Social and community factors associated with hypertension awareness and control among older adults in Tirana, Albania

Social and community factors associated with hypertension awareness and control among older... Abstract Background Determinants of hypertension diagnosis and/or awareness and control among older adults are understudied in Albania, a former communist country in South Eastern Europe, which is experiencing rapid demographic, socioeconomic and epidemiological transition. This paper examines the association of individual, interpersonal, organizational and community factors with hypertension awareness and control among older adults in Tirana, the Albanian capital. Methods Using 2012 International Mobility in Aging Study data on older adults from Albania’s capital city (n = 393) and the socioecological model as a conceptual framework, multinomial regression models identified factors associated with controlled, uncontrolled and undiagnosed hypertension. Results For hypertension, 17.3% participants had none, 23.4% were controlled, 48.4% were uncontrolled and 10.9% were undiagnosed/unaware. Compared to those with controlled hypertension, in multivariable models, a high level of friend support was negatively associated with uncontrolled (OR: 0.4; 95% CI: 0.2–0.9) and undiagnosed (OR: 0.2; 95% CI: 0.1–0.6) hypertension. A high level of perceived neighbourhood safety was negatively associated with uncontrolled (OR: 0.6; 95% CI: 0.3–1.0) and undiagnosed (OR: 0.4; 95% CI: 0.2–1.0) hypertension. Compared to those with no hypertension, children’s social support was positively associated with uncontrolled (OR: 2.2; 95% CI: 1.1–4.3) and undiagnosed (OR: 3.6; 95% CI: 1.3–9.6) hypertension. Conclusion This study provides new insights about distinct risk factors for inadequate hypertension management in Albania. It highlights the importance of community-level factors (safety) and interpersonal factors (family and friend ties) to hypertension diagnosis/awareness and control, which may provide novel intervention opportunities for hypertension programs. Introduction Non-communicable diseases (NCDs) account for 89% of deaths in Albania; cardiovascular diseases (CVD) alone account for 59% of total mortality.1 Albania, historically the most isolated former communist country in the Western Balkans, is currently experiencing swift demographic and socioeconomic transitions, including a rapid epidemiological shift from infectious diseases to NCDs.1 Between 1990 and 2010, the mortality rate from ischemic heart disease increased twofold in Albanian men and women.1,2 Mortality from ischemic heart disease in Albania is the highest in the South Eastern European region.1,2 Hypertension, a major CVD risk factor, is highly prevalent in Albania, especially among older adults whose proportion in the population is steadily increasing in line with global trends.1,3 Hypertension is the primary driver of death and disability in Albania.2 In the past two decades, the mortality rate from ischemic heart disease attributable to hypertension doubled.1 Nevertheless, hypertension awareness—e.g. knowledge that one has hypertension, which is predicated on diagnosis and adequate comprehension of that diagnosis—and control—e.g. successfully managing hypertension once diagnosed—are notably low in Albania,4,5 even though adequate hypertension control reduces the risk of CVD dramatically.6,7 Achieving hypertension control is an important health system goal given the opportunities for reduced disability and premature death. While average blood pressure values have declined in nearly all European countries since 1980, Albania has experienced an increase in mean systolic blood pressure over the same timeframe.6 There is wide acceptance that adherence to modern guidelines for the detection/treatment of hypertension has contributed to reduced hypertension levels in Western Europe.6 Yet, even in Western Europe, control among diagnosed patients appears elusive, with 30–50% of patients achieving hypertension control.8 There is a clear need to better understand risk factors for inadequate hypertension awareness and poor control, especially in Albania. Research on correlates of hypertension prevalence and control in Albania is largely limited to individual and behavioural predictors.4,5 Behaviours are influenced by social norms9 and other factors such as interpersonal relationships and community resources may also influence hypertension awareness and control. For example, social network ties and community integration are associated with hypertension-related outcomes elsewhere,10–12 but understudied in Albania. Yet, social factors and community connections have been demonstrated as important to health outcomes in the Balkan region.13 To better understand barriers to hypertension awareness and control in Albania, it is necessary to characterize broadly risk factors for hypertension from the individual to community, particularly given Albania’s unique political and social history. This approach can support the successful targeting and design of impactful public health programs.14,15 Guided by the socioecological model,16 this paper examines the association of factors, ranging from individual characteristics to social and community factors, with hypertension awareness and control in Albania. Our objective is to quantify the contribution of a more diverse set of predictive factors of poor hypertension awareness and control in Albania than previously studied. We are particularly interested in the contribution of individual-level risk factors when social and community factors are also considered in predictive models. Methods Data source and study population This study analyses 2012 baseline data from The International Mobility in Aging Study (IMIAS), which is a population-based longitudinal study conducted in five sites globally, including Tirana (700 000 inhabitants), the capital city of Albania.17 A detailed description of the study, including information about the study sites and cohort is available elsewhere.17 IMIAS randomly recruited equal numbers of community-dwelling men and women, aged 65–74 years, from the population registered at neighbourhood health centres in Tirana (N = 393). Over 90% of older adults in Tirana are registered at a health centre and the acceptance rate was 90%. Those with cognitive impairment were excluded from the study (one person in Tirana).17 Informed consent was obtained from all individual participants. Interviews were conducted at participants’ homes and administered by interviewers trained with standardized protocols. Additional procedures included taking the participants’ blood pressure and anthropometric measures, as well as a blood draw.17 Participants were also asked to bring the containers of all medications they were taking. The interviewer then recorded the medication names and a trained pharmacist coded these according to their approved uses.17 Conceptual framework In the socioecological model, individual-level health outcomes, knowledge and behaviours are influenced by interpersonal (e.g. family, friends), organizational (e.g. health system), community (e.g. neighbourhood safety) and policy levels.16 The socioecological model is well-used for studies of many health outcomes, including the understanding of factors predicting chronic disease.14 This model is particularly relevant to our work, as research from diverse locations demonstrate that social and environmental factors are strong determinants of older adults health.18 Importantly, the well-recognized role of social and community context in contributing to chronic disease among older adults is understudied in South Eastern Europe. Measure of hypertension awareness and control Blood pressure was measured three times with a validated automated blood pressure device after 5 min of rest. The mean value of the second and third blood pressure measurements was used in the analysis. Clinical hypertension was defined by a measured blood pressure of ≥140/90. Self-reported doctor-diagnosed hypertension was recorded based on an affirmative response to the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’. Self-reported hypertension has been validated in a number of studies and is highly specific (e.g. low probability that non-hypertensive persons are classified as hypertensive).19–21 We compared (i) controlled (no clinical hypertension, but self-reported hypertension), (ii) uncontrolled (clinical hypertension and self-reported hypertension) and (iii) undiagnosed/unaware of hypertension (clinical hypertension, but no self-reported hypertension), with (iv) no hypertension across factors from the socioecological model (described below). Table 1 depicts our classification of participants into hypertension categories based on the clinical and self-reported hypertension variables. Those in controlled and uncontrolled groups are aware of their hypertension status, while those in the undiagnosed category are considered unaware of their status. The later undiagnosed/unaware category captures participants who may not have been diagnosed with hypertension from a medical professional and/or who may not have been told or recall their diagnosis. Table 1 Classification of participants into hypertension categories   Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension    Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension  Table 1 Classification of participants into hypertension categories   Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension    Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension  Risk factors informed by the socioecological model Individual level: chronic conditions (diabetes and obesity), sociodemographic characteristics (sex/gender, education and income sufficiency) and health behaviours (smoking, alcohol consumption and exercise). Participants were coded as having diabetes if they responded affirmatively to, ‘Has a doctor or nurse ever told you that you have diabetes, that is to say, high blood sugar levels?’ or if they were taking medication for diabetes. For obesity, a BMI of ≥30 was considered obese. Education was self-reported based on the question, ‘What is the highest level of schooling that you have completed?’ and dichotomized into post-secondary education, yes/no. Income sufficiency was self-reported according to the question, ‘To what extent does your income allow you to meet your needs?’ Participants were coded as having sufficient incomes if they responded ‘Very well’ or ‘Suitably’. Participants who reported that they currently or formally smoked were categorized as smokers. For alcohol, those who reported ever drinking were coded as consumers, all others as not consuming alcohol. Finally, the amount of exercise engaged in by the participant was estimated using a validated computer animated assessment tool called the Mobility Assessment Tool for Walking—MAT-W.22 The tool assessed walking for leisure, as well as other purposes (e.g. transportation). We dichotomized the estimate into 60 min/day, yes/no. Interpersonal: Social network ties and community integration are associated with hypertension-related outcomes.10–12 These factors were assessed according to the level of social ties—partner, children, family and friends—reported by the participant using a validated social support and social network scale.23 We dichotomized responses into high social ties, yes/no. Institutional: These questions relate to the participants’ access and utilization of health care services. We recorded whether the participant had a usual source of care based on the question, ‘Do you have a regular medical doctor or clinic for medical care? Yes/No’. We also asked, ‘How many times have you gone to the doctor in the last year?’. Finally, we recorded whether among the medications presented to the interviewer there were containers of antihypertensive medications. Community and environment: We enquired if the participant engaged in religious activities, attended community or recreational centre(s) and/or was a member of a professional association. If the participant reported any such activities, this was coded as yes. A community barrier scale was employed based on explanatory and confirmatory factor analyses on eight items related to local community from the Home And Community Environment (HACE) scale.24 Three items—parks/walking areas that are easy to access and use; safe parks/walking areas, and places to sit and rest at bus stops; parks, or other places where people walk—loaded into a single factor.25 If a participant reported lacking one or more of these community attributes, he/she was assigned a score with the sum of the items ranging between three and nine. The higher the score, the more perceived barriers. Similarly, a safety scale was developed using explanatory and confirmatory factor analyses on 10 items on perceived safety developed by Sampson and Raudenbush.26 The final perceived safety scale included eight items with possible range of 8–24.27 The higher the score, the safer the participant perceived his/her community. Statistical methods We compared hypertension diagnosis/awareness and control across factors from the socioecological model. Factors were first compared across hypertension groups using chi-square test for independent proportions. All factors with P values of ≤0.20 from the descriptive analyses were included in multinomial logistic regression models (an extension of the binary logistic model). In our analysis, non-significant factors (only 2—exercise and strolling shops/store) were removed one-by-one. Education was entered into the semi-final model based on theory. Analyses were completed using STATA 13.0 (College Station, TX). Results Most (n = 282) participants reported doctor-diagnosed hypertension, and 59% (n = 234) had clinical hypertension as measured by a validated automated device. Sixty-eight (17.3%) participants had no hypertension, 92(23.4%) had controlled and 190(48.4%) had uncontrolled hypertension and 43(10.9%) had undiagnosed hypertension (table 2). Among those with diagnosed hypertension (controlled or uncontrolled), nearly all had antihypertensive medication in their home. A notable percentage of those with no or with undiagnosed hypertension also had antihypertensive medications in their possession. Table 2 Factors corresponding to levels of the socioecological (SE) model among older adults (64–75 years) from Tirana (N = 393), in the International Mobility in Aging Study, by hypertension status Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  a Self-reported doctor diagnosed diabetes or taking diabetes medication (all medications were shown to the interviewer and recorded). b Missing three values. c Defined as Body Mass Index (BMI) of 30 or greater. d This is defined by self-report as some post-secondary schooling (e.g. university, college, technical school) or more. e 1 missing value. f This is defined by self report. Current and former smokers are categorized as yes, while never smokers are categorized as no. g Missing three values. h This is defined according to self-report. Those who report ever drinking alcohol are coded as yes, all others as no. i Missing two values. j This is defined using a computer animated assessment tool k Missing seven values l All social support measures compare high to low or none. m Missing one value. n 12 or more reported visits to the doctor per year. o Missing two values. p Six visits or fewer reported visits to the doctor per year. q Missing two values. r Upper quartile of perception of safety scale. s 34 missing values. t Most people in Tirana had a score of 9. Scores of less than 9 were categorized as low community barriers. u Missing six values. Table 2 Factors corresponding to levels of the socioecological (SE) model among older adults (64–75 years) from Tirana (N = 393), in the International Mobility in Aging Study, by hypertension status Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  a Self-reported doctor diagnosed diabetes or taking diabetes medication (all medications were shown to the interviewer and recorded). b Missing three values. c Defined as Body Mass Index (BMI) of 30 or greater. d This is defined by self-report as some post-secondary schooling (e.g. university, college, technical school) or more. e 1 missing value. f This is defined by self report. Current and former smokers are categorized as yes, while never smokers are categorized as no. g Missing three values. h This is defined according to self-report. Those who report ever drinking alcohol are coded as yes, all others as no. i Missing two values. j This is defined using a computer animated assessment tool k Missing seven values l All social support measures compare high to low or none. m Missing one value. n 12 or more reported visits to the doctor per year. o Missing two values. p Six visits or fewer reported visits to the doctor per year. q Missing two values. r Upper quartile of perception of safety scale. s 34 missing values. t Most people in Tirana had a score of 9. Scores of less than 9 were categorized as low community barriers. u Missing six values. Factors that were statistically significant (P < 0.05) across study groups in descriptive models included, at the individual sociodemographic level, income sufficiency and exercise. At the interpersonal-level, strong social ties with friends varied significantly. Reported annual doctors’ visits varied significantly at the organizational-level. Access to care was very high across all groups. At the community-level, strolling shopping areas and high-perceived neighbourhood safety varied significantly across hypertension groups. Interestingly, there was low participation in religious activities (10%), community/recreational centres (6%) and professional associations (6%). Comorbidities were associated with hypertension as expected. Results from the multivariable models (table 3) reflect those in table 2; however, strolling shopping centres and exercise were no longer significant and were not retained in models. Income insufficiency was significantly associated with uncontrolled (OR: 3.55; 95% CI: 1.95–6.48) and undiagnosed (OR: 3.25; 95% CI: 1.31–8.08) hypertension compared to controlled hypertension. Fewer annual doctors’ visits were reported among those with undiagnosed hypertension. High level of support from friends was significantly negatively associated with uncontrolled (OR: 0.43; 95% CI: 0.21–0.86) and undiagnosed (OR: 0.23; 95% CI: 0.09–0.60) hypertension compared to controlled hypertension. High perceived neighbourhood safety was associated with less uncontrolled and undiagnosed hypertension, for both comparison groups. Unexpectedly, high support from children was positively associated with uncontrolled (OR: 2.17; 95% CI: 1.09–4.33) and undiagnosed (OR: 3.55; 95% CI: 1.32–9.56) hypertension compared to no hypertension. Table 3 Multinomial regression of factors associated with hypertension (HTN) status among older adults from Tirana (N = 393), in the International Mobility in Aging Studya   No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)    No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)  Note: Results are reported as odds ratios and 95% confidence intervals in parentheses. a All models are adjusted for the participant’s age. Table 3 Multinomial regression of factors associated with hypertension (HTN) status among older adults from Tirana (N = 393), in the International Mobility in Aging Studya   No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)    No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)  Note: Results are reported as odds ratios and 95% confidence intervals in parentheses. a All models are adjusted for the participant’s age. Discussion This study demonstrates that hypertension diagnosis/awareness and control are urgent issues for older adults in Albania despite strong access to care and high health service utilization. About half of study participants had uncontrolled hypertension and 10% had undiagnosed hypertension. Social and community factors associated with hypertension awareness and control in Albania were both expected and unexpected, highlighting the importance of this theoretically guided work in this unique context. Older adults with high support from friends were significantly less likely to have uncontrolled and undiagnosed hypertension than those with lower support. Similar results have been observed in Canada, where high friend support in older adults was associated with better self-rated health.28 Results from a population-based sample of older adults in neighbouring Kosovo also reported that participants who had contact with friends had a better self-perceived health status.13 Neighbourhood-level safety was also associated with hypertension awareness and control. These findings are congruent with research reporting associations between cohesiveness and connectivity in the social milieu and health outcomes as diverse as obesity, smoking, resistance to the common cold and health care utilization.29–31 In many studies, social support, community and network ties have been associated with successful ageing; that is, ageing in the absence of chronic disease and disability, with high cognitive and functional ability and active engagement in life,32,33 but the types of social support associated with better health outcomes may vary according to social context and not all social ties may benefit health.28 While cross-sectional data cannot make causal claims, our findings, taken within the context of the broader literature on successful ageing, suggest that interventions based around social capital, particularly those that build and/or leverage friends’ social support and neighbourhood safety may have potential to improve hypertension awareness and control.30 Unexpectedly, we observed that those with high support from children were more likely to have uncontrolled and undiagnosed hypertension than those with less support. Some argue that Albania’s communist past resulted in low levels of trust directly affecting the number of associations outside the close circle of family, kin and acquaintances.34 Although under transition, Albanian society retains strong traditional influences and the family structure and ‘familism’ remain important components of society. Children’s support may be taken for granted, while building and/or maintaining other friendships may specifically represent better social ties, less isolation and more independence. In a Canadian study of older adults that observed no association between family support and self-rated health, but a positive association between friends support and better health, the authors concluded that friends may be associated with leisure activities, while family ties may lead to unwanted responsibilities and potential conflicts.28 To achieve friends’ support in Albania, an older person may need a certain capacity of social adaptation and interaction. Thus, greater ‘friends support’ may reflect more successful ageing. Another explanation for the unexpected association between high support from children and uncontrolled and undiagnosed hypertension may be that declining health generates a higher need for care. Sicker individuals may claim their children’s support based on traditional values in which this support is expected. Thus, children support in this context may not represent an indicator of successful relationships in ageing, in contrast to the friends’ support variable. While this unexpected finding highlights the importance of social variables in the unique setting of this transitional society, it also underscores the limitation of cross-sectional analyses, especially when considering aetiological associations. Further, there is a possibility of measurement error in this variable. There is limited research outside of North America and Western Europe on what constitutes social support.28 While the measures of social support employed in this study were validated in middle-income settings,28 these were not validated specifically in Albania. A key strength of our study was the consideration of community and social variables along with the more commonly studied behavioural and personal risk factors for hypertension. When social and community context were included, behavioural correlates were poor predictors of hypertension awareness and control. In particular, exercise was not predictive once community and social factors were considered. As behaviours are strongly driven by social norms, considering upstream community and social factors with public health approaches to disease prevention and control are critical. We do highlight key individual-level factors that persisted in significance after other factors were considered. Those with insufficient incomes were more likely to have uncontrolled and undiagnosed hypertension than those with sufficient incomes. Previous work with this sample only explored associations between comorbidities and behaviours with hypertension status4 and thus, missed important determinants of awareness and control that were captured by this research. In fact, the previous study statistically adjusted for income sufficiency rather than explore it as correlate of hypertension status.4 Results from this study suggest high access to care in Tirana. From a programmatic perspective, health professionals may need to manage more intensely patients with low incomes and work carefully with them to address treatment adherence barriers. Cost is a well-known barrier to medication adherence and chronic disease management.35,36 Improved doctor-patient communication and/or access to trusted, high quality health care may also be critical.35 In a related finding, those with undiagnosed hypertension had significantly fewer annual doctors’ visits than those with no hypertension, suggesting that not all patients received optimal care. Also, we found that many participants in the no hypertension and undiagnosed categories had antihypertensive medications in their possession, raising concerns about poor communication to hypertensive patients of their diagnosis. While self-reported doctor diagnosed hypertension is well-validated elsewhere,19–21 some participants in this study may have misunderstood/misreported their hypertension status to the interviewers, especially if their cognitive function was declining. If all participants with anti-hypertensive medication indeed had hypertension, the hypertension prevalence in this sample would be 91%. Data from this sample of older adults in Tirana suggest that hypertension is essentially ubiquitous, highlighting a significant need for public health and medical intervention. This study has many strengths—including in-person hypertension measurement and comprehensive social and community factor assessment—and some limitations. We study hypertension in an urban setting; other locations may have distinct patterns. Given Albania’s unique history, there may be strong cohort effects. Still, findings of distinct roles for factors within the socioecological model are likely informative to other locations, especially those also experiencing social and demographic transitions. As previously stated, the relationships from cross-sectional studies are not assumed to be causal and are susceptible to reverse causality. For example, we cannot exclude that different levels of health status affect individual, interpersonal, institutional and community factors. This study provides new insights about social and community risk factors for inadequate hypertension control and diagnosis/awareness in Albania. We provide evidence to help direct public health interventions and policy around hypertension specifically and also identify new hypotheses to consider in future research. Acknowledgement The authors would like to thank the Canadian Institutes of Health Research who provided funding support for the International Mobility in Aging Study (IMIAS). Funding The Canadian Institutes of Health Research providing funding to support the International Mobility in Aging Study (IMIAS) through an Emerging Team Grant. The Fulbright Specialist Program also supported the collaboration (7361; T.L.S.). Conflicts of interest: None declared. Key points Hypertension is the primary driver of death and disability in Albania, but hypertension control is poor and the risk factors for inadequate management are insufficiently examined in this context. The socio-ecological model can be applied to identify a diverse array of factors associated with hypertension awareness and control, thereby increasing intervention opportunities for hypertension management. In this sample of older adults, income sufficiency, ties to friends and children and neighbourhood safety were all associated with hypertension diagnosis/awareness and/or control; in contrast, no behavioural factors were related to our hypertension outcomes. Programs and policies targeting a variety of risk factors, especially in and above the individual, may result in better hypertension control than a continued focus on behaviour. References 1 Burazeri G, Bregu A, Qirjako G. National Health Report: Health Status of the Albanian Population . Tirana, Albania: Institute of Public Health, 2014. 2 Institute for Health Metrics and Evaluation (IHME), US: Albania: Coutnry profile. Available at: http://www.healthdata.org/search? search_terms=Albania ( 14 August 2017, date last accessed). 3 World Health Organization. World report on ageing and health. Geneva, Switzerland:WHO, 2015. 4 Doulougou B, Gomez F, Alvarado B, et al.   Factors associated with hypertension prevalence, awareness, treatment and control among participants in the International Mobility in Aging Study (IMIAS). J Hum Hypertens  2016; 30: 112– 9. Google Scholar CrossRef Search ADS PubMed  5 Petrela E, Burazeri G, Pupuleku F, et al.   Prevalence and correlates of hypertension in a transitional southeastern European population: results from the Albanian Demographic and Health Survey. Arh Hig Rada Toksikol  2013; 64: 479– 87. Google Scholar CrossRef Search ADS PubMed  6 Mackenbach J, McKee M (eds). Successes and Failures of Health Policy in Europe: Four Decades of Divergent Trends and Converging Challenges. The European Observatory on Health Systems and Policies . Berkshire, England: McGraw Hill Education, Open University Press, 2013. 7 Turnbull F. Blood Pressure Lowering Treatment Trialists Collaboration. Effects of different blood-pressure-lowering regimens on major cardiovascular events: results of prospectively-designed overviews of randomised trials. Lancet  2003; 362: 1527– 35. Google Scholar CrossRef Search ADS PubMed  8 Banegas JR, Lopez-Garcia E, Dallongeville J, et al.   Achievement of treatment goals for primary prevention of cardiovascular disease in clinical practice across Europe: the EURIKA study. Eur Heart J  2011; 32: 2143– 52. Google Scholar CrossRef Search ADS   9 Stokols D. Translating social ecological theory into guidelines for community health promotion. Am J Health Promot  1996; 10: 282– 98. Google Scholar CrossRef Search ADS PubMed  10 Cornwell EY, Waite LJ. Social network resources and management of hypertension. J Health Soc Behav  2012; 53: 215– 31. Google Scholar CrossRef Search ADS PubMed  11 Fratiglioni L, Paillard-Borg S, Winblad B. An active and socially integrated lifestyle in late life might protect against dementia. Lancet Neurol  2004; 3: 343– 53. Google Scholar CrossRef Search ADS PubMed  12 Jones C, Simpson SH, Mitchell D, et al.   Enhancing hypertension awareness and management in the elderly: lessons learned from the Airdrie Community Hypertension Awareness and Management Program (A-CHAMP). Can J Cardiol  2008; 24: 561– 7. Google Scholar CrossRef Search ADS PubMed  13 Jerliu N, Burazeri G, Toci E, et al.   Social networks, social participation and self-perceived health among older people in transitional Kosovo. Eur J Public Health  2014; 24: 333– 7. Google Scholar CrossRef Search ADS PubMed  14 Balcazar H, Wise S, Rosenthal EL, et al.   An ecological model using promotores de salud to prevent cardiovascular disease on the US-Mexico border: the HEART project. Prev Chronic Dis  2012; 9: E35. Google Scholar PubMed  15 Stokols D, Allen J, Bellingham RL. The social ecology of health promotion: implications for research and practice. Am J Health Promot  1996; 10: 247– 51. Google Scholar CrossRef Search ADS PubMed  16 Centers for Disease Control and Prevention, USA: Social Ecological Model. Available at: http://www.cdc.gov/cancer/crccp/sem.htm ( 27 October 2015, date last accessed). 17 Zunzunegui MV, Alvarado BE, Guerra R, et al.   The mobility gap between older men and women: the embodiment of gender. Arch Gerontol Geriatr  2015; 61: 140– 8. Google Scholar CrossRef Search ADS PubMed  18 Mier N, Ory MG, Towne SD, Smith ML. Relative association of multi-level supportive environments on poor health among older adults. Int J Environ Res Public Health  2017; 14: 387. Google Scholar CrossRef Search ADS   19 Martin LM, Leff M, Calonge N, et al.   Validation of self-reported chronic conditions and health services in a managed care population. Am J Prev Med  2000; 18: 215– 8. Google Scholar CrossRef Search ADS PubMed  20 Molenaar EA, Ameijden EJCV, Grobbee DE, Numans ME. Comparison of routine care self-reported and biometrical data on hypertension and diabetes: results of the Utrecht Health Project. Eur J Public Health  2007; 17: 199– 205. Google Scholar CrossRef Search ADS PubMed  21 Vargas CM, Burt VL, Gillum RF, Pamuk ER. Validity of self-reported hypertension in the National Health and Nutrition Examination Survey III, 1988–1991. Prev Med  1997; 26: 678– 85. Google Scholar CrossRef Search ADS PubMed  22 Marsh AP, Janssen JA, Ip EH, et al.   Assessing walking activity in older adults: development and validation of a novel computer-animated assessment tool. J Gerontol A Biol Sci Med Sci  2015; 70: 1555– 61. Google Scholar CrossRef Search ADS PubMed  23 Ahmed T, Kone GK, Vafaei A. ea Social support and social network scale, psychometric properties of a new scale for international use in older adults: the IMIAS Study [Abstract]. Gerontologist  2015; 55: 453. 24 Keysor J, Jette A, Haley S. Development of the home and community environment (HACE) instrument. J Rehabil Med  2005; 37: 37– 44. Google Scholar CrossRef Search ADS PubMed  25 Alvarado B, Kerr H, Auais M, Vafaei A. The impact of neighborhood physical environment on cognitive function of older adults. In: Canadian Association on Gerontology/Association canadienne de gérontologie  2016. Fostering Innovation in Research on Aging 45th Annual Scientific and Educational Meeting held in Montréal, Québec Canada from October 20–22, 2016. http://cagacg.ca/asem/. 26 Sampson RJ, Raudenbush SW. Seeing disorder: neighborhood stigma and the social construction of “broken windows.” Soc Psychol Q  2004; 67: 319– 42. Google Scholar CrossRef Search ADS   27 McLeod J. Neighborhood perception of safety and the mobility gap between men and women in old age . PhD diss., Kingston, Ontario, Canada: Queen's University, 2016. 28 Belanger E, Ahmed T, Vafaei A, Curcio CL, Phillips SP, Zunzunegui MV. Sources of social support associated with health and quality of life: a cross-sectional study among Canadian and Latin American older adults. BMJ  2016; 6: e011503. Google Scholar CrossRef Search ADS   29 House JS, Landis KR, Umberson D. Social relationships and health. Science  1988; 241: 540– 5. Google Scholar CrossRef Search ADS PubMed  30 Lomas J. Social capital and health: implications for public health and epidemiology. Soc Sci Med  1998; 47: 1181– 8. Google Scholar CrossRef Search ADS PubMed  31 Smith KP, Christakis NA. Social networks and health. Annu Rev Soc  2008; 34: 405– 29. Google Scholar CrossRef Search ADS   32 Seeman TE, Lusignolo TM, Albert M, Berkman L. Social relationships, social support, and patterns of cognitive aging in healthy, high-functioning older adults: MacArthur studies of successful aging. Health Psychol  2001; 20: 243– 55. Google Scholar CrossRef Search ADS PubMed  33 Unger JB, McAvay G, Bruce ML, et al.   Variation in the impact of social network characteristics on physical functioning in elderly persons: macarthur studies of successful aging. J Gerontol B Psychol Sci Soc Sci  1999; 54B: S245– 51. Google Scholar CrossRef Search ADS   34 Shaqiri A. The Economic Well-Being of Albanians: Trust and Social Capital: How Do Cultural Values and the Levels of Trust Affect Social Capital and the Robustness of Economic Performances in the Albanian Societies ?. Saarbrucken, Germany: VDM Verlag Dr. Müller, 2008. 35 Gellad WF, Grenard JL, Marcum ZA. A systematic review of barriers to medication adherence in the elderly: looking beyond cost and regimen complexity. Am J Geriatr Pharmacother  2011; 9: 11– 23. Google Scholar CrossRef Search ADS PubMed  36 Njie GJ, Finnie RKC, Acharya SD, et al.   Reducing medication costs to prevent cardiovascular disease: a community guide systematic review. Prev Chronic Dis  2015; 12: E208. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

Social and community factors associated with hypertension awareness and control among older adults in Tirana, Albania

Loading next page...
 
/lp/ou_press/social-and-community-factors-associated-with-hypertension-awareness-AIIcu92UXc
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/cky036
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Determinants of hypertension diagnosis and/or awareness and control among older adults are understudied in Albania, a former communist country in South Eastern Europe, which is experiencing rapid demographic, socioeconomic and epidemiological transition. This paper examines the association of individual, interpersonal, organizational and community factors with hypertension awareness and control among older adults in Tirana, the Albanian capital. Methods Using 2012 International Mobility in Aging Study data on older adults from Albania’s capital city (n = 393) and the socioecological model as a conceptual framework, multinomial regression models identified factors associated with controlled, uncontrolled and undiagnosed hypertension. Results For hypertension, 17.3% participants had none, 23.4% were controlled, 48.4% were uncontrolled and 10.9% were undiagnosed/unaware. Compared to those with controlled hypertension, in multivariable models, a high level of friend support was negatively associated with uncontrolled (OR: 0.4; 95% CI: 0.2–0.9) and undiagnosed (OR: 0.2; 95% CI: 0.1–0.6) hypertension. A high level of perceived neighbourhood safety was negatively associated with uncontrolled (OR: 0.6; 95% CI: 0.3–1.0) and undiagnosed (OR: 0.4; 95% CI: 0.2–1.0) hypertension. Compared to those with no hypertension, children’s social support was positively associated with uncontrolled (OR: 2.2; 95% CI: 1.1–4.3) and undiagnosed (OR: 3.6; 95% CI: 1.3–9.6) hypertension. Conclusion This study provides new insights about distinct risk factors for inadequate hypertension management in Albania. It highlights the importance of community-level factors (safety) and interpersonal factors (family and friend ties) to hypertension diagnosis/awareness and control, which may provide novel intervention opportunities for hypertension programs. Introduction Non-communicable diseases (NCDs) account for 89% of deaths in Albania; cardiovascular diseases (CVD) alone account for 59% of total mortality.1 Albania, historically the most isolated former communist country in the Western Balkans, is currently experiencing swift demographic and socioeconomic transitions, including a rapid epidemiological shift from infectious diseases to NCDs.1 Between 1990 and 2010, the mortality rate from ischemic heart disease increased twofold in Albanian men and women.1,2 Mortality from ischemic heart disease in Albania is the highest in the South Eastern European region.1,2 Hypertension, a major CVD risk factor, is highly prevalent in Albania, especially among older adults whose proportion in the population is steadily increasing in line with global trends.1,3 Hypertension is the primary driver of death and disability in Albania.2 In the past two decades, the mortality rate from ischemic heart disease attributable to hypertension doubled.1 Nevertheless, hypertension awareness—e.g. knowledge that one has hypertension, which is predicated on diagnosis and adequate comprehension of that diagnosis—and control—e.g. successfully managing hypertension once diagnosed—are notably low in Albania,4,5 even though adequate hypertension control reduces the risk of CVD dramatically.6,7 Achieving hypertension control is an important health system goal given the opportunities for reduced disability and premature death. While average blood pressure values have declined in nearly all European countries since 1980, Albania has experienced an increase in mean systolic blood pressure over the same timeframe.6 There is wide acceptance that adherence to modern guidelines for the detection/treatment of hypertension has contributed to reduced hypertension levels in Western Europe.6 Yet, even in Western Europe, control among diagnosed patients appears elusive, with 30–50% of patients achieving hypertension control.8 There is a clear need to better understand risk factors for inadequate hypertension awareness and poor control, especially in Albania. Research on correlates of hypertension prevalence and control in Albania is largely limited to individual and behavioural predictors.4,5 Behaviours are influenced by social norms9 and other factors such as interpersonal relationships and community resources may also influence hypertension awareness and control. For example, social network ties and community integration are associated with hypertension-related outcomes elsewhere,10–12 but understudied in Albania. Yet, social factors and community connections have been demonstrated as important to health outcomes in the Balkan region.13 To better understand barriers to hypertension awareness and control in Albania, it is necessary to characterize broadly risk factors for hypertension from the individual to community, particularly given Albania’s unique political and social history. This approach can support the successful targeting and design of impactful public health programs.14,15 Guided by the socioecological model,16 this paper examines the association of factors, ranging from individual characteristics to social and community factors, with hypertension awareness and control in Albania. Our objective is to quantify the contribution of a more diverse set of predictive factors of poor hypertension awareness and control in Albania than previously studied. We are particularly interested in the contribution of individual-level risk factors when social and community factors are also considered in predictive models. Methods Data source and study population This study analyses 2012 baseline data from The International Mobility in Aging Study (IMIAS), which is a population-based longitudinal study conducted in five sites globally, including Tirana (700 000 inhabitants), the capital city of Albania.17 A detailed description of the study, including information about the study sites and cohort is available elsewhere.17 IMIAS randomly recruited equal numbers of community-dwelling men and women, aged 65–74 years, from the population registered at neighbourhood health centres in Tirana (N = 393). Over 90% of older adults in Tirana are registered at a health centre and the acceptance rate was 90%. Those with cognitive impairment were excluded from the study (one person in Tirana).17 Informed consent was obtained from all individual participants. Interviews were conducted at participants’ homes and administered by interviewers trained with standardized protocols. Additional procedures included taking the participants’ blood pressure and anthropometric measures, as well as a blood draw.17 Participants were also asked to bring the containers of all medications they were taking. The interviewer then recorded the medication names and a trained pharmacist coded these according to their approved uses.17 Conceptual framework In the socioecological model, individual-level health outcomes, knowledge and behaviours are influenced by interpersonal (e.g. family, friends), organizational (e.g. health system), community (e.g. neighbourhood safety) and policy levels.16 The socioecological model is well-used for studies of many health outcomes, including the understanding of factors predicting chronic disease.14 This model is particularly relevant to our work, as research from diverse locations demonstrate that social and environmental factors are strong determinants of older adults health.18 Importantly, the well-recognized role of social and community context in contributing to chronic disease among older adults is understudied in South Eastern Europe. Measure of hypertension awareness and control Blood pressure was measured three times with a validated automated blood pressure device after 5 min of rest. The mean value of the second and third blood pressure measurements was used in the analysis. Clinical hypertension was defined by a measured blood pressure of ≥140/90. Self-reported doctor-diagnosed hypertension was recorded based on an affirmative response to the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’. Self-reported hypertension has been validated in a number of studies and is highly specific (e.g. low probability that non-hypertensive persons are classified as hypertensive).19–21 We compared (i) controlled (no clinical hypertension, but self-reported hypertension), (ii) uncontrolled (clinical hypertension and self-reported hypertension) and (iii) undiagnosed/unaware of hypertension (clinical hypertension, but no self-reported hypertension), with (iv) no hypertension across factors from the socioecological model (described below). Table 1 depicts our classification of participants into hypertension categories based on the clinical and self-reported hypertension variables. Those in controlled and uncontrolled groups are aware of their hypertension status, while those in the undiagnosed category are considered unaware of their status. The later undiagnosed/unaware category captures participants who may not have been diagnosed with hypertension from a medical professional and/or who may not have been told or recall their diagnosis. Table 1 Classification of participants into hypertension categories   Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension    Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension  Table 1 Classification of participants into hypertension categories   Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension    Clinical hypertension, defined as a measured blood pressure value of ≥140/90 mm/Hg   Self-reported hypertension, based on the question, ‘Has a doctor or nurse ever told you that you have high blood pressure or hypertension’    Yes  No  Yes  Uncontrolled hypertension  Controlled hypertension  No  Undiagnosed/unaware of hypertension  No hypertension  Risk factors informed by the socioecological model Individual level: chronic conditions (diabetes and obesity), sociodemographic characteristics (sex/gender, education and income sufficiency) and health behaviours (smoking, alcohol consumption and exercise). Participants were coded as having diabetes if they responded affirmatively to, ‘Has a doctor or nurse ever told you that you have diabetes, that is to say, high blood sugar levels?’ or if they were taking medication for diabetes. For obesity, a BMI of ≥30 was considered obese. Education was self-reported based on the question, ‘What is the highest level of schooling that you have completed?’ and dichotomized into post-secondary education, yes/no. Income sufficiency was self-reported according to the question, ‘To what extent does your income allow you to meet your needs?’ Participants were coded as having sufficient incomes if they responded ‘Very well’ or ‘Suitably’. Participants who reported that they currently or formally smoked were categorized as smokers. For alcohol, those who reported ever drinking were coded as consumers, all others as not consuming alcohol. Finally, the amount of exercise engaged in by the participant was estimated using a validated computer animated assessment tool called the Mobility Assessment Tool for Walking—MAT-W.22 The tool assessed walking for leisure, as well as other purposes (e.g. transportation). We dichotomized the estimate into 60 min/day, yes/no. Interpersonal: Social network ties and community integration are associated with hypertension-related outcomes.10–12 These factors were assessed according to the level of social ties—partner, children, family and friends—reported by the participant using a validated social support and social network scale.23 We dichotomized responses into high social ties, yes/no. Institutional: These questions relate to the participants’ access and utilization of health care services. We recorded whether the participant had a usual source of care based on the question, ‘Do you have a regular medical doctor or clinic for medical care? Yes/No’. We also asked, ‘How many times have you gone to the doctor in the last year?’. Finally, we recorded whether among the medications presented to the interviewer there were containers of antihypertensive medications. Community and environment: We enquired if the participant engaged in religious activities, attended community or recreational centre(s) and/or was a member of a professional association. If the participant reported any such activities, this was coded as yes. A community barrier scale was employed based on explanatory and confirmatory factor analyses on eight items related to local community from the Home And Community Environment (HACE) scale.24 Three items—parks/walking areas that are easy to access and use; safe parks/walking areas, and places to sit and rest at bus stops; parks, or other places where people walk—loaded into a single factor.25 If a participant reported lacking one or more of these community attributes, he/she was assigned a score with the sum of the items ranging between three and nine. The higher the score, the more perceived barriers. Similarly, a safety scale was developed using explanatory and confirmatory factor analyses on 10 items on perceived safety developed by Sampson and Raudenbush.26 The final perceived safety scale included eight items with possible range of 8–24.27 The higher the score, the safer the participant perceived his/her community. Statistical methods We compared hypertension diagnosis/awareness and control across factors from the socioecological model. Factors were first compared across hypertension groups using chi-square test for independent proportions. All factors with P values of ≤0.20 from the descriptive analyses were included in multinomial logistic regression models (an extension of the binary logistic model). In our analysis, non-significant factors (only 2—exercise and strolling shops/store) were removed one-by-one. Education was entered into the semi-final model based on theory. Analyses were completed using STATA 13.0 (College Station, TX). Results Most (n = 282) participants reported doctor-diagnosed hypertension, and 59% (n = 234) had clinical hypertension as measured by a validated automated device. Sixty-eight (17.3%) participants had no hypertension, 92(23.4%) had controlled and 190(48.4%) had uncontrolled hypertension and 43(10.9%) had undiagnosed hypertension (table 2). Among those with diagnosed hypertension (controlled or uncontrolled), nearly all had antihypertensive medication in their home. A notable percentage of those with no or with undiagnosed hypertension also had antihypertensive medications in their possession. Table 2 Factors corresponding to levels of the socioecological (SE) model among older adults (64–75 years) from Tirana (N = 393), in the International Mobility in Aging Study, by hypertension status Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  a Self-reported doctor diagnosed diabetes or taking diabetes medication (all medications were shown to the interviewer and recorded). b Missing three values. c Defined as Body Mass Index (BMI) of 30 or greater. d This is defined by self-report as some post-secondary schooling (e.g. university, college, technical school) or more. e 1 missing value. f This is defined by self report. Current and former smokers are categorized as yes, while never smokers are categorized as no. g Missing three values. h This is defined according to self-report. Those who report ever drinking alcohol are coded as yes, all others as no. i Missing two values. j This is defined using a computer animated assessment tool k Missing seven values l All social support measures compare high to low or none. m Missing one value. n 12 or more reported visits to the doctor per year. o Missing two values. p Six visits or fewer reported visits to the doctor per year. q Missing two values. r Upper quartile of perception of safety scale. s 34 missing values. t Most people in Tirana had a score of 9. Scores of less than 9 were categorized as low community barriers. u Missing six values. Table 2 Factors corresponding to levels of the socioecological (SE) model among older adults (64–75 years) from Tirana (N = 393), in the International Mobility in Aging Study, by hypertension status Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  Level of the SE model  Variable name  No hypertension   Controlled hypertension   Uncontrolled hypertension   Undiagnosed/unaware of hypertension   P value  N (Overall %)    68(17.3)  92(23.4)  190(48.4)  43(10.9)        n (%)  n (%)  n (%)  n (%)    Individual-level  Comorbidities    Diabetesa,b              Yes  11 (16.2%)  25 (27.2%)  66 (35.1%)  11 (26.2%)  0.03    Obesityc              Yes  14 (20.6%)  39 (42.4%)  74 (39.0%)  14 (32.6%)  0.02  Socio-demographics    Gender              Female  30 (44.1%)  54 (58.7%)  99 (52.1%)  24 (55.8%)  0.32    Education              Post-secondaryd  47 (69.1%)  55 (59.8%)  119 (62.6%)  23 (53.5%)  0.39    Perceived income insufficiencye              Yes  41 (60.3%)  41 (44.6%)  133 (70.4%)  29 (67.4%)  <0.01  Behavioural    Smokef,g              Yes  25 (37.3%)  37 (40.2%)  78 (41.5%)  14 (32.6%)  0.72    Alcoholh,i              Yes  42 (61.8%)  50 (54.4%)  111 (59.0%)  31 (72.1%)  0.26    Exercise, 60min/dayj,k              Yes  25 (37.3%)  20 (22.2%)  40 (21.5%)  18 (41.9%)  0.01  Interpersonal    Social ties: partnerl              High  37 (54.4%)  43 (46.7%)  101 (53.2%)  25 (58.1%)  0.59    Social ties: children              High  42 (61.8%)  71 (77.2%)  130 (68.4%)  31 (72.1%)  0.19    Social ties: familym              High  44 (64.7%)  68 (74.7%)  125 (65.8%)  30 (69.8%)  0.44    Social ties: friends              High  46 (67.7%)  68 (73.9%)  120 (63.2%)  21 (48.8%)  0.04  Institutional    Usual Source of Care              Yes  68 (100%)  92 (100%)  188 (100%)  41 (95.4%)  <0.01    Monthly average doctor visitsn,o              Yes  30 (44.1%)  58 (63.0%)  131 (69.7%)  14 (32.6%)  <0.01    Lower quartile of visits to doctorp,q              Yes  33 (48.5%)  13 (14.1%)  36 (19.2%)  25 (58.1%)  <0.01    Has antihypertensive medication              Yes  36 (52.9%)  90 (97.8%)  182 (95.8%)  19 (44.2%)  <0.01  Community    Participate in Religious Activities              Yes  7 (10.3%)  12 (13.0%)  16 (8.4%)  6 (14.0%)  0.56    Attend community/recreation centre              Yes  5 (7.4%)  7 (7.6%)  11 (5.8%)  1 (2.4%)  0.66    Member of professional association              Yes  7 (10.5%)  7 (7.6%)  11 (5.8%)  0 (0.0%)  0.16    Stroll shops/stores              Yes  36 (52.9%)  68 (73.9%)  99 (52.1%)  19 (44.2%)  <0.01    Perception of safetyr,s              High  37 (55.2%)  37 (48.1%)  66 (37.7%)  12 (30.0%)  0.02    Community barrierst,u              Low  9 (13.6%)  13 (14.3%)  35 (18.6%)  4 (9.5%)  0.43  a Self-reported doctor diagnosed diabetes or taking diabetes medication (all medications were shown to the interviewer and recorded). b Missing three values. c Defined as Body Mass Index (BMI) of 30 or greater. d This is defined by self-report as some post-secondary schooling (e.g. university, college, technical school) or more. e 1 missing value. f This is defined by self report. Current and former smokers are categorized as yes, while never smokers are categorized as no. g Missing three values. h This is defined according to self-report. Those who report ever drinking alcohol are coded as yes, all others as no. i Missing two values. j This is defined using a computer animated assessment tool k Missing seven values l All social support measures compare high to low or none. m Missing one value. n 12 or more reported visits to the doctor per year. o Missing two values. p Six visits or fewer reported visits to the doctor per year. q Missing two values. r Upper quartile of perception of safety scale. s 34 missing values. t Most people in Tirana had a score of 9. Scores of less than 9 were categorized as low community barriers. u Missing six values. Factors that were statistically significant (P < 0.05) across study groups in descriptive models included, at the individual sociodemographic level, income sufficiency and exercise. At the interpersonal-level, strong social ties with friends varied significantly. Reported annual doctors’ visits varied significantly at the organizational-level. Access to care was very high across all groups. At the community-level, strolling shopping areas and high-perceived neighbourhood safety varied significantly across hypertension groups. Interestingly, there was low participation in religious activities (10%), community/recreational centres (6%) and professional associations (6%). Comorbidities were associated with hypertension as expected. Results from the multivariable models (table 3) reflect those in table 2; however, strolling shopping centres and exercise were no longer significant and were not retained in models. Income insufficiency was significantly associated with uncontrolled (OR: 3.55; 95% CI: 1.95–6.48) and undiagnosed (OR: 3.25; 95% CI: 1.31–8.08) hypertension compared to controlled hypertension. Fewer annual doctors’ visits were reported among those with undiagnosed hypertension. High level of support from friends was significantly negatively associated with uncontrolled (OR: 0.43; 95% CI: 0.21–0.86) and undiagnosed (OR: 0.23; 95% CI: 0.09–0.60) hypertension compared to controlled hypertension. High perceived neighbourhood safety was associated with less uncontrolled and undiagnosed hypertension, for both comparison groups. Unexpectedly, high support from children was positively associated with uncontrolled (OR: 2.17; 95% CI: 1.09–4.33) and undiagnosed (OR: 3.55; 95% CI: 1.32–9.56) hypertension compared to no hypertension. Table 3 Multinomial regression of factors associated with hypertension (HTN) status among older adults from Tirana (N = 393), in the International Mobility in Aging Studya   No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)    No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)  Note: Results are reported as odds ratios and 95% confidence intervals in parentheses. a All models are adjusted for the participant’s age. Table 3 Multinomial regression of factors associated with hypertension (HTN) status among older adults from Tirana (N = 393), in the International Mobility in Aging Studya   No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)    No HTN = REF   Controlled HTN = REF   Controlled HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Uncontrolled HTN  Undiagnosed/Unaware of HTN  Individual-level and comorbidity      Diabetes  1.36 (0.57–3.24)  2.09 (0.97–4.52)  2.11 (0.74–6.00)  1.54 (0.82–2.91)  1.56 (0.58–4.14)      Obesity  2.19 (0.99–4.86)  2.23 (1.09–4.55)  1.69 (0.64–4.43)  1.02 (0.56–1.83)  0.77 (0.31–1.91)  Individual-level socio-demographics      Post-secondary education  0.50 (0.23–1.09)  0.75 (0.38–1.48)  0.50 (0.20–1.23)  1.50 (0.81–2.75)  1.00 (0.41–2.41)      Insufficient income  0.47 (0.23–1.00)  1.69 (0.87–3.27)  1.54 (0.62–3.85)  3.55 (1.95–6.48)  3.25 (1.31–8.08)  Interpersonal-level      High support from children  2.14 (0.94–4.87)  2.17 (1.09–4.33)  3.55 (1.32–9.56)  1.02 (0.50–2.06)  1.66 (0.59–4.70)      High support from friends  1.44 (0.62–3.36)  0.62 (0.31–1.24)  0.33 (0.13–0.84)  0.43 (0.21–0.86)  0.23 (0.09–0.60)  Organizational-level      # visits to doctor/year  1.15 (1.07–1.23)  1.13 (1.06–1.20)  0.95 (0.87–1.03)  0.98 (0.92–1.04)  0.83 (0.76–0.90)  Community-level      High perceived safety  0.82 (0.40–1.67)  0.46 (0.25–0.86)  0.32 (0.14–0.78)  0.56 (0.31–1.01)  0.40 (0.16–0.96)  Note: Results are reported as odds ratios and 95% confidence intervals in parentheses. a All models are adjusted for the participant’s age. Discussion This study demonstrates that hypertension diagnosis/awareness and control are urgent issues for older adults in Albania despite strong access to care and high health service utilization. About half of study participants had uncontrolled hypertension and 10% had undiagnosed hypertension. Social and community factors associated with hypertension awareness and control in Albania were both expected and unexpected, highlighting the importance of this theoretically guided work in this unique context. Older adults with high support from friends were significantly less likely to have uncontrolled and undiagnosed hypertension than those with lower support. Similar results have been observed in Canada, where high friend support in older adults was associated with better self-rated health.28 Results from a population-based sample of older adults in neighbouring Kosovo also reported that participants who had contact with friends had a better self-perceived health status.13 Neighbourhood-level safety was also associated with hypertension awareness and control. These findings are congruent with research reporting associations between cohesiveness and connectivity in the social milieu and health outcomes as diverse as obesity, smoking, resistance to the common cold and health care utilization.29–31 In many studies, social support, community and network ties have been associated with successful ageing; that is, ageing in the absence of chronic disease and disability, with high cognitive and functional ability and active engagement in life,32,33 but the types of social support associated with better health outcomes may vary according to social context and not all social ties may benefit health.28 While cross-sectional data cannot make causal claims, our findings, taken within the context of the broader literature on successful ageing, suggest that interventions based around social capital, particularly those that build and/or leverage friends’ social support and neighbourhood safety may have potential to improve hypertension awareness and control.30 Unexpectedly, we observed that those with high support from children were more likely to have uncontrolled and undiagnosed hypertension than those with less support. Some argue that Albania’s communist past resulted in low levels of trust directly affecting the number of associations outside the close circle of family, kin and acquaintances.34 Although under transition, Albanian society retains strong traditional influences and the family structure and ‘familism’ remain important components of society. Children’s support may be taken for granted, while building and/or maintaining other friendships may specifically represent better social ties, less isolation and more independence. In a Canadian study of older adults that observed no association between family support and self-rated health, but a positive association between friends support and better health, the authors concluded that friends may be associated with leisure activities, while family ties may lead to unwanted responsibilities and potential conflicts.28 To achieve friends’ support in Albania, an older person may need a certain capacity of social adaptation and interaction. Thus, greater ‘friends support’ may reflect more successful ageing. Another explanation for the unexpected association between high support from children and uncontrolled and undiagnosed hypertension may be that declining health generates a higher need for care. Sicker individuals may claim their children’s support based on traditional values in which this support is expected. Thus, children support in this context may not represent an indicator of successful relationships in ageing, in contrast to the friends’ support variable. While this unexpected finding highlights the importance of social variables in the unique setting of this transitional society, it also underscores the limitation of cross-sectional analyses, especially when considering aetiological associations. Further, there is a possibility of measurement error in this variable. There is limited research outside of North America and Western Europe on what constitutes social support.28 While the measures of social support employed in this study were validated in middle-income settings,28 these were not validated specifically in Albania. A key strength of our study was the consideration of community and social variables along with the more commonly studied behavioural and personal risk factors for hypertension. When social and community context were included, behavioural correlates were poor predictors of hypertension awareness and control. In particular, exercise was not predictive once community and social factors were considered. As behaviours are strongly driven by social norms, considering upstream community and social factors with public health approaches to disease prevention and control are critical. We do highlight key individual-level factors that persisted in significance after other factors were considered. Those with insufficient incomes were more likely to have uncontrolled and undiagnosed hypertension than those with sufficient incomes. Previous work with this sample only explored associations between comorbidities and behaviours with hypertension status4 and thus, missed important determinants of awareness and control that were captured by this research. In fact, the previous study statistically adjusted for income sufficiency rather than explore it as correlate of hypertension status.4 Results from this study suggest high access to care in Tirana. From a programmatic perspective, health professionals may need to manage more intensely patients with low incomes and work carefully with them to address treatment adherence barriers. Cost is a well-known barrier to medication adherence and chronic disease management.35,36 Improved doctor-patient communication and/or access to trusted, high quality health care may also be critical.35 In a related finding, those with undiagnosed hypertension had significantly fewer annual doctors’ visits than those with no hypertension, suggesting that not all patients received optimal care. Also, we found that many participants in the no hypertension and undiagnosed categories had antihypertensive medications in their possession, raising concerns about poor communication to hypertensive patients of their diagnosis. While self-reported doctor diagnosed hypertension is well-validated elsewhere,19–21 some participants in this study may have misunderstood/misreported their hypertension status to the interviewers, especially if their cognitive function was declining. If all participants with anti-hypertensive medication indeed had hypertension, the hypertension prevalence in this sample would be 91%. Data from this sample of older adults in Tirana suggest that hypertension is essentially ubiquitous, highlighting a significant need for public health and medical intervention. This study has many strengths—including in-person hypertension measurement and comprehensive social and community factor assessment—and some limitations. We study hypertension in an urban setting; other locations may have distinct patterns. Given Albania’s unique history, there may be strong cohort effects. Still, findings of distinct roles for factors within the socioecological model are likely informative to other locations, especially those also experiencing social and demographic transitions. As previously stated, the relationships from cross-sectional studies are not assumed to be causal and are susceptible to reverse causality. For example, we cannot exclude that different levels of health status affect individual, interpersonal, institutional and community factors. This study provides new insights about social and community risk factors for inadequate hypertension control and diagnosis/awareness in Albania. We provide evidence to help direct public health interventions and policy around hypertension specifically and also identify new hypotheses to consider in future research. Acknowledgement The authors would like to thank the Canadian Institutes of Health Research who provided funding support for the International Mobility in Aging Study (IMIAS). Funding The Canadian Institutes of Health Research providing funding to support the International Mobility in Aging Study (IMIAS) through an Emerging Team Grant. The Fulbright Specialist Program also supported the collaboration (7361; T.L.S.). Conflicts of interest: None declared. Key points Hypertension is the primary driver of death and disability in Albania, but hypertension control is poor and the risk factors for inadequate management are insufficiently examined in this context. The socio-ecological model can be applied to identify a diverse array of factors associated with hypertension awareness and control, thereby increasing intervention opportunities for hypertension management. In this sample of older adults, income sufficiency, ties to friends and children and neighbourhood safety were all associated with hypertension diagnosis/awareness and/or control; in contrast, no behavioural factors were related to our hypertension outcomes. Programs and policies targeting a variety of risk factors, especially in and above the individual, may result in better hypertension control than a continued focus on behaviour. References 1 Burazeri G, Bregu A, Qirjako G. National Health Report: Health Status of the Albanian Population . Tirana, Albania: Institute of Public Health, 2014. 2 Institute for Health Metrics and Evaluation (IHME), US: Albania: Coutnry profile. Available at: http://www.healthdata.org/search? search_terms=Albania ( 14 August 2017, date last accessed). 3 World Health Organization. World report on ageing and health. Geneva, Switzerland:WHO, 2015. 4 Doulougou B, Gomez F, Alvarado B, et al.   Factors associated with hypertension prevalence, awareness, treatment and control among participants in the International Mobility in Aging Study (IMIAS). J Hum Hypertens  2016; 30: 112– 9. Google Scholar CrossRef Search ADS PubMed  5 Petrela E, Burazeri G, Pupuleku F, et al.   Prevalence and correlates of hypertension in a transitional southeastern European population: results from the Albanian Demographic and Health Survey. Arh Hig Rada Toksikol  2013; 64: 479– 87. Google Scholar CrossRef Search ADS PubMed  6 Mackenbach J, McKee M (eds). Successes and Failures of Health Policy in Europe: Four Decades of Divergent Trends and Converging Challenges. The European Observatory on Health Systems and Policies . Berkshire, England: McGraw Hill Education, Open University Press, 2013. 7 Turnbull F. Blood Pressure Lowering Treatment Trialists Collaboration. Effects of different blood-pressure-lowering regimens on major cardiovascular events: results of prospectively-designed overviews of randomised trials. Lancet  2003; 362: 1527– 35. Google Scholar CrossRef Search ADS PubMed  8 Banegas JR, Lopez-Garcia E, Dallongeville J, et al.   Achievement of treatment goals for primary prevention of cardiovascular disease in clinical practice across Europe: the EURIKA study. Eur Heart J  2011; 32: 2143– 52. Google Scholar CrossRef Search ADS   9 Stokols D. Translating social ecological theory into guidelines for community health promotion. Am J Health Promot  1996; 10: 282– 98. Google Scholar CrossRef Search ADS PubMed  10 Cornwell EY, Waite LJ. Social network resources and management of hypertension. J Health Soc Behav  2012; 53: 215– 31. Google Scholar CrossRef Search ADS PubMed  11 Fratiglioni L, Paillard-Borg S, Winblad B. An active and socially integrated lifestyle in late life might protect against dementia. Lancet Neurol  2004; 3: 343– 53. Google Scholar CrossRef Search ADS PubMed  12 Jones C, Simpson SH, Mitchell D, et al.   Enhancing hypertension awareness and management in the elderly: lessons learned from the Airdrie Community Hypertension Awareness and Management Program (A-CHAMP). Can J Cardiol  2008; 24: 561– 7. Google Scholar CrossRef Search ADS PubMed  13 Jerliu N, Burazeri G, Toci E, et al.   Social networks, social participation and self-perceived health among older people in transitional Kosovo. Eur J Public Health  2014; 24: 333– 7. Google Scholar CrossRef Search ADS PubMed  14 Balcazar H, Wise S, Rosenthal EL, et al.   An ecological model using promotores de salud to prevent cardiovascular disease on the US-Mexico border: the HEART project. Prev Chronic Dis  2012; 9: E35. Google Scholar PubMed  15 Stokols D, Allen J, Bellingham RL. The social ecology of health promotion: implications for research and practice. Am J Health Promot  1996; 10: 247– 51. Google Scholar CrossRef Search ADS PubMed  16 Centers for Disease Control and Prevention, USA: Social Ecological Model. Available at: http://www.cdc.gov/cancer/crccp/sem.htm ( 27 October 2015, date last accessed). 17 Zunzunegui MV, Alvarado BE, Guerra R, et al.   The mobility gap between older men and women: the embodiment of gender. Arch Gerontol Geriatr  2015; 61: 140– 8. Google Scholar CrossRef Search ADS PubMed  18 Mier N, Ory MG, Towne SD, Smith ML. Relative association of multi-level supportive environments on poor health among older adults. Int J Environ Res Public Health  2017; 14: 387. Google Scholar CrossRef Search ADS   19 Martin LM, Leff M, Calonge N, et al.   Validation of self-reported chronic conditions and health services in a managed care population. Am J Prev Med  2000; 18: 215– 8. Google Scholar CrossRef Search ADS PubMed  20 Molenaar EA, Ameijden EJCV, Grobbee DE, Numans ME. Comparison of routine care self-reported and biometrical data on hypertension and diabetes: results of the Utrecht Health Project. Eur J Public Health  2007; 17: 199– 205. Google Scholar CrossRef Search ADS PubMed  21 Vargas CM, Burt VL, Gillum RF, Pamuk ER. Validity of self-reported hypertension in the National Health and Nutrition Examination Survey III, 1988–1991. Prev Med  1997; 26: 678– 85. Google Scholar CrossRef Search ADS PubMed  22 Marsh AP, Janssen JA, Ip EH, et al.   Assessing walking activity in older adults: development and validation of a novel computer-animated assessment tool. J Gerontol A Biol Sci Med Sci  2015; 70: 1555– 61. Google Scholar CrossRef Search ADS PubMed  23 Ahmed T, Kone GK, Vafaei A. ea Social support and social network scale, psychometric properties of a new scale for international use in older adults: the IMIAS Study [Abstract]. Gerontologist  2015; 55: 453. 24 Keysor J, Jette A, Haley S. Development of the home and community environment (HACE) instrument. J Rehabil Med  2005; 37: 37– 44. Google Scholar CrossRef Search ADS PubMed  25 Alvarado B, Kerr H, Auais M, Vafaei A. The impact of neighborhood physical environment on cognitive function of older adults. In: Canadian Association on Gerontology/Association canadienne de gérontologie  2016. Fostering Innovation in Research on Aging 45th Annual Scientific and Educational Meeting held in Montréal, Québec Canada from October 20–22, 2016. http://cagacg.ca/asem/. 26 Sampson RJ, Raudenbush SW. Seeing disorder: neighborhood stigma and the social construction of “broken windows.” Soc Psychol Q  2004; 67: 319– 42. Google Scholar CrossRef Search ADS   27 McLeod J. Neighborhood perception of safety and the mobility gap between men and women in old age . PhD diss., Kingston, Ontario, Canada: Queen's University, 2016. 28 Belanger E, Ahmed T, Vafaei A, Curcio CL, Phillips SP, Zunzunegui MV. Sources of social support associated with health and quality of life: a cross-sectional study among Canadian and Latin American older adults. BMJ  2016; 6: e011503. Google Scholar CrossRef Search ADS   29 House JS, Landis KR, Umberson D. Social relationships and health. Science  1988; 241: 540– 5. Google Scholar CrossRef Search ADS PubMed  30 Lomas J. Social capital and health: implications for public health and epidemiology. Soc Sci Med  1998; 47: 1181– 8. Google Scholar CrossRef Search ADS PubMed  31 Smith KP, Christakis NA. Social networks and health. Annu Rev Soc  2008; 34: 405– 29. Google Scholar CrossRef Search ADS   32 Seeman TE, Lusignolo TM, Albert M, Berkman L. Social relationships, social support, and patterns of cognitive aging in healthy, high-functioning older adults: MacArthur studies of successful aging. Health Psychol  2001; 20: 243– 55. Google Scholar CrossRef Search ADS PubMed  33 Unger JB, McAvay G, Bruce ML, et al.   Variation in the impact of social network characteristics on physical functioning in elderly persons: macarthur studies of successful aging. J Gerontol B Psychol Sci Soc Sci  1999; 54B: S245– 51. Google Scholar CrossRef Search ADS   34 Shaqiri A. The Economic Well-Being of Albanians: Trust and Social Capital: How Do Cultural Values and the Levels of Trust Affect Social Capital and the Robustness of Economic Performances in the Albanian Societies ?. Saarbrucken, Germany: VDM Verlag Dr. Müller, 2008. 35 Gellad WF, Grenard JL, Marcum ZA. A systematic review of barriers to medication adherence in the elderly: looking beyond cost and regimen complexity. Am J Geriatr Pharmacother  2011; 9: 11– 23. Google Scholar CrossRef Search ADS PubMed  36 Njie GJ, Finnie RKC, Acharya SD, et al.   Reducing medication costs to prevent cardiovascular disease: a community guide systematic review. Prev Chronic Dis  2015; 12: E208. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The European Journal of Public HealthOxford University Press

Published: Mar 20, 2018

There are no references for this article.

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


DeepDyve is your
personal research library

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

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

All for just $49/month

Explore the DeepDyve Library

Search

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

Organize

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

Access

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

Your journals are on DeepDyve

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

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off