impact of medication adherence on work productivity in … · 2014-02-27 · n a reduction in work...

ABSTRACT

Objectives: To evaluate the impact of antihypertensive medication adherence on work productivity.

Study Design: Cross-sectional study.

Methods: Antihypertensive medication–treated respondents from the 2007 National Health and Wellness Survey (NHWS; n = 16,474) were included. Blood pressure measurements, medication adherence, and work productivity measures were obtained using subject self-reported data collected by the NHWS. Productivity and adherence were evaluated using the Work Productivity and Activity Impairment questionnaire and Morisky Medication Adherence Scale. Subjects were classifi ed as normotensive (systolic blood pressure [SBP] <120 mm Hg and diastolic blood pressure [DBP] <80 mm Hg), prehyper-tensive (SBP 120-139 mm Hg or DBP 80-89 mm Hg), stage 1 hypertensive (SBP 140-159 mm Hg or DBP 90-99 mm Hg), or stage 2 hypertensive (SBP >160 mm Hg or DBP >100 mm Hg). Multivari-ate linear regression was used to determine the relationship between antihypertensive medication adherence and work productivity loss, while controlling for important covariates.

Results: Among treated hypertensive subjects (n = 16,474), the mean age was 59.6 years, and 49% were female. Respondents employed full time (n = 3041) were younger (mean age = 51 years); 14%, 54%, 24%, and 8% were normotensive, prehypertensive, and stage 1 and 2 hypertensive, respectively. High adherence was repor-ted by 55% of employed respondents. Low adherence was associa-ted with more work productivity impairment (β = 2.12; P <.05). Stage 2 hypertension was associated with greater productivity im-pairment compared with other stages (β = –6.30 vs normotensives; β = –6.79 vs prehypertensives; β = –5.18 vs stage 1; all P <.05).

Conclusions: Low adherence to prescribed antihypertensive medi-cation regimens was associated with a reduction in work productivity. Programs to support antihypertensive medication adherence may present economic opportunities for employers by reducing work productivity impairment.

(Am J Pharm Benefi ts. 2012;4(4):e88-e96)

Origi

nal R

esea

rch

A n estimated 73.6 million adults 20 years and older

in the United States have hypertension.1 Individu-

als with uncontrolled blood pressure (BP) are at a

signifi cantly higher risk for cardiovascular-related morbidity,

including stroke, myocardial infarction, and kidney failure,

and subsequently have an increased mortality risk.2,3 Approx-

imately 65% of hypertensive persons in the United States are

treated with medication, and 57% of those who are antihyper-

tensive medication–treated have BP controlled4 to the recom-

mended target of <140/90 mm Hg for non-diabetic persons

or <130/80 mm Hg for those with diabetes.2 The prevalence

of hypertension increases steadily with age; from 13.4% and

6.2% of men and women aged 20 to 34 years, respectively,

to 53.7% of men and 55.8% of women aged 55 to 64 years.1

Uncontrolled hypertension, along with other cardiometa-

bolic risk factors, is associated with decreased work pro-

ductivity, poorer quality of life, and greater direct medical

costs.5,6 Previous research has linked better antihypertensive

medication adherence in persons with hypertension to a

greater probability of BP control,7-9 lower healthcare resource

use,10-13 and lower hospitalization rates.10 Poor medication

adherence has been cited as one of the primary causes of

hypertensive patients not achieving goal BP.2

Despite the importance of adherence to BP management,

there remain few published studies of the relationship be-

tween antihypertensive medication adherence and work

productivity among hypertensive employees. Our study

objective was to evaluate antihypertensive medication ad-

herence and work productivity impairment among treated

hypertensive adults who were employed full time. Second-

ary objectives were to examine the relationships between

both hypertensive disease classifi cation and comorbid con-

ditions with work productivity impairment.

METHODSThis cross-sectional study evaluated data from the 2007

National Health and Wellness Survey (NHWS). The NHWS

At a GlancePractical Implications e89

Author Information e94

Web Exclusive www.ajpblive.com

Impact of Medication Adherence on Work Productivity in Hypertension

Samuel Wagner, PhD, RPh; Helen Lau, MS; Feride Frech-Tamas, PhD, RPh; and Shaloo Gupta, MS

e88 TheAmericanJournalofPharmacyBenefits • July/August2012 www.ajpblive.com

P R A C T I C A L I M P L I C A T I O N S

This study included self-reported data on antihypertensive medica-tion–treated respondents from the 2007 National Health and Well-ness Survey (NHWS; n = 16,474) and a subset of subjects who were employed full time (n = 3041).

n A reduction in work productivity was reported by nonadherent sub-jects, primarily associated with productivity while at work.

n Stage 2 hypertensive respondents reported more work productivity impairment than other hypertensive subjects, and the number of comorbidities was associated with work productivity impairment.

n Since an association was found between nonadherence and poorer outcomes, programs to support antihypertensive medica-tion adherence present economic opportunities for employers by improving work productivity.

www.ajpblive.com Vol.4,No.4 • TheAmericanJournalofPharmacyBenefits e89

Hypertension Medication Adherence and Productivity

is a self-administered, Internet-based annual survey of

63,012 US adults 18 years and older which has been

conducted in the United States since 1998 by Consumer

Health Sciences.14 Survey participants provide informed

consent and are sampled to mirror generalized demo-

graphic characteristics (gender, age, and race/ethnicity)

of the US population. The survey sample is drawn from

an Internet panel maintained by Lightspeed Research

(Warren, New Jersey) and includes self-reported informa-

tion on participant demographic characteristics, medical

history, healthcare utilization, and healthcare attitudes,

behaviors, and outcomes. The protocol and informed

consent were reviewed and approved by Essex Institu-

tional Review Board, Inc, in Lebanon, New Jersey. NHWS

respondents were eligible if they had a self-reported diag-

nosis of hypertension and reported use of antihyperten-

sive prescription medication. While descriptive statistics

are reported for the entire eligible hypertensive popula-

tion with antihypertensive medication use (n = 16,474),

NHWS respondents were included in the final sample

used for productivity analyses if they also reported full-

time employment (n = 3041).

Medication adherence was estimated using the

Morisky Medication Adherence Scale (MMAS) as a proxy

for medication consumption.15-19 The MMAS has been

shown to be a reliable instrument (reliability α = 0.61),

and demonstrated both concurrent and predictive validity

with regard to BP control at both 2 and 5 years, respec-

tively.15 The MMAS consists of the following 4 questions

which are scored using a 0/1 response scale correspond-

ing to no/yes answers, respectively: “With regard to your

high blood pressure medications: 1) Do you ever for-

get to take your medicine? 2) Are you careless at times

about taking your medicine? 3) When you feel better do

you sometimes stop taking your medicine? and 4) Some-

times if you feel worse when you take your medicine,

do you stop taking it?” Respondent scores to the MMAS

are calculated as the sum of the 4 question responses;

the sum is used to categorize respondents as having high

adherence (MMAS = 0 “yes” responses) or low adher-

ence (MMAS = 1-4, or at least 1 “yes” response). Due to

a small percentage (<6%) of respondents reporting an

MMAS score >3, MMAS scores of 1 to 4 were collapsed

into 1 group. This categorization has been reported in

previous research.17,20-22

Work productivity and activity impairment were mea-

sured using the general health version of the Work Pro-

ductivity and Activity Impairment Questionnaire: General

Health (WPAI:GH).23 The WPAI:GH is a 6-item, quantita-

tive, self-reported evaluation of the level of absenteeism,

presenteeism, and daily activity impairment attributable

to general health during the prior 7 days. Activity im-

pairment was evaluated for all NHWS respondents, while

work productivity measures were assessed for respon-

dents who reported full-time employment. Among full-

time employed subjects, the following were evaluated:

absenteeism (the percent of work time missed due to

health reasons, or the number of hours missed during

the last 7 days as a percentage of the sum of the hours

missed plus the hours actually worked), presenteeism

(the percent of impairment while working due to health

reasons, or the degree that health affected productivity

while working as a percentage of the maximum possible

impairment), and overall work productivity loss (percent

of overall work impairment due to health, or absenteeism

plus presenteeism).24

Self-reported BP levels were also obtained as part of

the NHWS. Respondents were asked, “What was your last

blood pressure reading?” to obtain estimates of systolic BP

(SBP) and diastolic BP (DBP) measurements which were

used to classify participants according to the Seventh

Report of the Joint National Committee on Prevention,

Detection, Evaluation, and Treatment of High Blood Pres-

sure (JNC 7)–defined stages of hypertension, based on

highest reported SBP or DBP levels: normotensive (SBP

<120 mm Hg and DBP <80 mm Hg), prehypertensive

(SBP 120-139 mm Hg or DBP 80-89 mm Hg), stage 1 hy-

pertensive (SBP 140-159 mm Hg or DBP 90-99 mm Hg),

or stage 2 hypertensive (SBP >160 mm Hg or DBP >100

mm Hg).2 The presence of more than 140 other comor-

bid conditions, including diabetes, dyslipidemia, arthritis,

insomnia, and anxiety, was ascertained by respondent

self-report, and total number of comorbid conditions per


n Wagner•Lau•Frech-Tamas•Gupta

subject was calculated. Demographic information, includ-

ing age, gender, smoking history, race/ethnicity, income,

and marital status, was also obtained.

Statistical AnalysisUnivariate analyses were conducted for all study vari-

ables; mean, median, and standard deviations were com-

puted for all continuous variables, and frequency tables

were computed for all categorical variables. Bivariate

analyses of patient demographics, JNC 7 hypertensive

stages, antihypertensive treatment, work productivity

loss, and presenteeism were evaluated using χ2 tests for

categorical variables and analysis of variance for continu-

ous variables. Multivariate linear regression analyses were

performed to determine the relationship between antihy-

pertensive medication adherence and work productivity

loss, while controlling for important covariates including

age, gender, ethnicity, income, smoking, and subject co-

morbidities. Variables included in the final model were

selected based on results of bivariate analyses, as well

as clinical significance. SAS version 9.1 was used for all

study analyses.

RESULTSOverall, 19,874 of the 63,012 respondents to the

2007 NHWS reported a prior physician diagnosis of hy-

pertension (31.5%). About 83% (n = 16,474) of persons

with hypertension reported use of an antihypertensive

medication. The mean standard deviation age overall was

59.56 (11.99), and only 26% of our study population re-

ported being employed full time (Table 1). Subjects who

were classified as highly adherent to their antihyperten-

sive medication regimen were more likely to be female,

Caucasian, married, and non-smokers compared with

those reporting low adherence (P <.001, all comparisons).

Highly adherent subjects were older than low-adherence

subjects (mean age = 61.5 vs 55.8 years, P <.001); con-

sequently, subjects who were low adherence were also

more likely to report full-time employment (36% vs 22%,

P <.001). Antihypertensive medication adherence was re-

lated to BP, as high adherence individuals were also more

likely to be normotensive or prehypertensive, and less

likely to have stage 1 or 2 hypertension, compared with

their low-adherence counterparts.

Demographic and clinical characteristics of treated,

full-time employed respondents are summarized in the

Figure (n = 3041). Mean age of treated, full-time employed

hypertensive respondents was considerably younger than

the entire antihypertensive cohort using medication at

50.9 (9.91) years. More than half (55%) of these subjects

reported high adherence (eg, MMAS = 0) with their anti-

hypertensive medication regimen. More than half (56%)

self-reported having dyslipidemia as a comorbid condition,

and nearly 1 in 4 (24%) also reported having diabetes.

The relationship between antihypertensive medication

adherence, work productivity, and hypertensive stages is

Table 1. Demographic and Clinical Characteristics of Hypertensive Subjects Reporting Antihypertensive Medication Use (n = 16,474)

Total (n = 16,474)

Low Adherence (MMAS = 1-4) (n = 5580; 33.9%)

High Adherence (MMAS = 0) (n = 10,894; 66.1%)

P

Female, % 49.0% 46.5% 50.3% <.001

Age, mean (SD), years 59.6 (12.0) 55.8 (12.3) 61.5 (11.3) <.001

Caucasian, % 82.3% 75.5% 85.8% <.001

Married, % 59.8% 57.4% 61.1% <.001

Smokers, % 19.2% 21.3% 18.1% <.001

Total # of comorbidities, mean (SD) 8.1 (5.3) 8.2 (5.4) 8.0 (5.2) <.01

Hypertension stages, % <.001

Normotensive SBP <120 mm Hg and DBP <80 mm Hg

14.8% 11.3% 16.4%

Prehypertensive SBP 120-139 mm Hg or DBP 80-89 mm Hg

56.6% 54.1% 57.8%

Stage 1 hypertensive SBP 140-159 mm Hg or DBP 90-99 mm Hg

22.1% 25.4% 20.5%

Stage 2 hypertensive SBP >160 mm Hg or DBP >100 mm Hg

6.6% 9.2% 5.3%

Full-time employed, % 26.3% 35.8% 21.5% <.001

DBP indicates diastolic blood pressure; MMAS, Morisky Medication Adherence Scale; SBP, systolic blood pressure; SD, standard deviation.



summarized in Table 2. Compared with those with low

adherence (MMAS >0) to antihypertensive medications,

respondents who reported high adherence (MMAS = 0)

had significantly overall less work productivity loss (P

<.001) and were more likely to have no overall work

impairment. Among the full-time employed respondents

with high blood pressure treated with antihypertensive

medications, 41% of low-adherence respondents ex-

perienced no work impairment compared with 49% of

high-adherence respondents. Almost one-third (29%) of

individuals with low adherence experienced high levels

(>30%) of work impairment. Presenteeism was signifi-

cantly worse for subjects with low adherence versus high

adherence (P <.001). The stage of hypertension disease

was also significantly associated with self-reported adher-

ence (P <.001). Participants with more severe stages of

hypertension were less likely to be classified as highly

adherent.

Multivariate regression results are included in Table 3. After adjusting for demographics and comorbidities,

Figure. Demographic and Clinical Characteristics of National Health and Wellness Survey Hypertensive Subjects Reporting Antihypertensive Medication Use and Full-time Employment (n = 3041)

Male gender, Caucasian, married status, and self-reported dyslipidemia are the most prevalent characteristics of the full-time employed respondents.

100

90

80

70

60

50

40

30

20

10

0

High Adherence

%

55

Male

62

Caucasian

76

Married

63

CurrentlySmoking

20

Self-ReportedDiabetes

24

Self-ReportedDyslipidemia

56

Table 2. Relationship Between Medication Adherence, Work Productivity and Activity Impairment, and Hypertension Stages Among Treated, Full-time Employed Study Population (n = 3041)

Total (n = 3041)

Low Adherence (n = 1355)

High Adherence (n = 1686)

P

Overall work productivity loss, mean (SD) 21.39 (27.55) 23.75 (28.60) 19.50 (26.53) <.001

Overall work impairment .0002

None (0%) 45% 41% 49%

Low (1%-10%) 11% 11% 11%

Medium (11%-30%) 18% 19% 17%

High (>30%) 26% 29% 23%

Presenteeism, mean (SD) 18.42 (24.16) 20.51 (24.04) 16.74 (23.30) <.001

Hypertension stages <.001

Normal: SBP <120 mm Hg and DBP <80 mm Hg 14% 10% 16%

Prehypertension: SBP 120-139 mm Hg or DBP 80-89 mm Hg 54% 53% 55%

Stage 1: SBP 140-159 mm Hg or DBP 90-99 mm Hg 24% 27% 21%

Stage 2: SBP >160 mm Hg or DBP >100 mm Hg 8% 10% 7%

SBP indicates systolic blood pressure; SD, standard deviation.



low adherence was associated with more work produc-

tivity impairment (β = 2.12; P <.05), primarily related to

presenteeism. Stage 2 hypertension was associated with

greater productivity impairment compared with other hy-

pertensive stages (β = –6.30 vs normotensives; β = –6.79 vs

pre-hypertensives; β = –5.18 vs stage 1; all P <.05). A posi-

tive association between the number of comorbidities and

more productivity impairment was also noted (β = 1.86; P

<.05). Other comorbidities that significantly predicted work

productivity loss included depression, pain, high choles-

terol, and nasal allergies. Non-smokers were less likely to

experience work productivity loss compared with current

smokers (β = –2.82; P <.05). While patient antihypertensive

treatment was included as a covariate in the regression, it

was not a significant predictor of work productivity.

DISCUSSIONThis cross-sectional study found a relationship between

low self-reported antihypertensive medication adherence

and increased work productivity impairment, follow-

ing covariate adjustment, in a large, medication-treated,

employed hypertensive population. In our study, 31.5%

of subjects self-reported a hypertension diagnosis; this

figure is very close to national hypertension prevalence

estimates of 29.3%.4 We found that an advanced stage of

hypertension was an independent predictor of increased

work impairment while controlling for other clinical and

demographic factors. Stage 2 hypertensive participants

had a higher likelihood of work impairment compared

with subjects who were normotensive, prehypertensive,

or stage 1 hypertensive. Subjects who were highly ad-

herent to their antihypertensive medication regimen had

better BP control than those who were classified as low

adherence.

Among all of the hypertensive patients treated, those

reporting high adherence to their antihypertensive

medication regimen were significantly less likely to be

employed full time. In addition, consistent with previ-

ous research,25,26 we found that better antihypertensive

medication adherence was associated with increased age.

In our study, the mean age of highly adherent subjects

was 62 years (compared with 56 years for low-adherence

subjects) and consequently, a larger proportion of these

subjects would be expected to have retired. This is con-

sistent with US Census data, which indicate that 63% of

those aged 55 to 64 years, 24% aged 65 to 74 years, and

6% 75 years and older were in the labor force overall

during years 2006 to 2008.27

Table 3. Independent Effects of Hypertension Medication Adherence on Work Productivity Loss (n = 3041)a

Parameter Estimate Standard Error 95% CI Lower Limit 95% CI Upper Limit P

Intercept 35.571 5.488 24.810 46.332 <.001

Low adherence 2.120 0.923 0.310 3.929 .0217

High adherence (referent) 0.000

Age –0.278 0.049 –0.375 –0.181 <.001

Not currently smoking –2.815 1.134 –5.039 –0.591 .0131

Currently smoke (referent) 0.000

Normal: SBP <120 and DBP <80 –6.301 2.009 –10.240 –2.363 .0017

Prehypertension: SBP 120-139 or DBP 80-89 –6.787 1.701 –10.122 –3.453 <.001

Stage 1: SBP 140-159 or DBP 90-99 –5.182 1.812 –8.735 –1.628 .0043

Stage 2: SBP >160 or DBP >100 (referent) 0.000

No self-reported high cholesterol 2.203 0.973 0.295 4.111 .0236

Self-reported high cholesterol (referent) 0.000

No self-reported pain –4.278 1.048 –6.333 –2.223 <.001

Self-reported pain (referent) 0.000

No self-reported nasal allergies 3.563 1.080 1.446 5.680 .0010

Self-reported nasal allergies (referent) 0.000

No self-reported depression –4.164 1.286 –6.685 –1.643 .0012

Self-reported depression (referent) 0.000

Total comorbidities 1.858 0.192 1.482 2.234 <.001

CI indicates confidence interval; DBP, diastolic blood pressure; SBP, systolic blood pressure. aThe following covariates were included in the regression model, but were not significant predictors of productivity: gender, ethnicity, marital status, income, antihypertensive therapy regimen, diabetes, headache, arthritis, insomnia, sleep difficulties, heartburn, and anxiety.



Our findings are particularly important when consid-

ered in the context of the employer burden of hyper-

tension. Goetzel and colleagues studied the total cost of

health, absence, short-term disability, and productivity

losses for 10 chronic diseases using the MedStat Mar-

ketScan Health and Productivity Management database,

which contains information on 374,799 employees over

a 3-year time period.28 Hypertension ranked among the

10 chronic diseases with the highest total employer cost

burden in this study. The economic burden of illness for

hypertension, including inpatient and outpatient servi-

ces, prescription drugs, absenteeism, short-term disability

expenditures, and productivity losses, was estimated at

$392 per employee yearly. Annual productivity losses

due to hypertension were estimated at $247 per em-

ployee. Lamb and colleagues conducted a study of 8267

US employees from 47 employer locations to compare

productivity losses for allergic rhinitis with other condi-

tions, including hypertension. The authors found that ab-

senteeism plus presenteeism accounted for $105 yearly

per employee for hypertension.29 It is reasonable to infer

then that association between antihypertensive medica-

tion adherence and decreased work productivity found

in the current study may have cost implications from an

employer’s perspective.

Rizzo and colleagues, using data from the nationally

representative 1987 National Medical Care Expenditure

Survey, estimated the effects of prescription medication

on worker productivity for hypertension, along with dia-

betes, heart disease, and depression.30 The authors es-

timated that the net benefit to employers during 1987

amounted to $286 per hypertensive employee, $633 per

employee with heart disease, $822 per employee with

depression, and $1475 per diabetic employee. Average

compliance (63%) with antihypertensive medications

saved, on average, 3.5 days of work annually, compared

with 5.5 days lost for untreated patients. The authors con-

cluded that the observed benefits were due to reduced

absenteeism associated with prescription medication use

among employees with chronic illness.

Suboptimal adherence and persistence to prescribed

antihypertensive regimens has been documented by sev-

eral studies in “usual-care” settings,26,31,32 and long-term

persistence with antihypertensive therapies is poor.26,32

Antihypertensive medication adherence has been identi-

fied as an important contributor to BP goal attainment.7,33

Poor antihypertensive medication adherence is associ-

ated with higher healthcare resource use10-13 and higher

hospitalization rates.10 Previous research has documented

that medication adherence differs by therapeutic class,

with agents from the angiotensin-receptor blocker (ARB)

class generally associated with slightly higher adherence

and persistence.26,31,34 Although a few studies have found

a link between adverse effects from antihypertensive

medication and medication persistence,35,36 it is possible

that adverse effects from antihypertensive medications

could potentially impact employees to some degree at the

workplace. However, in the current study, antihyperten-

sive treatment was not associated with work productiv-

ity loss in the multivariate analysis. Among hypertensive

study subjects who had high adherence to their pre-

scribed antihypertensive medication regimen, 21% had

stage 1 hypertension and 7% had stage 2 hypertension.

Patients with hypertension usually require 2 or more dif-

ferent antihypertensive medications to attain goal BP.2

Addition of another class of antihypertensive medication

to the existing regimen or a dosage increase of the cur-

rent regimen would be warranted in these patients. The

failure of healthcare providers to intensify medication

regimens despite patients not achieving treatment goals

is often referred to as “clinical inertia,” which has been

well documented in usual care settings for hypertensive

patients.37-39 In addition to clinical inertia, other factors

in medication-compliant patients (such as dietary habits

and obesity) may play a role in failure to attain goal BP;

despite this, however, our findings suggest that a propor-

tion of our study population may require intensification

of their existing antihypertensive regimen, which is con-

sistent with the findings from these studies.

Sullivan and colleagues found that hypertension, in

conjunction with being overweight/obese and having

hyperlipidemia and/or diabetes, significantly impacted a

patient’s productivity.6 Controlling for other covariates,

overweight/obese patients with 2 of the 3 possible car-

diometabolic comorbidities (diabetes, dyslipidemia, and/

or hypertension) missed 179% more workdays and ad-

ditionally spent 147% more days in bed compared with

those without any cardiometabolic comorbidities. Fur-

thermore, the authors estimate that lost workdays and

bed days combined account for $17.3 billion annually

in the United States attributable to cardiometabolic risk

factor clusters and associated lost work productivity. Bur-

ton and colleagues studied 5512 employees, and similarly

found that as the number of metabolic risk factors in-

creased, the incidence of short-term disability increased,

as well as the increase in days missed due to illness.40

While to our knowledge, this is the first study to evalu-

ate the relationship between antihypertensive medication

adherence and work productivity limitations since that

of Rizzo and colleagues in 1996, this link has since been



demonstrated for other chronic diseases, including asth-

ma41 and depression.42 However, due to the recognized

symptomatic nature of these diseases, findings observed

for these chronic illnesses may not be generalizable to

hypertension, which is recognized as an asymptomatic

disease for most patients.

While our study provides a valuable contribution to the

antihypertensive medication adherence literature, several

limitations are important to consider when interpreting

our study’s findings. Our study was performed on a sam-

ple of subjects with hypertension who reported full-time

employment, and thus is representative of only those em-

ployed full time; there may be subject characteristics that

are associated with the likelihood of employment, medi-

cation adherence, and/or low productivity that resulted in

population selection bias and/or confounding. However,

the multivariate regression used for the productivity anal-

ysis should have reduced the effects of other factors for

those covariates that were available for analysis. As our

study sample was identified via an Internet-based survey,

our population may not be representative of the general

US population; due to the method of survey administra-

tion and the full-time employment status of our sample

it is plausible that our study sample represented persons

with higher income, education, and socioeconomic status

than the US population as a whole. However, our study’s

estimate of the prevalence of hypertension is consistent

with published estimates for a similar time period.4 Nev-

ertheless, for these reasons, our study’s findings may

not be generalizable to all hypertensive persons across

the United States. Some important information was not

available for inclusion as covariates, including duration

of hypertension, duration of medication use, medication

acquisition cost, insurance status, subject body mass in-

dex, and number and specific classes of antihypertensive

medications prescribed. In addition, while antihyperten-

sive therapy adherence was assessed, the reasons for an-

tihypertensive medication nonadherence are unknown.

Our study utilized self-reported data for all measures, in-

cluding BP measurements, productivity, and self-reported

antihypertensive agent use. No objective measurements

of BP and/or data to validate patient self-reported medi-

cation acquisition and consumption (such as prescription

claims information) were available. While prescription

claims data are usually considered the most accurate data

source to enable assessment of medication adherence, the

validated Morisky scale has been used across therapeutic

areas as a proxy to estimate medication adherence,15-19 al-

though admittedly, the MMAS may be somewhat less sen-

sitive and is subject to patient self-report bias. Previous

research has generally supported the validity of subject

self-report of blood pressure and/or hypertensive status

in various settings, with self-report correctly identifying

the majority of actual hypertensive persons, with higher

specificity than sensitivity generally reported.43-48 Previous

research has also supported the use of subject self report

of absenteeism based on validation using administrative

data.49 However, the WPAI is not condition-specific, and

productivity estimates may also reflect the impact of oth-

er comorbid conditions (although these were included

as covariates in multivariate analyses) as well as other

external unmeasured factors. Finally, it is possible that

our study’s findings as related to antihypertensive medi-

cation adherence may also reflect adherence to other

classes of medications, as patients might be expected to

have similar adherence results for medications for other

comorbid conditions; this may, in part, contribute to the

relationship between productivity and antihypertensive

medication compliance.

CONCLUSIONSA significant reduction in work productivity was re-

ported by participants who were classified as low ad-

herence with regard to their antihypertensive treatment,

primarily associated with productivity while at work (eg,

presenteeism). Stage 2 hypertensive respondents repor-

ted significantly more work productivity impairment than

respondents with less severe stages of hypertension, and

the number of comorbidities was also significantly as-

sociated with work productivity impairment. Our find-

ings suggest an association between low adherence to

antihypertensive treatment and poorer outcomes. Initia-

tives targeting improved adherence to medications and

improved BP control among patients with hypertension

may present economic opportunities for employers by

impacting work productivity.

Acknowledgment

Jenifer Wogen, MS, MedMentis Consulting, LLC, provided medical writing and editorial services in support of this manuscript, and received financial compensation from Novartis Pharmaceuticals Corporation.

Author Affiliations: From Novartis Pharmaceuticals Corporation (HL, FF-T), East Hanover, NJ; Kantar Health (SW, SG), Princeton, NJ (formerly known as Consumer Health Sciences).

Funding Source: Novartis Pharmaceuticals Corporation.

Author Disclosures: Dr Frech-Tamas and Ms Lau report employ-ment with Novartis Pharmaceuticals Corporation, the funder of the study, and stock ownership in the company. The other authors (SW, SG) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (SW, HL, FF-T, SG); acquisition of data (SW); analysis and interpretation of data (SW, HL, FF-T, SG); drafting of the manuscript (HL, FF-T); critical revision of the manuscript for important intellectual content (SW, HL, FF-T, SG);



statistical analysis (SW, SG); provision of study materials or patients (FF-T); obtaining funding (HL, FF-T); administrative, technical, or logistic support (FF-T); and supervision (FF-T).

Address correspondence to: Shaloo Gupta, MS, Health Economics and Outcomes Research, Kantar Health, 1 Independence Way, Ste 220, Princeton, NJ 08540. E-mail: [email protected].

REFERENCES1. Lloyd-Jones D, Adams R, Carnethon M, et al. Heart disease and stroke statis-tics 2009 update: a report from the American Heart Association Statistics Com-mittee and Stroke Statistics Subcommittee. Circulation. 2009;119(e):e21-e181.

2. Chobanian A, Bakris G, Black H. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA. 2003;289:2560-2572.

3. US Renal Data System, USRDS 2008 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. http://www .usrds.org/atlas08.aspx. Published 2008. Accessed July 17, 2009.

4. Ong KL, Cheung BM, Man YB, Lau CP, Lam KS. Prevalence, awareness, treatment, and control of hypertension among United States adults 1999-2004. Hypertension. 2007;49(1):69-75.

5. Kannan H, Thompson S, Bolge SC. Economic and humanistic outcomes associ-ated with comorbid type-2 diabetes, high cholesterol, and hypertension among individuals who are overweight or obese. J Occup Environ Med. 2008;50(5): 542-549.

6. Sullivan PW, Ghushchyan V, Wyatt HR, Wu EQ, Hill JO. Productivity costs associ-ated with cardiometabolic risk factor clusters in the United States. Value Health. 2007;10(6):443-450.

7. Bramley TJ, Gerbino PR, Nightengale BS, Frech-Tamas F. Relationship of blood pressure control to adherence with antihypertensive monotherapy in 13 managed care organizations. J Manag Care Pharm. 2006;12(3):239-245.

8. Elzubier AG, Husain AA, Suleiman IA, Hamid ZA. Drug compliance among hypertensive patients in Kassala, eastern Sudan. East Mediterr Health J. 2000; 6(1):100-105.

9. DiMatteo MR, Giordani PJ, Lepper HS, Croghan TW. Patient adherence and medical treatment outcomes: a meta-analysis. Med Care. 2002;40(9):794-811.

10. Sokol MC, McGuigan KA, Verbrugge RR, Epstein RS. Impact of medication adherence on hospitalization risk and healthcare cost. Med Care. 2005;43(6): 521-530.

11. McCombs JS, Nichol MB, Newman CM, Sclar DA. The costs of interrupting antihypertensive drug therapy in a Medicaid population. Med Care. 1994;32(3): 214-226.

12. Rizzo JA, Simons WR. Variations in compliance among hypertensive patients by drug class: implications for health care costs. Clin Ther. 1997(6);19: 1446-1457.

13. Caro JJ, Speckman JL. Existing treatment strategies: does noncompliance make a difference? J Hypertens. 1998;16(7):S31-S34.

14. Consumer Health Sciences. National Health and Wellness Survey. http://www .chsinternational.com/nhws.html. Published 2008. Accessed June 16, 2009.

15. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care. 1986;24(1):67-74.

16. Shalansky SJ, Levy AR, Ignaszewski AP. Self-reported Morisky score for identi-fying nonadherence with cardiovascular medications. Annals of Pharmacotherapy. 2004;38(9):1363-1368.

17. Kim E, Gupta S, Bolge S, Chen CC, Whitehead R, Bates JA. Adherence and outcomes associated with copayment burden in schizophrenia: a cross-sectional survey. J Med Econ. 2010;13(2):185-192.

18. Bates JA, Whitehead R, Bolge SC, Kim E. Correlates of medication adherence among patients with bipolar disorder: results of the bipolar evaluation of satisfac-tion and tolerability (BEST) study: a nationwide cross-sectional survey. Prim Care Companion J Clin Psychiatry. 2010;12(5).pii:PCC.09m00883.

19. Berni A, Ciani E, Cecioni I, Poggesi L, Abbate R, Boddi M. Adherence to antihy-pertensive therapy affects Ambulatory Arterial Stiffness Index. Eur J Intern Med. 2011;22(1):93-98.

20. George J, Munro K, McCaig DJ, Stewart DC. Prescription medications: beliefs, experiences, behavior, and adherence of sheltered housing residents. Ann Phar-macother. 2006;40(12):2123-2129.

21. Berry SD, Quach L, Procter-Gray E, et al. Poor adherence to medications may be associated with falls. J Gerontol A Biol Sci Med Sci. 2010;65(5):553-558.

22. Ivanova JI, Birnbaum HG, Hsieh M, et al. Adherence to inhaled corticosteroid use and local adverse events in persistent asthma. Am J Manag Care. 2008; 14(12):801-809.

23. Reilly MC, Zbrozek AS, Dukes EM. The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics. 1993;4(5): 353-365.

24. Reilly MA. Reilly Associates: WPAI Scoring. http://www.reillyassociates.net/WPAI_Scoring.html. Published 2002. Accessed July 21, 2009.

25. Pittman DG, Tao Z, Chen W, Stettin GD. Antihypertensive medication adher-ence and subsequent healthcare utilization and costs. Am J Manag Care. 2010; 16(8):568-576.

26. Conlin PR, Gerth WC, Fox J, Roehm JB, Boccuzzi SJ. Four-year persistence patterns among patients initiating therapy with the angiotensin II receptor antago-nist losartan versus other antihypertensive drug classes. Clin Ther. 2001;23(12): 1999-2010.

27. US Census Bureau. S2301. Employment Status: 2006-2008 American Com-munity Survey 3-Year Estimates. http://factfinder2.census.gov/faces/nav/jsf/ pages/index.xhtml. Published 2008. Accessed October 20, 2010.

28. Goetzel RZ, Long SR, Ozminkowski RJ, Hawkins K, Wang S, Lynch W. Health, absence, disability, and presenteeism cost estimates of certain physical and mental health conditions affecting U.S. employers. J Occup Environ Med. 2004;46(4):398-412.

29. Lamb CE, Ratner PH, Johnson CE, et al. Economic impact of workplace pro-ductivity losses due to allergic rhinitis compared with select medical conditions in the United States from an employer perspective. Curr Med Res Opin. 2006;22(6): 1203-1210.

30. Rizzo JA, Abbott TA 3rd, Pashko S. Labour productivity effects of prescribed medicines for chronically ill workers. Health Economics. 1996;5(3):249-226.

31. Patel BV, Remigio-Baker RA, Mehta D, Thiebaud P, Frech-Tamas F, Preblick R. Effects of initial antihypertensive drug class on patient persistence and compli-ance in a usual-care setting in the United States. J Clin Hypertens. 2007;9(9): 692-700.

32. Van Wijk BLV, Klungel OH, Heerdink ER, de Boer Ad. Rate and determinants of 10-year persistence with antihypertensive drugs. J Hypertens. 2005;23(11): 2101-2107.

33. Burnier M. Medication adherence and persistence as the cornerstone of ef-fective antihypertensive therapy. Am J Hypertens. 2006;19(11):1190-1196.

34. Wogen J, Kreilick CA, Livornese RC, Yokoyama K, Frech F. Patient adherence with amlodipine, lisinopril, or valsartan therapy in a usual-care setting. J Manag Care Pharm. 2003;9(5):424-429.

35. Burnier M, Hess B, Greminger P, Waeber B. Determinants of persistence in hypertensive patients treated with irbesartan: results of a postmarketing survey. BMC Cardiovasc Disord. 2005;5(1):13.

36. Gregoire JP, Moisan J, Guibert R, et al. Determinants of discontinuation of new courses of antihypertensive medications. J Clin Epidemiol. 2002;55(7): 728-735.

37. Andrade SE, Gurwitz JH, Field TS, et al. Hypertension management: the care gap between clinical guidelines and clinical practice. Am J Manag Care. 2004; 10(7, pt 2):481-486.

38. Oliveria SA, Lapuerta P, McCarthy BD, L’Italien GJ, Berlowitz DR, Asch SM. Physician-related barriers to the effective management of uncontrolled hyperten-sion. Arch Intern Med. 2002;162(4):413-420.

39. Heisler M, Hogan MM, Hofer TP, et al. When more is not better: treatment inten-sification among hypertensive patients with poor medication adherence. Circulation. 2008;117(22):2884-2892.

40. Burton WN, Chen CY, Schultz AB, Edington DW. The prevalence of metabolic syndrome in an employed population and the impact on health and productivity. J Occup Environ Med. 2008;50(10):1139-1148.

41. Joshi AV, Madhavan SS, Ambegaonkar A, Smith M, Scott VG, Dedhia H. Asso-ciation of medication adherence with workplace productivity and health-related quality of life in patients with asthma. J Asthma. 2006;43(7):521-526.



42. Burton WN, Chen CY, Conti DJ, Schultz AB, Edington DW. The association of antidepressant medication adherence with employee disability absences. Am J Manag Care. 2007;13(2):105-112.

43. Oksanen T, Kivimaki M, Pentii J, Virtanen M, Klaukka T, Vahtera J. Self-report as an indicator of incident disease. Ann Epidemiol. 2010;20(7):547-554.

44. Huerta JM, Tormo MJ, Egea-Caparrós JM, Ortolá-Devesa JB, Navarro C. Accuracy of self-reported diabetes, hypertension and hyperlipidemia in the adult Spanish population: DINO study findings [article in English, Spanish]. Rev Esp Cardiol. 2009;62(2):143-152.

45. Robinson JR, Young TK, Roos LL, Gelskey DE. Estimating the burden of dis-ease: comparing administrative data and self-reports. Med Care. 1997;35(9): 932-947.

46. Tormo MJ, Navarro C, Chirlaque MD, Barber X. Validation of self diagnosis of high blood pressure in a sample of the Spanish EPIC cohort: overall agreement and predictive values: EPIC Group of Spain. J Epidemiol Community Health. 2000; 54(3):221-226.47. Alonso A, Beunza JJ, Delgado-Rodríguez M, Martínez-González MA. Validation of self reported diagnosis of hypertension in a cohort of university graduates in Spain. BMC Public Health. 2005;12(5):94.48. Ahluwalia IB, Tessaro I, Rye S, Parker L. Self-reported and clinical measure-ment of three chronic disease risks among low-income women in West Virginia. J Womens Health. 2009;18(11):1857-1862.49. Short ME, Goetzel RZ, Pei X, et al. How accurate are self-reports? analysis of self-reported health care utilization and absence when compared with administra-tive data. J Occup Environ Med. 2009;51(7):786-796.

impact of medication adherence on work productivity in … · 2014-02-27 · n a reduction in work...

Documents