Concept: Revised-Graduated Prenatal Care Utilization Index (R-GINDEX)

Introduction

The Revised-Graduated Prenatal Care Utilization Index (R-GINDEX) is a measure of the adequacy of prenatal care provided to a woman by healthcare providers during the prenatal period. The R-GINDEX was developed by Alexander and Kotelchuck while they were reviewing and comparing five different prenatal care indices that measure prenatal care utilization. The R-GINDEX is a revised version of the GINDEX and is "useful for research focusing on birth outcomes and for monitoring trends in the proportion of cases with intensive use of prenatal care" (Alexander GR and Kotelchuck M., 1996). It relies on case-specific prenatal care information, including the number of prenatal visits, gestational age of the newborn, and the date when prenatal care began.

This concept describes the data required for creating the R-GINDEX including detailed information on how these data elements are derived, the R-GINDEX outcomes and categories, where and how the R-GINDEX has been used in MCHP research, and a list of limitations and cautions relating to the index.

This concept also provides a link to relevant SAS code that was used in Brownell et al. (2010) for calculating the R-GINDEX values. This code is based on the work by Alexander and Kotelchuck, 1996. To view the SAS code, please see the SAS code and formats section below and click on the following link: Revised-Graduated Prenatal Care Utilization Index (R-GINDEX) SAS Code.

Data Required for Creating the R-GINDEX

In order to create an R-GINDEX value for each prenatal case, knowledge of the following three variables is required:


1. gestational age of the newborn (GEST);
   
2. the trimester that prenatal care began (TPCB); and
   
3. the total number of prenatal care visits (PCV) during pregnancy

This data can be found in the following two MCHP data sets: the Medical Services data and the Hospital Abstracts data. Some of the variables are available directly from the data and some can be derived from the data in different ways. The methods will need to be decided on by the research team based on the goals and objectives of the research project.

1. GEST - Gestational Age of the Newborn

The Hospital Abstracts data contains data identifying the gestation period (GEST) (in weeks). The Medical Services data does not contain information on the gestational period.

2. TPCB - Trimester that Prenatal Care Began

The TPCB is calculated from three separate variables: gestational age of the newborn, date of birth, and date of the first prenatal care visit. As indicated above, the Hospital Abstracts data contains the gestation period (GEST) (in weeks). Date of birth can also be found in the Hospital Abstracts data.

The date of first prenatal care visit can either be found directly in the Hospital Abstracts data (OBS_VISIT1_DT) or calculated as the earliest date among all the relevant prenatal care visits identified in the Medical Services data. For more information on identifying the date of the first prenatal care visit, please read the Prenatal Care Visits concept.

If there are no prenatal care visits, the value of TPCB will be 0 (zero). If there are prenatal care visits, then the TPCB value (1, 2 or 3) can be calculated using the following table:

Trimester	Weeks	Days
1	0-13	1-91
2	14-27	92-189
3	28 +	190 +

3. PCV - Total Number of Prenatal Care Visits

The total number of prenatal care visits can be found directly in the Hospital Abstracts data (PREN_VISITS) or calculated by counting all the relevant prenatal care visits identified in the Medical Services data. For more information on calculating the total number of prenatal care visits from the Medical Services data, please read the Prenatal Care Visits concept.

Note: A value of 0 (zero) is possible, indicating that no prenatal care was received.

Exclusions

In Brownell et al. (2010) they applied the following exclusions to the data. The first step produced the cohort of newborns (live births) that were linked to an OBPHIN from the Hospital Abstracts data with the gestation variable included.

The second step excluded (removed from the analysis) all records that had at least one of the following conditions:


• the newborn has a missing or out of range gestational age (ie. < 18 weeks or > 45 weeks);
  
• the obphin's maternal age is missing or < 12 years old;
  
• the obphin gave birth to a set of twins, triplets, etc.;
  
• the newborn had a birth weight < 400 grams, but gestation > 22 weeks;
  
• the obphin did not have 100% Manitoba Health Coverage during her pregnancy period (ie. pregnancy period = nbgestpb * 7); or
  
• the newborn cannot be linked to an obphin

NOTE: After cleaning the data, every newborn record will have a value for gestation.

R-GINDEX Outcomes and Categories

The R-GINDEX developed by Alexander and Kotelchuck has six major categories of care and within each category there are subcategories based on the trimester prenatal care began. Placement into a category/subcategory, is dependent on the values of the three variables described above:


• gestational age (in weeks);
  
• the trimester than prenatal care began (TPCB); and
  
• the total number of prenatal v=care visits (PCV).

The six major categories and subcategories of the R-GINDEX are:


1. Inadequate Prenatal Care Utilization
   1. 1st Trimester
   2. 2nd Trimester
   
   
2. Intermediate Prenatal Care Utilization
   1. 1st Trimester
   2. 2nd Trimester
   
   
3. Adequate Prenatal Care Utilization
   1. 1st Trimester
   
   
4. Intensive Prenatal Care
   1. 1st Trimester
      
   2. 2nd Trimester
   3. 3rd Trimester
   
   
5. No Care - this indicates that no prenatal care was provided - see Limitations and Cautions below for an explanation of why no prenatal care may be recorded.
   
   
6. Missing Care

For detailed information on the values used to calculate each R-GINDEX outcome, please see the Coding Algorithms for Revised GINDEX in the Appendix of the full-text article titled Quantifying the adequacy of prenatal care: a comparison of indices. by Alexander and Kotelchuck, 1996. The full text article is available through a link on the publication abstract listing available from the PubMed website.

MCHP Research Using the R-GINDEX

The following section identifies the MCHP deliverables that have used the R-GINDEX in their research and describe how it was used. Links to report-specific discussion and findings are also provided.

Brownell et al. (2010)

In the deliverable Evaluation of the Healthy Baby Program by Brownell et al. (2010), they used the R-GINDEX to investiagte two different populations of women and evaluate whether they received adequate care (compared to all other categories) or inadequate care (a combination of inadequate and no care compared to all other categories). The first population included all women giving birth in 2004/2005 through 2007/2008 who had applied for the Healthy Baby Prenatal Benefit during pregnancy and whose incomes were less than $40,000. The second population included all women giving birth in 2004/2005 through 2007/2008 who received income assistance for at least one month during pregnancy.

For more information, please read the discussion and findings on Adequate Prenatal Care and Inadequate Prenatal Care in this deliverable.

Heaman et al. (2012)

In the deliverable Perinatal Services and Outcomes in Manitoba by Heaman et al. (2012) they used the R-GINDEX to determine the proportion of women with no or inadequate prenatal care for the time period 2001/02 to 2008/09. The calculated rates are presented by Regional Health Authority (RHA), Winnipeg Community Areas (CA), and Sociodemographic and Other Characteristics.

For more information, please read the section Inadequate Prenatal Care in this deliverable.

Ruth et al. (2015)

In the deliverable Long-Term Outcomes Of Manitoba's Insight Mentoring Program: A Comparative Statistical Analysis by Ruth et al. (2015), they used the R-GINDEX to investigate levels of adequate prenatal care in three different time periods, comparing the participants of the Insight Mentoring Program with a group of women reporting alcohol use during pregnancy and receiving income assistance.

For more information, please read the section Indicators of Prenatal Care in this deliverable.

Fransoo et al. (2019)

In the deliverable The 2019 RHA Indicators Atlas by Fransoo et al. (2019), they investigated the adequacy of prenatal care in two different five-year time periods; 2007/08–2011/12 and 2012/13–2016/17.

For more information, please read the section 10.2 Inadequate Prenatal Care in this deliverable.

SAS Code

The SAS code used in this research, to develop the count of prenatal visits, is included in the SAS code and formats section below (internal access only).

Limitations and Cautions

The following limitations and cautions have been identified when using the R-GINDEX:


• The R-GINDEX is based on the ACOG (American College of Obstetricians and Gynecologists) recommendations for number of visits for low risk pregnant women; the effectiveness of this standard has not been assessed through rigorous scientific testing, nor has adequacy of care for women with high risk pregnancies been operationalized (Alexander & Kotelchuck, 2001). For the purpose of calculating the indicators of prenatal care in Heaman et al. (2012), a prenatal visit was defined as a visit to a health professional (i.e., physician, midwife or nurse practitioner) where some kind of medical or healthcare was performed to take care of the pregnancy. Other forms of prenatal health services were not included in this definition, such as attendance at prenatal classes or Healthy Baby Community Support Programs.
  
  
• The Medical Services data may not contain all of the relevant prenatal care visit claims. For example, some mothers that "receive" no prenatal care could simply be seeing salaried physicians who neglected to submit claims for prenatal care visits (shadow-billing). In addition, their prenatal care could have been provided by midwives and/or at nursing stations, and these types of services are not recorded in the Medical Services data.
  
  
• A concern with the SAS code provided to calculate R-GINDEX is that it can easily categorize prenatal care from inadequate to intensive over 1 visit. For example:
  
  • Woman A: with 12 visits starting 3rd trimester = Inadequate care
    
  • Woman B: with 13 visits starting 3rd trimester = Intensive care