Teleconsultations by Primary Care Physicians of Belo Horizonte: Challenges in the Diffusion of Innovation

Abstract

Introduction: Telemedicine is the provision of health services by professionals who use information and communications technology in the exchange of information for the diagnosis, treatment, and prevention of diseases. Since 2004, the Municipal Department of Health of Belo Horizonte, MG, Brazil, has been implementing telehealth initiatives with the goal of increasing the ability of primary care providers to resolve problems by providing them with clinical support and with continuing education using information and communications technologies. This study characterized the primary care physicians who used the Belo Horizonte telehealth store-and-forward teleconsultation. Materials and Methods: A descriptive study was conducted of the frequency of use of store-and-forward teleconsultation and characteristics of physician users. Results: We analyzed 737 store-and-forward teleconsultations in the period 2006–2010. The number of store-and-forward teleconsultations grew from 2006 to 2009, and 85 physicians used the tool. The physician users were predominantly women, specialists in family and community medicine, those employed under a stable contract, and older and with more time in the profession and in public service. The adoption rate was 6.0%, 4.8%, and 1.2% in 2008, 2009, and 2010, respectively. Conclusions: Context variables and individual variables may be influencing the adoption of store-and-forward teleconsultation in Belo Horizonte. It would be useful to further study the factors that constitute barriers to using teleconsultations and to improve the training in use of the tool.

Introduction

Telemedicine emerged in the military and aerospace technology sectors in the late 1950s in order to provide remote medical support. Recent advances in telecommunications combined with the need to expand access, improve quality, and reduce the costs of health services have stimulated the diffusion of telehealth in all regions of the world as an integral part of traditional health services.¹ According to current trends, the telehealth arsenal has moved beyond its initial conception, which was primarily to support the physicians in rural or remote areas; today it is the basis for a new paradigm.^1,2

The Municipal Department of Health, Belo Horizonte (BH), MG, Brazil, structured the public healthcare system in accordance with the principles of primary care: first contact primary care providers assume longitudinal responsibility for the patient; consider organic, mental, and social aspects of health; and coordinate with other services of the municipal health system in order to deliver care that is comprehensive.³

In order to increase the ability of primary care providers to resolve patients' problems (without having to refer patients), the city health department established a telehealth program by offering specialist support and underwriting continuing medical education activities.^4,5 BH Telessaúde (BH Telehealth) is a program developed in partnership with the Federal University and is financed by the Ministry of Health and the European Union. In 2004, videoconferences were implemented, followed by teleconsultations in 2006. Next, digital electrocardiograms, tele-urgencies, and distance learning courses were incorporated.

Teleconsultation consists of an interaction between two professionals: a primary care physician and a specialist when the former seeks help to care for a patient. Teleconsultations serve two purposes—care and education—because in the course of supporting the care of the patient, the expertise that the interface provided is accumulated and consolidated. The interaction can be simultaneous (online teleconsultation) or asynchronous (store-and-forward teleconsultation).^4,5

The contribution of BH Telessaúde to the strengthening of primary care in BH has been recognized. For this reason, it constitutes a model for the National Telehealth Program,⁶ but still the teleconsultation approach has yet to win over most primary care physicians. From 2004 to 2010, 341 videoconferences were conducted, 13,167 interpretations of digital electrocardiograms were issued by the University's cardiologists, 350 physicians were trained in electrocardiogram interpretation, and 374 physicians and nurses received training in urgent and emergency care. There were 737 asynchronous (store-and-forward) and 90 synchronous (online) teleconsultations.

What was the rate of adoption of the store-and-forward teleconsultation tool in different areas of BH, and what are the characteristics of physicians who used it during the period studied? The aims of this study are to analyze the rate of adoption of teleconsultation tools in different areas in BH and the physicians users' profile.

Materials and Methods

This was a descriptive study of the frequency of utilization of the store-and-forward teleconsultations system and the characteristics of physician who used it from 2006 to 2010. Teleconsultations are provided by University faculty and specialists of a municipal specialty center responding to requests from physicians working in the primary care units.

With store-and-forward teleconsultations, the primary care physician sends the specialist a summary of the case and his or her questions using a system developed specifically for BH Telessaúde. The specialist receiving the consultation referral is expected to respond within 72 h. Suggestions and recommendations from the specialist can relate to the investigation, to the diagnosis, or to treatment options. Both the physician and the specialist are responsible for the patient, according to the rules of the Minas Gerais state physician licensing board (Conselho Regional de Medicina de Minas Gerais).

The primary care units in BH have personal computers in their consulting service with access to BH Telessaúde. After training, physicians are registered in the system and receive the password. Use is voluntary.

For this study the sources of the data included data captured and stored by the store-and-forward teleconsultation tool (BH Telessaúde), data from the personnel management system of the municipal Department of Health (Health System Management Network), and data from the Conselho Regional de Medicina de Minas Gerais.

In order to describe the profile of the physicians who used store-and-forward teleconsultation, we analyzed the records pertaining to a total of 737 store-and-forward teleconsultations submitted from 2006 to 2010. Of these, 30 (4.1%) were excluded: 11 teleconsultations were canceled by the requesting physician, 11 were teleconsultations requested by dental professionals and nurses, and 8 were teleconsultation tests generated as part of system maintenance.

Thus, 707 store-and-forward teleconsultations were analyzed using a specific protocol that included the following fields: number of store-and-forward teleconsultations performed per year (2006–2010), name of the primary care unit and the District of the requesting physician, and requesting physician's name, gender, date of birth, date when employment as a municipal physician began, date of termination (dismissal/resignation/contract expired), type of employment, specialty practiced at the primary care unit, year graduated from medical school, and medical license number. The data fields “requesting physician's name” and “number of store-and-forward teleconsultations performed” were obtained from BH Telessaúde.

Access to the personnel management system enabled us to extract data relating to gender, birth date, date of employment, date that public service ended, type of employment, primary care unit assigned to, medical specialty, and medical license number. The field “year of graduation” was obtained directly from the Conselho Regional de Medicina de Minas Gerais public database.

From the data obtained it was possible to calculate the number of years since graduating, length of service in the municipal health system, and the age of the requesting physician. Age and years since graduating were calculated as of December 31, 2010. The length of service was calculated by two different strategies: (a) for those physicians still in service, through December 31, 2010; and (b) for physicians who left the service during the study period, through the date their service concluded (because of termination, resignation, or dismissal).

For physicians who had more than one type of employment (e.g., as a civil servant at one facility and a contract employee at another) and therefore more than one hiring date during the period of the study, we considered only the civil service job and the earliest hiring date. For those who moved from one primary care unit to another, we considered their most recent assignment.

In order to calculate the rate of adoption in 2008, 2009, and 2010 the number of physicians working in the primary care unit was obtained from the municipal Department of Health's Office of Human Resources. Because of physician turnover it was not possible to obtain the exact number of physicians, so we elected to use as a proxy for the annual census of physicians the number of physicians assigned to each primary care unit during the month of June of each year. The period from 2006 to 2007 were not considered for adoption rate because it is the initial deployment and testing period.

In order to evaluate the frequency of store-and-forward teleconsultations by year and according to the characteristics of physicians who used store-and-forward teleconsultations, separate tables were prepared for the following categories: teleconsultations performed each year by sanitary district; primary care unit and physician users by year and by district; number of store-and-forward teleconsultations per user; annual adoption rate (number of physicians who actually used the tool in relation to potential number of physicians and the number who registered to use the system); and profile of the physician users (gender, age, years since graduation, time [in years] in public service, time [in years] in the specialty practiced in the municipal health system, and type of contract).

The category “potential physicians” refers to all active physicians in the month of June of each year assigned to primary care units in which the teleconsultations tool was functioning in 2008, 2009, and 2010. The category “physician users” refers to physicians who actually generated store-and-forward teleconsultations in the same period. The category “registered physicians” identifies the primary care unit physicians enrolled in BH Telessaúde in 2010. Registered physicians who left public service before 2010 were excluded.

The rate of adoption in relation to “registered physicians” was calculated only for 2010 because the objective was just to check that the training and the issuance of passwords reached 100% of “potential physicians.”

For the data analysis, concepts and methods of the Diffusion of Innovations Theory of Rogers⁷ were used.

The study was approved by the Research Ethics Committee of the Municipal Department of Health and the Federal University of Minas Gerais.

Results

The number of store-and-forward teleconsultations grew from 2006 to 2009, following the deployment process. In 2010 there was a marked decline in the number of store-and-forward teleconsultations and in the number of physician users (Tables 1 and 2).

Table 1.

Store-and-Forward Teleconsultations by Health District, 2006–2010

DISTRICT	2006	2007	2008	2009	2010	TOTAL
South Central	28	42	130	146	48	394
Barreiro	1	0	31	39	1	72
Northwest	0	0	23	38	9	70
Venda Nova	0	1	34	18	4	57
West	0	3	23	10	12	48
North	1	0	18	14	0	33
East	0	0	14	4	0	18
Pampulha	0	1	0	8	2	11
Northeast	0	0	2	2	0	4
Total	30	47	275	279	76	707

Table 2.

Adoption Rate in 2008–2010 by Percentage of Physicians Who Used the Store-and-Forward Teleconsultations

	2008			2009			2010
DISTRICT	POTENTIAL	USERS	%	POTENTIAL	USERS	%	POTENTIAL	USERS	%
Barreiro	155	10	6.5	160	7	4.4	165	1	0.6
South Central	108	20	18.5	100	13	13.0	106	5	4.7
East	105	2	1.9	122	3	2.5	131	0	0.0
Northeast	7	1	14.3	23	1	4.3	144	0	0.0
Northwest	105	6	5.7	130	9	6.9	161	5	3.1
North	118	5	4.2	121	4	3.3	138	0	0.0
West	135	2	1.5	135	2	1.5	134	1	0.7
Pampulha	0	0	0.0	41	4	9.8	63	1	1.6
Venda Nova	128	6	0.0	128	3	2.3	138	1	0.7
Total	861	52	6.0	960	46	4.8	1,180	14	1.2

The number of physicians registered as users was lower than the number of potential physician users (Table 3). The adoption rate in relation to registrants was higher than the rate of adoption in relation to the potential, but did not exceed 3.4%.

Table 3.

Adoption Rate in 2010 by Percentage of Physicians Registered in the Belo Horizonte Telehealth System Who Used Store-and-Forward Teleconsultations

	PHYSICIANS
DISTRICT	POTENTIAL	REGISTERED	USERS	% OF REGISTERED USERS
South Central	106	66	5	7.6
Northwest	161	67	5	7.5
Barreiro	165	50	1	2.0
Pampulha	63	26	1	3.8
Venda Nova	138	36	1	2.8
West	134	72	1	1.4
North	138	27	0	0.0
East	131	35	0	0.0
Northeast	144	33	0	0.0
Total	1,180	412	14	3.4

Overall, 85 physicians used the store-and-forward teleconsultations service from 2006 to 2010 (Table 4). Of these 85 physician users, 8 were responsible for 50.9% of all teleconsultation requests (Table 4). In 2010, the utilization remained tied to these “heavy users” of the service.

Table 4.

Number of Store-and-Forward Teleconsultations Made, 2006–2010

NO. OF STC/PHYSICIANS	NO. OF PHYSICIANS	%	NO. OF STC	%
1–2	44	51.8	65	9.2
3–4	8	9.4	27	3.8
5–6	4	4.7	22	3.1
7–8	9	10.6	67	9.5
9–10	3	3.5	27	3.8
11–20	9	10.6	139	19.7
21 or more	8	9.4	360	50.9
Total	85	100	707	100

STC, store-and-forward teleconsultations.

Some physicians stopped using BH Telessaúde in the years following their initial experience. For example, of the 15 physicians who started using the system in 2006, only 3 were still using the system in 2010.

Table 5 presents the profiles of the physicians who used the store-and-forward teleconsultation.

Table 5.

Characteristics of Physicians Who Used Store-and-Forward Teleconsultations, 2006–2010

VARIABLE	N	%
Gender
Female	49	57.6
Male	36	42.4
Total	85	100.0
Age (years)^a
24–30	11	12.9
31–35	17	20.0
36–40	16	18.8
41–45	13	15.3
46–50	8	9.4
51–55	13	15.3
56–61	7	8.2
Total	85	100.0
Specialty practiced
Family medicine	51	60.0
Internal medicine clinic	16	18.8
Pediatrics	9	10.6
Gynecology	3	3.5
Psychiatry	3	3.5
Others	3	3.5
Total	85	100.0
District located
South Central	31	36.5
Northwest	15	17.6
Barreiro	12	14.1
Venda Nova	8	9.4
North	7	8.2
Pampulha	5	5.9
East	3	3.5
West	3	3.5
Northeast	1	1.2
Total	85	100.0
Years since graduation^b
0–5	13	15.3
6–10	20	23.5
11–15	11	12.9
16–20	17	20.0
21–25	9	10.6
26–30	10	11.8
31–35	5	5.9
Total	85	100.0
Years of service^c
0–5	40	47.1
6–10	20	23.5
11–15	9	10.6
16–20	4	4.7
21–25	7	8.2
26–28	5	5.9
Total	85	100.0
Employment
Civil servant	66	77.6
Contract	16	18.8
UFMG	3	3.5
Total	85	100.0

Mean, 41.9 years; median, 40.4 years.

Mean, 15.1 years; median, 14.0 years.

Mean, 8.34 years; median, 6.1 years.

UFMG, Federal University of Minas Gerais.

Discussion

The growth in the extent and geographic use of BH Telessaúde observed in 2008 and 2009 may reflect the investments made by the Municipal Department of Health, which publicized the tools among municipal health workers during this period. In 2010 management changes led to a reduction of such efforts, which could explain the observed decline in teleconsultations that year. Other factors, however, may be involved.

It is known that the type of decision that gave rise to the innovation (authoritarian, collective, or voluntary), communication channels used its implementation, the social context (organizational environment, institutional norms, opinion leaders), and the performance of those individuals promoting the use of the tool are determinant factors in the process of diffusion of this innovation.⁷

According to Karahanna et al.,⁸ while in the early stages, the use of a new technology depends on institutional norms; subsequently the adherence to the system depends mainly on the perception of users regarding the characteristics of innovation (relative advantage, ease of use, usefulness, compatibility, possibilities for experimenting, ability to demonstrate results). Thus, one can assume that despite the adoption of the new tool when it was implemented, structural factors such as dissatisfaction with the results of the consultation, incompatibilities between the routine operations of primary care units, and the context of the use of the tool may have interfered with the continuity of its use. Some physicians stopped using BH Telessaúde in years following their initial experience, suggesting dissatisfaction with the system; however, this analysis lacks further elements, which this study cannot resolve.

According to the theory of the distribution of individuals in an organization of Rogers,⁷ the time it takes to adopt an innovation follows a standard normal distribution. This pattern allows classification of organization members into five groups:

1. Innovators (2.5%) are bold and use the innovation soon after its release.

2. Early users (13.5%) are persons who or organizations that exert leadership.

3. Early majority users (34%) are more practical and deliberate before adopting an innovation.

4. Late majority users (34%) are skeptical and generally adopt the innovation because of social pressure or economic necessity.

5. Laggards (16%) are individuals who are highly conservative and suspicious of change.

It is notable that few people use an innovation as soon as it is released (innovators and early users).⁹ It is also common to observe a “chasm” between the accession of the first users (innovators and early users) and the incorporation of a technology in the routine of an organization. Judging by the adoption rate, it is reasonable to observe that the use of the store-and-forward teleconsultations in the primary care units in BH never got across the “chasm.”¹⁰

The highest adoption rate was achieved by the South Central District. It is possible that context influenced adoption. The organizational environment, support of the organizational hierarchy, degree of network connectivity, action of the promoters of change, use of innovation by key individuals or opinion leaders, and local support of computer experts were factors identified in an earlier study.⁷

From the beginning, the South Central District stood out for having a team managing the use of teleconsultations and for its inclusion of physicians from the district's Medical Specialties Center in the role of teleconsultants.¹¹ The management team consists of representatives of the Medical Specialties Center, the Sanitary District, and the municipal Department of Health. It manages the supply of store-and-forward teleconsultations by the experts of the Medical Specialties Center, as well as the use of BH Telessaúde by physicians at the primary care units. Such procedures probably influenced the results because the context favored the formalization of telehealth activities among the professionals involved. In addition, in the South Central District, the primary care units and the Medical Specialties Center share patient care information through an electronic medical record; thus these physicians already had prior experience with information and computing technology (authors' unpublished data).

Despite successes in a few districts the overall results indicate weaknesses in the implementation process because the number of eligible physicians who registered was lower than the number of the potential users of the tool. Thus the training and the issuance of passwords did not reach 100% of physicians viewed as potential users of the tool.

Rogers' model⁷ is useful but is not sufficient to entirely explain the phenomenon of technology adoption. In addition to the external determinants described above, individual determinants also weigh upon the decision to use or not use a tool in the course of caring for patients. Although the highest adoption rate was achieved by the South Central District, “heavy users” were identified in other districts (Venda Nova and West). Such variation suggests that individual variables also come into play.

Women, family health specialists, physicians working under stable contracts, and older physicians with more time in the profession and in public service predominated among the BH Telessaúde users.

Hu et al.¹ studied the adoption of telemedicine technology in the public health sector in Hong Kong and identified the attitude of physicians as one of the most significant factors related to adoption. The authors emphasized that physicians' notions of professional autonomy, compared with that of technology users in other sectors, may explain the independence of an individual when electing to adopt a technology. In New Mexico, the authors associated difficulties implementing a program of rural telehealth with the optional nature of the adoption of the tool and lack of incentives for its use.¹² It is important to mention that in the health system analyzed in this study, the use of services by the physician is at his or her discretion.

In the United Kingdom, a survey by Snoden and Harrison¹³ compared general practitioner users and nonusers of telemedicine technologies considering the following variables: age, gender, length of experience, frequency of use of a computer, types of software applications used (registry of consultations, research, administrative applications, clinical information, prescription), and attitudes toward information and communication technologies. The authors observed a statistically significant difference between the experimental group of users (volunteer general practitioners in a telemedicine study) and the control group. The age variable stood out in its ability to explain adoption, with younger individuals more likely to use the tool.¹³ In contrast, in the United States,¹⁴ age, gender, and time since graduation were not associated with the greater or lesser use of the tool; instead, that study highlighted that adoption reflected the perception by users of the advantages and gains associated with its use.

The analysis performed here has limitations. It was not possible to analyze the dynamic relationships among different actors and social interests that underlie the municipal health system. All stakeholders (managers, physicians, patients) play an active role and can influence their community. The categorization of users (innovators, early, initial majority, late majority, and resistant) is not sufficient to explain the complex process of difussion.¹⁵

Future research is needed to examine the situations in which off-line teleconsulations are used and how the technology can be a central mediator in the construction and reproduction of new panoramas, including professional identities, work organization, and the management of services in the municipality.¹⁵

If the adoption rate is so low, what kind of mechanisms do primary care physicians use to handle complicated cases? Would it be plausible to suppose that either the physicians are sufficiently trained or many of the cases are not being resolved?

It is known that physician interest in new technologies depends on the appropriation of solutions previously established and considered satisfactory in the real workplace. Accumulation of tasks and deficiencies in access to training may hinder the adoption of technological tools.¹⁵

In summary, contextual variables and individual variables may be influencing the adoption of teleconsultations in BH. Despite the enthusiastic adoption by a group of physicians, teleconsultation has not yet reached a critical mass of users that will ensure the consolidation of the program. Understanding what factors constitute barriers to use and what can be done to improve education may contribute to reversing the situation.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

PJH

, Chau

PYK

, Sheng

ORL

. Investigation of factors affecting organization's Adoption of telemedicine technology. 33rd Hawaii International Conference on System Sciences. 5. 2000. www.computer.org/portal/web/csdl/doi/10.1109/HICSS.2000.926799. 2010 April .

Organização Pan-Americana de Saúde. Estratégia para o desenvolvimento de competências dos profissionais de saúde nos sistemas de saúde baseados na Atenção Primária. Washington, DC: 50° Conselho Diretor, 62°Sessão do Comitê Regional, 2010.

Departamento de Atenção Básica, Secretaria de Atenção à Saúde, Ministério da Saúde, Brazil. Política Nacional de Atenção Básica. Série Pactos pela Saúde. 2006. www.saude.gov.br/pactos pela saude.htm. 2012 January .

Santos

, Alkimim

MBM

, Souza

, Santos

, Alves

, Melo

MCB

. BH-Telessaúde: A experiência de um modelo de telessaúde de baixo custo voltado para área pública. Santos

, Souza

, Alves

, Santos

. Telessaúde: Um instrumento de suporte assistencial e educação permanente. Belo Horizonte, Brazil: Universidade Federal de Minas Gerais, 2006.

Santos

, Souza

, Santos

, Melo

MCB

, Trad

. Incorporação de recursos de telessaúde na atenção primária. Magalhães

Jr . Desafios e inovações na gestão do SUS em Belo Horizonte: A experiência de 2003 a 2008. Belo Horizonte, Brazil: Mazza, 2010.

Santos

, Santos

, Melo

MCB

, Silva

, Reis

, Souza

. Telehealth in primary healthcare: An analysis of Belo Horizonte's experience. Telemed J E Health, 2011; 17:25–29.

Rogers

. Diffusion of innovation. New York: Free Press, 2003.

Karahanna

, Straub

, Chervany

. Information technology adoption across time: A cross-sectional comparison of pre-adoption and post-adoption beliefs. MIS Q, 1999; 23:183–213.

Bateman

, Snel

. Administração: Construindo vantagem competitiva. São Paulo: Atlas, 1998.

10.

Clark

, Goodwin

. Sustaining innovation in telehealth and telecare. WSDAN briefing paper. 2010. www.technologyadoptionhub.nhs.uk/assets/_files/documents/apr_10/nhs__1271416227_Organisational_and_Behavioural.pdf. 2011 April .

11.

Reis

, Teixeira

DMC

, Lages

, Silva

RHLP

, Dornas

Jr , Santos

. A utilização da telessaúde em um centro de especialidades médicas. Santos

, Souza

, Alves

, Santos

. Telessaúde: Um instrumento de suporte assistencial e educação permanente. Belo Horizonte, Brazil: Universidade Federal de Minas Gerais, 2006.

12.

Helitzer

, Heath

, Maltrud

, Sullivan

, Alverson

. Assessing or predicting adoption of telehealth using the Diffusion of Innovations Theory: A practical example from a rural program in New Mexico. Telemed J E Health, 2003; 9:179–187.

13.

Snoden

, Harrison

. General practitioner participants in a telemedicine trial: Comparisons with their peers. J Telemed Telecare, 2001; 7:32–37.

14.

Barton

, Brega

, Devore

, Muller

, Paulich

, Floersch

, Goodrich

, Talkington

, Bontrager

, Grigsby

, Hrincevich

, Neal

, Loker

, Araya

, Bennett

, Krohn

, Grigsby

. Specialist physicians' knowledge and beliefs about telemedicine: A comparison of users and nonusers of the technology. Telemed J E Health, 2007; 13:487–499.

15.

Trindade

. A incorporação de novas tecnologias nos serviços de saúde: O desafio da análise dos fatores em jogo[online] Cad Saúde Pública, 2008; 24:951–964.