Impact of Internet on delivery of critical cardiac health care:
A Case Study from India

Corresponding author:
Sudeepa Banerjee
IISWBM, Management House, College Square West
Kolkata -700073, India
Email : sudeepab@iiswbm.edu
Telephone : 09830088873

Introduction

Ischmeic stroke and its relevance in India

Ischemic or heart strokes have grown to epidemic proportions in developing countries and two-thirds of global strokes occur in low and middle-income countries (Feigin, 2005). Cardio vascular diseases that are life threatening in most cases are spreading very fast too, both for genetic as well as lifestyle reasons in India. Unfortunately public healthcare facilities are woefully inadequate both in numbers and quality and there is a huge gap amongst the people who have access to healthcare and those who do not. India can be divided into two on this. There is one India that provides high-quality medical care to middle-class Indians and medical tourists, and the other India in which the majority of the population lives-a country whose residents have limited or no access to quality health care. The sophisticated and good facilities - both in terms of infrastructure and specialists - are all concentrated in the urban areas .For the majority of Indians, who live in rural areas, there is hardly even the basic healthcare centre and the ones that do exist perennially lack qualified doctors and nurses; therefore most people are treated by quacks and under qualified people.

According to the Medical Council of India, the doctor-patient ratio is 1 per 1634 (Press Information Bureau, Government of India, Press Release, August 20,2007). The nurse-patient ratio in India is 1 per 2250 (Press Information Bureau, Government of India, Press Release, May 12,2007). Moreover, according to the 1981 census of India, 73% of the doctors are located in cities, and that mainly in metropolitan cities.

Strokes and Stroke treatment

When the flow of blood is severely reduced or stopped due to a blockage in an artery a coronary attack or heart attack is likely to occur. Once a heart attack occurs there is damage or death of part of the heart muscle as a result of the ischemia. This is also known as myocardial infarction. Emergency treatment requires dissolving this clot to restore blood supply to the heart and can be treated by three methods according to the open-artery model: these include using Clot-busting drugs (thrombolytics) that can dissolve clots immediately; Angioplasty, a surgical procedure that can open the artery with a balloon and then, if indicated, install a stent that helps keep the vessel open; or Bypass surgery that provides an alternate route around the blockage using a blood vessel from the calf or elsewhere. The American College of Cardiology's guidelines now recommend that if transferring patients to an angioplasty facility delays treatment by one hour, doctors should consider first offering thrombolytics to open the arteries immediately, at least partially. Then, those patients who need further intervention can receive angioplasty or other procedures. That way, patients will benefit from both t-PA's (Tissue plasminogen activator) rapid rescue of blood flow and from PCI's complete reopening of the artery. Further Thrombolytic therapy is the only therapy approved by the US Food and Drug Administration for acute ischemic stroke. The number of stroke patients receiving r-tPA (recombinant tissue plasminogen activator) in third world countries is very low. However thrombolytic therapy is used in India (Pandian et. al., 2007) but only 14.7% of stroke patients in India reach hospital within the three-hour window period (Nandigam, 2003).

Research based on the concept of 'door-to-needle time' that indicates the time between the onset of symptoms in a patient until he or she receives the first dose of the thrombolytic, has shown that patients who receive thrombolytics that successfully opened their arteries within one hour have only a 1% mortality rate during the next 30 days. Waiting another half hour increases the mortality rate to 6.4%. When used after 3 hours, 80% of the benefits were found to have been lost and after 4 hours thrombolytics have virtually no effect in reducing mortality (Bergman et. al., 1982). However, the guidelines developed and updated in developed countries for treating these symptoms are unrealistic and impractical to adopt in developing countries (Hill and Lalor , 2008). When the treatment demands such prompt response, it means the community must have an emergency network that can transport patients to a stroke center quickly. In addition, a hospital must have 24-hour physicians on call who are trained and licensed to provide thrombolytics, as well as a 24-hour facility to detect whether a patient is having a hemorrhagic or an ischemic stroke.

Telemedicine

Telemedicine is the use of modern telecommunication and electronic information technologies (Perednia and Allen, 1995) to provide clinical healthcare to individuals at a distance from the healthcare provider or specialists (Bashshur et. al., 1997; Brown, 1995; Grigsby et. al. ,1993; Pushkin et. al, 1995). The feasibility of using telemedicine has been demonstrated by several studies. These include various specialties and subspecialties, including cardiology (Bird 1972 ;Finley et.al, 1989), critical care medicine (Grundy, Jones and Lovitt, 1982), neurology (Chaves-Carballo, 1992; Hubble 1992), pediatrics (Cunningham, Marshall & Glazer, 1978).
Telemedicine projects have also tested the use of various technologies from fax, telephone, radio, real-time interactive TV to Internet based exchanges (Bertera & Bertera 1981; Bertrand et.al ,1994; Padeken et al ,1995; Rinde , Nordrum & Nymo, 1993; Turner, Brick & Brick 1995). Telemedicine has been trialed in a wide range of situations and locations (Lovett & Bashshur, 1979) that include the army (Cawthon et. al, 1991; Spiller, Heelstein & Basquill 1990) as well as civil settings like hospitals, healthcare clinics and long term care facilities (Finkelstein et. al, 1993; Sparks et. al, 1993 ).

Telemedicine and strokes

Telemedicine has been found to be successful in congestive heart failure patients from rural regions in terms of three dimensions: clinical, financial and humanistic (Koch, Smith and Stanford, 1999). Telemedicine has also been found to offer long-term benefits for acute stroke patients in community hospitals (Audebert et. al, 2009) and has also been found to be beneficial to head and neck cancer patients (van den Brink, 2007). However, there is an urgent need for more evidence about the clinical impact of telemedicine applications (Brear, 2006). In most cases it has been reported that telemedicine-facilitated thrombolytic therapy for acute stroke patients was intended not to replace care provided by local providers but rather to address a time- and spatially related emergency need (Choi et.al, 2006). An Internet based system like the above, thus facilitates the safe administration of thrombolytic therapy to patients within rural communities suffering an acute ischemic stroke. Telemedicine can be used as a rapid consultation tool too that results in more accurate decision-making and may be used as a tool to provide the same quality of stroke care to patients in rural hospitals as is given in tertiary stroke centers (Switzer et. al, 2009). The success of using it as a rapid consultation tool has also been reported (Wang et. al, 2004). It was also reported that stroke telemedicine consultations result in more accurate decision-making compared with telephone consultations and can serve as a model for the effectiveness of telemedicine in other medical specialties (Meyer et. al, 2008). Since Thrombolytic therapy is the only proven therapy for acute ischemic stroke (Henninger et. al, 2009), use of Internet technology will thus enable greater use of thrombolytic therapy in strokes in a safe manner.

Relevance of Telemedicine in India

In this situation some of the obvious advantages of telemedicine from the patients' perspective could include improved access to quality and specialized health care; better access from underserved rural, semi-urban and remote regions; access to medical experts' services; reduced clinical travel and unnecessary patient transfers; cost effective care (Ohinmaa, Hailey & Roine, 2001); prevention of costly health complications resulting from delayed treatment(Schwamm et.al, 2004); and timely intervention.

Potential benefits for practitioners could include linkages with doctors and centers of excellence-education benefits; appropriate referrals and reduced professional isolation; ability to retain their local patients; access to comprehensive patient data - offline and online; quick and timely follow up of patients after treatment; training; retraining; educational facilities (Olver & Selva-Nayagam, 2000); and ability to cater to patients from a wide and varied geographical region or territory.

Other benefits to government and institutions from employment of telemedicine applications would be enhancement of timeliness, effectiveness and efficiency of health care delivery, which could produce cost savings to both the health system and individual patients. Telemedicine applications can bridge the gap in health service facilities between rural or remote and urban areas (Ellis, 2004) and commercially create a wider care base -therefore increasing viability of investments. Other benefits include low capital investment to create a facility, increased utilization of hospital resources, maintaining continuous relationships with old patients, and introducing a new model in health care -nurse- and technician-led clinics.

Objective of the study

There are several pressing health services issues in telemedicine that require specific studies (Bashshur 1980, 1995; Grigsby et. al, 1994c) and it is necessary to evaluate whether telemedicine applications are medically effective means of delivering healthcare and to measure effectiveness of telemedicine vis a vis conventional care. The studies mentioned have not tried to assess either the effect of telemedicine- based thrombolytic therapy or the effect of location on reducing the rate of expiry amongst cardiac patients. This study makes a modest attempt through outcome-based research to measure the impact of telemedicine in the delivery of critical cardiac healthcare through the following research questions:

R1: What is the impact of Internet based thrombolytic therapy on the probability of reducing the expiry rate of patients?
R2: Are there any location specific factors that have an impact on the expiry rate ?

Study design

The organization of this paper is as follows: the context of the study has been discussed first and then the model and hypotheses have been presented. This is followed by the methods of analysis, the results and the discussion. Finally, the conclusion and future directions are presented.

A case study method has been applied to carry out a qualitative and quantitative analysis by observation and data collection with an intention to study the impact of the Internet on critical cardiac health problems amongst the population located in areas without proper medical facilities in India.

The project evaluated

The Integrated Telemedicine Tele-health Project is a first-of-its-kind initiative jointly sponsored by the Indian Space Research Organization (ISRO), Hewlett Packard (HP), the Asia Heart Foundation at the Rabindranath Tagore International Institute of Cardiac Sciences (RTIICS) Kolkata (a metropolitan city and the capital of West Bengal, a state in the eastern part of India). It was initiated in 2001 along with the state governments of seven north-eastern states of India (Assam, Meghalaya, Tripura, Arunachal Pradesh, Manipur, Mizoram and Nagaland). The Rabindranath Institute at Kolkata is the main Telemedicine linking hub for the seven states.

The project tried to identify a 100-bedded hospital in each of these participating states and the hospitals were to be selected based on distance from the state capital and the lack of a coronary care unit. The first three centers along with the adjoining Coronary Care units (CCU's) were set up in Siliguri and Bankura in West Bengal and Udaipur in Tripura, with the first Coronary Care Unit inaugurated in Siliguri District Hospital, Siliguri, West Bengal on 24th June, 2001 followed by Bankura Sammilani Medical College and Hospital, Bankura, West Bengal on 21st July, 2001 and at Tripura Sundari District Hospital, Udaipur in Tripura on 22nd August, 2001. RTIICS was the referral hospital for all these centers.

Geography of West Bengal

West Bengal is the gateway to the east in India, stretching from the Himalayas in the north to the Bay of Bengal in the south. It is bounded on the north by Sikkim and Bhutan, on the east by Assam and Bangladesh, on the south by the Bay of Bengal and on the west by Orissa, Bihar and Nepal. It has therefore three international frontiers - one each to the north, east and west. The state lies between 27°13'15" and 21°25'24" north latitudes and 85°48'20" and 89°53'04" east longitudes.

Demography

The state of West Bengal has an area of 88,752 sq. km. and a population of 80.18 million. There are 19 districts, 341 blocks and 40,782 villages. The State has a population density of 903 per sq. km. (as against the national average of 312). The decadal growth rate of the state is 17.77% (against 21.54% for the country) and the population of the state is growing at a slower rate than the national rate.

Particulars	Required	In position	Shortfall
Sub-centre	12101	10356	1745
Primary Health Centre	1993	924	1069
Community Health Centre(CHC)	498	349	149
Multipurpose worker(MPW) (Female)/ANM at Sub Centres & PHCs	11280	6051	5229
Health Worker (Male) MPW(M) at Sub Centres	10356	4215	6141
Health Assistant (Female)/LHV at PHCs	924	300	624
Health Assistant (Male) at PHCs	924	225	699
Doctor at PHCs	924	810	114
Obstetricians & Gynaecologists at CHCs	349	38	311
Physicians at CHCs	349	107	242
Paediatricians at CHCs	349	25	324
Total specialists at CHCs	1396	186	1210
Radiographers	349	127	222
Pharmacist	1273	830	443
Laboratory Technicians	1273	441	832
Nurse/Midwife	3367	5215	-

Health Institution	Number
Medical College	9
District Hospitals	16
Referral Hospitals
City Family Welfare Centre
Rural Dispensaries
Ayurvedic Hospitals*	4
Ayurvedic Dispensaries*	295
Unani Hospitals*	1
Unani Dispensaries*	3
Homeopathic Hospitals*	12
Homeopathic Dispensary*	1220

Geography of Tripura

Tripura is located in the north eastern part of India, between 22 ° and 56 minutes and 24 ° and 32 minutes north latitude and between 90 ° and 09 minutes and 92 ° and 20 minutes east latitude. It is bounded on the north, west, south and south-east by Bangladesh whereas in the east it has a common boundary with the Indian states of Assam and Mizoram.

Demography

The population of Tripura is 3.20 million according to 2001 census, scattered over 4 districts, 40 blocks and 870 villages. The State has a density of 305 persons per sq. km. As against decadal growth rate of 21.54% at the national level, the population of the State has grown by 16.03% over the period 1991-2001. The sex ratio of Tripura at 948 females to 1000 males is higher than the national average of 933. Female literacy of the State rose to 65.41% from 49.65% in 1991.

Item	Required	In Position	Shortfall
Sub-centre	659	579	80
Primary Health Centre	104	76	28
Community Health Centre	26	11	15
Multipurpose Worker (Female)/ANM	655	638	17
Health Worker (Male)/ MPW(M)	579	436	143
Health Assistants (Female)/ LHV	76	62	14
Health Assistants(Male)	76	93	-
Doctor at PHCs	76	255	-
Surgeons	11	-	-
Obstetricians & Gynaecologists	11	-	-
Physicians	11	-	-
Paediatricians	11	-	-
Total specialists at Community Health Centres	44	-	-
Radiographers	11	1	10
Pharmacist	87	76	11
Laboratory Technicians	87	67	20
Nurse Midwife	153	235	-

Medical College	2
District Hospitals	2
Referral Hospitals
City Family Welfare Centre
Rural Dispensaries
Ayurvedic Hospitals*	1
Ayurvedic Dispensaries*	55
Unani Hospitals*	-
Unani Dispensaries*	-
Homeopathic Hospitals*	1
Homeopathic Dispensary*	93

From the data presented above, the situation is very clear about the shortfall of medical infrastructure in the two states under study. Further, in both states only the medical colleges have departments of cardiology and treat cardiac patients. Therefore the pressure on these colleges is considerable and in most cases patients have to travel a long distance and then wait to receive medical attention. It should also be noted that most of these patients are very poor and are not covered by any medical insurance.

The project considered is modeled on Fisher's (2005) hub-and-spoke model of telemedicine-delivered cardiac care designed to enhance the administration of acute stroke therapies. This project was unique since all previous telemedicine projects in India were based on an outpatient basis, where patients came for consultation through Internet technology and then left, as in other outpatient departments. Here patients were admitted to the cardiac care units on the basis of the Internet-based ECG report and condition of patient for further treatment. It was intended to provide patients with acute stroke a timely, adequate assessment and emergency stroke treatments.

The respective state governments in their local hospitals provided these spaces for the Critical Coronary Care units and the manpower was initially provided by the RTIICS. This manpower comprised trained staff - doctors, nurses and CCU technicians - who could take the special Electro Cardiogram (ECG), perform catheter-lab procedures, use defibrillators and administer life saving drugs in the critical hour of a heart attack. The doctors trained had the basic 5 years MBBS degree (bachelor of medicine and surgery) in medicine and the nurses also had the basic qualification (3 years-bachelors degree in nursing). They were further trained in IT technology to manage and run the services independently. After the initial handholding, in the Siliguri center, the nurses were provided by the government hospitals. In Udaipur, the doctors and the nurses were provided by the hospital and the technicians were provided by RTIICS. In Bankura all three types of staff were provided by RTIICS. The life -saving drug Streptokinase was also provided free of cost to the patients from that center by RTIICS. In the other two centers the respective state governments provided the same drug free of cost to its patients.

The project implementers have devised new ways and means to reduce the cost of installation of telemedicine equipment without compromising on the quality of care. They have used ISDN and telephone line connectivity that are easily available from BSNL (state run low cost ISP) even in very remote areas today. This technology supports easy data transfers and video calls over very long distances. Further, video-conferencing cameras available in the market at affordable prices, a computer (PIV or higher), an EECG device unit, which costs around 265 USD and a TV screen for video-conferencing are the basic equipments being used

Expensive telemedicine software or high-end gadgets like digital cameras, online microscopes, medical scanners that transfer and store medical data like X-Rays, CT Scans & MRIs have not been used in the local centers so far, as there is no trained manpower available to use these gadgets and provide high-tech examinations to patients at those remote locations. Each of these centers provide round-the-clock services every day of the year. The services provided are as follows:

• Tele-consultation in outpatient department
• Transtelephonic ECG (Electrocardiograph)
• Consultation on medical reports
• Emergency admissions and In-patients department
• Virtual rounds

Tele-consultation in Outpatient department by using video-conferencing facilities for emergency, non-emergency and follow-up cases. Patients who do not require immediate hospitalization can avail themselves of tele-consultation. In these cases all vital reports related to the patient are sent in advance (store and forward) to the referral hospital in a digital format (X-ray reports, Scan report, Tele-ECG and others) either as an email or through net-meeting. At the appointed time the specialist sits in the studio of the referral hospital while the patient along with a local doctor sits in the studio of the local hospital and the consultation takes place. There are certain doctors interested in tele-medicine who provide maximum consultations. However, the patient may also choose a doctor, who then gives an appointment. In case of follow up of patients, the same process as the OPD consultation is repeated. The other advantage of such a system is the reduction in waiting time of a patient. In most cases there is no waiting time unless there is another videoconference going on, because each center has only one videoconference unit at the moment. The local doctor provides the necessary clarification to both the specialist and the patient at the time of video-consultation. RTIICS provides both the tele-consultation and ECG services free of cost.

Transtelephonic ECG (Electrocardiograph) - A general physician cannot confirm a heart attack without performing an ECG and in most remote locations ECG clinics are not available. To counter this problem, each of these telemedicine units has a special six-lead ECG machine to conduct a transtelephonic electrocardiograph using special software. The ECG can be taken for 10 -20 seconds or as desired. Once an ECG is taken, the local center connects to the referral hospital through special software and a telephone line using Internet technology to transfer the data. At the referral hospital's monitor there is an alert every time a fresh ECG arrives, almost like an email alert. A specialist looks at the ECG and fills in an online form with his opinion and observation. Once completed, it is sent back to the local hospital through the Internet connection in the same way. This report is printed locally on arrival and the residing local doctor gets the opinion, observation and advice of his senior colleague from the referral centre along with the report and is able to treat the patient accordingly. This entire process takes a maximum of 15-20 minutes - whereas otherwise the complete activity- requires at least 6 hours. Further, no clinic works round the clock for these kinds of tests and consultations in the regions where these facilities have been made available.

Consultation on medical reports - Other medical reports like radiology reports, CT scans and ultrasound reports are scanned locally and sent to the referral hospital. However, new machines will soon be installed where the reports will be created digitally and sent automatically to the referral hospital.

Emergency admissions and In-patients department - When a patient arrives with a serious cardiac condition, the doctor on call can admit the patient and consult the specialist at the referral centre using the videoconference facility with the referral hospital in addition to performing the transtelephonic ECG and sending other basic test reports. The specialist then assesses the condition of the patient and decides on the line of treatment which is then administered locally, such as the use of defibrillators or performing catheter-lab procedures like thrombolysis. This facility is of tremendous help to patients and increases their chances of survival if they receive this medical attention within an hour of the heart attack. It also provides the patient with a specialist who is far more experienced in handling such cases than the local doctor, without having to go through either the trauma or the delay in travel to a specialty hospital.

Virtual rounds - A specialist from the referral center goes on virtual rounds through video-conferencing to check the condition and progress of admitted patients on a regular basis. He/she can in one session monitor the condition of patients in all the centers, while sitting at his own desk, since every bed has its own monitor and camera unit. This specialist consultant from the referral center takes virtual charge of the disease management of these patients and instructs doctors, technicians and staff to administer the treatment locally through a two-way Internet based videoconference. This continues until the patient is discharged from the CCU.

Methodology

To understand how the growing use of Internet technology is impacting the cardiac health treatment of critical patients, the performance of the ICUs at these local telemedicine centres will be measured in terms of the percentage of expiry of patients, which is a categorical variable. Every patient admitted has unique set of medical problems (such as high blood pressure or diabetes with varying degrees of duration and levels, progression of disease etc). However, all patients that have been admitted are at a critical state caused by a stroke. Under these circumstances, the study attempts to analyse the effectiveness of treatment and disease management from a distance using Internet technology.

Logistic regression analysis (LRA) extends the technique of multiple regression analysis to research situations in which the outcome variable is categorical. This formula has been taken into consideration since our dependent variable 'patient expiry' is categorical as the response can be 'yes' or 'no'. However, in the case of categorical dependent variable, the discriminant analysis can also be applied instead of logistic regression model (Hair et. al 2005). Perhaps, the former claims stricter assumptions than demanded by the latter with respect to predictor variables (Hosmer & Lemeshow,2000). Moreover, the interpretations of the results with respect to logistic regression analysis are much easier than discriminant one on the basis of these reasons, the logistic regression with maximum likelihood estimators (MLE) algorithm has been preferred to discriminant analysis in this study.

Further, the ordinary least square method cannot be used since probability can be more than than 1, whereas probability should lie between 0 and 1. To represent this probability in terms of thrombolysed patients and the location of these local telemedicine centers, logistic regression seems adequate. The mathematical representation of the model is Equation 1.1 as shown below:

Model for research question:

The simplified form of the logistic regression model (equation: 1.1) has been derived and presented as:

Equation: (1.1): In p ∕ (1-p)= α + βX₁+γX₂+δX₃+ε

Where,
p = 1 , when percentage of expiry is less than mean expiry in period t
p = 0 , when percentage of expiry is greater than or equal to mean expiry in period t
X1 = Number of patients treated by this method in period t
X2= ICU at the Bankura telemedicine centre
X3= ICU at the Udaipur telemedicine centre
ε = Standard error
Mean expiry = total number of expiry /total number of admissions across the data

Research Hypothesis

The admitted patients are normally in a very critical and damaged state and the disease may be complex. Hence it may not always be possible to save their life despite intervention by local doctors and specialists from the referral centre using Internet technologies. As expiry in the cardiac care unit is a negative outcome of treatment it should therefore affect the discharge rate negatively. Improvement in management of patients is associated with reduction in death rates (Fox et.al 2007).

The thrombolytic procedure that is used to dissolve the clots that cause strokes is often a life saving measure. The process involves insertion of a thrombolytic agent into the blood stream through a catheter-based procedure to dissolve the clot causing the stroke. At the local cardiac care units, this critical procedure is conducted on patients based on the assessment of the doctor from the referral hospital. The process is relayed through videoconference and monitored by the referral doctor while it is performed locally. The present researcher assessed whether telemedicine (real-time, two-way audio and video, and digital imaging and communications in medicine) was effective for decision making in acute telemedicine consultations. To test whether Thrombolytic therapy using Internet is effective in reducing the expiry of patients in cardiac care units of telemedicine centers the first hypothesis is: H1: Internet based Thrombolytic therapy will have a negative effect on expiry rate

The three centers studied varied in the type of nurses staffing these telemedicine ICUs. The Government hospital in Siliguri provided these nurses, who have been working a long time in the government hospital, whereas the nurses in Udaipur were also provided by the hospital but were younger and more rigorously trained by RTIICS, and finally the nurses in the Bankura centre were provided by RTIICS themselves. Further, the other difference between the three centers is that the life-saving drug streptokinase is provided by RTIICS in Bankura but is provided by the state governments in the other two centers.

The role of nursing in providing the necessary health care in ICUs is well known. The project led by the Johns Hopkins Safety and Quality Research in reducing catheter-related bloodstream infections in ICUs across the entire state of Michigan and many other parts of USA achieved extraordinary and sustained successes (Pronovost et. al 2006;Sawyer et.al. 2010). Itclearly established that these successes were the product of more consistent performance of evidence-based care processes, that was supported by a well-designed system for care and by a shared commitment to safety and quality improvement by all members of the ICU team (Pronovost and Vohr ,2010; Pronovost , Berenholtz & Needham,2008) and not the result of new research or more effective antiseptic strategies or the use of sophisticated technology or other expensive resources. Simple measures that can identify barriers to performance, such as lack of proximity of essential antiseptic supplies, or correct stock maintenance of all necessary equipment for sterile catheter insertion and authority to suspend a procedure until the physicians performed all necessary steps, such as strict hand washing and full barrier precautions, can produce extraordinary results.

In view of the above it is expected that these parameters will influence the performance of a telemedicine centre. To test the same the following hypothesis has been posited:
H2: Performance of a telemedicine centre may be location specific.

Study design

The data were collected by the present author from Rabindranath Tagore International Institute of Cardiac Sciences (RTIICS), Kolkata, India and the three active telemedicine units connected to it namely, Siliguri and Bankura in West Bengal and Udaipur in Tripura between 30/11/2008 and 10/02/2009. The data were collected month-wise (period t). The total number of patients admitted to the Siliguri centre was 3731, Bankura centre was 3402, and 1321 patients were admitted to the Udaipur centre in this period. Detailed interviews were conducted with the heads of these units and the head of the unit at RTIICS. An interview was also conducted with a telemedicine consultant specialized in cardiology at RTIICS and deeply involved in this project. Live sessions of teleconsultation through videoconferencing were observed and four trans-telephonic ECG reports were studied and the process observed.

The Grand Mean Expiry indicates the total percentage of patients who expired in spite of intervention by local doctors and specialists from the referral centre through telemedicine while being lodged in the cardiac care unit. Thrombolysed (X1) indicates the number of critical patients who have undergone thrombolytic therapy at the telemedicine centre. X2 indicates the Bankura centre and X3 indicates the Udaipur centre. The Siliguri center has been chosen as the base centre. The dependent variable Proportion of expiry has been divided into two groups. If the mean expiry in period t is less than the grand mean expiry, then the group is coded by 1, otherwise 0 (if mean ≥ grand mean).

Methods of analysis

The logistic regression method has been used to test the three hypotheses. The basic tenet of choosing the logistic regression equation is that the researcher wants to classify the patients discharged into two groups (yes or no) on the basis of expiry. Logistic regression models the probability of an outcome bounded by 0 and 1. The assumption is that each one unit increase in patient's thrombolysed will multiply the odd of expiry rate by a certain factor. Similar assumption is made in the case of location.

Results

The results of the analysis are presented in the Tables 1 and 2.

-2 LL	HL	df	P	Hit ratio
346.814	9.651	8	0.290	61.7

Note: LL: Log Likelihood , HL: Hosmer-Lemeshow goodness of fit

	B	S.E	Wald	df	Sig	Exp (B)
Thromb(X1)	-.100	0.62	2.586	1	0.108	0.905
Bankura(X2)	1.853	0.461	16.164	1	0.000	6.376
Udaipur(X3)	2.211	0.542	16.619	1	0.000	9.121
Constant	-2.183	0.699	9.741	1	0.002	0.113

Note: *p<0.05

It is observed from the results presented in Table 1, that the Log likelihood value and the Hosmer-Lemeshow Goodness-of-fit test results are insignificant (γ2=9.651, df =8, p=0.290). Hence the null hypothesis of the logistic regression, that, there is no difference between the observed and predicted values of the dependent; is not rejected in this case.

It is observed from Table 2 that the Wald statistic(1 tailed ) directional is significant and hence the parameter - thrombolysed -is significant in the model. The Wald statistic for the other two parameters X2 and X3 have also found to be significant and hence they are significant in the model. Here the Exp(B) or the odds ratio for the parameter X1 is less than 1 (0.905) indicating that there is a decreasing or negative impact on the dependent variable. Based on the above, the hypothesis H1 is confirmed in this study. It is also concluded from the study that if delivery of thrombolysis treatment can be increased, then the expiry in the ICU will decrease.

Our control variables capture effects due to location. Both Bankura (X2,0.00) and Udaipur (X3,0.00) have a high performance vis-a-vis the base location (Siliguri) in decreasing the probability of expiry. Thus the findings confirm that location specific factors should be taken into consideration and as such hypothesis H2 is also confirmed in this study. However these Beta coefficients do not indicate the relative importance of locations in terms of performance. To calculate the relative importance (rank) the estimated coefficients have been multiplied by the standard deviations. Accordingly it appears that Bankura (0.153) has performed better than Udaipur (0.188).

Discussion

The above study shows that the disease management by specialists from the referral center in RTIICS, during the period of hospitalization in the local centers, using Internet as the medium, at all the three centers, are yielding satisfactory results by increasing the chances of survival amongst critically ill cardiac patients from remote locations and providing sophisticated treatment at low cost and with less specialized manpower.

The findings confirm the benefit of using Internet technology-based interventions on critical patients admitted in the local cardiac care units. The decrease in their rate of expiry is positively related to the use of video-conference based thrombolytic therapy. It is confirmed that its use has been effective in decreasing the number of patients expiring.

This technology enables remote stroke physicians to direct the local staff by telementoring to administer tPA in rural settings where thrombolytics were not previously used. This project is also providing telementoring, which is an educational technique that involves real-time guidance of a less experienced physician through a procedure in which he or she has limited experience. This supports an endeavor to bring stroke expertise to rural community hospitals. The results also indicate that success of this model is location specific. In this study the similarities and differences found between these centers indicate that human factors do play an important role in reducing ICU deaths. In Bankura where the staff were fully provided by RTIICS the centre has produced the best results. The quality of training and the accountability associated with a private organization may have contributed to the quality of service. Therefore, telemedicine needs to be implemented with care and supervision and has to be administered well (Weaver and Spence, 2000). These three local centers, along with the referral centre in Kolkata, have handled 7405 admissions, 822 thrombolytic therapies and 12056 teleconsultations in this period. Thus it can be said that telemedicine is an effective method of providing quality health care to remotely located patients who have had no previous access to such quality treatment.

Conclusion

These centers are able to provide best quality treatment to their patients using lower level manpower in backward, underdeveloped and far-flung areas because of Internet technology through a virtual experience very close to the real practice. In the process, they are also being able to bridge the distance gap and reduce the traveling agony of their critical patients by bringing the service closer to home. It may also be noted that the cost of bridging the distance through this technology is quite minimal - at times only the cost of a local phone call.. Thus this model is effective in India and its impact on health care structures can be significant. Due to the existing poor doctor: patient ratio and an urban bias towards medical facilities, this model can be used to extend medical services to the rural and semi-urban parts of India by introducing telemedicine centers in other district level hospitals. The low cost of the infrastructure required for this model should also be helpful. In this respect, telemedicine can be seen as a tool that can be used to build up new health care models.

Future directions

The study could be replicated in the southern states of India and in other medical disciplines like ophthalmology, dermatology, cancer and tropical diseases to measure the effectiveness of telemedicine in these areas.

References