THE ASSOCIATION BETWEEN ROTATING SHIFT WORK AND METABOLIC SYNDROME AMONG HOSPITAL

(1)

THE ASSOCIATION BETWEEN ROTATING SHIFT WORK AND METABOLIC SYNDROME AMONG HOSPITAL

EMPLOYEES

A dissertation submitted in partial fulfilment of the rules and

regulations for the M.D. General Medicine Branch I Examination

of the Tamil Nadu Dr M.G.R. UNIVERSITY, CHENNAI to be

held in April, 2016.

(2)

Declaration Certificate

This is to certify that the dissertation “The association between rotating shift work and metabolic syndrome” is a bonafide work of Dr. Santhosh Kumar. E carried out under our guidance towards the M.D. Branch I (General Medicine) Examination of the Tamil Nadu Dr M.G.R. University, Chennai to be held in April, 2016

SIGNATURE:

Dr. Reginald Alex

Professor, Department of Medicine

Christian Medical College, Vellore, 632004, India

Dr. Anand Zachariah

Professor and Head of Department, Department of General Medicine Christian Medical College, Vellore, 632004, India

(3)

Declaration Certificate

This is to certify that the dissertation titled “The association between rotating shift work and metabolic syndrome” which is submitted by me in partial fulfilment towards M.D. Branch I (General Medicine) Examination of the Tamil Nadu Dr M.G.R.

University, Chennai to be held in April, 2016 comprises only my original work and due acknowledgement has been made in text to all material used.

SIGNATURE:

Santhosh Kumar. E

PG Registrar, Department of General Medicine Christian Medical College, Vellore, 632004, India

(4)

(5)

(6)

(7)

ACKNOWLEDGEMENTS

I would like to express my gratitude to Dr Reginald Alex for the opportunity to complete my post graduate thesis under his able guidance. Special thanks to Dr.

Reginald Alex for being so approachable, friendly and simple these qualities were of great support.

I would also like to thank my co- guides Dr. Rini.B and Dr. Henry Kirupakaran for their contribution. My sincerest gratitude goes to Dr. Antonisamy.B, my co-guide and the biostatistician, for his continued help. I would like to thank Mr. Prakash.R, junior statistician who has been very patient and kind in clearing all my doubts. I am grateful to the Dept of Medicine for their valuable suggestions and support. I would like to thank my immediate senior batch for their help, especially Dr. Deepti Bal for inspiring me to do a study on the staff.

I am forever indebted to the staff who agreed to be part of this thesis. Their contribution to the study is invaluable.

Lastly, I would like to thank God, family and my friends for the help.

(8)

Introduction:

There was an era where infections were ruining the health of mankind and that led to significant advancement in science and brought forth the germ theory, improved sanitation, and antibiotics. When science thought it had its supremacy over human well being, slowly and steadily were creeping in large numbers the non communicable diseases. The present generation is forced into a lifestyle where majority are involved in sedentary activities for livelihood and recreation. The general population is growing old posing new challenges such as diabetes, hypertension, ischemic heart disease, stroke and cancer.

Metabolic syndrome is a constellation of various non communicable diseases.

For the diagnosis of metabolic syndrome the necessary feature is central obesity along with at least two of the four other metabolic derangements such as elevated serum triglycerides, decreased HDL, elevated fasting glucose and elevated blood pressure(1).

Metabolic syndrome is a disease on the rise both in the developed countries like United States of America where the prevalence is 34.5%(NHANES 2002) and in developing Asian countries such as India where the prevalence is 41.4% (2).

Metabolic syndrome is a forerunner of various more serious metabolic diseases. There is also increasing evidence to suggest metabolic syndrome per se increases the risk of cardiovascular events, cerebrovascular events and also mortality

(11)

the ancient English thesaurus is becoming a common way of life. Shift work is increasing globally both in the West and also in the East(3)

In the European Union as per 2014 statistics between the ages of 15 years to 65 years 64.9% of people are employed(4) and 15% to 25% of them perform shift work(5). Shift work has been studied in detail recently in relation with health and well being. Shift work is the norm of work where different workers replace each other in the same place of work beyond the routine hours of work and work round the clock in rotation to keep the work place operating for 24 hours. There have been

associations made between shift work and weight gain, diabetes mellitus, hypertension, peptic ulcer, menstrual irregularities, metabolic syndrome,

cardiovascular events and even breast cancer(6). Association of shift work and metabolic syndrome has gained particular interest and there is data to suggest increased risk of metabolic syndrome among the shift workers (7)

Hospital is an ideal setting with employees representing both shift working and daytime working groups. Health care workers are also group of individuals who are at risk for developing metabolic syndrome. Sedentary life style, increased strenuous working conditions, less physical activity, disturbed sleep, improper eating habits and lack of exercise can eventually lead to developing metabolic syndrome. Hence, this adds to the economic burden for the individual as well as the institution for treatment of upcoming ominous diseases which requires life-long medications. The worry of unhealthy employee is also that more illness leads to work absenteeism and decreased productivity from the individual.

(12)

Thus, prevention of metabolic syndrome in the high risk group is needed. However, little is known about metabolic syndrome and its association with shift workers in India. There is also paucity of literature about metabolic syndrome among health care workers in the world(8)(9) and none in India. The questions whether the same

traditional risk factor for metabolic syndrome applies to health care workers or do they have different risk factors?

Hence, this study aims to look at the prevalence of metabolic syndrome among the health care workers in our tertiary level hospital, identify the risk factors and also to study the association between shift work and metabolic syndrome. The information will help us in health planning activities and strategies to curb the development of modifiable risk factors of metabolic syndrome.

(13)

Aims and objectives

AIM: To measure the association between rotating shift work and metabolic syndrome among hospital employees.

OBJECTIVES:

1. To study the prevalence of metabolic syndrome among shift working and daytime working employees in Christian Medical college, Vellore

2. To assess the association between rotating shift work and metabolic syndrome

(14)

Review of literature

Definitions of metabolic syndrome and shift work

Metabolic syndrome

Metabolic syndrome is a combination of medical disorders which, when occurring together, increase the risk of developing cardiovascular disease and diabetes mellitus.

The pivotal components are central obesity and insulin resistance.

International Diabetic Federation (IDF, 2005) new worldwide definition(1)

A person is defined as having metabolic syndrome if he/she has:

-Central obesity (defined by waist circumference as per ethnic group, South Asians:

Male-90cms or more and Female-80cms or more)/If BMI is more than 30 kg/m2 waist circumference need not be measured.

Plus any two out of the four of the following:

1. Raised trigylcerides-150 mg% or above or on treatment for this abnormality 2. Reduced HDL cholesterol- below 40 mg% in males or below 50 mg/dl for

females or on treatment for this abnormality

3. Raised blood pressure- Systolic blood pressure equal to or more than 130 mmHg or diastolic blood pressure equal to or more than 85 mmHg or on

(15)

Shift work

The International Labour Office (International Labour Organization, 1990) defines working in shifts as “a method of organizationof working time where workers

succeed one another at the same workplace so that the establishment can function longer than the hours of work of individual workers.”

The European Council Directive93/104 (1993) declares “regarding certain aspects of the organisation of working time, shift work is a method of organising work in shifts whereby workers succeed one other at the same work place according to a certain pattern. Shift worker shall mean any worker whose work schedule is part of shift work.”

In common terms somebody can be referred to do‘shift work’ or ‘rotating shift work’

if he or she does duties with timing such as 8 am to 4pm, 4pm to 11 pm, 11 pm to 8 am and rotates between these shifts in a scheduled manner.

Day worker is one who performs his occupational activities usually between the mornings to evening hours for example 7am _8am to 5pm _6pm.

(16)

History and co-evolution of shift work and metabolic syndrome

Shift work has been around for the most of human history, but it was mainly limited to soldiers and sailors.

In late 19^thCentury industrialisation lead to the surge of shift work in Europe.

Countries in the West ran operations around the clock to maximise production and profits.

During the First and Second World Wars shift work was a mean to help the war effort.

After the War, many different kinds of work plants continued to work for 24 hours, mostly the food and service industries. The latest growth industry using shift workers is "Call Centre" industry, where information and products ordered can be done

anytime of the day or night.

Healthcare system is no exception to the rule. Patient care, laboratories, pharmacies, and housekeeping have to run round the clock.

In recent times advancement in science and immediacy of human needs most occupations have turned into 24x7 pattern of work to cater the expectation of the consumers and to survive the competition in the market.

Metabolic syndrome, contradictory to the common belief is not a new entity. In the Mid 1700s, Italian physician Morgagni had made mention of the association between abdominal fat, systemic hypertension, atherosclerosis and elevated uric acid

(17)

physician, was the first to identify adiposity of the superior part of the body and its association with diabetes and cardiovascular diseases. The term plurimetabolical syndrome was in use in 1960’s. In 1980’s,hyperglycaemia, obesity, dyslipidemia and systemic hypertension was called as“X syndrome”. Reaven G., an

endocrinologist, established the pathogenic relationship with the peripheral insulin- resistance and it was addressed to as the“insulin– resistance syndrome”. This came to be known as metabolic syndrome only recently and World health Organisation in 1988 defined it for the first time(10).

It has to be more than mere co-incidence that the evolution of shift work and

metabolic syndrome follow the same timeline. The last two centuries have changed the lifestyle of human beings and their nature of disease and well being!

Magnitude of shift work and Metabolic Syndrome–Global perspective and Indian outlook

Shift work is the new norm of occupation which has influenced all nations of the world. Data from the west suggest 15-25 % of population in Europe are working in one or the other form of rotating shift work(5). In a report published by the

International Labour Organisation (ILO) in 2007 shift work was noted to be on a increasing trend with similar patterns in the East and the West. It was prevalent across all economies–developed, developing and under developed countries(3) . In Asia, data from China show that 36.1% of employees are involved in shift work.

(18)

The data on the percentage of employees involved in shift work is not available from India. However considering the growing industrialisation and work outsourcing to India from various Western nations there are no reasons why the rates of shift work should be any less in India as compared to our neighbour China.

Metabolic syndrome is the new epidemic of the 21^stcentury, the prevalence in US was 25.54 % in the year 2000 which with increasing intervention and awareness has

dropped to 22.90% by the year 2010(11). In the Asian countries the prevalence was comparatively low 10-15% on an average, however Koreans and Indians had

prevalence as high as in the US(12). Indian data from an urban population showed a strikingly high prevalence of 31.6% subjects, 22.9% in men and 39.9% in women (13).

India, a developing nation is challenged by this inevitable lifestyle change and the health hazards that come along with it. It is extremely pertinent that the two are studied together.

Pathophysiology

The pathophysiology of Metabolic syndrome revolves around central obesity. There are inherited, ethnic and acquired factors that lead to metabolic syndrome. Increased

(19)

There is increased lipolysis leading to increased free fatty acid in blood secondary to insulin resistance. Insulin resistance per se can lead to elevation of blood pressure. As a result there is a constellation of metabolic abnormalities that lead to increased cardiovascular events(14)

Shift work has been associated with adverse health effects. Shift work, no doubt affects the quality of sleep(15). Disturbed sleep in turn affects the human behaviour and the altered circadian cycle interferes with the various metabolisms(16)(17).

Duration of sleep as a risk factor for development of diabetes has been studied where both short (less than 5 hours) and long (more than 9 hours) have shown to increase the risk of developing diabetes(18). The theories proposed for shift work predisposing to ill health includes: changes in circadian rhythm, behavioural changes and social

factors. Change in circadian rhythm and artificial light at workplace at night have been shown to alter the satiety centre causing increased food intake(17). Change in the normal sleep-wake cycle also has shown to impact glucose and lipid metabolism adversely. Animal studies and voluntary human studies have revealed glucose

intolerance in sleep deprived subjects(19). Sleep deprivation studies have also shown increased circulation levels ghrelin and reduced levels of leptin(17). There is a

proportional rise in shift work and incidence of metabolic syndrome in industrialized countries such as USA. There is also arguing evidence that with time the body

accommodates and most of the hormonal disequilibrium is a transient process(20).

Refer figure 1 for summary of the pathophysiology of metabolic syndrome.

(20)

Figure 1: Pathophysiology of metabolic syndrome

Shift work could precipitate stress occupational and social, secondary to disturbed sleep, odd working hours and separation from family. There is evidence from Indian subcontinent that majority, of workers in rotating shift were stressed(21)(22).

However shift work per se as a stressor is still being debated, as larger studies have shown similar stress pattern among regular day workers and rotating shift workers(23)

The hypotheses of shift workers having lifestyle changes that predispose them to cardiovascular events have been well studied. These studies concluded that shift workers had higher odds of smoking, lower odds of smoking cessation, higher binge eating, lower physical activity and higher weight gain. Life style changes that

predispose to weight gain thereby metabolic syndrome and other direct risk factors

(21)

Figure 2: Shift work and related health issues(26)

Impact of metabolic syndrome

Though there are arguments about the existence of a syndrome such as ‘metabolic syndrome’, no doubts exist about the impact of the constituents of metabolic syndrome occurring together in an individual. The odds of having a myocardial infarction or stroke are doubled in people with metabolic syndrome(27). Figure 3 summarizes the pooled relative risk for a person with metabolic syndrome to develop various cardiovascular events. With the increasing prevalence of this epidemic not just in developed countries, but also in developing countries such as India, early

recognition and containment of the causative factors would decrease the mortality, morbidity and economical burden of these diseases. The economical impact of

(22)

metabolic syndrome is well studied. It is reported that employees with metabolic syndrome use more medical allowances and have decreased productivity owing to the diseases(28)(29)(30).

Figure 3(2)

Metabolic syndrome in shift workers: truth or myth ?

The idea of shift work predisposing to metabolic syndrome, though relatively new has been studied in the past. Shift work and its relation to weight gain, increasing waist circumference, diabetes mellitus, hypertension and dyslipidemia have been

documented in prior studies from Belgium, Finland and Korea(31)(32)(33). The study from Korea investigated the shift duration per day and the individual parameters of

(23)

These inconsistent results create doubts whether it is really the shift work that predisposes to metabolic syndrome among the shift workers?(33)

Large longitudinal studies have shown that rotating shift work is an individual risk factor for weight gain, with shift workers having a significant rise in their BMI(34).

Rotating shift work has been associated with increased incidence of obesity and diabetes mellitus as shown in two large prospective cohorts in Nurses’Health Study(35). Most recent study done in South Korea on female shift workers show an odds ratio of 6.30(CI 95 % 1.24 to 32.15) for shift workers developing metabolic syndrome(36). There is a temporal association between duration of exposure to shift work with higher odds of developing metabolic syndrome(37).

More and more studies are looking at associations and etiological factors for metabolic syndrome as it is a predecessor for more sinister diseases such as myocardial infarction or stroke. When any of these vascular events have once occurred they have already caused a marked damage to the end organ. Hence the challenge now is to curb the precursor of such evils that is metabolic syndrome.

The systemic review that searched for the association between shift work and metabolic syndrome was published in 2013. A total of 10 studies were available for systematic review. The results were varied. Out of the ten studies eight had results showing positive association. However it is to be noted seven out of the eight had not accounted for sleep and other confounding factors. The three studies that had looked at sleep and other confounding factors had not shown any significant association(38).

(24)

A more recent publication in this area of research was a meta-analysis which had included 13 studies and the pooled relative risk was 1.57 (95% 1.25-1.98) which demonstrated that there is a risk of the rotating shift workers to develop metabolic syndrome(7).

Cardiovascular events in shift workers

Well before the association between metabolic syndrome and shift work was studied, the association between myocardial infarction and shift work was well established. In 1999 a case control study showed increased odds ratio 1.5( 1.3 to 1.9) in men and 1.7 (1.3 to 2.4) in women shift workers for developing myocardial infarction(39).

Following which there were multiple cross sectional and prospective studies showing an increased odds ratio and relative risk respectively for cardiovascular events among shift workers.

In 2012 a systematic review and meta-analysis was done which showed a positive association between vascular events among shift workers with risk ratio 1.23, 95%

confidence interval 1.15 to 1.31(40).

Biomarkers of predicting metabolic syndrome

(25)

biomarkers studied so far include alanine amino transferase (ALT), aspartate amino transferase (AST), gamma glutamyl transferase (GGT) and uric acid. They were studied in relation with normal individuals who were followed up prospectively. It was shown that men with higher GGT had higher incidence of metabolic syndrome and type II diabetes mellitus(41)(42). There have also been studies proving the association between ALT and metabolic syndrome in patients with diabetes without sonographically evident fatty liver(43). Though the exact mechanism by which these enzyme elevation leads on to metabolic syndrome still remains to be elucidated there have been various mechanisms proposed. The most accepted among them is that the fat deposition in the liver leads to enzyme elevation which therefore becomes a surrogate marker for forth coming insulin resistance. The other explanations is that enzyme elevation correlate with subclinical inflammation which correlates with altered insulin signalling and thereby insulin resistance.

Relevance of the topic and justification for the study

The magnitude of the threat to health and well being that a seemingly benign disease such as metabolic syndrome is causing is quite humongous. The rapidity at which it is evolving across the continents is just a mere reflection of the changes in the lifestyle of people. Though it seems quite established in the light of the above discussion that shift work has an association with metabolic syndrome, it is to be noted that most of the data so far have been from the West and industrialized Asian countries.

(26)

the world, there is paucity of literature with respect to the relation of shift work and metabolic syndrome. To our best knowledge, this study would be the first of its kind in our country.

Considering that health care system has a higher population of rotating shift workers as compared to other industries it was planned that this study be done in Christian Medical College and Hospital, Vellore. Shift work is an inevitable system of work in the present setting. If it poses any health risks to the workers it is important to identify and bring about measures to limit the same.

The study also hopes to add valuable information with regard to “what is the

prevalence of metabolic syndrome among our staff?” The prime objective isto know

“if our staffworking in shift system are at a higher risk of developing metabolic syndrome?” The investigators sincerely hope that this information would help the hospital administration in its health planning and prevention schemes towards its staff.

(27)

Methodology

Setting

The study was conducted in Christian Medical College, Vellore (CMC). It a tertiary care level hospital situated in Vellore, Tamil Nadu. According to CMC, Vellore 2011 statistics there are 8078 employees. These staff are spread over 3 campuses and 2 secondary level hospitals. The staff employed in CMC were as follows:

1. Doctors–1187

2. Nursing Staff–2491

3. Technicians /Non-Medical workers–1860

4. Housekeeping staff and Attendants–1486

The study will be done among employees in Christian Medical College and Hospital. The participants would be those who have functioned in their respective norm of work, shift or daytime for a period of more than 5 years.

Time of recruitment

The study participants were recruited between the months of August 2014 to April 2015.

(28)

Study design

A cross sectional study to compare the prevalence of Metabolic syndrome among Shift workers and Daytime workers between the age group of 25 years to 50 years. Figure 4 is the schematic representation of the study design.

Figure 4: Schematic representation of the study design

Total hospital attendants with 5 years of experience

and between age group 25-50 years

Shift working staffs Randomly selected group 1(79)

Self adminstered questionnaire

Height, weight, waist circumference measurements

Fasting blood glucose, triglycerides and HDL

Prevalence of metabolic syndrome among the shift

working staff

Day time working staffs Randomly selected group 2(79)

Self adminsitered questionnaire

Height, weight, waist circumference measurements

Fasting blood glucose,triglycerides and

HDL

Prevalence of metabolic syndrome among the daytime working staff

(29)

Inclusion criteria and Exclusion criteria

Inclusion criteria

1. Staff of Christian Medical College, Vellore who have five or more years of continuous work experience in shift work pattern or daytime pattern between the age group of 25 years to 50 years

Exclusion criteria

1. Not willing to consent

2. Pregnant women

(30)

Shift working group:

This group comprises of the following staff:

a. Attendants working in ICUs, wards and Emergency department

b. Laboratory technicians working in Blood bank, Virology, Clinical pathology and Clinical biochemistry

c. Anaesthesia technicians

d. Radiology technicians

e. Receptionists

f. Critical care therapists working in all the ICUs

f. Pharmacists

Note: Nurses were excluded as the Nursing Superintendent’s office refused permission.

Daytime working group:

(31)

b. Laboratory technicians working in Microbiology, Neurophysiology, Neurochemistry, haematology, Wellcome laboratory and histopathology

c. Secretaries of varies offices

d. Office staff of Accounts, Billing section,

e. Doctors in non clinical specialties

f. Staff of transport department

g. Librarians

h. Staff of Chaplaincy

i. Physiotherapists and occupation therapists

j. Medical records department staff

Note: Nurses were excluded as the Nursing Superintendent’s office refused permission.

Sample size calculation

Sample size was calculated based on values from the result of a study done in Japan which studied the association of shift work and metabolic syndrome(36).

The sample size was calculated for the study to have Type I (α) error of 5% and a power of 80%(β=0.2).

(32)

The calculation was done using the formula for sample size calculation for comparing two proportions.

n = (Zα/2+Zβ)2 * (p1(1-p1)+p2(1-p2)) / (p1-p2)2

Where,

n is the sample size

Zα/2 is the critical value of the Normal distribution at α/2(for a confidence level of 95%, α is 0.05 and the critical value is 1.96)

Zβ is the critical value of the Normal distribution at β (for a power of 80%, β is 0.2 and the critical value is 0.84)

p1 and p2 are the expected sample proportions of the two groups.

In the above formulae the values used are

P1 is the proportion of cases among the study group- shift workers=15.3%

P2 is the proportion of cases among the comparison group-daytime workers=2.8%

The sample size thus calculated from the above formula is 79 in each group

A sample size of 158 is needed to detect a difference of 12.5% of metabolic

(33)

Sample selection

The study planned initially included nurses. However the Nursing Superintendent’s office refused permission for their participation. Therefore the study was planned to be conducted among the hospital attendants and housekeeping staff. The General Superintendent permitted their participation in the study if there is participation from other grades of workers as well. Therefore the Medical superintendent’s permission was also taken. The study population was modified to include all the grade of workers from doctors, technicians, attendants, housekeeping staff, pharmacists and office staff.

Doctors who were in clinical specialties had a pattern of work which could not clearly be classified as shift or daytime working as per our definition and therefore were excluded. The non clinical doctors were included.

The list of the employees between the age group of 25 years to 50 years from the above mentioned pool, who have served in Christian Medical College and Hospital (CMC), Vellore for 5 years or more consecutively in the same department, was collected from the CHIPS (Computerized Health Information Processing System) department with permission from appropriate authorities.

The list of the CMC staff who have served for 5 years or more in the same pattern of work and within the age group of 25 years to 50 years will be divided into two groups:

shift workers(group 1) and daytime workers(group 2).

The samples for the study were chosen proportionately from their respective designations using systematic random sampling technique.

(34)

When a selected participant fails to consent or cannot participate in the study due to any reason the person serially next will be approached for his/her participation in the study.

Institutional review board and ethics committee approval

The protocol was submitted before the initiation of study and was approved by the Institutional Review Board and Ethics Committee (IRB Min No: 8888, dated

9/6/2014). A copy of the approval letter is attached in the beginning of the dissertation (kindly refer page 5).

Informed consent was obtained from all the participants prior to their entry into the study (Appendix C). All the variables of interest were collected through a data abstraction form (Appendix B) which was duly filled by the participant.

Funding

The study is funded by the institution, Christian Medical College and Hospital, Vellore. The funding was referred as the‘Fluid Research grant’. The Hospital institution and the management had no role in the data collection, data analysis or interpretation of the results.

(35)

Data collection

The protocol was submitted before the initiation of study and was approved by the Institutional Review Board and Ethics Committee (IRB Min No: 8888). Patients were selected as mentioned above. The selected patients were approached for participation in the study in their respective work place by the principal investigator.

Informed consent was obtained from all the participants prior to their entry into the study. All the variables of interest were collected through a data abstraction form (Appendix B) which was filled by the participant.

The following measurements were taken as per the below mentioned manner by the principal investigator

 Height

 Weight

 Body mass index(BMI)–derived value

 Waist circumference

 Blood pressure

Measurement of Weight:

Weight was measured using a calibrated weighing machinefrom the company ‘Krups’

which had a maximum capacity of 120 Kilograms with 500 gm divisions. The same weighing scale was used for all the participants through the study. The participant’s

(36)

weight was measured without foot wear and with the clothing that the participant had worn for work.

Measurement of Height:

Height was measured using a metallic measuring tapefrom the company ‘Hi-Wide’

which had a length of 5 metres and 1mm divisions. The participant was asked to remove foot wear and stand with his/her back against the wall with feet together .The back of the feet, buttocks, shoulders, and the back of the head was touching the wall and the participant was asked to look straight ahead with his/her chin tucked in. Once in good position, a wooden ruler was pushed against the wall in perpendicular

direction, and then pushed down to the participants head. The head level was marked with pencil on the wall.

Now using the metallic measuring tape from the ground up to the mark was measured to find the height of the patient. Care was taken to make sure the tape was straight during the measurement.

Body Mass index was thus calculated from the weight and height.

The formula used:

BMI = Weight in Kilograms / (Height in metres x Height in metres)

(37)

Measurement of Waist circumference:

Waist circumference was measured when the study participant was standing,

breathing normally and with clothing in the abdominal region. The waist measurement was taken midway between the costal margin and the iliac crest, using a 1.0-m flexible non-stretch inch tape measure with 1mm divisions.

Measurement of Blood Pressure:

Blood pressure of the participant was measured after a10 minutes interaction with the participants with a duly calibrated‘Heine’ non-mercury sphygmomanometer with the subject sitting on a chair. The same sphygmomanometer was used for all the

participants through the study. The blood pressure was measured on the right arm with the arm rested on a firm surface such as a table. The systolic blood pressure was

estimated first by palpatory method and the exact systolic and diastolic blood pressure was recorded using the auscultatory method.

The data abstraction form had incorporated the following standardized pre- validated questionnaires:

1. General practice physical activity questionnaire (GPPAQ) for measuring the physical activity

2. Perceived stress scale (PSS) for assessment of level of stress

3. Pittsburgh sleep quality index (PSQI) for measuring the quality of sleep

(38)

Level of physical activity measurement:

General practice physical activity questionnaire (GPPAQ) was developed by NHS (National Health services), UK. The tool has been validated for use in

- Patients in the age group 16–74 years.

- Provides 4-level Physical Activity Index (PAI) categorizing patients as:

o Inactive

o Moderately Inactive

o Moderately active

o Active.

The questionnaire has been proven valid and the results are reproducible(44)(45).

This questionnaire was used for the assessment of the level of physical activity among the participants of this study.

Assessment of stress:

Cohen perceived stress scale (PSS) was used in the study data abstraction form as a tool to assess stress. PSS which included 10–item questionnaire in 5-point Likert

(39)

The Perceived Stress Scale (PSS) is a popular scale to measure stress and is used as a psychological instrument for measuring the level of stress.

It is a measure of the amount a person perceives a given situations in his/her life as stressful.

Questions are structured to measure how‘unpredictable’,‘uncontrollable’, and

‘overloaded’ the participantsfind their day to day situations and responsibilities.

The PSS is validated for use in the general community with education level of junior high school level.

The questions are simple to understand (46) .

Evidence for Validity: High scores in PSS have correlated well in prior studies in patients with the following problems

- Inability to quit smoking

- In diabetics with poor control of blood sugar levels

- Increased vulnerability to stressful life-event-elicited depressive symptoms

The below mentioned is the interpretation of the PSS score and the level of stress:

0-7 Much Lower than Average 8-11 Slightly Lower than Average

12-15 Average

(40)

16-20 Slightly Higher than Average 21 and over Much Higher than Average

Assessment of sleep quality:

The Pittsburgh Sleep Quality Index (PSQI) is an instrument designed to measure the quality and patterns of sleep in the adults. It differentiates between“poor”and“good”

sleep by measuring seven domains: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction over the last month. The participants rate each of these seven areas of sleep. Scoring of each of the answers is on a 0 to 3 scale, whereby 3

represents negative extreme on the Likert Scale. A sum of greater than 5 indicates a

“poor” sleeper. The PSQI has shown to have internal consistency and a reliability coefficient (Cronbach’s alpha) of 0.83 for its seven components.Numerous studies using the PSQI in a variety of adult subjects worldwide have supported high validity and reliability(47).

(41)

will not give their blood sample following night shift, but on the days when they perform day or evening shift work.

Laboratory analysis of the blood for the necessary parameters

The following biochemical methods were used to measure the following values in the Clinical Biochemistry lab, CMC, Vellore.

FASTING TRIGLYCERIDES (in mg/dl): enzymatic colorimetric

FASTING HDL (in mg/dl): enzymatic colorimetric, direct quantitative method, Roche automated clinical chemistry analysis

FASTING GLUCOSE (in mg/dl): hexokinase method, UV enzymatic colorimetric

(42)

Outcomes studied

Primary outcome

To calculate the prevalence of metabolic syndrome among shift working and daytime working employees.

Secondary

1. To calculate the odds of developing metabolic syndrome among shift working staff.

2. To individually assess the odds of developing hypertension, obesity, central obesity, impaired fasting glucose, decreased HDL cholesterol, elevated triglycerides and diabetes mellitus among the shift working staff.

(43)

Potential bias and methods to reduce the bias

Selection bias

The problem of selection bias was anticipated and addressed by using random sampling technique.

Information bias

The same data abstraction forms were used for data collection for all the participants of the study and there was no difference between the two groups in the method of data collection.

Confounding factors

1. Past history of shift work and its lasting effect on the metabolic status of the participant–was expected and patients who were employed only in a similar pattern of work for 5 or more years were included.

2. Sleep duration and quality These parameters were assessed 3. Physical activity in each participants and was adjusted 4. Stress levels for, by multiple logistic regression 5. Age was considered as an important parameter for metabolic diseases and the

baseline characteristics and outcome were assessed separately stratified for age.

Age was also adjusted for the outcomes by multiple logistic regressions.

(44)

Reporting bias

The response to specific questions of behaviour including alcohol consumption, smoking, stress and physical activity of the participants might be a more socially acceptable answer as perceived by them. To decrease the reporting bias the

participants were reassured about the confidentiality about their participation and the details they yield during the study.

Statistical analysis

Data was entered using Epidata version 3.1. The data was exported and analysed using Stata software version 13.

Baseline characteristics were summarized as counts and percentage for categorical variables and mean and standard deviations for continuous variables according to shift and daytime workers.

Chi square test was used to test the difference in the categorical variables according to the groups. Similarly independent student t test was used to test the mean difference in the continuous variables according to the groups.

Univariate analysis and multivariable logistic regression was used to test the

significance of demographic, lifestyle and clinical variables on metabolic syndrome.

(45)

Results

The study was done after obtaining the approval of the Institution review board. The study participants were recruited between the time period August 2014 and April 2015.

The total CMC employees’ list from the CHIPS (Computerized Health Information Processing System) department was obtained and the list of participants eligible for the study was processed. The participants were grouped as per their department and designation. The samples for the study were chosen by systematic random sampling.

The chosen employees were approached in both of the groups, 97 in the shift work group and 99 in the daytime working group. 17 from the shift working group and 19 from the daytime group were excluded for the reasons mentioned in the figure 5. It is to be noted that the list according to CHIPS had few participants aged 50 years who reported to be over 50 years at the time of approaching them for participation and hence they were excluded. Once a participant could not be included in the study the immediate next person from the list was chosen and approached for participation, thereby achieving the sample size.

There were eighty participants in both the groups with complete data forms,

anthropometric measurements and blood pressure reading for analysis. However there were 6 participants, 3 from each group who had incomplete laboratory parameters.

The baseline characteristics for the study participants comparing both the groups is represented in Table 1.

(46)

Figure 5: Study flow diagram

Baseline

Number of staff employed in CMC = 8078 (2011-12 statistics)

Number of staff fulfilling the study inclusion criteria = 2021

No. of daytime workers

included in the study =80

No. of daytime workers approached = 99

No. of shift workers

included in the study = 80

No. of shift workers approached = 97

Total withdrawals -19

8 staffs refused consent 4 staffs were pregnant 5 staffs age had recently crossed 50 years 1 staff resigned

1 staff was on long leave

Total

withdrawals -17

4 staffs refused consent

1 staff was pregnant 8 staffs stopped doing shift work

2 staff was on long leave

2 staffs resigned

Systematic random sampling done and study samples chosen

(47)

Table 1 represents the baseline characteristics of the participants of the study on shift work and metabolic syndrome among hospital employees.

Variable Total

(N =160) n(%)

Shift workers (N=80) n(%)

Daytime workers (N= 80) n(%)

P value

Age*(years) 38.24(7.02) 35.25(6.21) 41.23(6.52) <0.001

Sex Male Female

92(57.5) 68(42.5)

46(57.5) 34(42.5)

46(57.5)

34(42.5) 0.99

Designation

1. Attendants and house keeping

2. Technical staff 3. Office staff 4. Pharmacist

57(35.63)

42(26.25) 44(27.5) 17(10.63)

29(36.25)

27(33.75) 7(8.75) 17(21.25)

28(35)

15(18.75) 37(46.25) 0

<0.001

Smoker 13(8.1) 9(11.2) 4(5) 0.148

Alcohol Consumer 25(15.6) 15(18.7) 10(12.5) 0.28

Timing of meal Always regular Often regular Sometime regular Never regular

90(56.3) 29(18.1) 32(20) 9(5.6)

46(57.5) 13(16.3) 17(21.3) 4(5)

44(55) 16(20) 15(18.8) 5(6.3)

0.89

Junk food eating Never Seldom Sometime Often

41(25.8) 24(15.1) 83(52.2) 11(6.9)

16(20) 10(12.5) 47(58.7) 7(8.8)

25(31.7) 14(17.7) 36(45.6) 4(5)

0.18

Status

Vegetarian Non-Vegetarian

22(13.8) 138(86.2)

6(7.5) 74(92.5)

16(20) 64(64)

0.02

(48)

Diabetes mellitus┼ 11(6.9) 3(3.8) 8(10.1) 0.11

Systemic hypertensionᴬ 22(13.8) 9(11.3) 13(16.5) 0.34

Dyslipidemia 16(10.1) 4(5) 12(15.2) 0.03

Physical activity#

Inactive

Moderately inactive Moderately active Active

34(21.3) 64(40) 36(16.2) 36(22.5)

7(8.8) 39(48.7) 13(16.3) 21(26.2)

27(33.8) 25(31.3) 13(16.2) 15(18.7)

0.001

Stress ##

Much lower than average Slightly lower than average Average

Slightly higher than average Much higher than average

8(5) 18(11.3) 38(23.7) 58(36.3) 38(23.7)

5(6.3) 7(8.7) 15(18.7) 30(37.5) 23(28.8)

3(3.8) 11(13.7) 23(28.8) 28(35) 15(18.7)

0.31

Sleep ###

Poor sleep Good sleep

71(44.4) 89(55.6)

42(52.5) 38(47.5)

29(36.2)

51(63.8) 0.04

BMI* 26.3(4.6) 25.9(4.4) 26.8(4.8) 0.29

Note: * represents mean and standard deviation

┼ : Represents the participants who were known to have diabetes meelitusat the onset of the study

A : Represents the participants who were known to have systemic hypertension at the onset of the study

#Physical activity was graded based on GPPAQ(General practice physical activity questionnaire)

##Stress was scored based on PSS(Perceived stress scale)

###Sleep was classified as good or bad based on PSQI(Pittsburgh sleep quality index)

(49)

Baseline characteristics

AGE OF THE PARTICIPANTS

Mean age of the study population was 38.24(SD ±7.02). Mean age of the shift workers was 35.25(SD ±6.21). Mean age of the daytime workers is 41.23(SD ±6.52). Age was significantly different between the two groups with daytime workers being older than the shift workers by mean of 5.98 years. Increasing age had linear correlation with increasing metabolic syndrome as per univariate analysis with Odds ratio of 1.05 with 95% CI(1.002-1.10) with a p=0.04. However on multiple logistic regression there was no statistically significant relation noted between age and metabolic

syndrome OR-1.04 with 95% CI(0.98–1.10) and p=0.205. However as metabolic diseases increase with increasing age and at baseline there was found to be significant difference between the two groups in terms of age, the study population was again stratified into two age groups: Age≤40 years and Age>40 years. The baseline

characteristics and the outcomes were studied for the two age groups to understand the impact of age on the study if any. Kindly refer below to section on ‘Age and its effect on the baseline characteristics and outcomes of the study’.

(50)

GENDER OF THE PARTICIPANTS

There was a male predominance with 92/160 (57.5%) of the participants being males and 68/160 (42.5%) being females. The distribution of the genders was similar among the shift workers and the daytime workers.

Figure 6 is a graphical representation of the distribution of men and women employees in the study groups.

Figure 6 represents the distribution of the gender between the shift work and the

0 50 100

Male

N-92

Total Shift workers

Figure 6 represents the distribution of the gender between the shift work and the

Male Female

N-92

N-68

N-46 N-46 N-34

N-34

Gender distribution (N- 160) Male-57.5%

Female-42.5%

Shift workers Daytime Workers

Figure 6 represents the distribution of the gender between the shift work and the N-34

Gender distribution (N- 160) Male-57.5%

Female-42.5%

Daytime Workers

(51)

DESIGNATION OF THE PARTICIPANTS

Study population comprises of staff with different designations. Hospital attendants and hospital housekeeping staff together account 57/160(35.6%) of the population, almost equally distributed between the shift workers and daytime workers. Office workers were the second most common group comprising predominantly of office secretaries and medical records

professionals. The group also had small numbers of receptionists and cash counter staff. There were 44/160(27.5%) office workers. Office workers were predominantly daytime workers than shift workers. Lab technicians were the next group of employees. They contributed to 42/160(26.25%) of the study population. Lab technicians were predominantly shift workers. The other non classifiable group of employees were the pharmacists. They contributed to 17/160(10.63%) of the study population. All of them were shift workers.

Figure 7 is a graphical representation of the employees in both groups distributed as per their designation.

(52)

Figure 7 represents distribution of the study participants among various

SMOKING AMONG THE PARTICIPANTS

There were 13/160(8.13%) smokers in total. 9/13 were shift workers and only 4/13 were daytime workers. The difference between the rates of smoking between the shift workers and daytime workers was not statistically significant

29 27 28

15 7

37

17 0 0 5 10 15 20 25 30 35 40

Shift workers N-80 Daytime workers N- 80

N u mb er of staff

Type of work

Distribution of Designation N-160

Hospital attendants and house keeping Technicians

Office staff

Pharmacists

(53)

ALCOHOL CONSUMPTION HABIT AMONG THE PARTICIPANTS

There were 25/160(15.6%) people reportedly consume alcohol among the participants. Of the alcohol consumers 15 were shift worker and 10 were daytime workers, there was no statistical difference between the two groups p=

0.27. 24/25 of them were men and there was one woman alcohol consumer.

16/25(64%) of the alcohol consumers were hospital attendants or housekeeping staff. 11/25(44%) of the alcohol consumers were smokers as well. 13/25(52%) of the alcohol consumers consumed alcohol less than a week. 7/25(28%) of the alcohol consumers consumed alcohol once a week. 5/25(20%) of the alcohol consumers consumed alcohol more than once week. None reportedly consume alcohol on a daily basis. 17/25(68%) of the alcohol consumers had reported increased stress levels. 18/25(72%) of the alcohol consumers had reported poor sleep quality.

TIMING OF FOOD

More than half of the study population 90/160(56.25%), reportedly always eating food at regular timings. This was not different between the shift workers and the daytime workers. 29/160(18.1%) reported to often eat food at regular timings and this group did not differ between the shift workers and daytime workers. There were 32/160(20%) who reported that only sometimes they ate

(54)

9/160(5.6%) reported to never eat food at regular timing; again there was no difference between the two groups for this behaviour.

JUNK FOOD EATING HABIT

44/160(25.8%) of the study population reported that they never eat unhealthy junk food and there was no statistical difference between the shift workers and daytime workers for the behaviour. 24/160(15.1%) of the study population reported that they seldom eat unhealthy junk food and there was no statistical difference between the shift workers and daytime workers for the behaviour. 83/160(52.2%) majority of the study population reported that they eat unhealthy junk food sometimes. However there was no statistical difference between the shift workers and daytime workers for the behaviour. 11/160(6.92%) of the employees reported to eat junk and unhealthy food often. There was no difference between the two groups for the behaviour. The mere numbers were showing a trend of shift workers having a habit of eating junk food often than the daytime workers.

(55)

DIETARY STATUS OF THE PARTICIPANTS

There was 22/160(13.8%) vegetarians and the rest were non-

vegetarians138/160(86.25%). The vegetarians were higher in the daytime working group compared to the shift group, the difference being statistically significant, P value–0.022. Figure 8 is a graphical representation of the number of vegetarians and non vegetarians in the study population and the difference in their distribution among the shift workers and daytime workers.

Figure 8 shows the dietary status of the study population and their distribution between the shift working group and daytime group

6 16

74 64

0 20 40 60 80 100

Shift workers, N=80

Daytime workers,

N=80

N u mb er

Type of work

Status of food

Non-Vegetarians, N=138

Vegetarians,

N=22

(56)

PHYSICAL ACTIVITY OF THE PARTICIPANTS

The General Practice Physical Activity Questionnaire (GPPAQ) was used to assess the level of physical activity of the population studied.

GPPAQ divided level of physical activity into 4 categories– a. Inactive

b. Moderately Inactive c. Moderately active d. Active

Majority of the study population were in the ‘moderately inactive’ groupwhich accountedfor 64/160(40%) followed by ‘active’ 36/160(22.5%), ‘inactive’

34/160(21.25%) and lastly ‘moderately active’ group 26/160(16.25%). There was statistically significant difference between the shift workers and daytime workers with the shift work group being more physically active (p=0.001). Represented in Figure 9 is the graphical representation of the physical activity of the staff of both of the study groups.

(57)

Figure 9 shows the percentage of staff in each of the levels of physical activity and also the difference between the shift working and daytime working groups.

STRESS LEVELS OF THE PARTICIPANTS

Perceived stress scale (PSS) was used to assess the stress levels of the study participants

The following scores represent the following stress levels as per PSS a. Much lower than average stress–0 to 7

b. Lower than average stress- 8 to 11 c. Average stress–12- to 15

d. Slightly higher than average stress- 16 to 20 e. Much higher than average stress–more than 20

8.8

48.8

16.2

26.2

33.8 31.2

16.3 18.7

0 10 20 30 40 50 60

Inactive Moderately

Active Active

Pe rc en tag e

Physical activity

Level of Physical activity

Shift workers, N-80 Daytime workers, N-80

(58)

Majority of the study participants were in the ‘slightly higher than average’ stress group with 58/160(36.3%) followed by ‘much higher than average stress’

38/160(23.8%) ,‘average’ stress group 38/160(23.8%), ‘lower than average stress’

18/160(11.3%) and lastly ‘much lower than average stress’ consisting of 8/160(5%).

The shift workers and daytime workers had similar scores for stress and there was no statistical difference between the two groups. Figure 10 is a graphical representation of the different levels of stress and the percentage of shift worker and daytime workers in each category.

Figure 10 shows the percentage of staff at each level of stress and the difference between the shift workers and the daytime workers

6.3 8.7

18.8

37.5

28.7

3.7

13.8

28.7

35

18.8

0 10 20 30 40

P er ce n tage

Stress level

Level of Stress

Shift workers, N=80

Daytime workers,

N=80

(59)

SLEEP QUALITY OF THE PARTICIPANTS

Pittsburgh Sleep Quality Index (PSQI) was used to assess the quality of sleep of the study participants

PSQI score of 5 or less indicate good quality sleep quality and more than 5 indicate poor sleep quality. The maximum score was 20.

Majority of the study participants reported to have good quality sleep 89/160(55.6%).

The quality of sleep was found to be different between the shift workers and the daytime workers. 38/80(47.5%) of the shift workers had reported good sleep quality. 42/80(52.5%) of the shift workers had reported poor sleep quality. On the contrary 51/80(63.8%) of the daytime workers reported good quality sleep and 29/80(36.2%) of the daytime workers reported poor sleep quality. The difference was statistically significant with a P value of 0.04. Figure 11 represents in graphical form the number of shift workers and the number of daytime workers and the quality of their sleep.

(60)

Figure 11 shows the quality of sleep among the shift workers and daytime workers.

Outcomes

The prevalence of metabolic syndrome among the study participants was 33.1%, it was 25% among the shift workers and 41.3% among the daytime workers.

The table 2 below represents the primary and secondary outcomes of the study.

The outcomes are also represented in a graphical form in Figure 12 showing the odds

42 29

38 51

0 20 40 60 80 100

Shift wokers, N=80

Daytime workers, N=80

F re q u en cy i n N u mb er

Type of work

Good sleep

Poor sleep

(61)

Outcome Total N= 160 (%)

Shift workers N=80 (%)

Daytime workers N=80 (%)

Unadjusted Odd’s ratio (95% CI)

P value

Odds ratio after adjusting for covariates#

(95% CI)

P value

Primary Outcome Metabolic

syndrome

53(33.1) 20(25) 33(41.3) 0.47 (0.24-0.93)

0.03 0.55 (0.24-1.29)

0.17 Secondary

Outcomes 1. Systemic Hypertension

35(21.8) 16(20) 19(23.8) 0.8 (0.4-1.7)

0.56 1.08 (0.4-2.9)

0.88 2.Impaired Fasting

glucose 36(22.5) 11(13.8) 25(31.3) 0.35

(0.16-0.77)

0.01 0.58 (0.21-1.58)

0.29 3.Central Obesity 104(65) 49(61.3) 55(68.8) 0.72

(0.37-1.38)

0.32 0.68 (0.27-1.67)

0.39

4.Obesity 101(63.1) 49(61.3) 52(65) 0.93

(0.48-1.79)

0.8 0.98 (0.4-2.2)

0.96 5.Elevated

triglycerides

43(26.9) 23(28.8) 20(25) 1.21 (0.6-2.4)

0.59 0.95 (0.35-2.6)

0.93 6.Decreased HDL 95(61.7) 47(58.8) 48(60) 0.95

(0.51-1.78)

0.87 0.66 (0.27-1.58)

0.35 7.Diabetes mellitus 20(12.5) 8(10) 12(15) 0.63

(0.24-1.63)

0.34 1.12 (0.3-4.1)

0.87

# odds ratio adjusted for age, sex, diet, physical activity, stress, sleep and alcohol consumption.

(62)

Figure 12 is a box plot representation of the outcomes of the study on shift work and metabolic syndrome among hospital employees

METABOLIC SYNDROME

Metabolic syndrome prevalence in the study population was 53/160(33.1%). Majority of the participants with metabolic syndrome belonged to the daytime working group compared to the shift working group.

(63)

Multiple logistic regression was performed and odds ratio was calculated adjusting for age, sex, diet, physical activity, stress, sleep and alcohol consumption. The odds ratio thus calculated was 0.55 (95% CI 0.24–1.29) and a P value =0.17.

Number of night shifts and metabolic syndrome

The number of night shifts per month was assessed for its effect on metabolic syndrome among the shift workers. As presented in Table 3 the number of night shifts per month was divided into up to 5 night shifts per month, 6-10 night shifts per month and more than 10 night shifts per month. The prevalence of metabolic

syndrome did not differ statistically depending on the number of night shifts.

Table 3 represents the distribution of staff in the shift working group and the prevalence of metabolic syndrome among them.

Variables

Total n=80(%)

Metabolic syndrome

p-value Yes

N=20(%) No n=60(%) No of night

shifts

0.821

<=5 27(33.3) 8(38.1) 19(31.7) 6-10 43(54.3) 10(52.4) 33(55.0)

>10 10(12.4) 2(9.5) 8(13.3)

Secondary analysis was done to test the association with metabolic syndrome for the following parameters:

1. Smoking had no significant relation with metabolic syndrome on univariate analysis with odds ratio of 0.58 and 95% CI(0.15-2.21) and p value of 0.427

2. Alcohol: On performing the univariate analysis alcohol consumption had inverse relationship with metabolic syndrome. The odds ratio of 0.33 with

THE ASSOCIATION BETWEEN ROTATING SHIFT WORK AND METABOLIC SYNDROME AMONG HOSPITAL