Common use of Variance Estimation Clause in Contracts

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 and VARPSU97, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 and VARPSU97 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 for the estimated mean of out-of-pocket payment.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1998 data. Variables needed to implement a Xxxxxx series estimation approach are provided in the file and are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR98 and VARPSU97VARPSU98, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR98 and VARPSU97 VARPSU98 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the a computer software package SUDAAN will yield an estimate of standard error estimates of $6.01 48.38 and 0.0359 for the estimated mean of out-of-pocket paymentpayment and the estimated mean proportion of total expenditures paid by private insurance respectively.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1996 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR96 and VARPSU97VARPSU96, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section section 4.2 Using a Xxxxxx Series series approach, specifying VARSTR97 VARSTR96 and VARPSU97 VARPSU96 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 0.59 for the estimated mean of out-of-pocket payment. Example 3 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR96 and VARPSU96 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of 0.0091 for the weighted mean proportion of total expenditures paid by private insurance.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1996 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR96 and VARPSU97VARPSU96, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR96 and VARPSU97 VARPSU96 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 18.96 for the estimated mean of out-of-pocket payment. Example 3 from section 4.2 Using a Xxxxxx Series approach, specifying VARSTR96 and VARPSU96 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of 0.0159 for the weighted mean proportion of total expenditures paid by private insurance.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPSMEPS for both person-level and family-level analyses. Various approaches can be used to develop such estimates of variance including use of the a Xxxxxx series Series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables We will describe the variables needed to implement a Xxxxxx series Series estimation approach are described in the paragraph belowapproach. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARST13 and PSU13 on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates of variancebased on this MEPS database. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approachIn the past, specifying VARSTR97 and VARPSU97 as the MEPS variance estimation strata and PSUs (within these strata) respectively were developed independently from year to year. However, beginning with the 2002 Point-in-Time PUF, the variance strata and specifying a “PSUs have been developed to be compatible with replacement” design in all future PUFs. Thus, data from future years can be pooled and the computer software package SUDAAN will yield an estimate variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of standard error of $6.01 for the estimated mean of out-of-pocket paymentdata.

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Prior to 2002, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the year. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUFs until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The following illustrates these concepts using two examples from Section 4.2current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2015 dataset. Example 2 from Section 4.2 Using a Xxxxxx Series approachSpecifically, specifying VARSTR97 and VARPSU97 as the 165 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001- 1165 for the estimated mean 2015 PIT file. Beginning with the 2002 PIT database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC- 036BRR.)

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1999 data. Variables needed to implement a Xxxxxx series estimation approach are provided in the file and are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR99 and VARPSU97VARPSU99, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR99 and VARPSU97 VARPSU99 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the a computer software package SUDAAN will yield an estimate of standard error estimates of $6.01 0.4609 and 0.0067 for the estimated mean of out-of-pocket paymentpayment and the estimated mean proportion of total expenditures paid by private insurance respectively.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1999 data. Variables needed to implement a Xxxxxx series estimation approach are provided in the file and are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR99 and VARPSU97VARPSU99, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR99 and VARPSU97 VARPSU99 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the a computer software package SUDAAN will yield an estimate of standard error estimates of $6.01 5.31 and 0.0128 for the estimated mean of out-of-pocket paymentpayment and the estimated mean proportion of total expenditures paid by private insurance respectively.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPSMEPS for both person and family level analyses. Various approaches can be used to develop such estimates of variance including use of the a Xxxxxx series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables We will describe the variables needed to implement a Xxxxxx series estimation approach are described in the paragraph belowapproach. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARST13 and PSU13 on this MEPS data file (updated versions of corresponding variables on provided in previously released MEPS public use files) serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 and VARPSU97 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 for the estimated mean of out-of-pocket paymentbased on this MEPS database.

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Prior to 2002, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the year. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The following illustrates these concepts using two examples from Section 4.2current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2013 dataset. Example 2 from Section 4.2 Using a Xxxxxx Series approachSpecifically, specifying VARSTR97 and VARPSU97 as the 165 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001- 1165 for the estimated mean 2013 PIT file. Beginning with the 2002 PIT database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC- 036BRR.)

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 and VARPSU97, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series series approach, specifying VARSTR97 and VARPSU97 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 85.51 for the estimated mean of out-of-pocket payment.

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family-level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx- series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the rounds. The following illustrates these concepts using However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. Thus, when pooling data across years 2002 through the Panel 11 component of 2007 files, the variance strata and PSU variables provided can be used without modification (except for the renumbering necessary of the 2007 variance strata and PSU values, as discussed below) for variance estimation purposes for estimates covering multiple years of data. There were 203 variance estimation strata, each stratum with either two examples from Section 4.2or three variance estimation PSUs. Example 2 from Section 4.2 Using For the 2007 PIT data a Xxxxxx Series approachnew set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, specifying VARSTR97 and VARPSU97 as the 125 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. With MEPS Panels 11 and PSUs (within these strata) respectively 12 based on two NHIS independent samples, 328 variance strata are available for variance estimation purposes for the 2007 PIT data. Those numbered 1-125 are associated with Panel 12 while those numbered 126 through 328 are associated with Panel 11. This numbering system was used to make analyses across multiple years of MEPS data straightforward. Those using data from Panel 12 and specifying a “with replacement” design MEPS panels in the computer software package SUDAAN future can use the variance strata numbering from Panel 12 onward. For those using data from Panel 11 and earlier, the numbering from 1 to 203 can be used for years 2006 and earlier. To include Panel 11 data from 2007 MEPS databases, subtracting 125 from the values identifying the variance strata appearing on the 2007 databases will yield an estimate of standard error of $6.01 for produce the estimated mean of out-of-pocket paymentvariance strata that corresponded to those from earlier years.

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The following illustrates these concepts using two examples from Section 4.2current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2012 dataset. Example 2 from Section 4.2 Using a Xxxxxx Series approachSpecifically, specifying VARSTR97 and VARPSU97 as the 165 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001- 1165 for the estimated mean 2012 PIT file. Beginning with the 2002 PIT database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family-level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx- series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates of variancebased on this MEPS database. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approachIn the past, specifying VARSTR97 and VARPSU97 as the MEPS variance estimation strata and PSUs (within these strata) respectively were developed independently from year to year, and specifying a “the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with replacement” design the 2002 Point-in-Time PUF, the variance strata and PSUs have been developed to be compatible with all future PUFs, subject to changes in the computer software package SUDAAN will yield an estimate of standard error of $6.01 NHIS sample design (there has been a design change for the estimated mean 2006 NHIS which will affect MEPS for the first time in 2007). Thus, when pooling data across years (2002 and forward), the variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of out-of-pocket paymentdata. There are 203 variance estimation strata, each stratum with either two or three variance estimation PSUs.

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The following illustrates these concepts using two examples from Section 4.2current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2011 dataset. Example 2 from Section 4.2 Using a Xxxxxx Series approachSpecifically, specifying VARSTR97 and VARPSU97 as the 165 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001- 1165 for the estimated mean 2011 PIT file. Beginning with the 2002 PIT database 203 variance strata were formed for use in developing variance estimates for all subsequent years and databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC- 036BRR.)

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 and VARPSU97, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 and VARPSU97 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 23.08 for the estimated mean of out-of-pocket payment. Example 3 from section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 and VARPSU97 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of 0.0132 for the weighted mean proportion of total expenditures paid by private insurance.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 2000 data. Variables needed to implement a Xxxxxx series estimation approach are provided in the file and are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR00 and VARPSU97VARPSU00, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR00 and VARPSU97 VARPSU00 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the a computer software package SUDAAN will yield an estimate of standard error estimates of $6.01 0.4237 and 0.0074 for the estimated mean of out-of-pocket paymentpayment and the estimated mean proportion of total expenditures paid by private insurance respectively.

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Prior to 2002, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the year. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The following illustrates these concepts using two examples from Section 4.2current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2014 dataset. Example 2 from Section 4.2 Using a Xxxxxx Series approachSpecifically, specifying VARSTR97 and VARPSU97 as the 165 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001- 1165 for the estimated mean 2014 PIT file. Beginning with the 2002 PIT database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC- 036BRR.)

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 and VARPSU97, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series series approach, specifying VARSTR97 and VARPSU97 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 0.92 for the estimated mean of out-of-pocket payment.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPSMEPS for both person-level and family-level analyses. Various approaches can be used to develop such estimates of variance including use of the a Xxxxxx series Series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables We will describe the variables needed to implement a Xxxxxx series Series estimation approach are described in the paragraph belowapproach. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates of variancebased on this MEPS database. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approachIn the past, specifying VARSTR97 and VARPSU97 as the MEPS variance estimation strata and PSUs (within these were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2003 Point-in-Time PUF, the variance strata and PSUs have been developed to be compatible with all future PUFs. Thus, data from future years can be pooled and the variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of data. There are 203 variance estimation strata) respectively and specifying a “, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 for the estimated mean of out-of-pocket paymenteither two or three variance estimation PSUs.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPSMEPS for both person-level and family-level analyses. Various approaches can be used to develop such estimates of variance including use of the a Xxxxxx series Series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables We will describe the variables needed to implement a Xxxxxx series Series estimation approach are described in the paragraph belowapproach. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates of variancebased on this MEPS database. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approachIn the past, specifying VARSTR97 and VARPSU97 as the MEPS variance estimation strata and PSUs (within these were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs have been developed to be compatible with all future PUFs. Thus, data from future years can be pooled and the variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of data. There are 203 variance estimation strata) respectively and specifying a “, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 for the estimated mean of out-of-pocket paymenteither two or three variance estimation PSUs.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1999 data. Variables needed to implement a Xxxxxx series estimation approach are provided in the file and are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR99 and VARPSU97VARPSU99, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR99 and VARPSU97 VARPSU99 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” Awith replacement@ design in the a computer software package SUDAAN will yield an estimate of standard error estimates of $6.01 2.50 and 0.0075 for the estimated mean of out-of-pocket paymentpayment and the estimated mean proportion of total expenditures paid by private insurance respectively.

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the rounds. The following illustrates these concepts using two examples from Section 4.2However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. Example 2 from Section 4.2 Using As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a Xxxxxx Series approachnew set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, specifying VARSTR97 and VARPSU97 as the 125 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001-1125 for the estimated mean 2008 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The following illustrates these concepts using two examples from Section 4.2current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2010 dataset. Example 2 from Section 4.2 Using a Xxxxxx Series approachSpecifically, specifying VARSTR97 and VARPSU97 as the 165 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001- 1165 for the estimated mean 2010 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey dataestimates, one needs analysts need to take into account the complex sample design of MEPSMEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the Xxxxxx-series linearization method, balanced repeated replication, and jackknife replication. Various approaches can be used to develop such estimates software packages provide analysts with the capability of variance including use of the Xxxxxx series or various replication implementing these methodologies. Replicate weights have not been developed for the MEPS 1997 data. Variables Instead, the variables needed to implement a Xxxxxx calculate appropriate standard errors based on the Xxxxxx-series estimation approach linearization method are described in included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the paragraph belowuse of the Xxxxxx-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using a Xxxxxx Series approachthe Xxxxxx-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables VARSTR and VARPSU on this MEPS data file serve to identify the MEPS full year utilization database are VARSTR97 sampling strata and VARPSU97, respectivelyprimary sampling units required by the variance estimation programs. Specifying a “with replacement” design in a one of the previously mentioned computer software package such as SUDAAN (Xxxx, 1996) should packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally variables of interest distributed throughout the country (and thus the MEPS sample PSUs), there are over one can generally expect to have at least 100 degrees of freedom associated with the corresponding estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of variancethe strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The following illustrates these concepts using two examples from Section 4.2current set of variance strata and PSUs were re-established for the 2008 PIT data, and were carried over to the PIT 2009 data. Example 2 from Section 4.2 Using a Xxxxxx Series approachSpecifically, specifying VARSTR97 and VARPSU97 as the 165 variance estimation strata and PSUs (within these strata) respectively and specifying a “were created, each stratum with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 either two or three variance estimation PSUs. These have been numbered 1001-1165 for the estimated mean 2009 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of outthe variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-of2005, and thus associated with MEPS samples for MEPS Panels 1-pocket payment11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1996 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR96 and VARPSU97VARPSU96, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR96 and VARPSU97 VARPSU96 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 2.71 for the estimated mean of out-of-pocket payment. Example 3 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR96 and VARPSU96 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of 0.0118 for the weighted mean proportion of total expenditures paid by private insurance.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1998 data. Variables needed to implement a Xxxxxx series estimation approach are provided in the file and are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR98 and VARPSU97VARPSU98, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (XxxxXxxx et al, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR98 and VARPSU97 VARPSU98 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” replacement design in the a computer software package SUDAAN will yield an estimate of standard error estimates of $6.01 2.93 and 0.0080 for the estimated mean of out-of-pocket paymentpayment and the estimated mean proportion of total expenditures paid by private insurance respectively.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 1996 data. Variables needed to implement a Xxxxxx series estimation approach are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR96 and VARPSU97VARPSU96, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series series approach, specifying VARSTR97 VARSTR96 and VARPSU97 VARPSU96 as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the computer software package SUDAAN will yield an estimate of standard error of $6.01 136 for the estimated mean of out-of-pocket payment.

Variance Estimation. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take into account the complex sample design of MEPS. Various approaches can be used to develop such estimates of variance including use of the Xxxxxx series or various replication methodologies. Replicate weights have not been developed for the MEPS 1997 2003 data. Variables needed to implement a Xxxxxx series estimation approach are provided in the file and are described in the paragraph below. Using a Xxxxxx Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR97 VARSTR and VARPSU97VARPSU, respectively. Specifying a “with replacement” design in a computer software package such as SUDAAN (Xxxx, 1996) should provide standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples examp les from Section section 4.2. Example 2 from Section 4.2 Using a Xxxxxx Series approach, specifying VARSTR97 VARSTR and VARPSU97 VARPSU as the variance estimation strata and PSUs (within these strata) respectively and specifying a “with replacement” design in the a computer software package SUDAAN will yield an estimate of standard error estimates of $6.01 0.3927 and 0.0052 for the estimated mean of out-of-pocket paymentpayment and the estimated mean proportion of total expenditures paid by private insurance respectively.