Adherence and Persistence Among Statin Users Aged 65 Years and Over: A Systematic Review and Meta-analysis

Adherence and Persistence Among Statin Users Aged 65 Years and Over: A Systematic Review and... Abstract Background Older people (aged ≥ 65 years) have distinctive challenges with medication adherence. However, adherence and persistence patterns among older statin users have not been comprehensively reviewed. Methods As part of a broader systematic review, we searched Medline, Embase, PsycINFO, CINAHL, Database of Abstracts of Reviews of Effects, CENTRAL, and the National Health Service Economic Evaluation Database through December 2016 for English articles reporting adherence and/or persistence among older statin users. Data were analyzed via descriptive methods and meta-analysis using random-effect modeling. Results Data from more than 3 million older statin users in 82 studies conducted in over 40 countries were analyzed. At 1-year follow-up, 59.7% (primary prevention 47.9%; secondary prevention 62.3%) of users were adherent (medication possession ratio [MPR] or proportion of days covered [PDC] ≥ 80%). For both primary and secondary prevention subjects, 1-year adherence was worse among individuals aged more than 75 years than those aged 65–75 years. At 3 and ≥10 years, 55.3% and 28.4% of users were adherent, respectively. The proportion of users persistent at 1-year was 76.7% (primary prevention 76.0%; secondary prevention 82.6%). Additionally, 68.1% and 61.2% of users were persistent at 2 and 4 years, respectively. Among new statin users, 48.2% were nonadherent and 23.9% discontinued within the first year. The proportion of statin users who were adherent based on self-report was 85.5%. Conclusions There is poor short and long term adherence and persistence among older statin users. Strategies to improve adherence and reduce discontinuation are needed if the intended cardiovascular benefits of statin treatment are to be realized. HMG-CoA reductase inhibitors, Discontinuation, Cardiovascular disease prevention Background The efficacy of statins in the prevention of cardiovascular diseases (CVDs) has been well documented (1). Statins are used by over a billion people globally (2) and are largely considered to be standard-of-care such that new lipid lowering agents are investigated as add-ons to statins rather than as stand-alone treatments (3). However, despite the evidence supporting their clinical benefits, as well being well tolerated by most patients and having a low cost (for generic substitutes), many statin users fail to follow the prescribed dosing regimens (4,5). Studies have linked poor statin adherence and discontinuation to worse clinical outcomes, with significant economic implications (5,6). Statin use is widespread among older populations, being taken by half of American men aged 65–74 years (7), as well as by 40% of Australians aged 65 years and over (8). Further increases in statin utilization are expected due to population ageing, and also as a result of recent changes in cholesterol treatment guidelines recommending the adoption of risk-based management approach (9). To achieve the desired statin effect, optimal adherence is necessary. However, in older patients, adherence to medication is particularly challenging due to factors such as polypharmacy and increased susceptibility to adverse events (10). The evidence supporting the use of statins in the elderly, particularly for primary prevention, is also less clear, and a matter of controversy (11). This is likely to further impact older patients’ willingness to continue and adhere with treatment. Despite the above, synthesized data on adherence and persistence patterns among older statin users are limited; we have not identified any published systematic reviews that specifically address this topic. While previous general reviews provided important information on statin adherence (4,5), they often focused on 12 months of patient follow-up. It is unclear what the patterns of adherence and persistence beyond this period are. In some instances, the distinction between adherence and persistence has not been clearly made (12). In the present study, we sought to (i) characterize the patterns of adherence and persistence among older statin users and (ii) compare adherence and persistence among primary and secondary prevention users. This study was part of a larger review on the patterns and barriers to statin use among older people (13). Methods Search Strategy The detailed protocol for our review on adherence/persistence and their predictors among older statin users has been published (13). We searched Medline, Embase, CINAHL, PsycINFO, Cochrane CENTRAL, Database of Abstracts of Reviews of Effects (DARE), and the National Health Service Economic and Evaluation Database (NHSEED) through December 2016 to identify English articles reporting adherence and/or persistence among older statin users. The search terms adopted involved a combination of those related to the intervention (ie, statins or HMG-CoA reductase inhibitors) and outcomes (ie, adherence, persistence or discontinuation, etc.) (Supplementary Table S1). Study Selection and Evaluation Articles reporting on adherence and/or discontinuation among older statin users (aged ≥65 years) were eligible for inclusion. Studies utilizing any of a variety of methods including pill count, prescription refill records or patient’s self-reports/recalls assessed via validated scales were considered. For studies that measured adherence via the medication possession ratio (MPR), proportion of days covered (PDC) and proportion of doses taken (PDT), we included only those in which adherence was dichotomized with an 80% cutoff. While the 80% threshold is arbitrary, it is often cited to correspond to the minimum adherence level required to achieve a satisfactory clinical effect of statin treatment (12). Persistence data were collected from studies that reported statin discontinuation. Studies on medication discontinuation usually adopt the “permissible-gap” method whereby patients are considered to have discontinued treatment after exceeding the maximum allowable period of no refill (12). The methodological quality of the studies included in this report were assessed via a set of questions formulated with reference to the National Institute of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies (14). This tool was selected based on its robustness yet easy adaptability. Data Extraction and Analysis Two reviewers (R.O. and A.J.) extracted study data and authors were contacted for unpublished data if necessary. When appropriate, meta-analysis using random-effect modeling techniques was performed. Heterogeneity across studies was quantified via the I2 statistic (15). Assessment of publication bias was by direct observation of funnel plots and Egger’s regression test was used to assess for statistical significance (16). The robustness of pooled estimates was tested through leave-one-out sensitivity analyses (15). Subgroup analyses based on anticipated sources of heterogeneity (eg, measurement technique, region of study, etc.) were performed. Statistical significance was set at p less than .05. Analyses were undertaken using StatsDirect Statistical Software (version 3.0, StatsDirect Ltd, Cheshire, UK) and MetaXL. Results Study Characteristics A total of 93 articles were included in the current report (Figure 1). We successfully linked 20 articles (as 9 studies) resulting in 82 unique studies that reported data among 3,048,563 older statin users sampled from over 40 countries (Supplementary Table S2). The included articles were published in the years 2002–2016, with the majority (72.0%) published within the last 5 years (2012–2016). Among 45 studies involving 1,470,650 subjects for which gender breakdown was available, 58.6% of the subjects were females. Per regional distribution of studies, 45% were from North America, 40% from Europe, 14% from the rest of the world, and 1% were cross-regional. The studies included in the review were of reasonable quality, and the majority (72%) were graded as good (Supplementary Figure S1), and the remaining studies were graded as medium. Figure 1. View largeDownload slide Flow chart of systematic review process. Figure 1. View largeDownload slide Flow chart of systematic review process. Patterns of Statin Adherence Fifty-eight studies involving 2,274,890 subjects reported adherence via assessment of database records (prescription, refill, or pharmacy claims data) with duration of follow-up ranging from 6 months to ≥12.6 years. Of the 58 studies, 76% used the PDC method and 24% used the MPR approach. The most frequent follow-up period was 12 months; 39 studies (60 datasets) reported 1-year adherence among 1,194,395 subjects. The pooled proportion (Figure 2) of subjects adherent at 1-year was 59.7% (95% confidence interval [CI] 55.6–64.1). The proportion of primary prevention subjects who were adherent at 1-year was 47.9% (95% CI, 39.8–56.1), compared to 62.3% (95% C1 58.2–66.4) of secondary prevention subjects (p value for difference <.0001). The 1-year adherence was higher among subjects aged 65–75 years than those aged >75 years irrespective of statins being taken for primary (65–75 years = 49.6%, >75 years = 37.3%: p value for difference <.0001) or secondary prevention (65–75 years = 62.6%, >75 years = 58.3%: p value for difference <.0001). Among studies that used the MPR method, 58.5% (95% CI 48.9–67.7) of subjects were adherent at 1 year, whereas 60.4% (95% CI 56.7–64.1) of subjects in studies that utilized the PDC approach were adherent (p value for difference <.0001). For new statin users, 51.8% (95% CI 42.6–61.0) were adherent within the first treatment year. Figue 2. View largeDownload slide Proportion of subjects adherent at 1-year follow-up. Figue 2. View largeDownload slide Proportion of subjects adherent at 1-year follow-up. From the pooled data (Table 1), 59.9%, 59.6%, 55.3%, 35.9%, 35.7%, and 28.4% of older statin users were adherent at 6 months, 2, 3, 4, 5, and ≥10 years, respectively. The 6-month, 1, 2, and 3-year pooled adherence were unchanged by a leave-one-out sensitivity analyses (plots not shown). Funnel plot assessments showed no obvious evidence of publication bias (Supplementary Figure S2) and this was confirmed via Egger’s regression results for 6 months (p value = .661), 1 year (p value = .6263), 2 years (p value = .5061), 3 years (p value = .4933), 4 years (p value = .4493), 5 years (p value = .4987), and ≥10 years (p value = .2442). Table 1. Proportion of Older Statin Users Adherent at Various Follow-up Periods‡ Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Note: CI = Confidence interval; MPR = Medication possession ratio; PDC = Proportion of days covered. ‡Only studies that reported adherence via assessment of database records (MPR or PDC) are included in this table. *The majority of studies involved mix of patients but did not stratify results according to indication nor by age group (65–75 vs >75); †no specific defined year (eg, 18 mo, 4.3 y, etc.). View Large Table 1. Proportion of Older Statin Users Adherent at Various Follow-up Periods‡ Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Note: CI = Confidence interval; MPR = Medication possession ratio; PDC = Proportion of days covered. ‡Only studies that reported adherence via assessment of database records (MPR or PDC) are included in this table. *The majority of studies involved mix of patients but did not stratify results according to indication nor by age group (65–75 vs >75); †no specific defined year (eg, 18 mo, 4.3 y, etc.). View Large Three other observational studies involving 190 subjects measured adherence using validated self-reporting scales; 85.5% (95% CI 77.6–92.1, I2 = 0.99) of the subjects were adherent based on self-reports. Patterns of Statin Persistence Thirty-three studies involving 872,098 subjects reported proportion of patients persisting with treatment. The majority (67%) of studies adopted the permissible gap method, that is, they defined a maximum break in refill beyond which patients were deemed to have discontinued treatment (12). However, significant disparities in the criteria applied were evident (eg, the permissible gap varied from 30 days to 365 days). The inconsistencies precluded the conduct of a meta-analysis. Instead, we summarized the data descriptively using median and interquartile range (IQR) as adopted in other medication utilization reviews (17). We did not report the summary data in the form of range because the intervals were extremely wide (Supplementary Figure S3). Statistical estimates of differences between groups were not assessed as variance would have been underestimated (18). Median was not computed when data were available from less than four studies. Twenty-six studies (49 datasets) involving 742,691 subjects reported persistence for 1 year. The overall proportion of older statin users persistent at 1 year was 76.7% (IQR 71.4–87.4). Among primary prevention subjects, 76.0% (IQR 39.5–85.0) were persistent at 1 year compared with 82.6% (65.0–88.8) of secondary prevention subjects. For new statin users, 76.1% (IQR 71.1–82.8) persisted with treatment at the first anniversary. The median proportion of users persistent at 6 months, 2, 3, and 4 years was 82.6%, 68.1%, 63.3%, and 61.2%, respectively (Table 2). Few studies reported persistence at 5 and ≥10 years follow-up, 2 and 1, respectively. In all of these studies, the proportion of patients persistent was less than 60%. Three studies did not involve a specific year of follow-up and median was not computed. Table 2. Proportion of Older Statin Users Persistent at Various Follow-up Periods Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Note: IQR = Interquartile range. *The permissible gap varied from 30 days to 365 days across studies; †Other includes for instance reliance on patient reports; ‡The majority of studies did not stratify results according to statin indication and were excluded from this analysis. View Large Table 2. Proportion of Older Statin Users Persistent at Various Follow-up Periods Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Note: IQR = Interquartile range. *The permissible gap varied from 30 days to 365 days across studies; †Other includes for instance reliance on patient reports; ‡The majority of studies did not stratify results according to statin indication and were excluded from this analysis. View Large Discussion We report suboptimal statin use among older patients, with 40% nonadherence at 1 year and more than one-third discontinuing treatment by 3 years. Although recent (2012–2016) publications reported slightly higher adherence and persistence than earlier ones, the trend is notable for a population at increased risk of cardiovascular events in whom more than 80% of CVD-related deaths occur (19). It appears that the first 180 days of follow-up is the most critical period when many patients become nonadherent or discontinue treatment. Nonetheless, we observed no clear evidence of plateau in the adherence or persistence rates during follow-up, suggesting the need to provide long-term treatment support. The high self-reported adherence observed in our study is consistent with previous research, that suggest that adherence measured via self-reporting scales or patient interviews tend to overestimate levels of medication use for multiple reasons that include recall and social desirability bias (20). Thus, while assessment of adherence via self-reports may be time saving, inexpensive and easy to apply in routine clinical practice (20), the value of using them to understand how well patients are following the statin dosing regimen is low. There is strong evidence supporting the long-term use of statins to achieve the desired effects both on cardiovascular morbidity and mortality. In the Scandinavian Simvastatin Survival Study (4S) trial, the biochemical effect of simvastatin was observed after a few weeks of treatment but the effects on coronary artery disease morbidity and mortality were only observed after approximately 24 months (21). However, as per our analysis, nearly one in four patients newly initiated on statin discontinued within the first 12 months of treatment. It is possible that patients stopped treatment as a result of poor drug effectiveness or due to adverse effects especially as ageing may result in altered pharmacokinetics and pharmacodynamics. Nonetheless, an analysis of the tolerability of 4,924 older people in 50 statin trials showed that only 2.1% of treatment discontinuation was due to adverse events (22). Similarly, in a retrospective cohort study with over 100,000 patients, more than half discontinued statin treatment but less than 4% of discontinuers cited adverse effects as the reason for discontinuation (23). The perceived benefits or susceptibility to disease as explained by the Health Belief Model has been shown to positively correlate with patients’ behavior (24). Since dyslipidaemia is an asymptomatic condition, patients may be unaware of their risk of adverse health outcomes. Statin users who have experienced an event (eg, myocardial infarction [MI], ie, secondary prevention) may consider themselves to be at higher risk, as well as exhibit greater belief in the severity of the diseases being prevented and may therefore be more committed to treatment (24). Kronish and colleagues (25) reported that among older patients who were nonadherent to statins, 38% became adherent following hospitalization for acute MI. Thus, perception of low risk may partly explain the poorer adherence and persistence observed among the primary prevention subjects. Nonetheless, considering that only 62% of the secondary prevention subjects in our study were adherent at 1 year, other factors beyond probabilistic information (ie, benefit or risk) may additionally contribute to determining patients’ behavior. Polypharmacy is prevalent among older populations and the risk of drug–drug interactions (DDIs) increases with the number of medications. For example, a patient on five to nine medications has about 50% probability of experiencing a DDI and this increases to 100% when the number of medications exceeds 20 (26). In addition, metabolism enzymes become less functional at advanced ages resulting in increased statin area under the curve (AUC) and higher likelihood of DDIs. Side effects are also common when more medications are used, with as much as 40% of hospitalizations in older patients attributed to adverse drug events (26). As a result, increasing polypharmacy has frequently been correlated with poor statin adherence (6). Older patients may also experience significant cognitive decline as well as face other barriers (eg, difficulty in swallowing) which may all affect adherence (10). In recent years also, there has been intensified debate around the use of statins in older adults (11). Analyses from several countries including the United Kingdom (27) and Denmark (28) indicate that the polarized views on the interpretation of the effects of statin treatment have had adverse impacts on patients’ adherence and continuation rates. In the very old (>75 years) particularly, guidelines suggest a lack of evidence to support the use of statins for primary prevention (29). This may partly explain the worst adherence observed among subjects aged more than 75 years who were taking statins for primary prevention. Ongoing trials such as STAREE (NCT02099123) and SITE (NCT02547883) will provide important evidence about the use of statin for primary prevention in the elderly. STAREE will examine whether treatment with atorvastatin 40mg/day versus placebo in healthy elderly individuals (≥70 years) will prolong overall survival or disability-free survival. SITE will evaluate the cost-effectiveness of stopping statins in people aged ≥75 years—the primary endpoints being overall mortality and incremental cost per quality-adjusted life-year (QALY) gained. The results of both STAREE and SITE are expected after 2020. The clinical implications of our findings are important. There is increasing evidence linking poor statin adherence and discontinuation to worse clinical outcomes (5). For example, patients who discontinue statin treatment following MI are about 3 times more likely to die than those continuing treatment (30). Compared with adherent patients, those nonadherent to statins and antihypertensive medications also have more than sevenfold increased risk of fatal stroke (31). Unsurprisingly, a meta-analysis attributed nine excess CVD deaths per 100,000 to nonadherence to cardiovascular medications including statins (5). The observed poor statin adherence and persistence may also be indicative of widespread suboptimal preventive medication use among the older population. Indeed, some studies suggest that nonadherence to one medication class may predict suboptimal use of another. Among 2,695 older Americans who initiated statin treatment following hospitalization for acute MI, low adherence to antihypertensive medications was associated with 62% increased risk of statin nonadherence (32). The economic impact of not using prescribed medications is significant, with studies frequently associating poor statin adherence and discontinuation with lower cost-effectiveness (33). In most developed countries, CVDs attract higher costs than any other disease class. In six European countries (France, Germany, Spain, Italy, Sweden, and the United Kingdom), the total financial impact of CVDs exceeded €102 billion in 2014 (34). The use of evidence-based pharmacological therapies such as statins remains the cornerstone of CVD prevention and treatment. Thus, poor adherence/persistence represents a missed opportunity to realize the full clinical potential of preventive therapies that are intended to reduce CVD burden. Consequently, in the United States alone, the yearly economic loss due to medication nonadherence has been estimated as $290 billion (35). It has been noted that if adherence to prescribed medications were to improve, it would have far greater impact on population health than improvement in specific medical treatments (36). Interventions targeted at patients and health professionals are necessary to address the poor statin adherence and persistence. Patient counseling/education, reinforcement and reminders have been observed to improve adherence (37). A Cochrane review found reinforcement and reminders increased adherence by about 24% (38). Strategies that seek to enhance patient–physician communication or those aimed at minimizing regimen complexity have also shown some degree of success (37). Other interventions that target the health delivery system such as through provision of extended care via ancillary health staff or those meant to eliminate health system cost-containment measures such as copayments have been considered necessary (6). Ito and colleagues for instance demonstrated that providing full medication reimbursement (for cardiovascular drugs including statins) to post-MI patients is cost-effective (39). However, in view of the complexity of the drivers of suboptimal adherence/persistence, the greatest improvements are likely to be achieved via the adoption of a mix of patient-tailored long-term interventions (6,36–38). This review has some limitations. First, studies do not report in detail actual reasons for nonadherence or discontinuation. It is possible that some discontinuation was due to adverse effects or initiated by clinicians. Second, adherence/persistence was based on indirect measurements and we cannot ascertain whether patients actually took the medication (12). Third, we restricted our study to English articles, and this is likely to limit the generalizability. Fourth, patients who discontinue statins may reinitiate treatment and this has not been incorporated in our analysis. Fifth, although some publications have offered guidance regarding the definitions and measurement of medication adherence and persistence (12), we noticed significant inconsistencies in the literature particularly with regards to the assessment of statin discontinuation. Persistence was therefore analyzed descriptively, although, meta-analysis would have provided a more robust outcome. Finally, we observed heterogeneity (I2 > 90 in many cases) which may be attributed to multiple factors including patient characteristics and data sources. Our study has important strengths worth mentioning. We applied a specific adherence cutoff ensuring consistency across studies. No geographic or time restrictions were imposed allowing the inclusion of a large body of evidence. Contact made with study authors ensured the inclusion of a large amount of unpublished data allowing us to provide for the first time a pooled estimate of self-reported statin adherence. To our knowledge, ours is the first systematic review to offer some understanding of the patterns of long-term (>1 year) adherence and persistence among statin users. Our pooled adherence estimates are robust and the absence of publication bias supports the reliability of our findings. Lastly, considering that >70% of the included studies were recently published, our reported adherence and persistence estimates may speak to current trends. Conclusions There is poor short and long-term adherence and persistence among older statin users. Strategies to improve adherence and reduce discontinuation are needed to close the gap if the intended cardiovascular benefits of statin treatment are to be realized. Supplementary Material Supplementary data is available at The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences online. Funding R.O. is sponsored by a Monash Graduate Scholarship and Monash International Postgraduate Research Scholarship for his doctoral studies. S.Z. is funded by a National Health and Medical Research Council Senior Research Fellowship. No other funding has been required to undertake this work. Conflict of Interest S.Z. reports past participation in advisory boards and/or receiving honoraria from: Amgen Australia; AstraZeneca/Bristol-Myers Squibb Australia; Janssen-Cilag; Merck, Sharp, and Dohme (Australia); Novartis Australia; Novo Nordisk; Sanofi; Servier Laboratories; Takeda Australia; and Monash University (undertaking contract work for AstraZeneca Pty Limited/Bristol-Myers Squibb Australia Pty Limited) for work unrelated to this study. Acknowledgments We are grateful to authors who generously provided us with unpublished data. We also thank Lorena Romero, a Senior Medical Librarian (Ian Potter Library, Alfred Hospital, Melbourne, Australia) who reviewed our initial search strategy and provided useful feedback. References 1. Naci H , Brugts JJ , Fleurence R , Tsoi B , Toor H , Ades AE . Comparative benefits of statins in the primary and secondary prevention of major coronary events and all-cause mortality: a network meta-analysis of placebo-controlled and active-comparator trials . Eur J Prev Cardiol . 2013 ; 20 : 641 – 657 . doi: 10.1177/2047487313480435 Google Scholar CrossRef Search ADS PubMed 2. Ioannidis JP . More than a billion people taking statins?: potential implications of the new cardiovascular guidelines . JAMA 2014 ; 311 : 463 – 464 . doi: 10.1001/jama.2013.284657 Google Scholar CrossRef Search ADS PubMed 3. Sabatine MS , Giugliano RP , Keech AC et al. Evolocumab and clinical outcomes in patients with cardiovascular disease . New EngJ Med . 2017 ; 376 : 1713 – 1722 .doi: 10.1056/NEJMoa1615664 Google Scholar CrossRef Search ADS 4. Lemstra M , Blackburn D , Crawley A , Fung R . Proportion and risk indicators of nonadherence to statin therapy: a meta-analysis . Can J Cardiol . 2012 ; 28 : 574 – 580 . doi: 10.1016/j.cjca.2012.05.007 Google Scholar CrossRef Search ADS PubMed 5. Chowdhury R , Khan H , Heydon E et al. Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences . Eur Heart J . 2013 ; 34 : 2940 – 2948 . doi: 10.1093/eurheartj/eht295 Google Scholar CrossRef Search ADS PubMed 6. Phan K , Gomez YH , Elbaz L , Daskalopoulou SS . Statin treatment non-adherence and discontinuation: clinical implications and potential solutions . Curr Pharm Des . 2014 ; 20 : 6314 – 6324 . doi:10.2174/1381612820666140620162629 Google Scholar CrossRef Search ADS PubMed 7. National Center for Health Statistics: National center for health statistics. Health, united states, 2010 : With Special Feature on Death and Dying . Hyattsville, MD: US Department of Health and Human Services ; 2011 . 8. Morgan TK , Williamson M , Pirotta M , Stewart K , Myers SP , Barnes J . A national census of medicines use: a 24-hour snapshot of Australians aged 50 years and older . Med J Aust . 2012 ; 196 : 50 – 53 . doi:10.5694/mja11.10698 Google Scholar CrossRef Search ADS PubMed 9. Pencina MJ , Navar-Boggan AM , D’Agostino RB , Sr. et al. Application of new cholesterol guidelines to a population-based sample . New Eng J Med . 2014 ; 370 : 1422 – 1431 . doi: 10.1056/NEJMoa1315665 Google Scholar CrossRef Search ADS PubMed 10. Yap AF , Thirumoorthy T , Kwan YH . Systematic review of the barriers affecting medication adherence in older adults . Geriatr Gerontol Int . 2016 ; 16 : 1093 – 1101 . doi: 10.1111/ggi.12616 Google Scholar CrossRef Search ADS PubMed 11. Stone NJ , Intwala S , Katz D . Statins in very elderly adults (debate) . J Am Geriatr Soc . 2014 ; 62 : 943 – 945 . doi: 10.1111/jgs.12788_1 Google Scholar CrossRef Search ADS PubMed 12. Cramer JA , Roy A , Burrell A et al. Medication compliance and persistence: terminology and definitions . Value Health . 2008 ; 11 : 44 – 47 . doi: 10.1111/j.1524-4733.2007.00213.x Google Scholar CrossRef Search ADS PubMed 13. Ofori-Asenso R , Zomer E , Curtis A et al. Patterns and predictors of adherence to statin therapy among older patients: protocol for a systematic review . JMIR Res Protoc . 2017 ; 6 : e39 . doi: 10.2196/resprot.7195 Google Scholar CrossRef Search ADS PubMed 14. National Institute of Health: Quality assessment tool for observational cohort and cross-sectional studies. NIH 2014 . https://www.nhlbi.nih.gov/health-pro/guidelines/in-develop/cardiovascular-risk-reduction/tools/cohort ( Accessed 13 November 2016 ). 15. Higgins J , Green S . Cochrane handbook for systematic reviews of interventions . Wiley Online Library , 2008 . http://au.wiley.com/WileyCDA/WileyTitle/productCd-0470699515.html# (Accessed 11 November 2016). 16. Egger M , Davey Smith G , Schneider M , Minder C . Bias in meta-analysis detected by a simple, graphical test . BMJ 1997 ; 315 : 629 – 634 . doi: 10.1136/bmj.315.7109.629 Google Scholar CrossRef Search ADS PubMed 17. World Health Organization, Harvard Medical School and Harvard Pilgrim Health: Medicines Use in Primary Care in Developing and Transitional Countries Fact Book Summarizing Results from Studies Reported Between 1990 and 2006 . Geneva : World Health Organization ; 2009 . http://www.who.int/medicines/publications/who_emp_2009.3/en/ ( Accessed 20 February 2017 ). 18. McDonald J : Handbook of Biological Statistics . 2nd ed . Baltimore, MD : Sparky House Publishing ; 2009 . 19. Yazdanyar A , Newman AB . The burden of cardiovascular disease in the elderly: morbidity, mortality, and costs . Clin Geriatr Med . 2009 ; 25 : 563 – 77, vii . doi: 10.1016/j.cger.2009.07.007 Google Scholar CrossRef Search ADS PubMed 20. Nunes V , Neilson J , O’Flynn N et al. Clinical Guidelines and Evidence Review for Medicines Adherence: Involving Patients in Decisions About Prescribed Medicines and Supporting Adherence . London : National Collaborating Centre for Primary Care and Royal College of General Practitioners ; 2009 . https://www.nice.org.uk/guidance/cg76/evidence/full-guideline-242062957 ( Accessed 05 February 2017 ). 21. Pedersen TR , Kjekshus J , Berg K et al. Scandinavian Simvastatin Survival Study G: randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the scandinavian simvastatin survival study (4s) . Atherosclerosis Supplements 2004 ; 5 : 81 – 87 . doi: 10.1016/j.atherosclerosissup.2004.08.027 Google Scholar CrossRef Search ADS PubMed 22. Hey-Hadavi JH , Kuntze E , Luo D , Silverman P , Pittman D , Lepetri B . Tolerability of atorvastatin in a population aged > or =65 years: a retrospective pooled analysis of results from fifty randomized clinical trials . Am J Geriatr Pharmacother . 2006 ; 4 : 112 – 122 . doi: 10.1016/j.amjopharm.2006.06.001 Google Scholar CrossRef Search ADS PubMed 23. Zhang H , Plutzky J , Skentzos S et al. Discontinuation of statins in routine care settings: a cohort study . Ann Intern Med . 2013 ; 158 : 526 – 534 . doi: 10.7326/0003-4819-158-7-201304020-00004 Google Scholar CrossRef Search ADS PubMed 24. DiMatteo MR , Haskard KB , Williams SL . Health beliefs, disease severity, and patient adherence: a meta-analysis . Med Care . 2007 ; 45 : 521 – 528 . doi: 10.1097/MLR.0b013e318032937e Google Scholar CrossRef Search ADS PubMed 25. Kronish IM , Ross JS , Zhao H , Muntner P . Impact of hospitalization for acute myocardial infarction on adherence to statins among older adults . Circ Cardiovasc Qual Outcomes . 2016 ; 9 : 364 – 371 . doi: 10.1161/CIRCOUTCOMES.115.002418 Google Scholar CrossRef Search ADS PubMed 26. Maher RL , Hanlon J , Hajjar ER . Clinical consequences of polypharmacy in elderly . Expert Opin Drug Saf . 2014 ; 13 : 57 – 65 . doi: 10.1517/14740338.2013.827660 Google Scholar CrossRef Search ADS PubMed 27. Matthews A , Herrett E , Gasparrini A et al. Impact of statin related media coverage on use of statins: interrupted time series analysis with UK primary care data . BMJ . 2016 ; 353 : i3283 . Google Scholar CrossRef Search ADS PubMed 28. Nielsen SF , Nordestgaard BG . Negative statin-related news stories decrease statin persistence and increase myocardial infarction and cardiovascular mortality: a nationwide prospective cohort study . Eur Heart J . 2016 ; 37 : 908 – 916 . doi: 10.1093/eurheartj/ehv641 Google Scholar CrossRef Search ADS PubMed 29. Stone NJ , Robinson JG , Lichtenstein AH et al. ; American College of Cardiology/American Heart Association Task Force on Practice Guidelines . 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines . J Am Coll Cardiol . 2014 ; 63 ( 25 Pt B ): 2889 – 2934 . doi: 10.1016/j.jacc.2013.11.002 Google Scholar CrossRef Search ADS PubMed 30. Ho PM , Spertus JA , Masoudi FA et al. Impact of medication therapy discontinuation on mortality after myocardial infarction . Arch Intern Med . 2006 ; 166 : 1842 – 1847 . doi: 10.1001/archinte.166.17.1842 Google Scholar CrossRef Search ADS PubMed 31. Herttua K , Martikainen P , Batty GD , Kivimäki M . Poor adherence to statin and antihypertensive therapies as risk factors for fatal stroke . J Am Coll Cardiol . 2016 ; 67 : 1507 – 1515 . doi: 10.1016/j.jacc.2016.01.044 Google Scholar CrossRef Search ADS PubMed 32. Muntner P , Yun H , Sharma P et al. Ability of low antihypertensive medication adherence to predict statin discontinuation and low statin adherence in patients initiating treatment after a coronary event . Am J Cardiol . 2014 ; 114 : 826 – 831 . doi: 10.1016/j.amjcard.2014.06.009 Google Scholar CrossRef Search ADS PubMed 33. Aarnio E , Korhonen MJ , Huupponen R , Martikainen J . Cost-effectiveness of statin treatment for primary prevention in conditions of real-world adherence–estimates from the Finnish prescription register . Atherosclerosis . 2015 ; 239 : 240 – 247 . doi: 10.1016/j.atherosclerosis.2014.12.059 Google Scholar CrossRef Search ADS PubMed 34. Centre for Economics and Business Research: The Economic Cost of Cardiovascular Disease from 2014–2020 in six European Economies, Cebr , 2014 . https://cebr.com/reports/the-rising-cost-of-cvd/ ( Accessed 02 February 2017 ). 35. NEHI Institute: Thinking Outside the Pillbox a System-Wide Approach to Improving Patient Medication Adherence for Chronic Disease . New England Healthcare Institute Web site , 2009 . http://www.nehi.net/writable/publication_files/file/pa_issue_brief_final.pdf ( Accessed 05 February 2017 ). 36. Haynes RB , Ackloo E , Sahota N , McDonald HP , Yao X . Interventions for enhancing medication adherence . Cochrane Database Syst Rev . 2008 : CD000011 . doi: 10.1002/14651858.CD000011.pub3 37. Maningat P , Gordon BR , Breslow JL . How do we improve patient compliance and adherence to long-term statin therapy ? Curr Atheroscler Rep . 2013 ; 15 : 291 . doi: 10.1007/s11883-012-0291-7 Google Scholar CrossRef Search ADS PubMed 38. Schedlbauer A , Davies P , Fahey T . Interventions to improve adherence to lipid lowering medication . Cochrane Database Syst Rev . 2010 : CD004371 . doi: 10.1002/14651858.CD004371.pub3 39. Ito K , Avorn J , Shrank WH et al. Long-term cost-effectiveness of providing full coverage for preventive medications after myocardial infarction . Circ Cardiovasc Qual Outcomes . 2015 ; 8 : 252 – 259 . doi: 10.1161/CIRCOUTCOMES.114.001330 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences Oxford University Press

Loading next page...
 
/lp/ou_press/adherence-and-persistence-among-statin-users-aged-65-years-and-over-a-G07qzmYq6b
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1079-5006
eISSN
1758-535X
D.O.I.
10.1093/gerona/glx169
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Older people (aged ≥ 65 years) have distinctive challenges with medication adherence. However, adherence and persistence patterns among older statin users have not been comprehensively reviewed. Methods As part of a broader systematic review, we searched Medline, Embase, PsycINFO, CINAHL, Database of Abstracts of Reviews of Effects, CENTRAL, and the National Health Service Economic Evaluation Database through December 2016 for English articles reporting adherence and/or persistence among older statin users. Data were analyzed via descriptive methods and meta-analysis using random-effect modeling. Results Data from more than 3 million older statin users in 82 studies conducted in over 40 countries were analyzed. At 1-year follow-up, 59.7% (primary prevention 47.9%; secondary prevention 62.3%) of users were adherent (medication possession ratio [MPR] or proportion of days covered [PDC] ≥ 80%). For both primary and secondary prevention subjects, 1-year adherence was worse among individuals aged more than 75 years than those aged 65–75 years. At 3 and ≥10 years, 55.3% and 28.4% of users were adherent, respectively. The proportion of users persistent at 1-year was 76.7% (primary prevention 76.0%; secondary prevention 82.6%). Additionally, 68.1% and 61.2% of users were persistent at 2 and 4 years, respectively. Among new statin users, 48.2% were nonadherent and 23.9% discontinued within the first year. The proportion of statin users who were adherent based on self-report was 85.5%. Conclusions There is poor short and long term adherence and persistence among older statin users. Strategies to improve adherence and reduce discontinuation are needed if the intended cardiovascular benefits of statin treatment are to be realized. HMG-CoA reductase inhibitors, Discontinuation, Cardiovascular disease prevention Background The efficacy of statins in the prevention of cardiovascular diseases (CVDs) has been well documented (1). Statins are used by over a billion people globally (2) and are largely considered to be standard-of-care such that new lipid lowering agents are investigated as add-ons to statins rather than as stand-alone treatments (3). However, despite the evidence supporting their clinical benefits, as well being well tolerated by most patients and having a low cost (for generic substitutes), many statin users fail to follow the prescribed dosing regimens (4,5). Studies have linked poor statin adherence and discontinuation to worse clinical outcomes, with significant economic implications (5,6). Statin use is widespread among older populations, being taken by half of American men aged 65–74 years (7), as well as by 40% of Australians aged 65 years and over (8). Further increases in statin utilization are expected due to population ageing, and also as a result of recent changes in cholesterol treatment guidelines recommending the adoption of risk-based management approach (9). To achieve the desired statin effect, optimal adherence is necessary. However, in older patients, adherence to medication is particularly challenging due to factors such as polypharmacy and increased susceptibility to adverse events (10). The evidence supporting the use of statins in the elderly, particularly for primary prevention, is also less clear, and a matter of controversy (11). This is likely to further impact older patients’ willingness to continue and adhere with treatment. Despite the above, synthesized data on adherence and persistence patterns among older statin users are limited; we have not identified any published systematic reviews that specifically address this topic. While previous general reviews provided important information on statin adherence (4,5), they often focused on 12 months of patient follow-up. It is unclear what the patterns of adherence and persistence beyond this period are. In some instances, the distinction between adherence and persistence has not been clearly made (12). In the present study, we sought to (i) characterize the patterns of adherence and persistence among older statin users and (ii) compare adherence and persistence among primary and secondary prevention users. This study was part of a larger review on the patterns and barriers to statin use among older people (13). Methods Search Strategy The detailed protocol for our review on adherence/persistence and their predictors among older statin users has been published (13). We searched Medline, Embase, CINAHL, PsycINFO, Cochrane CENTRAL, Database of Abstracts of Reviews of Effects (DARE), and the National Health Service Economic and Evaluation Database (NHSEED) through December 2016 to identify English articles reporting adherence and/or persistence among older statin users. The search terms adopted involved a combination of those related to the intervention (ie, statins or HMG-CoA reductase inhibitors) and outcomes (ie, adherence, persistence or discontinuation, etc.) (Supplementary Table S1). Study Selection and Evaluation Articles reporting on adherence and/or discontinuation among older statin users (aged ≥65 years) were eligible for inclusion. Studies utilizing any of a variety of methods including pill count, prescription refill records or patient’s self-reports/recalls assessed via validated scales were considered. For studies that measured adherence via the medication possession ratio (MPR), proportion of days covered (PDC) and proportion of doses taken (PDT), we included only those in which adherence was dichotomized with an 80% cutoff. While the 80% threshold is arbitrary, it is often cited to correspond to the minimum adherence level required to achieve a satisfactory clinical effect of statin treatment (12). Persistence data were collected from studies that reported statin discontinuation. Studies on medication discontinuation usually adopt the “permissible-gap” method whereby patients are considered to have discontinued treatment after exceeding the maximum allowable period of no refill (12). The methodological quality of the studies included in this report were assessed via a set of questions formulated with reference to the National Institute of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies (14). This tool was selected based on its robustness yet easy adaptability. Data Extraction and Analysis Two reviewers (R.O. and A.J.) extracted study data and authors were contacted for unpublished data if necessary. When appropriate, meta-analysis using random-effect modeling techniques was performed. Heterogeneity across studies was quantified via the I2 statistic (15). Assessment of publication bias was by direct observation of funnel plots and Egger’s regression test was used to assess for statistical significance (16). The robustness of pooled estimates was tested through leave-one-out sensitivity analyses (15). Subgroup analyses based on anticipated sources of heterogeneity (eg, measurement technique, region of study, etc.) were performed. Statistical significance was set at p less than .05. Analyses were undertaken using StatsDirect Statistical Software (version 3.0, StatsDirect Ltd, Cheshire, UK) and MetaXL. Results Study Characteristics A total of 93 articles were included in the current report (Figure 1). We successfully linked 20 articles (as 9 studies) resulting in 82 unique studies that reported data among 3,048,563 older statin users sampled from over 40 countries (Supplementary Table S2). The included articles were published in the years 2002–2016, with the majority (72.0%) published within the last 5 years (2012–2016). Among 45 studies involving 1,470,650 subjects for which gender breakdown was available, 58.6% of the subjects were females. Per regional distribution of studies, 45% were from North America, 40% from Europe, 14% from the rest of the world, and 1% were cross-regional. The studies included in the review were of reasonable quality, and the majority (72%) were graded as good (Supplementary Figure S1), and the remaining studies were graded as medium. Figure 1. View largeDownload slide Flow chart of systematic review process. Figure 1. View largeDownload slide Flow chart of systematic review process. Patterns of Statin Adherence Fifty-eight studies involving 2,274,890 subjects reported adherence via assessment of database records (prescription, refill, or pharmacy claims data) with duration of follow-up ranging from 6 months to ≥12.6 years. Of the 58 studies, 76% used the PDC method and 24% used the MPR approach. The most frequent follow-up period was 12 months; 39 studies (60 datasets) reported 1-year adherence among 1,194,395 subjects. The pooled proportion (Figure 2) of subjects adherent at 1-year was 59.7% (95% confidence interval [CI] 55.6–64.1). The proportion of primary prevention subjects who were adherent at 1-year was 47.9% (95% CI, 39.8–56.1), compared to 62.3% (95% C1 58.2–66.4) of secondary prevention subjects (p value for difference <.0001). The 1-year adherence was higher among subjects aged 65–75 years than those aged >75 years irrespective of statins being taken for primary (65–75 years = 49.6%, >75 years = 37.3%: p value for difference <.0001) or secondary prevention (65–75 years = 62.6%, >75 years = 58.3%: p value for difference <.0001). Among studies that used the MPR method, 58.5% (95% CI 48.9–67.7) of subjects were adherent at 1 year, whereas 60.4% (95% CI 56.7–64.1) of subjects in studies that utilized the PDC approach were adherent (p value for difference <.0001). For new statin users, 51.8% (95% CI 42.6–61.0) were adherent within the first treatment year. Figue 2. View largeDownload slide Proportion of subjects adherent at 1-year follow-up. Figue 2. View largeDownload slide Proportion of subjects adherent at 1-year follow-up. From the pooled data (Table 1), 59.9%, 59.6%, 55.3%, 35.9%, 35.7%, and 28.4% of older statin users were adherent at 6 months, 2, 3, 4, 5, and ≥10 years, respectively. The 6-month, 1, 2, and 3-year pooled adherence were unchanged by a leave-one-out sensitivity analyses (plots not shown). Funnel plot assessments showed no obvious evidence of publication bias (Supplementary Figure S2) and this was confirmed via Egger’s regression results for 6 months (p value = .661), 1 year (p value = .6263), 2 years (p value = .5061), 3 years (p value = .4933), 4 years (p value = .4493), 5 years (p value = .4987), and ≥10 years (p value = .2442). Table 1. Proportion of Older Statin Users Adherent at Various Follow-up Periods‡ Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Note: CI = Confidence interval; MPR = Medication possession ratio; PDC = Proportion of days covered. ‡Only studies that reported adherence via assessment of database records (MPR or PDC) are included in this table. *The majority of studies involved mix of patients but did not stratify results according to indication nor by age group (65–75 vs >75); †no specific defined year (eg, 18 mo, 4.3 y, etc.). View Large Table 1. Proportion of Older Statin Users Adherent at Various Follow-up Periods‡ Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Follow-up Period Number of Studies Sample Size % Adherent (95% CI) I2 Statistic 6 mo 9 929,815 59.9 (57.3–66.0) 0.96 1 y Overall 39 1,194,395 59.7 (55.6–64.1) 0.99 Assessment method  MPR 9 348,233 58.5 (48.9–67.7) 0.96  PDC 30 846,162 60.4 (56.7–64.1) 0.99 Statin indication*  Primary prevention 6 91,233 47.9 (39.7–56.1) 0.97   65–75 y 5 80,821 49.6 (43.5–55.7) 0.99   >75 y 2 6,124 37.3 (20.0–56.5) 0.99  Secondary prevention 21 172,004 62.3 (58.2–66.4) 0.99   65–75 y 9 46,204 62.6 (52.6–71.9) 0.99   >75 y 6 11,205 58.3 (44.4–71.5) 0.99 Participant Type  New Users 12 177,021 51.8 (42.6–61.0) 0.99  Mixed patients 27 1,017,374 63.5 (59.5–63.7) 0.99 Region of Study  North America 22 949,083 61.5 (57.1–65.8) 0.99  Rest of the world 17 245,312 57.9 (50.7–64.9) 0.99 Publication period  <2012 10 163,608 54.1 (41.9–66.0) 0.98  2012–2016 29 1,030,787 61.9 (58.3–65.3) 0.97 2 y 15 134,800 59.6 (49.6–61.9) 0.96 3 y 6 121,962 55.3 (46.6–63.9) 0.99 4 y 3 45,666 35.9 (14.3–61.0) 0.99 5 y 3 14,747 35.7 (19.2–54.0) 0.99 ≥10 y 3 9,600 28.4 (13.3–46.3) 0.98 Other† 8 103,745 57.5 (50.3–64.5) 0.98 Note: CI = Confidence interval; MPR = Medication possession ratio; PDC = Proportion of days covered. ‡Only studies that reported adherence via assessment of database records (MPR or PDC) are included in this table. *The majority of studies involved mix of patients but did not stratify results according to indication nor by age group (65–75 vs >75); †no specific defined year (eg, 18 mo, 4.3 y, etc.). View Large Three other observational studies involving 190 subjects measured adherence using validated self-reporting scales; 85.5% (95% CI 77.6–92.1, I2 = 0.99) of the subjects were adherent based on self-reports. Patterns of Statin Persistence Thirty-three studies involving 872,098 subjects reported proportion of patients persisting with treatment. The majority (67%) of studies adopted the permissible gap method, that is, they defined a maximum break in refill beyond which patients were deemed to have discontinued treatment (12). However, significant disparities in the criteria applied were evident (eg, the permissible gap varied from 30 days to 365 days). The inconsistencies precluded the conduct of a meta-analysis. Instead, we summarized the data descriptively using median and interquartile range (IQR) as adopted in other medication utilization reviews (17). We did not report the summary data in the form of range because the intervals were extremely wide (Supplementary Figure S3). Statistical estimates of differences between groups were not assessed as variance would have been underestimated (18). Median was not computed when data were available from less than four studies. Twenty-six studies (49 datasets) involving 742,691 subjects reported persistence for 1 year. The overall proportion of older statin users persistent at 1 year was 76.7% (IQR 71.4–87.4). Among primary prevention subjects, 76.0% (IQR 39.5–85.0) were persistent at 1 year compared with 82.6% (65.0–88.8) of secondary prevention subjects. For new statin users, 76.1% (IQR 71.1–82.8) persisted with treatment at the first anniversary. The median proportion of users persistent at 6 months, 2, 3, and 4 years was 82.6%, 68.1%, 63.3%, and 61.2%, respectively (Table 2). Few studies reported persistence at 5 and ≥10 years follow-up, 2 and 1, respectively. In all of these studies, the proportion of patients persistent was less than 60%. Three studies did not involve a specific year of follow-up and median was not computed. Table 2. Proportion of Older Statin Users Persistent at Various Follow-up Periods Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Note: IQR = Interquartile range. *The permissible gap varied from 30 days to 365 days across studies; †Other includes for instance reliance on patient reports; ‡The majority of studies did not stratify results according to statin indication and were excluded from this analysis. View Large Table 2. Proportion of Older Statin Users Persistent at Various Follow-up Periods Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Follow-up Period Number of Studies Sample Size % Persistent (IQR) 6 mo 13 539,188 82.6 (71.1–89.0) 1 y Overall 26 742,691 76.7 (71.4–87.4) Assessment method Permissible-gap* 17 702,335 76.0 (71.0–82.4)  Other† 9 40,356 87.7 (80.2–88.8) Statin indication‡  Primary prevention 5 266,114 76.0 (39.5–85.0)  Secondary prevention 13 164,896 82.6 (65.0–88.8) Participant Type  New Users 15 702,448 76.1 (71.1–82.8)  Mixed patients 11 40,243 83.0 (73.5–88.6) Region of Study  North America 8 196,926 65.0 (57.7–88.0)  Rest of the world 18 545,765 76.8 (74.1–85.6) Publication period  <2012 6 193,464 61.0 (41.3–73.9)  2012–2016 20 549227 80.2 (76.3–88.3) 2 y 12 453,830 68.1 (46.7–76.3) 3 y 6 196,716 63.3 (60.3–68.7) 4 y 4 171,005 61.2 (58.4–70.6) Note: IQR = Interquartile range. *The permissible gap varied from 30 days to 365 days across studies; †Other includes for instance reliance on patient reports; ‡The majority of studies did not stratify results according to statin indication and were excluded from this analysis. View Large Discussion We report suboptimal statin use among older patients, with 40% nonadherence at 1 year and more than one-third discontinuing treatment by 3 years. Although recent (2012–2016) publications reported slightly higher adherence and persistence than earlier ones, the trend is notable for a population at increased risk of cardiovascular events in whom more than 80% of CVD-related deaths occur (19). It appears that the first 180 days of follow-up is the most critical period when many patients become nonadherent or discontinue treatment. Nonetheless, we observed no clear evidence of plateau in the adherence or persistence rates during follow-up, suggesting the need to provide long-term treatment support. The high self-reported adherence observed in our study is consistent with previous research, that suggest that adherence measured via self-reporting scales or patient interviews tend to overestimate levels of medication use for multiple reasons that include recall and social desirability bias (20). Thus, while assessment of adherence via self-reports may be time saving, inexpensive and easy to apply in routine clinical practice (20), the value of using them to understand how well patients are following the statin dosing regimen is low. There is strong evidence supporting the long-term use of statins to achieve the desired effects both on cardiovascular morbidity and mortality. In the Scandinavian Simvastatin Survival Study (4S) trial, the biochemical effect of simvastatin was observed after a few weeks of treatment but the effects on coronary artery disease morbidity and mortality were only observed after approximately 24 months (21). However, as per our analysis, nearly one in four patients newly initiated on statin discontinued within the first 12 months of treatment. It is possible that patients stopped treatment as a result of poor drug effectiveness or due to adverse effects especially as ageing may result in altered pharmacokinetics and pharmacodynamics. Nonetheless, an analysis of the tolerability of 4,924 older people in 50 statin trials showed that only 2.1% of treatment discontinuation was due to adverse events (22). Similarly, in a retrospective cohort study with over 100,000 patients, more than half discontinued statin treatment but less than 4% of discontinuers cited adverse effects as the reason for discontinuation (23). The perceived benefits or susceptibility to disease as explained by the Health Belief Model has been shown to positively correlate with patients’ behavior (24). Since dyslipidaemia is an asymptomatic condition, patients may be unaware of their risk of adverse health outcomes. Statin users who have experienced an event (eg, myocardial infarction [MI], ie, secondary prevention) may consider themselves to be at higher risk, as well as exhibit greater belief in the severity of the diseases being prevented and may therefore be more committed to treatment (24). Kronish and colleagues (25) reported that among older patients who were nonadherent to statins, 38% became adherent following hospitalization for acute MI. Thus, perception of low risk may partly explain the poorer adherence and persistence observed among the primary prevention subjects. Nonetheless, considering that only 62% of the secondary prevention subjects in our study were adherent at 1 year, other factors beyond probabilistic information (ie, benefit or risk) may additionally contribute to determining patients’ behavior. Polypharmacy is prevalent among older populations and the risk of drug–drug interactions (DDIs) increases with the number of medications. For example, a patient on five to nine medications has about 50% probability of experiencing a DDI and this increases to 100% when the number of medications exceeds 20 (26). In addition, metabolism enzymes become less functional at advanced ages resulting in increased statin area under the curve (AUC) and higher likelihood of DDIs. Side effects are also common when more medications are used, with as much as 40% of hospitalizations in older patients attributed to adverse drug events (26). As a result, increasing polypharmacy has frequently been correlated with poor statin adherence (6). Older patients may also experience significant cognitive decline as well as face other barriers (eg, difficulty in swallowing) which may all affect adherence (10). In recent years also, there has been intensified debate around the use of statins in older adults (11). Analyses from several countries including the United Kingdom (27) and Denmark (28) indicate that the polarized views on the interpretation of the effects of statin treatment have had adverse impacts on patients’ adherence and continuation rates. In the very old (>75 years) particularly, guidelines suggest a lack of evidence to support the use of statins for primary prevention (29). This may partly explain the worst adherence observed among subjects aged more than 75 years who were taking statins for primary prevention. Ongoing trials such as STAREE (NCT02099123) and SITE (NCT02547883) will provide important evidence about the use of statin for primary prevention in the elderly. STAREE will examine whether treatment with atorvastatin 40mg/day versus placebo in healthy elderly individuals (≥70 years) will prolong overall survival or disability-free survival. SITE will evaluate the cost-effectiveness of stopping statins in people aged ≥75 years—the primary endpoints being overall mortality and incremental cost per quality-adjusted life-year (QALY) gained. The results of both STAREE and SITE are expected after 2020. The clinical implications of our findings are important. There is increasing evidence linking poor statin adherence and discontinuation to worse clinical outcomes (5). For example, patients who discontinue statin treatment following MI are about 3 times more likely to die than those continuing treatment (30). Compared with adherent patients, those nonadherent to statins and antihypertensive medications also have more than sevenfold increased risk of fatal stroke (31). Unsurprisingly, a meta-analysis attributed nine excess CVD deaths per 100,000 to nonadherence to cardiovascular medications including statins (5). The observed poor statin adherence and persistence may also be indicative of widespread suboptimal preventive medication use among the older population. Indeed, some studies suggest that nonadherence to one medication class may predict suboptimal use of another. Among 2,695 older Americans who initiated statin treatment following hospitalization for acute MI, low adherence to antihypertensive medications was associated with 62% increased risk of statin nonadherence (32). The economic impact of not using prescribed medications is significant, with studies frequently associating poor statin adherence and discontinuation with lower cost-effectiveness (33). In most developed countries, CVDs attract higher costs than any other disease class. In six European countries (France, Germany, Spain, Italy, Sweden, and the United Kingdom), the total financial impact of CVDs exceeded €102 billion in 2014 (34). The use of evidence-based pharmacological therapies such as statins remains the cornerstone of CVD prevention and treatment. Thus, poor adherence/persistence represents a missed opportunity to realize the full clinical potential of preventive therapies that are intended to reduce CVD burden. Consequently, in the United States alone, the yearly economic loss due to medication nonadherence has been estimated as $290 billion (35). It has been noted that if adherence to prescribed medications were to improve, it would have far greater impact on population health than improvement in specific medical treatments (36). Interventions targeted at patients and health professionals are necessary to address the poor statin adherence and persistence. Patient counseling/education, reinforcement and reminders have been observed to improve adherence (37). A Cochrane review found reinforcement and reminders increased adherence by about 24% (38). Strategies that seek to enhance patient–physician communication or those aimed at minimizing regimen complexity have also shown some degree of success (37). Other interventions that target the health delivery system such as through provision of extended care via ancillary health staff or those meant to eliminate health system cost-containment measures such as copayments have been considered necessary (6). Ito and colleagues for instance demonstrated that providing full medication reimbursement (for cardiovascular drugs including statins) to post-MI patients is cost-effective (39). However, in view of the complexity of the drivers of suboptimal adherence/persistence, the greatest improvements are likely to be achieved via the adoption of a mix of patient-tailored long-term interventions (6,36–38). This review has some limitations. First, studies do not report in detail actual reasons for nonadherence or discontinuation. It is possible that some discontinuation was due to adverse effects or initiated by clinicians. Second, adherence/persistence was based on indirect measurements and we cannot ascertain whether patients actually took the medication (12). Third, we restricted our study to English articles, and this is likely to limit the generalizability. Fourth, patients who discontinue statins may reinitiate treatment and this has not been incorporated in our analysis. Fifth, although some publications have offered guidance regarding the definitions and measurement of medication adherence and persistence (12), we noticed significant inconsistencies in the literature particularly with regards to the assessment of statin discontinuation. Persistence was therefore analyzed descriptively, although, meta-analysis would have provided a more robust outcome. Finally, we observed heterogeneity (I2 > 90 in many cases) which may be attributed to multiple factors including patient characteristics and data sources. Our study has important strengths worth mentioning. We applied a specific adherence cutoff ensuring consistency across studies. No geographic or time restrictions were imposed allowing the inclusion of a large body of evidence. Contact made with study authors ensured the inclusion of a large amount of unpublished data allowing us to provide for the first time a pooled estimate of self-reported statin adherence. To our knowledge, ours is the first systematic review to offer some understanding of the patterns of long-term (>1 year) adherence and persistence among statin users. Our pooled adherence estimates are robust and the absence of publication bias supports the reliability of our findings. Lastly, considering that >70% of the included studies were recently published, our reported adherence and persistence estimates may speak to current trends. Conclusions There is poor short and long-term adherence and persistence among older statin users. Strategies to improve adherence and reduce discontinuation are needed to close the gap if the intended cardiovascular benefits of statin treatment are to be realized. Supplementary Material Supplementary data is available at The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences online. Funding R.O. is sponsored by a Monash Graduate Scholarship and Monash International Postgraduate Research Scholarship for his doctoral studies. S.Z. is funded by a National Health and Medical Research Council Senior Research Fellowship. No other funding has been required to undertake this work. Conflict of Interest S.Z. reports past participation in advisory boards and/or receiving honoraria from: Amgen Australia; AstraZeneca/Bristol-Myers Squibb Australia; Janssen-Cilag; Merck, Sharp, and Dohme (Australia); Novartis Australia; Novo Nordisk; Sanofi; Servier Laboratories; Takeda Australia; and Monash University (undertaking contract work for AstraZeneca Pty Limited/Bristol-Myers Squibb Australia Pty Limited) for work unrelated to this study. Acknowledgments We are grateful to authors who generously provided us with unpublished data. We also thank Lorena Romero, a Senior Medical Librarian (Ian Potter Library, Alfred Hospital, Melbourne, Australia) who reviewed our initial search strategy and provided useful feedback. References 1. Naci H , Brugts JJ , Fleurence R , Tsoi B , Toor H , Ades AE . Comparative benefits of statins in the primary and secondary prevention of major coronary events and all-cause mortality: a network meta-analysis of placebo-controlled and active-comparator trials . Eur J Prev Cardiol . 2013 ; 20 : 641 – 657 . doi: 10.1177/2047487313480435 Google Scholar CrossRef Search ADS PubMed 2. Ioannidis JP . More than a billion people taking statins?: potential implications of the new cardiovascular guidelines . JAMA 2014 ; 311 : 463 – 464 . doi: 10.1001/jama.2013.284657 Google Scholar CrossRef Search ADS PubMed 3. Sabatine MS , Giugliano RP , Keech AC et al. Evolocumab and clinical outcomes in patients with cardiovascular disease . New EngJ Med . 2017 ; 376 : 1713 – 1722 .doi: 10.1056/NEJMoa1615664 Google Scholar CrossRef Search ADS 4. Lemstra M , Blackburn D , Crawley A , Fung R . Proportion and risk indicators of nonadherence to statin therapy: a meta-analysis . Can J Cardiol . 2012 ; 28 : 574 – 580 . doi: 10.1016/j.cjca.2012.05.007 Google Scholar CrossRef Search ADS PubMed 5. Chowdhury R , Khan H , Heydon E et al. Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences . Eur Heart J . 2013 ; 34 : 2940 – 2948 . doi: 10.1093/eurheartj/eht295 Google Scholar CrossRef Search ADS PubMed 6. Phan K , Gomez YH , Elbaz L , Daskalopoulou SS . Statin treatment non-adherence and discontinuation: clinical implications and potential solutions . Curr Pharm Des . 2014 ; 20 : 6314 – 6324 . doi:10.2174/1381612820666140620162629 Google Scholar CrossRef Search ADS PubMed 7. National Center for Health Statistics: National center for health statistics. Health, united states, 2010 : With Special Feature on Death and Dying . Hyattsville, MD: US Department of Health and Human Services ; 2011 . 8. Morgan TK , Williamson M , Pirotta M , Stewart K , Myers SP , Barnes J . A national census of medicines use: a 24-hour snapshot of Australians aged 50 years and older . Med J Aust . 2012 ; 196 : 50 – 53 . doi:10.5694/mja11.10698 Google Scholar CrossRef Search ADS PubMed 9. Pencina MJ , Navar-Boggan AM , D’Agostino RB , Sr. et al. Application of new cholesterol guidelines to a population-based sample . New Eng J Med . 2014 ; 370 : 1422 – 1431 . doi: 10.1056/NEJMoa1315665 Google Scholar CrossRef Search ADS PubMed 10. Yap AF , Thirumoorthy T , Kwan YH . Systematic review of the barriers affecting medication adherence in older adults . Geriatr Gerontol Int . 2016 ; 16 : 1093 – 1101 . doi: 10.1111/ggi.12616 Google Scholar CrossRef Search ADS PubMed 11. Stone NJ , Intwala S , Katz D . Statins in very elderly adults (debate) . J Am Geriatr Soc . 2014 ; 62 : 943 – 945 . doi: 10.1111/jgs.12788_1 Google Scholar CrossRef Search ADS PubMed 12. Cramer JA , Roy A , Burrell A et al. Medication compliance and persistence: terminology and definitions . Value Health . 2008 ; 11 : 44 – 47 . doi: 10.1111/j.1524-4733.2007.00213.x Google Scholar CrossRef Search ADS PubMed 13. Ofori-Asenso R , Zomer E , Curtis A et al. Patterns and predictors of adherence to statin therapy among older patients: protocol for a systematic review . JMIR Res Protoc . 2017 ; 6 : e39 . doi: 10.2196/resprot.7195 Google Scholar CrossRef Search ADS PubMed 14. National Institute of Health: Quality assessment tool for observational cohort and cross-sectional studies. NIH 2014 . https://www.nhlbi.nih.gov/health-pro/guidelines/in-develop/cardiovascular-risk-reduction/tools/cohort ( Accessed 13 November 2016 ). 15. Higgins J , Green S . Cochrane handbook for systematic reviews of interventions . Wiley Online Library , 2008 . http://au.wiley.com/WileyCDA/WileyTitle/productCd-0470699515.html# (Accessed 11 November 2016). 16. Egger M , Davey Smith G , Schneider M , Minder C . Bias in meta-analysis detected by a simple, graphical test . BMJ 1997 ; 315 : 629 – 634 . doi: 10.1136/bmj.315.7109.629 Google Scholar CrossRef Search ADS PubMed 17. World Health Organization, Harvard Medical School and Harvard Pilgrim Health: Medicines Use in Primary Care in Developing and Transitional Countries Fact Book Summarizing Results from Studies Reported Between 1990 and 2006 . Geneva : World Health Organization ; 2009 . http://www.who.int/medicines/publications/who_emp_2009.3/en/ ( Accessed 20 February 2017 ). 18. McDonald J : Handbook of Biological Statistics . 2nd ed . Baltimore, MD : Sparky House Publishing ; 2009 . 19. Yazdanyar A , Newman AB . The burden of cardiovascular disease in the elderly: morbidity, mortality, and costs . Clin Geriatr Med . 2009 ; 25 : 563 – 77, vii . doi: 10.1016/j.cger.2009.07.007 Google Scholar CrossRef Search ADS PubMed 20. Nunes V , Neilson J , O’Flynn N et al. Clinical Guidelines and Evidence Review for Medicines Adherence: Involving Patients in Decisions About Prescribed Medicines and Supporting Adherence . London : National Collaborating Centre for Primary Care and Royal College of General Practitioners ; 2009 . https://www.nice.org.uk/guidance/cg76/evidence/full-guideline-242062957 ( Accessed 05 February 2017 ). 21. Pedersen TR , Kjekshus J , Berg K et al. Scandinavian Simvastatin Survival Study G: randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the scandinavian simvastatin survival study (4s) . Atherosclerosis Supplements 2004 ; 5 : 81 – 87 . doi: 10.1016/j.atherosclerosissup.2004.08.027 Google Scholar CrossRef Search ADS PubMed 22. Hey-Hadavi JH , Kuntze E , Luo D , Silverman P , Pittman D , Lepetri B . Tolerability of atorvastatin in a population aged > or =65 years: a retrospective pooled analysis of results from fifty randomized clinical trials . Am J Geriatr Pharmacother . 2006 ; 4 : 112 – 122 . doi: 10.1016/j.amjopharm.2006.06.001 Google Scholar CrossRef Search ADS PubMed 23. Zhang H , Plutzky J , Skentzos S et al. Discontinuation of statins in routine care settings: a cohort study . Ann Intern Med . 2013 ; 158 : 526 – 534 . doi: 10.7326/0003-4819-158-7-201304020-00004 Google Scholar CrossRef Search ADS PubMed 24. DiMatteo MR , Haskard KB , Williams SL . Health beliefs, disease severity, and patient adherence: a meta-analysis . Med Care . 2007 ; 45 : 521 – 528 . doi: 10.1097/MLR.0b013e318032937e Google Scholar CrossRef Search ADS PubMed 25. Kronish IM , Ross JS , Zhao H , Muntner P . Impact of hospitalization for acute myocardial infarction on adherence to statins among older adults . Circ Cardiovasc Qual Outcomes . 2016 ; 9 : 364 – 371 . doi: 10.1161/CIRCOUTCOMES.115.002418 Google Scholar CrossRef Search ADS PubMed 26. Maher RL , Hanlon J , Hajjar ER . Clinical consequences of polypharmacy in elderly . Expert Opin Drug Saf . 2014 ; 13 : 57 – 65 . doi: 10.1517/14740338.2013.827660 Google Scholar CrossRef Search ADS PubMed 27. Matthews A , Herrett E , Gasparrini A et al. Impact of statin related media coverage on use of statins: interrupted time series analysis with UK primary care data . BMJ . 2016 ; 353 : i3283 . Google Scholar CrossRef Search ADS PubMed 28. Nielsen SF , Nordestgaard BG . Negative statin-related news stories decrease statin persistence and increase myocardial infarction and cardiovascular mortality: a nationwide prospective cohort study . Eur Heart J . 2016 ; 37 : 908 – 916 . doi: 10.1093/eurheartj/ehv641 Google Scholar CrossRef Search ADS PubMed 29. Stone NJ , Robinson JG , Lichtenstein AH et al. ; American College of Cardiology/American Heart Association Task Force on Practice Guidelines . 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines . J Am Coll Cardiol . 2014 ; 63 ( 25 Pt B ): 2889 – 2934 . doi: 10.1016/j.jacc.2013.11.002 Google Scholar CrossRef Search ADS PubMed 30. Ho PM , Spertus JA , Masoudi FA et al. Impact of medication therapy discontinuation on mortality after myocardial infarction . Arch Intern Med . 2006 ; 166 : 1842 – 1847 . doi: 10.1001/archinte.166.17.1842 Google Scholar CrossRef Search ADS PubMed 31. Herttua K , Martikainen P , Batty GD , Kivimäki M . Poor adherence to statin and antihypertensive therapies as risk factors for fatal stroke . J Am Coll Cardiol . 2016 ; 67 : 1507 – 1515 . doi: 10.1016/j.jacc.2016.01.044 Google Scholar CrossRef Search ADS PubMed 32. Muntner P , Yun H , Sharma P et al. Ability of low antihypertensive medication adherence to predict statin discontinuation and low statin adherence in patients initiating treatment after a coronary event . Am J Cardiol . 2014 ; 114 : 826 – 831 . doi: 10.1016/j.amjcard.2014.06.009 Google Scholar CrossRef Search ADS PubMed 33. Aarnio E , Korhonen MJ , Huupponen R , Martikainen J . Cost-effectiveness of statin treatment for primary prevention in conditions of real-world adherence–estimates from the Finnish prescription register . Atherosclerosis . 2015 ; 239 : 240 – 247 . doi: 10.1016/j.atherosclerosis.2014.12.059 Google Scholar CrossRef Search ADS PubMed 34. Centre for Economics and Business Research: The Economic Cost of Cardiovascular Disease from 2014–2020 in six European Economies, Cebr , 2014 . https://cebr.com/reports/the-rising-cost-of-cvd/ ( Accessed 02 February 2017 ). 35. NEHI Institute: Thinking Outside the Pillbox a System-Wide Approach to Improving Patient Medication Adherence for Chronic Disease . New England Healthcare Institute Web site , 2009 . http://www.nehi.net/writable/publication_files/file/pa_issue_brief_final.pdf ( Accessed 05 February 2017 ). 36. Haynes RB , Ackloo E , Sahota N , McDonald HP , Yao X . Interventions for enhancing medication adherence . Cochrane Database Syst Rev . 2008 : CD000011 . doi: 10.1002/14651858.CD000011.pub3 37. Maningat P , Gordon BR , Breslow JL . How do we improve patient compliance and adherence to long-term statin therapy ? Curr Atheroscler Rep . 2013 ; 15 : 291 . doi: 10.1007/s11883-012-0291-7 Google Scholar CrossRef Search ADS PubMed 38. Schedlbauer A , Davies P , Fahey T . Interventions to improve adherence to lipid lowering medication . Cochrane Database Syst Rev . 2010 : CD004371 . doi: 10.1002/14651858.CD004371.pub3 39. Ito K , Avorn J , Shrank WH et al. Long-term cost-effectiveness of providing full coverage for preventive medications after myocardial infarction . Circ Cardiovasc Qual Outcomes . 2015 ; 8 : 252 – 259 . doi: 10.1161/CIRCOUTCOMES.114.001330 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The Journals of Gerontology Series A: Biomedical Sciences and Medical SciencesOxford University Press

Published: Sep 2, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off