Example: National Ambulatory Medical Care Survey (NAMCS) tables

This example uses the National Ambulatory Medical Care Survey (NAMCS) 2019 Public Use File (PUF) to replicate certain tables from the National Ambulatory Medical Care Survey: 2019 National Summary Tables. NAMCS is “an annual nationally representative sample survey of visits to non-federal office-based patient care physicians, excluding anesthesiologists, radiologists, and pathologists.” Note that the unit of observation is visits, not patients – this distinction is important since a single patient can make multiple visits.

Selected variables from NAMCS 2019 come with the surveytable package, for use in examples, in an object called namcs2019sv.

Begin

Begin by loading the surveytable package.

library(surveytable)

Now, specify the survey that you’d like to analyze.

set_survey(namcs2019sv)

Survey info {NAMCS 2019 PUF}
Variables	Observations	Design
33	8,250	Stratified 1 - level Cluster Sampling design (with replacement) With (398) clusters. namcs2019sv = survey::svydesign(ids = ~CPSUM, strata = ~CSTRATM, weights = ~PATWT , data = namcs2019sv_df)

Check the survey name, survey design variables, and the number of observations to verify that it all looks correct.

For this example, we do want to turn on certain NCHS-specific options, such as identifying low-precision estimates. If you do not care about identifying low-precision estimates, you can skip this command. To turn on the NCHS-specific options:

set_opts(mode = "NCHS")
## * Mode: NCHS.

Table 1

Counts and percentages

This table shows the overall estimated count as well as the counts and percentages by type of doctor, physician specialty, and metropolitan statistical area.

The variables that are necessary for creating this table are already in the survey, making the commands very straightforward.

total()

Total {NAMCS 2019 PUF}
n	Number (000)	SE (000)	LL (000)	UL (000)
8,250	1,036,484	48,836	945,014	1,136,809
N = 8250. Checked NCHS presentation standards. Nothing to report.

tab("MDDO", "SPECCAT", "MSA")

Type of doctor (MD or DO) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
M.D. - Doctor of Medicine	7,498	980,280	48,388	889,842	1,079,910	94.6	0.7	93.1	95.8
D.O. - Doctor of Osteopathy	752	56,204	6,602	44,597	70,832	5.4	0.7	4.2	6.9
N = 8250. Checked NCHS presentation standards. Nothing to report.

Type of specialty (Primary, Medical, Surgical) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Primary care specialty	2,993	521,466	31,136	463,840	586,252	50.3	2.6	45.1	55.5
Surgical care specialty	3,050	214,832	31,110	161,661	285,490	20.7	3.0	15.1	27.3
Medical care specialty	2,207	300,186	43,497	225,806	399,067	29.0	3.6	22.1	36.6
N = 8250. Checked NCHS presentation standards. Nothing to report.

Metropolitan Statistical Area Status of physician location {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
MSA (Metropolitan Statistical Area)	7,496	973,676	50,515	879,490	1,077,947	93.9	1.7	89.6	96.8
Non-MSA	754	62,809	17,549	36,249	108,830	6.1	1.7	3.2	10.4
N = 8250. Checked NCHS presentation standards. Nothing to report.

Rates

The published table also shows several rates. To calculate rates, in addition to the survey, we need a source of information with population estimates. You would typically use a function such as read.csv() to load the population estimates and get them into the correct format. The surveytable package comes with an object called uspop2019 that contains several population estimates for use in these examples.

class(uspop2019)
## [1] "list"
names(uspop2019)
## [1] "total"       "MSA"         "AGER"        "Age group"   "SEX"        
## [6] "AGER x SEX"  "Age group 5"

Here is the overall population estimate:

uspop2019$total
## [1] 323186697

Once we have the overall population estimate, the overall rate is:

total_rate(uspop2019$total)

Total (rate per 100 population) {NAMCS 2019 PUF}
n	Rate	SE	LL	UL
8,250	321	15	292	352
N = 8250. Checked NCHS presentation standards. Nothing to report.

To calculate the rates for a particular variable, we need to provide a data frame with a variable called Level that matches the levels of the variable in the survey, and a variable called Population that gives the population size (which is assumed to be a constant rather than a random variable).

For MSA, we can see the levels of the variables just by using the tab() command, just as we did above. Thus, to calculate rates, we need a data frame as follows:

uspop2019$MSA
##                                 Level Population
## 1 MSA (Metropolitan Statistical Area)  277229518
## 2                             Non-MSA   45957179

Now that we have the appropriate population estimates, the rate is:

tab_rate("MSA", uspop2019$MSA)

Metropolitan Statistical Area Status of physician location (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
MSA (Metropolitan Statistical Area)	7,496	351	18	317	389
Non-MSA	754	137	38	79	237
N = 8250. Checked NCHS presentation standards. Nothing to report.

We can also calculate rates of a specific variable based on the entire population:

tab_rate("MDDO", uspop2019$total)
## * Rate based on the entire population.

Type of doctor (MD or DO) (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
M.D. - Doctor of Medicine	7,498	303	15	275	334
D.O. - Doctor of Osteopathy	752	17	2	14	22
N = 8250. Checked NCHS presentation standards. Nothing to report.

tab_rate("SPECCAT", uspop2019$total)
## * Rate based on the entire population.

Type of specialty (Primary, Medical, Surgical) (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
Primary care specialty	2,993	161	10	144	181
Surgical care specialty	3,050	66	10	50	88
Medical care specialty	2,207	93	14	70	124
N = 8250. Checked NCHS presentation standards. Nothing to report.

Table 3

Counts and percentages

This table presents estimates for each age group, as well as for each age group by sex.

var_list("age")

Variables beginning with ‘age’ {NAMCS 2019 PUF}
Variable	Class	Long name
AGE	numeric	Patient age in years (raw - use caution)
AGER	factor	Patient age recode

The survey has a couple of relevant age-related variables. AGE is the patient age in years. AGER is a categorical variable based on AGE. However, for this table, in addition to AGER, we need another age group variable, with different age categories. We create it using the var_cut function.

var_cut("Age group", "AGE"
        , c(-Inf, 0, 4, 14, 64, Inf)
        , c("Under 1", "1-4", "5-14", "15-64", "65 and over") )

Now that we’ve created the Age group variable, we can create the tables:

tab("AGER", "Age group", "SEX")

Patient age recode {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 15 years	887	117,917	14,097	93,229	149,142	11.4	1.3	8.9	14.2
15-24 years	542	64,856	7,018	52,387	80,292	6.3	0.6	5.1	7.5
25-44 years	1,435	170,271	13,966	144,925	200,049	16.4	1.1	14.3	18.8
45-64 years	2,283	309,506	23,290	266,994	358,787	29.9	1.4	27.2	32.6
65-74 years	1,661	206,866	14,366	180,481	237,109	20.0	1.2	17.6	22.5
75 years and over	1,442	167,069	15,179	139,746	199,735	16.1	1.3	13.7	18.8
N = 8250. Checked NCHS presentation standards. Nothing to report.

Age group {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 1	203	31,148	5,282	22,269	43,566	3.0	0.5	2.1	4.1
1-4	281	38,240	5,444	28,864	50,662	3.7	0.5	2.7	4.8
5-14	403	48,529	5,741	38,430	61,282	4.7	0.5	3.7	5.9
15-64	4,260	544,632	36,082	478,254	620,223	52.5	2.0	48.6	56.5
65 and over	3,103	373,935	24,523	328,777	425,296	36.1	1.9	32.3	40.0
N = 8250. Checked NCHS presentation standards. Nothing to report.

Patient sex {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Female	4,609	605,045	34,866	540,377	677,452	58.4	1.9	54.6	62.1
Male	3,641	431,439	27,664	380,436	489,280	41.6	1.9	37.9	45.4
N = 8250. Checked NCHS presentation standards. Nothing to report.

tab_cross("AGER", "SEX")

(Patient age recode) x (Patient sex) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 15 years: Female	434	59,958	7,206	47,318	75,974	5.8	0.7	4.5	7.3
15-24 years: Female	346	41,128	4,532	33,066	51,156	4.0	0.4	3.2	4.9
25-44 years: Female	923	113,708	11,461	93,256	138,646	11.0	1.0	9.0	13.2
45-64 years: Female	1,253	175,978	16,009	147,153	210,450	17.0	1.1	14.8	19.3
65-74 years: Female	891	120,099	11,066	100,171	143,992	11.6	1.0	9.7	13.7
75 years and over: Female	762	94,173	11,085	74,682	118,751	9.1	0.9	7.3	11.1
Under 15 years: Male	453	57,959	7,728	44,570	75,371	5.6	0.7	4.3	7.2
15-24 years: Male	196	23,728	4,344	16,457	34,210	2.3	0.4	1.6	3.2
25-44 years: Male	512	56,562	7,277	43,861	72,942	5.5	0.6	4.3	6.8
45-64 years: Male	1,030	133,528	12,956	110,319	161,619	12.9	1.0	10.9	15.1
65-74 years: Male	770	86,766	6,767	74,409	101,176	8.4	0.6	7.2	9.7
75 years and over: Male	680	72,896	6,661	60,872	87,296	7.0	0.6	5.9	8.3
N = 8250. Checked NCHS presentation standards. Nothing to report.

Rates

tab_rate("AGER", uspop2019$AGER)

Patient age recode (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
Under 15 years	887	195	23	154	246
15-24 years	542	156	17	126	192
25-44 years	1,435	199	16	169	234
45-64 years	2,283	375	28	323	435
65-74 years	1,661	662	46	577	758
75 years and over	1,442	776	70	649	928
N = 8250. Checked NCHS presentation standards. Nothing to report.

tab_rate("Age group", uspop2019$`Age group`)
## * Population for some levels not defined: 15-64

Age group (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
Under 1	203	824	140	589	1,152
1-4	281	242	34	183	321
5-14	403	118	14	94	150
15-64	4,260	NA	NA	NA	NA
65 and over	3,103	708	46	623	806
N = 8250. Checked NCHS presentation standards. Nothing to report.

tab_rate("SEX", uspop2019$SEX)

Patient sex (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
Female	4,609	366	21	327	410
Male	3,641	273	18	241	310
N = 8250. Checked NCHS presentation standards. Nothing to report.

To calculate the rates for one variable (AGER) by another variable (SEX), we need population estimates in the following format:

uspop2019$`AGER x SEX`
##                Level Subset Population
## 1     Under 15 years Female   29604762
## 2        15-24 years Female   20730118
## 3        25-44 years Female   43192143
## 4        45-64 years Female   42508901
## 5        65-74 years Female   16673240
## 6  75 years and over Female   12421444
## 7     Under 15 years   Male   30921894
## 8        15-24 years   Male   20988582
## 9        25-44 years   Male   42407267
## 10       45-64 years   Male   40053148
## 11       65-74 years   Male   14586962
## 12 75 years and over   Male    9098236

Once we have these population estimates, the rates are:

tab_subset_rate("AGER", "SEX", uspop2019$`AGER x SEX`)

Patient age recode (Patient sex = Female) (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
Under 15 years	434	202	24	160	257
15-24 years	346	198	22	160	247
25-44 years	923	263	26	216	321
45-64 years	1,253	414	38	346	495
65-74 years	891	720	66	601	864
75 years and over	762	758	89	601	956
N = 4609. Checked NCHS presentation standards. Nothing to report.

Patient age recode (Patient sex = Male) (rate per 100 population) {NAMCS 2019 PUF}
Level	n	Rate	SE	LL	UL
Under 15 years	453	187	25	144	244
15-24 years	196	113	21	78	163
25-44 years	512	133	17	103	172
45-64 years	1,030	333	32	275	404
65-74 years	770	595	46	510	694
75 years and over	680	801	73	669	960
N = 3641. Checked NCHS presentation standards. Nothing to report.

Table 5

This table gives the expected sources of payment. We use the PAY* variables to create several new variables that are required by the table. Note that the PAY* variables are logical (TRUE or FALSE), which simplifies the workflow. (The survey was imported into R using the importsurvey package, which automatically detects binary variables and imports them as logical variables.)

#
var_all("Medicare and Medicaid", c("PAYMCARE", "PAYMCAID"))

#
var_any("Payment used", c("PAYPRIV", "PAYMCARE", "PAYMCAID"
  , "PAYWKCMP", "PAYOTH", "PAYDK"))
var_not("No other payment used", "Payment used")

var_all("Self-pay", c("PAYSELF", "No other payment used"))
var_all("No charge", c("PAYNOCHG", "No other payment used"))
var_any("No insurance", c("Self-pay", "No charge"))

#
var_case("No pay", "NOPAY", "No categories marked")
var_any("Unknown or blank", c("PAYDK", "No pay"))

##
tab("PAYPRIV", "PAYMCARE", "PAYMCAID", "Medicare and Medicaid"
  , "No insurance", "Self-pay", "No charge"
  , "PAYWKCMP", "PAYOTH", "Unknown or blank")

Expected source of payment for visit: Private insurance {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
FALSE	3,395	436,331	38,863	366,358	519,669	42.1	2.8	36.5	47.8
TRUE	4,855	600,153	35,912	533,694	674,888	57.9	2.8	52.2	63.5
N = 8250. Checked NCHS presentation standards. Nothing to report.

Expected source of payment for visit: Medicare {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
FALSE	5,539	717,764	40,947	641,784	802,739	69.2	2	65.1	73.2
TRUE	2,711	318,721	24,814	273,557	371,341	30.8	2	26.8	34.9
N = 8250. Checked NCHS presentation standards. Nothing to report.

Expected source of payment for visit: Medicaid or CHIP or other state-based program {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
FALSE	7,223	894,590	45,335	809,962	988,061	86.3	1.7	82.6	89.5
TRUE	1,027	141,894	19,039	108,992	184,728	13.7	1.7	10.5	17.4
N = 8250. Checked NCHS presentation standards. Nothing to report.

Medicare and Medicaid {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
FALSE	8,126	1,016,202	47,395	927,389	1,113,520	98	0.5	96.9	98.9
TRUE	124	20,282	5,177	12,120	33,941	2	0.5	1.1	3.1
N = 8250. Checked NCHS presentation standards. Nothing to report.

No insurance {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL	Flags
FALSE	7,945	994,579	50,446	900,421	1,098,583	96	2.2	89.1	99.1	Pc
TRUE	305	41,906	22,733	14,133	124,256	4	2.2	0.9	10.9	Cx Px
N = 8250. Checked NCHS presentation standards: Cx: suppress count (and rate); Px: suppress percent; Pc: footnote percent - complement.

Self-pay {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL	Flags
FALSE	7,951	994,770	50,446	900,611	1,098,772	96	2.2	89.1	99.1	Pc
TRUE	299	41,715	22,734	13,995	124,337	4	2.2	0.9	10.9	Cx Px
N = 8250. Checked NCHS presentation standards: Cx: suppress count (and rate); Px: suppress percent; Pc: footnote percent - complement.

No charge {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL	Flags
FALSE	8,243	1,036,081	48,837	944,610	1,136,409	100	0	99.9	100.0
TRUE	7	404	238	38	4,293	0	0	0.0	0.1	Cx
N = 8250. Checked NCHS presentation standards: Cx: suppress count (and rate).

Expected source of payment for visit: Workers Compensation {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL	Flags
FALSE	8,148	994,493	47,949	904,779	1,093,103	95.9	3.2	84.1	99.7	Pc
TRUE	102	41,991	33,472	8,413	209,596	4.1	3.2	0.3	15.9	Cx Px
N = 8250. Checked NCHS presentation standards: Cx: suppress count (and rate); Px: suppress percent; Pc: footnote percent - complement.

Expected source of payment for visit: Other {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
FALSE	8,077	1,017,442	47,898	927,725	1,115,836	98.2	0.3	97.5	98.7
TRUE	173	19,042	3,156	13,641	26,581	1.8	0.3	1.3	2.5
N = 8250. Checked NCHS presentation standards. Nothing to report.

Unknown or blank {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
FALSE	7,879	1,004,918	49,232	912,871	1,106,245	97	0.6	95.4	98.1
TRUE	371	31,567	6,533	20,907	47,661	3	0.6	1.9	4.6
N = 8250. Checked NCHS presentation standards. Nothing to report.

Check the presentation standards flags! Under NCHS presentation standards rules, some of these estimates should not be shown.

Table 6

This table shows the primary care provider and referral status, by prior-visit status.

In the table, the “Unknown” and “Blank” values are collapsed into a single value. We can collapse two or more levels of a factor into a single level using the var_collapse function.

var_collapse("PRIMCARE", "Unknown if PCP", c("Unknown", "Blank"))
var_collapse("REFER", "Unknown if referred", c("Unknown", "Blank"))

Now, for the table:

tab("PRIMCARE", "REFER", "SENBEFOR")

Are you the patient’s primary care provider? {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Unknown if PCP	316	40,669	9,479	25,619	64,560	3.9	0.9	2.4	6.1
Yes	2,278	383,481	28,555	331,362	443,798	37.0	2.6	31.9	42.3
No	5,656	612,335	43,282	533,050	703,413	59.1	2.5	53.9	64.1
N = 8250. Checked NCHS presentation standards. Nothing to report.

Was patient referred for visit? {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Unknown if referred	874	87,560	14,725	62,867	121,951	8.4	1.5	5.7	11.9
Not applicable	2,278	383,481	28,555	331,362	443,798	37.0	2.6	31.9	42.3
Yes	2,134	264,044	34,909	203,623	342,394	25.5	2.8	20.1	31.5
No	2,964	301,400	30,918	246,436	368,622	29.1	2.6	24.1	34.4
N = 8250. Checked NCHS presentation standards. Nothing to report.

Has this patient been seen in your practice before? {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Yes, established patient	6,771	862,626	43,142	782,038	951,518	83.2	1.4	80.3	85.8
No, new patient	1,479	173,859	16,338	144,536	209,130	16.8	1.4	14.2	19.7
N = 8250. Checked NCHS presentation standards. Nothing to report.

The percentages within each subset that is defined by SENBEFOR add up to 100% – for this reason, we want to use tab_subset(), not tab_cross().

tab_subset("PRIMCARE", "SENBEFOR")

Are you the patient’s primary care provider? (Has this patient been seen in your practice before? = Yes, established patient) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Unknown if PCP	236	28,494	7,930	16,343	49,679	3.3	0.9	1.8	5.6
Yes	2,127	359,164	26,752	310,334	415,677	41.6	2.8	36.2	47.3
No	4,408	474,967	36,421	408,617	552,092	55.1	2.7	49.6	60.4
N = 6771. Checked NCHS presentation standards. Nothing to report.

Are you the patient’s primary care provider? (Has this patient been seen in your practice before? = No, new patient) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL	Flags
Unknown if PCP	80	12,174	4,817	5,488	27,005	7	2.6	2.8	14.0	Cx Px
Yes	151	24,317	4,402	16,964	34,855	14	2.6	9.2	20.0
No	1,248	137,368	14,568	111,497	169,241	79	3.2	71.8	85.1
N = 1479. Checked NCHS presentation standards: Cx: suppress count (and rate); Px: suppress percent.

tab_subset("REFER", "SENBEFOR")

Was patient referred for visit? (Has this patient been seen in your practice before? = Yes, established patient) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Unknown if referred	606	58,208	11,757	39,052	86,761	6.7	1.4	4.2	10.1
Not applicable	2,127	359,164	26,752	310,334	415,677	41.6	2.8	36.2	47.3
Yes	1,324	172,899	28,658	124,751	239,629	20.0	2.9	14.7	26.4
No	2,714	272,354	27,253	223,783	331,468	31.6	2.7	26.3	37.3
N = 6771. Checked NCHS presentation standards. Nothing to report.

Was patient referred for visit? (Has this patient been seen in your practice before? = No, new patient) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Unknown if referred	268	29,351	6,343	19,106	45,091	16.9	3.5	10.6	25.0
Not applicable	151	24,317	4,402	16,964	34,855	14.0	2.6	9.2	20.0
Yes	810	91,145	13,090	68,656	121,000	52.4	5.1	41.9	62.8
No	250	29,046	7,757	17,084	49,382	16.7	4.0	9.6	26.2
N = 1479. Checked NCHS presentation standards. Nothing to report.

Table 11

This table shows the same information as Table 3, but only for preventive care visits. That is, estimates for each age group, as well as for each age group by sex, but only for preventive care visits.

Let’s create Age group from AGE and cross AGER and SEX to create a variable called Age x Sex:

var_cut("Age group", "AGE"
        , c(-Inf, 0, 4, 14, 64, Inf)
        , c("Under 1", "1-4", "5-14", "15-64", "65 and over") )
## Warning in var_cut("Age group", "AGE", c(-Inf, 0, 4, 14, 64, Inf), c("Under 1",
## : Age group: overwriting a variable that already exists.
var_cross("Age x Sex", "AGER", "SEX")

To see the possible values of MAJOR (Major reason for this visit), and to estimate the total count for preventive care visits:

tab("MAJOR")

Major reason for this visit {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Blank	175	15,887	3,354	10,335	24,419	1.5	0.3	1.0	2.3
New problem (less than 3 mos. onset)	2,193	275,014	19,691	238,955	316,514	26.5	1.5	23.7	29.5
Chronic problem, routine	2,861	380,910	35,080	317,916	456,386	36.8	2.5	31.8	41.9
Chronic problem, flare-up	635	74,017	9,329	57,706	94,939	7.1	0.9	5.5	9.1
Pre-surgery	159	12,864	2,151	9,188	18,010	1.2	0.2	0.9	1.7
Post-surgery	659	54,170	6,749	42,350	69,289	5.2	0.7	4.0	6.7
Preventive care	1,568	223,624	18,520	190,068	263,103	21.6	1.7	18.3	25.1
N = 8250. Checked NCHS presentation standards. Nothing to report.

To create the tables of age, sex, and their interaction, and limit them to only the preventive care visits:

tab_subset("AGER", "MAJOR", "Preventive care")

Patient age recode (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 15 years	300	50,701	8,556	36,352	70,714	22.7	3.5	16.1	30.4
15-24 years	121	18,196	2,889	13,246	24,996	8.1	1.2	5.9	10.9
25-44 years	370	50,573	6,835	38,749	66,005	22.6	2.5	17.8	28.0
45-64 years	355	53,805	9,478	37,982	76,218	24.1	3.2	17.9	31.1
65-74 years	225	27,985	4,669	20,073	39,017	12.5	1.8	9.2	16.5
75 years and over	197	22,363	3,805	15,925	31,404	10.0	1.7	6.9	13.8
N = 1568. Checked NCHS presentation standards. Nothing to report.

tab_subset("Age group", "MAJOR", "Preventive care")

Age group (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 1	118	19,094	3,996	12,601	28,932	8.5	1.7	5.4	12.6
1-4	86	14,819	3,149	9,676	22,695	6.6	1.3	4.3	9.7
5-14	96	16,788	3,524	11,035	25,542	7.5	1.5	4.9	11.0
15-64	846	122,574	13,515	98,688	152,242	54.8	3.3	48.1	61.4
65 and over	422	50,349	6,451	39,083	64,861	22.5	2.4	17.8	27.8
N = 1568. Checked NCHS presentation standards. Nothing to report.

tab_subset("SEX", "MAJOR", "Preventive care")

Patient sex (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Female	1,014	139,091	11,845	117,664	164,421	62.2	2.9	56.2	68.0
Male	554	84,532	10,594	66,039	108,204	37.8	2.9	32.0	43.8
N = 1568. Checked NCHS presentation standards. Nothing to report.

tab_subset("Age x Sex", "MAJOR", "Preventive care")

Age x Sex (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 15 years: Female	157	28,138	4,766	20,106	39,378	12.6	2.0	8.9	17.1
15-24 years: Female	96	12,866	2,336	8,936	18,525	5.8	1.0	3.9	8.1
25-44 years: Female	305	40,612	6,177	30,087	54,819	18.2	2.5	13.5	23.7
45-64 years: Female	233	31,373	4,986	22,889	43,001	14.0	1.8	10.7	17.9
65-74 years: Female	119	13,842	2,948	9,006	21,275	6.2	1.2	4.1	9.0
75 years and over: Female	104	12,259	2,228	8,497	17,687	5.5	1.0	3.6	7.9
Under 15 years: Male	143	22,563	4,501	15,195	33,505	10.1	1.8	6.7	14.4
15-24 years: Male	25	5,330	1,615	2,751	10,328	2.4	0.7	1.2	4.3
25-44 years: Male	65	9,961	2,607	5,849	16,964	4.5	1.1	2.6	7.1
45-64 years: Male	122	22,432	5,911	13,196	38,130	10.0	2.3	6.0	15.5
65-74 years: Male	106	14,143	2,611	9,716	20,587	6.3	1.0	4.4	8.7
75 years and over: Male	93	10,104	2,704	5,828	17,518	4.5	1.2	2.5	7.4
N = 1568. Checked NCHS presentation standards. Nothing to report.

As each of the above commands is similar, and differs only in the first variable that is passed to the tab_subset() function, this code can be streamlined with a for loop:

for (vr in c("AGER", "Age group", "SEX", "Age x Sex")) {
    print( tab_subset(vr, "MAJOR", "Preventive care") )
}

Patient age recode (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 15 years	300	50,701	8,556	36,352	70,714	22.7	3.5	16.1	30.4
15-24 years	121	18,196	2,889	13,246	24,996	8.1	1.2	5.9	10.9
25-44 years	370	50,573	6,835	38,749	66,005	22.6	2.5	17.8	28.0
45-64 years	355	53,805	9,478	37,982	76,218	24.1	3.2	17.9	31.1
65-74 years	225	27,985	4,669	20,073	39,017	12.5	1.8	9.2	16.5
75 years and over	197	22,363	3,805	15,925	31,404	10.0	1.7	6.9	13.8
N = 1568. Checked NCHS presentation standards. Nothing to report.

Age group (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 1	118	19,094	3,996	12,601	28,932	8.5	1.7	5.4	12.6
1-4	86	14,819	3,149	9,676	22,695	6.6	1.3	4.3	9.7
5-14	96	16,788	3,524	11,035	25,542	7.5	1.5	4.9	11.0
15-64	846	122,574	13,515	98,688	152,242	54.8	3.3	48.1	61.4
65 and over	422	50,349	6,451	39,083	64,861	22.5	2.4	17.8	27.8
N = 1568. Checked NCHS presentation standards. Nothing to report.

Patient sex (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Female	1,014	139,091	11,845	117,664	164,421	62.2	2.9	56.2	68.0
Male	554	84,532	10,594	66,039	108,204	37.8	2.9	32.0	43.8
N = 1568. Checked NCHS presentation standards. Nothing to report.

Age x Sex (Major reason for this visit = Preventive care) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL
Under 15 years: Female	157	28,138	4,766	20,106	39,378	12.6	2.0	8.9	17.1
15-24 years: Female	96	12,866	2,336	8,936	18,525	5.8	1.0	3.9	8.1
25-44 years: Female	305	40,612	6,177	30,087	54,819	18.2	2.5	13.5	23.7
45-64 years: Female	233	31,373	4,986	22,889	43,001	14.0	1.8	10.7	17.9
65-74 years: Female	119	13,842	2,948	9,006	21,275	6.2	1.2	4.1	9.0
75 years and over: Female	104	12,259	2,228	8,497	17,687	5.5	1.0	3.6	7.9
Under 15 years: Male	143	22,563	4,501	15,195	33,505	10.1	1.8	6.7	14.4
15-24 years: Male	25	5,330	1,615	2,751	10,328	2.4	0.7	1.2	4.3
25-44 years: Male	65	9,961	2,607	5,849	16,964	4.5	1.1	2.6	7.1
45-64 years: Male	122	22,432	5,911	13,196	38,130	10.0	2.3	6.0	15.5
65-74 years: Male	106	14,143	2,611	9,716	20,587	6.3	1.0	4.4	8.7
75 years and over: Male	93	10,104	2,704	5,828	17,518	4.5	1.2	2.5	7.4
N = 1568. Checked NCHS presentation standards. Nothing to report.

Note that when called from inside a for loop, the print() function needs to be called explicitly.

More advanced coding

In addition, for each age-sex category, the published table shows the percentage of preventive care visits made to primary care physicians.

To calculate these percentages, a slightly more involved for loop is needed. Below is the code, followed by an explanation:

set_opts(output = "csv", file = "my_output.csv")

## * Printing to CSV file my_output.csv.

for (vr in c("AGER", "Age group", "SEX", "Age x Sex")) {
    var_cross("tmp", "MAJOR", vr)
    for (lvl in levels(surveytable:::env$survey$variables[,vr])) {
        print( tab_subset("SPECCAT", "tmp", paste0("Preventive care: ", lvl)) )
    }
}
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: Under 15 years) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 15-24 years) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 25-44 years) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 45-64 years) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 65-74 years) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 75 years and over) {NAMCS 2019 PUF} to CSV file my_output.csv.
## Warning in var_cross("tmp", "MAJOR", vr): tmp: overwriting a variable that
## already exists.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: Under 1) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 1-4) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 5-14) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 15-64) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 65 and over) {NAMCS 2019 PUF} to CSV file my_output.csv.
## Warning in var_cross("tmp", "MAJOR", vr): tmp: overwriting a variable that
## already exists.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: Female) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: Male) {NAMCS 2019 PUF} to CSV file my_output.csv.
## Warning in var_cross("tmp", "MAJOR", vr): tmp: overwriting a variable that
## already exists.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: Under 15 years: Female) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 15-24 years: Female) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 25-44 years: Female) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 45-64 years: Female) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 65-74 years: Female) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 75 years and over: Female) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: Under 15 years: Male) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 15-24 years: Male) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 25-44 years: Male) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 45-64 years: Male) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 65-74 years: Male) {NAMCS 2019 PUF} to CSV file my_output.csv.
## * Printing Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: 75 years and over: Male) {NAMCS 2019 PUF} to CSV file my_output.csv.
set_opts(output = "auto")
## * Printing with huxtable for screen, gt for HTML, or kableExtra for PDF.

• Because so many tables will be printed, we don’t want to print them to the screen. Rather, we print them to a CSV file called my_output.csv.
• Since tab_subset() is called from within a for loop, to print the tables, we need to call the print() function explicitly.
• As before, the loop goes through the age, sex, and age / sex interaction variables, calling each of these variables vr.
• MAJOR and vr are crossed, with the result stored in a variable called tmp.
• Next, the inner loop goes through all levels of vr, calling each of these levels lvl.
• The code tabulates SPECCAT (Type of specialty – Primary, Medical, Surgical) on a subset in which tmp (which is MAJOR crossed with vr) is restricted to "Preventive care: " followed by lvl, which is some level of vr, such as “Under 15 years” for AGER.
• Finally, CSV output is turned off.

If you run this code, all of the tables should be stored in the CSV file. To give you an idea of what the tables should look like, here is just one of the tables:

vr = "AGER"
var_cross("tmp", "MAJOR", vr)
## Warning in var_cross("tmp", "MAJOR", vr): tmp: overwriting a variable that
## already exists.
lvl = levels(surveytable:::env$survey$variables[,vr])[1]
tab_subset("SPECCAT", "tmp", paste0("Preventive care: ", lvl))

Type of specialty (Primary, Medical, Surgical) (tmp = Preventive care: Under 15 years) {NAMCS 2019 PUF}
Level	n	Number (000)	SE (000)	LL (000)	UL (000)	Percent	SE	LL	UL	Flags
Primary care specialty	289	49,978	8,682	35,483	70,395	98.6	1.2	93.3	99.9	Pc
Surgical care specialty	4	149	74	28	790	0.3	0.2	0.0	1.8	R Cx
Medical care specialty	7	574	574	50	6,628	1.1	1.2	0.0	7.0	Cx Px
N = 300. Checked NCHS presentation standards: R: If the data is confidential, suppress all estimates, SE’s, CI’s, etc.; Cx: suppress count (and rate); Px: suppress percent; Pc: footnote percent - complement.

To match the percentage in the published table, see the “Primary care specialty” row. Be sure to check the presentation standards flags.