A RANDOMIZED CONTROLLED TRIAL TO REDUCE LIFESTYLE RELATED RISK FACTORS FOR NON-COMMUNICABLE DISEASES THROUGH FUNERAL SOCIETIES IN SRI LANKA

A cluster randomized controlled trial was conducted to change diet, physical activity, smoking and alcohol consumption among participants through funeral societies in the Western Province. Twenty one administrative subdivisions in the Ragama area were randomized into intervention and control groups. Ten administrative subdivisions in the North Western Province were selected as an additional control group. The primary outcome was change in the proportion of individuals who consumed 5 servings or more of fruits and vegetables per day. The study showed only a modest, not significant increase in the proportion of people consuming 5 servings of fruits and vegetables or more per day. Of the secondary outcomes assessed, their intake of green leaves and mean MET minutes spent on leisure activities increased significantly more in the intervention group than in the control groups, but the differences were small. To effectively reduce non-communicable diseases (NCD) in Sri Lanka, community-based organizations could be utilized to deliver prevention programs.

In response, the Sri Lankan Ministry of  2008), in which 19.6% of the males and 30.4% of females aged 15 -64 years were found to be overweight or obese, as many as 51.5% of males and 66.0% of females were overweight or o bese in the preliminary analysis of the Ragama Health Study (Pathmeswaran, et al, 2009).Ragama is

MATERIALS AND METHODS
The      2 and 3, respectively.In the intervention group, the proportio n meeting their fruit and vegetable consumptio n goals increased by 1.9% from baseline to follo wup.This change was not significantly different from the change in the co ntro l groups (0% and 0.9%, respectively In secondary analyses of those who were part of the intervention group, assessing the relationship between particip ation in the workshops and diet, 79 (37%) participants attended 4-6 workshops and, 116(55%) did not attend any workshops.Sixteen (7.5%) participants attended 1 -3 workshops.Among those who attended 4 -6 workshops, there was a net increase of 18.8% in the i ntake of green leaves, compared to a 4.9% increase among those who did not attend any | workshops.This difference was statistically significant (X2=10.14, p<0.01).There were no significant differences by participation for the other dietary primary outcomes.
Change in mean phy sical activ ity levels in leisure domain (primary outcome) were greater among participants who attended 4 -6 workshops than among those with zero participation in the intervention group (figure 3), but this difference was also not statisti cally significant.The change in mean physical activ ity in work, transport and domestic domains were greater among participants who attended 4 -6 workshops, but again these differences were not statistically significant.review (Merzel and D'Afflitti, 2003).These The results were not adjusted for multiple comparisons, so the effects that are marginally significant should be interpreted with caution to avoid type 1 errors.

CONCLUSION
Our study demonstrates that funeral societies provide a promising vehicle to mobilize and address community for Rev i sta El etr ôn ic a Ge st ão & Soc ied ad e v. 12 , n .31, p . 2 19 4 -2 21 2 | Jan e iro /Ab ri l -20 18 IS SN 1 9 80 -57 56 | D OI: 1 0. 21 17 1/ g e s. v 1 which showed that 25% of the study population was overweight or obese.In addition, 11.5% were current daily smokers and 82.4% ate les s than 5 servings of fruits and vegetables per day.Twenty fiv e percent of the study population had a lo w level of physical activity with <6 are key to the preventio n of NCDs.The main risk factors associated with NCDs include an unhealthy diet, physical inactiv ity, tobacco use and harmful use of alco hol.Significant successes have been realized with lifestyle modifications addressing these risk factors in several studies in v arious populations(Hu et al, 2011; Simoes et al, 2009).The population strategy is the attempt to control the determinants of incidence, to lower the mean level of risk factors, and to shift the whole distribution of exposure in a favorable direction(Ro se, 2001).Although public health programs and policy are often defined at regio nal and national levels, the community is where preventio n and intervention activities actually take place(MacQueen et al, 2001)CBOs), which are notfor-profit nongovernmental, civil society and grassroots organizations are important stakeholders and have been increasingly called upo n to use research evidence to inform their advocacy, program planning, and service delivery efforts(Wilson et al, 2010).Experience fro m HIV/AIDS prevention programs has demonstrated that CBOs can successfully reach o ut to the most -at-risk populations.The utilization of CBOs has also Rev i sta El etr ôn ic a Ge st ão & Soc ied ad e v. 12 , n .31, p . 2 19 4 -2 21 2 | Jan e iro /Ab ri l -20 18 IS SN 1 9 80 -57 56 | D OI: 1 0. 21 17 1/ g e s. v 1 2 i3 1. 22 8 7 | 2196 been recommended for NCD prevention and treatment (Lamptey et al, 2011).In the US, community interventions aimed at reducing NCDs in minority gro ups hav e successfully delivered their programs through CBOs such as the Young Women's Christian Association (YWCA) and churches (Dodani, 2011; King, 2 007).Sri Lanka as a developing nation needs to identify feasible and cost effective measures to control the growing epidemic of NCD s and experience gained in more developed countries may be of value in this regard (Sarrafzadegan et al, 2 006).Sri Lanka has numero us CBOs at t he village level (Leitan, 2010).However, there is lack of research on whether the use of CBOs fo r delivering interventio ns to reduce NCD risk factors is effective.In Sri Lanka, funeral societies are common CBOs that are found in many villages and they organize and take part in many health -related activities.Funeral societies in Sri Lanka evolved fro m facilitating gatherings following death o r bereavement of a family member.These societies provi de financial and other services to the family of the deceased and incorporate a number of social services into their mandates.Today these societies take initiatives to uplift the status of the community including organizatio n of free health clinics, distr ibution of free spectacles and many other social and health activities.The societies are formed by the community itself with an elected committee and families are enrolled as members.Some of the societies are even registered with the government civil adm inistratio n system.These societies provide a promising context for successful community -based health promotion, especially as the capacity of communities to m obilize and address community issues is increasingly reco gnized and utilized.The objective of t his study was to evaluate the effects of a lifestyle modificatio n program targeting 25-60 year old adults delivered through funeral societies in Ragama, Sri Lanka.Out comes (change from baseline in diet, physical activity, smoking and alcohol use) were as sessed six months after the start of the program .Compared to the natio nal survey in 2007 (Sri Lanka Directorate Of Non Communicable D iseases, Rev i sta El etr ôn ic a Ge st ão & Soc ied ad e v. 12 , n .31, p . 2 19 4 -2 21 2 | Jan e iro /Ab ri l -20 18 IS SN 1 9 80 -57 56 | D OI: 1 0. 21 17 1/ g e s. v 1 2 i3 1. 22 8 7 | 2197 therefore identified as a high -risk area for NCD in the report.

2198 Figure 1
Figure 1 CONSORT flow diagram displaying clusters and subjects recruited into the community based intervention study
El etr ôn ic a Ge st ão & Soc ied ad e v. 12 , n .31, p . 2 19 4 -2 21 2 | Jan e iro /Ab ri l -20 18 IS SN 1 9 80 -57 56 | D OI: 1 0. 21 17 1/ g e s. v 1 2 i3 1. 22 8 7 | 2206 Table 3. Change in Physical activity (metabolic equivalent (MET) minutes) among participants in a community-based lifestyle modification intervention study in Western and North Western Provinces in Sri Lanka.Mean MET minutes at baseline and mean change at follow-up are shown for both control and intervention groups.The results of the multiple regression analysis showing the effects of the intervention by group is also shown.

Figure 3 .
Figure 3. Association of physical activity with workshop participation in intervention group.Change in mean MET minutes of activity by workshop attendance by domain in shown.
Rev i sta El etr ôn ic a Ge st ão & Soc ied ad e v. 12 , n .31, p . 2 19 4 -2 21 2 | Jan e iro /Ab ri l -20 18 IS SN 1 9 80 -57 56 | D OI: 1 0. 21 17 1/ g e s. v 1 of theory.The short duration of t he lifestyle modification program, limited access to resources and facilities for physical activity, time constraints participants may have faced and availability of healthy food choices are all possible reasons why the intervention failed to show more sig nificant improvements in diet and physical activity outcomes.There was no significant change in smoking and alcohol use follo wing the intervention in the present study .The result was no t unanticipated in that there was limited focus on these behaviors in our study -even large community trials conducted for many years have shown very small reduction in smoking and alcohol use(Gnich et al, 2002).001).The sitting time questio n in the IPAQ long version is an additional indicato r variable to assess sedentary behavio r.Sedentary behavior refers to a group of behaviors that occur whilst sitting or lying down while awake and typically require ve ry low energy expenditure.There is growing evidence that sedentary behavior may be adversely associated with chro nic disease in adults and risk factors for chronic disease in children and adolescents (British Heart Foundation National Centre (BHFNC), 2012 ).While the effects were small, co nsidering the relatively short duration of the intervention and limited reso urces used in the implementation, these changes are still promising for future prevention pro gram s for NCDs.The study demonstrates that funeral societies can be used successfully to deliver interventions to reduce NCD risk factors in the com munity, which could impact on the population level.Strengths of this study include use of a cluster randomized control design, the support of the funeral soci eties, the inclusio n of an additional control area from a separate province to assess potential contamination between the primary intervention and co ntrol areas and low numbers lost to follo w up.The main limitatio n o f the study is that the assessment of t he out comes were measured through self-reported change in diet, physical activity, alco hol use and smoking, and some ascertainm ent bias may therefore be present.Although administered by trained data collectors, the questionnaires like lymay not be very ac curate in measuring Rev i sta El etr ôn ic a Ge st ão & Soc ied ad e v. 12 , n .31, p . 2 19 4 -2 21 2 | Jan e iro /Ab ri l -20 18 IS SN 1 9 80 -57 56 | D OI: 1 0. 21 17 1/ g e s. v 1 .The participants in the intervention group would be more likely to remember and report behavior changes following the intervention than those in the control gro ups.The short interventio n period of 6 months is another limitation as it is difficult to make any reliable and valid evaluation of impact of behavioral intervention on outcomes.to attend any of the interactive worksho ps and exercise sessions which were part of the intervention.In fact, only 44% of the participants in the intervention area attended at least one workshop.However, all participants in the intervention area were provided with the educational materials and instructions for lifestyle modification on the first day o f contact and monthly reminders were sent through society members.
more intensive, and conducted as a long term interventio n.Rev i sta El etr ôn ic a Ge st ão & Soc ied ad e v. 12 , n .31, p . 2 19 4 -2 21 2 | Jan e iro /Ab ri l -20 18 IS SN 1 9 80 -57 56 | D OI: 1 0. 21 17 1/ g e s. v 1 an i M ar i e E li zab et h Dab r er a , M in is tr y of Hea lth car e an d Nu tr it io n , Sr i L an k a Wo r k a s a con su ltan t co m m u n it y p h y s ic ian an d r eg ion al ep id em io lo g is t for M i n i str y o f H ea lth c ar e an d Nu tr it ion , Sr i Lan k a .M y in ter e sts ar e ep id em i olo g y of com m u n i cab le an d n on com m u n icab le d is ea se s an d h e al th p o l ic y an d p lan n in g in p u b l ic h ea lth .E -m a i l: thushanidabrera@yahoo.comAr u n a sa l em Path m es wa r an Dep ar tm en t o f P u b l ic H ea lth , F acu lt y o f M ed ic i n e, Un i ver si ty of K e lan i ya , Sr i Lan k a .Pr o f es sor in Pu b l ic He a lth , Dep ar tm en t o f Pu b l ic Hea lth , Facu lt y o f M ed ic in e, Un i ver s it y o f K el an i y a , Sr i L an k a. M a in in ter est s d ise a se ep id e m i o l og y an d s tat i stc al m eth od s E -m a i l: pathmes@gmail.comAn u r ad h an i Ka stu r ir a th n a Dep ar tm en t o f P u b l ic H ea lth , F acu lt y o f M ed ic i n e, Un i ver si ty of K e lan i ya , Sr i Lan k a. Sen io r L ectu r er in Pu b l i c Hea lt h , Dep ar tm en t of Pu b l ic He a lth , F ac u lt y o f M ed ic in e , Un iv er s it y of Ke la n i y a, Sr i Lan ka .In t er est ed in ch r o n ic d i se a se ep id em i ol og y .E -m a i l: akasturiratne@yahoo.com Ga yan i T il l eker atn e Du k e G lob al Hea lth In st itu t e, D u ke Un i ver si ty , US A Du k e G lob al Hea lth In st itu t e, D u ke Un i ver si ty , US A .In f ect iou s d is ea se sp ec ia l i st a n d ep id em io lo g ist E -m a i l: ga y an i .ti l lek er a tn e@ d u k e.ed u Tr u l s Ø stb ye Du k e G lob al Hea lth In st itu t e, D u ke Un i ver si ty , US A Du k e G lob a l He a lth In st itu te , Du ke U n i ver s it y , U SA .A ch r on ic d ise ase ep id em io lo g i st an d p u b l i c h ea lth r e se ar ch er , h as a sp ec ia l in ter est in ob es it y , d ise as es of th e el d er l y an d gl ob a l h e a lth .E -m a i l: tr u ls .ostb ye@ d u ke.ed u study was co nducted in the Ragam a

Table 1
describes the socio demographic characteristics of participants in the intervention and control groups.In Ragama