Piloting Collection Model of Health Insurance Contributions for

Informal Sector Members

Mahlil Ruby

, Eka Pujiyanti

, Euis Ratna Sari

, Gemala Chairunissa Puteri

, Riza Maulana

Putra

, Unun Hamida Qodarina

, Dinda Srikandi

, Mutia Anggun Sayekti

, Dedy Revelino

Pardamean Sirega

, Rizki Lestari

and Welly Gadistina

Center for Health Economics and Policy Studies (CHEPS) Faculty of Public Health, Universitas Indonesia, Depok,

Indonesia

The Indonesia Health Social Security Institution (BPJS Kesehatan), Central Jakarta, Jakarta, Indonesia

euisratnasari.jasmin@gmail.com

Keywords: Health insurance contributions, Informal sector, Promotor agent, JKN.

Abstract: Studies conducted by BPJS and CHEPS, 2015 showed that an active collection model through a third

party (a promoter agent) can improve the collectability status of informal sector members. This study

aimed to do a piloting collection model by a promoter agent (individuals, primary care facilities (PCF),

and local leaders) as well as informed about the ideal criteria of being agents with its challenges.

Methods: A quasi experimental design with non-equivalent group design (NEGD) was implemented

on two provinces with 1,509 households’ participants in Semarang and 1,193 households in

Balikpapan. Results: an individual promoter agent had 37.5% lowest proportion of being delinquent.

Multivariate regression analysis revealed that participants fostered by PCF and local leader as their

promoter agents had higher risk for being delinquent than those facilitated by individual promoter

agents. The effectiveness of the model differs between Semarang and Balikpapan with Individual

Agent Model showed a significant effect in Semarang. While in Balikpapan, there were no meaningful

models to improve the collectability. At last, the ideal individual agents have to hold: an experience

working in the community, have highly social skills as well as highly understanding of their

intervention regions.

1 INTRODUCTION

The number of participants BPJS Health up to

December 2016 has reached 171,048,734

participants equal BPJS Health achieves 68.5% of

the target of UHC and BPJS Health should be able

to cover about 75 millions of residents in the

remaining 2 years to 2019. From the contribution

aspect, the growth of informal sector members who

exceed this target ideally can support revenue

contribution of JKN, but in fact the realization of

contributions of informal sector members until the

end of 2015 and then reached 2.8 trillion rupias. This

amount is far above the initial target of the

determination of the receivables of informal sector

members estimated by BPJS Health at 1.8 trillion

rupias. The high receivables are contributed by the

low collectability of informal sector contributions

that only reached 60%. This will certainly affect the

national health financing cash flow, especially for

BPJS Health which can ultimately have implications

for financial management and services in health care

providers.

In the context of informal sector members, there

were four main reasons impact delinquency rate:

erratic income (23.6%), reluctant to queue (15.8%),

others reasoning (16.8%) such as (ATM Offline,

took a long time to pay at bank, forgot to pay

(12.8%), disappointed with provider or BPJS Health

(6,8%) (Ruby, 2016).

In Kusumasari & Widiastuti (2013), there are

four element can influence person’s behaviour to

pay health insurance contributions. Action is built on

the condition that a person wants or feels the need

for a health service guarantee, especially when they

sick (desired). However, when people health, it is

necessary to build an interest in informal sector

worker that make them interested in joining JKN

program, which in their perception can be a

protector of their risk from illness or risk of financial

burden due to illness (interest). To achieve these

perceptions, it is necessary to build awareness of in

informal sector members in the existence of JKN in

protecting themselves and their health (awareness).

242

Ruby, M., Pujiyanti, E., Sari, E., Puteri, G., Putra, R., Qodarina, U., Srikandi, D., Sayekti, M., Siregar, D., Lestari, R. and Gadistina, W.

Piloting Collection Model of Health Insurance Contributions for Informal Sector Members.

In Proceedings of the 4th Annual Meeting of the Indonesian Health Economics Association (INAHEA 2017), pages 242-247

ISBN: 978-989-758-335-3

If the AIDA link to informal sector members has

been established it is not impossible that the level of

collectability of health insurance contributions in

this type of participants may increase. It takes a

method or model of collection contributions that

more than just provision of payment channels that

tend to be passive. It takes an officer or individual

who actively and directly interacts with the

participants to guide, educate and build awareness of

participants' behaviour in paying health insurance

contributions, which will be known as BPJS Health

Promoter Agent.

The general objective of this activity is the

implementation of the Piloting Collection Model of

Health Insurance Contributions for Informal Sector

Members to enhance the collectiveness and

sustainability of group contributions of informal

workers in the achievement of the National Health

Insurance (JKN) program.

2 METHODS

The design of this study used quasi-experimental

method. Quasi experiments are experiments that

have treatments, outcome measures, and

experimental units but do not use random

placements.

The quasi experimental design used in this study

is Non-Equivalent Group Design (NEGD). This is

because in the assessment to be done on the level of

collectability status that will be compared not only

between groups of intervention with the control

group alone, but also among fellow intervention

groups with different promoter model models.

Baseline collectability level compared to the end line

collectability level on each model as an intervention

agency and the intervention by the control.

The selection of both groups was done

purposively. A total of 300 informal worker

households were designated as intervention targets,

but due to field dynamics faced by agents, the

number of households reached for each agency was

at least 150 households. The number of households

that can be visited is 6 - 8 households per day per

promoter agent, so that within 1 month (25 working

days) will be achieved the visit of 150 households of

informal workers per month. While the control areas

were not intervened, 150 informal workers'

households in the control areas were adjusted to the

number of participants for each intervention and

observed their initial contribution rate (as baseline

data) and at the end of the program (end line).

The unit of analysis in this study was the

household of informal worker participants who

participated in the BPJS Health in one cluster (1

cluster, consisting of 150 households). Interventions

conducted at Central Java Province, Semarang City

and East Kalimantan Province, Balikpapan City

The population in this study was all households

of informal workers in the experimental model of

active dues collection model, namely Central Java

and East Kalimantan provinces. While the sample in

this study is the head of the family who became the

respondent survey of the collection model of

contribution stage 2 in 6 cluster points per province

(1 point cluster contains 150 households).

This trial was conducted 2 (two) stages. Phase 1

is held on October 20, 2015 until December 23,

2015. Stage 2 is held from April 2016 to 31 October

2016.

3 RESULTS

Data was analysed by univariate, bivariate, and

multivariate analysis. Variables were used consist of

number of family members, children under 5 years

old and oldest of family members, services class,

worker number of family members, sex, age,

education background, married status, smoking

behaviour, sector of work, head of household

income, health status at first registration, economic

status, catastrophic status, and respondents’

delinquency status in the baseline and end-line of the

research. Data was category and used SPSS software

to analyse.

The data was analysed by univariate to show the

frequency of all variables. Bivariate analysis was

done to show the relationship between collectability

status baseline and end-line and others variables

which had represented social economic, and

demography aspect. Bivariate analysis used chi

square test by city, rural urban cluster, and village

sample.

Besides, chi square was used to show the

differences increase of delinquency rate between

intervention and control members. So, it could show

the effectivity of intervention. Effectivity analysis

also described based on promoter agent model in

this intervention which shown the lowest proportion

of delinquency rate by the promoter agent model.

After found the affecting factor of collectability

rate, the data was analysed multivariate to show the

most affecting factor of collectability rate.

Multivariate analysed used logistic regression test by

looking the highest Odds Ratio (OR) value of

independent variables. Multivariate analysed was

done on city area sample level (Semarang and

Balikpapan)

Piloting Collection Model of Health Insurance Contributions for Informal Sector Members

243

In this study there are a total of 2702 household

heads of informal sector members who responded to

the trial of applying an active JKN dues collection

model through a promoter agent. This number

consisted of 1509 household heads in Semarang city

(55.8%) and 1193 household heads in Balikpapan

(44.2%).

1. Demography profile

Table 1: Demography Profile of Respondents in Semarang

and Balikpapan 2016

Variables Semarang Balikpapan

Sex

Male 75.2% 80.9%

Female 24.6% 19.1%

<26 years ol

2.8% 3.5%

26-40 years ol

28.9% 27.4%

41-58 years ol

46% 47.3%

> 58

ears ol

22.4% 21.8%

Majority of the respondents are male which

were 75.2% in Semarang and 80.9% in Balikpapan.

But there are around twenty percent respondents are

female who act as head of household for informal

workers who become their household members. This

data show that the role family health insurance payer

majority are heads of families, fathers or sons who

are in the household. However, the table above

showed women role as family health insurance payer

is rising.

Most (3 of 4) payer contributions in the

households of informal workers in Semarang are at

working age (26 - 58 years). However, not a few

workers in informal workers' salaries are more than

58 years old (retirement age), 22%. This trend

occurs in all villages sample. Not much different

from the characteristics of the age of underwriters in

the city of Semarang, 3 of 4 underwriters in the city

of Balikpapan are also at the age of work and the

rest are at retirement age.

2. Social and Economic Profile

Table 2: Social Economic Profile of Respondents in

Semarang and Balikpapan 2016

Variables Semaran

Balik

Married status

Single 5.5% 4.3%

Marrie

82.5% 79%

Divorce 11.5% 16.5%

Education back

roun

< SHS 23.1% 48.4%

SHS

–

Vocation/ Bachelo

49.4% 48.9%

> Vocation- Doctoral 27.3% 2.7%

Most of the respondents are married 83% in the

city of Semarang and 79% of Balikpapan. Both in

Kota Semarang and Balikpapan, underwriters in

households of informal sector member are

dominated by heads of households with high school

education – Diploma (48%). This indicates that most

of the contributors in informal sector member

households are educated, although there are still

29% of respondents who have junior high school

education. In this group, potential contribution

arrears most occur.

From Semarang data in Figures and Balikpapan

in Figures 2015 indicates that 60% of the workforce

population in Semarang and 27% of the workforce

population in Balikpapan work in the informal sector

(PBPU). In workforce aspect, most of them work in

employment in services (71%) and 15% in trading.

Only 3% of them work in agricultural. Of those who

work, 58% work more than 40 hours a week.

Nevertheless, there are still 5% health insurance

payers are unemployed and 37% work less than 40

hours a week. In this group of households also the

potential contributions of arrears most occur.

Table 3: Income Profile of Respondents in Semarang and

Balikpapan 2016

Area

sample

Income Profile Category Percentage

Semarang < Rp 1,900,000 27.6%

1,900,000-R

4,500,000 52.4%

> R

4,500,000 6.4%

Balikpapan < Rp 2,200,000 53.2%

Rp 2,200,000-Rp 4,500,000 36.5%

> Rp 4,500,000 10.2%

84% of the participants of the informal sector

who responded to the pilot project in Semarang are

worked. Of those who worked, more than half of the

respondents earned between Rp 1,900,000 to Rp

4,500,000, only 6.4% of respondents earned more

than Rp 4,500,000 while 4 out of 10 respondents

were still earning less than the minimum regional

wage (UMR) of Semarang, Rp 1.900.000 (27.6%).

Meanwhile, in Balikpapan city, 86% of family

members of informal sector who participated in the

pilot project in this city are worked. Of those who

worked, 36.5% of the respondents earned between

Rp 2,200,000 to Rp 4,500,000, only 10.2% of

respondents earned more than Rp 4,500,000 while

almost 5 out of 10 respondents still earned less than

UMR in Balikpapan at Rp 2,200,000 (53.2%).

INAHEA 2017 - 4th Annual Meeting of the Indonesian Health Economics Association

244

3. Other Social Economics Variables

Table 4: Economic Status and Catastrophic Rate in

Semarang and Balikpapan 2016

Variables Semaran

Balik

Catastro

hic rate

Yes 10.6% 22.1%

No 89.4% 77.9%

Economic status

Poo

12.7% 35.4%

No Poo

87.3% 64.6%

In this study also identified poverty status of

underwriters in informal sector members. Poverty

criterions were based on Social Ministry Policy

No.46 of 2013 about Fakir dan Orang Tidak

Mampu. The data showed that 12.7% of informal

sector member in the pilot project model of health

insurance contributions collection in Semarang

categorized as poor household in Semarang. In

addition, there are 10.6% of informal sector

household falls into the category of catastrophic

conditional, which is a household financial condition

where 30% of the income is charged to pay dues.

In Balikpapan city 35.4% of respondents in this

city are identified as having poor status based on

Social Ministry Policy No.46 of 2013 about Fakir

dan Orang Tidak Mampu.. In addition, 22.1% of

informal sector member fall into the category of

catastrophic conditional, which was, the financial

condition of informal sector members with a

proportion of 30% contribution to income.

Percentage of participants in the informal workers

who fall into the poor category and experience

greater condensed catastrophic conditions in

Balikpapan than the percentage in Semarang. This is

allegedly because of the cost of living (living cost)

in the city of Balikpapan. Balikpapan is currently a

costly living city according to various references.

4. Healthy profile

Table 5: Healthy profile of Respondents in Semarang and

Balikpapan 2016

Variables Semarang Balikpapan

Health status at re

istration

Health 89.8% 96.6%

Sic

10.2% 3.4%

Services class

class 20.2% 16.8%

class 36% 25.4%

class 43.8% 57.8%

Most of the respondents in Semarang who

registered as member of BPJS Health were good

health (8.89%). This means, 1 out of 10 respondents

registered as BPJS health’s members were taken

sick. Based on the class of care taken, 43.8%

informal sector members who as respondents in

Semarang used 3

class of services. In Balikpapan,

96.6% of informal sector members who participated

in the pilot project admitted being in good health at

the first registration as a participant and 57.8%

respondents chosen 3

class of services.

5. Level of Contribution Collectability

In this pilot study conducted collection of dues

through the placement of promoter agencies BPJS

Health in some sub-districts which has been

established into the intervention area of study. In

each city, six sub-districts have participating

households in the largest informal worker group. In

the sub-district, each village was selected to

intervene. In each village a promoter agent is placed,

with one type between individual promotional

agents, promoter agencies working with FKTP and

promoter agencies in collaboration with village

officials.

To see the level of contribution collectability, the

researcher uses the respondent analysis unit is the

individual with a total of 7856 individuals

participating in the informal sector members who

participated as the respondents of this pilot project

actively through the promoter agent. This number

consisted of 4,282 individuals participating in

informal sector members in Semarang, Central Java

(54.5%) and 3574 members in Balikpapan, East

Kalimantan (45.5%).

In Semarang, respondents with informal sector

members who chosen third class of services were the

most participants (44%) except in Tlogosari Kulon

(30%) and the lowest participants chose first class

(19%). The amount of delinquency against total

participants at the baseline was 31% and increased

by 17% to 48% on the end-line. Participants who

chose first class of services became the highest

arrears followed by 2

and 3

class. In the baseline

data, villages with the highest total arrears were

Pudak Payung (38%) and lowest Krobokan (22%).

Then when viewed on the end-line data, the highest

arrears remain in Pudak Payung (54%) and lowest

Tlogosari (39%) and Krobokan (40%).

The rate of contribution collectability in the

Balikpapan city intervention area showed that

respondents in the informal sector members with the

third class of services were the largest participants

(61%) and the same for all villages, while the lowest

Piloting Collection Model of Health Insurance Contributions for Informal Sector Members

245

participants were in 1

class (16%) except in Batu

Ampar village by 23 %. Total Arrears to total

participants at the 24% baseline increased 22% to

45% on the end-line. Participants with 3

class

(23.54%) became the highest arrears and followed

by 2

class and 3

class of services. The highest

arrears on baseline were Batu Ampar village (36%)

and lowest Karang Rejo (19%) became the lowest

percentage of influence was Karang Joang (18%).

4 DISCUSSION

Based on the result of chi square test with 95%

confidence level, it is found that the condition of the

respondent when registering as a participant of BPJS

Health in Semarang has an effect on the incidence of

arrears (P value = 0,002) with the highest proportion

57,9% respondents in this city registered in sick

condition. Respondents in sick conditions tended to

have dues delinquency, they do not continue to pay

dues when they have received services and returned

to health. This is allegedly because participants who

enrolled in good health tended to have sufficient

knowledge of the concept and philosophy of health

insurance for their lives, while those in sickness

tended to register due to their current needs. This

variable does not provide a meaningful relationship

to contribution in Balikpapan, however there is an

odds ratio value that can illustrate the role of this

variable against the risk of contributions arrears.

From the results of the odd ratios, the households of

informal workers in Semarang who are underwriters

are enrolled and / or their members are sickly, 2.1

times are at higher risk of delinquent JKN

contributions compared to those enrolled when they

are healthy. While in Balikpapan, informal sector

household whose underwriters are enrolled and/ or

their family members are sick, 1.4 times higher risk

of arrears JKN contributions compared to those

enrolled when healthy.

The contribution delinquency in Balikpapan also

was affected by number of family member (p values

0.000) and the highest proportion is households with

more than three persons in one family. So,

households with more than three persons in family

2.8 times higher risk of arrears JKN contributions.

Besides that, in Semarang the presence of

elderly has a significant relationship with the

incidence of arrears (P value = 0.0001), as well as in

the city of Balikpapan (P value = 0.0001). Therefore,

the informal sector households with elderly

household members or underwriters are heads of

families over the age of 58 should receive

government assistance. Not only about the

ownership of the elderly in the household, the

arrears of contribution in the informal sector

members in Balikpapan is also affected by the

ownership of children under five (P value = 0.006).

Toddlers are also a group of people who are

vulnerable to health problems from outside so that

the risk of exposure by disease agents to be high.

From the socioeconomic point of view, the

informal employment households in Semarang who

have catastrophic rates have a significant

relationship with incidents of arrears (P value =

0.034), as well as respondents in Balikpapan (P

value = 0.008).

Based on the intervention of the promoter agent,

it can be seen that the type of promoter agent has a

significant effect on delinquency status of informal

sector members in Semarang City with p values

0.000 and the proportion of participants in the lowest

arrears is the individual promoter agent of 37.5%.

Based on the analysis of multivariate regression,

participants of informal workers are supervised by

agents promoters working with primary health care

is 1.4 times more at risk for delinquent dues than

agents promoter individually while participants of

the informal workers who scouted agent promoter

who worked with village officers at risk 1.7 times

greater for delinquent JKN contributions compared

to individual promoter agencies.

Meanwhile, in Balikpapan the lowest proportion

of participants in arrears exists in individual

promoter agent by 44.4%. Based on a regression

analysis, informal sector members who had

supervised by promoter agent working with health

facilities and village officials 1 times greater risk of

delinquency than individual promoter agent.

One of the achievements of the promoter

agency's performance is the role of the agent in

maintaining the level of college participants'

collective contribution, both routine and routine (not

in arrears). Below shows the result that the presence

of a promoter agency in the middle of the informal

worker of BPJS Health in Semarang made almost

61% of the informal workers' households of the pilot

project participants still pay regularly. While in the

control area, where the fee payment model is

passively conditioned on the basis of the existing

payment channel, there are 30.77% of the informal

worker's households who remain delinquent in

payment of their health insurance contributions.

While in Balikpapan, the presence of promoter

agency in the middle of informal sector members of

BPJS Health keeps more than 50% of informal

sector households routinely pay from informal

INAHEA 2017 - 4th Annual Meeting of the Indonesian Health Economics Association

246

workers. Whereas in the control area, where the fee

payment model is passively conditioned on the basis

of the existing channelling payment channel

(existing channel), there are 22.54% of the informal

worker's households who remain delinquent

payment of their health insurance contributions.

In terms of cost ratio in the implementation of

contribution collection by promoter agent, it can be

seen that the model of individual promoter agent in

Semarang City has the lowest cost ratio that is 1.57

whereas the biggest cost ratio in Semarang City is

promoter agent in cooperation with village officer

with the amount of 2.44. In the city of Balikpapan,

lowest expense ratio is the promoter agent who

cooperated with village officials with a ratio of 3.51

while the ratio of the largest costs in Balikpapan is

the agents who cooperate with healthcare facilities

with the amount of 3.98.

5 CONCLUSION

1. In general, all areas of both intervention and

control occur to increase arrears;

2. The effectiveness of models differs between two

areas: Individual Agent Model is more

meaningful in Semarang. While Balikpapan

does not have a meaningful model in the

increase of collectability;

3. Costs ratio is greater than the benefits so that

the cost of benefits is not efficient;

4. There is no general profile of agents that can

indicate performance fee collection: Semarang:

women and agents as main income, while in

Balikpapan: Male, age above 30 years, married,

has experience and this job as the main income.

5. Profile agent required to perform their functions

are have active experience in social activities so

that they can formulate strategies in the field;

having a high social spirit and good

communication skill; know the area very well,

so it can map the participants quickly and can

be accepted by the local community well.

6. The main obstacle is the presence of household

heads who are hard to find because of work

and are not willing to meet, the difficulties of

the agents coming to the night especially for

the female agent, for the housing area are not

given access to meet the families of informal

worker participants, resident participants, and

in rural areas difficult geographically

reachable.

REFERENCES

DJSN. 2012. Peta Jalan Menuju Jaminan Kesehatan

Nasional Tahun 2012-2019.

Fraenkel, Jack R dan Norman E. Wallen. 2006. How

to Design and Evaluate Research in Education.

New York: McGraw-Hill Inc.

Green, L. The PRECEDE-PROCEED Framework.

Retrieved March 21, 2016, from Faculty of

Medicine, Ottawa University:

http://www.med.uottawa.ca/courses/epi6181/im

ages/PRECEDE.pdf

Hasbullah Thabrany, Materi Sosialisasi Jaminan

Kesehatan, FKM UI, 2013.

ILO., 2009. Memperluas Cakupan Jaminan Sosial

Bagi Pekerja Sektor Perekonomian Informal. ILO.

(2010). Jaminan Sosial Bagi Tenaga Kerja di

Sektor Perekonomian Informal di Indonesia.

Kusumasari, A., & Widiastuti, N. E., 2013.

Kesadaran Wajib Pajak dari Sudut Tingkat

Pendidikan, Jenis Pekerjaan, Tingkat

Penghasilan, dan Moderenisasi Sistem

Perpajakan. Prosiding Simposium Perpajakan

Nasional 4. Jakarta.

Kwon, Soonman. 2008. Thirty Years of National

Health Insurance in South Korea: Lessons for

Achieving Universal Health Care Coverage.

http://heapol.oxfordjournals.org/

Law no. 24 Year 2011 on the Social Security

Administering Agency (BPJS).

Law no. 40 of 2004 on the National Social Security

System (SJSN). Law no. 3 Year 2011 on Fund

Transfers. World Bank. (2010). Employment

Reports in Indonesia

Margono, S., 2005. Metodologi Penelitian

Pendidikan. Jakarta: PT. Rineka Cipta

Pemaparan “Progress Kepesertaan BPJS Kesehatan

Sektor Informal” oleh Plt Direktur Perencanaan

Pengembangan dan Manajemen Risiko BPJS

Kesehatan, 13 Januari 2016 di Jakarta

Peraturan Presiden Nomor 111 tahun 2013 tentang

Jaminan Kesehatan.

Sedarmayanti and Syarifudin Hidayat. 2002.

Research methodology. Bandung: Mandar

Swaminathan 2009. Measuring The Informal

Economy in Developing Countries.

Zuriah, Nurul. 2006. Social and Educational

Research Methodology. Jakarta: Earth Literacy

Piloting Collection Model of Health Insurance Contributions for Informal Sector Members

247