|Year : 2014 | Volume
| Issue : 2 | Page : 154-160
Package of essential noncommunicable disease (PEN) interventions in primary health-care settings of Bhutan: a performance assessment study
Dukpa Wangchuk1, Navkiran Kaur Virdi2, Renu Garg3, Shanthi Mendis4, Nani Nair5, Dorji Wangchuk6, Rajesh Kumar2
1 Division of Noncommunicable Diseases, Department of Public Health, Ministry of Health, Royal Government of Bhutan, Thimphu, Bhutan
2 School of Public Health, Postgraduate Institute of Medical Education and Research, Chandigarh, India
3 World Health Organization, Regional Office for South-East Asia, New Delhi, India
4 World Health Organization (WHO) headquarters, Geneva, Switzerland
5 WHO Representative, Bhutan
6 Director General Health Services, Bhutan
|Date of Web Publication||22-May-2017|
Professor of Community Medicine, School of Public Health, Postgraduate Institute of Medical Education and Research, Chandigarh 160012
Background: A World Health Organization (WHO) package of essential noncommunicable (PEN) disease interventions was piloted in two districts of Bhutan by non-physician health workers. They conducted risk assessment among patients aged over 40 years who visited the outpatient department of health institutions. Blood glucose was also measured among those who were overweight/ obese (body mass index ≥23 kg/m2) or had a high waist circumference (>80 cm in women and >90 cm in men). Appropriate counselling, treatment and referral were provided to the patients. The performance of the PEN project in detecting and managing noncommunicable diseases (NCDs) and their risk factors was assessed.
Methods: All health institutions of Paro (one district hospital and three basic health units [BHUs]) and Bumthang districts (one district hospital and four BHUs), were included in the PEN pilot assessment study. All patients who had presented to the clinics in the pilot districts from 1 June to 31 August 2012 constituted the study population. The data were collected from the clinical form, supervisor’s report and monthly report of the PEN project. The characteristics of patients with an NCD at registration and at the third follow-up visit were compared in a before–after analysis. Absolute changes in the characteristics of patients were computed for those who had completed the required followups during a 3-month assessment period.
Results: In a 3-month period, 39 079 patients had attended clinics in the pilot districts. About 10% of the clinic attendees (3818/39 079) were aged over 40 years; of these, 22.6% (864/3818) had a high blood pressure, and 49.7% (1896/3818) were overweight/obese or had a high waist circumference. Screening of overweight/ obese/high waist circumference cases revealed that 26.1% (494/1896) had high blood sugar levels. Out of the 896 patients who were registered on PEN protocols, 13% had >20% risk of developing cardiovascular diseases (CVDs) in next 10 years as per the WHO/International Society of Hypertension risk-assessment charts. Among 444 who had three follow-up visits, high 10-year-CVD risk (>20%) had declined from 13% to 7.3%. Among 400 persons with hypertension, use of medication increased and high blood pressure declined from 42.3% to 21.5%. Among 115 persons with diabetes, use of anti-diabetes medication increased and high blood sugar declined from 68/100 to 51/100.
Conclusion: Implementation of the PEN intervention in the primary health-care setting of Bhutan led to improvement in blood pressure and diabetes control, and reduction in CVD risk.
Keywords: Bhutan, intervention, noncommunicable diseases, primary health care
|How to cite this article:|
Wangchuk D, Virdi NK, Garg R, Mendis S, Nair N, Wangchuk D, Kumar R. Package of essential noncommunicable disease (PEN) interventions in primary health-care settings of Bhutan: a performance assessment study. WHO South-East Asia J Public Health 2014;3:154-60
|How to cite this URL:|
Wangchuk D, Virdi NK, Garg R, Mendis S, Nair N, Wangchuk D, Kumar R. Package of essential noncommunicable disease (PEN) interventions in primary health-care settings of Bhutan: a performance assessment study. WHO South-East Asia J Public Health [serial online] 2014 [cited 2019 Jul 17];3:154-60. Available from: http://www.who-seajph.org/text.asp?2014/3/2/154/206731
| Introduction|| |
Bhutan has a significant public health burden of noncommunicable diseases (NCDs), similar to other low-and middle-income countries undergoing epidemiological transition. Hospital-based statistics show a trend of rising NCDs. A community-based survey in the capital Thimphu revealed a high prevalence of NCD risk factors; 38.4% of the population in the 25–74 years age group had 3–5 risk factors in 2009.
To curb the rising trend and limit the growing burden of NCDs, the Royal Government of Bhutan has adopted a number of population-based strategies. These include the formulation of a national policy for the prevention and control of NCDs, banning the sale of tobacco and smoking in public places, and policies on promotion of a healthy diet and physical activity. In addition to interventions targeted at the population level, the government has also piloted a Package of Essential Noncommunicable (PEN) Disease Interventions for individual-level risk reduction and risk management.
The World Health Organization (WHO) developed the Package of essential noncommunicable (PEN) disease interventions for primary health care, in low-resource settings. PEN identifies core technologies, medicines and risk-prediction tools; discusses protocols required for implementation of a set of essential NCD interventions; and provides technical and operational outlines for integration of essential NCD interventions into primary care and for evaluation of their impact. The pilot project on the use of PEN interventions in two districts of Bhutan started in 2009. The results of a 3-month performance assessment study of this pilot in 2012 are reported here.
Background to the Bhutan PEN pilot
The Ministry of Health (MoH) initiated the implementation of PEN interventions in two pilot districts (Paro and Bunthang) in 2009. The populations of Para and Bumthang were approximately 32 000 and 17 000, respectively. Before implementation, WHO PEN protocols were adapted for Bhutan by an expert group, staff were trained, and basic diagnostic equipment and essential drugs were provided in the pilot districts.
All patients attending the outpatient clinics of health institutions located in the pilot districts who are found to have an NCD are registered by the health workers. In addition, all patients who are aged over 40 years are also screened, by asking about their lifestyles, and by measuring their blood pressure, height, weight and waist circumference, using standard methods.,, Those who are found to be overweight or obese or have a high waist circumference (body mass index [BMI] ≥23 kg/m2 or waist circumference >80 cm in women and >90 cm in men) undergo blood glucose testing (random or fasting), using glucometers. The 10-year risk of developing cardiovascular disease (CVD) is assessed using the WHO/International Society of Hypertension (ISH) risk-assessment charts. A PEN clinical form is filled in for those presenting with an NCD and for those who are found to have high blood pressure (>140/90 mmHg) or high blood glucose (blood sugar random >140 mg/dL or fasting >110 mg/ dL) during the screening. Clinical, laboratory, counselling and medication data are recorded in the clinical form. This form has three sections; section I records identification and sociodemographic data, and clinical diagnosis; section II records the results of investigations; and section III records medication and counselling data. A definitive date for a follow-up visit is entered in the clinical form. Patients are referred to a district hospital or national hospital, if required, as per the Bhutan adaptation of PEN protocols. A referral form is used when patients are referred, which has patients’ identification details, diagnosis and reasons for referral, and instructions from the referral facility to the basic health unit (BHU) for followup and refill medications.
During the follow-up visit, patients are asked whether they had been registered earlier; if so, the clinical record is retrieved, clinical and laboratory investigations are performed, counselling is provided and the data are recorded on the clinical form. Patients are asked whether they have taken all medications regularly since their last visit, and their prescriptions are refilled, if necessary. Health workers check patients’ blood pressure and BMI, and enter these data in the clinic registers; the blood glucose test results are entered in the laboratory registers as well. A note is also made on the clinical record of patients who have missed their follow-up visit and, when they visit the next time, the reason for missing the earlier follow-up visit is inquired into and recorded. Patients are expected to visit once a month or more if needed. On the last day of the month, health workers use tally sheets to abstract data from the outpatient register, laboratory register and clinical records, to prepare a monthly report that is sent to the district health office by the fifth day of the next month.
Supervisors conduct weekly supportive supervision visits. They check the accuracy of the data in the clinical form by comparing these with the laboratory register, clinic register and medicine stock register. They also conduct observation of the facility, validate the data sent in the monthly report, and provide on-the-spot feedback to staff. Wherever they observe deficiencies in knowledge, skills, counselling or recording, they conduct continuing education of staff. The supportive supervision checklists are used by the supervisor to report, each week, the adequacy of human resources, their training status and skills in diagnosing and managing NCDs; the status of equipment and laboratory reagents; the clinical protocols being followed; the outpatient register; the stock register; and clinic records, among others. The checklist also has a “remarks” column to record reasons for inadequacy and steps taken to correct the deficiencies. Reports of supervisory visits are sent, by email/fax, to the nodal office of the MoH on the same day the visit is conducted, or by the seventh day of each month. The MoH nodal officer is responsible for the overall coordination and management of the pilots, and for arranging supportive supervisory visits to each district every 2 weeks.
| Methods|| |
This study utilized only routinely collected health-care data and therefore did not require ethical clearance.
A performance assessment study of the PEN pilot was conducted by the MoH over a 3-month period during 2012, as part of routine monitoring and evaluation of health-care services. All health institutions of Paro (one district hospital and three BHUs) and Bumthang districts (one district hospital and four BHUs) were included in the PEN pilot evaluation study.
All patients who had presented to the clinics in the pilot districts from 1 June to 31 August 2012 constituted the study population, except those who were not likely to be staying in the area for at least 3 months. Before the start of the PEN evaluation study, clinical and managerial staff (physicians, nurses, basic health workers and district health officers) working in the health institutions of the pilot districts were oriented in 3-day workshops on data-collection and data-transmission procedures.
The data for the performance assessment were collected from the clinical form and supervisor’s report and monthly report. An electronic spreadsheet was prepared for data entry in the computer from the clinical form. Monthly reports provided an abstract of the data from the clinic register and clinical forms. The data were checked every month and feedback was provided to the district health officers about missing values, suspected wrong entries (range checks) or inconsistency in the data.
All registered patients with NCD were counselled about a healthy diet and physical activity. Those who consumed alcohol were counselled on alcohol harm reduction and smokers were advised to stop smoking. Medicines were prescribed as per the protocol.
Descriptive analysis included tabulation of monthly report data and clinic record data. The characteristics of patients with an NCD at registration and at the third follow-up visit were compared in a before–after analysis, especially for selected conditions (hypertension and diabetes), to estimate changes in behavioural (e.g. regular intake of medication) and biological characteristics (e.g. blood pressure control, blood sugar control, weight control). Absolute decrease or increase in selected behavioural and biological characteristics was computed among those who had completed three follow-up visits during the assessment period.
| Results|| |
During the 3-month period, 39 079 patients attended the clinics in the pilot districts; of these, 2568 (6.6%) had NCD. Those above 40 years of age were screened for high blood pressure and high BMI/waist circumference. About 10% of the clinic attendees (3818/39 079) were aged over 40 years; of these, 22.6% (864/3818) had high blood pressure, and 49.7% (1896/3818) were overweight/obese or had a high waist circumference. Those found to be overweight/obese/ have high waist circumference were further screened for high blood sugar and 26.1% (494/1896) of them were found to have high blood sugar levels. PEN clinical forms were filled in for 960 patients (who had an NCD or high blood pressure or high blood sugar, as per the PEN protocols), but the data of only 896 patients were complete and available for analysis; 640 of them were registered for the study in the first month of the 3-month assessment period.
Out of the 896 patients with complete data, 344 (38.4%) were enrolled in six BHUs and 552 (61.6%) in two district hospitals. Most were in the age group 40–59 years (47.4%), 60–79 years (37.4%) and 80 years or older (5.1%). Women comprised 60% of the patient population, illiterate persons 66.6%, farmers 52.7%, and 61.7% had an income below Nu 3000 (US$ 50) per month.
The most common conditions diagnosed were hypertension (802/896; 89.5%) and diabetes (185/896; 20.6%) (see [Table 1]). Some patients had more than one NCD (16.2%; 145). About 13% (115/896) of the patients had a 10-year CVD risk of >20%. Only 9.7% of patients reported that they consumed alcohol and 4.2% smoked; 57.1% were overweight (BMI = 23–30 kg/m2), 17.6% were obese (BMI >30 kg/m2) and 65.2% had a high waist circumference. High blood pressure was recorded in 41.5% and high blood glucose in 30.9%. Total cholesterol was measured only in those who attended hospitals (N= 109); of these, 20.2% were found to have high cholesterol levels (>200 mg/dL).
|Table 1: Types of noncommunicable disease cases registered in health institutions of Paro and Bumthang districts, Bhutan in a 3-month period (June to August 2012)|
Click here to view
Urine albumin testing was positive in 36/446 (8.1%) patients (trace = 16, 1+ = 12, 2+ = 5, and 3+ = 3). Peak expiratory flow rate was measured, to look for chronic obstructive pulmonary disease in 100 patients; 40 had a flow of >80% (green =, normal), 24 had a flow of 50%–80% (yellow = airway narrowing), and 36 had flow of <50% (red = severe airway narrowing). Foot examination was done in 172/185 patients with diabetes (93%), and eye check-up for blurred vision in 159/185 (86%).
Of the 896 patients, 720 (80.3%) and 451 (50.3%) attended the second and third follow-up visits. Eighty-six (9.2%) patients visited four times during the 3-month study period. At the first clinic visit, 43/896 (4.8%) patients were referred to a higher facility. At their second visit, 11/720 (1.5%), and at the third visit 3/451 (0.7%) were referred. Feedback was provided by the referral health facility to the primary health-care provider for 27 of the 43 patient who were referred after their first consultation visit. Similarly, feedback was provided for 4 of the 11 patients and 2 of the 3 patients who were referred after their second and third consultations, respectively.
Changes in the condition of patients between the first clinic visit and third clinic visit were assessed by comparing the characteristics among 444 patients who had completed three clinic visits during the assessment period (see [Table 2]). The proportion of those in the 10-year CVD risk category of >20% declined from 13% to 7.3%. Alcohol and smoking consumption had also declined. However, the level of obesity or high waist circumference did not decline much. Regular intake of medicine increased slightly. There was some decline in the proportion of those who had albumin in their urine. Peak expiratory flow rates also improved.
|Table 2: Change in characteristics of patients at various clinic visits in Paro and Bumthang districts, Bhutan, in a 3-month period (June to August 2012)|
Click here to view
Among 400 patients with hypertension, use of antihypertensive medication had increased and the proportion of those with high blood pressure declined from 42.3% to 21.5%. The various antihypertensive drugs prescribed and the proportion of patients taking each of these are detailed in [Table 2]. A small proportion of patients were also on anti-platelet (2%) and anti-cholesterol drags (4.5%); they may have had a cardiovascular event in the past, such as stroke or coronary artery disease.
Out of the 115 persons with diabetes followed up for 3 months, 97 (84.4%) were prescribed medicines. Anti-diabetic medication prescription increased slightly, from 95.9% to 97.9% during the period of the study. The proportion of patients who underwent examination of the feet and eyes for vision increased from the first to the third visit. The proportion of patients with albumin in their urine declined over the period of study. Although the proportion of patients with high blood sugar at the third follow-up visit remained high, a decline was observed from 68% to 51%.
Overall, during the study period, only 89 (10%) patients missed the followup visit. The reasons recorded for missing the followup were inability to walk for long distances by old people, disability following stroke, and not wanting to travel by bus due to motion sickness. A few patients shifted to indigenous medical practitioners, and some died during the followup period. Loss to followup was higher among patients in the hospitals, as many took treatment from other sources such as the army and other employers.
| Discussion|| |
A unique feature of the PEN interventions implemented in Bhutan was that they were implemented by the health services in real-life conditions through non-physician health workers. The proportion of patients who were followed up was high, owing to the involvement of health workers, who also conducted home visits. Implementation of PEN interventions led to the identification of new NCD cases. On an average, every month, each BHU registered five new patients and each hospital registered 15 new patients with NCD. Moreover, opportunistic screening of those aged over 40 years resulted in the identification of patients with high blood pressure and high blood sugar. These patients, and those with an NCD (at any age), were counselled on risk reduction (smoking cessation, alcohol harm reduction, healthy diet and physical activity). The major NCD identified was diabetes and the predominant risk factors identified were hypertension, overweight/obesity and high waist circumference.
Changes in the behavioural and biological risk factors could be assessed, as patient data were recorded at each clinic visit in individual case records. Comparisons of the characteristics of patients revealed a significant improvement in risks, such as reduced CVD risk due to better control of hypertension and diabetes by regular intake of medications; reduction in alcohol and smoking and albumin in the urine; and improvement in the peak flow expiratory rate (see [Table 2]). Similar outcomes have been achieved in projects conducted in controlled settings in high-income countries and in some low- and middle-income countries. ,,,,,,,However, significant change did not occur in the BMI or waist circumference during the 3-month observation period. Other studies have also reported difficulty in reducing overweight and obesity without intensive interventions.,” This finding indicates that health-promotion activities should focus on population-level strategies for healthy diets and physical activity, so that people can attain their optimum weight.
The changes measured by a before–after study design may not entirely reflect the effect of programme interventions. A comparison of data with those from other health facilities would have provided a better estimate of effect, but in view of the proven efficacy of NCD interventions in other settings,,,,,,,, it would not be ethical to collect data without the provision of interventions. That is why a cluster randomized trial was not considered for this study. However, the logical framework of input–process–output–outcome shows that it is reasonable to conclude that PEN interventions were responsible, to a large extent, for the observed changes. As the assessment was conducted only over a 3-month period, the sustainability of observed changes need to be studied over a longer time frame. Another limitation of the study was that the data on the standardization of equipment were not recorded and reported, although health workers had mentioned that all measuring equipment and measurement procedures were standardized periodically using protocols. The changes in this study were measured in patients rather than in the source population. The WHO recommended surveillance (STEPS) should be conducted periodically to measure the effect of PEN interventions at the population level.
The project faced several challenges, including those of following up patients and filling in lengthy forms, especially in a busy district hospital. Modifications are needed in the manuals, protocols and data-recording forms, to make these shorter and more user friendly. Establishing linkages between the district health officer who looked after the BHUs and medical officers of the district hospitals was also a challenge.
In future, efforts should be made to integrate monitoring of the PEN project within the health management information system, keeping in view the fact that PEN interventions also require monitoring of outcomes. Software needs to be developed for recording the clinical data of patients, to monitor outcomes. The national citizenship number can be included in the clinical records, so that the data of patients with NCD who seek care in multiple institutions are not duplicated at the national level. Patients should also be given a record that they can use while visiting other health facilities. As the majority of patients with NCD who sought care from the government clinics were poor, middle-aged women, integrating the identification and treatment of NCDs in the primary health-care setting would benefit vulnerable sections of society. The fact that no change in the BMI or waist circumference was found during the 3-month observation period highlights the need for intensive population-based strategies and interventions.
Secondary- and tertiary-care facilities should be made aware of the drug formulary available at primary health-care level, so that drugs available at primary health-care level can be utilized more efficiently. Linkages should also be built with indigenous health practitioners, as some patients stop taking allopathic drugs on the advice of indigenous medicine practitioners. To encourage and conduct followup of those who find it difficult to attend clinics, such as elderly patients and those with disabilities, mobile phones and home visits can be used. Village health workers can be trained and utilized for this purpose.
PEN protocols should have benchmarks for followup, i.e. the interval at which followup should be considered adequate for various categories of patients, and the changes that are expected in behavioural and biological risk factors over that period (quarterly/biannually/annually, etc.), so that health workers can aim to achieve definite outcomes. Costing of PEN interventions in varying contexts also needs to be done, so that sufficient resources are allocated for scale-up.
| Conclusions|| |
The Bhutan PEN pilot project empowered non-physician primary health-care workers to extend screening, diagnostic, treatment and counselling services to patients with NCD from a health facility close to their home. The availability of medicines free of cost and close to home, and counselling by health workers who were well known in the community improved behaviours and encouraged the regular intake of medicines, which led to better control of high blood pressure and high blood sugar, thereby reducing the CVD risk. Overall, there were several markers of successful outcomes; hence, PEN interventions should be extended not only to all districts in Bhutan but also to other low- and middle-income countries.
| Acknowledgements|| |
We thank the World Health Organization for extending financial and technical support, and Dr Lungten Z Wangchuk from the Division of Noncommunicable Diseases, Department of Public Health, Ministry of Health, Royal Government of Bhutan for her contribution to planning, implementation, monitoring and evaluation of the PEN project.
Source of Support: World Health Organization.
Conflict of Interest: None declared.
Contributorship: DW, planning of study, collection and interpretation of data and writing of manuscript; NKV, data editing, analysis, interpretation of data and writing of manuscript; RG, planning of study, data interpretation and writing of manuscript; SM, planning of study, standardization of data-collection instruments, data interpretation and writing of manuscript; NN, planning of study, data interpretation and writing of manuscript; DW, planning of study, data interpretation and writing of the manuscript; RK, planning of study, standardization of data-collection instruments, data interpretation and writing of the manuscript.
| References|| |
Omran AR. The epidemiologic transition: a theory of the epidemiology of population change. Millbank Mem Fund Q. 1971;49:509–38.
Royal Government of Bhutan, Ministry of Health. Annual Health Bulletin. Thimphu: Policy and Planning Department, 2012.
Royal Government of Bhutan, Ministry of Health. Report on 2007 STEPS survey for risk factors and prevalence of non-comunicable diseases in Thimphu. Thimphu: MOH, 2009. http://www.who.int/chp/ steps/2007NCDreport.pdf
- accessed 23 September 2014.
World Health Organization. Package of essential noncommunicable (PEN)p disease interventions for primary health care in low-resource settings. Geneva: WHO, 2010.
Royal Government of Bhutan, Ministry of Health. Package of essential NCD (PEN) interventions for basic health units. Thimphu: Non-Communicable Division, Department of Public Health, 2011.
Beevers G, Lip GYH, O’ Brien E. ABC of hypertension: blood pressure measurement. BMJ. 2001; 322:981–5.
Adams C, Burke V, Beilin LJ: Accuracy of blood pressure measurement and anthropometry among volunteer observers in a large community survey. J Clin Epidemiol. 2002;55:338–44.
Ulijaszek SJ, Kerr DA. Anthropometric measurement error and the assessment of nutritional status. Br J Nutr. 1999;82:165–77.
Mendis S, Lindholm LH, Mancia G, Whitworth J, Alderman M, Lim S, Heagerty T. World Health Organization (WHO) and International Society of Hypertension (ISH) risk prediction charts: assessment of cardiovascular risk for prevention and control of cardiovascular disease in low- and middle-income countries. J Hypertens. 2007; 25:1578–82.
Kar SS, Thakur JS, Jain S, Kumar R: Cardiovascular disease risk management in a primary health care setting of north India. Indian Heart J. 2008;60:19–25’
Puska P, Vartiainen E, Tuomilehto J, Salomaa V, Nissinen A. Changes in premature deaths in Finland: successful long-term prevention of cardiovascular diseases. Bull World Health Organ. 1998;76:419–25.
Farquhar JW, Maccoby N, Wood PD, Alexander JK, Breitrose H, Brown BW, Jr, Haskell WL, McAlister AL, Meyer AJ, Nash JD, Stern MP. Community education for cardiovascular health. Lancet. 1977Jun 4; 1(8023): 1192–5.
Multiple Risk Factor Intervention Trial Research Group. Multiple risk factor intervention trial. Risk factor changes and mortality results. JAMA. 1982;248:1465–77.
Farquhar JW, Fortmann SP, Flora JA, Taylor CB, Haskell WL, Williams PT, Macoby N, Wood PD. Effects of community wide education on cardiovascular disease risk factors. The Stanford Five-City Project. JAMA. 1990;264:359–65.
Luepker RV, Murray DM, Jacob DR Jr, Mittelmark MB, Bracht N, Carlaw R, Crow R, Elmer P, Finnegan J, Folsom AR. Community education for cardiovascular disease prevention: risk factor changes in the Minnesota Heart Health Program. Am J Public Health. 1994;84:1383–93.
World Health Organization, Regional Office for Europe. Countrywide integrated non-communicable disease intervention programme (CINDI). Copenhagen: WHO-EURO, 1995.
Hardeman W, Griffin S, Johnston M, Kinmonth AL, Wareham NJ. Interventions to prevent weight gain: a systematic review of psychological models and behaviour change methods. Int J Obes Relat Metab Disord. 2000;24:131–43.
Goodpaster BH, DeLany JP, Otto AD, Kuller L, Vockley J, South-Paul JE, Thomas SB, Brown J, McTigue K, Hames KC, Lang W, Jakicic JM. Effects of diet and physical activity interventions on weight loss and cardio-metabolic risk factors in severely obese adults: a randomized trial. JAMA. 2010;304:1795–802.
[Table 1], [Table 2]
|This article has been cited by|
||Public health and health systems: implications for the prevention and management of type 2 diabetes in south Asia
| ||Andrew P Hills,Anoop Misra,Jason M R Gill,Nuala M Byrne,Mario J Soares,Ambady Ramachandran,Latha Palaniappan,Steven J Street,Ranil Jayawardena,Kamlesh Khunti,Ross Arena |
| ||The Lancet Diabetes & Endocrinology. 2018; |
|[Pubmed] | [DOI]|