compliance antihipertensi

8/6/2019 Compliance Antihipertensi

1/15

Mental Health, Binge Drinking, andAntihypertension Medication AdherenceJim E. Banta, PhD, MPH; KellyB. Haskard, PhD; Mark G. Haviland, PhD;Summer L. Williams, MA; Leonard S. Werner, MD; Donald L. Anderson, MD;M. Robin DiMatteo, PhD

Objectives: To evaluate the rela-t i ons h ip bet w een s e lf -report edmental health and binge drinking,a s w e l l a s h e a l t h s t a t u s ,sociodemographic, social support,eco no m ic resource, and health careaccess indicators to antihyperten-s i o n m e d i c a t i o n a d h e r e n c e .Method: Analysis of 2003 Califor-nia Health Interview Survey data.Results: Having poor m enta l hea lthdays predicted medication nonad-herence, whereas binge drinkingdid not. Nonadherence predictorsincluded you nger age. L atino, non-

US citizen, uninsured, less eduction, and no regular medical carAdherence predictors were oldage, African Am erican, hav ing prscription insurance, a college dgree, poor health, comorbid diabte s or heart disease, and overw eigor obe se. Conclusion: Better metal hea lth may improve m edica tiadherence among hyp ertensive idividuals.Key words: hypertension, medcat ion adherence , mental heal tbinge drinkingAm J Health Behav. 2009;33(2): 158-1

P revalence estimates of high bloodpressure or systemic hypertensionvary but are generally in the rangeJim E. Banta, Assistant Professor, Departmentof Health P olicy and Manag ement, School of PublicHealth; Mark G. Haviland, Professor of Psychiatry,Department of Psychiatry, L oma Linda UniversitySchool of Medicine; Leonard S. Werner, Professor,Department of Internal Medicine, Loma Linda Uni-versity School of Medicine; Donald L. Anderson,Associate Professor, Department of Psychiatry,Loma Linda University School of Medicine, Lom aLinda, CA. Kelly B. Haskard, Assistant Professor,Department of Psychology, Texas State University,San Marcos, TX. Summer L. Williams, Doctoral Can-didate; M . Robin DiMatteo, Distinguished Profes-

of 20-30% in ad ult A mericans, ' '^ an d 2004, hiperten sin w as the cause of defor more than 50,000.' Approximately 2of US adults do not recognize that thhave hypertension; of those who habeen diagnosed, 59% receive treatmeand 34% are both receiving treatmand controlling their blood pressure acceptable levels (untreated diastolic prsure


2/15

Banta et al

Patient AdherencePat ient adherence i s " the extent ton, following a d iet, or execu ting lifestyle

25% of patients are nonadherent,

Adherence typically requires that pa-

rces. ' Predictably, nonad herence rates

0% in nontrial patients. '" In treated hy-In a Canadian study of patientsas 23%.'^ A study involving multiple ad-

self-rt) showed tha t 15-25% of patien ts in

mens may be particularly challenging forpatients with poor mental health. Meta-analyses have shown that the r isk ofnonadherence to medical t rea tment i s27% higher when pat ients are depressedthan when they are not.'"* Moreover, somecardiovascular conditions are associatedwi th an increased r i sk of depress ion,which can pose a substantial threat tomedication and lifestyle regimen adher-ence by decreasing pat ient sat isfact ionwith medical care and altering thoughtprocesses '^ and increasing hop elessne ssand reducing social support. '^ A study ofnear ly 30 ,000 veterans wi th hyper ten-sion demonstrated that depression wasassociated with nonadherence to a medi-cation regimen.'^ Another study of Afri-can American men wi th hyper tens ionfound a significant relationship betweenmore serious depression and medicationnonadherence. '^ One must interpret thesestudies cautiously, for there is not neces-sarily agreement among different adher-ence measures; for example, self-reportedantihypertension medicat ion use in pa-tients with depression has been shownnot to agree with their prescription reflllrates (self-reported adherence is lowerthan the rates at which patients refillprescriptions). '^ Associations among de-pression, substance use, and nonadher-ence to medicat ions are impor tant toexam ine because the r e l a t ionsh ip be -tween menta l d isorders and nonadher-ence may be complicated further by prob-lematic heal th-r isk behaviors . '^

Nonadherence in Hypertension andBinge DrinkingExcessive alcohol use, including bingedrinking, is another risk factor that pre-dicts nonadherence to ant ihypertensivemedication (as well as general cairdiovas-cular harm).^'^' This nonadherence riskfactor has not received the same level ofresearch at tent ion as has poor mentalhealth. Binge drinking is defined as 5drinks or more at one time.^^ Excessivedrinking may cause patients to forget totake their medication or take it incor-rectly. Alcohol use also has been associ-ated with nonadherence to an antihyper-


3/15

Antihypertension Medicat ion Adherence

was a negative relationship between ad-here nce and alcohol consumption.^^ AFinnish population study of hypertensiveindividuals revealed a positive relation-ship between cilcohol use and blood pres-sure level, and in males particularly, al-cohol use was associated with lower medi-cation adherence rates.^'^ Other researchhas shown that urban African Americanmen who abuse alcohol and take illegaldrugs are less likely to be taking theiran t ihyper tens ion medica t ion and morelikely to have uncontrolled blood pres-sure.^^ Not all research, however, hasdemonstrated that alcohol use decreasesm edic ation adherence.^^ Clearly, the rela-tionship between alcohol use, and par-t icular ly heavy/excessive use, and ad-herence in the context of hypertensionrequires additional study.

Other Predictors of Nonadherence inHypertens ionOther variables found to be related tonon adh eren ce in hype r tens ion inc ludesmoking, '^ nonmarried status, '^ and in-surance changes , inc luding co-pay in-creases.^^'^^ Across medical conditions,prescription nonadherence has increasedslightly over time and is relatively loweramong Medicaid and Medicare beneficia-ries and th ose with low incom es.^' Age isan oth er pred ic tor of no nad her en ce toantihypertension regimens, with the ex-tremes (younger and older) associatedwith nonadherence.^ Obesity also may bea nonadherence predictor in hyperten-sion, especially with respect to lifestylemodifications.^' Minority ethnic groupshave a higher prevalence of hyperten-sion, and being African American or non-white has been associated with nonad-he rence . "'

Barriers to adherence that are particu-lsirly problematic for vulnerable popula-tions include poor physician-patient com-m unication, lack of un der stan din g of healthinformation, lack of social support, fewerresources, and greater disease severity.^^'^^In terms of factors within an individual,limits in health literacy or health/medi-cation knowledge have been shown to beassociated with hypertension nonadher-

factorcosts associated with drug benecapsalso is associated with reduced uof medications.^^ Reg imen -related factosuch as medication side effects or lack improvement in day-to-day well-being, tomay be other factors infiuencing adhence in hypertension.^^ It is important explore and understand these risk factoand barriers because outcomes of nonaherence in hj^ertension and other contions include poor disease managemeincreased risk of morbidity and mortaliand higher costs to the health care stem.28

Objectives of the Present StudyOur main objective was to determthe re la t ionship be tween se l f - repor tmenta l hea l th , b inge dr inking , andm easu re of antihyp ertension med icatiuse (adherence) in a large and ethnicadiverse sample, Cal i fornia adul ts wh3rpertension. To better un de rsta nd tsample and these relat ionships, we dtermined the percentage of Californiawith hypertension and compared thowith and without the disease on heastatus, sociodemographic, social suppoeconomic resource, and health care acess indicators. Our secondary objectwas to compare those adhering and nadhering to antihypertension medication these same factors.METHODDataData for this study were from the ad2003 California Health Interview Surv

(CHIS; www.chis.ucla.edu). CHIS is largest state health survey in the UniStates and has been conducted biannally since 20 01 . This comprehensive svey covers a number of topics, includaccess to health care, health insurancoverage, heal th behaviors , and heastatus, with the average adult intervilasting 33 minutes,^^ altho ug h proxy terviews were allowed for frail and persons over the age of 60. CHIS ispopulation-based, random-digit-dial tephone survey of California householinterviews were conducted by staff fromprivate f i rm special iz ing in s tat is t i


4/15

Banta et al

which inc ludes many e thnic mi-

As with most survey data, the publicly

t r i b u t i o n s of a g e , g e n d e r , r a c e /

h. By usi ng the 80 p oststratificationm e tha t the 42,04 4 completed re-For CHIS 2003 , the overall responsefor. ad ul ts wa s 33.5%, based on asuccess rate in initial screening

most current populat ion characteris t icsas maintained by the California Depart-ment of Finance. The content of the sur-vey results also have been compared toother large population surveys, such asthe National Health Interview Survey,the Medical Expenditure Panel survey,and the BRFSS. As a result of these com-parisons, i t general ly is bel ieved thatCHIS has done an adequate job of mea-suring what it intended to measure.^'MeasuresRespondents who had been told by aphysician that they had high blood pres-

sure ( ie , answering "yes" to a yes/noquestion) were included in the hyperten-sion grou p. For individua ls in the hyp er-tension group, those who reported thatthey were currently taking medication tocontrol their high blood pressure wereincluded in the adherent group (also ayes/no question). Although this item failsto capture the multifaceted nature of ad-herence (and was not designed by theauthors because th is was a secondarydata analysis), the term adherence is usedhere because of the theoretical and con-ceptual underpinning of the research lit-erature, and it is used with awareness ofthose caveats. For the purposes of thisstudy, we assumed that al l respondentsanswering "yes" to the hypertension ques-tion should have been taking medication.Although this may be a reasonable as-sumption based on practice guidelines,^the inferences that one can make aboutphysician adherence to pract ice guide-lines a re limited,' ' an d it is poss ible th atsom e responden t s w ere a t t em pt ing tocontrol their hypertension through dietand exercise.Mental health was defined by responsesto the question "number of days mentalhealth not good in the last 30 days," avariable that ranged from 0 to 30. Afternoting this variable 's nonnormal distr i -butio n, it w as recoded into 2 catego ries: 0poor me ntal health day s and 1 to 30 daysof poor mental heal th. Binge drinking

(yes/no) was defined in CHIS as drinking5 or more drinks at one time during thelast 30 days. (There are several other


5/15

Antihypertension Medication Adherence

Estimated Population

Binge Drinking During Last MonthNo Poor Mental Health Days Last 30 Daysi-30 Poor Mentai Heaith DaysDoctor Told Had CancerDoctor Told H ad Heart DiseaseDoctor Told Had DiabetesDoctor Toid H ad AsthmaFair / Poor Health StatusGood HealthVery Good / Excellent HealthBMI Underweight / NormaiO v erw eig htObeseMaie18-44 Years of Age45-64 Years of Age65-75 Years of Age76 + Years of AgeWhiteLatinoAfrican AmericanAsian/Pacific IslanderA I A N / OtherLess than 12 Years of EducationHigh Scbooi GraduateSome CoiiegeCoiiege GraduateMarr iedSociai Support, Mean (Std Error)Never Smoked RegulariyCurrent SmokerQuit SmokingNo'Usuai Source of CareHousebold Income, Mean (Std Error)WorkingFuii-Time,21+HoursAVeekWorking Part-time, 0-20 HoursAVeekNot Working / Missing DataEmpioyment-Based InsuranceMedicaidUninsuredOther, Including Medieare

Table 1Characterist icsNo Hypertension (%)est. n=19,585,48530,577 surveys

16.358.042.0

6.23.73.45.9

15.527.457.148.734.6i 6. 749.364.427.34. 8

3.550.128.15.412.43 .918.624.425.031.954.0

15.37 (0.02)62.117.020.818.7

$62,353 ($321)60.86. 832.554.210.318.616.9

of California AdultsHypertension (%)est. n=6,011,57711,467 surveys

I I . 160.239.815.217.417.08.9

36.731.132.230.537.032.548.023.840.319.316.656.519.19.111.44. 021.426.025.926.757.2

15.25(0 .05)51.914.533.6

7.6$52,635 ($626)

41.74.753.738.110.39. 742.0

in 2003P value


6/15

Banta et al

by (1) ye s/n o respon ses to

very good/ excellent he alth) ;

Sociodemographic measures includedtion. Social sup port was represen ted

working statu s, insuran ce, and

Analytic PlanStatistical analyses included bivariatel t s w h o w e r e t a k i n g

F values , was used to examine

tivariable regression were poor mentalheal th and b inge dr inking. Covar ia teswere the health status, sociodemographic,social support, economic resource, andhealth care access variables describedabove. (Models with interactions of agewith mental heal th and binge drinkingalso were tested.) Finally, a sensitivityanadysis based on a "standard patient"was run to better describe the effect ofpoor mental health on medication nonad-herence, holding all other variables con-stant. Statistical analyses were conductedusing Stata/SE 9.2 for Windows software(StataCorp LP, College Station, TX).

RESULTSTable 1 show s th at slightly more th an 6million adults had hypertension, which is23.5% (95% C.I. 23.0 to 24.0%) of all 25.6million California ad ul ts. Com pared tothose without hj^Dertension, those withhypertension were less likely to reportbinge drinking (11.1 vs 16.3%, P


7/15


Table 2Population Characteristics of California

Binge Drinking During Last MonthNo Poor Mental Health Days Last 30 Days1-30 Poor Mental Health DaysDoctor Told Had CancerDoctor Told Had Heart DiseaseDoctor Told Had DiabetesDoctor Told Had AsthmaFair / Poor Health StatusGood HealthVery Good / Excellent HealthBMI Underweight / NormalOverweightObeseMale18-44 Years of Age45-64 Years of Age65-75 Years of Age76 + Years of AgeWhiteLatinoAfrican AmericanAsian/Pacific IslanderAIAN / OtherLess than 12 Years of EdueationHigh School GraduateSome CollegeCollege GraduateMarriedSocial Support, Mean (Std E rror)Never Smoked RegularlyCurrent SmokerQuit SmokingNo Usual Source of CareHousehold Ineome, Mean (Std Error)Working Full-Time, 21+ HoursAVeekWorking Part-time, 0-20 HoursAVeekNot Working / M issing DataEmployment-Based InsuranceMedieaidUninsured

Nonadherent - nottaiiing medications (%)est. n=l,941,1383094 surveys

17.449.950.2

8.67.87.17.9

29.531.838.730.436.633.051.652.635.36.65.549.529.17.09.64.824.325.925.324.552.6

15.00(0.09)54.620.924.516.7

$53.933 ($1.260)56.25.138.847.714.019.5

Adults with HypertensionAdherent - takingmedications (%)est n=4,070,4398373 surveys

8.165.234.818.322.021.7

9.340.230.829.030.637.232.246.310.1 .42.725.321.959.814.410.012.33.619.926.126.327.859.3

15.37(0.06)50.711.437.9

3.2$52.016 ($759)

34.84.560.833.58.55.0

P vaiue

< 0 . 0 0 1

< 0 . 0 0 1< 0 . 0 0 1< 0 . 0 0 1


8/15

Banta et al

and less likely to have poor EnglishIn summary, those with hyperten-

Table 2 presents the distributions of

Those taking hyper tension-re la ted

Males were moderately less adherent;ences; only 10.1% of ad-

Although there were no significant dif-

who were not.Table 3 presents the results of a multi-variable logistic regression model. Aftercovariate adjustment, there was a posi-tive relationship between having poormental health days and nonadherence.Compared to those with no poor mentalhealth days, the odds ratio for adherenceamong those with any poor mental healthdays during the las t 30 days w as 0.74 (95%CI 0.64 to 0.85). Binge drinking, on theother heind, was not a significant nonad-herence predictor (O.R. 0.99, 95% CI 0.78to 1.25). We ran several more models,models with additional alcohol-relatedvariables such as number of days in last30 days in which alcohol was consumed,average number of drinks per day, andnum ber of days in which 5 or more drinkswere consumed, as well as models withage-binge drinking and age-mental healthinteraction terms. These additional alco-hol-related variables and interactionterms were not significant, and enteringthese variables did not significantly af-fect the results (odds ratios) from theoriginal model.

Consistent with the bivariate analy-sis, factors associated with nonadher-ence in the multivariable model (P


9/15

Antihypertension Medication Adherence

Table 3Odds Ratios for Adherence to Antihypertension Medication,Multivariable Logistic Regression

Binge Drinking During Last Monthi or More Poor Mentai Health DaysDoctor Told Had CancerDoctor Toid Had Heart DiseaseDoctor Toid Had DiahetesDoctor Toid Had AsthmaFair / Poor Health StatusGood Heaith (Ref.)Very Good / Exceiient HeaithBMI Underweight / Normai (Ref.)OverweightObeseMaie18-44 Years of Age45-64 Years of Age (Ref.)65-75 Years of Age76 + Years of AgeWhite (Ref.)LatinoAfrican AmericanAsian/Pacific IsianderAIAN / OtherLess than 12 Years of EducationHigh School Graduate (Ref.)Some CoiiegeCoiiege GraduateMarriedSociai SupportNever Smoked Reguiariy (Ref.)Current SmokerQuit SmokingNo Usuai Source of CareHousehoid Income (log)Working Fuii-Time, 21+ HoursAVeek (Ref.)Working Part-Time, 0-20 Hours/WeekNot Working / Missing DataEmployment-Based Insuranee (Ref.)MedicaiduninsuredOther, Including MedicareCovered for Prescription DrugsRural

O d d sRat io

0.990.741.051.462.451.201.370.73

1.201.280.910.222.182.38

0.751.701.120.730.731.081.271.151.01

0.82I.I20.370.97

0.981.22

1.080.721.121.401.03

9S% Confidence SignificanceInterval Level(0.78, 1.25)(0.64, 0.85) (0.88, 1.26)(1.18, 1.82) (1.89, 3.17) (0.97, 1.49)(1.13, 1.67) (0.62,0.85) * "

(1.02, i.41) (1.08, 1.52) *(0.80, 1.04)(0.18, 0.26) (1.60,2.98) (1.66,3.40) (0.60, 0.94) (1.30,2.21) (0.88, 1.43)(0.52, 1.02)(0.59, 0.91) (0.92, 1.28)(1.08, i .5 l ) (0.98, 1.35)(0.99, 1.03)

(0.66, 1.02)(0.95, 1.30)(0.27,0.50) (0.92, 1.03)

(0.70, 1.39)(1.03, 1.44)

(0.83, 1.42)(0.52,0.99) (0.85, 1.48)(1.13, 1.73) '(0.85, 1.26)


10/15

Banta et al

g all other variables constant. A stan-is defined by using the meanfor numeric variables (eg, house-and social support) and thefor each categorical variable (eg,The sensitivity analy-the logit specification of thein Table 3 to calculateof adherence under differentFor this analysis, a standardwas taken to be an urban adultwas18 and 44 years of age,i attended some college, was work-in-and who had a usual source ofit was assumed that sheas a US citizen who spoke English well,and did not have comorbidor cancer; it was as-she was overweightto poor health. Usinga hypothetical person as an ex-the estimated likelihood of adher-was calculated varying only racial/and mental health status.

Table 4 presents the results of sensi-67.7% ofno poor

had the highest estimated prob-of adhering (53.0%) and Latinothe lowest (33.3%), highlightingTheof the6.4 toof poor mentalto those with noIn the case of Africanthe estimated adher-om 53.0 to 45 .4.is a 14.3% decrease in adherence

was a7.8% decrease among white women, a9.4% decrease among Latino women,nd a 17.1% decrease among Asian/Pa-DISCUSSION

Table 4Proportion of Persons withHypertension Est imated to beTaking HypertensionMedicat ionsNo Poor 1 to 30Mental Poor MentalHealth Health

Days (%) Days (%)White 40.0Latino 33.3African American 53.0Asian / Paciflc Islander 42.8

32.926.945.435.5Note.Used mean household income and socialsupport. Assumed fair/poor health,overweight, female, 18-44 years of age, somecollege, married, never smoked, has usualsource of care, working full-time,employment-hased insurance, prescriptiondrug coverage, US citizen, urban, speaksEnglish well, nocomorbid asthma, diabetes, orcancer.

tension, whereas binge drinking was not.Nearly 40% of adults with h5^ertensionreported having some poor mental healthdays in the prior month. Medication ad-herence improved as expected with in-creasing age, education, and insurance.Poor health status also was positivelyassociated with adherence; this findingparallels the results of meta-analytic re-search demonstrating that in the case ofhypertension, there is a positive rela-tionship between adherence and poorerhealth status.^^ Not surprisingly, lack ofinsurance and lack of a usual source ofcare were associated with medicationnonadherence. Although non-US citizenswere much less likely to have been givena diagnosis of hypertension, noncitizenswith hypertension were less likely to betaking medications to control their highblood pressure.Approximately one quarter of adu lt Cali-


11/15


dium intake.^' Of those in the presentstudy with hypertension, however, roughlyone third were nonadherent to a medica-tion regimen. These results are compat-ible with a previously cited meta-analysisin which nonadherence to treatments forhyper tens ion and other cardiovasculardiseases was approximately 24%.^ Adher-ence can be a part icular chal lenge inh3q3ertension because patients often donot have noticeable symptoms. Patientsalso may have medication side effects,which make adherence diff icul t . Suchfactors may be coupled with poor mentalhealth in addition to other barriers asdescribed in this study, and it is difficultto untangle and fully understand thesevarious connections. As predicted, how-ever, there was a significant relationshipbetween having poor mental health daysand medication adherence in this study.The results of the sensitivity analysis areconsistent with previously noted research,too, which h as shown a 27% higher risk ofnonadherence among depressed pat ientscompared to those who are not depressed.'"*Poor mental health can threaten adher-ence for a variety of reasons, includinghopelessness, lowered optimism, socialisolation, withdrawal from friends andfamily, and cognitive deficits that tend tobe associated with depression. ' ' '

Although binge drinking was not a sig-nificant predictor of medication nonad-herence, alcohol use does contribute tothe pathogenes is of hyper tens ion. De-creasing (and for some people, eliminat-ing) alcohol consumption is essential tohyper t ens ion management . The C ana-dian Hyper tens ion Educat ion Programrecommends, for example, limiting alco-hol use to a maximum of 14 drinks perweek in men and 9 drinks per week inwomen.* In the present study, binge drink-ing was defined as 5 or more drinks in onesitting, although elsewhere, binge drink-ing is defined as 8 or more drinks at onetime."*^ In studies of other diseases (eg,HIV), alcohol consumption has been linkedto poor medication adherence with strik-ing frequency.''^"''^ Th e positive as so cia-tion of alcohol consumption with poor

drinking and adherence relat ionship the pre sen t stud y. Next, disea se severmay be confounding the re la t ionshiRather than using HIV-positive patienthe present s tudy uses a populat ion-bassample of hypertension pat ients l ikewith varying degrees of associated seveity, which may affect adherence.^^These findings suggest that improvments in mental health and detection poor mental health may lead to increasmedicat ion adherence among hypertesive ad ul ts. It ma y be beneficial for hea lcare providers to screen for and addremental health problems among their hpertensive patients. Although this studid not link binge drinking to medicatinonadherence when controlling for othfactors, the findings are provocative asuggest a need for more studies designto understand and reduce the detr imetal effects of alcohol use, particularly ecessive use, in hypertensive patients. 'These findings also suggest that inteventions to improve adherence should directed tow ards young adults a nd Latinand further underscore the importance health care access, that is, health insuance coverage and having a usual sourof medical care. Research efforts must directed to the relationship between ptient health literacy and its effects patient adherence, as well as the ways which lower socioeconomic status ptients may be at risk for poor adherenan d poor he alth literacy.''* Also deservifurther research are methods of improing adherence among non-US citizens

Study LimitsThis study suffers from the weaknessassocia ted wi th analys is of secondadata, particularly lack of clinical specifity. Furthermore, as a state survey, rsul ts may not be general izable to te n t i r e n a t i o n . T h e h y p e r t e n s i o n a nm e d i c a t i o n a d h e r e n c e c r i t e r i a w etaken from a self-report interview abased on single, dichotomous questioa b o u t h y p e r t e n s i o n d i a g n o s i s aant ihyper tens ion medicat ion use . Nether was optimal; for example, we weunable to account for physician adhe


12/15

Banta et al

or psy chia tr ic-rela ted disab i l i ty.

f binge drin kin g in California co un -ed s in California counties.^" As a large

a lu a te the e ffec ts o f m an y m ore

CONCLUSIONDespite the study's l imits, and particu-

AcknowledgmentThe authors thank Fores t Labora to-Inc, for providing project funding. Ado, Florida.

1.American Heart Associat ion. Hea rt disea seand s t roke s ta t i s t i cs2007 update (onl ine) .Avai labl e a t : h t t p : / /ww w. am er i canh ea r t . o rg /Accessed June 20, 2007.2.Prevalence of act ions to control high bloodp r e s s u r e 2 0 s t a t e s , 2 0 0 5 . MMWR. 2 0 0 7 ;

for clinicians. Nurs Clin North Am. 2004;39(4):7 9 3 - 8 1 4 .5.U.S. Departm ent of Heal th and H um an Ser-vices. CDC Coordinat ing Center for Heal thPromot ion Address ing the Nat ion 's LeadingKil lers (2006) (onl ine) . Available at : ht tp: / /www.cdc.gov/ . Accessed June 20, 2007..Sabate E. Adherence to long-term therapies:evidence for ac t ion. Genev a, Swi tzer land:World Heal th Organizat ion 2 0 0 3 .7.DiMatteo MR. Variat ions in pat ients ' adher-ence to medical recommendat ions : a quant i ta-tive review of 50 years of research. Med Care.2 0 0 4 ; 4 2 ( 3 ) : 2 0 0 - 2 0 9 .8.Garber MC, Nau DP, Erickson SR, et al. Theconcordance of se l f - repor t wi th other mea-

sures of medicat ion adherence: a summary ofthe l i terature. Med Care. 2004;42(7) :649-652.9.DiMatteo MR, DiNicola DD. Achieving PatientCom pliance: The Psychology of the MedicalPra ct i t io ne r 's Role. New York: Perg am on1 9 8 2 .lO.Burnier M, Santschi V, Favrat B, et al. Moni-tor ing compl iance in res i s tant hyper tens ion:an impor tant s tep in pat ient management . JHypertens. Suppl 2003;21(2) :S37-S42.l l .Schroeder K, Fahey T, Ebrahim S. Interven-t ions for improving adherence to t reatment inpat ients with high blood pressure in ambula-tory se t t ings . Cochrane Datahase Syst R ev.2 0 0 4 ; ( 2 ) : C D 0 0 4 8 0 4 .12.Grgoire J, Moisan J, Guibert R, et al. Predic-tors of self-reported noncompliance with an-t ihypertensive drug treatment: a prospect ivecohort s tudy. Can J Cardiol. 2006;22(4):323-3 2 9 .13.1skedjian M, Einarson TR, MacKeigan LD, etal. Relationship between daily dose frequencyand adherence t o an t i hyper tens ive pha rma -co the rapy : ev idence f rom a me ta - ana lys i s .Clin Ther. 2002;24(2) :302-316.14.DiMatteo MR, Lepper HS, Croghan TW. De-pression is a r isk factor for noncompliancewith medical t reatment: meta-analysis of theeffects of anxiety and depression on patienta d h e r e n c e . Arch Intern Med. 200 0; 160(14) :2 1 0 1 - 2 1 0 7 .15.Spernak SM, Moore PJ, Hamm LF. Depres-sion, construct ive thinking and pat ient sat is-fact ion in cardiac t reatment adherence. PsycholHealth Med. 2007;12(2) :172-189.16.Uchino BN, Gacioppo JT, Kiecolt-Glaser JK.The relat ionship between social support andphys iological processes : a review wi th em-phas is on under lying mechanisms and impl i -cat ions for heal th. Psychol Bull. 1996 ; 119(3):4 8 8 - 5 3 1 .17.Siegel D, Lopez J, Meier J. Antihypertensive


13/15


19.Morris AB, Li J, Kroenke K, et al. Factorsassociated with drug adherence and bloodpressure control in patients with hyperten-sion. Pharmacotherapy. 2006;26(4):483-492.2O.Tykarski A. Resistant hypertension. BloodPress. 2005; 14 (Suppl 2):42-45.21.Winickoff RN, Murphy PK. The persistentproblem of poor blood pressure control. ArchIntern Med. 1987; 147(8): 1393-1396.22.U.S. Department of Health and Human Ser-vices, Nationed Institute on Alcohol Abuse andAlcoholism Newsletter. NIAAA Council ap-proves definition of binge drinking - 2004(onl ine). Avai lable a t : ht tp: / /pub s. niaaa. nih. gov/ publications / Newsletter/w i n t e r 2 0 0 4 / N e w s l e t t e r _ N u m b e r 3 . p d f .Accessed June 1, 2007.23.Lang T, Degoulet P, Aime F, et al. Relation-ship between alcohol consumption and hy-pertension prevalence and control in a Frenchpopulation. J Chronic Dis. 1987;40(7):713-720.24.Tuomilehto J, Enlund H, Salonen JT, et al.Alcohol, patient compliance and blood pres-sure control in hypertensive patients. Scand JSoc Med. 1984;12(4):177-181.25.Kim MT, Dennison CR, Hill MN, et al. Rela-tionship of alcohol and illicit drug use withhigh blood pressure care and control amongurban hypertensive Black men. Ethn Dis.2000;10(2):175-183.26.VoUcer L, Volicer BJ. Self reports on alcoholuse and compliance with antihypertensivemedication. Clin Exp Hypertens A. 1982;4(6):849-865.27.Goldman DP, Joyce GF, Escarce JJ, et al.Pharmacy benefits and the use of drugs by thechronically ill. JAMA. 2004;291(19):2344-2350.28.Krousel-Wood M, Hyre A, Muntner P, et al.Methods to improve medication adherence inpatients with hypertension: current statusand future directions. Curr Opin Cardiol.2005;20(4):296-300.

29.Kennedy J, Coyne J, Sclar D. Drug affordabilityand prescription noncompliance in the UnitedStates: 1997-2002. Clin Ther. 2004;26(4):607-614.30.Clark LT. Improving compliance and increas-ing control of hypertension: needs of specialhype r t e ns ive popu la t i ons . Am Heart J.1991;121(2 Pt 2):664-669.31.Xu KT, Ragain RM. Effects of weight status onthe recommendations of and adherence tolifestyle modifications among hypertensiveadults. J Hum Hypertens. 2005;19(5):365-371.32.DiMatteo MR, Haskard KB, Williams SL. Healthbeliefs, disease severity, and patient adher-ence : a meta -ana lys i s . Med Care.2007;45(6):521-528.

2006;21(12):1215-1221.35.Hong TB, Oddone EZ, Dudley TK, et Medication barriers and anti-hypertensimedication adherence: the moderating role locus of control . Psychoi Health Me2006; l l ( l ) :20-28 .36.Osterberg L, Blaschke T. Adherence to mecation. JV Engl J Med. 2005;353(5):487-49737.California Health Interview Survey. CH2003 Adult Public Use File, Release 1 (copu ter file). Los Angeles, CA: UCLA Center Health Policy Research 2005.38.Ponce NA, Lavarreda SA, Yen W, et al. TCalifornia Health Interview Survey 200translation of a major survey for Californimultiethnic population. Public Health R2004;119(4):388-395.39.California Health Interview Survey. Weiging and Estimation of Variance in the CHPublic Use Files . Los Angeles, CA: UCCenter for Health Policy Research 2005.4O.Milchak JL, Carter BL, Ardery G, et Measuring adherence to practice guidelinfor the management of h3T3ertension: an evaat ion of the l i tera ture . Hypertensio2004;44(5):602-608.41.Khan NA, McAlister FA, Rabkin SW, et al. T2006 Canadian Hjrpertension Education Pgram recommendations for the management hypertension: part II - therapy. Can J Card2006;22(7):583-593.42.Murray RP, Connett JE, Tyas SL, et Alcohol volume, drinking pattern, ind cardvascular disease morbidity and mortality: there a U-shaped function? Am J Epidem2002;155(3):242-248.43.Arnsten JH, Demas PA, Grant RW, et Impact of active drug use on antiretrovitherapy adherence and viral suppression HIV-infected drug users. J Gen Intem M2002;17(5):377-381.44.Cook RL, Sereika SM, Hunt SC, et al. Probldrinking and medication adherence amopersons with HIV infection. J Gen Intem M2001;16(2):83-88.45.Golin CE, Liu H, Hays RD, et al. A prospectstudy of predictors of adherence to combition antiretroviral medication. J Gen IntMed. 2002;17(10):756-765.46.Wagner JH, Justice AC, Chesney M, et Patient- and provider-reported adherentoward a clinically useful approach to measing antiretroviral adherence. J Clin Epidem2001;54(Suppl):S91-S98.47.Miller PM, Stockdell R, Nemeth L, et al. Inisteps taken by nine primary care practicesimplement alcohol screening guidelines whypertensive patients: the AA-Trip proje


14/15

Banta et al

49.Hays RD, DiMatteo MR. Key issue s and sug- 50.California De par tm ent of Health Services,gestions for patien t compliance assessme nt: Office of Health Information and Researchsources of information, focus of m easu res, dat a repo rts (online). Available at: h t tp : //and nature of response options. J Compliance www.d h s . ca .g o v /h i sp /ch s /OHIR/ r ep o r t s / .Health Care. 1987;2(l):37-53. Accessed October 16, 2007 .


15/15

compliance antihipertensi

Documents