Smoking cessation among diabetes patients: results of a pilot randomized controlled trial in Kerala, India.

0.22 MB·English

by  Thankappan, KR

Thankappanetal.BMCPublicHealth2013,13:47
http://www.biomedcentral.com/1471-2458/13/47
RESEARCH ARTICLE Open Access
Smoking cessation among diabetes patients:
results of a pilot randomized controlled trial in
Kerala, India
KR Thankappan1*, GK Mini2, Meena Daivadanam1, G Vijayakumar3, PS Sarma1 and Mark Nichter4
Abstract
Background: India has thesecond largest diabetic population (61 million) and tobacco users (275million) inthe
world. Data on smoking cessation among diabetic patients are limited in low and middleincome countries. The
objective of the study was to document the effectiveness of diabetic specific smoking cessation counseling bya
non-doctor health professionalin addition to a cessation advice to quit, delivered bydoctors.
Methods: In our parallel-grouprandomizedcontrolled trial,we selected 224 adult diabetes patients aged 18 years
or older who smoked inthe last month,from two diabetes clinics in South India. Using a computer generated
random sequence withblock size four; the patients were randomizedequallyinto intervention-1and intervention-2
groups. Patients in both groups were asked and advised to quit smoking by a doctor and distributed diabetes
specific educationmaterials. The intervention-2group received an additional diabetes specific 30 minutes
counseling sessionusing the5As (Ask, Advise, Assess, Assist and Arrange), and 5 Rs(Relevance, Risks, Rewards,
Roadblocks and Repetition)from a non-doctor health professional. Follow updata were available for 87.5% of
patients atsixmonths. The Quit Tobacco International Project is supported by a grant from theFogarty
International Centre of theUS National Institutes of Health(RO1TW005969-01).
The primary outcomes were quit rate (seven day smoking abstinence) and harm reduction (reduction of the
number ofcigarettes /bidis smoked perday > 50% ofbaselineuse) atsix months.
Results: Inthe intention to treat analysis, theodds for quitting was 8.4 [95%confidence interval (CI): 4.1-17.1] for
intervention-2 group comparedto intervention-1group. Even among high level smokers theodds of quitting was
similar. The odds of harm reduction was 1.9(CI: 0.8-4.1) for intervention-2 group compared to intervention-1group.
Conclusions: The value addition of culturally sensitive diabetic specific cessation counseling sessions delivered by
non-doctor health professionalwas animpressive and efficacious way of preventing smoking related diabetic
complications.
Trial Registration: Clinical Trial Registry of India (CTRI/2012/01/002327)
Keywords: Diabetes, Smokingcessation, Counseling, Kerala, India
*Correspondence:[email protected]
1AchuthaMenonCentreforHealthScienceStudies,SreeChitraTirunal
InstituteforMedicalSciencesandTechnology,Trivandrum,Kerala,695011,
India
Fulllistofauthorinformationisavailableattheendofthearticle
©2013Thankappanetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsofthe
CreativeCommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,
distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
Thankappanetal.BMCPublicHealth2013,13:47 Page2of8
http://www.biomedcentral.com/1471-2458/13/47
Background in peri-urban areas of two south Indian cities located in
India has the second largest population (1210 million) [1] Kerala state were screened for smoking through the use
and number of people with diabetes (61 million) [2] and of a brief instrument provided to patients at the clinic
tobacco users (275 million) [3] after China. Both diabetes registration counter from December 2008 to April 2011.
prevalence and tobacco use are increasing rapidly in India. Among them 14.6% (n=363) were current smokers.
Although the proportion of smokers among all types of Being a pilot study we decided to include all the patients
tobaccousersinIndiawasonly40.5%in2010,incontrastto who satisfied the inclusion criteria during the recruit-
other countries, smoking was predicted to cause about one ment period. Inclusion criteria for the study were: male
million deaths [4]. Smoking is strongly linked to the risk of diabetes patients aged 18 years or above, literate, native
diabetes morbidity as well as mortality [5-7]. The Inter- to the clinic catchment area, high probability that they
national Diabetes Federation in 2003 [8] and the American would be treated at the clinic for the next six months
Diabetes Association in 2004 [9] have both strongly recom- and willingness to participate in the study. After exclud-
mended that people with diabetes not to smoke because of ing patients who did not meet inclusion criteria, or did
increased risk of diabetes complications. The major compli- not agree to participate in the study (n=139) a total of
cations are cardiovascular diseases [10], stroke [11], diabetic 224 male diabetes patients (mean age 53 years) who
retinopathy[12],andperipheralarterialdisease[13]. smoked in the previous month were selected for the
Kerala,theIndianstatemostadvancedinepidemiological study. Female patients were excluded since the smoking
transition [14] and the state with the highest prevalence of rate among females in Kerala was zero percent [3].
diabetes, was reported as the harbinger of what is going to Ethical clearance for the study was given by the Sree
happen to the rest of India in the near future [15,16]. Chitra Tirunal Institute for Medical Sciences and
Current smoking prevalence of 27.9% among Kerala men Technology, Trivandrum. Written consent to partici-
washigherthanthe24.3%forthewholeofIndia[3].Apre- pate was obtained after patients were informed about
vious Quit Tobacco International (QTI) study in Kerala the purpose of the study. Patients were a mix of newly
found that 59% of diabetes patients were tobacco users diagnosed and long time patients (Figure 1).
(43.5%exclusivesmokers)priortodiagnosisandmorethan
half of these users continued to use tobacco, many daily, Studyprocedure
even after diagnosis. Notably, 52% had not been advised to The procedure followed was minimally invasive. A
quitsmokingbytheirdoctorsanddidnotassociatesmoking screening instrument was kept at the hospital reception
with diabetes complications [17]. Given the prevalence of where all the patients had to register after entering the
smoking among diabetics, there was clearly a need for pro- hospital. Smokers were identified by the counselor from
active cessation efforts. Results of a randomized controlled this screening tool which inquired the patient’s smoking
trialfromtheUSfoundthatsmokingcessationintervention status. Patients attending the clinic routinely go to the
using motivational interviewing integrated into an estab- lab for a blood glucose examination and then have a
lished diabetes self management training program curricu- waiting period before lab results are ready and they can
lum resulted in a trend towards greater abstinence at three be seen by the doctor. During this time, the counselor
monthsoffollow-upinthosereceivingthedirectedsmoking met patients who had indicated a history of smoking on
cessation intervention [18]. Data on smoking cessation the screening instrument. After being informed about
among diabetic patients are limited in low and middle in- the study details written consent from the patients was
comecountries.AnIndonesianstudyof71diabeticpatients obtained. From those who gave consent, the counselor
demonstratedthefeasibilityofdisease-centreddoctors’mes- collected baseline information using a pre tested struc-
sages about smoking cessation for these patients in a clinic tured interview schedule. Details about basic demo-
setting [19]. Considering the limited access to doctors in graphic information, smoking history, current smoking
India, particularly in rural areas, [20] there is a need to patterns and presence of any other chronic diseases
utilize the services of non-doctor health professionals for (hypertension, cardiovasculardisease,stroke,chronicob-
smoking cessation more frequently than doctors. The ob- structive pulmonarydisease,andcancer)were collected.
jectiveofthestudywastodocumenttheeffectivenessofdia- Subsequently the counselor randomized the patients
betic specific cessation counseling by a non-doctor health equally into two groups; intervention–1 and interven-
professionalinadditiontoadiabeticspecificcessationmes- tion–2 groups, with block size four. Sequentially, every
sagetoquit,deliveredbydoctors. four patients enrolled were randomized into the two
intervention groups using a computer generated random
Methods sequence to achieve a block size of four, to facilitate in-
Participants terim analysis. Their medical records were then flagged
All of the 2490 male diabetic patients (aged 18 years and with different colored stickers by the counselor in order
above) who attended two referral diabetes clinics located to identify group assignment. After the interview, the
Thankappanetal.BMCPublicHealth2013,13:47 Page3of8
http://www.biomedcentral.com/1471-2458/13/47
Enrollment
Assessed for eligibility (n=2490)
Excluded (n= 2266)
Not meeting inclusion criteria (n=2238)
Declined to participate (n=28)
Randomized (n=224)
Allocation
Allocated to Intervention group-1 (n=112) Allocated to Intervention group-2 (n=112)
Received allocated intervention (n=112) Received allocated intervention (n=112)
Follow-Up at 6
Drop out (n=14) Drop out (n=14)
months
Migration –6, Inconvenience to attend on the Migration –4, Inconvenience to attend on the
follow-up day -5, Lost to follow up-3 follow-up day -5, Lost to follow up -5
Analysis
Analysed (n=98)
Analysed (n=98)
Figure1Patientflowdiagram.
patient consulted the doctor. The doctor gave the patient relevance and support the doctor’s advice) the counselor
a standard diabetes specific tobacco cessation message. assessed each patient’s readiness to quit. If ready to quit,
Thedoctoralsoshowedthepatientvisualimagesofcom- the counselor assisted him by discussing practical quit
mon diabetes complications exacerbated by smoking. At tips, how to get through an initial period of withdrawal,
the end of consultation the doctor instructed the patient andhowtodealwithcommonwithdrawalsymptoms,em-
to meet with the counselor. The counselor provided edu- phasizingthattheseonlylastedforafewdays.Ifnotready
cational materials on tobacco and diabetes developed by to quit, the counselor briefly identified roadblocks and
theQTIontheharmoftobaccofordiabetespatientsthat challengestoquitting,andencouragedthepatienttothink
built on formative research and followed a question an- aboutquittingafterreconsideringtherisksofsmokingfor
swer format for all the patients [21] and gave follow up developing diabetes complications and the benefits of
datesforconsultationtoallpatients. quitting as a means of preventing complications as a
Intervention–2 group patients received three diabetic primemotivator.
specific tobacco counseling sessions (at first contact, at All patients were given smoking cessation advice on
onemonthandatthreemonths)lastingabout30minutes each visit by the doctor for the next six months. Partici-
in each session following the 5 ‘A’s (Ask, Advise, Assess, pants in intervention–2 group additionally received face
AssistandArrange)and5‘R’s(Relevance,Risks,Rewards, to face counseling sessions on each visit for the next six
Roadblocks and Repetition) [22]. In this session, after months. Thus patients in the intervention group-2
goingovertheeducationalmaterial,developedbytheQTI receivedthreecounselingsessions:firstatbaseline,second
for smoking cessation, with the patient (to establish atmonthoneandthirdatmonththreeoffollow-up.
Thankappanetal.BMCPublicHealth2013,13:47 Page4of8
http://www.biomedcentral.com/1471-2458/13/47
Training on the number of cigarettes/bidis smoked on an average
The doctors and diabetes educators selected to counsel dayofuse.
patients in the study sites were initially given training on
theharmoftobaccofordiabetespatientsincluding:1)are- Statisticalanalysis
view of epidemiological data on smoking as a diabetes risk Statistical comparisons of means and proportions were
factor, 2) complications strongly associated with smoking made using Student’s t-tests, Chi-square tests, Chi-square
among those afflicted with diabetics, and 3) the mechan- tests for trend or Fisher’s exact test. The relative risk was
isms through which smoking contributes to vascular con- estimated by computing odds ratios (OR). Multivariable
striction and obstructed blood flow. Educational materials models using multiple logistic regression analyses were
developed by the QTI for diabetes patients that explain used to identify the correlates of quit rates. A complete
these factsin simple terms were provided to the counselor case analysis and intention to treat analysis were done. In
to give to patients. Doctors and the counselors were also order to test our hypothesis that high level smokers are
trained in basic brief intervention cessations skills. Doctors more addicted to smoking and less likely to quit smoking
wereinstructedtoaskallpatientsabouttheirsmokingsta- comparedtotheirlowlevelsmokingcounterparts,wedid
tus and to strongly advise patients not to smoke using a astratifiedanalysisof baselinelevelsmokingandquitrate
standardized diabetes specific cessation message linking at sixmonths.AlltheanalysesweredoneusingSPSSver-
smoking to the complications of diabetes. Doctors were sion 17.0 and statistical significance was set at two tailed
provided a visual aid illustrating how tobacco narrows the p<0.05.Thestatisticianwasblindedtogroupassignment.
passage of blood in the vascular system and pictures illus-
tratingdiabetescomplicationsatdistalpointsofthe vascu- Results
larsystemsuchaseyes,feet,fingers,andpenis.Insum,the We screened 2490 male diabetic patients. Among them
doctors were instructed to actively deliver two of the Five 363 (14.6%) were current smokers. Of these 363 patients,
As, Ask and Advise, using illness specific visual aids. The after excluding patients who were not willing to partici-
counselors were given additional training in tobacco cessa- pate in the study (n=31), who were not the natives to the
tioncounselingandinstructedtoactivelyconductallofthe cliniccatchmentarea(n=89)andwhocouldnotcomefor
five ‘A’s with patients in intervention group–2 each time follow ups for the next six months (n=19), a total of 224
they attended the clinic and 5 ‘R’s when necessary. The patientswereincludedinthefinalstudy.
counselors were instructed to document the details of ces- Average age of the study patients was 53 years (range
sation offered to at least 15 patients using five ‘A’s and five 28–75). Under the age of 40 years there were 17 patients
‘R’s. An examination was conducted by one of the authors (7.6%), confirming previous findings of the early onset of
(MN) based on these 15 brief interventions to assess their diabetesinIndia[23].Mediandurationofdiabeteswassix
cessation skills. A certificate titled “basic tobacco cessation yearscomparedtothemeandurationofeightyears.Close
competency” was issued on successful performance in the to three-fourths of the patients were subjectively assessed
examinationbytheUniversityofArizona. by the counselors as belonging to the middle socioeco-
nomicstatus(SES)group,whichwassimilartotheSESof
Follow-upofpatients thegeneralpopulationinKerala[24].Baselinecharacteris-
Followupinterviews were conducted withall studyparti- tics in both the intervention groups were comparable
cipants in both groups at one month, three months and (Table1).
six months. All the follow-up interviews were conducted The mean age of initiation of smoking was 21 years (SD
in person, although some of them were reminded by 6.9, range: 8–56 years). Around 44% of patients initiated
phone calls to come for the follow-up visit. Patients were smokingintheiradolescentyears(<20years).Twentyeight
askedabouttheirsmokinginthelastsevendaysaswellas patients(12.5%)werediagnosedinthelasttwoyearspriorto
questionsrelatedtothenumberofcigarettes/bidisusedin the study. Mean age of diagnosis of diabetes was 45.4 years
an average day. We reminded all the patients in both the (SD: 10.1, range: 22–71 years). Thus on an average these
groups by phone about the six month follow up in order patientssmoked24yearsbeforethediagnosisofdiabetes.
to get maximum response. All the diabetic patients fol- In the first follow-up wave (month one) we were able
lowed up at one, three and six months were seen by the to contact 173 (77.2%) patients, in the second follow-up
doctorandwereadvisedtoquitsmoking. (month three) 163 (72.8%) patients and in the third
follow-up(monthsix) 196(87.5%)patients.
Outcomemeasure
The primary outcome measure was a seven day smoking Quitrateandharmreduction
abstinence (quit rate) measured by a question: “During Smoking status of the patients at the six-month follow
the past seven days, did you smoke even a puff?” Other up based on complete case analysis is given in Table 2
smoking outcomes gathered included patients’ reports andthatbasedonintentiontotreat analysis inTable 3.
Thankappanetal.BMCPublicHealth2013,13:47 Page5of8
http://www.biomedcentral.com/1471-2458/13/47
Table1Baselinecharacteristics
Backgroundcharacteristics Interventiongroup-1 Interventiongroup-2 P
value
N=112 N=112
Meanage(years)±SD 54.2±8.8 52.5±9.9 0.193
Meanageofinitiationofsmoking(years)±SD 20.9±8.1 21.2±5.6 0.723
Meanageatdiagnosisofdiabetes(years)±SD 46.3±9.2 44.5±10.7 0.193
Meandurationofdiabetes(years)±SD 7.9±6.1 8.0±6.6 0.897
Meannumberofsticksusedperdayatbaseline±SD 15.0±14.6 14.1±13.2 0.640
CurrentlyMarried(%) 98.2 94.6 0.140
Others(%) 01.8 05.4
<10yearsofschooling(%) 27.7 20.5 0.137
≥10yearsofschooling(%) 72.3 79.5
Working(%) 61.6 66.1 0.289
Notworking(%) 38.4 33.9
LowSES(%) 24.1 17.9
MiddleSES(%) 72.3 75.9
UpperMiddleSES(%) 03.6 06.3 0.376
Presenceofanyotherchronicdiseases(%) 39.3 38.4 0.500
SD=StandardDeviation.SES=SocioeconomicStatus.
Data from the six-month follow-up was available for six-month follow-up the quit rate further increased to
196 patients (87.5%). The odds for quitting was 10 times 51.8% in the intervention group–2. Among those who
higher for intervention–2 group compared to interven- cameforallthethreefollowupvisitsinthisgroup,statis-
tion–1 group in the complete case analysis and close to tically significant (p=0.007) positive trend in quit rate was
nine times higher in the intention to treat analysis. seen with increase in the number of counseling sessions
Harm reduction (defined as > 50% reduction in the attended.
number of cigarettes/bidis used per day compared to Readiness to quit was assessed only for intervention
baseline use), which was significantly higher in the inter- group-2 as part of the intervention strategy. Out of the
vention–2 group, was not found to be significant in the 112 patients in the intervention – 2 groups 77 reported
intention to treat analysis. The mean number of cigar- that they were ready to quit at baseline. At six month
ettes/bidissmokedperdayatmonthsixwas4(SD8.2)in follow-up, 40 patients (51.9%) out of these 77, quit
the intervention–2 group, significantly lower (p < 0.001) smoking where as among the 35 patients who were not
than that of 10 (SD 13.7) in the intervention–1 group in ready to quit at baseline, 18 (51.4%) quit smoking at six
completecaseanalysis. month follow-up.
The quit rate based on intention to treat analysis at Quit Rate at six months by baseline level of smoking
the one-month follow-up between the intervention–1 is given in Table 4. Although the quit rates among low
group (11.6%) and the intervention–2 group (19.6%) was and high level smokers significantly increased in the
not statistically significant (p = 0.09). At three months intervention-2 group, the increase in quit rate among
followupthequitrateremainedatalmostthesamelevel the medium level smokers did not achieve statistical sig-
intheintervention–1group(10.7%),whereasintheinter- nificance probably due to small sample size. However,
vention–2 group the quit rate increased to 28.6% and the 72% of the medium level smokers shifted to low level
difference was statistically significant ( p <0.001). At the smokingat theendofsix months.
Table2Smokingstatusatsixmonthsfollowupusingcompletecaseanalysis
Outcome Interventiongroup–1(n=98) Interventiongroup–2(n=98) Unadjusted Adjusted pvaluefor
OR(95%CI) OR*(95%CI) adjustedOR
n(%) n(%)
Quitrate 14(14.3) 58(59.2) 8.7(4.3-17.4) 10.7(5.1-22.7) <0.001
Harmreduction 25(29.8) 20(50.0) 2.3(1.1-5.1) 2.6(1.1-5.8) 0.025
*Adjustedforage,education,occupation,presenceofanyotherchronicdisease,durationofdiabetes,volumeofcounselingsessionsreceivedandnumberof
sticksperday.QuitRate=Pointprevalenceabstinenceofnosmokinginthelastsevendays.Harmreduction=Reductionofsmoking(numberofsticksperday)
morethan50%ofbaselineuse.OR=OddsRatio.CI=ConfidenceInterval.
Thankappanetal.BMCPublicHealth2013,13:47 Page6of8
http://www.biomedcentral.com/1471-2458/13/47
Table3Smokingstatusatsixmonthsfollowupusingintentiontotreatanalysis
Outcome Interventiongroup-1(n=112) Interventiongroup-2(n=112) Unadjusted Adjusted pvaluefor
OR(95%CI) OR*(95%CI) adjustedOR
n(%) n(%)
Quitrate 14(12.5) 58(51.8) 7.5(3.8-14.7) 8.4(4.1-17.1) <0.001
Harmreduction 25(25.5) 20(37.0) 1.71(0.84-3.5) 1.9(0.8-4.1) 0.101
*Adjustedforage,education,occupation,presenceofanyotherchronicdisease,durationofdiabetes,volumeofcounselingsessionsreceivedandnumberof
sticksperday.QuitRate=Pointprevalenceabstinenceofnosmokinginthelastsevendays.Harmreduction=Reductionofsmoking(numberofsticksperday)
morethan50%ofbaselineuseOR=OddsRatio.CI=ConfidenceInterval.
Discussion the repeated 30 minutes counseling sessions for quitting
The study found that both the doctor’s message alone each time the patient visited the clinic. High quit rates
and counseling lead many patients to quit or signifi- of 50% at one year follow-up was reported by chronic
cantly reduce their smoking habit. This was true of both obstructive pulmonary disease (COPD) patients in a re-
low level smokers at baseline and high level smokers. cent study on smoking cessation buddies in COPD indi-
Quit rates in the intervention–1 group were 12.5% at cating that high quit rates are possible in chronic disease
month six compared to the baseline, indicating the im- patients[25].
portance of routine smoking cessation advice by doctors Generally,people includinghealth professionalsdonot
to all diabetes patients. Doctor’s cessation advice that is associate smoking with diabetes. Awareness of the asso-
disease specific is responded to better by patients than ciation between smoking and cancer, cardiovascular
general advise [21]. Our finding of a nearly nine times diseases and respiratory diseases are generally higher
higher quit rate (seven day abstinence from smoking) of [26]. However, it was reported in a previous study from
smokingintheintervention–2groupofdiabetespatients Kerala that close to two thirds (64%) of diabetes patients
compared to the intervention–1 group indicates that reported that smoking will not affect the disease and
trained non-doctor health professional increases the only 10% reported that smoking causes a lot of aggrava-
chancesapatient willquit significantly. tion of diabetes [27]. In our study both the doctor and
The only prior study from a developing country on the counselor used visual aids and diabetes specific
cessation among diabetes patients from Indonesia smoking cessation materials developed by the QTI to
reported a quit rate of 30% in the group that received motivatepatientstoconsiderquitting topreventcompli-
doctor’s advice and 37% in the group that received doc- cations fromdiabetes.
tor’s advice and counseling. Although this difference in Thisstudy found a dose response relationship between
quit rate was not statistically significant, the quit rate of counseling and quit rate. The quit rate increased signifi-
both groups’ was significantly higher at the six-month cantly from the one-monthfollow-uptothethird month
follow-up compared to the base line. The quit rate of and again at the six-monthfollow-up.Thisdemonstrates
52% in our intervention–2 group was much higher than the significance of repeat counseling at frequent inter-
the 37% in Indonesia. This could be due to several fac- vals for increasing quit rates and probably sustaining it.
tors, including the lower average number of cigarettes/ It is important to treat smoking as a chronic disease
bidis smoked per day in Kerala compared to Indonesia, understanding the nature of addiction, possibility of
the highly educated population in Kerala, better imple- relapse and the need for continuum of care [28]. The
mentation of the national tobacco control program and doctor employed the 2As (ask and advise) in their brief
intervention lasting three minutes. Non-doctor health
professionals who were trained as cessation counselors
employed all five ‘A’s and 5 ‘R’s adapted to the Indian
Table4QuitRateatsixmonthsbybaselinelevelof
context during their 30 minute counseling sessions.
smoking:Intentiontotreatanalysisresults
Theirassistingpatientstorecognizetherisksofsmoking
Baseline Quitrate Pvalue
levelof and benefits of quitting, and to face physical, psycho-
Interventiongroup-1 Interventiongroup-2
smoking logical and social roadblocks to quitting and plan quits
N(%) N(%)
resultedinhigher quit ratesovertime.
Low1 5/32(15.6) 25/39(64.1) <0.001 The Indian Institute of Diabetes in Kerala, one of our
Medium2 4/30(13.3) 8/24(33.3) 0.105 study sites for this study, has taken note of the outcome
High3 5/50(10.0) 25/49(51.0) <0.001 of this study and is currently planning to incorporate
smoking cessation counseling as a routine activity in
Total 14/112(12.5) 58/112(51.8) <0.001
their diabetic clinics and advise the State Government to
1Smoked1–5sticks(cigarettes/bidis)perday,2smoked6–10sticksperday,
3smokedmorethan10sticksperday. follow thispractice.
Thankappanetal.BMCPublicHealth2013,13:47 Page7of8
http://www.biomedcentral.com/1471-2458/13/47
Limitationsofthestudy Acknowledgements
Our study followed the patients only for six months, the TheQuitTobaccoInternationalProjectissupportedbyagrantfromthe
FogartyInternationalCentreoftheUSNationalInstitutesofHealth
outcomes wereselfreported and we did not conductbio-
(RO1TW005969-01).WewouldliketothanktheIndianInstituteofDiabetes,
chemical verification due to resource constraints. How- TrivandrumandtheMedicalTrustHospitalPandalam,Pathanamthittafor
ever, it has been reported by the Society for Research on helpingustoconductthisstudyintheirdiabeticclinics.WealsothankMr.
MadhuandMs.Lavanya,thecounselorsinthisstudy,fortheircounseling
NicotineandTobaccosubcommitteeonbiochemicalveri-
servicesanddatacollection.
fication that for population based trials with low demand
situation biochemical verification may not be necessary Authordetails
1AchuthaMenonCentreforHealthScienceStudies,SreeChitraTirunal
[29].Thecounselorswhoassessedtheoutcomeswerenot
InstituteforMedicalSciencesandTechnology,Trivandrum,Kerala,695011,
blindedtotheallocationgroupswhilecollectingfollowup India.2QuitTobaccoIndiaProject,AchuthaMenonCentreforHealthScience
data,althoughthestatisticianswhoanalyzedthedatawere Studies,SreeChitraTirunalInstituteforMedicalSciencesandTechnology,
Trivandrum,Kerala,695011,India.3MedicalTrustHospital,Pandalam,
blinded. Since zero percent of women in Kerala smoked
Pathanamthitta,Kerala,India.4FamilyMedicineandPublicHealth,University
they were excluded from this study [3]. With a large pro- ofArizona,Tucson,Arizona,USA.
portion of diabetic smokers seemingly ineligible for the
Received:16July2012Accepted:15January2013
study, another limitation is that the results may apply to
Published:18January2013
only specific smokers (male, literate, clinical care at the
samesiteoversixmonths).
References
1. RegistrarGeneralofIndia:CensusofIndia.2011.http://www.censusindia.gov.in.
Conclusion 2. InternationalDiabetesFederation(IDF):IDFDiabetesatlas.5thedition.
Brussels:InternationalDiabetesFederation;2011.
All doctors should routinely ask and advise diabetes
3. InternationalInstituteforPopulationSciences(IIPS):GlobalAdultTobacco
patients to quit smoking, including calling attention to SurveyIndia(GATSIndia),2009–2010.Mumbai:MinistryofHealthandFamily
the complications that are more likely if they continue Welfare,GovernmentofIndia,NewDelhi;2010.
4. JhaP,JacobB,GajalakshmiV,GuptaPC,DhingraN,KumarR,SinhaDN,
to smoke. We have demonstrated that a brief interven-
DikshitRP,ParidaDK,KamadodR,BorehamJ,PetoR:RGI-CGHR
tion by doctors is likely to result in a quit rate of about Investigators:anationallyrepresentativecase–controlstudyofsmoking
10–13%. If this brief interventionis further supported by anddeathinIndia.NEnglJMed2008,358:1137–1147.
5. WilliC,BodenmannP,GhaliWA,FarisPD,CornuzJ:Activesmokingand
counseling sessions by a trained non-doctor health pro-
theriskoftype2diabetes:asystematicreviewandmeta-analysis.JAMA
fessional, the patients were able to reach a 52% quit rate 2007,298:2654–2664.
at the six-month follow-up. Even among the high level 6. EliassonB:Cigarettesmokinganddiabetes.ProgCardiovascDis2003,
45:405–413.
smokers the quit rate at six months was 51%. And of
7. FoyCG,BellRA,FarmerDF,GoffDCJr,WagenknechtLE:Smokingand
those who didnot quit,atleast25%engaged inharmre- incidenceofdiabetesamongU.S.adults:findingsfromtheinsulin
ductionbyreducing their former smoking levels to more resistanceatherosclerosisstudy.DiabetesCare2005,28:2501–2507.
8. InternationalDiabetesFederation:Positionstatement-diabetesandtobacco
than 50% of baseline. Counseling sessions using disease
use.2003.http://www.idf.org/position-statement-diabetes-and-tobacco-use.
specific diabetes messages and culturally sensitive use of 9. AmericanDiabetesAssociation:Smokinganddiabetes:position
the five ‘A’s and five ‘R’s cessation protocol is an effica- statement.DiabetesCare2004,27:S74–S75.
10. Al-DelaimyWK,MansonJE,SolomonCG,KawachiI,StampferMJ,Willett
cious way of reducing smoking, an important risk factor
WC,HuFB:Smokingandriskofcoronaryheartdiseaseamongwomen
that significantly increases the chances of life threaten- withtype2diabetesmellitus.ArchInternMed2002,162:273–279.
ing,debilitatingandcostly diabetescomplications. 11. GiordaCB,AvogaroA,MagginiM,LombardoF,MannucciE,TurcoS,
AlegianiSS,RaschettiR,VelussiM,FerranniniE,DAIStudyGroup:Incidence
If this system of brief intervention by the doctors sup-
andriskfactorsforstrokeintype2diabeticpatients:theDAIstudy.
ported by a counseling session by a non-doctor health Stroke2007,38:1154–1160.
professional can be incorporated in the national health 12. PaetkauME,BoydTA,WinshipB,GraceM:Cigarettesmokinganddiabetic
retinopathy.Diabetes1977,26:46–49.
system, a substantial proportion of diabetes complica-
13. EasonSL,PetersenNJ,Suarez-AlmazorM,DavisB,CollinsTC:Diabetes
tionsdue tosmoking canbeprevented. mellitus,smoking,andtheriskforasymptomaticperipheralarterialdisease:
whomshouldwescreen?JAmBoardFamPract2005,18:355–361.
Competinginterests 14. PetersDH,RaoKS,FryattR:Lumpingandsplitting:thehealthpolicy
Theauthorsdeclarethattheyhavenocompetinginterests. agendainIndia.HealthPolicyPlan2003,18:249–260.
15. ThankappanKR,ShahB,MathurP,SarmaPS,SrinivasG,MiniGK,
Authors’contributions DaivadanamM,SomanB,VasanRS:Riskfactorprofileforchronicnon-
Thecontributionofeachoftheauthorsislistedbelow:KRT:Conception, communicablediseases:resultsofacommunity-basedstudyinKerala.
design,revisionofthemanuscriptwithcriticalcontributionforintellectual India.IndianJMedRes2010,131:53–63.
content,andgivingfinalapprovaloftheversiontobepublished,GKM: 16. ReddyKS,PrabhakaranD,ChaturvediV,JeemonP,ThankappanKR,
Conception,design,analysisandinterpretationofdata,andrevisionofthe RamakrishnanL:Methodsforestablishingasurveillancesystemfor
manuscriptforintellectualcontent,MD:Conception,design,analysisand cardiovasculardiseasesinIndianindustrialpopulations.BullWorldHealth
interpretationofdataanddraftingofthemanuscript,GV:Conception, Organ2006,84:461–469.
design,andrevisionofthemanuscriptwithcriticalcontributionfor 17. ThresiaCU,ThankappanKR,NichterM:Smokingcessationanddiabetes
intellectualcontent,PSS:Analysis,interpretationofdata,andcontributionto controlinKerala,India:anurgentneedforhealtheducation.HealthEduc
criticalrevisionofthemanuscriptforstatisticalaccuracy,MN:Conception, Res2009,24:839–845.
design,analysisandinterpretationofdata,andrevisionofthemanuscriptfor 18. HohansonJM,AndersonRL,HennrikusDJ,LandoHA,KendallDM:
intellectualcontent.Allauthorsreadandapprovedthefinalmanuscript. Integratedtobaccocessationcounselinginadiabetesself-management
Thankappanetal.BMCPublicHealth2013,13:47 Page8of8
http://www.biomedcentral.com/1471-2458/13/47
trainingProgram:arandomizedtrialofdiabetesandreductionof
tobacco.DiabetesEduc2006,32:562–570.
19. NgN,NichterM,PadmawatiRS,PrabandariYS,MuramotoM,NichterM:
BringingsmokingcessationtodiabetesclinicsinIndonesia.ChronicIlln
2010,6:125–135.
20. RaoM,RaoKD,KumarAK,ChatterjeeM,SundararamanT:Human
resourcesforhealthinIndia.Lancet2011,377:587–598.
21. NichterM,ProjectQuitTobaccoInternationalGroup:Introducingtobacco
cessationindevelopingcountries:anoverviewofProjectQuitTobacco
International.TobControl2006,15:12–17.
22. FioreMC,JaénCR,BakerTB,BaileyWC,BenowitzN,CurrySJ,DorfmanSF,
FroelicherES,GoldsteinMG,HealtonCG,HendersonPN:Treatingtobacco
useanddependence:2008Update:QuickReferenceGuideforClinicians.
Rockville,MD:U.S.DepartmentofHealthandHumanServices.PublicHealth
Service;2009.
23. RamachandranA,MaRC,SnehalathaC:DiabetesinAsia.Lancet2010,
375:408–418.
24. AravindanKP:KeralaPadanam(inMalayalam).Kozhikodu,India:Kerala
SastraSahityaParishad;2006.
25. CoxK:SmokingcessationbuddiesinCOPD.NursTimes2011,107:22–23.
26. PokhrelBR,ThankappanKR,MiniGK,SarmaPS:Tobaccouseamonghealth
professionalsandtheirroleintobaccocessationinNepal.PrevControl
2006,2:117–125.
27. ThankappanKR,ThresiaCU:TobaccouseandsocialstatusinKerala.
IndianJMedRes2007,126:300–308.
28. SteinbergMB,SchmelzerAC,LinPN,GarciaG:Smokingasachronic
disease.CurrCardioRiskRep2010,4:413–420.
29. SRNTSubcommitteeonBiochemicalVerification:Biochemicalverification
oftobaccouseandcessation.NicotineTobRes2004,4:149–159.
doi:10.1186/1471-2458-13-47
Citethisarticleas:Thankappanetal.:Smokingcessationamong
diabetespatients:resultsofapilotrandomizedcontrolledtrialinKerala,
India.BMCPublicHealth201313:47.
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit