SUBMITTED 18 MAR 23 1 REVISION REQ. 7 MAY 23; REVISION RECD. 25 MAY 23 2 ACCEPTED 2 JUL 23 3 ONLINE-FIRST: JULY 2023 4 DOI: https://doi.org/10.18295/squmj.7.2023.045 5 6 Adherence to Medications in Patients with Ischemic Heart Disease in Oman 7 Ahmed Al-Maskari,1 Qasim Al-Maamari,1 Mariya Al-Abdali,2 Hajer Al-8 Shaaibi,2 Sunil K. Nadar3 9 10 1Directorate of Nursing and 3Department of Medicine, Sultan Qaboos University Hospital, 11 Muscat, Oman; 2College of Medicine and Health Sciences, Sultan Qaboos University, Muscat, 12 Oman. 13 *Corresponding Author’s e-mail: sunilnadar@gmail.com 14 15 Abstract 16 Objectives: The aim of this study was to evaluate the level of adherence to medications in 17 patients with IHD in Oman and assess the factors influencing it. Methods: This was a cross 18 sectional questionnaire-based study among patients with IHD. Results: A total of 105 patients 19 (Mean age 49.9+11.1 years; 78.1% male) were recruited. Most of the patients take the 20 medications by themselves (84 or 80%). 77 patients (73.3%) said that over the preceding two 21 weeks, they missed at least three doses. The reasons for missing included forgetting (100%), too 22 many tablets (57%), not effective (48%) and too many times a day (23%). There were no factors 23 that could be identified that made patients prone to not taking medications. Conclusion: 24 Medication adherence was low among patients with IHD in Oman with high pill burden the most 25 common reason for non-adherence. Physicians must keep this in mind when patients are 26 reviewed. 27 Keywords: Cardiovascular disease, adherence, myocardial infarction. 28 29 mailto:sunilnadar@gmail.com Introduction 30 Cardiovascular disease are a major cause of morbidity and mortality worldwide. It is estimated 31 that 17.9 million deaths (representing 32% of all deaths) in 2019 were attributed to CVD.1 A 32 major part of the management of CVD lie in adherence to medications and lifestyle changes with 33 the intention of lowering future cardiovascular events and for symptom control.2 Medications are 34 prescribed either as primary prevention for those at high risk for future cardiovascular events or 35 secondary prevention of future events for those who have already sustained a cardiovascular 36 event. 37 38 Current evidence based practice has led to patients with CVD and those at high risk, being 39 initiated on an increasing number of medications. This can affect adherence and as with other 40 chronic illnesses, non-adherence to medication and lifestyle modifications, remains a major 41 issue.3 Studies have demonstrated that adherence is often an issue when given as a long-term 42 preventative strategy rather than for symptom control. This is especially true for hypertensive 43 patients who are often asymptomatic but experience various side effects to medications.4 44 The reported rates of adherence with cardiovascular medications range from 30-70%, with 45 patients often not taking all or part of their prescribed medications.5 Poor adherence is a major 46 public health issue such that, the World Health Organisation (WHO) in their report on adherence 47 to long term therapies, state that interventions to improve medication adherence might have a far 48 greater impact on the health of the population than any improvement in a specific medical 49 treatment itself.6 50 51 There are many factors that are responsible for poor adherence and understanding these factors in 52 a particular population is key to improving medication adherence and ultimately improving 53 patient care and outcomes. The WHO has identified five broad categories of factors that affect 54 adherence: 1) socio-economic factors 2) patient-related, 3) therapy-related, 4) co-morbid 55 conditions, and 5) healthcare system-related.6 The first four factors are mainly related to the 56 individual patient and it is important to understand these factors which are very individualised 57 and often related to patient preferences and social and cultural behaviours. These include costs, 58 beliefs regarding medications and treatment, number of pills, side effects of pills, understanding 59 the need for medications, the benefits of it, forgetfulness and overall wellbeing.7-9 60 61 Improving medication adherence helps improve cost savings and helps to put into practice 62 lessons learned from clinical trials and ultimately helps to reduce the burden of chronic 63 illnesses.10 Early detection of non-adherence can prevent expensive investigations, hospital 64 admissions and unnecessary additional medications and interventions. It is important to ensure 65 high levels of adherence to medications to improve cardiovascular outcomes in the population as 66 a whole.2 67 68 In Oman, IHD is a leading cause of mortality and morbidity accounting for around 35% of all 69 deaths in 2020.11 There is dearth of information regarding adherence to medication in the 70 Middle-east. The aim of this study was to assess the levels of medication adherence in patients 71 with IHD and the patient related factors affecting it in Oman. 72 73 Methods 74 This was a cross-sectional study questionnaire based on patients who attended the outpatient 75 clinic at a tertiary hospital in Muscat, Oman. This was performed between January to December 76 2021. Patients aged 18 or over, who were diagnosed to have IHD with a previous MI or 77 undergone PCI for stable angina more than a year earlier were included in the study. Patients had 78 to be independent and fully alert and oriented to be included in the study, They could take 79 medications either by themselves or be given by a carer. Patients who were not able to give 80 consent to the study, or those who had a recent cardiac event or who had a hospitalisation for a 81 cardiovascular event within the preceding 12 months were not included in the study. We chose a 82 sample size of 100 as this is a pilot study. 83 84 The questionnaire was self-developed. It was based in two parts. The first part was about the 85 general demographics of the patients, the number of medications and the total pill burden. The 86 second part of the questionnaire was about their practice of taking medications. It included a 87 question regarding whether they missed more than three doses in the previous two weeks. If they 88 did, they were considered as non-adherent. This was because 3 days out of 14 would work out to 89 an adherence rate of 80% which is the rate accepted by many studies for adequate adherence. 90 This was initially trialled on a few volunteers to adjust the wording, to ensure that there were no 91 difficult or confusing words. The questionnaire was developed in Arabic by native Arabic 92 speaking staff. 93 94 Ethical approval was obtained from the research committee of the Sultan Qaboos University 95 (MREC #1550). All patients gave informed consent before answering the questionnaire. 96 Statistical analysis was performed using SPSS version 21. Data was presented as number (%) or 97 mean + standard deviation (SD) or median (Interquartile range- IQR). Data analysis was done by 98 chi-square test or student t-test as appropriate. A p value of <0.05 was considered to be 99 statistically significant. 100 101 Results 102 A total of 105 patients (Mean age 49.9+11.1 years; 82 or 78.1% male) filled in the 103 questionnaires. The patients were taking a median of 9 (interquartile range (IQR) of 6-10) tablets 104 per day. They were taken a median 3 (IQR 2-3) times a day. Most of the patients had a previous 105 MI (100 or 95.2%) with the remaining 5 having chronic stable angina. 87 (82.9%) of patients had 106 a previous PCI, while 12 (11.4%) had previous CABG. Hypertension was the commonest risk 107 factor (97 or 92.4%) followed by diabetes (87 or 82.9%). 14 patients (13.3%) had a previous 108 stroke. Table 1 summarises their demographic features. 109 110 Most of our patients take the medications by themselves (84 or 80%), while for the remaining 21 111 (20%) a family member or carer administers the medications. Twenty-two patients (%%) use a 112 reminder to help them remember to take medications. This includes a phone app for 12 patients, 113 and a family member for 10. 114 115 77 patients (73.3%) said that over the preceding two weeks, they would have missed at least 116 three doses. The main reason given by our patients for not taking medications regularly are that 117 they simply forgot (77 or 100%), while 44(57%) patients felt that there were too many [Table 2]. 118 Thirty-seven (48%) felt that they were not effective and 18 didn’t take all the medications as 119 prescribed as it was too many times a day (23%). A further 41 (39%) said that they stop taking 120 the tablets when they feel well. A majority of patients (80 or 76.2%) felt that taking tablets was a 121 burden and 63 (60%) said that they sometimes do not take tablets while travelling. 122 123 Table 3 summarises the differences between those who missed more than 3 doses of tablets over 124 the previous 2 weeks with those who said they never missed a single dose. There was no 125 difference between the two groups. There was no difference in age, gender, educational status, or 126 the number of pills or the number of times a day. The cardiovascular risk factors were similar, 127 apart from those who had a previous stroke had a higher number of those who missed tablets (p-128 0.01). Although there did not appear to be any difference between the two groups, by binary 129 logistic regression, the only factor that predicted non-adherence was the number of pills taken 130 daily (Odds ratio 1.26, 955 confidence interval 1.001-1059, p=0.04) 131 132 Discussion 133 The rates of non-adherence were high in our group, with more than three-quarters of those 134 surveyed saying that they had missed multiple doses over the previous two weeks. The findings 135 are similar to those from other studies. Adherence to medications in CVD in are low worldwide. 136 In a meta-analysis of around 20 observational studies involving more than 300,000 patients with 137 CVD, it was estimated that the prevalence of poor adherence was as high as 43%.12 For 138 individual risk factors that rate can be even higher. It has been demonstrated that at the end of 139 end of 6 months one-third of patients discontinue their antihypertensive medications, and only 140 around half of all patients persist with their initial therapy at one year.4 141 142 The rates of adherence for primary prevention of CVD are generally lower than those for 143 secondary prevention. Patients who have suffered a myocardial infarction or a stroke are more 144 likely to take their medication regularly than those who have not suffered an event.12 However, 145 even for those on secondary prevention, adherence is sub-optimal as demonstrated on a study on 146 4591 post MI patients, where around 18% of patients did not collect their prescriptions even once 147 in the four months following the MI.13 In in a separate cohort of 22,379 post-ACS patients, 60% 148 discontinued their statin medication within 2 years of hospitalization.14 149 150 There are many factors that affect adherence to medications.5 These include physician/healthcare 151 related factors, patient factors and socio-economic factors. Healthcare factors include busy 152 outpatient consultations, pill burden, access to pharmacies and ease of refills and costs. Patient 153 factors include socio-economic factors such as affordability, access to health care, social 154 circumstances, forgetfulness, and understanding of the disease process.15 155 156 Beliefs regarding medications, the disease process and its management are an important part and 157 determinant of medication adherence. It has been shown that these beliefs play an important part 158 in medication adherence even among people of middle-eastern origin who live in Australia.16 159 Beliefs regarding medications include the importance and usefulness of taking medications, and 160 the concern regarding the illness. Horne et al, into one of four subgroups according to their 161 attitudes towards medication; Sceptical (low necessity, high concerns), Ambivalent (high 162 necessity, high concerns), Indifferent (low necessity, low concerns), and Accepting (high 163 necessity, low concerns).17 The sceptical and ambivalent patients have been shown to have low 164 adherence rates as compared to the other two groups.16;18 Education is therefore an important part 165 of ensuring adherence and empowering the patient to take care of their own health by explaining 166 the need for adherence and the benefits. In our cohort of patients, we did not specifically study 167 the beliefs of the patients, though we had asked them whether or not they felt the medications 168 were effective. 169 170 Other cultural factors such as the use of traditional medicine, which is high in Oman, could also 171 play a role in non-adherence. 19 In this study Al-Riyami et al found that a high proportion of 172 people prefer to take herbal medications and undergo traditional practices to taking medications 173 for chronic illnesses. Scicchitano et al have described the role of “functional foods” or 174 “nutraceuticals” in cardiovascular disease.20 Functional foods are described as any food or food 175 ingredient that may provide a health benefit beyond the traditional nutrients it contains,21 while 176 nutraceuticals have been defined as food (or part of a food) that provides medical or health 177 benefits, including the prevention and/or treatment of a disease.22 This concept might explain the 178 benefits of herbal diets and the willingness of the population to try these remedies rather than 179 modern medications and thereby affecting adherence. 180 181 The findings of our study are also in keeping with data from the middle east, where adherence 182 rates ranging from 1.4-88% have been demonstrated in a variety of conditions.23 A previous 183 study from Oman on patients with hypertension also revealed that less than 50% of the patients 184 were adherent to medications.24 Al-Qasem et al performed a systemic review of studies from the 185 middle east and found that the reasons reported by patients for non-adherence in the Middle East 186 are similar to those reported in the international literature.23 However, some of the reasons 187 mentioned were unique to the middle-east such as patient dissatisfaction and/or lack of trust in 188 health care providers and lack of social support. Poor physician support and explanations and 189 lack of understanding of the need for taking medications has also been described previously from 190 the region.25 191 192 In our study, we could not identify any factor that could predict non-adherence apart from the 193 number of pills. The age, gender, educational status, or marital status did not affect adherence. 194 The use of aids or whether the patient self-medicated or whether it was given by a carer did not 195 affect adherence. It is possible that the small sample size was not able to differentiate between 196 those who were adherent and those who were not. The mean age of our patients was 49 years 197 with only a small proportion of patients above 60 years. This could explain why age did not 198 appear to be a contributing factor. Previous studies on adherence to medications for other 199 conditions in the region demonstrated that gender, educational status affected adherence, with 200 female patients and those with lower educational achievement had poorer adherence. In our 201 study we did not demonstrate any difference, possibly again due to the small sample size and 202 general over all poor adherence among all groups. 23;26;27 203 204 The only factor that predicted low adherence was the pill burden. Most patients with IHD are on 205 a large number of pills for secondary prevention. After coronary intervention, or on adequate 206 medical therapy, many patients have adequate control of symptoms which then becomes a 207 determinant of poor adherence.8;13;14;28 The average number of pills per patient in our study was 208 around 9 which is a large number. A large proportion of patients felt that they were taking too 209 many tablets and felt burdened by it. It is therefore important for physicians to review patients’ 210 medications at each visit to ensure that patients are not on any medications that they do not need 211 to be on, such as stopping dual antiplatelet therapy after one year (if indicated). The use of 212 combination pills reduces the pill burden and has been demonstrated to improve adherence in 213 patients with CVD with improved clinical outcomes.29 Perhaps local health authorities should 214 consider using more combination pills in routine practice. 215 216 Assessment of medication adherence is not easy There are many questionnaires that are validated 217 for a variety of conditions.15 However, all of them have their own advantages and disadvantages 218 and has been described elsewhere. We did not use any of these questionnaires, but instead chose 219 to directly ask the patients whether or not they missed any doses in the previous two weeks. 220 Although direct questioning has its limitations, we felt this would help us explore reasons behind 221 non-adherence.30 222 223 There were a few limitations to our study. Medication adherence is very variable over time, with 224 patients having periods of strict adherence followed by periods of mild to severe nonadherence.31 225 Our study only investigated a snapshot of the two weeks preceding the questionnaire. This is 226 therefore not representative of their overall long-term adherence. This study was performed in 227 the outpatient setting of a tertiary hospital in Muscat, Oman which is a large city. The results are 228 not generalisable to the whole of Oman and the rural areas. Access to health care, beliefs 229 traditions and customs are different in the different parts of Oman, along with different social 230 and economic conditions, all of which play a role in medication adherence and have not been 231 fully assessed in this study. Another limitation was the lack of follow up. It would have been 232 useful to have additionally studied the effects of intervention and co-operation between primary 233 and secondary care on the levels of medication adherence as this has been previously 234 demonstrated to be useful in an Italian population.32 235 236 Our study was limited to Omani population and the adherence rates among the expat population 237 might be different due to the different socio-economic conditions. Our study sample size was 238 small. This study was conducted just when social restrictions related to the COVID pandemic 239 were being lifted. Access to healthcare and patients was still not as free as prior to the pandemic. 240 However, this is the first study of its type on adherence to medications among patients with IHD 241 from Oman. Larger studies are required to fully assess the barriers to optimal medication 242 adherence among patients with IHD from different parts of Oman. 243 244 Conclusion 245 The rates of medication non-adherence in patients with IHD is high in Oman. Number of pills 246 was the major determinant of non-adherence. Lowering pill burden and frequent review of 247 medications, use of combination pills must be considered in these patients. Additionally, more 248 needs to be done to assess the needs of each individual patient in order to improve their 249 understanding of the disease and to improve overall adherence. 250 251 Conflicts of Interest 252 The authors declare no conflict of interests. 253 254 Funding 255 No funding was received for this study. 256 257 Authors’ Contribution 258 AA, QA, MA, HA were involved in collection of data and contributing to writing the 259 manuscript. SKN-data analysis and manuscript writing. All authors approved the final version of 260 the manuscript. 261 262 References 263 1. Burden of cardiovascular diseases. Available at https://www.who.int/news-room/fact-264 sheets/detail/cardiovascular-diseases-(cvds) Accessed on April 2023 265 2. Roth GA, Mensah GA, Fuster V. The Global Burden of Cardiovascular Diseases and Risks: A 266 Compass for Global Action. J Am Coll Cardiol 2020 Dec 22;76(25):2980-1. 267 3. Miller NH. Compliance with treatment regimens in chronic asymptomatic diseases. Am J Med 268 1997 Feb 17;102(2A):43-9. 269 4. Hill MN, Miller NH, Degeest S, Materson BJ, Black HR, Izzo JL, Jr., et al. Adherence and 270 persistence with taking medication to control high blood pressure. J Am Soc Hypertens 271 2011 Jan;5(1):56-63. 272 5. Baroletti S, Dell'Orfano H. Medication adherence in cardiovascular disease. Circulation 2010 273 Mar 30;121(12):1455-8. 274 6. Adherence to Long Term Therapies: Evidence for Action; Geneva: World Health 275 Organization, 2003. Available at 276 https://apps.who.int/iris/bitstream/handle/10665/42682/9241545992.pdf?sequence=1&is277 Allowed=y accessed on April 2023 278 7. Gast A, Mathes T. Medication adherence influencing factors-an (updated) overview of 279 systematic reviews. Syst Rev 2019 May 10;8(1):112. 280 8. Kvarnstrom K, Westerholm A, Airaksinen M, Liira H. Factors Contributing to Medication 281 Adherence in Patients with a Chronic Condition: A Scoping Review of Qualitative 282 Research. Pharmaceutics 2021 Jul 20;13(7). 283 9. Magrin ME, D'Addario M, Greco A, Miglioretti M, Sarini M, Scrignaro M, et al. Social 284 support and adherence to treatment in hypertensive patients: a meta-analysis. Ann Behav 285 Med 2015 Jun;49(3):307-18. 286 10. Axon DR, Vaffis S, Chinthammit C, Lott BE, Taylor AM, Pickering M, et al. Assessing the 287 association between medication adherence, as defined in quality measures, and disease-288 state control, health care utilization, and costs in a retrospective database analysis of 289 Medicare supplemental beneficiaries using statin medications. J Manag Care Spec Pharm 290 2020 Dec;26(12):1529-37. 291 11. Cardiovascular disease in Oman. Available at https://www.worldlifeexpectancy.com/oman-292 coronary-heart-disease accessed on April 2023 293 12. Naderi SH, Bestwick JP, Wald DS. Adherence to drugs that prevent cardiovascular disease: 294 meta-analysis on 376,162 patients. Am J Med 2012 Sep;125(9):882-7. 295 13. Jackevicius CA, Li P, Tu JV. Prevalence, predictors, and outcomes of primary nonadherence 296 after acute myocardial infarction. Circulation 2008 Feb 26;117(8):1028-36. 297 14. Jackevicius CA, Mamdani M, Tu JV. Adherence with statin therapy in elderly patients with 298 and without acute coronary syndromes. JAMA 2002 Jul 24;288(4):462-7. 299 15. Lam WY, Fresco P. Medication Adherence Measures: An Overview. Biomed Res Int 300 2015;2015:217047. 301 16. Shahin W, Kennedy GA, Cockshaw W, Stupans I. The Role of Medication Beliefs on 302 Medication Adherence in Middle Eastern Refugees and Migrants Diagnosed with 303 Hypertension in Australia. Patient Prefer Adherence 2020;14:2163-73. 304 17. Horne R, Parham R, Driscoll R, Robinson A. Patients' attitudes to medicines and adherence 305 to maintenance treatment in inflammatory bowel disease. Inflamm Bowel Dis 2009 306 Jun;15(6):837-44. 307 18. Unni E, Shiyanbola OO. Clustering medication adherence behavior based on beliefs in 308 medicines and illness perceptions in patients taking asthma maintenance medications. 309 Curr Med Res Opin 2016;32(1):113-21. 310 19. Al-Riyami H, Al-Maskari A, Nadar SK, Almaskari M. Knowledge, Attitudes and Practices 311 Regarding Traditional and Complimentary Medicine in Oman. Sultan Qaboos Univ Med 312 J 2023 Feb;23(1):90-8. 313 20. Scicchitano P, Cameli M, Maiello M, Modesti PA, Muiesan ML, Novo S, et al. 314 Nutraceuticals and dyslipidemia: Beyond the common therapeutics. J functional foods 6, 315 11-32. 2014. 316 21. Ross S. Functional foods: the Food and Drug Administration perspective. Am J Clin Nutr 317 2000 Jun;71(6 Suppl):1735S-8S. 318 22. Kalra EK. Nutraceutical--definition and introduction. AAPS PharmSci 2003;5(3):E25. 319 23. Al-Qasem A, Smith F, Clifford S. Adherence to medication among chronic patients in 320 Middle Eastern countries: review of studies. East Mediterr Health J 2011 Apr;17(4):356-321 63. 322 24. Al-Noumani H, Wu JR, Barksdale D, Alkhasawneh E, Knafl G, Sherwood G. Relationship 323 Between Medication Adherence and Health Beliefs Among Patients with Hypertension in 324 Oman: Pilot study. Sultan Qaboos Univ Med J 2017 Aug;17(3):e329-e333. 325 25. Bassam M, Behbehani N, Farouk H, Alsayed M, Montestruc F, Al-Jahdali H, et al. 326 Adherence to medication among adult asthma patients in the Middle East and North 327 Africa: results from the ESMAA study. Respir Med 2021 Jan;176:106244. 328 26. Al-Jahdali H, Wali S, Salem G, Al-Hameed F, Almotair A, Zeitouni M, et al. Asthma control 329 and predictive factors among adults in Saudi Arabia: Results from the Epidemiological 330 Study on the Management of Asthma in Asthmatic Middle East Adult Population study. 331 Ann Thorac Med 2019 Apr;14(2):148-54. 332 27. Almazrou SH, Alfaifi SI, Alfaifi SH, Hakami LE, Al-Aqeel SA. Barriers to and Facilitators 333 of Adherence to Clinical Practice Guidelines in the Middle East and North Africa 334 Region: A Systematic Review. Healthcare (Basel) 2020 Dec 15;8(4). 335 28. Saag KG, Bhatia S, Mugavero MJ, Singh JA. Taking an Interdisciplinary Approach to 336 Understanding and Improving Medication Adherence. J Gen Intern Med 2018 337 Feb;33(2):136-8. 338 29. Castellano JM, Pocock SJ, Bhatt DL, Quesada AJ, Owen R, Fernandez-Ortiz A, et al. 339 Polypill Strategy in Secondary Cardiovascular Prevention. N Engl J Med 2022 Sep 340 15;387(11):967-77. 341 30. Anghel LA, Farcas AM, Oprean RN. An overview of the common methods used to measure 342 treatment adherence. Med Pharm Rep 2019 Apr;92(2):117-22. 343 31. Mathews R, Wang W, Kaltenbach LA, Thomas L, Shah RU, Ali M, et al. Hospital Variation 344 in Adherence Rates to Secondary Prevention Medications and the Implications on 345 Quality. Circulation 2018 May 15;137(20):2128-38. 346 32. Locuratolo N, Scicchitano P, Antoncecchi E, Basso P, Bonfantino VM, Brescia F, et al. 347 [Follow-up of patients after an acute coronary event: the Apulia PONTE-SCA program]. 348 G Ital Cardiol (Rome) 2022 Jan;23(1):63-74. 349 350 Table 1: Demographics of the participants. 351 Number (n=105) percentage Age 49.9 ± 11.1 years Sex Male Female 82 23 78.1% 21.9% Marital status Never married Married Separated 7 81 17 6.7% 77.1% 16.2% Educational status Less than Primary school Secondary school Graduate or more 35 49 21 33.7% 46.7% 20% Diabetes 87 82.9% Hypertension 97 92.4% Smoker 46 43.8% Previous stroke 46 43.8% Previous MI 100 95.2% Previous CABG 12 11.4% Previous PCI 87 82.9% MI = myocardial infarction; CABG = coronary artery bypass grafting; PCI = percutaneous 352 coronary intervention. 353 354 Table 2: Reasons for non-compliance. 355 Number (n=77) Percentage Just forget to take 57 74.02% Too many tablets 44 57.1% Too many side effects 18 23.3% Not effective 37 48.1% Many times a day 15 19.4% 356 Table 3: Differences between those who missed tablets and those who did not. 357 Those who did not miss (n=28) Those who missed more than 3 doses in last two weeks (n=77) p-value Age (years) 49.01 ± 10.1 50.2 ± 11.5 0.61* Gender Male Female 23 (82.1%) 5(17.9%) 59(76.6%) 18(23.4%) 0.60 Marital status Married Single/Divorced/Widowed 21(75%) 7(25%) 60(77.9%) 17(22.1%) 0.54 Educational status Primary school or less Completed secondary school Postgraduate studies 9(32.1%) 12(42.8%) 7(25%) 26(33.7%) 37(48.1%) 14(18.1%) 0.7 Number of pills 8(5-10) 9(6-10) 0.24* Number of times per day 3(2.5-3) 3(3-3) 0.9* Diabetes 21(75%) 66(85.7%) 0.19 Hypertension 26(92.8%) 71(92.2%) 0.91 Smoker 11(39.2%) 35(45.4%) 0.67 Previous cva 0 14(18.1%) 0.01 Previous CABG 1(3.5%) 11(14.2%) 0.12 Previous PCI 21 66(85.7%) 0.19 Who gives the pills Self Carer 22(78.5%) 6(21.5%) 62(80.5%) 15(19.5%) 0.52 Use reminder 7(25%) 15(19.5%) 0.53 CVA = cerebrovascular accident; PCI = percutaneous coronary intervention; CABG = 358 coronary artery bypass grafting. 359 Analysis by chi-square test apart from *students t-test and **Mann Whitney U test. 360