Sample
This cross-sectional study was conducted between August 2014 and June 2016 in a primary care clinic which is part of the National Healthcare Group (NHG), serving the northern part of Singapore with an average daily attendance of about 1,400 patients. The primary care clinics referred to as ‘polyclinics’ in Singapore provide a comprehensive range of health services, such as providing treatment for acute medical conditions, management of chronic diseases, women and child health services, and dental care. All patients who were: (i) aged 21 years and above (ii) diagnosed with current co-existence of three most prevalent chronic conditions, i.e., hyperlipidaemia, hypertension, and diabetes mellitus Type 1 or 2 (i.e., diagnosed with all three conditions) (iii) able to understand spoken English, Mandarin, Malay or Tamil and (iv) seen at the Polyclinic at least twice in the six months prior to recruitment were included in the study.
The current study was part of a larger study examining multimorbidity in a primary care setting, to ensure the achievement of the individual aims of the study, various sample sizes were calculated. The largest sample size was used to ensure that the study had enough power to answer all the research questions. Taking into account 5% missing data whereby listwise deletion could be safely practiced, a sample size of 892 was considered desirable. Further, assuming a 50% response rate from the patients approached for the study, a sample size of 1800 was considered reasonable. A random sample of 1800 patients who met the inclusion criteria was drawn from the patient population and tagged using the clinic list. The sample was released in 4 replicates as only one research assistant worked full time on the project, and this ensured a good outreach. In all, 1,366 patients were approached of whom 932 patients agreed to participate in the study – resulting in an acceptable response rate of 68.2%.
Clinicians and front-line staff referred the patients to the research assistant. Potential participants from the sample were approached before/during/after their scheduled appointments (i.e., regular follow up for their chronic conditions) at the Polyclinic and invited to participate in the study. Participants included were clinically stable (not acutely ill) and determined to be cognitively capable of providing informed consent and participating in the research which took about 45 minutes overall. Trained research assistants conducted the interviews in the language preferred by the respondent. The study questionnaire was programmed on the QuickTapSurvey (www.quicktapsurvey.com) app on a tablet computer. Each interview took approximately 30 minutes. On completion of the study respondents were paid SGD 30 as inconvenience fee. The Domain Specific Review Board, NHG, Singapore (Ethics Committee) approved the conduct of the study, and all respondents provided written informed consent before participating in the study.
Questionnaires
International Physical Activity Questionnaire Short Form (IPAQ-SF)
The 7-item IPAQ- SF questionnaire assesses a person’s physical activity undertaken as part of their daily life [38]. The first six questions of IPAQ-SF asks about three specific types of activity in the last seven days, namely: walking, moderate-intensity activities, and vigorous-intensity activities. Respondents are then asked about the specific number of days and amount of time in minutes which they spend doing these respective activities. The last question deals with the amount of time ‘spent sitting’ on workdays. Using the criteria provided in the IPAQ scoring protocol [39], daily and weekly metabolic equivalents of a task (MET) values were calculated using the same formulas recommended as follows:
Walking MET-minutes/week = ×3.3 walking minutes × walking days
Moderate MET-minutes/week = ×4.0 moderate-intensity activity minutes × moderate days
Vigorous MET-minutes/week = ×8.0 vigorous-intensity activity minutes × vigorous-intensity days
Total PA MET-minutes/week = sum of walking + moderate + vigorous MET-minutes/week scores.
Three levels of physical activity have been proposed by the IPAQ group [39]:
Low: Those not meeting criteria for either moderate or high physical activity as defined below were categorized in this group.
Moderate: This includes achieving a minimum of at least 600 MET-min/week OR 3 or more days of vigorous activity of at least 20 minutes per day OR
5 or more days of moderate-intensity activity OR walking of at least 30 minutes
every day in a week.
High: Comprises achieving a minimum of at least 3000 MET-minutes/week OR vigorous-intensity activity on at least 3 days and accumulating at least 1500 MET-minutes/week.
The American College of Sports Medicine and American Heart Association have recommended (both for adults aged 18-64 years and older adults aged 65 years and above as well as those with chronic physical conditions) moderate intensity aerobic activity for a minimum of 5 days per week of or vigorous intensity aerobic activity for a minimum of 3 days per week of to promote and maintain health [40]. These values correspond to the ‘moderate’ category of the IPAQ -SF. Thus, for the purposes of this study we have reclassified IPAQ categories into two groups – (i) Low physical activity group as ‘Insufficently active’and (ii) Moderate / High activity group as ‘Sufficently active’ [41].
The EuroQol-5 Dimension (EQ-5D-3L)
The EQ-5D-3L is an instrument that evaluates the generic quality of life and comprises a descriptive system and a Visual Analogue Scale (VAS) [42]. The descriptive system has five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Respondents were asked to rate their health on a three-point scale (no problem/moderate problem/extreme problem). The answers given by the respondents result in 243 unique health states and can be converted into a utility score (EQ-5D index) anchored at 0 for death and 1 for perfect health. The utility index used in this study was based on Singapore time trade-off values [43]. The VAS records the patient’s self-rated health on a vertical visual analogue scale that ranges from 0 for ‘Best imaginable health state’ and 100 for ‘Worst imaginable health state.’ In addition to examining the EQ-5D index and VAS, the five dimensions of the EQ-5D were also dichotomized (having no problem vs. having moderate/extreme problems) to identify whether respondents had endorsed having issues with said dimensions.
Socio-demographic collection form
Demographic information was obtained during the interview. These included the year of birth (age was calculated from the interview date), gender, ethnicity, marital status, education level, income and type of housing. These correlates have been found to be associated with physical activity in previous studies [41, 44-47]. The age of the respondents was grouped into four categories – ‘< 55’, ‘55-64’, ‘65-74’, and ‘≥ 75’, ethnicity was classified as Chinese, Malay, Indian and Others, Marital status was grouped into two categories – Married, and Single/Separated/Divorced/Widowed. Education level was grouped into four categories – No formal education, Primary, Secondary, and Post-Secondary. Monthly household income was classified into five categories – < SGD 2,000, SGD 2,000-3,999, SGD 4,000-5,999, SGD ≥ 6,000. Type of housing was grouped into four categories – 1/2/3 room Housing Development Board (HDB) (public housing) flats, 4-room HDB flats, 5-room/exceutive HDB/ Condominium/Private flats/Landed Property/Private Terrace/Bungalow.
Clinical Data
Data on hemoglobin A1c (HbA1C; cut off at 7% and above) [48], low density lipoprotein cholesterol (LDL-C; cut off at 2.60 mmol/L and above) [49] and blood pressure (Systolic blood pressure cut off at 140 mm/Hg; Diastolic blood pressure cut off at 90 mmHg) [50] were collected from patient records based on routine clinical monitoring of the patients as a measure of diabetes, dyslipidemia and hypertension control. We obtained the body mass index (BMI) by collecting height and weight of the respondents before the clinic appointment. Based on World Health Organization (WHO) [51] International cut-offs, respondents who had BMI ≥ 25 kg/m2 were classified as ‘Overweight/Obese’, while those whose BMI fell within the range of 18.5 to less than 25 were classified as having BMI in the ‘normal range.’ Additionally, we also analysed the BMI based on WHO guidelines for Asian populations, those who had BMI ≥ 23 kg/m2 were categorized as ‘Overweight/Obese’, and those in the range of 18.5 to less than 23 were classified to be in the ‘normal range’ [52].
Statistical analysis
Means and standard deviations were calculated for continuous variables, whereas frequencies and percentages were calculated for categorical variables. A multivariable logistic regression analysis was performed to determine the sociodemographic correlates of IPAQ. A series of simple logistic regression analyses were used to determine the association between IPAQ and clinical health outcomes, and the five subscales of the EQ-5D. Simple linear regression analyses were conducted to investigate the association between the IPAQ and EQ-5D index, as well as EQ-VAS scores. Statistical significance was set at the conventional level of p < 0.05, using two-sided tests. All statistical analyses were conducted with SPSS version 23.