Risk assess risk of diabetes, originally developed from community-based

RiskAssessment of Pakistani individual for Diabetes in a resource constraintsociety KhalidAbdul Basit, Asher Fawwad, Musarrat Riaz, Bilal Tahir, Maria Khalid, AbdulBasitABSTRACTAim: Assessment of Pakistani individuals who areat risk of developing diabetes using a RAPID (risk assessment of Pakistaniindividuals for diabetes) score.Methodology: This observational study was a sub-analysisof National Diabetes Survey of Pakistan (NDSP)conducted from 2016-2017 in allprovinces of Pakistan. Ethical approval was obtained from National BioethicsCommittee (NBC) Pakistan RAPID score, a validated and published scoring scaleto assess risk of diabetes, originally developed from community-based surveyswas used. The risk score is assessed by parameters namely; age, waist to hipratio and positive family history of the disease.

Subjects with score greater ?4 is considered at risk of DM. Information regarding social-demography andanthropometric measures, were obtained by a designed questionnaire on one-to-oneinterview bases by survey officers. Data was analyzed using Statistical Packagefor Social Sciences (SPSS) version 20.Results: A total of 4905 individuals were assessed,of which (n=2205) were male and (n=2700) were female. Mean age of participantswas 41.82 ± 14.19 years.

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

Significant differences between males andfemales were observed in BMI (p< 0.0001). RAPID score predicted that 1268 (25.9%) individuals would have risk ofdiabetes while 3637(74.1%) individuals had no risk of diabetes witha sensitivity of 45.

36% and a specificity of 76.39%. The comparisonof oral glucose tolerance test (OGTT) with RAPID score estimated that 20.2% subjects with normal glucose tolerance and 33.

7%prediabetics were at risk of DM. It also estimated that 54.6% subjects who werenewly diagnosed with DM by WHO criteria were not at risk.Conclusion:A simple diabetes risk score, based on a setof variables can be used for the identi?cation of high risk individuals forearly intervention to delay or prevent type 2 diabetes. Community basedawareness programs are needed to educate people regarding healthy lifestyle inorder to reduce the risk of diabetes.

IntroductionType 2 Diabetes is amongst the most commonchronic disease and a serious public health challenge of 21stcentury for both the developed and developing world. (1-2). The increasing prevalenceis not only contributing to health burden; a significant economic impact isalso noted (3). According to International Diabetes Federation (IDF), it isanticipated to foresee people with diabetes in 2011 will rise approximately bytwo-fold in 2030, about 366 to 552 million (4) very old reference bhaiya should be used 2017 atlasreference. The 2nd National Diabetic Surveyof Pakistan (NDSP) showed that approximately 26% of population is sufferingfrom diabetes (5).  This high prevalenceis fueled by ageing, physical inactivity, unhealthy food intake and stress ofdiverse origins (6-9). However, the effect of these environmental factorsvaries with variance in genome (10). With maternal gene predominance, thelifetime risk of developing diabetes to that of background individual,escalates to 40% in single parental disease, and about 70% if both parents areaffected (11).

Recent statistics has revealed that more thanone fourth of total diabetic population are unaware of their disease (12,13).Laboratory screening modalities including fasting blood glucose, HbA1c, OGTTare efficient in detecting diabetes. (14-15) However, mass screening isessential for early detection and appropriate intervention of the disease, acost-effective measure in resource limited health care system (16-17). For thisreason, The American Diabetes Association (ADA) recommends regular screeningfor type 2 diabetes beginning at the age of 45 years repeating subsequentlyafter every 3 years (18). Various models have been proposed for riskassessment of diabetes, but results are almost always heterogenous. Thisincludes the British, Canadian, Australian, German, Chinese and Indian riskassessment models comprising of designed questionnaires, anthropometric,demographic, family history and elementary lifestyle information (19-24) A similar algorithm was designed named, RAPID(risk assessment of Pakistani individuals for diabetes) for identification ofhigh risk individuals through readily available variables withoutpharmacological invention or physician interpretation (25).

This studydemonstrates the validation of preformed scoring system in the epidemiologicaland population-based survey conducted in Pakistan in 2016-2017. Secondly, thepredictability of future diabetes through the risk stratification score anddiagnostic modality is also assessed.  MethodologyThis community based observational study was asub-analysis of National Diabetes Survey of Pakistan (NDSP) conducted from February2016 to August 2017 in all 4 provinces of Pakistan. Ethical approval wasobtained from National Bioethics Committee (NBC) of Pakistan. All Pakistaniindividuals aged 20 years or more, were eligible to participate after obtaininginformed consent and informed about the purpose of the survey. Detailedinformation regarding demographic, anthropometric and medical examination wereobtained with the help of pre-designed questionnaire. All information wasgathered by one-to-one based interview by a trained survey officers. Height was measured to the nearest of 0.

1cm instanding erect posture vertically and weight was recorded with nearest of 0.1 kgwith participants in light clothes and without shoes by paramedical staff.Waist circumference was measured at the midpoint between the lower margin ofthe least palpable rib and the top of the iliac crest and hip circumference was measured around the widest portion ofthe buttocks, with the tape parallel to the floor.  Centralobesity was diagnosed as waist circumference ? 90 cm and ? 80 cm and / orwaist-to-hip ratio (WHR) ? 0.9 and ? 0.8 in males and females respectively (25).Detailed methodology has been published earlier (5).

RAPID score was used in the study to estimate therisk of diabetes. RAPID score, a validated and published scoring scale toassess risk of diabetes, originally developed from community-based surveys wasused. The risk score is assessed by parameters namely; age, waist to hip ratioand positive family history of the diabetes. Subjects with score greater ? 4 areconsidered at risk of diabetes (25).

                 Figure1: Flow diagram Total sample n = 10834 Already known diabetic   Excluded from the study   After given 75 gm of glucose Normal population   Pre-diabetic   Diabetic Known Diabetic Missing any RAPID score parameter Complete RAPID score parameter Missing any RAPID score parameter   Complete RAPID score parameter Missing any RAPID score parameter     Complete RAPID score parameter Excluded from the study   Excluded from the study   Excluded from the study   Final analysis                        StatisticalanalysisContinuous variables like age, weight, height, BMI,systolic and diastolic blood pressure were presented in the form of mean andstandard deviation. Categorical variables i.e. gender, family history ofdiabetes and hypertension presented as frequency with percentage. Independentt-test was used for continuous variables and chi-square test was used forcategorical variables. Parameters of age, waist to hip ratio and positivefamily history of diabetes was examined using multivariate logistic regression,p-value < 0.05 was considered statistically significant. Statistical Packagefor Social Sciences (SPSS) version 20.

0 was used for analyses.ResultsA total of 4905 participants were screenedfor RAPID score. The baseline characteristics of males (n=2205) and females(n=2700) of the study participants are shown in table-1. Mean age of theparticipants was 41.

82 ± 14.19 years. Mean BMI and waist-hip ratio (WHR) were26.84 ± 5.84 kg/m2 and 0.94 ± 0.

2, respectively. Significantdifferences between males and females were observed in BMI (p< 0.0001). Majorityof subjects were married and non-users of tobacco. No significant differencewas found in biochemical characteristics between males and females except forHDL-cholesterol (mg/dl) and triglycerides (mg/dl) (p<0.

0001). RAPID score predicted that 1268 (25.9%)individuals would have risk of diabetes while 3637(74.1%) individuals had norisk of diabetes with a sensitivity of 45.36% and a specificity of76.39% (table 2). Figure 2 shows comparison of oral glucosetolerance test (OGTT) with RAPID score.

RAPID scoreestimated that 20.2% subjects with normal glucose tolerance and 33.7%prediabetics were at risk of DM. It also estimated that 54.6% subjects who werenewly diagnosed with DM by WHO criteria were not at risk.

Table 1: Demographic andbiochemical characteristics of study population Variables Male Female P-value Total n 2205 2700 4905 Age (years) 42.96±15.03 40.89±13.4 <0.0001 41.82± 14.

19 Marital status Single 379(17.5%) 384(14.6%) <0.0001 763(15.9%) Married 1787(82.

5%) 2240(85.4%) 4027(84.1%) Tobacco addiction Yes 606(27.8%) 144(5.5%) <0.0001 750(15.

6%) No 1572(72.2%) 2486(94.5%) 4058(84.4%) Family history of diabetes Yes 644(29.2%) 735(27.2%) 0.124 1379(28.

1%) No 1561(70.8%) 1965(72.8%) 3526(71.9%) Weight (kg) 73.

64±14.87 65.73±14.

7 <0.0001 69.3±15.29 Height (cm) 168.45±9.98 154.9±9.

64 <0.0001 160.99±11.89 Body mass index (kg/m2) 26±5.23 27.52±6.2 <0.

0001 26.84±5.84 Waist circumference (cm) 92.13±13.

69 91.64±14.59 0.228 91.86±14.19 Hip circumference (cm) 101.18±14.

75 102.09±17.66 0.106 101.

68±16.43 Waist-to-hip ratio 0.94±0.15 0.94±0.23 0.559 0.94±0.

2 Systolic blood pressure (mmHg) 125.19±16.87 124.

3±19.23 0.093 124.7±18.

22 Diastolic blood pressure (mmHg) 83.07±12.04 83.31±13.73 0.535 83.2±13.01 Fasting blood sugar (mg/dl) 93.

59±38.84 92.56±36.99 0.345 93.02±37.83 Random blood sugar (mg/dl) 129.

7±52.78 130.06±48.18 0.806 129.9±50.29 Cholesterol (mg/dl) 192.

43±58.46 191.46±59.57 0.626 191.85±59.12 Triglycerides (mg/dl) 191.9±128.

8 169.03±109.68 <0.0001 178.

2±118.23 High density lipoprotein (mg/dl) 30.75±11.52 34.17±14.96 <0.

0001 32.8±13.78 Low density lipoprotein (mg/dl) 121.76±38.37 121.52±39.26 0.

857 121.62±38.9         Table 2: Sensitivity and specificity analysisof RAPID score  RAPID Score WHO GTT Classification DM NGT or PDM Total At risk 230 1038 1268 Not at risk 277 3360 3637 Total 507 4398 4905 Sensitivity 45.36% Specificity 76.39% Here,   DM= Diabetes Mellitus            PDM=Pre-Diabetes Mellitus (Includes Impaired & Fasting Glucose Tolerance)            NGT= Normal Glucose Tolerance                                 Figure 2: Comparison of OGTT with RAPID Score      DiscussionThis study demonstrates the simplified andconvenient diabetes risk score (RAPID), an efficient screening tool used forearly detection of type 2 diabetes among Pakistani population. This risk scoreincludes the basic parameters such as age, waist to hip ratio and familyhistory. It is noninvasive, convenient and cost-effective score for an ordinaryindividual to access a risk of diabetes.

It helps to identify high riskindividuals for developing diabetes.Among the modifiable risk factors that playeda considerable role in earlier studies was obesity, measured by BMI or waistcircumference (27). In the present study, both BMI and waist circumferenceincreases the risk of diabetes at cutoff points which was similar to the studyconducted by Aekplakorn W. et.al. (27) Moreover,most of our diabetic patients fall in the category of obesity, according to newstandardization of WHO for overweight (23-27kg/m²) and obesity (?27.5kg/m²) (28).

  Strong risk factor of type 2 diabetes is afamily history of diabetes. (29) In one of the earlier study, reduced physicalactivity is the convincing finding of family history of diabetes. (29) In thisstudy, family history was non-significant which was similar to a studyconducted in Canada. (19) Robust impact of BMI and family history leading toincreased risk of diabetes was observed in English young individuals. (32) Ageplayed a significant role in our study as seen in other risk prediction models(19,22,27) and increase incidence in middle aged population was noted earlierin the past three decades. (30)This risk prediction score of diabetes hasspecificity 76.39% and a sensitivity 45.

36% which is opposite to a surveyconducted in Oman with specificity of 65% and sensitivity of 79%. (26) Inearlier population-based study use of oral glucose tolerance test (OGTT) isimpractical to identify high risk individuals. Furthermore, nearly 40% subjectswith former normal OGTT 3-5 years earlier develops incident diabetes (31).Similarly, in our study almost one fifth of subjects with normal OGTT are atrisk of developing diabetes in their future.

 Conclusion: A simple diabetes risk score, based on a setof variables can be used for the identi?cation of high risk individuals forearly intervention to delay or prevent type 2 diabetes. Hence, community-basedawareness programs are needed to educate people regarding healthy lifestyle inorder to reduce the risk of diabetes. References: 1.     Gujral UP, Pradeepa R, WeberMB, Narayan KM, Mohan V. Type 2 diabetes in South Asians: similarities anddifferences with white Caucasian and other populations.

Annals of the New YorkAcademy of Sciences. 2013 Apr 1;1281(1):51-63.2.     Narayan KV.

Type 2 diabetes:why we are winning the battle but losing the war? 2015 Kelly West AwardLecture. Diabetes Care. 2016 May 1;39(5):653-63.3.     Hex N, Bartlett C, Wright D,Taylor M, Varley D. Estimating the current and future costs of Type 1 and Type2 diabetes in the UK, including direct health costs and indirect societal andproductivity costs.

Diabetic Medicine. 2012 Jul 1;29(7):855-62.44.     .WhitingDR, Guariguata L, Weil C, Shaw J. IDF diabetes atlas: global estimates of theprevalence of diabetes for 2011 and 2030.

Diabetes research and clinicalpractice. 2011 Dec 31;94(3):311-215.     NDS6.     Lin EH, Katon W, Von KorffM, Rutter C, Simon GE, Oliver M, Ciechanowski P, Ludman EJ, Bush T, Young B.

Relationship of depression and diabetes self-care, medication adherence, andpreventive care. Diabetes care. 2004 Sep 1;27(9):2154-60.

7.      MyttonOT, Clarke D, Rayner M. Taking unhealthy food and drinks to improve health.BMJ: British Medical Journal (Online). 2012;344.8.     Healy GN, Wijndaele K,Dunstan DW, Shaw JE, Salmon J, Zimmet PZ, Owen N.

Objectively measuredsedentary time, physical activity, and metabolic risk. Diabetes care. 2008 Feb1;31(2):369-71.9.     de Miguel-Yanes JM, ShraderP, Pencina MJ, Fox CS, Manning AK, Grant RW, Dupuis J, Florez JC, D’agostinoRB, Cupples LA, Meigs JB.

Genetic risk reclassification for type 2 diabetes byage below or above 50 years using 40 type 2 diabetes risk single nucleotidepolymorphisms. Diabetes care. 2011 Jan 1;34(1):121-5.10.  Sim X, Ong RT, Suo C, TayWT, Liu J, Ng DP, Boehnke M, Chia KS, Wong TY, Seielstad M, Teo YY.Transferability of type 2 diabetes implicated loci in multi-ethnic cohorts fromSoutheast Asia. PLoS genetics. 2011 Apr 7;7(4):e1001363.

11.  Ahlqvist E, Ahluwalia TS,Groop L. Genetics of type 2 diabetes.

Clinical chemistry. 2011 Feb 1;57(2):241-54.12.  Zhang Y, Dall TM, Mann SE,Chen Y, Martin J, Moore V, Baldwin A, Reidel VA, Quick WW. The economic costsof undiagnosed diabetes. Population health management. 2009 Apr 1;12(2):95-101.

13.  Abdulaziz Al Dawish M, AlwinRobert A, Braham R, Abdallah Al Hayek A, Al Saeed A, Ahmed Ahmed R, Sulaiman AlSabaan F. Diabetes mellitus in Saudi Arabia: a review of the recent literature.Current diabetes reviews.

2016 Dec 1;12(4):359-68.14.  Nolan CJ, Damm P, Prentki M.Type 2 diabetes across generations: from pathophysiology to prevention andmanagement. The Lancet. 2011 Jul 15;378(9786):169-81.15.

  Bennett CM, Guo M, DharmageSC. HbA1c as a screening tool for detection of type 2 diabetes: a systematicreview. Diabetic medicine. 2007 Apr 1;24(4):333-43.16.  LiR, Zhang P, Barker LE, Chowdhury FM, Zhang X. Cost-effectiveness ofinterventions to prevent and control diabetes mellitus: a systematic review.

Diabetes care. 2010 Aug 1;33(8):1872-94.17.  Zhou X, Pang Z, Gao W, WangS, Zhang L, Ning F, Qiao Q.

Performance of an A1C and fasting capillary bloodglucose test for screening newly diagnosed diabetes and pre-diabetes defined byan oral glucose tolerance test in Qingdao, China. Diabetes care. 2010 Mar1;33(3):545-50.18.  American DiabetesAssociation. Screening for type 2 diabetes. Diabetes care. 2004 Jan 1;27(suppl1): s11-4.

19.  Rowan CP, Miadovnik LA,Riddell MC, Rotondi MA, Gledhill N, Jamnik VK. Identifying persons at risk fordeveloping type 2 diabetes in a concentrated population of high riskethnicities in Canada using a risk assessment questionnaire and point-of-carecapillary blood HbA 1c measurement.

BMC public health. 2014 Sep 8;14(1):929.20.  BuijsseB, Simmons RK, Griffin SJ, Schulze MB.

Risk assessment tools for identifyingindividuals at risk of developing type 2 diabetes. Epidemiologic reviews. 2011May 27;33(1):46-62.21.  Noble D, Mathur R, Dent T,Meads C, Greenhalgh T.

Risk models and scores for type 2 diabetes: systematicreview. Bmj. 2011 Nov 28;343:d7163.22.  Chen L, Magliano DJ, BalkauB, Colagiuri S, Zimmet PZ, Tonkin AM, Mitchell P, Phillips PJ, Shaw JE.AUSDRISK: an Australian Type 2 Diabetes Risk Assessment Tool based ondemographic, lifestyle and simple anthropometric measures.

Medical Journal ofAustralia. 2010 Feb 15;192(4):197.23.  GaoWG, Dong YH, Pang ZC, Nan HR, Wang SJ, Ren J, Zhang L, Tuomilehto J, Qiao Q.

Asimple Chinese risk score for undiagnosed diabetes. Diabetic Medicine. 2010 Mar1;27(3):274-81.24.  Joshi SR. Indian diabetesrisk score. JAPI.

2005 Sep 13;53:755-7.25.  Riaz M, Basit A, Hydrie MZ,Shaheen F, Hussain A, Hakeem R, Shera AS. Risk assessment of Pakistaniindividuals for diabetes (RAPID). Primary care diabetes.

2012 Dec31;6(4):297-302.26.   Al-Lawati JA, Tuomilehto J. Diabetes risk score inOman: a tool to identify prevalent type 2 diabetes among Arabs of the MiddleEast. Diabetes research and clinical practice. 2007 Sep 30;77(3):438-44.

27.  Aekplakorn W, Bunnag P,Woodward M, Sritara P, Cheepudomwit S, Yamwong S, Yipintsoi T, Rajatanavin R. Arisk score for predicting incident diabetes in the Thai population. Diabetescare. 2006 Aug 1;29(8):1872-7.

28.   Jih J, Mukherjea A, Vittinghoff E, Nguyen TT,Tsoh JY, Fukuoka Y, Bender MS, Tseng W, Kanaya AM. Using appropriate body massindex cut points for overweight and obesity among Asian Americans. Preventive medicine.2014 Aug 31;65:1-6.29.   Isomaa B, Forsén B, Lahti K, Holmström N,Wadén J, Matintupa O, Almgren P, Eriksson JG, Lyssenko V, Taskinen MR, Tuomi T.

A family history of diabetes is associated with reduced physical fitness in thePrevalence, Prediction and Prevention of Diabetes (PPP)–Botnia study.Diabetologia. 2010 Aug 1;53(8):1709-13.30.

   FoxCS, Pencina MJ, Meigs JB, Vasan RS, Levitzky YS, D’Agostino RB. Trends in theincidence of type 2 diabetes mellitus from the 1970s to the 1990s. Circulation.2006 Jun 27;113(25):2914-8.31.

   Unwin N, Shaw J, Zimmet P, Alberti KG.Impaired glucose tolerance and impaired fasting glycaemia: the current statuson definition and intervention. Diabetic medicine. 2002 Sep 1;19(9):708-23.

32.   Hippisley-Cox J, Coupland C, Robson J, SheikhA, Brindle P. Predicting risk of type 2 diabetes in England and Wales:prospective derivation and validation of QDScore. Bmj. 2009 Mar 18;338:b880.