Exploring the Patterns of Use and the Feasibility of Using Cellular Phones for Clinic Appointment Reminders and Adherence Messages in an Antiretroviral Treatment Clinic,Durban,South Africa

Abstract

In preparation for a proposed intervention at an antiretroviral therapy (ART) clinic in Durban, South Africa, we explored the dynamics and patterns of cellular phone use among this population, in order to ascertain whether clinic contact via patients' cellular phones was a feasible and acceptable modality for appointment reminders and adherence messages. Adults, who were more than 18 years old, ambulatory, and who presented for treatment at the clinic between October-December 2007, were consecutively recruited until the sample size was reached (n = 300). A structured questionnaire was administered, including questions surrounding sociodemographics, cellular phone availability, patterns of use, and acceptability of clinic contact for the purpose of clinic appointment reminders and adherence support. Most respondents (n = 242; 81%) reported current ownership of a cellular phone with 95% utilizing a prepaid airtime service. Those participants who currently owned a cellular phone reported high cellular phone turnover due to theft or loss (n = 94, 39%) and/or damage (n = 68, 28%). More females than men switched their cell phones off during the day (p = 0.002) and were more likely to not take calls in certain social milieus (p ≤ 0.0001). Females were more likely to share their cell phone with others (p = 0.002) or leave it in a place where someone could access it (p = 0.005). Most respondents were willing to have clinic contact via their cellular phones, either verbally (99%) or via text messages (96%). The use of cellular phones for intervention purposes is feasible and should be further investigated. The findings highlight the value of gender-based analyses in informing interventions.

Introduction

V arious technological approaches have been developed to expand the health encounter and improve health promotion in a range of medical conditions. In the absence of land lines and because of the high mobility of the population, cellular phones are used widely throughout South Africa. The utilization of cellular phones in the HIV/AIDS-related health care setting has, in particular, received attention; telecommunication strategies in supporting health care services and systems in resource-limited settings are on the international public health agenda.¹ The effectiveness of telecommunications technology has been demonstrated in a variety of related areas; supporting antiretroviral treatment (ART) adherence,^2
–4 facilitating HIV testing⁵ and monitoring and follow-up of HIV-infected patients by clinicians.⁶ There is ongoing cellular phone research in a range of HIV-related health care services in South Africa⁷ and the first multisite randomized clinical trial in Kenya, looking at various patient outcomes including ART adherence via a short message service (SMS) intervention, is well underway.⁸

Sub-Saharan Africa is at the epicenter of the global HIV epidemic. In the South Africa, 2009 national HIV prevalence and syphilis survey, an estimated 29.3% of antenatal attendees were HIV-infected. KwaZulu-Natal, home to 20% of the South African population, has an extremely high antenatal HIV prevalence of 38.7%.⁹ Ethekwini (Durban) is the district in KwaZulu-Natal with the highest prevalence of all; 40.3%.⁹ Despite the high national unemployment rate, use of cellular phones is widespread among South Africans.¹⁰ The South African cellular network covers over 71% of the population, including the remoter regions of the country. Most urban areas and national roads have full network coverage.

In preparation for an intervention to improve ART adherence and clinical attendance rates at a busy urban ART clinic in Durban, South Africa, we investigated the dynamics and patterns of cellular phone use among this patient population to assess whether clinic contact via patient's cellular phones (either verbal contact or text messaging services) was a feasible and acceptable modality for clinic appointment reminders and adherence messages.

Methods

Setting and participants

This cross-sectional study was conducted at the McCord Hospital antiretroviral therapy (ART) clinic, KwaZulu-Natal, South Africa. McCord is a state-subsidized, district-level hospital that serves an urban and peri-urban population in the Durban Metropolitan catchment area and has extensive experience in ART services for both adults and children. As of April 30, 2007, a total of 2408 adults (cumulative total, 3340) 18 years and older were receiving treatment at the ART program. The Hospital ART Clinic charged a monthly user fee of ZAR140 (approximately $20 USD).

Recruitment

Adults who were 18 years of age or older, ambulatory, and who presented for treatment at the ART clinic between October and December 2007 were consecutively recruited and enrolled until the sample size of 300 was reached. All participants were interviewed by a female research assistant who had extensive qualitative and quantitative fieldwork experience. This research assistant was not a staff member of the hospital and received additional training regarding the study and the research instrument.

Measures

Participants were interviewed by the same interviewer using a structured questionnaire and administered in English or isiZulu, depending on respondent preference. The interview included questions about sociodemographic characteristics, cellular phone availability and patterns of use, and opinions about the acceptability of the clinic contacting the individual by voice or text messaging for the purpose of clinic appointment reminders as well as adherence support.

Analyses

Our primary analysis compared patterns of cellular phone use and willingness for clinic contact via cellular phone by gender. Odds ratios (OR) and 95% confidence intervals (CI) for associations between gender and cellular phone availability and patterns of use were calculated by logistic regression. Analyses were conducted using SAS 9.1 (SAS Institute, Cary, NC).

Ethical approval

Ethical approval for the study was obtained from the McCord Hospital Research Ethics Committee at which site the study was conducted and the Institutional Review Board of the Spaulding Rehabilitation Hospital. Respondents were asked to provide verbal consent to indicate willingness to participate.

Results

Between October and December 2007 we consecutively enrolled 300 individuals 18 years of age or older who presented for treatment at the ART clinic. There were 2 refusals. Most respondents (n = 242; 81%) reported current ownership of a cellular phone (median time of ownership = 3 years). Older age (years) (adjusted odds ratio [AOR] = 1.05, 95% CI 1.01–1.09, p < 0.01) and being currently employed (AOR = 3.6, 95% CI 1.8–7.7, p < 0.001) were independently associated with cellular phone ownership. Gender was not associated with cellular phone ownership (p = 0.97). The following analyses are based on the 242 participants who reported current ownership of a cellular phone.

Characteristics of participants who owned a cellular phone

The majority of participants were female (67%), with median age of 35 years (interquartile range, 30–40 years), and were black Africans (96%; Table 1). The male/female ratio as well as ethnicity and age group was representative of the patient profile accessing services at the ART clinic. The majority of respondents (60%) were unemployed, and approximately half had at least a secondary level education (52%). Of all respondents, 32% reported currently receiving financial support from government social grants (e.g., child care or disability grants); a significantly greater proportion of females reported accessing these grants than males (43% versus 10%, p < 0.0001). The sample was evenly balanced with respect to time since HIV diagnosis and time on ART indicating a representative selection of the patient population with regard to these dimensions.

Table 1.

Characteristics of Study Participants Who Currently Owned a Cellular Phone (N = 242)

	Male N = 80	Female N = 162	Total N = 242
	n (%)	n (%)	n (%)
Age group
Median years (IQR)	36 (31.0–41.5)	34 (30.0–40.0)	35 (30–40)
<30 years	16 (20%)	36 (22.2%)	52 (21.5%)
30 to 35 years	16 (20%)	47 (29.0%)	63 (26.0%)
36 to 40 years	24 (30%)	37 (22.8%)	61 (25.2%)
≥40 years	24 (30%)	42 (25.9%)	66 (27.3%)
Ethnicity
Black	77 (96.3%)	156 (96.3%)	233 (96.3%)
White	0 (0.0%)	1 (0.62%)	1 (0.41%)
Colored	0 (0.0%)	3 (1.9%)	3 (1.2%)
Indian	3 (3.7%)	2 (1.2%)	5 (2.1%)
Educational level
Primary or less	45 (56.3%)	70 (43.2%)	115 (47.5%)
Secondary	34 (42.5%)	75 (46.3%)	109 (45.0%)
Postsecondary	1 (1.3%)	17 (10.5%)	18 (7.4%)
Unemployed	50 (62.5%)	92 (56.8%)	142 (58.7%)
Receive government grants ^a	8 (10.0%)	70 (43.2%)	78 (32.2%)
Time since HIV diagnosis
0 to 6 months	22 (27.5%)	36 (22.2%)	58 (24.0%)
7 to 2 months	20 (25.0%)	35 (21.6%)	55 (22.7%)
>12 months	38 (47.5%)	91 (56.2%)	129 (53.3%)
Time on antiretroviral therapy
0 to 6 months	38 (47.5%)	61 (37.7%)	99 (40.9%)
7 to 12 months	18 (22.5%)	46 (28.4%)	64 (26.5%)
>12 months	24 (30.0%)	55 (34.0%)	79 (32.6%)

χ² = 27.0, p < 0.0001.

IQR, interquartile range.

High cellular phone turnover due to theft, loss and/or damage

The majority of participants (95%) utilized a prepaid airtime service. This route to accessing airtime allows the user to buy airtime on an ad hoc basis, allowing for financial constraints without being tied into a standard 2-year contract with a set monthly fee. There was high cellular phone turnover due to theft, loss, and/or damage. Close to 40% of respondents reported that they had previously owned one or more cellular phones that had been lost or stolen, and 28% reported that they had previously owned a cellular phone that had been damaged to the extent that they could not use it.

Patterns of cellular phone use

Twenty-three percent of respondents reported that they switched their cell phones off during the day (Table 2). Females had almost four times the odds of reporting that they switched their cell phone off during the day than males (p = 0.002), citing attendance at church, prayer meetings, funerals, and doctor visits as typical reasons. Fifty-nine percent of respondents reported that they would not answer their phone in certain contexts or locations. These contexts included busy streets, unsafe areas, funerals, church, work, schools. Females had almost five times the odds of reporting this than males (p < 0.0001).

Table 2.

Experience of Phone Theft, Loss or Damage, Patterns of Cellular Phone Use and Willingness to be Contacted by HIV Clinic Among Participants (N = 242)

	N (%)	OR (95% CI)	p Value
High cellular phone turnover due to theft, loss and/or damage
Has had a cellular phone lost or stolen in past
Male	27 (33.8)	ref	p = 0.21
Female	67 (42.1)	1.4 (0.82–2.5)
Cellular phone has been damaged in past such that it could not be used			p = 0.31
Male	19 (25.3)	ref
Female	49 (31.8)	1.4 (0.74–2.6)
Patterns of cellular phone use
Switch cellular phone off during day
Male	8 (10.0)	ref	p = 0.002
Female	47 (29.0)	3.7 (1.6–8.2)
There is sometimes a time or place where no calls are taken
Male	28 (35.0)	ref	p < 0.0001
Female	114 (70.4)	4.4 (2.5–7.8)
Shares cellular phone with another person(s)
Male	12 (15.0)	ref	p = 0.002
Female	56 (34.6)	3.0 (1.5–6.0)
Phone is sometimes in a place where someone could pick it up and use it or read messages on it			p = 0.005
Male	17 (21.3)	ref
Female	64 (39.5)	2.4 (1.3–4.5)
Believe that one or more persons has ever read SMSs without your permission
Male	17 (21.3)	ref	p = 0.32
Female	44 (27.5)	1.4 (0.73–2.6)
Willingness for clinic to contact by cell phone
Typically answers calls from unknown number
Male	72 (92.3)	ref	p = 0.21
Female	139 (86.9)	0.55 (0.21–1.4)

OR, odds ratio; CI, confidence interval; SMS, short message service.

Privacy

Twenty-eight percent of respondents reported that they shared their cell phone with one or more other people (Table 2). The odds of sharing a phone were significantly greater for females compared to males (OR = 3.0, 95% CI 1.5–6.0). Household members were the predominant category of person with whom cell phones were shared but with important differences between males and females. Whereas men most commonly reported sharing their cell phones with their wives (n = 5/12, 42%), females more commonly reported sharing their phones with their family members (n = 28/56, 50%) or children (n = 22/56, 39%).

One third of respondents reported that they sometimes left their cell phone in a place that allowed the potential for someone to pick it up and access it (e.g., to use it or read messages on it; Table 2). The practice was significantly more common among females than males (OR = 2.4, 95% CI 1.3–4.5).

One quarter (25%) of respondents who currently owned a cellular phone believed that their text messages had at some point been read without their permission (Table 2). However, those who elaborated on this could identify the family member/s concerned and did not appear to be unduly disturbed by this.

Willingness for clinic to contact by cell phone

The majority (87%) of participants indicated that they usually answered calls that displayed “private number” (which is how the Hospital number would be displayed) on their caller identification screen (Table 2). Among the 31 respondents who reported that they routinely did not answer calls from unknown numbers, the main reasons offered included concerns about sexual rivals phoning to insult them, crime, or being contacted by a service provider requiring debt to be paid.

Most respondents were willing for the clinic to contact them on their cellular phones either verbally (99%) or via text messages (96%). Of three persons (all women) who stated that they would not be willing for the clinic to contact them verbally, one expressed concern that she did not want to be “chased” by the clinic and two explained that they did not need reminders because they knew their appointment dates. Of the nine participants (three men, six women) who stated that a text message from the clinic would be unacceptable, concerns included their not knowing how to access the message, not being able to read messages immediately, not being able to read, that someone else might see the message, and that a reminder was not necessary because other reminder strategies were already used.

Use of cell phones as reminders

Use of the cell phone alarm function was a commonly mentioned strategy for remembering to take medication on time. Most of the participants (79%) were already utilising the alarm clock feature on their cellular phone and some regularly received text messages from treatment supporters.

Discussion

Research focusing on the efficacy and effectiveness of interventions that utilize communication technology in facilitating ART adherence is being strongly supported by various international public health agencies. In resource-limited settings, in particular, communication technology offers promising possibilities in addressing current HIV-related health care services infrastructural and human resource constraints. The purpose of this research was to understand existing patterns of cell phone use in a population on ART, as well as assess attitudes regarding the possibility of the clinic using cell phones to communicate with clients.

Among patients attending an urban ART clinic in Durban, South Africa, we found that the majority (81%) of patients currently owned a cellular phone. As such, for most patients, contact by cellular phone is a viable option. However, our findings highlight several issues that should be considered in the design of an intervention involving patient contact by cellular phones. For example, the issue of loss of cell phone devices due to theft and/or damage was common, and would have implications in terms of sustainability of such an intervention, as well as issues of confidentiality with lost devices. While the device itself could be relatively easily replaced, the majority of participants used prepaid airtime which meant that they would lose their original phone number once they had acquired their new cell phone with another Subscriber Identity Module (SIM) card which is preallocated the number. Regular updating of patient contact details at each clinic visit was therefore demonstrated as imperative.

We found differences in the patterns of use of cellular phones between men and women. Women were significantly more likely to have their cellular phones used by others in their lives and to not be contactable either due to refusing to answer their phone or to switching it off in certain social settings. Text messaging would address the problem of cell phones users not being available at the time of contact, but the issues around confidentiality need to be carefully considered. Gendered patterns of cellular phone use need to be taken into account in the design and implementation of clinical interventions that rely on cellular phone use. In particular, considerations around the framing of the specifics of the clinic message need to be carefully planned, with the input of clinic patients and counselors.

Despite the gender differences in patterns of cellular phone use, the majority of participants, regardless of gender, indicated willingness to be contacted by the clinic, either in the form of a phone call or via a text message. Interestingly, while receiving a text message treatment reminder was considered helpful to the majority of participants, it was not viewed by patients as critical to the success of their treatment schedule. A recent Canadian study, on the other hand, reported that electronic reminder devices were identified as helpful by respondents.¹¹ Median adherence with a variety of reminder tools was 95% and specifically with electronic devices was reported as 76%. In another study involving use of the internet to support persons living with HIV, 500 caregivers were recruited to complete an online survey of Internet use and self-efficacy. ¹² The findings suggest that most respondents (72%) used Internet websites for information and resources, thus lending credence to the important role of Internet-type or electronic reminder technologies in both HIV adherence and knowledge dissemination.

One recent study¹³ in a sample of perinatally infected youth found that self-efficacy and outcome expectancy were significantly higher in adherent versus nonadherent subjects. For those subjects with low self-efficacy and outcome expectancy, adherence results differed based on the presence or absence of either mental health or structural barriers. Thus, as the global diffusion of mobile and Internet-based systems expands, the need for further exploration of these technologies and their ability to augment HIV-related behavioural interventions, self-efficacy, and motivational readiness for adherence to ART should be examined.¹³

This study has several important strengths. It included a representative sample of ART clinic patients who were well distributed in terms of sex ratio, time since HIV diagnosis, and time on ART. While most participants indicated high acceptability of an intervention such as this, disaggregating the data by sex drew attention to the often hidden gender considerations and the long term implications of interventions such as the one proposed. This kind of analysis allows for a more nuanced understanding of generalized assumptions surrounding privacy and confidentiality in the context of ART care and support.

There were some limitations to this study as well. Because of the cross-sectional study design, we cannot examine patterns of cellular phone usage over time. In addition, it is possible that we have overestimated patient willingness to receive reminders via cell phones given the possibility of courtesy bias. Specifically, since information was collected from patients in the clinic where they received care, it is possible that participants were reluctant to express negative opinions to an interviewer who may be perceived to be associated with the facility from which they receive care. Furthermore, the information collected was based on participants' recall of aspects of their prior cellular phone usage. It is possible that recall bias could result in inaccurate information being inadvertently provided or collected. Furthermore, generalizability of our results to other populations should be made with some caution. Given the clinic entrance fee, the study may have selected a slightly different sociodemographic sample of patients who may not be representative of HIV-infected individuals with cell phones in the study catchment area.

While the study shows that the potential value of using cell phones to improve adherence in patients on ART, it also highlights the need for nuanced interventions which respond to contextual dynamics as well as gender considerations. Given the substantial loss and sharing of cell phones, technology that requires a password to access text messages should protect the confidentiality of the recipient with regard to the message content. Carefully framed clinic messages which are ambiguous enough not to generate suspicion if accessed by an unintended recipient, but at the same time being directed enough to be recognizable to the intended recipient also need to be developed with the input of patients. With regard to gender considerations, disaggregating the study data by sex provides the opportunity for a gender-based analysis. This study underlines the value of an exploratory feasibility study prior to interventions being implemented. In addition, the role of self-efficacy and its link with electronic reminder devices including cell phones requires further study.

Footnotes

Acknowledgments

This study was carried out with financial support from the Maurice Webb Trust Fund.

Ethical approval for the study was obtained from the McCord Research Ethics Committee and the Institutional Review Board of the Spaulding Rehabilitation Hospital.

The authors would like to thank and acknowledge Sibongile Maimane for conducting the interviews.

Presented in part at the 3rd International Conference on HIV Treatment Adherence, Jersey City, New Jersey, March 17–18, 2008.

Author Disclosure Statement

No competing financial interests exist.

References

UNAIDS. Resource needs for an expanded response to AIDS in low- and middle-income countries. Geneva: World Health Organisation, August 2005.

Andrade

, Wu

, Skolasky

et al.

Randomized feasibility study of a cell phone and a pager as potential adherence enhancing devices [abstract]

XV Intermational AIDS Conference. Bangkok, Thailand: eJIAS. eJournal of the International AIDS Society, 2004. www.aegis.com/conferences/iac/2004/B12064.html. 2010 April 7.

Puccio

, Belzer

, Olson

et al. The use of cell phone reminder calls for assisting HIV-infected adolescents and young adults to adhere to highly active antiretroviral therapy: A pilot study. AIDS Patient Care STDs, 2006; 20:438–444.

Rueda

, Park-Wyllie

, Bayoumi

et al. Patient support and education for promoting adherence to highly active antiretroviral therapy for HIV/AIDS. Cochrane Database Syst Rev, 2006; 19:CD001442.

Morin

, Khumalo-Sakutukwa

, Chalebois

et al. Removing barriers to knowing HIV Status: Same day mobile HIV testing in Zimbabwe. J Acquir Immune Defic Syndrome, 2006; 41:218–224.

Kaiser Daily HIV/AIDS Report. Reuters profiles South African program using text messages to monitor, track HIV/AIDS patients. www.kaiserhealthnews.org/daily-reports/2005/august/19/dr00032102.aspx?referrer = search. 2010 April 6.

The Cellphone for HIV project. www.cell-life.org. 2010 April 6.

Lester

, Mills

, Kariri

et al. The HAART cell phone adherence trial (WelTel Kenya1): A randomised controlled trial protocol. Trials, 2009; 10:1–10.

Department of Health. National HIV and Syphilis Antenatal Sero-Prevalence Survey in South Africa, 2009. www.doh.gov.za/docs/index.htm. 2010 April 6.

10.

Stats SA (2006) General Household Survey, 2005. www.statssa.gov.za/Publications/P0318/P0318July2005.pdf. 2010 April 6.

11.

Ostrop

, Gill

. Antiretroviral medication adherence and persistence with respect to adherence tool usage. AIDS Patient Care STDs, 2000; 14:351–358.

12.

Horvath

, Courtenay-Quirk

, Harwood

et al. Using the internet to provide care for persons living with HIV. AIDS Patient Care STDs, 2009; 23:1033–1041.

13.

Rudy

, Murphy

, Harris

, Muenz

, Ellen

. Adolescent Trials Network for HIV/AIDS Interventions. AIDS Patient Care STDs, 2010; 24:97–104.

14.

Swendeman

, Rotheram-Borus

. Innovation in sexually transmitted disease and HIV prevention: Internet and mobile phone delivery vehicles for global diffusion. Curr Opin Psychiatry, 2010; 23:139–144.