherence, older adults who were newly prescribed one of the drugs of interest during years 2011, 2012, and 2013 were identified, and the date of first drug dispensing was defined as the index date. Patients were considered as newly treated if they had no dispensing of any drug of the same class over the 6 months before the index date. Adherence to antidepressants and to other chronic treatments, including antihypertensive drugs, statins, psychiatric drugs, antidiabetics, antiaggregants, anticoagulants, drugs for COPD, and antiosteoporotics, was estimated by calculating the proportion of days covered by drugs dispensed during a period of 365 days. The number of days covered by each dispensing was calculated as the ratio between the total amount of active drug in each dispensed package and the recommended daily dose. In accordance with previously adopted thresholds, adherence was defined as a proportion of days covered more than 80%.12,13 Adherence was defined as adherence to the broad class of drugs (both SSRIs and SNRIs); this made it possible to calculate adherence even in the event of switching from one drug to another.

Data collection was in full compliance with the Italian law on personal data protection.

Statistical Analysis

Prevalence of SSRIs and SNRIs prescriptions was calculated across 4 age groups (65-74, 75-84, 85-94, and ≥ 95 years), gender, and 3 geographic areas (North, Central, and South Italy). Categorical and continuous variables were expressed as frequency and percentages, and as mean (SD), respectively. Multivariate logistic regression models were run to analyze the association of adherence to SSRIs and SNRIs with adherence to the other therapeutic categories in 2013. In these models, adherence to SSRIs and SNRIs, age, gender, and total number of prescribed drugs were independent variables.

RESULTS

During 2011, 2012, and 2013, a database was compiled of data collected on drug prescriptions. During 2013, the database included 17,846,127 individuals, who represented 30.0% of the Italian population (n = 59,394,207). Of these, 3,762,299 were aged 65 years or older (21.1% of the database sample).

Prevalence of SSRI and SNRI prescription ranged from 11.4 in 2011 to 12.1% of the whole sample in 2013 and was higher in the North than in the South of Italy across the 3 years (Table 1). The most frequently identified SSRIs and SNRIs in 2013 were paroxetine (25.6%), escitalopram (20.2%), sertraline (19.1%), citalopram (15.6%), venlafaxine (8.8%), duloxetine (7.5%), fluoxetine (2.6%), and fluvoxamine (0.6%).

Prevalence was more than double in women than in men. At analysis by age, the lowest prescription prevalence was found in the age group of ≥ 95 years (ranging from 8.6% to 8.9%) and the highest in the 2 age groups of 75-84 and 85-94 years (ranging from 13.1% to 15.6%) (Table 1). Table 2 presents characteristics of persons receiving a prescription of 1 or more SSRIs or SNRIs in 2013. Persons prescribed SSRIs or SNRIs were more likely to be women, younger, and prescribed a higher mean number of drugs (the mean [SD] number of drugs in persons prescribed with SSRIs and SNRIs was 8.5 [5.8], compared to 5.1 [4.9] in those not prescribed SSRIs and SNRIs) and a higher number of drugs of other therapeutic categories.

Adherence to SSRI and SNRI prescription ranged from 40% in 2011% to 41% in 2013. Adherence was higher in the North and Central regions of Italy compared to the South (Table 3). It was also slightly higher in women than in men and lower in the oldest age group (Table 3 shows proportion of adherence to SSRIs and SNRIs during the 3 years).

Persons adherent to SSRI and SNRI prescription were more adherent to the other therapeutic categories (Table 4). Logistic regression models showed that adherence to antidepressants was associated with increased odds of adherence to all the other therapeutic categories with the exception of anticoagulants. Models were adjusted for age, gender, and total number of prescribed drugs (Table 4). The highest association was found between adherence to SSRIs and SNRIs and adherence to other psychiatric drugs (OR = 1.9; 95% CI, 1.8-2.0).

DISCUSSION

Principal Findings

Findings of this study on a large administrative database of older Italian persons showed that the prevalence of SSRI and SNRI prescription varied from 11.4% in 2011 to 12.1% in 2013. Adherence to SSRIs and SNRIs, calculated according to 4 age groups and during 3 years of observations, ranged from 31.2% in persons older than 95 years in 2011 to 41.8% in 75- to 84-year-old persons in 2013. Older persons adherent to these classes of antidepressants were more likely to be adherent to the other therapeutic categories for chronic diseases they were prescribed.

Comparison With Other Studies

Over the past few years, in both clinical and research settings, increasing attention has been devoted to the study of depression in the expanding population of older individuals, in whom a variety of topics, including prevalence, possible treatments, and consequences have been examined extensively. One of the most relevant issues in this area—the problem of nonadherence to antidepressants—still awaits further clarification.14 Although drugs are commonly considered a cornerstone in the treatment of depression, evidence from epidemiologic studies confirms that about 1 in 3 patients does not complete treatment.15 Previous studies showed that 50% of patients cease antidepressant treatment as early as within 3 months, a proportion peaking to over 70% at 6 months.16 Trifirײ and colleagues17 reported that 27.5% of Italian general practitioner adult patients who were prescribed antidepressants discontinued therapy during the first year, mostly within 3 months. In a secondary analysis of results of the Clinical Outcomes in MEasurement-Based Treatment (COMET) trial, Chang and colleagues18 showed the importance of the frequency of monitoring of symptoms in depressive disorders. More frequent measurement of depression symptoms was associated with greater medication persistence; this finding may explain the lower adherence that we observed among those aged ≥ 95 years, since this patient population may be less likely and able to be monitored frequently.

A comparison with previous studies on the association between adherence to antidepressants and adherence to other therapeutic categories for chronic diseases is not possible due to the lack of previous research in this field.

Strengths and Limitations

Assessment of adherence to treatment from administrative data is a potential key limitation to this study. Until now, several indicators of adherence have been proposed, each presenting advantages and limitations; rate of prescriptions has the advantage to be objective and relatively easy to measure. However, it has the limitation that a prescription is not equivalent to ingestion of medication. Thus, our findings may slightly overestimate the real adherence to treatment, though identification of repeated dispensing from pharmacy can be taken as an indirect proof that the drug is effectively ingested.19 Finally, we cannot exclude that our results might be explained by residual confounding. People adhering to antidepressants might have higher education, cultural status, and health literacy and for this reason be more likely to adhere to other treatments. In addition, adherence reflects individual and behavioral factors such as cognitive and functional status, self-efficacy, and motivation for self-care as well as factors related to the nature of a condition or coexisting diseases, complexity and duration of the treatment regimen, possible adverse drug reactions, and cost of treatment. Studying factors associated with attrition during treatment for depression, Warden and colleagues20 showed that patients’ pretreatment concerns about continuing antidepressant treatment in the presence of side effects signals challenges to the completion of acute phase treatment. Social factors are also important, not only individual relationships, but also patient-provider relationship and different access to social support.21 Thus, the causal pathway of such an association cannot be explored in our study, and we might argue that higher adherence to antidepressants simply reflects an individual tendency toward good of adherence in general. However, when we checked the association of adherence to categories of non-antidepressant drugs with all the other drugs, we did not find consistent results. For example, adherence to antihypertensive drugs was inversely correlated with adherence to other psychiatric drugs (OR = 0.88); adherence to other psychiatric drugs and antidiabetics, with adherence to antiosteoporotics (OR = 0.78 and 0.74, respectively); and adherence to antiplatelets, with adherence to anticoagulants (OR = 0.50). Thus, despite these limitations, our results may suggest that adherence to drugs that reduce depressive symptoms translates into greater adherence to overall therapeutic prescriptions; in fact, depressed patients’ improvement in motivation with treatment may lead to enhanced self-care and motivation to improve physical health. This hypothesis, to be proven in prospective trials, further reinforces the indication to actively treat depression in older persons to improve their quality of life and general state of health.

A strength of the study is the very high number of prescriptions evaluated over 3 years across different Italian regions. Another strength is the novelty of the topic, as no study has evaluated whether adherence to a specific class of drugs is associated with adherence to other chronically prescribed medications in older persons.

Conclusions and Policy Implications

The burden of illness in both the general and the elderly population has shifted from acute to chronic diseases, making the problem of poor adherence to long-term treatments of major concern to all stakeholders in the health care system. The risk of poor adherence increases with the duration and complexity of therapeutic regimens, which are by definition inherent to chronic diseases. Adherence is the single most important modifiable factor that compromises treatment outcome.

In the older population, depression is frequent and also commonly coexists with other chronic medical conditions requiring several co-prescriptions, and this association can be interactive. On the one hand, chronic diseases increase the risk of depression, with the prevalence of depression being up to 5 times higher in persons with chronic medical conditions.22 This strong association can be explained by the presence of disability, pain, and polypharmacy in older persons affected by multiple diseases. On the other hand, depression can delay diagnosis of other diseases and negatively affect adherence to medications and a healthy lifestyle that are needed to prevent other clinical conditions.23 In this study, persons prescribed SSRIs and SNRIs were more likely to be prescribed a greater number of drugs, which indirectly confirms the association between depression and the burden of chronic diseases. Moreover, persons adherent to antidepressants were more likely to be adherent to the other drugs they were co-prescribed. A strong association has been shown between late-life depression and poor adherence to medical treatment for diabetes and cardiovascular disease.24 Moreover, older adults with chronic depression with 80% or greater antidepressant adherence had significantly lower risk of hospitalizations for cardiovascular diseases than those with lower adherence.25 The Prevention of Suicide in Primary Care Elderly Collaborative Trial (PROSPECT) project demonstrated that older adults with major depression provided with additional resources to intensively manage depression had a mortality risk lower than that observed in usual care and similar to that of older adults without depression.26 Possible explanations for all of the previously described findings are that persons adherent to antidepressants have a higher rate of remission of depressive symptoms, which makes them more interested in and attentive to their global health status. They might also have more trust in other medical treatments and have greater awareness of their diseases. Another hypothesis could be that persons who are adherent to antidepressants have a higher health literacy. The result of these attitudes and behaviors is the higher likelihood of being adherent to the overall therapeutic regimen.

Social and economic factors and factors related to the health care team/system, the characteristics of the disease, the type of therapies, and patient-related factors may all influence adherence. Indeed, patients’ personal attributes are likely to have the strongest influence on adherence. No single intervention strategy or package of strategies has been shown to be effective in improving adherence across all patients, conditions, and settings. Consequently, interventions that target adherence must be tailored to the particular illness-related demands experienced by the patient. To accomplish this, health systems and providers need to assess not only adherence, but also those factors that influence it. Treatment guidelines for chronic conditions, for instance, should recommend screening for depression, which can be an indicator of poor adherence.27 When benefits and harms in treating depression are balanced, adherence to treatment should be taken into account as a potential benefit.

Results from this study suggest the need for further evaluation of possible interventions aimed at increasing adherence to antidepressants, which may translate into improved overall adherence and patient outcomes, with potential substantial overall cost savings. Underdiagnosis and undertreatment of depression, which are very frequent in older adults,28 should be considered as factors contributing to poor drug adherence in older persons.

Submitted: October 30, 2015; accepted February 11, 2016.

Drug names: citalopram (Celexa and others), duloxetine (Cymbalta), escitalopram (Lexapro and others), fluoxetine (Prozac and others), fluvoxamine (Luvox and others), paroxetine (Paxil, Pexeva, and others), sertraline (Zoloft and others).

Disclosure of off-label usage: The authors have determined that, to the best of their knowledge, no investigational information about pharmaceutical agents that is outside US Food and Drug Administration-approved labeling has been presented in this article.

OsMed Health-DB Network: Local Health Unit Asti (Piemonte), Representatives: S. Martinetti, P. Mero, L. Raeli; Local Health Unit Pavia (Lombardia), Representatives: S. Migliazza, M. Dellagiovanna, C. Cerra; Local Health Unit Bergamo (Lombardia), Representatives: M. Gambera, R. Piccinelli, M. Zambetti, F. Atzeni; Local Health Unit Lecco (Lombardia), Representatives: V. Valsecchi, P. DeLuca, E. Scopinaro, D. Moltoni; Local Health Unit Varese (Lombardia), Representatives: E. Pini, O. Leoni, C. Oria, M. Papagni, G. Nosetti, E. Caldiroli; Local Health Unit Alto Adige (Trentino-Alto Adige), Representative: V. Moser; Local Health Unit Trento (Trentino-Alto Adige), Representatives: R. Roni, A. Polverino; Local Health Unit ULSS20 Verona (Veneto), Representatives: C. Bovo, L. Mezzalira, M. Andretta, L. Trentin; Local Health Unit n. 1 Triestina (Friuli Venezia Giulia), Representatives: S. Palcic, A. Pettinelli; Local Health Unit n. 2 Isontina (Friuli Venezia Giulia), Representatives: A. Arbo, A. Bertola, G. Capparoni; Local Health Unit n. 4 Medio Friuli (Friuli Venezia Giulia), Representatives: C. Cattaruzzi, L. Marcuzzo; Local Health Unit n. 6 Friuli Occidentale (Friuli Venezia Giulia), Representatives: F.V. Rosa, B. Basso; Local Health Unit USL N° 1 Imperiese (Liguria), Representatives: M. Saglietto, S. Delucis, M. Prioli, R. Filippi; Local Health Unit n. 3 Genovese (Liguria), Representatives: A. Coccini, M. Ghia, F. Sanfelici; Local Health Unit Piacenza (Emilia Romagna), Representative: S. Radici; Local Health Unit Ferrara e University Hospital S. Anna Ferrara (Emilia Romagna), Representatives: P. Scanavacca, A. Campi, S. Bianchi, A. Verzola; Local Health Unit Bologna (Emilia-Romagna), Representatives: M. Morini, M Borsari, A. Danielli; Local Health Unit USL 1 Massa Carrara (Toscana), Representatives: M. Dal Maso, B. Marsiglia; Local Health Unit USL 8 Arezzo (Toscana), Representative: B. Vujovic; Local Health Unit USL 9 Grosseto (Toscana), Representatives: M. Pisani, P. Bonini, F. Lena; Local Health Unit Marche (Marche), Representatives: P. Aletti, A. Marcobelli, S. Sagratella; Local Health Unit USL Umbria 2 (Umbria), Representatives: S.Fratini, F. Bartolini; Local Health Unit USL Roma A (Lazio), Representatives: G. Riccioni, A. Meneghini; Local Health Unit USL Roma D (Lazio), Representatives: R. Di Turi, V. Fano, A. Blasi, E. Pagnozzi; Local Health Unit USL Roma F (Lazio), Representatives: G. Quintavalle, P. D’ Avenia, M.C. De Matthaeis; Local Health Unit Frosinone (Lazio), Representatives: F. Ferrante, S. Crescenzi, L.Marziale, P.Venditti, C.Bianchi; Local Health Unit AUSL 4 Teramo (Abruzzo), Representatives: I. Senesi, R. Baci, I. De Carlo; Regional Health Unit Molise (Molise), Health General Direction—Pharmaceutical Programming and Assistance—Representatives: A. Lavalle, G. Trofa; Local Health Unit Caserta (Campania), Representatives: G. Marcello, C. Pagliaro, C. Troncone, G. Farina, M. G. Tari; Local Health Unit Potenza (Basilicata), Representatives: G. Motola, F. De Luca, M. L. Saltarelli, C. Granieri; Provincial Health Unit Cosenza (Calabria), Representatives: M. Vulnera, L. Palumbo, F. La Viola, L. Florio, A. E. De Francesco (on behalf of Calabria Region); Provincial Health Unit Reggio Calabria (Calabria), Representatives: D. Costantino, A. E. De Francesco (on behalf of Calabria Region); Provincial Health Unit 3 Catania (Sicilia), Representatives: F. Rapisarda, P.L. Lazzaro; Provincial Health Unit Palermo (Sicilia), Representatives: M. Pastorello, M. Parelli, M. Visconti, I. Uomo; Local Health Unit Cagliari (Sardegna), Representatives: P. Sanna, F. Lombardo.

Financial disclosure: The authors have no personal affiliations or financial relationships with any commercial interest to disclose relative to the article.

Funding/support: None.

Additional information: The OsMed Health-DB database is stored at CliCon Health Economics and Outcomes Research, Via Salara, 36, 48121, Ravenna, Italy.

REFERENCES

1. Fischer MA, Stedman MR, Lii J, et al. Primary medication non-adherence: analysis of 195,930 electronic prescriptions. J Gen Intern Med. 2010;25(4):284-290. PubMed doi:10.1007/s11606-010-1253-9

2. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;353(5):487-497. PubMed doi:10.1056/NEJMra050100

3. WHO World Health Organization (WHO). Adherence to Long-term Therapies: Evidence for Action. WHO Web site. http://www.who.int/chp/knowledge/publications/adherence_report/en/ Accessed February, 2015.

4. Peterson AM, Takiya L, Finley R. Meta-analysis of trials of interventions to improve medication adherence. Am J Health Syst Pharm. 2003;60(7):657-665. PubMed

5. Onder G, van der Cammen TJ, Petrovic M, et al. Strategies to reduce the risk of iatrogenic illness in complex older adults. Age Ageing. 2013;42(3):284-291. PubMed doi:10.1093/ageing/aft038

6. Volkert J, Schulz H, Hפrter M, et al. The prevalence of mental disorders in older people in Western countries—a meta-analysis. Ageing Res Rev. 2013;12(1):339-353. PubMed doi:10.1016/j.arr.2012.09.004

7. Kessler RC, Birnbaum H, Bromet E, et al. Age differences in major depression: results from the National Comorbidity Survey Replication (NCS-R). Psychol Med. 2010;40(2):225-237. PubMed doi:10.1017/S0033291709990213

8. Isacsson G, Bo׫thius G, Henriksson S, et al. Selective serotonin reuptake inhibitors have broadened the utilisation of antidepressant treatment in accordance with recommendations: findings from a Swedish prescription database. J Affect Disord. 1999;53(1):15-22. PubMed doi:10.1016/S0165-0327(98)00083-4

9. Thompson C, Peveler RC, Stephenson D, et al. Compliance with antidepressant medication in the treatment of major depressive disorder in primary care: a randomized comparison of fluoxetine and a tricyclic antidepressant. Am J Psychiatry. 2000;157(3):338-343. PubMed doi:10.1176/appi.ajp.157.3.338

10. Demyttenaere K. Noncompliance with antidepressants: who’s to blame? Int Clin Psychopharmacol. 1998;13(suppl 2):S19-S25. PubMed doi:10.1097/00004850-199802002-00004

11. Demyttenaere K. Risk factors and predictors of compliance in depression. Eur Neuropsychopharmacol. 2003;13(suppl 3):S69-S75. PubMed doi:10.1016/S0924-977X(03)00095-6

12. Rasmussen JN, Chong A, Alter DA. Relationship between adherence to evidence-based pharmacotherapy and long-term mortality after acute myocardial infarction. JAMA. 2007;297(2):177-186. PubMed doi:10.1001/jama.297.2.177

13. Mazzaglia G, Ambrosioni E, Alacqua M, et al. Adherence to antihypertensive medications and cardiovascular morbidity among newly diagnosed hypertensive patients. Circulation. 2009;120(16):1598-1605. PubMed doi:10.1161/CIRCULATIONAHA.108.830299

14. Stein-Shvachman I, Karpas DS, Werner P. Depression treatment non-adherence and its psychosocial predictors: differences between young and older adults? Aging Dis. 2013;4(6):329-336. PubMed doi:10.14336/AD.2013.0400329

15. Pampallona S, Bollini P, Tibaldi G, et al. Patient adherence in the treatment of depression. Br J Psychiatry. 2002;180:104-109. PubMed doi:10.1192/bjp.180.2.104

16. Pomerantz JM, Finkelstein SN, Berndt ER, et al. Prescriber intent, off-label usage, and early discontinuation of antidepressants: a retrospective physician survey and data analysis. J Clin Psychiatry. 2004;65(3):395-404. PubMed doi:10.4088/JCP.v65n0316

17. Trifirײ G, Tillati S, Spina E, et al. A nationwide prospective study on prescribing pattern of antidepressant drugs in Italian primary care. Eur J Clin Pharmacol. 2013;69(2):227-236. PubMed doi:10.1007/s00228-012-1319-1

18. Chang TE, Jing Y, Yeung AS, et al. Depression monitoring and patient behavior in the Clinical Outcomes in MEasurement-Based Treatment (COMET) trial. Psychiatr Serv. 2014;65(8):1058-1061. PubMed doi:10.1176/appi.ps.201300326

19. Grymonpre R, Cheang M, Fraser M, et al. Validity of a prescription claims database to estimate medication adherence in older persons. Med Care. 2006;44(5):471-477. PubMed doi:10.1097/01.mlr.0000207817.32496.cb

20. Warden D, Trivedi MH, Wisniewski SR, et al. Identifying risk for attrition during treatment for depression. Psychother Psychosom. 2009;78(6):372-379. PubMed doi:10.1159/000235977

21. Costa E, Giardini A, Savin M, et al. Interventional tools to improve medication adherence: review of literature. Patient Prefer Adherence. 2015;9:1303-1314. PubMed doi:10.2147/PPA.S87551

22. Moussavi S, Chatterji S, Verdes E, et al. Depression, chronic diseases, and decrements in health: results from the World Health Surveys. Lancet. 2007;370(9590):851-858. PubMed doi:10.1016/S0140-6736(07)61415-9

23. Prince M, Patel V, Saxena S, et al. No health without mental health. Lancet. 2007;370(9590):859-877. PubMed doi:10.1016/S0140-6736(07)61238-0

24. Ciechanowski PS, Katon WJ, Russo JE. Depression and diabetes: impact of depressive symptoms on adherence, function, and costs. Arch Intern Med. 2000;160(21):3278-3285. PubMed doi:10.1001/archinte.160.21.3278

25. Cooper DC, Trivedi RB, Nelson KM, et al. Antidepressant adherence and risk of coronary artery disease hospitalizations in older and younger adults with depression. J Am Geriatr Soc. 2014;62(7):1238-1245. PubMed doi:10.1111/jgs.12849

26. Gallo JJ, Morales KH, Bogner HR, et al. Long term effect of depression care management on mortality in older adults: follow-up of cluster randomized clinical trial in primary care. BMJ. 2013;346:f2570. PubMed doi:10.1136/bmj.f2570

27. Cutler DM, Everett W. Thinking outside the pillbox—medication adherence as a priority for health care reform. N Engl J Med. 2010;362(17):1553-1555. PubMed doi:10.1056/NEJMp1002305

28. Allan CE, Valkanova V, Ebmeier KP. Depression in older people is underdiagnosed. Practitioner. 2014;258(1771):19-22, 2-3. PubMed